Site Map

Please note:

You are viewing archival ICANN material. Links and information may be outdated or incorrect. Visit ICANN's main website for current information.

ICANN Meetings in Cape Town

Workshop: Internationalized Domain Name

Wednesday, December 1, 2004

Note: The following is the output of the real-time captioning taken during the IDN Workshop held on 1 December, 2004 in Cape Town, South Africa . Although the captioning output is largely accurate, in some cases it is incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the session, but should not be treated as an authoritative record.

>>PAUL TWOMEY: COLLEAGUES, I SUPPOSE, LET US CONVENE THE WORKSHOP ON INTERNATIONALIZED DOMAIN NAMES.
THERE ARE MORE PEOPLE MOVING THROUGH. OBVIOUSLY SOME OF THE OTHER MEETINGS ARE RUNNING A LITTLE BEHIND TIME, BUT IF WE DON'T START NOW WE WON'T BE FINISHED IN TIME FOR DINNER THIS EVENING.
SO WE HAVE, ACTUALLY -- I'LL BE ASKING VINT CERF, THE CHAIRMAN OF ICANN, TO GIVE US A GENERAL INTRODUCTION, BUT PERHAPS I CAN JUST TAKE US TO THE AGENDA HERE TODAY.
FIRST OF ALL, AS I THINK VINT WILL MAKE CLEAR, INTERNATIONALIZED DOMAIN NAMES IS A KEY ISSUE AND ONE WHICH WE ARE PARTICULARLY INTERESTED IN AND LOOKING TO CONTINUE TO ENGAGE THE COMMUNITY AND RAISE THE PROFILE OF AND ALSO ENGAGE ON WHAT THE ISSUES ARE FOR IMPLEMENTATION GOING FORWARD.
WE HAVE A THREE-PART SESSION HERE THIS AFTERNOON, AND ALL PARTS SORT OF FOCUS ON IDNS, PARTICULARLY IN AN AFRICAN CONTEXT.
THE FIRST WILL BE A SESSION ON THE DEVELOPMENT OF IDNS IN THE AFRICAN REGION, AND WE'LL HAVE AN EXPERT SERIES OF SPEAKERS SPEAKING TO US.
SECONDLY, THERE WILL BE A PANEL SPECIFICALLY DEALING WITH APPLICATION DEVELOPMENT, WHICH WAS ONE OF THE ISSUES BROUGHT UP QUITE A LOT IN THE KUALA LUMPUR WORKSHOP ON IDNS. AND THE -- THERE WILL BE A DISCUSSION BY A CORE PANEL ON THAT ACTUAL ISSUE OF APPLICATION DEVELOPMENT.
AND THEN THERE'S GOING TO BE A FINAL PANEL ON POLICY ISSUES AND PROCESSES.
I'D LIKE TO THANK ALL THE PEOPLE WHO WERE INVOLVED IN PUTTING THIS TOGETHER. THEY'VE ALL DONE A GREAT JOB. PARTICULARLY I'D LIKE TO THANK TINA DAM WHO IS IN THE AUDIENCE FOR THE WORK SHE'S DONE ON THE STAFF SIDE, ALSO THERESA SWINEHART.
PERHAPS, THEN, I'LL OPEN UP THE SESSION NOW TO THE FIRST INTRODUCTORY SPEECH FROM VINT.

(APPLAUSE.)
>>VINT CERF: WELL, THANK YOU VERY MUCH. I'M NOT SURE THAT I DESERVE ANY APPLAUSE AT ALL. OF COURSE, ONE IS TEMPTED, IF YOU'RE APPLAUDED AS YOU STAND UP, TO SIMPLY SIT DOWN BECAUSE IT DOESN'T GET ANY BETTER THAN THAT.
I CONSIDER MYSELF TO BE SOMETHING OF AN IMPOSTER STANDING UP HERE SPEAKING ON IDNS. THERE ARE PEOPLE IN THIS AUDIENCE AND PEOPLE WHO WILL BE ON THE PANELS WHO ARE INFINITELY BETTER PREPARED THAN I TO MAKE ANY PRONOUNCEMENTS AT ALL ABOUT IDNS. BUT THAT NEVER HAS STOPPED ME BEFORE, SO I DO HAVE A FEW REMARKS THAT I WOULD LIKE TO SHARE.
FIRST OF ALL, THERE IS STILL A GREAT DEAL OF TECHNICAL WORK THAT NEEDS TO BE DONE TO IMPLEMENT THE STANDARDS THAT HAVE BEEN DEVELOPED IN THE IETF TO SUPPORT UNICODE OR ITS ENCODING IN ASCII ON THE INTERNET. THERE ARE A LOT OF DIFFERENT APPLICATIONS THAT WE'LL NEED TO BE COGNIZANT OF THESE NEW REPRESENTATIONS OF DOMAIN NAMES. AND I DON'T WANT TO UNDERESTIMATE THE EFFORT INVOLVED IN GETTING THAT SOFTWARE INTO EVERYONE'S HANDS. THIS IS, OF COURSE, A VERY IMPORTANT OPPORTUNITY FOR THE OPEN-SOURCE NOTION, TO TAKE ADVANTAGE OF IT BY SHARING IMPLEMENTATIONS OF APPLICATIONS THAT CAN USE THESE NEW IDN-BASED DOMAIN NAMES.
I ALSO THINK, THOUGH, THAT WE'RE AT A POINT WHERE WE CAN GATHER AND BENEFIT FROM REAL FEEDBACK FROM ACTUAL IMPLEMENTATION OF IDNS, BOTH ON THE REGISTRATION SIDE WITH REGARD TO POLICIES THAT MIGHT BE REQUIRED TO RESTRICT WHICH KINDS OF TERMS CAN BE REGISTERED, AND ALSO JUST PLAIN EXPERIENCE IN USE, WHAT SORTS OF PROBLEMS HAVE ARISEN IN THE COURSE OF TRYING TO USE IDNS, TO OPERATE A REGISTRY AND/OR A REGISTRAR TRYING TO ALLOW PEOPLE TO REGISTER THESE NAMES AND THEN WHAT SORT OF APPLICATION PROBLEMS HAVE ARISEN.
IT SEEMS TO ME THAT WE HAVE A CONSIDERABLE AMOUNT TO LEARN FROM PEOPLE WHO HAVE ACTUALLY TRIED IT. I HAD A CONVERSATION WITH SABINE DOLDERER NOT TOO LONG AGO, AND I LEARNED SOMETHING VERY INTERESTING THAT I HADN'T THOUGHT ABOUT. I'M A STUDENT OF GERMAN, BUT NOT A VERY GOOD SPEAKER OF IT, AND I HAD THOUGHT IT WAS A REASONABLE SPECULATION THAT ANYTHING THAT WAS SPELLED WITH AN UMLAUT OF THE OE TYPE COULD ALSO BE SPELLED WITH AN O AND THE TWO LITTLE DIERESIS AT THE TOP. SHE QUICKLY MENTIONED AN IMPORTANT NAME, GOETHE, HE DID NOT SPELL NOR DOES ANYONE ELSE SPELL IT G-O UMLAUT T-H-E. SO IT'S NOT TRUE THAT EVERY DIPHTHONG OF THAT TYPE MAPS INTO THE OTHER.
THAT MAY NOT BE TRUE OF OTHER LANGUAGES. BUT IT WAS AN IMPORTANT EXAMPLE OF THE NEED TO BE CONSCIOUS OF LANGUAGE AS WE APPROACH IDN USE.
ALTHOUGH DOMAIN NAMES ARE JUST LABELS IN THE COMPUTER SCIENCE SENSE OF THE TERM, THE WAY WE USE THEM OFTEN INVOLVES LANGUAGE BECAUSE WE'RE TRYING TO MAKE THOSE DOMAIN NAMES MEAN SOMETHING TO THE READER.
SO ONCE AGAIN, THERE IS A STRONG NECESSITY TO TURN TO LANGUAGE SPEAKERS FOR GUIDANCE AND ADVICE ABOUT HOW TO FORMULATE POLICIES WITH REGARD TO IDNS.
ONE INTERESTING THING THAT I THINK I HAVE LEARNED FROM THE CHINESE, JAPANESE, AND KOREAN COMMUNITY IS THAT IN THEIR CASE, THEY ACTUALLY HAVE A MORE CANONICAL FRAMEWORK THAT HAS BEEN HELPFUL IN DEALING WITH TRADITIONAL AND SIMPLIFIED CHINESE CHARACTERS WHICH APPEAR IN MORE THAN ONE LANGUAGE. THERE, THE ATTEMPT IS TO MAKE MORE CANONICAL THE REPRESENTATIONS THAT ARE ACTUALLY REGISTERED. I'M SURE JAMES SENG AND OTHERS WILL DO A BETTER JOB THAN I HAVE JUST DONE REPRESENTING THE HARD WORK THAT THEY'VE GONE THROUGH TO PREPARE THEIR RESTRICTION FRAMEWORKS.
SO I GUESS THE POINT I WANT TO MAKE HERE IS THAT WE WOULD, I THINK, BE FOOLING OURSELVES IF WE THOUGHT THAT THERE WAS A SINGLE SET OF POLICIES AND PROCEDURES FOR DEALING WITH IDNS WHEN WE'RE LOOKING AT THEM FROM A LANGUAGE-SPECIFIC POINT OF VIEW AND THAT WE ACTUALLY HAVE TO TAKE INTO ACCOUNT THAT LANGUAGES REALLY DO VARY IN THEIR STRUCTURE AND, THEREFORE, POLICIES MAY HAVE TO VARY ALONG WITH THAT.
THERE ARE AN ENORMOUS NUMBER OF TECHNICAL SIDE EFFECTS WHICH IDNS PRESENT. FOR EXAMPLE, WHEN THEY APPEAR IN TEXT AND YOU CUT AND PASTE THE TEXT, THERE IS AN INTERESTING QUESTION ABOUT WHAT IS DONE WITH THE -- WITH THAT REPRESENTATION, WHETHER THERE IS AN AWARENESS IN THE PROGRAM THAT'S DOING THE CUTTING AND PASTING OF WHETHER THIS IS A SPECIAL ENTITY WHICH MAY NEED TO BE INTERPRETED IN DIFFERENT SCRIPTS THAN JUST THE UNDERLYING ASCII. WE ALMOST CERTAINLY WILL ENCOUNTER SIMILAR KINDS OF PROBLEMS WITH FORWARDING. ONE OF MY FAVORITE EXAMPLE OF AN UNEXPECTED SIDE EFFECT IN FORWARDED E-MAIL WAS AN E-MAIL I GOT FROM VENI MARKOVSKI WHO HAPPENS TO BE OUR BULGARIAN ON THE BOARD OF ICANN, AND HE GOT QUITE EXCITED WHEN HE SAW THAT DOT DE WAS REGISTERING GERMAN UMLAUTED DOMAIN NAMES, SO HE FORWARDED SOME EXAMPLES TO ME.
I SHOULD HAVE BROUGHT THE EXAMPLE TO SHOW YOU BUT I'LL HAVE TO SKETCH IT IN WORDS FOR YOU.
THE E-MAIL THAT I GOT SAID, LOOK, DNIC IS NOW REGISTERING GERMAN DOMAIN NAMES WITH UMLAUTS IN THEM. BUT THE E-MAIL THAT I GOT FROM HIM SHOWED CYRILLIC CHARACTERS IN THE PLACE WHERE WE WOULD HAVE EXPECTED TO SEE UMLAUTS. AND SO I FORWARDED THE MESSAGE BACK TO HIM SAYING SURELY YOU UNDERSTAND THE DIFFERENCE BETWEEN A CYRILLIC CHARACTER AND A GERMAN UMLAUT. AND HIS REACTION WAS FAIRLY INTERESTING, BECAUSE HE SAID, WELL, IT DIDN'T LOOK THAT WAY WHEN I SENT IT TO YOU.
WHAT HAD HAPPENED WAS THE WAY IN WHICH THE MESSAGE WAS ENCODED, THERE WAS A CHARACTER SET ASSOCIATED WITH THE ENTIRE MESSAGE. AND SINCE VENI'S E-MAIL SOFTWARE WAS ROOTED IN BULGARIAN, IT WAS ASSUMED ANYTHING HE SENT WOULD INCLUDE A SCRIPT WITH THE CHARACTERS HE SENT AND IN PARTICULAR THE GERMAN UMLAUTS WERE ENCODED INTO THE CHARACTERS. SO THAT'S WHY I GOT THE WRONG CHARACTER SET.
IT'S EXAMPLES LIKE THAT, AND I'M SURE YOU CAN MAKE UP MUCH MORE ELABORATE ONES WHEN YOU DEAL WITH WORD PROCESSING PROGRAMS RATHER THAN E-MAIL, THINGS GET COMPLEX WHEN DEALING WITH HTML COMPONENTS. SO I BRING THIS UP, AS WE TRY TO INTRODUCE INTERNATIONALIZED DOMAIN NAMES, ALTHOUGH ICANN ISN'T RESPONSIBLE FOR DEALING WITH THE PROBLEMS OF APPLICATIONS, WE SHOULD BE AT LEAST CONSCIOUS OF THE FACT THAT AS WE INTRODUCE THESE THINGS AT THE DOMAIN NAME LEVEL THAT WE WILL ALSO INTRODUCE SIDE EFFECTS, RIPPLING SIDE EFFECTS IN OTHER APPLICATIONS.
THERE ARE ISSUES I THINK WE HAVE NOT ADDRESSED. ONE HAS TO DO WITH IDN ELEMENTS AT THE TOP LEVEL. TODAY MOST OF THE MECHANISMS WE HAVE IN PLACE THAT I'M AWARE OF TEND TO DEAL WITH TOP-LEVEL DOMAINS THAT ARE STILL IN ASCII, AND I UNDERSTAND THAT THERE ARE IMPLEMENTATIONS OF TOP-LEVEL DOMAINS THAT INCLUDE IDN ELEMENTS. THEY'RE NOT NECESSARILY PART OF THE GLOBAL DOMAIN NAME SYSTEM YET.
THERE ARE SOME INTERESTING POLICY ISSUES. ONE OF THEM IS WHETHER TWO TOP-LEVEL DOMAINS THAT MEAN THE SAME THING BUT ARE REPRESENTED IN DIFFERENT CHARACTER SETS SHOULD BE TREATED AS DISTINCT OR THE SAME.
IF THEY'RE DISTINCT, THEN THE SECOND LEVEL DOMAINS THAT ARE REGISTERED IN EACH NEED TO BE TREATED AS DISTINCT. ON THE OTHER HAND, IT MIGHT BE PREFERRED IN SOME COMMUNITIES THAT THE TOP-LEVEL DOMAIN IN IDN FORM AND THE TOP-LEVEL DOMAIN IN ASCII FORM IN FACT FORM A COHERENT WHOLE, AND, THEREFORE, THE SECOND LEVEL DOMAINS THAT ARE EDGE REGISTERED AND ANY THIRD AND FOURTH LEVEL DOMAINS ARE INTENTIONALLY SUPPOSED TO BE BOUND TO EACH OTHER. I'M NOT PROPOSING ANY ONE OF THOSE POLICIES PARTICULARLY, BUT I AM SAYING ONE NEEDS TO UNDERSTAND WHETHER THERE WILL BE A BINDING OR NOT IN THE CASES WHERE WE START INTRODUCING TOP-LEVEL DOMAINS IN IDN FORM THAT ARE INTENDED TO BE ANALOGS OF THEIR ASCII COUNTERPARTS. THIS ALSO RAISES A POTENTIAL PROBLEM, THE NUMBER OF TLDS THAT MIGHT BE INTRODUCED AS A CONSEQUENCE OF INTRODUCING TOP-LEVEL DOMAINS IN IDN FORM. THE WORST CASE SCENARIOS, OF COURSE, MIGHT INVOLVE TENS OF THOUSANDS OF ENTRIES IF EVERY EXISTING ASCII-BASED TOP-LEVEL DOMAIN !
NEEDED ALSO TO BE REPRESENTED IN EVERY POSSIBLE LANGUAGE THAT'S AVAILABLE THROUGH THE UNICODE ENCODING. THAT'S A SERIOUS POTENTIAL PROBLEM FOR THE ROOT OPERATION.
SO ONCE AGAIN, A TECHNICAL CONSIDERATION WOULD INFLUENCE POLICY DECISIONS.
AND THE FINAL POINT I'D LIKE TO MAKE IS THAT IN OUR PASSION TO INTRODUCE IDNS, I THINK WE CAN EASILY CONFUSE TWO DIFFERENT IMPORTANT OBJECTIVES THAT I THINK INTERNET SHOULD HAVE. ONE OF THEM, OF COURSE, IS TO SUPPORT THE USE OF LOCAL LANGUAGES TO MAKE THE INTERNET MORE ACCESSIBLE TO PEOPLE WHOSE LANGUAGES ARE NOT YET REPRESENTED IN THE DOMAIN NAME SYSTEM. ON THE OTHER HAND, WE ALSO HAVE AN OBLIGATION TO PAY ATTENTION TO INTERNATIONALIZATION; THAT IS TO SAY, THE ABILITY OF COMMUNITIES TO COMMUNICATE WITH EACH OTHER EVEN IF THEIR BASE LANGUAGES ARE NOT THE SAME.
IF IT'S IMPOSSIBLE FOR SOMEONE TO ENTER ONTO THEIR KEYBOARD AN IDN IN A LANGUAGE THAT THEIR KEYBOARD ISN'T SET FOR, THAT'S A PROBLEM.
SO I HOPE THAT WE DON'T LOSE TRACK OF ONE OF THE IMPORTANT VALUES OF THE INTERNET, WHICH IS TO PERMIT COMMUNITIES TO COMMUNICATE WITH EACH OTHER. WHEN WE HEARD MR. DESAI THIS MORNING POINT OUT THAT THE TREATY AGAINST THE PROLIFERATION OF LAND MINES, HIS BELIEF IS THAT TREATY COULD NOT HAVE BEEN FORMED IF NOT FOR THE EFFECTIVE INTERNET ALLOWING COMMUNITIES TO COMMUNICATE WITH EACH OTHER. IF THEY COULDN'T EVEN ENTER EACH OTHER'S E-MAIL ADDRESSES BECAUSE OF AN IDN BARRIER, THEN THAT OUTCOME MIGHT BE IN QUESTION.
SO WE WANT TO BALANCE, I THINK, THE UTILITY OF LOCALIZATION WITH THE IMPORTANCE OF INTERNATIONALIZATION SO AS TO NOT LOSE THAT IMPORTANT VALUE.
SO I, FIRST OF ALL, COMMEND TO YOUR ATTENTION THE EXPERTS THAT YOU'LL BE HEARING FROM IN THE REMAINDER OF THIS PROGRAM AND THANK EACH OF YOU FOR ATTENDING AND EACH OF THE PANELISTS FOR TAKING TIME TO MAKE THEIR PRESENTATIONS.
I SHOULD TELL YOU THAT PAUL AND I MAY HAVE TO DISAPPEAR ABOUT 2:00, SO PLEASE DON'T TAKE OUR DISAPPEARANCE AS A LACK OF INTEREST.
THANK YOU VERY MUCH.
PAUL, I TURN THIS BACK OVER TO YOU.
>>PAUL TWOMEY: THANK YOU, VINT. IT'S A BIT LIKE AN ICANN BOARD SHUFFLE UP HERE BUT I'M GOING TO INTRODUCE MOUHAMET DIOP, SENEGALESE MEMBER OF THE BOARD, AND I'LL INTRODUCE ONE OF OUR OTHER KEY SPEAKERS WHO IS ADAMA SAMASSEKOU.
MOUHAMET.
>>MOUHAMET DIOP: THANK YOU VERY MUCH, AND THANK YOU, VINT CERF, FOR TAKING THE TIME WHAT YOU REALLY FEEL ABOUT THE IDN IMPLEMENTATION AND THE EVOLUTION OF THE IDN FOR THE NEXT STEP.
BEFORE I GIVE THE FLOOR TO ADAMA SAMASSEKOU TO COME AND SHARE WITH US HIS VIEW ABOUT THE IDN IN AFRICA AND WHAT'S GOING TO BE THE NEXT STEP FOR HIM, I WANT TO SHARE MAYBE SOME HISTORY ABOUT IDN IN ICANN, BECAUSE I THINK IT'S VERY IMPORTANT TO RECORD FOR WHAT HAPPENED, BECAUSE YOU KNOW THAT AT ANY ICANN MEETING, WE HAVE NEW FACES AND PEOPLE WANT TO KNOW WHAT HAPPENED AND WHERE WE ARE AND WHAT ARE THE NEXT STEP THAT WE HAVE TO ACHIEVE IN ORDER TO GET THINGS MOVING.
SO I THINK THAT THERE WAS SOMETHING REALLY INTERESTING THAT HAPPENED THE LAST THREE YEARS IN ICANN WHERE WE JUST RECOGNIZED THE IMPORTANCE OF THE DOMAIN NAME EXPANSION, ESPECIALLY ABOUT IDN. AND I THINK THAT THERE WAS SOME STRONG VOICE COMING FROM ASIA-PACIFIC, AND LATIN COUNTRY, SOUTH AMERICAN COUNTRY IN ORDER TO ENABLE, I MEAN, THE LANGUAGE IN THE DOMAIN NAME SPACE.
AND I THINK THAT THERE'S A COUPLE INITIATIVES THAT HAVE BEEN SET UP IN ORDER TO ENABLE THAT INITIATIVE. I JUST RECALL THE IDN COMMITTEE THAT HAD BEEN SET UP BY ICANN AND CHAIRED BY ONE OF THE ICANN BOARD DIRECTORS WHOSE NAME IS KATOH-SAN FOR MANY YEARS.
HE TRIED TO SET UP, I MEAN FROM SCRATCH, WHAT NEED TO BE DONE. I MEAN, JUST HAVE A CLEAR UNDERSTANDING OF WHAT THE ISSUE ARE, BECAUSE WHEN WE'RE TALKING ABOUT MULTILINGUALISM, THERE ARE MANY ISSUES THAT COVER THOSE ITEMS. AND FROM ICANN'S PERSPECTIVE, WE'RE JUST TRYING TO UNDERSTAND WHAT IS OUR ROLE AND WHAT CAN BE DEALT WITH AT THE ICANN LEVEL. MAYBE THERE ARE SOME ISSUES THAT HAVE TO BE DEALT WITH AT ANOTHER LEVEL. BUT ON THE PURE ICANN RESPONSIBILITY LEVEL, WE'RE TRYING TO SEE HOW COMPLEX THE ISSUE ARE. AND THE IDN COMMITTEE HAVE WORKED WITH MANY EXPERTS. JOHN KLENSIN WAS ONE OF THEM, MANY PEOPLE FROM DIFFERENT CONSTITUENCY WHO JOINED THE COMMITTEE. WE'VE GOT REPRESENTATIVE FROM THE ADDRESSING SUPPORT COMMUNITY, AND I HAVE THE CHANCE AT THAT TIME TO REPRESENT THE (INAUDIBLE) SUPPORT ORGANIZATION INSIDE THAT COMMITTEE.
WE HAVE ALSO ELISABETH PORTENEUVE WHO REPRESENT THE CCNSO AT THAT TIME. JUST TO LET YOU KNOW THAT THERE WAS MANY PEOPLE WHO ARE INVOLVED AT THAT TIME IN THE PROCESS, JUST TO UNDERSTAND THE ISSUE ABOUT IDN AND WHAT HAVE -- WHAT ICANN HAVE TO BE DONE. AND JUST TO UNDERSTAND WHAT ISSUES HAVE BEEN RAISED BY THE INTERNATIONALIZED DOMAIN NAME.
SO THIS COMMITTEE CAME OUT WITH SOME RECOMMENDATION, AND WE SEE THAT THERE WAS MANY TOPICS THAT HAVE TO BE COVERED, LIKE THE COMMUNITY-BASED, WHAT WILL BE THE IMPLICATION OF LANGUAGE, WHAT'S GOING TO BE THE IMPLICATION ON THE CYBERSQUATTING, AND THE GEOGRAPHICAL IMPLICATION OF THIS ISSUE, THE CULTURAL IMPLICATION.
AND WHEN THE -- WHEN THE TERM OF THIS IDN COMMITTEE CAME TO AN END, MANY PEOPLE ASKED THE QUESTION, DOES THAT MEAN THAT WE HAVE FINISHED THE JOB OR HAS ICANN SET UP EVERYTHING FOR IDN TO BE IMPLEMENTED?
AND WE KNOW THERE WAS ANOTHER TECHNICAL COMMITTEE CALLED THE IDN-RIC COMMITTEE, IT WAS REGISTRY IMPLEMENTATION COMMITTEE IN WHICH WE GOT ALL THE REGISTRY. VERISIGN WAS PART OF IT, AFILIAS, WE GOT ALL THE REGISTRY. WE GOT SOME COUNTRY CODE, TOP-LEVEL DOMAIN WHO ARE REALLY INTERESTED IN IMPLEMENTING A SECOND-LEVEL REGISTRATION FOR IDN. WE GOT ALSO A VERY IMPORTANT WORK THAT HAVE BEEN DONE BY THE SGKS, CHINESE, KOREAN, AND JAPANESE GROUP THAT HELP US A LOT. I MEAN, UNDERSTAND THE ISSUE. AND WHAT WAS REALLY IMPORTANT IS WE COME UP WITH IETF STANDARDS THAT HELP UNDERSTAND THAT IF YOU WANT TO IMPLEMENT IDN, THESE ARE THE STANDARDS THAT HAVE TO BE USED IN ORDER TO AVOID A LOCALIZATION OR KEY-WORD APPLICATION TO BE THE ONLY CHOICE GIVEN TO PEOPLE WHEN THEY WANT TO DEAL WITH OTHER LANGUAGES.
SO IETF ISSUED THE STANDARD LAST YEAR, AND WE NOW UNDERSTAND THAT TECHNICALLY WE DON'T HAVE ANY BARRIER FOR THE IMPLEMENTATION OF IDN, BUT WE ALL CONVENED AND UNDERSTAND THAT THERE WAS NEED FOR POLICY, I MEAN, TO BE REALLY WELL DEFINED IN ORDER TO AVOID ALL THE PROBLEM THAT HAVE BEEN RAISED BY THE INTERNATIONALIZATION OF DOMAIN NAME. AND WE HAVE A CHANCE AT THE LAST ICANN MEETING IN KUALA LUMPUR TO HAVE AN IDN SESSION LED BY JOHN KLENSIN WITH ALL THE OTHER EXPERTS, AND WE HAD A CHANCE TO GET EXPERIENCE FROM VARIOUS COUNTRY. AND WE ALSO HAD THE OPPORTUNITY TO SEE TECHNICALLY WHAT ARE THE REAL CHALLENGES WE ARE FACING WHEN WE'RE TALKING ABOUT IDN.
AND EVEN IF YOU'RE REALLY AFRAID WHEN WE'RE TALKING ABOUT THE IDN IMPLEMENTATION OF THE TOP-LEVEL DOMAIN NAME, I JUST WANT TO SHARE SOME OF MY FEELING ABOUT THAT.
IT'S REALLY IMPORTANT THAT WE MAKE THE INTERNET WORLD AS -- AS WE FEEL IT IN OUR NORMAL LIFE. LIKE IF I TALK TO MY SON, IF I TALK TO FRIENDS, I JUST WANT TO DO IT IN MY OWN NATIVE LANGUAGE.
AND WHAT I'M EXPECTING IS ALL THE EVOLUTION WE'RE MAKING WITH THE TECHNOLOGY IS TO ENABLE PEOPLE TO DO THAT IN THAT WAY.
AND IT'S A VERY CHALLENGING ISSUE FOR US, IS HOW WE CAN MAKE THE INTERNET JUST LIKE AN IDEAL TECHNOLOGICAL ENVIRONMENT THAT WILL HAVE PEOPLE JUST DUPLICATE EXACTLY WHAT THEY HAVE IN THEIR NORMAL LIFE IN THAT ENVIRONMENT.
I'M NOT AFRAID ABOUT THE TECHNICAL IMPLICATION, EVEN IF I SHOULD, JUST BECAUSE THIS TECHNICAL COMMUNITY HAS MADE CHALLENGES ALL OVER THEIR LIFE AND THIS IS JUST A NEW ONE. SO IT'S NOT TIME TO GIVE US. IT'S JUST TIME TO SAY WE MAKE IT, WE ALREADY MADE IT, AND WE'RE GOING TO MAKE IT.
AND THAT'S THE ONLY THING THAT GIVE US ENOUGH RESOURCES TO SAY THAT. IDN IMPLEMENTATION IS A CHALLENGE. IT'S REALLY HARD IF WE LISTEN TO PEOPLE. THEY JUST EXPLAIN TO US EVERY DAY THAT IT WILL NOT BE AN EASY PART, BUT I FEEL CONFIDENT THAT WE SHOULD MAKE IT AND WE'RE GOING TO MAKE IT BECAUSE IT'S PART OF THE CULTURAL DIVERSITY PROGRAM. IT'S JUST THE FIRST STEP THAT WILL ALLOW PEOPLE TO FEEL THEMSELVES IN AN ENVIRONMENT WHERE THE TECHNOLOGY WILL ENABLE PEOPLE TO LIVE EXACTLY THE SAME WAY THEY'RE DEALING NOW.
SO I WOULD JUST CALL FOR ICANN TO HAVE ENOUGH COURAGE TO TACKLE THAT PROBLEM AND TO LAUNCH, MAYBE JUST AS SOMETHING LIKE THE FIRST EXPERIMENTATION AT THE TOP-LEVEL DOMAIN. IT'S NOT THE ICANN BOARD OF DIRECTORS WHO ARE SPEAKING; IT'S JUST THE AFRICAN NATIVE SPEAKER. (INAUDIBLE) SPEAKING IS MY NATIVE LANGUAGE, AND I'M CALLING TO LOOK AT, AS A NEXT PHASE, AFTER THE SECOND-LEVEL IMPLEMENTATION OF IDN, TO TALK WITH PEOPLE AND HAVE THAT TOP LEVEL IMPLEMENTATION OF IDN TO OCCUR. WE NEED THAT AND I THINK IT WILL COME VERY SOON.
THE OTHER SIGNAL WE GET FROM ICANN AS A VERY STRONG ONE IS DURING THE LAST KUALA LUMPUR MEETING, IT HAD BEEN SET UP A PRESIDENTIAL COMMITTEE. PRESIDENTIAL COMMITTEE MEANS, THAT'S THE CHAIR, THAT THE CEO OF ICANN WILL HAVE TO POPULATE THAT COMMITTEE. AND I DON'T THINK THAT HE WILL TAKE AS LONG AS THE WORKING GROUP FOR U.N. TO POPULATE, BECAUSE AFTER THE WSIS PROCESS, SO WE HAVE WAITED FOR MORE THAN 11 MONTHS BEFORE WE HAVE THE COMMITTEE POPULATED.
SO FROM KUALA LUMPUR, WE HAVE SPENT FOUR MONTHS, SO PLEASE, CEO, WE ARE EXPECTING THAT COMMITTEE TO BE POPULATED VERY SOON. AND I THINK THAT IF YOU COMMIT YOURSELF ON IT MAYBE BEFORE WE LEAVE THAT MEETING IN CAPE TOWN, WE WILL GET THAT COMMITTEE POPULATED.
SO I WILL NO LONGER KEEP THE FLOOR. I JUST WANT TO COME AND HAVE ADAMA SAMASSEKOU SHARE WITH US, HIS EXPERIENCE AS THE PRESIDENT OF THE AFRICAN ACADEMY OF LANGUAGE. HE IS ONE OF THE SUPPORTERS OF THE CULTURAL DIVERSITY. HE WAS ALSO THE FORMER CHAIR OF THE WSIS PROCESS, THE PREPARATORY WSIS SUMMIT IN GENEVA. AND HE WAS VERY PLEASED TO ACCEPT TO COME TO AN ICANN MEETING AND SHARE WITH US HIS EXPERIENCE, AND I THINK THAT WE SHARE A LOT OF THINGS, AND I'M GOING TO CALL HIM FOR THE FLOOR. THANK YOU VERY MUCH. ADAMA.
(APPLAUSE.)
>>ADAMA SAMASSEKOU: THANK YOU VERY MUCH, MOUHAMET.
I DON'T SEE A LOT OF KOSA SPEAKING PEOPLE HERE, BUT IN KOSA IT MEANS GOOD AFTERNOON, LADIES AND GENTLEMEN. WHILE WE ARE IN SOUTH AFRICA, I THINK WE NEED TO PAY TRIBUTE TO ONE OF OUR GREAT LANGUAGES, AFRICAN LANGUAGES.
LET ME CONTINUE IN FRENCH.
THANK YOU VERY MUCH, MY BROTHER, FOR GIVING ME THIS OPPORTUNITY TO SHARE WITH YOU THE VISION THAT I HAVE PREPARED. I HAVE HAD AN ACADEMIC PRESENTATION, BUT AS WE HAVE SEEN THE PRESENTATION THIS MORNING, I, AFTER VINT CERF, WHAT HE HAS SAID SO CLEARLY, THE FIRST THING THAT I'D LIKE TO SAY WHEN I TALK ABOUT IDN -- DON'T WORRY, I'M NOT GOING TO TALK ABOUT IT AS A TECHNICIAN. I DON'T KNOW ANYTHING. MOUHAMET KNOW BETTER THAN ME. AND I ALSO HAVE A BROTHER HERE WHO WILL TALK MUCH BETTER, MAXIME SOMÉ. HE'S PRESENT HERE. AND PIERRE OUEDRAOGO, WHO KNOW THIS DOMAIN.
WHAT IS IMPORTANT FOR ME TO SHARE WITH YOU IS REALLY THE SITUATION OF OUR CONTINENT.
CLEARLY, THE FIRST QUESTION THAT ONE WOULD ASK HIMSELF, WHY MOUHAMET TALKED WITH SUCH KIND OF SENSITIVITY OF HIS MOTHER TONGUE, WHICH HE WOULD LIKE IT TO BE USED IN THE INTERNET AS HE USE IT WITH HIS CHILDREN. THIS IS A QUESTION OF OUR CONTINENT.
AFRICA IS THE ONLY CONTINENT IN THE WORLD WHERE A CHILD GOES TO SCHOOL, STARTS LEARNING A LANGUAGE OTHER THAN WHAT HE'S USED IN HIS FAMILY. THAT'S THE ONLY CONTINENT WHERE IN THE CONTEXT OF JUSTICE, IN MOST OF THE CASE IN ALL THE STATES, THEY USE INTERPRETERS WHO SPEAK EITHER FRENCH, ENGLISH, PORTUGUESE, AND AT THE SAME TIME KNOW THE MOTHER TONGUE OF THE ACCUSED.
AFRICA IS THE ONLY CONTINENT WHICH HAS TWO OR THREE PERCENT OF ITS ROLE PLAYERS OF DEVELOPMENT AT THE COMMUNITY LEVEL WHO ARE IN THE MARGIN OF EXCHANGE OF INFORMATION, BECAUSE INFORMATION, THE PUBLIC MANAGEMENT OF BUSINESS IS DONE IN OFFICIAL LANGUAGES WHICH ARE NOT MASTERED BY THE POPULATION, THE AFRICAN POPULATION. THIS IS THE SITUATION.
TODAY WE ARE IN THE CONTEXT WHERE THE WORLD IS CHANGING AND WE HAVE THE POSSIBILITY AT THE TECHNICAL LEVEL TO DO ANYTHING; TO TAKE IN CHARGE ALL THE LANGUAGES IN THE WORLD. BUT THIS SITUATION IS STILL STAGNANT. AND THIS IS A BIG ISSUE, AND WHEN THIS MORNING WE SAID IT WAS TECHNICALLY POSSIBLE. BUT WHAT WE ARE WAITING FOR, TECHNICIANS ARE WAITING FOR LINGUISTS TO SAY WHAT ARE POSSIBLE FOR THE CORRECT (INAUDIBLE) IN OUR LANGUAGES AND THEY ARE CALLED BILINGUISTS AND WE SAY THEY HAVE TO PUT EVERYTHING IN PLACE REALLY RAPIDLY. BUT MY FIRST QUESTION AT THE LEVEL OF AFRICAN INITIATIVES THAT I WOULD LIKE TO TALK ABOUT HERE, IF YOU WANT TO TALK ABOUT THE INSTRUMENT WHICH IS AFRICAN ACADEMY OF LANGUAGES, WE WOULD LIKE TO PUT IN PLACE, SINCE THREE YEARS NOW, THE IDEA OF THE ACADEMY WAS STARTED THREE YEARS AGO, BUT IT WAS STARTED THREE YEARS BEFORE. AND WHEN THE HEAD OF STATES OF AFRICA DECIDED TO CREATE AN AFRICAN INSTITUTION WHICH WOULD BE IN CHARGE OF VALIDATING AFRICAN LA!
NGUAGES TO BE USED AS WORKING LANGUAGES ON A DAILY BASIS IN ALL THE LIFE ACTIVITIES OF THE STATES AND AFRICAN COMMUNITIES. AND THEY TOOK THIS DECISION ON THE PROPOSAL OF THE MOKINARI, THE OLD FORMER PRESIDENT OF MALI. BUT IT'S A SITUATION, IT'S A POLITICAL WILL THAT IS VERY OLD AT THE LEVEL OF HEAD OF STATE, SINCE OAU IN 1964 IN THE CHARTER OF THE ORGANIZATION, THE WORKING LANGUAGES OF THE OAU SHOULD BE EIGHT POSSIBLE AFRICAN LANGUAGES AS SUCH. AND ENGLISH AND ARABIC AND FRENCH AND PORTUGUESE.
IF POSSIBLE, AFRICAN LANGUAGES. IT STARTED IN 1963, AND FROM 1963 WE HAD TO WAIT, JULY 2004 TO HAVE IN OAU, WHICH IS AU TODAY, A PRESIDENT, AN AFTERNOON CAN PRESIDENT, MAKING A SPEECH IN AN AFRICAN LANGUAGE, IN SWAHILI AS SUCH, THE LANGUAGE WHICH WAS ADOPTED AS A WORKING LANGUAGE OF OAU SINCE 1986, WHICH WAS IMPLEMENTED THEN.
THAT SUGGESTS TO SAY THAT THIS WILL EXIST, BUT THEN IT WASN'T IMPLEMENTED. AND TODAY THE ACADEMY HAS BEEN CHARGED OF PUTTING IN PLACE THIS VISION AND TO REALIZE AT THE ACADEMY WHAT IS IT ALL ABOUT. THE AFRICAN ACADEMY OF LANGUAGES IS A SCIENTIFIC INSTITUTION SPECIALIZED -- THE AU SPECIALIZED FROM MALI (INAUDIBLE), IT HAS TO PROMOTE AFRICAN LANGUAGE TO PROMOTE THE USE AND FACILITATE EXCHANGES BETWEEN STATES AT AN ECONOMICAL AND SOCIAL LEVEL AND TO RECOGNIZE THE AU IN TERMS OF THE DEVELOPMENT OF THE CONTINENT. AT THE CONTINENT LEVEL, (INAUDIBLE) IS THE HIGHER INSTITUTE STRUCTURE WHICH HAS IDEAS AND WHICH HAS TO TRANSLATE IDEAS TO MAKE THEM REALIZABLE, AND TO IMPLEMENT LINGUISTIC POLITICS IN A NORMALIZED WAY AND IN A CONSENSUAL WAY.
THE FINAL REPORT OF THE STATES ORGANIZED BY THE AU SECRETARY ON 5TH OF FEBRUARY LAST ON LANGUAGES AS THE OBJECTIVE OF PROMOTING AFRICAN LANGUAGES AND CROSS-LANGUAGES, AND I WOULD LIKE TO STOP THERE BECAUSE WHAT IS IMPORTANT IS FOR AFRICA TODAY TO GIVE PRIORITY TO THE LANGUAGES BETWEEN BORDERS, BETWEEN SENEGAL, MAURITANIA AND OTHERS. AND IT IS SHARED BY NINE STATES AND WESTERN AFRICA AND FRANCOPHONE STATES, AND I THINK THAT THIS WILL ALLOW EVEN SWAHILI IN THE REGION WE ARE IS SHARED BY 15 STATES WITHOUT LOOKING AT GUNI IN WESTERN (INAUDIBLE) AFRICA. THE KNOWLEDGE OF LANGUAGES IS VERY IMPORTANT AS AN INTEGRATION FACTOR FOR INITIAL UNDERSTANDING AND PROGRESS AND RESOLUTION OF CONFLICT TO REALIZE WE HAVE TO PUT OBJECTIVES ON COORDINATION OF ACTIVITIES, OF RESEARCH ACTIVITIES ON LANGUAGE TO PUT IN PLACE A FRAMEWORK OF RESEARCH FOR CONSULTATION AND RESEARCH AND MODERNIZATION OF INSTRUMENTS AND THE I.T. AND HARMONIZING TEACHING AND ALSO HAVING DOCUMENTATIONS AND PUTTING THE!
DATABASES AND REDEFINING THE RELATIONSHIP BETWEEN LANGUAGES, WHICH IS IMPORTANT BECAUSE AFRICAN LANGUAGES SHOULD DEVELOP A PARTNERSHIP RELATIONSHIP WITH EUROPEAN LANGUAGES, WHICH ARE PART OF OFFICIAL LANGUAGES AT THE LEVEL OF EDUCATION, JUSTICE AND THE PRESS. AND ALL THE USE, EQUITABLE USE OF THESE MEDIUMS IN USING AFRICAN LANGUAGES AND EUROPEAN LANGUAGES WHICH ARE LANGUAGES WHICH ARE USED TOO AS WORKING LANGUAGES.
THE MAIN BODY OF THE ACADEMY IS A BOARD, ASSEMBLY OF THE ACADEMY, THE TECHNICAL SECTOR AND THE COMMISSION OF LANGUAGES AS FAR AS LANGUAGE COMMISSIONS, EACH LANGUAGE IS A VEHICLE AND WILL BE A COMMISSION OF LANGUAGES, THERE WILL BE A (INAUDIBLE) COMMISSION, A MUNDANE COMMISSION, AND (INAUDIBLE) COMMISSION, ET CETERA. AND THESE ARE THE COMMISSIONS, LANGUAGE COMMISSIONS, WHICH WILL HAVE THE BASIC STRUCTURE SCIENTISTS WILL WORK ON AND WE WILL USE THESE LANGUAGES AS WORKING LANGUAGES FROM THE AFRICA REGIONS, AND AFRICA IS DIVIDED INTO FIVE BIG REGIONS. AND WE HAVE THE SATIC REGION, AND THE NEXT REGION, WE HAVE THE ECONOMIC AND SOCIAL COMMISSION TO USE LANGUAGE AS WORKING LANGUAGES, LANGUAGES WHICH WILL BE USED ACROSS BORDERS FOR THE ECONOMIC DEVELOPMENT.
AND THE TWO WORDS ON THE ACTIVITY OF THE ACADEMY. SO FAR, IN FACT, WHEN WE ANALYZE THE ACTIVITIES THAT HAVE BEEN UNDERTAKEN AND WE CAN SAY THAT THE ACADEMY IS FUNCTIONAL, WHICH IT IS NOT OPERATIONAL SO FAR, WE HAVE TO WAIT FOR THE PLANNING TO LAUNCH THE ACADEMY IN 2005. LEGALLY THE STRUCTURES WILL BE PUT IN PLACE. AND THIS WILL MAKE IT OPERATIONAL, BUT WE HAVE HAD ACTIVITIES, A LOT OF ACTIVITIES, AND I DON'T WANT TO GO IN DETAILS ON THESE ACTIVITIES, BUT I WOULD LIKE TO LET YOU KNOW ABOUT THE THREE MAIN ACTIONS OF THE ACADEMY, JUST TO GIVE YOU AN IDEA WHICH WILL LET US UNDERSTAND THE WHOLE THING.
FIRST OF ALL, THE ACADEMY HAS TRIED TO PUT IN PLACE AN ACTION PLAN OF THE LINGUISTICS FIELD WHICH WAS ACCEPTED BY THE HEAD OF STATE SINCE 1986. THERE IS A PLAN OF ACTION WHICH WASN'T IMPLEMENTED, SO THEN THE ACADEMY, WITH THE EXPERT OF THE LINGUISTS, HAS GOT A STEERING COMMITTEE WHICH IS CALLED ELBA AND THESE IMPLEMENTATIONS OF THE PLAN OF AFRICA AND THIS STEERING COMMITTEE WILL HAVE TO MEET IN A FEW DAYS AT THE END OF THE WEEK IN CAPE TOWN TO TACKLE THE DIFFERENT CHANNELS WHICH WILL BE USED IN THE FUTURE.
AND THE FIRST CHANNEL IS FOR US TO TACKLE FIVE OR SIX PROJECTS OF THE COMMITTEE WHICH WILL BE PUT IN PLACE ALL OVER THE CONTINENT, INCLUDING TRAINING OF HUMAN RESOURCES IN LINGUISTICS, AFRICAN LINGUISTICS, TO THE I.T.
THE SECOND POINT IS THAT THE AU COMMISSION HAS DECIDED TO CONSIDER 2006 AS THE YEAR OF AFRICAN LANGUAGES.
THE ACADEMY HAS GOT TO PREPARE FOR THAT.
UNTIL 2006, THE TECHNICIANS WILL HELP US TO HAVE IN CYBERSPACE THE ELEMENTS THAT WILL ALLOW US TO POPULARIZE THE MASTERY OF THE -- OF POPULATIONS WITH THIS TOOL THAT WE CALL THE INTERNET.
AND SO AS WE COULD -- THE CREATORS AT THEIR OWN LEVEL AND ALL ESPECIALLY IN THE REMOTE AREAS WILL BE ABLE TO USE THESE TOOLS.
ANOTHER CHANNEL IS THAT THE ACADEMY IS WORKING WITH THE HELP OF THE ECONOMIC COMMISSION OF THE UNITED NATIONS FOR AFRICA, ELEMENTS WHICH WILL HELP THE WHOLE THING BE OPERATIONAL IN CREATING, YOU KNOW, PUT THE CYBERSPACE IN AFRICAN LANGUAGES.
AND THERE IS A NETWORK THAT HAS BEEN PUT IN PLACE.
AND IN THIS CONTEXT, MAXIME SOMÉ IS HERE, IS A MEMBER OF THIS NETWORK.
AND WE ARE TRYING TO WORK ON THE SITUATION OF AFRICAN LANGUAGES TO HAVE TECHNICALLY THE POSSIBILITY OF HAVING THEM IN THE -- AT THE INTERNATIONAL LEVEL.
ANOTHER CHANNEL THAT -- IN THE FRAMEWORK OF THE WORLD SUMMIT, WE HAVE TALKED ABOUT DIVERSITY OF -- CULTURAL DIVERSITY, AND THE ACADEMY IS GOING TO ORGANIZE WITH UNESCO, WITH ICANN, AND MICROSOFT AND OTHER ACTORS, ALL OF THE AGENCIES WHO SPEAK EUROPEAN LANGUAGES, A BIG CONFERENCE, THEMATIC CONFERENCE IN THE SECOND PART OF THE SUMMIT AT HEADQUARTERS OF MARCO.
AND THAT CHANNEL WILL DEFINITELY INCLUDE EXPERTS OF ICANN.
AND THEY WILL HAVE THE OPPORTUNITY TO SHOW HOW WE COULD GO FASTER THAN -- EMPOWERING THE AFRICANS IN THEIR OWN LANGUAGES.
THE LAST CHANNEL THAT I WOULD LIKE TO RAISE HERE IS THE -- IN THIS THEMATIC CONFERENCE IS THAT THERE WILL BE AN INTERNATIONAL CONFERENCE THAT WILL ALLOW EUROPEAN AGENCIES WORKING ON AFRICAN LANGUAGES, THE FRANCOPHONIE AGENCIES, SECRETARY OF -- IN CHARGE OF PORTUGUESE LANGUAGES, THAT WILL HAVE TO MEET AND TO WORK WITH THE ACADEMY TO SHOW HOW WE COULD DEVELOP IN AFRICA A REAL PARTNERSHIP BETWEEN AFRICAN LANGUAGES AND THE EUROPEAN LANGUAGES, AS I OFTEN SAID, AFRICA, BEFORE BEING AN ENGLISH-SPEAKING -- AND FRENCH SPEAKING IS AFRICAN SPEAKING AS SUCH.
AND THIS IS WHAT WE WILL HAVE TO PUT IN PLACE, ALL OF THIS PROCESS, WHICH WILL ALLOW US TO -- WAITING FOR THE WORLD SUMMIT, TO HELP US IN REGIONAL CONFERENCE IN ACCRA IN FEBRUARY TO TACKLE THIS QUESTION OF AFRICAN LANGUAGES IN CYBERSPACE, WHILE PREPARING CONFERENCE IN MAY AND WORKING ON THE FURTHER PROGRAM OF TUNIS.
AFRICA SHOULD BE PRESENT NOT ONLY WITH THE PROPOSALS THAT I MENTION OF THE FUNDING AND -- BUT ALSO GIVING YOU THE IDEA, WITH CONCRETE PROPOSALS AT THE LEVEL OF CONTENT, BECAUSE WE NEED THE INFORMATION AND THIS IS THE CONTENT WHICH IS IMPORTANT.
WE HAVE TO MASTER INTERNET, USING OUR LANGUAGES.
AND THIS IS A WAY FOR AFRICA TO BE PRESENT IN THIS FRAMEWORK INSTEAD OF BEING MARGINALIZED AS A MARGINAL SOCIETY IN A SOCIETY WHERE WE ARE.
TO CONCLUDE, AS I SEE THE WHOLE THING, AS MOUHAMET ASKED ME THE QUESTION, WELL, ICANN HAS ALREADY THOUGHT ON THE IDN AND THE TECHNICAL QUESTIONS ARE UNDERWAY, AND WE HAVE TO HAVE THE POLICIES TO IMPLEMENT THE WHOLE THING.
THE ACADEMY WILL BE OPERATIONAL IN 2005, AND WE WILL BE READY IN THE NAME OF THE CONTINENT TO PROMOTE THE REFLECTION ON THE CREATION OF IDN IN OUR LANGUAGES AND TO DEVELOP WITH ICANN A REAL PARTNERSHIP IN THIS QUESTION, WHICH WILL BE CONSTRUCTIVE AND WHICH WILL ALLOW US TO FACE THE SITUATION AS OTHER CONTINENTS.
AND THAT'S WHAT I WANT TO SHARE WITH YOU IN TERMS OF THREE VISIONS, PERSPECTIVES, AND DETERMINATION, BECAUSE NO ONE ELSE, APART FROM US, WILL HAVE TO CHANNEL THE WHOLE THING.
WE NEED YOU TO GO ALONG WITH US, AND WE ARE ALREADY WORKING.
AND I WAS JUST TRYING TO CONFIRM IT.
THANK YOU VERY MUCH FOR YOUR ATTENTION.

(APPLAUSE.)
>>MOUHAMET DIOP: THANK YOU, PRESIDENT.
I THINK THAT AT THE END OF THIS SESSION WE WILL HAVE THE OPPORTUNITY TO GET THE FLOOR FOR PEOPLE WHO WANT TO INTERVENE OR I DON'T KNOW WHAT CAN BE THE MOST INTERACTIVE WAY THAT YOU WANT THINGS TO HAPPEN.
THE NEXT SPEAKER WILL BE MAXIME SOMÉ.
HE IS FROM THE UNIVERSITY.
HE IS A LINGUIST.
SO HE IS RUNNING DIFFERENT PROJECT NOW WITH THE FRANCOPHONIE AGENCE, BECAUSE THEY ARE REALLY CONCERNED ABOUT THE IDN ISSUE.
AND WE HAVE PIERRE OUEDRAOGO ALSO FROM THE FRANCOPHONIE AGENCE, FROM INTIF, WHO ARE REALLY CONCERNED ABOUT THE UNICODE AND THE IDN PROJECT IN AFRICA.
SO NEXT SPEAKER WILL BE MAXIME SOMÉ FROM THE UNIVERSITY, ACADEMY OF VERSAILLES.
MAXIME, YOU HAVE THE FLOOR.
>>MAXIME SOME: LET ME THANK ALL THE ORGANIZERS WHO ALLOW ME TO BE HERE.
ICANN, AND THE AFRICAN ACADEMY OF LANGUAGES AND ALL THE ORGANIZATIONS WHO ARE WORKING.
AND THAT'S WHY I'M HERE.
I'M VERY HAPPY TO BE HERE, YOU KNOW.
I'M HERE AS JUST A SINGLE PIECE OF HAIR IN THE SUIT.
BUT I HAVE HEARD -- FROM WHAT I HAVE HEARD, I HAVE CONFIDENCE.
AND IT DOESN'T CHANGE ANYTHING THAT I HAVE -- I WANTED TO SAY.
BUT I MAY PERHAPS FRUSTRATE SOME PEOPLE, COMPUTER SCIENTISTS OR LINGUISTS.
BUT I'D LIKE TO SAY WHAT I HAVE TO SAY.
AND SO WE CAN WONDER TODAY WHAT THE REFLECTION ON AFRICAN LANGUAGES IN THE DEBATE OF IT TODAY AS -- IN THE SITUATION WHERE WE ARE WITH THE CONVERSATION THAT WE HAVE THIS AFTERNOON AROUND THE DEVELOPMENT OF THE CONTINENT.
TALKING ABOUT ECONOMY, DOMINANT ECONOMY, AND WE -- AS WE CAN SEE, WE ARE WORKING ON DUPLICATING WHAT IS HAPPENING AT THE INTERNATIONAL LEVEL.
WE HAVE THE COMPUTER SCIENTISTS WHO ARE IGNORANT OF WHAT IS -- THEY ARE DOING.
AND THEY ARE NEGLECTING LINGUISTS.
AND THEY ARE RIGHT, AND THEY ARE WORKING IN THEIR RESERVED DOMAIN AND THEY ARE EXPERTS IN THEIR DOMAIN AND THEY HAVE SECRETS OF THE SCIENCE.
AND WHEN THE PEOPLE WHO ARE NOT EXPERTS TRY TO COME CLOSER, THEY RUN THE RISK TO BURN THEMSELVES.
AND THE COMPUTER SCIENTISTS ARE RIGHT, BECAUSE THEY ARE CONSUMERS OF NEW TECHNOLOGY ARE NOT TERRIBLY LINGUISTS.
LINGUISTS, AS I CAN SAY, ARE RESPONSIBLE -- ADAMA SAMASSEKOU HAS HIGHLIGHTED THE ISSUE AS LINGUISTS.
YOU NEED TO GIVE THE LINGUISTS THE OPPORTUNITY TO WORK ON THE LANGUAGES, BECAUSE THEY WILL GIVE YOU THE TIME AND THEY WILL GIVE YOU THE ABILITY TO WORK AND IT WILL PROBABLY LAST LONGER.
LINGUISTS ARE RESPONSIBLE, THEY DON'T MASTER THE SCIENCE OF IT.
BUT WHEN THEY ASK TO -- THEM TO GET CLOSER OR ASK TO EXPLAIN SIMPLY THINGS TO MAKE THEM CLEAR BY SO MANY PEOPLE, THEY WASTE THEIR TIME IN THE REFLECTION THINKING ON THE PEOPLE ARE TIRED, THOSE WHO WANT TO GO FASTER.
I AM NOT A PARTISAN, I AM A LINGUIST.
BUT LINGUISTS ARE ALSO RESPONSIBLE FOR THE DELAY WHEN THEY HAVE TO WORK WITH THE EXPERTS.
BUT THIS VISION THAT I WANTED TO GIVE, WHICH IS A KIND OF DICHOTOMY, SHOULDN'T MAKE US -- DISCOURAGE US.
WE HAVE GOT POSSIBILITIES, ENORMOUS POSSIBILITIES.
AND THIS IS A HOPE THAT I WOULD LIKE TO SHARE WITH YOU THIS AFTERNOON.
AND ON THE ONE HAND, I CAN SAY THAT WITH PASSION, THERE ARE PEOPLE WHO THINK THAT IT'S POSSIBLE TO PUT TOGETHER COMPUTER SCIENTISTS AND LINGUISTS WHO CAN MAKE THINGS WORK.
AND THIS VISION, IF IT'S SHARED, IS THE REASON WHY I AM HERE TO SHARE WITH YOU HOW WE COULD WORK TOGETHER, WHY ARE WE WORKING, YOU KNOW, IN A SEPARATE WAY.
WE HAVE TO WORK TOGETHER TO RENEW AFRICAN RENAISSANCE, THE RENAISSANCE.
THERE ARE A LOT OF RESOLUTIONS THAT HAVE BEEN TAKEN FROM INDEPENDENTS AND WHICH HAVEN'T BEEN IMPLEMENTED BECAUSE PEOPLE ARE WORKING, YOU KNOW, IN ISOLATION, AND THEY ARE DOING WONDERS BUT IGNORING WHAT OTHER PEOPLE ARE DOING.
WE NEED TO BREAK THE WALLS, AND WE HAVE TO WORK IN TRANSPARENCY AND WORK TOGETHER AND WORK -- AND TO REPORT SO THAT EVERYONE BENEFITS FROM OUR WORK.
SO TODAY, YOU KNOW, IT IS EMERGING, AND WE HAVE GOT PRESSURE, WE HAVE GOT TO MAKE PRESSURE ON GOVERNMENT.
AND IF IT HAD BEEN MASTERED LONG BEFORE THAN TODAY WE TALK ABOUT CIVIL SOCIETY, WHICH HAS A VERY IMPORTANT ROLE TO PLAY, THEN EXPERTS HAVE GOT A ROLE TO PLAY, NOT ONLY THE MASTERY, WHERE WE SAY EVERYTHING IS FINE, GOVERNMENT IS GOING TO DO EVERYTHING, AS HAS BEEN SAID.
ONLY WHEN GOVERNMENT WILL FEEL THE PRESSURE OF CIVIL SOCIETY AND EXPERTS WITH CONCRETE THINGS, THEN THE POLITICIANS WILL BE ABLE TO REACT, BECAUSE THEY WANT TO SEE CONCRETE THINGS.
BUT IF IT'S IDEAL, THEY WILL HAVE TO WAIT.
THEN THERE'S SOMETHING THAT PEOPLE DON'T LIKE.
WE ARE HAPPY TO BE COMPLACENT WITH OUR ROLE AND OUR TRADITIONS.
THIS IS REALLY GOOD.
BUT THIS IDEA IS AMUSING FOR EVERYONE.
BUT WE HAVE TO QUESTION THE SYSTEM, BECAUSE AS FAR AS WE ARE -- WE HAVE GOT TO KIND OF -- A KIND OF DELAY NOW ON THE CONTINENT, WHERE OLD TRADITION IS GOOD, BUT WE HAVE TO BE AWARE THAT IF WE -- THE CONTINENT HAD THE TOOLS SO RAPIDLY USED IN THE FORM OF WRITING, WHICH WOULD BE THERE EVERYWHERE, THEY WOULD HAVE DONE IT IN A DIFFERENT WAY.
IF I MAKE THIS COMPARISON, TODAY, WE HAVE TO PUT SOMETHING IN PLACE.
THIS IS THE IT.
AND IF WE DON'T THINK TOGETHER HOW THE LANGUAGES SHOULD NORMALLY BE -- ENRICH THEMSELVES, WE'LL BE REALLY FAR BEHIND.
I HAD A FRIEND, PIERRE OUEDRAOGO, WHO SAID THE GENERATION WILL BE ASKING US QUESTIONS WHEN WE HAVE TO IMPLEMENT THE TECHNOLOGY WHAT WE HAD DONE, AND WE WILL BE HELD RESPONSIBLE, BECAUSE WE HAVE NOT DONE ANYTHING, AND THE LANGUAGES WILL NEVER BE USED IN THE IT.
AND THIS SOCIETY OF IT SHOULD BE A SOCIETY WHERE THERE IS CULTURAL DIVERSITY.
WE HAVE TO FIGHT FOR CULTURES AND LANGUAGES NOT ONLY FACE TO FACE, WHICH IS BETWEEN FRANCOPHONE AND ENGLISH-SPEAKING COUNTRIES, BUT WE HAVE TO LOOK INTO OUR LANGUAGES AS WELL.
WE CAN'T THINK WITHOUT TRYING TO ANALYZE THE SITUATION OF THE -- THE REAL SITUATION ON THE GROUND OF OUR LANGUAGES IN CYBERSPACE.
AND WE SUGGEST THAT WE HAVE TO LINK -- TO MAKE A LINK BETWEEN LANGUAGES AND THE SCIENCE INDUSTRIES.
I TOOK THE CRITERIA SUCH AS IN THE IT DOCUMENTATION, EQUIPMENT, AND AUDIOVISUAL EQUIPMENT. AND WE'RE NOT GOING TO MAKE A LIST OF THE WHOLE THING RAPIDLY.
WE CAN MAKE A ROUND AND SEE IN THESE DOMAINS WHICH WE CAN DO AND HOW OUR LANGUAGES ARE REPRESENTED.
AND THIS IS NOT TO REDUCE THE QUESTION TO A DOMAIN WHERE -- INTERNET DOMAIN.
WE HAVE TO UNDERSTAND THAT IT'S A GLOBAL VISION AND THE VALUE IT HAS AND HOW WE VALUE THESE OTHER LANGUAGES IN THE DOMAIN WHICH IS THE INDUSTRY OF LANGUAGES, DO WE TAKE INTO ACCOUNT OUR LANGUAGES.
AND THIS IS A QUESTION -- IN THE BUSINESS WORLD, THIS IS A REFLECTION THAT WE HAVE TO DRAW FROM.
AND I'VE GOT A TABLE, BUT FOR TECHNICAL REASONS, I AM NOT A SPECIALIST, I DIDN'T PUT IT IN A POWERPOINT, I CAN'T PRESENT THIS TO YOU.
BUT I AM JUST GOING TO MAKE A SUMMARY, SAYING THAT WHAT IS THE LEGAL STATUTE IN THE 24 COUNTRIES, FRANCOPHONE COUNTRIES.
AND IN 24 COUNTRIES, SIX COUNTRIES DON'T HAVE LINGUISTIC LAW IN -- WANT TO TALK ABOUT AFRICAN LANGUAGES, AND WANT TO TALK ABOUT NATIONAL LANGUAGES, BECAUSE WE HAVE GOT CROSS LANGUAGES.
AND IN THE 24 COUNTRIES, SIX HAVEN'T GOT ANY LAW WHATSOEVER.
BUT THIS DOESN'T MEAN THAT IN THE OTHER COUNTRIES THERE ARE -- IT'S WORKING PERFECTLY.
YES, THERE ARE LAWS, YOU KNOW.
WE DON'T NEED TO HAVE A GOOD LINGUISTIC LAW.
BUT WE NEED TO HAVE THE MECHANISM OF IMPLEMENTATIONS.
AND THE LAW DOESN'T COUNT, BUT THE WAY YOU APPLY THE LAW IS VERY IMPORTANT.
AND IF YOU LOOK AT IT, THE LEGAL STATUTE IS -- THEY WOULD NEED TO LOOK AT THE LEVEL OF IMPLEMENTATION.
AND WITHOUT THIS IMPLEMENTATION, WE CAN'T GO ANYWHERE.
WE CAN'T REALLY BLAME THE COMPUTER SCIENTISTS ONLY.
BUT, LEGALLY, WE HAVEN'T TAKEN -- WE WERE NOT PREPARED TO REALLY USE THESE LANGUAGES.
WE CAN'T COMPLAIN THAT ALL THE PEOPLE DON'T TAKE INTO ACCOUNT OUR LANGUAGES, SINCE LEGAL PEOPLE DON'T REALLY -- THE LEGAL FRAMEWORK IS NOT REALLY OPERATIONAL AT ALL.
WE HAVE TO LOOK INTO THE SITUATION IN EACH COUNTRY WHERE THERE ARE SO MANY LANGUAGES.
WE REALIZE THAT WE HAVE DOMINANT LANGUAGES.
IT CAN'T BE THE LANGUAGES OF THE MAJORITY SOMETIMES, BUT FOR SOME SIMPLY OTHER REASONS, THIS LANGUAGE MAY NOT BE SPOKEN BY SO MANY PEOPLE, YOU KNOW, BUT I CAN'T -- I'VE -- ALL THE EXAMPLES WE HAVE IN SENEGAL, THE WOLOF CASE, FOR EXAMPLE, THIS IS SPOKEN BY A FEW PEOPLE, BUT THERE ARE SO MANY PEOPLE SPEAKING THAT LANGUAGE.
AND MORE THAN 36 PEOPLE SPEAKING WOLOF, AND IT GOES BEYOND.
ANOTHER EXAMPLE, THE BAMBARA EXAMPLE, WE -- NOT MORE THAN 30,000 PEOPLE SPEAK THIS LANGUAGE.
BUT WE'VE GOT SO MANY -- MORE THAN 3 MILLION PEOPLE SPEAKING THESE LANGUAGES.
SO THAT MEANS THAT THERE HASN'T -- THE LANGUAGES HAVE GOT THEIR OWN DYNAMICS.
WE HAVE, JUST WITHOUT WASTING TIME, TO USE A LINGUIST TO REALLY DETERMINE THE INFLUENCE OF LANGUAGES.
AND THAT WILL ALLOW TO REALLY TACKLE THE ISSUE.
AND WE CAN SAY THERE ARE SO MANY LANGUAGES WE CAN'T USE THEM.
BUT WE REALLY CATEGORIZE THE WHOLE THING.
WE CAN SEE WHAT CAN BE DONE EVEN FOR COUNTRIES WHO HAVE GOT EVEN MORE THAN 60 LANGUAGES, AND WE CAN SEE WHAT LANGUAGES ARE SPOKEN.
AND THIS IS A STARTING POINT WITH THIS KIND OF SCHEME IN AFRICA, THAT'S A STARTING POINT.
WE HAVE TO LOOK AT THE CROSS-BORDER LANGUAGES.
AND THIS IS A KEY AS A STARTING POINT FROM EACH LANGUAGE, THOUGH EACH COUNTRY WILL LOOK AT THE LANGUAGES WHICH ARE THE MOST SPOKEN AND ACROSS-BORDER LANGUAGES, THERE ARE SO MANY THINGS THAT CAN BE DONE.
WE CAN'T JUST CONSIDER ALL THE LANGUAGES, BUT WE HAVE TO USE IN INDIAN CASE, FOR EXAMPLE, WHERE WE HAVE, FOR EXAMPLE, 18 CONSTITUTIONAL LANGUAGES AND HAVE GOT MORE THAN 1,000 DIALECTS.
BUT THEY'VE TRIED TO WORK ON THIS SITUATION.
AND I'LL COME BACK TO THIS SITUATION LATER.
BUT IN THESE COUNTRIES, WE TALK ABOUT THE SPIRIT AND LEGISLATIVE LAW.
AND THE SPIRIT IS THAT THE LANGUAGE SHOULD BE -- CONTRIBUTE TO THE DEVELOPMENT OF THE COUNTRY.
BUT NO EFFORT HAS BEEN MADE SO FAR.
AND THAT MEANS THAT THESE LANGUAGES ARE JUST NOT MOVING ANYWHERE.
THEY ARE STAGNANT.
AND WE -- IF THE PEOPLE REALIZE THAT, BUSINESSMAN, POLITICIANS, AND THE BIG DECISION-MAKERS ARE NOT INTERESTED IN THE QUESTION, THEY WILL PRODUCE WHAT HAS BEEN REPRODUCED IN THE EDUCATIONAL SYSTEM, AND WE WORK IN THE LANGUAGE, THE FOREIGN LANGUAGE, THE DOMINANT LANGUAGE, WHICH IS THE OFFICIAL LANGUAGE.
SO WHEN I AM TALKING ABOUT THE IDN IN A SIMPLE TABLE, WHEN WE TALK ABOUT THE INDUSTRY OF LANGUAGES, WE HAVE THREE DOMAINS, SIMPLY: COMPUTER SCIENCE, DO WE HAVE A SITE, DO WE HAVE EDOCUMENTATION, OR DO WE HAVE EQUIPMENT?
AND I HAVEN'T FOUND ANYTHING MORE IN -- FOR THE TABLE IN THESE 24 COUNTRIES.
SO WE DON'T FIND IN ANY LANGUAGE IN ANY REALLY SATISFACTORY WAY A BENCHMARK FOR THE SITES WHICH EXIST.
AND THERE ARE A LOT OF SITES, BUT THERE ARE SO MANY ON AFRICAN LANGUAGES.
BUT THE SITES WHICH ARE FUNCTIONAL, IN PERCENTAGE, ARE IN ENGLISH-SPEAKING COUNTRIES.
IN THE FRENCH-SPEAKING COUNTRIES, THERE ARE NOT MANY SITES AT ALL.
THE PRESENCE OF IT IS NONEXISTENT IN A FRANCOPHONE SITUATION.
WE HAVE GOT TO WORK VERY HARD.
THE OBSTACLE WE HAVE FACED, WE HAVE TO TALK ABOUT THEM, WHAT ARE THE STUMBLING BLOCKS IN THIS FRAMEWORK?
WE CRITICIZED THE COLONIAL SYSTEM A LOT.
BUT IN THIS SYSTEM, YOU KNOW, PUT IN PLACE, THE KIND OF TYPE OF SCHOOLS AND THE TYPE OF PEOPLE, THE ADMINISTRATION AFTER INDEPENDENCE DIDN'T KNOW EXACTLY WHAT TYPE OF PEOPLE AND TYPE OF SOCIETY WE HAVE TO PUT IN PLACE.
UNDER THIS ARE REALLY STUMBLING BLOCKS, BECAUSE THEY HAVE GOT SIDE EFFECTS ON OUR EDUCATIONAL SYSTEM.
AND SOCIAL SYSTEMS AS WELL.
AND AS YOU CAN SAY, FRANCOPHONE COUNTRIES HAVE GOT 20% OF PEOPLE SPEAKING FRENCH AND 10% OF PEOPLE SPEAKING PORTUGUESE, AND WE REPRODUCED THESE MISTAKES INDEFINITELY.
AND IN THIS CONTEXT, WE CAN WONDER WHAT TYPE OF SOCIETY OF IT WE WANT TO PUT IN PLACE, BECAUSE IT'S A QUESTION, ISSUE WHICH WAS NOT REALLY ANSWERED ON 3,000 SITES WHICH WERE UNDER STUDY, WHICH WERE VISITED.
HALF, YOU KNOW, OF THEM SPEAK TWO LANGUAGES.
AND 84 ONLY IN AFRICAN LANGUAGES.
AND I WOULD SAY, YOU KNOW, CONGRATULATIONS TO THE ANGLOPHONE SIDE, BECAUSE THEY'VE GOT THIS SITE IN SENEGAL.
ON THE FRANCOPHONE SIDE, THERE IS NO SITE WHICH IS OPERATIONAL SO FAR IN A LANGUAGE IN A FRANCOPHONE COUNTRY.
84, AS I SAID, OUT OF 3,000 CALCULATED.
AND, YOU KNOW, IF -- FOR ME, IT'S NONEXISTENT.
AND THE OBSTACLES ARE NOT HUMAN ONLY.
WE HAVE GOT THEM ON THE TECHNICAL SIDE.
WE HAVE LACK OF SOFTWARE AND, YOU KNOW, IT SOLUTIONS, AND WE WOULD SAY THE COMPUTER SCIENTISTS AND OTHERS HAVE TO REALLY WORK ON THIS QUESTION, ON THE QUESTION OF NATIONAL ALPHABETS AND AFRICAN ALPHABETS EXIST.
BUT IT WAS ADOPTED.
BUT I WOULD LIKE TO SAY THAT IF THERE IS SOMETHING MISSING, WE HAVE TO COMPLETE IT.
WE DON'T HAVE TO WAIT INDEFINITELY AND SAYING THAT THERE WERE ALPHABET WHICH WERE ENCODED, THE AFRICAN ALPHABET WHICH WERE ENCODED.
BUT THE ONLY OBSTACLE IS THE QUESTION OF UNCODING THE ENCODING SYSTEM.
AND -- OF THE ALPHABET WHICH IS IN LATIN CHARACTER, WE HAVE TO WORK WITH THAT TO MOVE FORWARD, OR WE WORK WITH RESEARCH AND ACADEMICS TO REWORK ON THE ALPHABETS.
AND I SEE SPECIFIC CHARACTERS AND REINVENTED ALPHABETS, AND WE COULD JUST USE THEM.
AND WE ARE WAITING FOR THE YEAR BEFORE WE MOVE ON.
THIS IS KIND OF A STRATEGIC CHOICE THAT WE HAVE TO MAKE.
THIS IS JUST FOR THE CASE OF RESEARCH.
WE HAVE TO BE CONCRETE AND ADOPT SOME ALPHABETS.
WE HAVE TO ENCODE THEM USING THE EXISTING SYSTEM.
AND IF THERE IS NO ENCODING SYSTEM, THEN WE WON'T RECOGNIZE ANYTHING.
SO AFTER HIGHLIGHTING HUMAN AND TECHNICAL ASPECTS WHICH ARE THE STUMBLING BLOCK, THERE ARE SOME OTHERS.
AND TO START WITH, OR TO HELP THE AFRICAN LANGUAGES TO BE PRESENT, WE DO NEED TO LOOK AT THE LINK BETWEEN THE ETHICS AND THE EDUCATIONAL SYSTEM.
WITHOUT ANY REFORM OF THE EDUCATIONAL SYSTEM, WE ARE USING BILINGUAL SYSTEM TO ENSURE AFRICAN LANGUAGES IN THE SYSTEM, THEN THERE IS NO VALIDATION OF THE LANGUAGE, WE CAN'T REALLY ENSURE PRESENCE OF THE LANGUAGES IN THE SYSTEM.
THERE IS AN EXAMPLE WHICH WAS GIVEN BY MAURICE DEPRIVIEU AND (INAUDIBLE) TO SHOW THAT THERE ARE SITES, THERE ARE SITES AND THERE ARE PEOPLE UNDERGOING DISCUSSION IN SWAHILI, LINGALA, FOR EXAMPLE, AND IF YOU SEE THAT, THERE ARE SITES WHERE DISCUSSIONS ARE CARRIED OUT IN THOSE LANGUAGES.
AND THESE LANGUAGES ARE TAKEN INTO ACCOUNT IN THE EDUCATIONAL SYSTEM.
AND THESE LANGUAGES HAVE BEEN REALLY VALUED.
AND WE DON'T HAVE TO WORK ON ONE SIDE OF SAYING THAT OTHER LANGUAGES ARE JUST THERE TO HAVE SITES AND WE IGNORE THEM IN OUR DAILY LIFE.
IT'S NOT ONLY IN THE SOCIETY OF INFORMATION, BUT WE HAVE TO IMPLY ALL THE EDUCATIONAL SYSTEM AND THEN WE HAVE TO WORK VERY HARD AND MOVE FORWARD FURTHER.
AND WHAT SHOULD WE DO?
WE ARE NOT JUST GOING TO BE INTIMIDATED BY DIFFICULTIES, UNDERSTANDING AND STANDSTILL.
WE HAVE TO REALLY TRY HOW TO WORK HARD, WE HAVE TO SEE WHAT -- TO ANSWER THE QUESTION.
PEOPLE WONDER OH, WE COULD HAVE A SITE IN AFRICAN LANGUAGES, DOES IT WORK?
I THINK THAT IT DOES.
I THINK THAT IT WORKS.
WE HAVE TO REALLY ALLOW EACH COUNTRY TO PUT -- LINGUISTS SHOULD COME TOGETHER AND WORK ON THIS ISSUE, AND THEY HAVE TO CREATE SITES, TO WORK ON PRESENTATION OF A PAGE, FOR EXAMPLE, AND CHOOSE THEIR OWN CONTENT TO SHOW THAT IT WORKS.
ENTIRELY IN THE LANGUAGE OF THEIR CHOICE TO SEE THAT IT WORKS AND POLITICIANS TO SEE THAT IT WORKS, THEN IT WILL MAKE THINGS CHANGE.
ANOTHER ELEMENT THAT NEEDS -- THERE IS THE QUESTION OF HOW TO -- THE LOCATION OF THE SITES.
WE COULD JUST NORMALLY SEE A URL, THE SECRETARY OR THE ORGANIZATIONS.
AND TO HAVE WE HAVE THE SOLUTIONS TO LOOK AT THE SITES.
AND THIS HERE, ALL THE VISITED SITES IN IDN HAVEN'T GOT A NAME, AN IDN, AND WE HAVE -- AND THE NEW SITES THAT WE HAVE TO CREATE SHOULD BE -- SHOULD HAVE SPECIFIC NAMES ON THE LINGALA, FOR EXAMPLE, SWAHILI, AND POINT OF DOMAIN WHICH IS CLEAR. NOT A SITE WHICH GIVES US A DIFFERENT COMBINATIONS THAT ARE VERY DIFFICULT AND WE DON'T KNOW WHETHER THEY ARE REALLY, YOU KNOW, SATISFYING OUR CONCERNS, ET CETERA.
SO CREATORS ARE FREE TO DO ANYTHING, BUT TO ENSURE THE PRESENCE OF VISIBILITY IN THE SAME WAY AS WE SAY, WE DON'T HAVE TO HAVE FIGURES, JUST FIGURES, BUT WE HAVE TO HAVE VERY SIMPLE THINGS THAT HELP US TO RESPECT THINGS.
ACALAN AS I'VE HEARD, THE AFRICAN COMMISSION, ECONOMIC COMMISSION, HAS GOT ECAP, FOR EXAMPLE, AND HAVE WRITTEN THAT PRESENCE, ADAMA SAMASSEKOU TALKED ABOUT EIGHT, WE HAVE TO COORDINATE THE ACTIVITIES AND MAINLY TO HAVE OBJECTIVES THAT ARE CLEAR AND HOW TO MEET THE OBJECTIVES. AND WITH THIS, IT SHOWS A VISIBILITY THAT WE CAN REALLY HAVE. AND WE HAVE -- WE WON'T LOSE ANYTHING, AND PEOPLE ARE WONDERING WHETHER IT WORKS, AND WE WILL HAVE ONE ADDRESS, FOR EXAMPLE, AT THE ACADEMY OF LANGUAGES AND SEE WHAT HAPPENS AND WHAT IS GOING ON ON AFRICAN LANGUAGES AND WITH THE I.T. AND WHAT IS BEING DONE ET CETERA.
AND AS I JUST SAID, YOU KNOW, WE DON'T HAVE TO BE JUST LIKE THE PEOPLE HAVE GOT TO REINVENT THE WHEEL. WE CAN'T SAY THERE IS NO ALPHABET ADOPTED OR ENCODED IN EACH COUNTRY. THE THINGS EXIST, BUT THEY HAVEN'T BEEN ENCODED, AND THE STRUCTURES, AND WE HAVE TO STANDARDIZE THINGS.
THERE ARE POSSIBILITIES. THE PROBLEM IS THAT IS AN ISSUE TODAY. WE DON'T HAVE TO GO THAT WAY.
IF WE DON'T WANT TO GO THAT WAY, WHAT I WOULD SAY IS THE DECISION, DO WE HAVE TO CREATE ALPHABETS? IS IT AFRICAN? I DON'T THINK IT HELPS THE DEVELOPMENT OF AFRICA TO SLOW DOWN IN THAT WAY.
IN CONCLUSION, AFRICAN LANGUAGES ARE NOT A BREAK TO THE DEVELOPMENT, THE ETHICAL DEVELOPMENT. THOSE WHO ARE WORKING ON THE QUESTION SAY IT'S NOT GOING TO DELAY THE PROCESS. PEOPLE ARE WORKING IN THE I.T. SHOULD BE CONVINCED THAT AFRICAN LANGUAGES, ON THE CONTRARY, BY STREAMING THEM IN THE VISION OF THE SITUATION AND THE PRACTICAL SIDE, WILL ALLOW A SHARE OF KNOWLEDGE, AND WILL ALLOW ACCESS OF A LARGE NUMBER OF PEOPLE IN THIS TECHNOLOGY. IT'S A QUESTION OF ORGANIZING THE WHOLE THING, AND THAT WILL ALLOW THE DEVELOPMENT OF EVERYONE. IT'S NOT THE DREAM OF LINGUISTS, BUT IT'S GOOD TO CLEARLY SHOW THE WHOLE THING, TO SHOW THAT EACH AFRICAN LANGUAGES, ESPECIALLY CROSS BORDER LANGUAGES BE ABLE TO SHOW THE CULTURAL DIVERSITY INSTEAD OF HAVING STERILE FACE TO FACE. THANK YOU VERY MUCH.
>>MOUHAMET DIOP: THANKS VERY MUCH, MAXIME. IT'S A VERY INTERESTING SESSION. WE'RE JUST TRYING TO PROVE OURSELVES THAT WE CAN SPEAK DIFFERENT LANGUAGES AND TRY TO UNDERSTAND EACH OTHER.
QUITE A BIG CHALLENGE.
THE NEXT SPEAKER WILL BE ALIREZA SALEH, ROOZBEH POURNADER FROM THE PERSIAN EXPERIENCE.
SO WE WILL HAVE HERE A DESCRIPTION OF A PROJECT THEY HAVE ALREADY, AN IDN PRESENTATION, AND THEY HAVE BEEN THE DESIGNER OF THE PROJECT FOR THE LANGUAGE, AND IT'S VERY INTERESTING EXPERIENCE THAT WE'RE GOING TO LEARN FROM THEM.
SO YOU'VE GOT THE FLOOR.
>>>: OKAY. GOOD AFTERNOON, LADIES AND GENTLEMEN, MY NAME IS ALIREZA SALEH FROM IRNIC AND WE ARE REPRESENTING THE IDN PROJECT IN IRNIC AND THE WORK WITH FARSIWEB COMPANY.
WE ARE JUST GOING TO SKIP THE INTRO ABOUT IDN BECAUSE WE SUPPOSE EVERYONE IS FAMILIAR WITH IDN AND ITS CONCEPTS, SO WE START THE PRESENTATION.
ACTUALLY, THIS PROJECT HAS BEEN STARTED ON 2002, BUT IT'S DELAYED FOR A LITTLE BECAUSE OF CHANGING SOME POLICIES FOR REGISTERING DOT IR DOMAINS. IT'S FREE TO EVERYONE AND WE HAVE A DIFFERENT DISPUTE RESOLUTION POLICY ABOUT IT.
WE HAVE JUST FINISHED THE POLICIES OF THE TEST BED AND NOW THEY ARE FINALIZED AND WRITTEN, AND THEY WILL BE SOON AVAILABLE ON THE WEB SITE OF IRNIC TO BE -- THAT WILL BE USED IN -- THAT WILL BE USED FOR THAT.
SO WE ARE PLANNING TO START OUR IDN PROJECTS AT THE LATE IN DECEMBER. WE HAVE SOME TECHNICAL PROBLEMS WITH MERGING OUR IDN CODES WITH OUR REGISTRATION SOFTWARE. SO WE ARE NOW AT THE POSITION OF MERGING THESE CODES, AND WE HOPE THAT WE CAN START THINGS VERY SOON IN THIS MONTH. AND I THINK THAT -- WE THINK THAT MAYBE THE DURATION OF THE TEST BED WOULD BE AROUND SIX MONTHS. THAT IS STARTED RIGHT AFTER WE STARTING THE TEST.
THE OFFICIAL LANGUAGE IN IRAN IS PERSIAN, AND THIS IS THE ONLY OFFICIAL LANGUAGE THAT WE ARE USING IS PERSIAN, BUT WE HAVE DIFFERENT KINDS OF LANGUAGE ALSO THAT WE ARE -- THAT THEY ARE LIKE AZARBAIJANI, KURDISH, AND BALOCHI AND ARABIC. SO WE ARE WORKING ON ALL OF THESE LANGUAGES, AND NOW WE ARE TECHNICALLY BREAK OUT THE DRAFT OF HOW WE HAVE TO APPLY THESE LANGUAGES.
BUT FOR OUR RECENT TEST BED AND TEST PERIOD, WE OFFICIALLY START WITH PERSIAN LANGUAGE. THAT IS OUR OFFICIAL LANGUAGE. AND WE ARE THINKING ABOUT SOME ZONES THAT WE -- SOME ZONES THAT WE HAVE TO USE AT TOP LEVEL FOR IR.
THE IDEA FOR THE TOP-LEVEL DOMAINS IS THAT THE IDN ZONES IN PERSIAN, THE DOT COM, DOT NET AND DOT BIZ, THE IDEA IS THAT THE COMPANIES OR ORGANIZATION USING THIS IN -- IN IRAN. AND ALSO WE DO OUR RESEARCH ON THE MAJOR NAMES THAT CAN BE USED AS A IDN ZONES. AND WE CALCULATE IT AROUND 30 PERSIAN, THAT MOST OF THEM -- THAT ALL OF THEM ARE TEMPORARY AND JUST USE AT THE BEGINNING AND WE WILL CONTINUOUSLY -- WE WILL CONTINUOUSLY PROCESS THE FEEDBACK AND STATISTICS ABOUT THE THINGS THAT COMES TO OUR -- THE ENTRIES THAT PEOPLE LIKE AND THE SUGGESTION ABOUT THE PEOPLE OF IRAN.
ACTUALLY, I THINK THAT'S HOW WE CAN START THE -- HOW WE CAN -- HOW THE CLIENTS CAN USE THIS TOP LEVEL -- THIS PERSIAN ZONES. WE ARE SETTING UP SOME RESOLVERS TO RESOLVE FULLY PERSIAN DOMAINS. THE USERS CAN SET THEM AS THEIR RESOLVER AND THEN USE THAT FULLY IDN URL.
ALSO, THERE IS A POSSIBILITY FOR THEM TO ADD THE DOT -- THE DOT IR AT THE END OF THE URL, AND THEN THEY CAN USE ANY RESOLVER IN THE WORLD FOR RESOLVE THEIR DOMAINS. BUT ACTUALLY, THE IDEA BEYOND -- THE IDEA BEHIND THE IDN IS TO USE THE FULLY ONE SCRIPT, THAT THE URL SHOULD BE WRITTEN IN JUST ONE SCRIPT LANGUAGE, NOT JUST SWITCH THE KEYBOARD, BECAUSE WE ASSUME THE USER IS JUST USING THE MOTHER TONGUE OR ITS ORIGINAL LANGUAGE.
ALSO, AS YOU KNOW, MOZILLA-BASED BROWSERS LIKE NETSCAPE SUPPORT IDN INTERNALLY AND THE USERS CAN USE IT. WE ARE JUST -- WE -- IN OUR FAQ, IT WILL BE SOON AVAILABLE ON OUR WEB SITE, WE TRY THE CUSTOMERS TO USE IT, AND ALSO THEY CAN DOWNLOAD AND INSTALL THE PLUG-IN FOR IE. AND WE HOPE THAT MICROSOFT WILL SOON SUPPORT THE IDN.
OKAY. ACTUALLY, THE TEST BED IS FREE, AND NO MONEY EXCHANGED, WE HAD DURING THE TEST BED AND EVERYBODY CAN JUST COME AND SIGN UP FOR IDN DOMAIN. BUT WE JUST LIMITED THIS TO THE USER THAT HAD ALREADY -- THAT IS ALREADY -- IS A CONTACT OF AN ACTIVE DOMAIN, SUCH AS TECHNICAL HANDLE, BILLING HANDLE, OR ADMINISTRATIVE HANDLE OF THAT DOMAIN.
AND IT JUST SHOWS THAT WE CAN TRACK HIM AND WE CAN FIND HIM IF THEY DO SOMETHING THAT IS NOT -- IT JUST MAKES SOMETHING BAD FOR OUR SERVICE.
WE ASSUME THAT WE WILL DEFINITELY PURGE ALL THE DOMAINS THAT HAS BEEN REGISTERED DURING THE TEST BEDS, AND THEN WE ARE JUST USING IT FOR -- WE ARE JUST USING IT FOR DECISION. AND AFTER REAL -- AFTER REAL LABELS THAT WE HAVE -- WE HAVE TO USE. BUT IT'S JUST FOR AN EXPERIMENT, AND MAYBE THEY WERE NOT USEFUL AT ALL. BUT WE HAVE TO GO AND GO FOR IT.
THE PEOPLES THAT ARE PARTICIPATING IN THIS TEST BED HAS OFFICIALLY NO -- NO CHANCE BETTER THAN THE OTHERS, BUT MAYBE THEY HAVE SOME PRIVILEGES DURING THE SUNRISE PERIOD THAT WE PLAN TO HAVE A SUNRISE PERIOD FOR A SPECIAL WELL-KNOWN COMPANIES, AND ALSO THEY WILL HAVE THE SAME DOMAIN AND ASCII RIGHTS NOW. THEY WERE INFORMED BEFORE THE OTHERS, AND THIS IS THE CHANCE FOR THEM.
OKAY. WE ARE JUST REACHING TO THE TECHNICALITIES THAT I ASK MR. POURNADER TO CONTINUE THERE.
>>ROOZBEH POURNADER: OKAY. THANK YOU. IS THIS TOO HIGH? OKAY.
SO I'M GOING TO FOLLOW ON WITH THE TECHNICALITIES INVOLVED IN THE DETAILS.
THE BASIC LABELS, AND THIS IS RELATED TO LINGUISTIC ISSUES INVOLVED THAT WE'RE TRYING TO SOLVE HERE, WE DECIDED TO LOAD JUST PERSIAN LETTERS, PERSIAN DIGITS AND THE FAMOUS HYPHEN MINUS CHARACTER SO PEOPLE CAN HAVE SOMETHING LIKE A SPACE FOR THE DOMAINS.
OF COURSE, IN PERSIAN THERE'S ALSO ANOTHER CHARACTER LOOKING LIKE A SPACE, BUT NOT A SPACE. WE CALL IT A ZERO WIDTH NONJOINER. SOME PEOPLE CALL IT A PSEUDO SPACE OR ZERO WIDTH SPACE. THERE'S NOT ACTUALLY A CHARACTER KNOWN AS THAT BUT THE NAME IS NONJOINER FOR THAT.
BUT STILL, HYPHEN MINUS WOULD BE GOOD. PEOPLE ARE USED TO SEEING IT IN DOMAIN NAMES.
WE WILL NOT BE ALLOWING DIACRITICAL MARKS IN THE PERSIAN LANGUAGE. IT IS STILL TECHNICALLY CORRECT. AND IT'S LIKE THAT FOR HEBREW.
AND THERE ARE TWO CHARACTERS I TOLD YOU ABOUT, ZERO WIDTH JOINER AND ZERO WIDTH NONJOINER. IT'S NOT REALLY IGNORING, BUT YOU CAN ASSUME THAT IT'S IGNORED. AND THERE WILL BE THE IDNA BIDIRECTIONAL RESTRICTIONS WHICH DICTATE WEIRD THINGS, REALLY. FOR EXAMPLE, THAT'S LABELS SHOULD START AND FINISH WITH LETTERS, IF THEY ARE RIGHT TO LEFT, WHICH IS THE CASE OF THE PERSIAN LANGUAGE.
SO WE ARE FOLLOWING THAT.
WE'VE ADDED A CERTAIN RESTRICTION ABOUT LABELS WITHOUT ANY LETTERS. SO LABELS ONLY MADE OF PERSIAN DIGITS AND HYPHENS WOULD NOT BE ALLOWED. THIS IS SOMETHING WE CAME BY AFTER LOTS OF DISCUSSIONS, AND IT LOOKS SIMPLE BUT IT WASN'T REALLY THAT SIMPLE. EVEN DETERMINING WHAT IS THE REAL RANGE OF PERSIAN LETTERS WASN'T EASY FOR US. WE NEEDED TO DISCUSS MATTERS. THERE WERE PEOPLE WHO WERE AGAINST CERTAIN LETTERS BECAUSE THE LETTERS HAD CERTAIN FOREIGN FEEL TO IT. YOU WON'T BELIEVE THE POLITICAL PROBLEMS. FOR EXAMPLE, THERE IS THIS LETTER WHICH IS -- WHICH PEOPLE USUALLY CONSIDER ONLY AN ARABIC LETTER, AND THAT -- IT'S ONLY USED IN A VERY FEW PERSIAN WORDS, AND THOSE WORDS, BECAUSE OF THEIR ARABIC ORIGIN, THEY'RE ALSO CONSIDERED A LITTLE ARABIC. BUT THEY'RE PROPER PERSIAN WORDS, THE PERSON SHOULD HAVE THE RIGHT TO USE THEM. THEY'RE EVEN USED IN CERTAIN ORGANIZATION NAMES. SO WE CAN'T IGNORE THEM. ORGANIZATION NAMES IN IRAN. BUT PEOPLE USUALLY DON'T LIKE TO -- LETT!
ERS, I MEAN LIKE THE LETTER IN THE ALPHABET. YOU GET WHAT I MEAN.
SO THE REASON FOR NOT ALLOWING LABELS WITH NO LETTERS, I MEAN, THEY WILL BE PERSIAN DIGITS ONLY, AND THAT WILL BE A LITTLE PROBLEM BECAUSE THERE MAY BE A CONFUSION WITH THE SAME NUMBER WRITTEN WHOLLY IN ASCII, WHICH WILL BE NON-ALLOWED IN ANOTHER WAY.
SO WHAT WE'RE DOING HERE IS WE'RE TRYING TO DISCONNECT ANY CONNECTION BETWEEN THE DOMAIN WHICH IS IN LATIN SCRIPT AND THE PERSIAN DOMAIN NAME WHICH IS IN THE ARABIC SCRIPT.
SO WE'RE PROBABLY THE FIRST COUNTRY REGISTRY TO IMPLEMENT BUNDLES FOR AN ALPHABETICAL LANGUAGE. I'M NOT TALKING ABOUT JAPANESE, KOREAN. PEOPLE USUALLY TRY TO CALL THIS THE ABJABS, AND IT'S SOMETHING LIKE AN ALPHABET. AND OUR BUNDLES IS MAINLY TRYING TO SOLVE A CERTAIN PROBLEM. THE PROBLEM IS THAT MANY OF OUR USERS HAVE NONSTANDARD KEYBOARDS, NONSTANDARD KEYBOARDS FOR THE PERSIAN LANGUAGE, BECAUSE THEY'VE NOT BEEN FOLLOWING UNICODE PROPERLY AND THEY'VE BEEN USING CUSTOMIZED CHARACTER SETS OR CUSTOMIZED FONTS, WHICH ARE NOT STANDARD.
FOR EXAMPLE, THEY USE A CERTAIN PERSIAN LETTER INSTEAD OF -- THEY USE A CODE FOR A CERTAIN ARABIC LETTER INSTEAD OF THE PROPER PERSIAN LETTER USED FOR THAT THING.
I WILL GIVE YOU MORE VISUAL EXAMPLES FOR THIS, BUT THIS IS THE GENERAL IDEA THAT PEOPLE ARE USING NON-STANDARD FONTS AND KEYBOARDS VERY FREQUENTLY. WE WILL TRY TO USE THE IDN OPPORTUNITY IN ORDER TO MAKE THEM MORE CONFORMING TO THE STANDARDS. I MEAN, INSTALL PROPER FONTS, INSTALL UNICODE CONFORMING FONTS OR OF COURSE RENDERING TECHNOLOGY. BUT WE WILL TRY TO MAKE SURE THAT IF THE USER IS NOT AWARE OF THE SITUATION AND -- LET'S SEE, HE MAY SEE SOMETHING ON A BOSS OR ON A VOL AND IT MAY BE A DOMAIN NAME IN VERSION. AND THEN HE GOES INTO HIS COMPUTER AND TIES THAT IN AND HE'S USING A STANDARD KEYBOARD. OKAY. NOTHING COMES OUT AND THE USER DOESN'T KNOW ANYTHING ABOUT IT. WHY IS THIS FAILING? HE CAN'T GET A CLUE.
SO WHAT WE ARE TRYING TO DO HERE IS FIX THE SITUATION FOR THE END USER BUT NOT THE MIDDLE -- NOT POSSIBLY THE REGISTRANT OR PEOPLE LIKE THAT WHO MAY FIND OUT ABOUT THE SITUATION BY VISITING OUR WEB SITE.
SO THIS IS A BUNDLE, AND THERE'S ALSO THE SECURITY PROBLEM. BECAUSE AN ARABIC LETTER OF YE, FOR EXAMPLE, WHICH IS USED IN CERTAIN FONTS, IN CERTAIN NONSTANDARD FONTS, MAY LOOK EXACTLY LIKE A VERSION YE BECAUSE THE FONT IS NOT A STANDARD. SO IT'S REALLY SOMEHOW USING A "B" TO DISPLAY A "C." AND IF THE USER IS SEEING "C"S EVERYWHERE, HE MAY THINK THIS IS REALLY A C. BUT TO THE COMPUTER AND UNDERNEATH THAT, IT'S A B.
SO WE WILL TRY TO MAKE CERTAIN THERE IS NO SECURITY PROBLEM AND THE USERS WILL BE ABLE TO USE THESE DOMAIN NAMES. WE ARE USING THESE TABLES FOR ARABIC INDIC DIGITS WHICH ARE DIFFERENT FROM PERSIAN DIGITS AND IN SHAPE OF COURSE AND FOR EUROPEAN DIGITS WHICH IS WHAT YOU KNOW IN THE WHOLE WORLD AS DIGITS. BUT THEY'RE ONLY USED RARELY IN PERSIAN CONTEXT. ONLY MOSTLY IN SCIENTIFIC CONTEXT.
SO IF THOSE THINGS ARE USED IN THE DOMAIN, WE WILL TRY TO NORMALIZE THEM FIRST. SO IF YOU DO ARABIC DIGIT OR A EUROPEAN DIGIT 4, PEOPLE CONVERT IT TO A PERSIAN DIGIT 4 IN ORDER TO MAKE SURE THAT IT'S THE REGISTERED THING. AND IN ORDER TO ALLOW THAT THING, WE'LL ALLOW A BUNDLE OF AT MOST SIX LABELS SO WE CAN HELP THE INTEGRATION. THIS LIMITATION WAS MAINLY DONE TO MAKE SURE THAT WE ARE NOT TALKING ABOUT HUGE -- I MEAN, SIZES OF BUNDLES. AND I WILL EXPLAIN IT TO YOU A LITTLE LATER.
SO THIS IS THE TABLE I'M TALKING ABOUT, AND THIS IS THE SIX THINGS YOU SEE HERE.
DO YOU HAVE ME? OKAY.
SO THIS, ACTUALLY, IS THE PROPER THING (INDICATING). THE ROWS AND COLUMNS, YOU CAN SEE, THESE ARE THE TWO LETTERS KOF AND YE WHICH ARE A LITTLE DIFFERENT IN THEIR SHAPE AND BEHAVIOR IN PERSIAN AND IN ARABIC. BUT THESE DIFFERENCES ARE SO IMPORTANT THAT THE UNICODE HAS ASSIGNED DIFFERENT CHARACTER CODES TO THESE TWO.
I REALLY BELIEVE THAT MANY OF THE COUNTRIES WHO HAVE NOT BEEN -- I MEAN, IN REALLY GOOD CONNECTION WITH THE BIGGER WORLD WILL HAVE SOME OF THESE PROBLEMS. I WOULD REALLY EXPECT THIS WITH SOME AFRICAN COUNTRIES, SOME EASTERN EUROPEAN COUNTRIES. I'M NOT SURE WHERE THE WHOLE LIST WILL GO, BUT THIS HAS CREATED -- THESE HAVE BEEN PROBLEMS. IT'S NOT ONLY, OF COURSE, THE COMMUNICATION PROBLEM WITH THE WHOLE WORLD. IT SOLVES THE COMMUNICATION PROBLEM INSIDE THE COMMUNITY.
SO IF YOU HAVE SEPARATE COMMUNITIES USING DIFFERENT KIND OF COMPUTER SOFTWARE WHICH MAY HAVE BECOME NOT A STANDARD IN THE MEANWHILE, YOU'LL POSSIBLY HAVE SUCH KIND OF PROBLEMS.
SO LET ME GET BACK TO THE SHAPES.
THIS IS ALSO -- AND THESE ROWS, AGAIN, YOU CAN SEE THE NUMBERS IN PERSIAN, ARABIC. THEY SOMEHOW LOOK THE SAME, BUT FOUR OF THEM HAVE DIFFERENT SHAPES, ACTUALLY. ALTHOUGH THESE WILL ALSO BE ACCEPTABLE IF THESE SHAPES TO A PERSIAN READER OR USER, THE UNICODE CODE POINTS FOR THOSE SIMPLY ARE DIFFERENT BECAUSE THEY'RE ALSO BEHAVIORAL DIFFERENCES BETWEEN THESE TWO SETS. AND THE LAST ROW IS OF COURSE THE EUROPEAN NUMBERS THAT YOU ALL KNOW.
SO THIS IS MAINLY -- WE WILL CONSIDER THIS -- WE'LL CONSIDER THIS THE MAIN -- DOMAIN NAME BUT WE WILL ALSO GIVE YOU THESE OTHER FIVE WHICH ARE THE COMBINATION OF ROWS AND TABLES, TO MAKE SURE THAT ALL OF YOUR USERS, DEPENDING ON THEIR CERTAIN CONFIGURATIONS, CAN WORK PROPERLY.
AS AN EXAMPLE, MICROSOFT CURRENTLY HAS THIS ON THEIR KEYBOARD. (INDICATING). UNFORTUNATELY, BECAUSE OF TECHNOLOGY RESTRICTION, THEY DON'T HAVE PERSIAN DIGITS ON THEIR KEYBOARD. ON A PERSIAN KEYBOARD LAYOUT THEY USE EUROPEAN NUMBERS. SO IF SOMEBODY IS USING A MICROSOFT KEYBOARD ON SAY WINDOWS, FOR EXAMPLE, THEY WILL BE ONLY ALLOWED -- THEY WILL ONLY BE ABLE TO ENTER THIS (INDICATING).
MOST OF THE NONSTANDARD PERSIAN IMPLEMENTATIONS OR SOMETHING WE CALL FARSI-FICATION PACKS WILL BE USED ON THE COMPUTER. THEY HAVE NO IDEA THEY ARE USING ARABIC KOFS AND YES AND EUROPEAN DIGITS. SO WE HAVE TO MAKE SURE THIS TRANSLATES TO THIS AND THEY BELONG TO THE SAME GUY. OKAY. SO THE PROCESS OF DOMAIN NAME IS SOMETHING LIKE THIS. WE FIRST CHECK THE VALIDITY. IF THIS IS A VALID INTERNATIONALIZED DOMAIN NAMES. WE WILL BE ADDING MORE NAMES TO IT WHEN WE FIND INFORMATION.
AND THERE ARE ALSO OTHER ISSUES, WHEN WE MIX LANGUAGES, WE WILL TRY TO DO SOMETHING ABOUT THEM.
BUT WE HAVE THE INFRASTRUCTURE FOR THAT.
BUT SINCE WE ONLY HAVE THE INFORMATION FOR PERSIAN, WE'RE NOT DOING MUCH IN THAT YET.
OKAY.
AND CHECKING THE VALIDITY IS VERY EASY.
AND WE ARE ONLY HERE IF THE LABEL IS NOT IN THE FAMOUS LDH FORM, FAMOUS LETTER, DIGIT, HYPHEN.
IF IT'S NOT LDH, WE ALREADY HAVE A GOOD PROCESS FOR TAKING CARE OF THAT.
WE ONLY GET HERE IF THE LABEL IS NOT LDH.
CHECKING IS EASY BECAUSE POLICY TAKES CARE OF MOST OF THAT.
WE CAN MAKE SURE THAT, I MEAN, IF ALL THE LETTERS ARE IN A CERTAIN SET THAT WE'VE MENTIONED, THEY WILL POSSIBLY MEET ALL OF THE STRINGPREP AND NAMEPREP REQUIREMENTS. AND THE CHECKS ARE MAINLY FIRST WE CHECK THAT ALL OF THE LETTERS IN THE LABEL ARE IN THE PROPER ORDER AND THAT THE LETTERS SHOULD START AND END WITH LETTERS.
THE PROCESS, FIRST OF ALL, WHEN NORMALIZED TO PROPER LETTERS AND DIGITS, SO IF YOU'RE ANYWHERE IN THIS TABLE, FOR EXAMPLE, OR EVEN YOU'RE MIXING DIFFERENT THINGS, FOR EXAMPLE, AN ARABIC KOF WITH A PERSIAN YE, YOU WILL GET TO THIS THING AS THE MAIN REGISTERED LABEL.
AND THIS THING WILL BE CONSIDERED THEIR MAIN REGISTERED LABEL.
AND THEN WE'LL REMOVE WHAT STRINGPREP REMOVES, WHICH ARE MAIL TO (INAUDIBLE) WHICH ARE USED IN PERSIAN.
BUT THEY'RE ALLOWED IN THE LABELS.
SOMETIMES THESE ARE REQUIRED IN THE SPELLING.
SPELLING IN A LABEL OR NAME WITHOUT THIS WILL BE A MISSPELLING.
AND THEN WE WILL GO TO COMPUTING THE BUNDLE, WHICH IS AT MOST 6, BUT USUALLY LESS.
THERE MAY BE A CASE THAT YOU DON'T HAVE ANY KOF OR YES OR ANY DIGITS IN YOUR LABEL.
SO THE BUNDLE MAY ONLY BE ONE LABEL.
AND THEN WE WILL COMPUTE THE ASCII EQUIVALENTS USING THE TOASCII THING INTO IDN, IDNA PROCESS.
IF THE MAIN REGISTERED LABEL IS TOO LONG AFTER THIS TOASCII, THAT, ACTUALLY, DNS WILL REJECT IT.
AND IF THE ALTERNATE BUNDLE MEMBERS ARE TOO LONG, WE WILL DO A CASE-BY-CASE BASIS.
IF EACH OF THEM FAILS, WE WILL REJECT THAT.
FINALLY, JUST TO MAKE SURE THAT THE USER IS SEEING THE RIGHT THING AND IS DOING WHAT HE OR SHE WANTS TO DO, WE WILL SHOW HIM A PICTURE, AN IMAGE OF WHAT IS BEING REGISTERED, NOT ONLY AS TEXT.
SO WE CAN MAKE SURE THAT THIS IS -- THEY CAN MAKE SURE THAT THIS IS WHAT THEY ARE REALLY REGISTERING AND THIS IS WHAT THEY REALLY WANT, SO THEY CAN STOP IT IF THAT'S NOT IT.
AND THIS IS ALMOST ALL OF THE TECHNICAL DIFFICULTIES.
THERE ARE LOTS OF OTHER DETAILS INVOLVED THAT I CAN GET IN LATER MEETINGS OR AT THE QUESTION PART.
SO AFTER THE TEST BED, WE WILL BE -- WHAT WE ARE REALLY PLANNING TO DO, BASED ON THE EXPERIENCE, WE'LL POSSIBLY CHOOSE SOME OF THESE LABELS AND THEN START TALKING TO ICANN TO GET SOME OF THOSE ACCEPTED AS PERSIAN GTLDS.
THIS IS A CHOICE, OF COURSE.
OR WE MAY TRY TO LOBBY FOR A DOT IRAN THING, WRITTEN IN ARABIC SCRIPT, OF COURSE, OR SHORTHAND OR ABBREVIATION FOR THAT.
WE DON'T KNOW.
WE WILL DECIDE ON THAT BASED ON THE FEEDBACK WE GET FROM THE USERS.
AND THOSE ZONES THAT WE GAVE TO OUR USERS, THEY COULD NOW BE USED UNDER THE -- SUCH A COUNTRY TLD, NOT A COUNTRY CODE TLD REALLY.
IT'S WHAT I LIKE TO CALL A CTLD.
OR WE DECIDE, NO, IT'S NOT WORKING.
OUR USER WOULDN'T LIKE THAT.
IT'S NOT WORTH OF EFFORT.
I MEAN, SOMEBODY ELSE SHOULD START THE WORK BASED ON WHAT THEY DO.
USUALLY THESE ARE PEOPLE WHO DO GTLD -- TRY TO HAVE GTLDS IN LOCAL NAMES.
WE DON'T KNOW YET, OF COURSE.
AND EVERYTHING WILL DEPEND ON THAT EXPERIENCE, WHAT WE FIND ABOUT THE USERS AND THE STATISTICS WE ARE GOING TO GATHER DURING THE WHOLE PROCESS.
WE WILL BE TRYING TO FOLLOW EVERYTHING, I MEAN, WHATEVER USER MISTAKES, WHAT ARE THEY TRYING TO DO, AND WHAT HAPPENS, REALLY.
AND I GUESS IT WILL BE EASY MONITORING.
OKAY.
I BELIEVE I'M FREE FOR QUESTIONS NOW, IF ANYTHING -- COULD WE HAVE QUESTIONS?

>>MOUHAMET DIOP: YES.
>>ROOZBEH POURNADER: AND HOW MUCH TIME DO WE HAVE FOR THAT?
SORRY?
I'M ALREADY OVER TIME?
OH, SORRY.
>>MOUHAMET DIOP: WHAT I SUGGEST IS WE HAVE -- WE HAD A PRESENTATION.
WE'RE GOING TO TAKE 15 MINUTES FOR QUESTIONS AFTER THE PRESENTATION.
AND WE'RE GOING TO GO FOR THE BREAK BEFORE THE SECOND SESSION ON THE IDN APPLICATION DEVELOPMENT.
THANK YOU VERY MUCH.
AND, SO -- SO PIERRE OUEDRAOGO WILL MAKE A PRESENTATION ON THE UNICODE, AS THE NEXT SPEAKER.

(APPLAUSE.)
>>MOUHAMET DIOP: I GUESS I'M GOING TO ASK TINA TO HAVE THE PRESENTATION AFTER THAT TO MAKE THEM AVAILABLE ON THE WEB SITE FOR PEOPLE WHO....
>>PIERRE OUEDRAOGO: OKAY.
THANK YOU.
I'LL BE SPEAKING OF DOING MY PRESENTATION IN FRENCH.
AND I HOPE THAT YOU'VE GOT YOUR HEADSETS FOR INTERPRETING.
NO ONE HAS GOT ANY.
IS WE'VE GOT HEADSETS OUTSIDE.
I'M GOING TO GIVE FOUR POINTS THAT ARE IMPORTANT IN THE ORGANIZATION PROCEDURE UNDER -- AT THE LEVEL OF THE MEMBERSHIP, THE TIP TOP OF MEMBERS THAT WE HAVE, AND IN THE CONTEXT -- IN THE AFRICAN CONTEXT.
I WAS A LITTLE BIT EMBARRASSED BY THE OFFER THAT WAS MADE TO ME TO DO MY PRESENTATION, BECAUSE HERE THERE ARE BETTER SPECIALISTS THAN ME WHO KNOW BETTER ABOUT THIS ISSUE.
I AM A NEW MEMBER OF THE CONSORTIUM, AND I JOINED IN JUNE 2004 WHEN WE STARTED PARTICIPATING ACTIVELY IN THE MEETINGS.
BUT I UNDERSTAND THAT I'M ALSO THE FIRST AFRICAN TO SIT IN THE PRELIMINARY CONTACT.
THAT'S WHY THEY ASKED ME TO DO THE PRESENTATION.
BUT I KNOW THAT THERE ARE PEOPLE IN THIS ROOM WHO KNOW NOT MUCH BETTER THAN ME.
THAT'S WHY I AM GOING TO TALK ABOUT THE PRESENTATION BEFORE TALKING ABOUT WHAT IS INTERESTING FOR AFRICA ON THE QUESTION DEALING WITH THIS ORGANIZATION.
WE KNOW THAT THERE ARE FIGURES, WHAT WE HAVE AS A SIGN, AS THEY ARE NOT FIGURES, THEY ARE NOT UNDERSTOOD AT ALL.
BUT WE KNOW THAT THERE WAS A KIND OF ANARCHY SOMETIMES BECAUSE SOMETIMES THERE WAS SOME NUMBER WHO COULD NOMINATE, YOU KNOW, A LETTER OR A LETTER WHICH COULD INDICATE TWO LETTERS, ACCORDING TO SYSTEMS OF -- ENCODING SYSTEM, THERE WAS A RISK THAT -- FROM A SYSTEM TO ANOTHER PLATFORM, THERE WAS A REALLY DISTORTION OF THE DATA.
THEN IT WAS CLEAR THAT COORDINATION BE MADE IN THIS DOMAIN.
AND THAT'S WHY UNICODE WAS BORN, UNDER THE IMPULSION OF THE INDUSTRY, THE IT INDUSTRY.
MAINLY, THERE ARE BIG INDUSTRY -- IBM, MICROSOFT, ET CETERA -- AND WE'VE GOT IN THIS ROOM ANOTHER MEMBER OF THE CONSORTIUM WHO IS PRESENT, MICHEL FROM MICROSOFT, WHOM WE MET WHEN WE HAD A MEETING DEALING WITH (INAUDIBLE) UNICODE.
IT'S USELESS TO TALK ABOUT THE IMPORTANCE OF UNICODE.
BUT WE KNOW THAT IF WE HAVE TO COMMUNICATE, WE NEED TO HAVE THE INTEROPERABILITY.
THAT IS VERY IMPORTANT TO HAVE THE SAME CODE.
AND THERE ARE, YOU KNOW, A REDUCTION OF COST BECAUSE IT'S LESS COMPLEX, AND WE DO LESS TO GO FROM ONE SYSTEM TO ANOTHER.
THERE IS SOMETHING THAT IS ADDED, AND WE HAVE THE OPPORTUNITY TO SEE THAT THERE WERE ADDITIONS EVEN IN LANGUAGE THAT ARE CONSIDERED TODAY AS NOT BEING USED, THERE WERE ADDITIONS STILL.
AT THE LEVEL OF THE ORGANIZATION, RAPIDLY, IT IS AN ORGANIZATION WHICH IS A NONPROFIT ORGANIZATION-MAKING WHICH HAS GOT A BOARD WITH EIGHT DIRECTORS, AND ARE COMING FROM BIG COMPANIES.
AND THESE ARE THE PEOPLE WHO REALLY WORK IN THE UNIFORMIZATION OF THE WHOLE THING.
AND THE HEADQUARTER IS IN CALIFORNIA.
AND THE OBJECTIVE OF UNICODE IS TO DEVELOP AND TO PROMOTE THE USE OF THE LISTENING STANDARD.
AND AS FAR AS THE STRUCTURE IS CONCERNED, WE HAVE TWO TECHNICAL COMMITTEES, UTC, WHICH IS IN CHARGE OF THE NORMS AND THE STANDARDS OF UNICODES, AND THE LDC, WHICH IS IN CHARGE OF LOCATION.
AND WE ARE GOING TO BE INTERESTED IN THE UTC.
THIS IS WHERE DECISIONS ON CODES OF CHARACTER -- CHARACTER CODES ARE MADE.
AND THE UTC ARE NOT ONLY IN CHARGE OF DEVELOPMENT, BUT ALSO THE MAINTENANCE OF TECHNICAL STANDARD OF THE UNICODE.
AND ALSO THEY PUBLISH REGULARLY.
THEY MEET FOUR TIMES A YEAR, AND THE REPORTS ARE ONLINE.
AND THE PARTICULARITY IS THAT MEMBERS CAN VOTE, NOT ONLY FULL MEMBERS, AS IS WRITTEN IN ENGLISH, BUT ALSO THE ASSOCIATED MEMBERS.
UTC RECEIVED PROPOSALS OF NEW CHARACTERS.
AND THESE ARE SUBJECT TO -- THEY SHOULD COME IN A FORM WHICH IS PREDEFINED.
I GIVE YOU THE WEB SITE TO SEE WHERE THEY ARE INTERESTED, K 1 SCRIPT PUBLISHED IN UNICODE, WE DON'T REMOVE IT, THIS CAUSES PROBLEMS FOR STABILITY, AND THERE IS SOME POSSIBILITY THERE IS NOT BETTER THAT WE'VE GOT 4.0, THERE IS A BOOK WHICH IS PUBLISHED, AND IT'S A BIG BOOK, GOOD WE HAVEN'T BROUGHT, IT'S TOO BIG.
AND WE ALSO HAVE A BOOK WHICH IS IN A FORM OF FILES.
SO WE HAVE AT THE LEVEL OF REPORTS THREE TYPES OF REPORTS.
WE'VE GOT THE SAME JARGON THAT WE FIND IN THE -- FROM ICANN.
AND THIS IS THE TECHNICAL REPORTS.
THESE -- OUR UNICODE STANDARD ANNEXES THAT IS PART OF THE STANDARD, UNICODE STANDARD, AND ALL THE ACRONYMS ARE THERE.
AND WE'VE GOT XUTS AND WE'VE GOT A URL WHICH WILL BE AVAILABLE IN THE PRESENTATION, BECAUSE IT WILL BE ONLINE TO ALLOW YOU IF YOU WANT TO DEEPEN -- YOU KNOW, TO KNOW MORE.
AND THE QUESTION THAT I'M ASKING MYSELF IS, WHEN I JOIN THEM, I WONDER HOW IT WAS WORKING.
WE HAVE TO LOOK AT THE CODES.
AND I DISCOVERED THAT, WORKING TOGETHER, IT WAS VERY GOOD.
AND THE VERSION WAS SYNCHRONIZED.
ALSO, UNICODE ADDS SOME CONSTRAINTS OF SPECIFICATIONS WHICH ALLOW THE CHARACTERS TO HAVE A UNIFORM -- A KIND OF UNIFORM APPLICATION.
AND THESE, THE APPLICATIONS WHICH ARE REALLY CLEAR, AND YES.
NOW, AS FAR AS THE TYPE OF MEMBERS THAT THERE ARE OR ARE CONCERNED, THE MEMBERS WHO CAN PARTICIPATE, IT'S AN OPEN ORGANIZATION, AND PEOPLE SOMETIMES DON'T KNOW.
THIS IS WHAT I DISCOVERED WHEN I DISCUSSED WITH MANY PEOPLE IN AFRICA, THEY DIDN'T KNOW THAT IT WAS OPEN.
THEY DIDN'T KNOW THAT IT EXISTS.
THEY SAW IT IN THE MACHINE BUT THEY DIDN'T KNOW WHAT WAS BEHIND IT.
IT'S AN ORGANIZATION WHERE WE HAVE DIFFERENT ORGANIZATIONS, AND THERE ARE SOCIETIES, THERE ARE GOVERNMENTS, AND INTERNATIONAL ORGANIZATIONS.
AND RESEARCH INSTITUTIONS AND INDIVIDUALS AS WELL.
AND TO BE A MEMBER, WHATEVER CATEGORY, YOU HAVE TO PAY A PARTICIPATION FEE FOR -- TO ALLOW CONSORTIUM TO BE RUN.
THERE ARE FULL MEMBERS, THERE ARE 16 AT THE MOMENT.
AND AMONG THEM, THERE ARE THREE ONLY WHICH ARE NOT A CORPORATION IN THE SECTOR OF IT.
AND THEY ARE GOVERNMENT OF PAKISTAN AND INDIA, AND THE INTERGOVERNMENTAL AGENCE OF FRANCOPHONIE, WHICH I HAVE PRESENTED HERE. AND THESE ALL HAVE GOT THE RIGHT, AND THEY HAVE NO RESTRICTION.
THERE IS A SMALL BUSINESS WHERE THERE IS ONLY ONE MEMBER.
I DON'T KNOW WHY.
AT THE LEVEL OF ASSOCIATE MEMBERS, THERE ARE 36 MEMBERS.
AND INDIVIDUALS, THERE ARE EVEN MORE.
IN THE LAST CATEGORY, WE HAVE THE LIAISON CATEGORY, WHICH ALLOWS THE ORGANIZATION, THE TECHNICAL ORGANIZATION, LOOKING AT THE STANDARDS IN THE UNICODE, THE DOMAIN, TO PARTICIPATE IN THE MEETING AND THEIR INVITATIONS.
AND IT'S JUST LIKE ANY OTHER ORGANIZATION WITH THE TEXT, ET CETERA.
AND AT THE LEVEL OF MEETINGS, WE HAVE -- I HAVE NOTED THREE TYPES OF MEETINGS: ANNUAL MEETINGS.
THIS IS STATUTORY MEETINGS WHERE MEMBERS, FULL MEMBERS, PARTICIPATE.
WE'VE GOT TECHNICAL COMMITTEE, WHERE EVERYONE CAN PARTICIPATE, AND NOT EVERYONE VOTE IN THE SAME WAY.
WHEN YOU HAVE FULL MEMBERS, YOU'VE GOT FULL VOTES, POWER.
IF YOU COME FROM A SMALL BUSINESS, YOU'VE GOT HALF A VOTE.
AND THE VOTE IS PROPORTIONAL TO THE CONTRIBUTION.
THEN THIS IS WHAT HAPPENS.
IT WORKS AS THE SHARES, HAVING SHARES.
THERE ARE PUBLIC MEETINGS, NOTABLY, A CONFERENCE, INTERNATIONAL CONFERENCE OF UNICODE WHICH TAKES PLACE IN AMERICA, NORTH AMERICA.
AND THE 27TH ONE WILL TAKE PLACE IN BERLIN, IN GERMANY, FROM 5TH TO 8TH APRIL.
WE HAVE TO PAY A MEMBERSHIP IF YOU HAVE TO ATTEND THERE TO HELP.
THERE ARE A LOT OF SERVICES WHICH ARE INCLUDED IN THE FEE.
THOSE WHO WANT TO GO THERE WILL GET INFORMED ABOUT THE PARTICIPATION.
AND NOW I'D LIKE TO TALK ABOUT THE AFRICAN PERSPECTIVE, AND NOTABLY, THE INITIATIVE OF INTERGOVERNMENTAL AGENCE OF THE FRANCOPHONIE.
THE FRANCOPHONIE, AS YOU KNOW, IS AN AFRICAN INITIATIVE.
IT'S NOT A FRENCH OR BELGIAN INITIATIVE.
THE THREE HEAD OF STATES, AFRICAN STATES (INAUDIBLE) FROM TUNISIA WHO REALLY LAID A FOUNDATION OF FRANCOPHONIE, AND IT'S IN -- THE IN THE FRANCOPHONIE CHARTER, IT'S CLEARLY STIPULATED THAT NATIONAL LANGUAGES OF THE COUNTRY MEMBERS ARE PARTNERS OF FRENCH.
IT'S NOT RESERVED TO FRENCH PEOPLE IN ISOLATION; IT'S VALID FOR COUNTRY MEMBERS.
TAKING THIS PRINCIPLE INTO ACCOUNT, THERE IS A SUPPORT TO THESE LANGUAGES TO DEVELOP THEIR SUPPORT ON INTERNET AS FAR AS ENCODING IS CONCERNED.
UNDER THE NAME OF THIS PRINCIPLE, SO THIS IS THE ACTION OF FRANCOPHONIE IN THE CONTEXT OF AFRICA.
WE HAVE, THEN, HIGHLIGHTED ON THE NETWORK, WHICH WAS BAPTIZED IN A (INAUDIBLE) AND CONCENTRATE ON THE ISSUE OF ENCODING.
FRANCOPHONIE HAS BEEN A MEMBER NOW TO ALLOW AFRICAN, YOU KNOW, AGENDA AT THE LEVEL OF UNICODE AND INCREASE THE TECHNOLOGICAL AUTONOMY OF AFRICA IN THE FIELD OF THE ENCODING.
THE FIRST AFRICANS TO PARTICIPATE IN THE UTC HAVE BEEN STARTED IN 2004 WITH THE HELP OF FRANCOPHONIE.
AND IT WAS IN WASHINGTON.
AND THE UTC AT TORONTO FOR THE FIRST TIME, THE AFRICANS WERE PRESENT IN THIS FORUM. AND IT WAS VERY SURPRISING FOR AFRICANS, WHEN AFRICA IS INTERESTED IN THESE KINDS OF ACTIVITIES ON THIS NETWORK, WORK CLOSELY IN RELATION WITH ICANN TO FACILITATE THE UPDATING AND FINALIZATION OF THE ALPHABET, AFRICAN ALPHABET, BASED ON THE LATIN CHARACTER, WITH THE BASIS -- AS BASIS OF WORK REALIZED IN 1998 IN NIAMEY WITH THE AFRICAN ORGANIZATION WHICH HELPED AFRICA TO HAVE ALPHABET, WHICH WAS ON A TEMPORARY BASIS, WHICH IS THE ALPHABET THAT YOU HAVE.
THERE IS A NUMBER OF CHARACTER THAT IS YOU HAVE SEEN IN THE LAST UTC COMMITTEE WHICH ARE NOT IN.
Q WITH A KIND OF TAIL, INVERTED TAILS, AND ALL SORTS OF CHARACTERS.
AND I WOULD LIKE TO STOP AT THIS LEVEL AND -- PERHAPS TO SAY THAT WE ARE LUCKY IN AFRICA, BECAUSE THERE IS NOTHING -- NOTHING HAS BEEN DONE FULLY.
OUR FRIEND FROM IRAN HAS TALKED ABOUT DIFFICULTIES THAT WE HAVE WITH THE CHARACTERS, AND WE'RE AMAZED IN JULY IN KUALA LUMPUR TO SEE THE MALAY WRITING.
IT'S VERY IMPORTANT, WRITING IN A SPECIFIC WAY.
WE HAVE TO USE CHARACTERS, AND THEY ARE NOT BIGOTRY, THEY ARE USED TO EXPRESS PHONEMES, I THINK THIS IS THE EXPRESSION USED BY LINGUISTS THEMSELVES.
IF THAT WAS THE CASE, FOR EXAMPLE, IN THE ENGLISH, FOR EXAMPLE, WHEN WE LEARNED ENGLISH, THERE WAS PHONETICS IN ENGLISH.
AND WITH CHARACTERS, THAT WAS -- THEY WERE VERY DIFFICULT.
AND WE WOULDN'T BE ABLE TO WRITE ENGLISH IF WE HAD TO WRITE WITH PHONETICS.
AND AS I USUALLY SAY, IF IN ENGLISH WE TRIED TO LOOK AT ALL OF THE DIFFERENCES IN THE WAY THE PEOPLE SPEAK AND THE WAY IN ASIA PEOPLE SPEAK, WE WOULD BE ABLE TO REALLY GET TO A VERY COMPLEX, I THINK, AND VERY COSTLY, AND SIMPLIFICATION HAS ALLOWED TO HELP LEARNING, AND ALSO AT THE LEVEL OF INDUSTRY, THERE IS A DIMENSION, THERE IS A SCALE OF ECONOMY WHICH HAS BEEN PERCEIVED.
AND THIS IS OUR DEBATES THAT SHOULD NORMALLY BE TACKLED NEXT YEAR IN THE AFRICAN LEVEL.
AND THEN WE HAVE TO TAKE INTO ACCOUNT TO CONTRIBUTE TO THE PLAN OF ACTION AT THE LEVEL OF UTC IN BRINGING ALL THE PROBLEMATICS OF AFRICA AT THIS LEVEL.
AND ALSO WE HAVE TO ORGANIZE TO SENSITIZE EVERY -- IF POSSIBLE, ALL THE FORUM, AFRICAN FORUM. AND THERE IS A PROJECT TO HAVE WORKSHOP WITH ACALAN AND THE REGIONAL CONFERENCE AT THE PREPARATORY MEETING AT THE WORLD SUMMIT.
AND THIS WILL TAKE PLACE IN ACCRA.
AND FROM THE 29TH TO THE 4TH OF FEBRUARY, NO -- NO, 24TH JANUARY TO 4TH FEBRUARY, THAT WILL BE -- THERE IS ANOTHER ONE WHICH WILL BE ORGANIZED IN APRIL.
AND WE'LL JUST CARRY OUT THIS ACTION WITH ALL THE EXPERTS AND THE DECISION-MAKERS.
AND WE ALSO HAVE TO PUT -- TO PUT -- TO MAKE -- TO PUT THE FOCUS ON THE ALPHABET.
THAT WON'T COMPLICATE THE WHOLE THING.
THESE ARE THE POLITICAL DECISIONS, AS SUCH, AND THE -- SOME TECHNIQUES WILL BE DONE WITH THE AU.
AND WE HOPE THAT THIS WILL HELP THE STREAMLINING TO AVOID DIVIDING AFRICA, IN A DOMAIN WHERE WE CAN AVOID IT.
BORDERS EXIST, BUT WE CAN'T REALLY DIVIDE AFRICA WITH THE KEYBOARDS.
AND THE TECHNICAL SUPPORT IS VERY IMPORTANT, AFRILANG, WITH WHOM WE WORK.
AND WE'VE GOT TO SUPPORT AND TO IMPROVE THE AFRICAN INITIATIVES WITH THE SOFTWARE, WHICH SHOULD BE ADAPTED BASED ON THE CODE.
AND WE HAVE TO CONTRIBUTE TO THE DEVELOPMENT OF -- THE REINFORCEMENT OF THE CAPACITY IN THE IDN AND ON THE GROUND WITH THE ORGANIZATIONS THAT HAVE, YOU KNOW, MENTIONED PREVIOUSLY, ICANN, FOR EXAMPLE.
IN CONCLUSION, I WOULD LIKE TO SAY FOUR POINTS.
FIRST OF ALL, UNICODE IS VERY IMPORTANT.
IT'S VERY IMPORTANT FOR THE STREAMING OF REGIONAL INTEGRATION OF AFRICA.
WE DON'T HAVE TO CREATE DIVISIONS.
WE HAVE TO MAKE DECISIONS, AND IN FRONT OF HISTORY WE HAVE TO DO A FOLLOW-UP.
AND THE AFRICAN CHARACTERS NEED TO BE NORMALIZED.
BECAUSE FROM 1978, THERE WERE WORK IN DIFFERENT COUNTRIES, AND THERE WAS NO SUMMARY TO GIVE A NEW VERSION WHICH COULD BE ACCEPTABLE BY ALL AFRICANS.
AND I THINK THAT RESEARCHERS HAVE WORKED VERY HARD, AND IT'S TIME THAT THEY PUT EVERYTHING TOGETHER AND TO TRY TO WORK AND TO -- THIS IS MY OWN WISH.
AND I WOULD LIKE TO CONVINCE OUR BROTHERS AND SISTERS, AFRICAN SISTERS WHO ARE HERE TO HELP US IN THIS VISION.
AND WE'VE GOT THE IP, AND IT IS BASED ON CODES.
AND WE HAVEN'T FINISHED YET AT ALL.
WE ARE NOT SURE.
AND MAYBE WHEN WE DECIDE ABOUT THE ALPHABET, WE WILL HAVE THE CHARACTERS OF UNICODE IN THESE CONDITIONS AND THE PROBLEM WILL BE DEFINITELY SOLVED.
BUT IF THERE ARE SOME WHICH ARE NOT, THEN WE HAVE TO DO THE WORK.
AND THIS IS WHAT I WANTED TO SAY.
AND AT OUR LEVEL, WE ARE OPEN.
AND WE WANT ALL THE VOLUNTEERS TO JOIN US TO HELP US IN THIS ACTION IN THE IT DEVELOPMENT IN AFRICA.
AND I THANK YOU VERY MUCH.

(APPLAUSE.)
>>MOUHAMET DIOP: THANK YOU VERY MUCH, PIERRE.
I THINK THAT WE GOT, SINCE THE BEGINNING OF THIS PRESENTATION, EVERYTHING IS GOOD NEWS.
I THINK THAT WE'LL HAVE ENOUGH RESOURCES TO INTERACT AFTER THIS SESSION.
VALENTIN NEMETH FROM SOUTH AFRICA, YOU ALSO HAVE A PRESENTATION TO MADE ON CHARACTERS AND CERTIFICATES.
>>VALENTIN NEMETH: HELLO.
VERY GOOD AFTERNOON, EVERYONE.
WE HAD A VERY LONG AFTERNOON SO FAR.
AND I THINK IT'S TIME TO HAVE SOME FUN AND TALK ABOUT SOME SERIOUS SUBJECT.
LET'S TALK ABOUT SECURITY.
AND WE'RE GOING TO SET UP THE MACHINE IN A SECOND.
OKAY. WELL, WHILE THE MACHINE IS BOOTING UP, CAN EVERYONE HEAR ME OKAY? IS IT ALL RIGHT?
WE HAVEN'T REALLY -- WE TALKED ABOUT A LOT OF ASPECTS AROUND IDN, HOW DOES IT WORK, WHAT ARE THE LANGUAGE ISSUES, WHAT OTHER ISSUES ARE THERE. WE HAVEN'T REALLY SPOKEN ABOUT SECURITY.
IF WE LOOK AT THE INTERNET, THE BUSINESS ON THE INTERNET IS REALLY DRIVEN BY SECURITY. YOU WANT SECURE TRANSACTIONS, YOU WANT YOUR INTERNET BANKING SECURE, E-COMMERCE NEEDS TO BE SECURE.
SO HOW DOES IDN FIT INTO THIS PICTURE?
WHAT DO THE -- WHAT DO WE NEED TO LOOK AT AND TALK ABOUT?
OUR MACHINE IS STILL BOOTING UP.
RIGHT. SO....
YEAH, WE HAD A LITTLE -- YEAH. DO WE HAVE A REPRESENTATIVE OF MICROSOFT HERE WHO COULD COMMENT ON IT?
(LAUGHTER.).
>>VALENTIN NEMETH: OKAY. SO A LITTLE BIT OF A BACKGROUND ON IDN. NONE OF THE TALKS UNTIL NOW HAVE REALLY SPOKEN ABOUT HOW IDN WORKS AND WHAT IT IS. SHALL WE HAVE A TWO-MINUTE SORT OF BACKGROUND ON IT?
THE BASIC IDEA IS IDN MEANING INTERNATIONALIZED DOMAIN NAME, IT ALLOWS YOU TO HAVE THE INTERNET IN YOUR OWN LANGUAGE, MEANING THAT THE DOMAIN NAME, THE HOST NAME AND THE URL, THE WEB SITE, IT CAN BE IN YOUR OWN NAME. IT DOESN'T HAVE TO BE ENGLISH, AND IT CAN BE CHARACTERS THAT ONLY PEOPLE LIVING IN ASIA AND OTHER PLACES KNOW.
NOW, HOW DOES THAT WORK?
THE TECHNICAL REALIZATION IS REALLY A TRANSFORMATION. THE IDN LIVES IN TWO FORMS. IT LIVES IN THE NATIVE LANGUAGE FORM, WHICH IS WHAT PEOPLE TALK ABOUT AND WHAT PEOPLE USE, AND THERE'S A TECHNICAL REPRESENTATION AS WELL, A MACHINE READABLE REPRESENTATION, WHICH IS REALLY USED FOR TRANSACTIONS.
SO THAT'S THE KEY POINT, IN MY OPINION, WHAT MAKES -- NO, THAT'S NOT THE SAME. I THINK I'VE GOT A --
>>ROOZBEH POURNADER: IT'S HERE NOW.
>>VALENTIN NEMETH: FANTASTIC.
SO AFTER ALL THE GETTING STARTED, TREMENDOUS.
SO JUST TO CLOSE OFF THE CHAIN, WITH IDN WE HAVE TWO REPRESENTATIONS. WE HAVE THE NATIVE LANGUAGE REPRESENTATION AND WE HAVE GOT THE MACHINE READABLE REPRESENTATION WHICH DOESN'T LOOK ANY SIMILAR TO THE NATIVE LANGUAGE REPRESENTATION. AND THAT IS NEEDED FOR THE NAME RESOLUTION, THE DNS TRANSACTION, AND FOR THE INTERNET TO WORK, REALLY.
SO I'VE BEEN TALKING SO MUCH, BUT I HAVEN'T REALLY INTRODUCED MYSELF, AND OUR COMPANY. SO WHO ARE WE?
WE ARE THAWTE CONSULTING BASED IN CAPE TOWN, SOUTH AFRICA, HENCE THE AFRICAN CONTEXT. WE ARE THE SECOND LARGEST CERTIFICATE AUTHORITY WORLDWIDE. WE STARTED OPERATING IN 1995 AND ISSUED THE FIRST DIGITAL CERTIFICATE IN 1996.
WE ARE THE FIRST CERTIFICATE AUTHORITY TO ISSUE A CERTIFICATE OUTSIDE OF THE UNITED STATES. AND SO FAR, IN THE LAST EIGHT YEARS, WE ISSUED ALMOST HALF A MILLION SSL CERTIFICATES, INCLUDING CODE SIGNING CERTIFICATES AND THREE QUARTER MILLION PERSONAL CERTIFICATES IN 172 COUNTRIES.
SO I THINK THAT PROVES THAT WE ARE PRETTY INTERNATIONAL.
WE EMPLOY OVER 100 STAFF. WE SUPPORT 28 LANGUAGES ACTIVELY, AND WE ARE DEDICATED TO THE OPEN SOURCE COMMUNITY.
AND IN 1999, THE COMPANY HAS BEEN ACQUIRED BY VERISIGN, BUT WE STILL OPERATE AS THAWTE AND AS AN INDEPENDENT BUSINESS UNIT.
AND OUR MARKETING FOLKS PUT TOGETHER A VERY NICE MISSION STATEMENT, WHICH EVERYONE CAN READ. I WOULD LIKE JUST TO MOVE ON.
SO WE TALKED ABOUT IDN. LET'S TALK ABOUT SECURITY.
HOW DOES TRUST WORK? HOW DOES SECURITY WORK TODAY? HOW DO WE TRUST THE WEB SITE WHERE WE PAY MONEY, WHERE WE BUY THINGS, WHERE WE DO OUR BANKING?
WELL, THE NUTS AND BOLTS ARE THAT AN ORGANIZATION, BE IT A BANK, BE IT A COMPANY, AN E-TAILOR, YOU WOULD GO TO A CERTIFICATE AUTHORITY AND REQUEST A DIGITAL CERTIFICATE. THE CA, THE CERTIFICATE AUTHORITY, WILL VERIFY THAT THE DOMAIN THAT YOU WOULD LIKE TO SECURE INDEED BELONGS TO HIM AND THE CA ALSO MAKES SURE THAT THIS PERSON REPRESENTS A VALID ORGANIZATION AND HE IS REALLY THE PERSON OR THE ORGANIZATION HE SAYS HE IS, WHICH IS CALLED AUTHENTICATION.
ONCE A DIGITAL CERTIFICATE IS ISSUED, IT'S INSTALLED ON A WEB SITE, AND USERS CAN VISIT THE SITE.
WHEN A USER ACTUALLY HITS THE SITE, WHAT HAPPENS? THE BROWSER WILL DOWNLOAD THE DIGITAL CERTIFICATE AND THE DIGITAL CERTIFICATE HAS THE DOMAIN NAME EMBEDDED IN IT. THE BROWSER COMPARES THE EMBEDDED DOMAIN NAME WITH THE DOMAIN NAME OF THE DOWNLOAD PATH. SHOULD THERE BE A MISMATCH, THEN THE BROWSER WILL SAY, HEY, I HAVE A PROBLEM. I DON'T THINK WE SHOULD TRUST. AT THAT POINT, THE USER TYPICALLY DOESN'T TRUST. THE BROWSER DOESN'T SHOW THE CLOSED PADLOCK AS YOU'VE PROBABLY SEEN, AND THERE'S NO TRANSACTION, WHICH MEANS NO BUSINESS, REALLY.
AND SHOULD THERE BE A MATCH, SHOULD THE DOWNLOAD AND THE DOMAIN NAME PART AND THE DOMAIN NAME IN THE CERTIFICATE MATCH, THEN THE BROWSER IS HAPPY, NO WARNING, CLOSED PADLOCK AND THE USER CAN MOVE ON AND DO HIS TRANSACTION.
NOW, WHAT IS REALLY THE PROBLEM WITH IDN? I MENTIONED THAT IDN, INTERNATIONALIZED DOMAIN NAMES, LIVE OR HAVE TWO FORMS. THE NATIVE LANGUAGE FORM AND THE PUNYCODE ENCODED FORM WHICH IS THE MACHINE READABLE, YOU COULD CALL IT -- THERE IS THAT WORD BUT THERE WE ARE.
SO THE COMPARISON BETWEEN EASTERN EUROPEAN, WESTERN EUROPEAN LANGUAGES, ASIAN, AND WHAT IS THE NATIVE LANGUAGE FORM AND WHAT IS THE PUNY ENCODED FORM. IF YOU LOOK AT IT WHICH ONE CAN YOU REALLY UNDERSTAND OR READ?
I WOULD SERIOUSLY ARGUE THAT IT'S THE LEFT-HAND SIDE, NOT THE RIGHT-HAND SIDE.
SO IF WE MAKE SECURITY AND TRUST ON THE INTERNET WORK, AND KNOWING THAT THE TRANSACTIONS ON THE INTERNET NEED TO USE THE PUNYCODE ENCODED FORM, THAT WOULD MEAN THAT WE NEED TO PUT THE PUNYCODE ENCODED FORM INTO A DIGITAL CERTIFICATE WHICH IS NOT READABLE TO THE USER. THE USER DOESN'T RECOGNIZE IT AS A VALID DOMAIN NAME AND HE DOESN'T TRUST. AND IF HE DOESN'T TRUST HE DOESN'T DO THE TRANSACTION AND WHERE IS THE BUSINESS GROWTH?
SO JUST LET'S LOOK AT AN EXAMPLE.
RIGHT. SO LET'S LOOK AT -- I THINK THINGS ARE JUMPING HERE, BUT WE CAN JUMP, TOO.
SO LOOKING AT WHAT IS THE PROBLEM, AGAIN, THE TWO FORMS, WE CAN PUT THE TWO FORMS INTO THE DIGITAL CERTIFICATE AS WELL. WE CAN MAKE THE HUMAN READABLE FORM OF THE CERTIFICATE OR THE MACHINE READABLE PART OF THE CERTIFICATE.
NOW, WHAT HAPPENS? IF YOU MAKE THE NATIVE LANGUAGE FORM PART OF THE CERTIFICATE, THE USER WILL TRUST. THE PROBLEM IS IT DOESN'T WORK. WE NEED THE PUNYCODE FORM IN THE CERTIFICATE FOR IT TO WORK. BUT IF WE DO THAT, THE USER DOESN'T TRUST BECAUSE HE DOESN'T RECOGNIZE THE DOMAIN NAME. SO AGAIN, NO TRUST, NO TRANSACTION.
SO HERE IS THE EXAMPLE I WAS LOOKING FOR. THIS IS A WEB SITE, THE USER TYPED IN THE NATIVE LANGUAGE NAME. THE BROWSER HAD THE NECESSARY PLUG-IN. IT RESOLVED, LOADED THE PAGE, AND WE INSTALLED THE DIGITAL CERTIFICATE THAT WORKS SO THE PADLOCK IS ON.
BUT LET'S GO AND CHECK THE DETAILS.
IF YOU LOOK AT THE DETAILS, THE CERTIFICATE HAS BEEN ISSUED TO THE ENCODED FORM, WHICH WE DON'T RECOGNIZE. QUITE FRANKLY, THE NAME I TYPED IN AND THIS ARE DIFFERENT, SO I WOULDN'T TRUST. WOULD YOU?

SO CLEARLY THERE'S A PROBLEM IN THE SECURITY SPACE WITH IDN. NOW, IS IT REALLY WORTHWHILE TO LOOK AT IT AND MAKE IT WORK? WELL, I WOULD ARGUE THAT IT IS WORTH OUR WHILE. LOOKING AT SOME FACTS, 92% OF THE WORLD SPEAKS A PRIMARY LANGUAGE OTHER THAN ENGLISH. 48% OF INTERNET USERS ARE NON-NATIVE ENGLISH SPEAKERS.
IN 2003, NON-ENGLISH SPEAKERS REPRESENTED TWO-THIRDS OF ALL INTERNET USERS, AND A PROJECTION THAT BY 2007, CHINESE WILL BE THE NUMBER ONE WEB LANGUAGE.
SO DO WE NEED IDN? WELL, I THINK WE DO.
AND LOOK AT THE TRUST FACTOR AGAIN. WE SEE IT'S A TRUST THING. IT'S REALLY ALL ABOUT TRUST. IF YOU'RE IN A FOREIGN LANGUAGE ENVIRONMENT, YOU TRAVEL, YOU CERTAINLY DON'T UNDERSTAND AND CAN'T READ EVERYTHING.
NOW, IF YOU ARE AT HOME IN YOUR ARM CHAIR SURFING THE INTERNET, YOU ARE WITH IDN IN A FOREIGN LANGUAGE ENVIRONMENT. YOU NEED TO TRUST. AND YOU CAN TRUST USING SECURITY AND YOU CAN TRUST USING DIGITAL CERTIFICATES.
AND A HOT SUBJECT TODAY IS DEFINITELY PHISHING. I HOPE EVERYONE KNOWS WHAT PHISHING IS, USING A CLEVER ENGINEERING TO GET PRIVACY DATA. AND WHAT THESE GUYS DO IS THEY SET UP WEB SITES WITH A SIMILAR URL, A SIMILAR DOMAIN NAME TO A LEGITIMATE SITE, BE IT A BANK, BE IT GOVERNMENT, AND TRY TO CAPTURE USER DATA.
NOW, AGAIN, WE NEED TO TRUST. THE DIGITAL CERTIFICATE IS A FANTASTIC WAY TO ESTABLISH THE TRUST AND THE CERTIFICATE IS THERE TO MAKE SURE -- WE HAVE TECHNOLOGY TODAY, DON'T WE?
SO DO WE NEED TO SECURE IDN? I THINK WE DO.
HAVING SPOKEN TO A COUPLE OF FOLKS IN THE PASSAGE, THEY ALREADY MENTIONED THAT IDN IS AN EARLY ADOPTER MARKET. PROCESSES AREN'T NECESSARILY SUPPORTED, NOT EVERYONE KNOWS ABOUT IT. NONETHELESS, WE CAN SEE THAT THERE ARE ROUGHLY ONE MILLION DELEGATIONS WORLDWIDE TODAY. A MILLION IDN DOMAINS ACTIVE ON THE INTERNET. THAT'S QUITE A NUMBER. SOME STUDIES HAVE SHOWN THAT ABOUT TWO TO THREE PERCENT OF ALL INTERNET DOMAIN NAMES ARE SECURED WITH SSL CERTIFICATES, SO TWO PERCENT OF A MILLION IS A GOOD NUMBER. I THINK IT'S A GOOD NUMBER.
WHY DO ORGANIZATIONS USE IDN? THE ONE FACTOR WHICH HAS BEEN DISCUSSED A LOT IS THE LANGUAGE, NATIVE LANGUAGE AND UNDERSTANDING THE LANGUAGE AND READING IT. BUT THERE'S ANOTHER VERY IMPORTANT FACTOR WHICH IS INTELLECTUAL PROPERTY.
THINK OF -- THINK OF COCA-COLA. IT DOESN'T REALLY SOUND IDN BUT IT'S A VERY GOOD EXAMPLE OF A NAME BEING A TRADEMARK AND BEING USED AS A TRADEMARK ON THE INTERNET AS A DOMAIN NAME.
THINK OF CYBERSQUATTERS. THEY ARE TRYING TO GET COCA-COLA.COM.
WELL, IN LANGUAGES THAT ARE NOT USING THE ENGLISH OR LATIN ALPHABET, THERE ARE A LOT OF BRAND NAMES AND THE COMPANIES VERY MUCH WANT TO KEEP THOSE BRAND NAMES AND TRADEMARK THEM. AND REPRESENT THEM ON THE INTERNET. IN ORDER TO DO THAT THEY NEED TO REGISTER IDNS AND IN ORDER TO USE THEM AND MAKE PEOPLE TRUST, THEY NEED CERTIFICATES.
SO I THINK EVERYONE AGREES WITH ME THAT SECURITY IS A GOOD IDEA. DO I SEE A YES?
OKAY. SOUNDS GOOD.
RIGHT. SO WE HAVE LOOKED AT THE WHOLE PROBLEM AND WE DECIDED LET'S GO FOR IT. LET'S SECURE IDN. WHAT IS REALLY THE CORNER CASES HERE? WHAT ARE THE BASICS THAT WE CAN BUILD ON, BUILD THE SECURE IDN ON?
WELL, FIRST OF ALL, VERY OBVIOUS, THE USER INSTINCTIVELY TRUSTS THE NATIVE LANGUAGE IDN. THAT'S ONE FACT.
NETWORK TRANSACTIONS, HOWEVER, NAME RESOLUTION, ARE DONE IN THE PUNYCODE FORM. THAT'S ANOTHER FACT.
THOSE ARE THE BAD NEWS, REALLY. THE GOOD NEWS IS THE CERTIFICATE DEFINED BY THE X 509 STANDARD, THEY DO AND WE CAN ALSO PUT DOMAIN NAMES INTO X 509 CERTIFICATES. SO WE CAN DEAL WITH THE MULTIPLE FORM.
WE SET OURSELVES A NUMBER OF DESIGN GUIDELINES. NUMBER ONE IS GET TRUST. GET PEOPLE TRUST THE IDN, THE SECURE IDN. AND WE CAN DO THAT BY MAXIMIZING THE REPRESENTATION OF THE NATIVE LANGUAGE NAME, AND MINIMIZING THE MACHINE READABLE OR PUNYCODE ENCODED FORM. AND OF COURSE WE NEED TO PROVIDE SUFFICIENT PUNYCODE, IF YOU WILL, TO MAKE THE TRANSACTION WORK.
AND A VERY IMPORTANT POINT THAT SECURE IDN NEEDS TO WORK IN THE CURRENT EXISTING BROWSER ENVIRONMENT. IT SHOULD NOT RELY ON ADDITIONAL SOFTWARE, IT SHOULD NOT RELY ON ADDITIONAL PLUG-IN, BECAUSE IF IT DOES, THEN IT WON'T WORK FOR EVERYONE IMMEDIATELY.
NOW, HAVING SAID THAT, TODAY THE CURRENT BROWSER ENVIRONMENT DOES REQUIRE ADDITIONAL SOFTWARE. AND I'M PRETTY SURE THERE'S GOING TO BE LOTS OF DISCUSSION ABOUT THAT, WHO SUPPORTS WHAT, THE MOZILLA FOLKS AND (INAUDIBLE) FOLKS AND THE MARKET LEADER.
SO I DON'T GO INTO THAT, BUT THE POINT IS THAT ONCE IDN WORKS IN YOUR BROWSER, SECURE IDN WILL WORK, TOO.
SO LET'S LOOK AT AN EXAMPLE, BECAUSE THE GOOD NEWS IS THAT WE MADE SECURE IDN WORK. IS IT GOOD NEWS?
SO LET'S LOOK AT THIS SITE WE VISITED. THIS IS A TEST BED, OBVIOUSLY, USED AN IDN DOMAIN TO GET TO A SITE. SSL CERTIFICATE WITH OUR NEW IDN CAPABLE CERTIFICATE INSTALLED. PADLOCK IS DISPLAYED, WHICH IS GOOD NEWS. THE FIRST STEP OF THE TRUST HAS BEEN ESTABLISHED. AND LET'S DO OUR DOUBLE CLICK ON THE PADLOCK, AND LOOK AT THE DETAIL.
DETAIL SHOWS THAT THE CERTIFICATE HAS BEEN ISSUED TO THE IDN DOMAIN NAME, WHICH IS THE SAME WHAT WE HAVE ENTERED, SO DO WE HAVE TRUST? I THINK WE DO HAVE TRUST. AND WE TOOK IT A STEP FURTHER, AND BECAUSE THERE ARE PEOPLE OUT THERE WHO ARE VERY INTERESTED IN SECURITY, AND REALLY WANT TO KNOW WHAT THEY'RE DEALING WITH AND WHO THEY'RE DEALING WITH. SO LET'S LOOK AT THE CERTIFICATE DETAILS. AND THE INTERESTING PART IS THE SUBJECT NAME, WHICH REALLY CONTAINS THE EMBEDDED DOMAIN NAME, AND WE STILL HAVE THE NATIVE LANGUAGE FORM IN THERE. AND WE ALSO ADDED A LITTLE NOTE THAT YOU'RE LOOKING AT AN IDN CERTIFICATE, SO THAT THE USER CAN BE EDUCATED WHAT IT IS AND CAN LEARN MORE ABOUT IT.
SO I HOPE YOU LIKE THIS NEWS. WE HAVE GOT SECURE IDN WORKING, AND IF YOU WOULD LIKE TO FIND OUT MORE OR GET ONE, VISIT WWW.THAWTE.COM/IDN AND YOU WILL FIND FULLY INTERNATIONALIZED SSL CERTIFICATE ENROLLMENTS, CODE SIGNING CERTIFICATES.
THANK YOU THANKS VERY MUCH, AND IN CASE OF ANY QUESTIONS WE WILL HAVE TIME FOR THAT LATER AND WE ALSO HAVE A STAND AND YOU'RE MORE THAN WELCOME TO VISIT US THERE AND ASK ME OR THE FOLKS WHO ARE THERE TO HELP. THANK YOU.
>>MOUHAMET DIOP: THANK YOU VERY MUCH VALENTIN.
I'M GOING TO CALL THE LAST SPEAKER, MOEGSHIEN JABAAR, FROM IZIKO MUSEUMS OF CAPE TOWN. I'LL CALL UP, AFTER HE HAS FINISHED, ALL THE PREVIOUS SPEAKERS TO BE AVAILABLE FOR 15 MINUTES EACH, AND MAYBE WE'LL GIVE THE FLOOR AND PEOPLE WILL ASK QUESTIONS ABOUT THAT.
SO ALL THE SPEAKERS PLEASE MAKE YOURSELF AVAILABLE FOR THE NEXT COMING MINUTES.
>>MOEGSHIEN JABAAR: GOOD AFTERNOON, LADIES AND GENTLEMEN. I WAS ASKED TO DO THE PRESENTATION ON IZIKO JUST TO COVER THE BASIS OF HOW DOES MUSEUMS FIT IN THE FRAMEWORK OF IDN, AND I'VE DONE SOME RESEARCH JUST TO LOOK AT LANGUAGES THAT WE'VE GOT IN OUR COLLECTION, AND I CAME ACROSS SOMETHING INTERESTING AND I WOULD LIKE TO SHARE THAT WITH YOU THIS AFTERNOON.
JUST TO INTRODUCE MYSELF, THE NAME IS MOEGSHIEN JABAAR, AND WITH THE ASSISTANCE ON THIS PRESENTATION WAS DONE BY PROFESSOR BRELEKAMP. HE COULDN'T MAKE IT THIS AFTERNOON SO HE ASKED ME TO STAND IN FOR HIM. THE AGENDA WILL COVER WE'LL COVER THE IZIKO MUSEUM MISSION STATEMENT AND I'LL TALK ABOUT THE BOKAAP MUSEUM AND HOW WE ARE SITUATED, I'LL TALK ABOUT ARABIC DOCUMENTS WE'VE GOT IN OUR COLLECTION AND THEN I'LL BRIEFLY SPEAK ABOUT OUR INTRANET, OUR KIOSK THAT WE'VE GOT IN ONE OF THE MUSEUMS AND THE INTERNET FACILITIES THAT WE DO PROVIDE. AND THEN I WOULD LIKE TO SPEAK ABOUT THE WAY FORWARD. FOR IDSN.
THE MISSION STATEMENT, IZIKO MUSEUMS OF CAPE TOWN ARE AFRICAN MUSEUMS OF EXCELLENCE THAT EMPOWER AND INSPIRE ALL PEOPLE TO CELEBRATE AND RESPECT OUR DIVERSE HERITAGE. WE'VE SEEN THE DIVERSE HERITAGE THIS AFTERNOON IN QUITE A FEW SPEAKERS, AND THAT IS SOMETHING THAT WE WANT TO DO AS WELL AS A MUSEUM.
THE DIVERSE HERITAGE WITHIN CAPE TOWN ITSELF IN TERMS OF LANGUAGES, WE'VE GOT 11 OFFICIAL LANGUAGES, AND THAT CREATES IN ITSELF A HUGE CHALLENGE FOR ALL OF US. IF WE LOOK AT THE MISSION, TO MANAGE AND PROMOTE IZIKO'S UNIQUE COMBINATION OF SOUTH AFRICAN HERITAGE COLLECTIONS, SITES AND SERVICES FOR THE BENEFIT OF THE PRESENT AND FUTURE GENERATIONS.
THE SUM OF OUR COLLECTIONS THAT SERVES ALL OF US THAT SITS HERE, YOU WILL SEE IT IN A DIFFERENT LIGHT ONCE YOU HAVE SEEN THE PRESENTATION FURTHER ON.
THIS IS THE BOKAAP MUSEUM, IT'S SITUATED UP WELL STREET, AND THIS IS PART OF 15 MUSEUMS OF IZIKO. MORE FAMILIAR TO YOU WOULD BE THE CASTLE BECAUSE THERE'S A PRESENTATION HAPPENING THERE TOMORROW NIGHT. BUT THIS MUSEUM IS SITUATED IN THE CENTER OF A HUGE ISLAMIC COMMUNITY, AND THIS COMMUNITY, THEY HAVE DEALT WITH ARABIC SCRIPT IN A DIFFERENT FORM, AND THAT ARABIC SCRIPT I WILL EXPLAIN TO YOU AFTER THE SLIDE.
THE ARABIC THAT THEY HAVE DEALT WITH IN TERMS OF SLAVERY, THEY CAME ACROSS THE SLAVES, AND A LOT OF THEM HAD A KNOWLEDGE OF ARABIC BUT THE LANGUAGE WAS AFRIKAANS. THEY COULDN'T RIGHT IN THE LANGUAGE THEY SPEAK TODAY.
THIS IS AN EXAMPLE I HAVE EXTRACTED FROM OUR COLLECTION. THE FIRST FEW LINES, ACTUALLY THE SCRIBES FORMATTED IT ON A FRIDAY FOR THOSE ISLAMIC PEOPLE THAT IS MORE FAMILIAR AND THAT'S SITTING AROUND IN THE AUDIENCE, THEY WOULD BE VERY FAMILIAR.
NOW, IF YOU LOOK AT THE ARABIC SCRIPT, YOU WILL HAVE TO BEAR WITH ME, I'M NOT A GOOD READER OF ARABIC BUT I'LL TRY TO READ JUST THE FIRST FEW LINES.
(SPEAKING ARABIC.)
BASICALLY WHAT IT'S SAYING IS THAT THE PRAYER OF FRIDAY, THERE IS ANOTHER PRAY CONNECTED TO THAT. THIS IS A DOCUMENT WRITTEN WAY BACK IN THE SLAVERY TIME AND THIS DOCUMENT IS HIDDEN IN OUR COLLECTION BECAUSE WE CANNOT EXPOSE IT TO THE PUBLIC. WE CANNOT LAY THIS DOCUMENT OUT TO THE REST OF THE WORLD BECAUSE WE ARE FORCED TO EXTEND OUR KNOWLEDGE ONLY IN THE ENGLISH LANGUAGE ON THE INTERNET AND THAT IS SOMETHING WE WANT TO LOOK AT IN THE NEAR FUTURE. AND THIS IDSN SEEMS TO BE THE SOLUTION TO THIS PROBLEM. I'VE SEEN OTHER SPEAKERS EXPLAIN HOW ARABIC LANGUAGE AND ARABIC SCRIPT CAN BE EXTENDED TO THE REST OF THE WORLD. AND I APPLAUD THEM FOR THE EFFORTS.
IF WE LOOK AT OUR FACILITIES, IZIKO'S POLICIES NEEDS TO BE EXPOUNDED IN 11 LANGUAGES. AT THE MOMENT, BECAUSE OF FINANCIAL CONSTRAINTS WITHIN SOUTH AFRICA, BUDGETARY CONSTRAINTS, WE CANNOT DELIVER ALL THESE LANGUAGES ON OUR INTERNETS, ON OUR LOCAL-AREA NETWORKS, SO BASICALLY WE ARE FORCED TO STICK TO ONE LANGUAGE, WHICH IS THE ENGLISH LANGUAGE. THE KOSA LANGUAGE WHICH IS PREDOMINANTLY SPOKEN IN OTHER AREAS OF AFRICA AND SOUTH AFRICA, WE CANNOT SUPPORT THESE PEOPLE. AND IN TERMS OF LEGISLATION, WE HAVE TO COMMUNICATE IN THE LANGUAGE PREFERRED BY THE INDIVIDUAL, AND THAT IS SOMETHING THAT WE CANNOT SUPPORT AT THIS STAGE.
WE'RE LOOKING -- JUST TO SPEAK ABOUT THE IBM KIOSK, WE'VE GOT IT SITUATED AT THE NATIONAL GALLERY ART MUSEUM THAT IS IN THE GARDENS. NOW, WHAT IS THE RUSSIANS HAVE GONE, IS IBM IN CONNECTION OR IN RELATION WITH THE HERMITAGE MUSEUM, THEY'VE GIVEN US A KIOSK WITH THE DIFFERENT LANGUAGES EXPOSED IN THIS KIOSK. PEOPLE CAN SELECT THE DIFFERENT LANGUAGES. MORE RECENTLY, THEY'VE ALSO GONE INTO THE AREA WHERE THEY LOOK AT THOSE LANGUAGES AND GIVE PEOPLE THE OPPORTUNITY TO SELECT THE LANGUAGE WHICH IS MORE PREFERRED TO THEM, AND THIS IS SOMETHING THAT DEFINITELY WE NEED TO TALK ABOUT.
IF WE LOOK AT THE EDUCATION, WE HAD AN INCIDENT AT OUR MUSEUMS WHERE WE DELIVERED A TRAINING SESSION ON HOW TO SEARCH THE WEB. THERE WAS AN INDIVIDUAL WHO COULD ONLY SPEAK KOSA. SHE WAS AN EDUCATOR AT THE SCHOOL, BUT SHE WAS TEACHING KOSA AS A LANGUAGE. AND THIS PERSON WAS SITTING THERE EVENTUALLY NOT INTERESTED IN THIS PARTICULAR SUBJECT, BUT LATER ON, BECAUSE HE COULD COMMUNICATE TO COLLEAGUES THAT WAS ALSO SEARCHING THE INTERNET AT THAT STAGE, SHE STARTED CRYING BECAUSE SHE SAW THE BENEFITS OF THE INTERNET AND HOW SHE CAN ACTUALLY USE IT TO EDUCATE THE PEOPLE WITHIN HER AREA.
AND THIS IS SOMETHING WE NEED TO THINK ABOUT AS WELL. ACCESS.
ARE WE GOING THOROUGH ACCESS TO OUR PEOPLE? ARE WE DEPRIVING OUR VERY COMMUNITY THE ACCESS THAT THEY SHOULD BE GIVEN?
WHAT IS OUR WAY FORWARD?
WE DEFINITELY, IZIKO NEEDS TO DEVELOP OUR INTERNAL INTRANET, AND THAT IS SOMETHING THAT WE'VE SORT OF ENGAGED IN AT THE LOCAL LEVEL. BUT BECAUSE OF FINANCIAL CONSTRAINTS, WE HAD TO CUT BACK AND JUST FOCUS ON THE ENGLISH LANGUAGE. IF WE LOOK AT COMMUNITY DEVELOPMENT, YES, DEFINITELY, WE NEED TO GO OUT TO COMMUNITIES TO EXPRESS WHAT WE ARE DISCUSSING NOW TODAY IN TERMS OF IDSN AND HOW WE CAN BENEFIT PEOPLE IN THE FUTURE.
EMBRACING ALL OUR CULTURAL ACTIVITIES, WE'VE SEEN PREVIOUSLY THE CULTURAL ACTIVITIES THAT HAPPENED DURING THIS PERIOD OF SLAVERY. SHOULDN'T WE EXPLAIN THIS TO OUR CHILDREN AND TO THE FUTURE THAT'S COMING. WE NEED TO INFORM THEM OF THE STRUGGLES HERITAGE, NOT JUST OF SOUTH AFRICA. OF AFRICA AS A WHOLE, AND AFRICA TRULY HAS A GREAT HERITAGE TO SPEAK ABOUT.
REPRESENTATION -- REPRESENTING THE LANGUAGES IN WHICH IT IS EXPRESSED, THAT IS CRITICAL FOR US, THAT WE STAY IN THE LANGUAGE AND WE COMMUNICATE IN THE LANGUAGE THAT IS HOME TO THE PEOPLE. NOT HOME TO THOSE THAT WRITES THE LANGUAGES ON THE INTERNET.
WE NEED TO TAKE THE MESSAGE HOME AND WE NEED TO KEEP IT AT HOME. AND I THINK THAT WAS EXPRESSED BY THE PREVIOUS SPEAKER AS WELL.
TO BUILD REPOSITORIES AND LOCALIZE OUR CONTENT. WE'VE TRIED TO A HUGE EXTENT AS THE MUSEUM TO DO THAT FOR THE COMMUNITY. WE ARE PROUD TO SAY THAT WE'VE GOT A HUGE COLLECTION, A SOCIALISTIC COLLECTION, AN NATURAL COLLECTION, AN ART COLLECTION, AND THESE ARE IN LANGUAGES THAT CAN BE MADE AVAILABLE, BUT THEIR AVAILABILITY IS DENIED TO SOME EXTENT.
IDENTIFY EACH DOCUMENT WITH ITS OWN LANGUAGE. THAT IS CRITICAL. WE'VE SEEN THE ARABIC LANGUAGE IN AFRIKAANS SCRIPT. NOW, I'M SURE THERE'S VERY FEW PEOPLE IN THIS AUDIENCE WHO HAVE SEEN A DOCUMENT LIKE THAT, AND THAT IS THE TYPE OF DOCUMENT THAT NEEDS TO BE GIVEN TO EVERYBODY IN THEIR OWN LANGUAGE; NOT IN A LANGUAGE THAT REDUCES THE QUALITY OF THE INFORMATION. THE QUALITY NEEDS TO BE EXTENDED IN TERMS OF THE LANGUAGE THAT IT WAS CREATED IN.
IDN IS REQUIRED. THE REQUIREMENT IS TO DO ALL OF THIS. I WOULD LIKE TO SPEAK TO THOSE THAT ARE CAPABLE OF SETTING UP THIS FOR US, AND TO ASK THEM THAT THEY DO GIVE US THE OPPORTUNITY TO GO FORWARD WITH THIS.
I THANK YOU.
(APPLAUSE.)
>>MOUHAMET DIOP: SO I CALL FOR THE PREVIOUS SPEAKER TO COME AROUND THE TABLE FOR A COUPLE OF 15 MINUTES OF EXCHANGE.
SO PIERRE OUEDRAOGO, MAXIME SOMÉ, CAN YOU JUST JOIN THE FLOOR.
AND TAKE A SEAT, PLEASE.
PRESIDENT ADAMA.
SO NOW I CALL FOR 15 MINUTES OF QUESTIONS, AND EVEN PEOPLE WHO WANT TO INTERVENE IN ORDER TO ADD SOME COMMENTS.
MAYBE I WILL TAKE ALSO ON MY PRESIDENCY POWER TO CALL MAYBE FOR SOME COMMENT, BECAUSE WE HAVE A CHANCE TO GET SOME REALLY KNOWLEDGEABLE PEOPLE WHO ARE ALREADY INVOLVED IN THE IDN ISSUE FOR MANY YEARS.
AND ESPECIALLY JOHN KLENSIN AND THE OTHER PEOPLE HERE.
OKAY?

>> GOOD EVENING.
MY NAME IS STEPHANE BRUNO.
I REPRESENT (INAUDIBLE) FROM HAITI.
I HAVE TWO QUESTIONS, ONE FOR PIERRE OUEDRAOGO AND ONE FOR --
TALKING ABOUT THE IDN, SOMETIMES I -- IT SEEMS THAT WE CONFUSE THE ISSUE OF USING NATIVE LANGUAGE TO TYPE DOMAIN NAME ADDRESSES, FOR EXAMPLE, TO FIND A WEB SITE.
WE SEEM SOMETIMES TO CONFUSE THIS ISSUE WITH THE ISSUE OF CONTENT OF THE WEB SITE IN A NATIVE LANGUAGE.
IS IT BECAUSE THOSE TWO THINGS ARE RELATED OR IS THERE A MISTAKE?
BECAUSE I BELIEVE THAT EVEN WITH ASCII DOMAIN NAMES, WE COULD HAVE WEB SITES WITH CONTENT WRITTEN IN NATIVE LANGUAGES.
SO I WOULD LIKE TO HAVE SOME QUESTIONS ON THAT.
THE SECOND QUESTION I HAVE IS ABOUT THE UNICODE ORGANIZATION.
YOU SAID THAT DEPENDING ON THE AMOUNT OF MEMBERSHIP FEES THAT YOU PAY, YOU HAVE A CERTAIN AMOUNT OF VOTE.
HOW DO YOU PREVENT SOMEONE TO JUST BUY, YOU KNOW -- BUY THE ORGANIZATION TO MAKE HIS VOTE PASS, FOR EXAMPLE?
>>JOHN KLENSIN: SINCE YOU ASKED ME TO COME UP AND SAY SOMETHING, LET ME ADDRESS THE FIRST QUESTION THAT WAS JUST ASKED.
YOU'RE QUITE CORRECT, THE ISSUE OF REPRESENTATION OF CHARACTERS ON A WEB PAGE AND TYPING INTO IT AND SENDING A NATIVE CHARACTER E-MAIL IS QUITE SEPARATE FROM THE DOMAIN NAMES ASSOCIATED WITH THAT MATERIAL.
WE'VE HAD A FAIRLY DECENT, IF NOT PERFECT, CAPABILITIES OF DOING THAT IN BOTH E-MAIL AND THE WEB FOR ABOUT TEN YEARS.
AND ITS ADVANTAGE OVER THE DNS IS THAT WE'VE GOT ENOUGH INFORMATION IN AN E-MAIL MESSAGE OR A WEB PAGE THAT WE CAN ACTUALLY SPECIFY THE CHARACTER SET THAT'S BEING USED.
SO, FOR EXAMPLE, GETTING TO SOME OF THE COMMENTS WHICH WERE MADE IN SOME OF THE TALKS TODAY, IF WE'VE GOT A SITUATION IN WHICH UNICODE IS NOT QUITE RIGHT BUT THERE IS A NATIONAL CHARACTER SET OR A LOCAL CHARACTER CODING SYSTEM THAT IS WIDELY USED BETWEEN GROUPS WHO ARE COMMUNICATING WITH EACH OTHER IN A PARTICULAR LANGUAGE, IT'S QUITE EASY WITH E-MAIL OR A WEB PAGE TO ACCOMMODATE THAT CHARACTER SET, EVEN THOUGH IT MIGHT NOT BE UNIVERSALLY READABLE.
ITS NOT BEING UNIVERSALLY READABLE IS ONE OF THOSE INTERESTING ISSUES FOR US, BECAUSE IF I CAN'T READ THE LANGUAGE, WHETHER OR NOT I CAN RENDER THE CHARACTER SET IS PROBABLY NOT VERY IMPORTANT TO ME.
SO THE FIRST ANSWER IS, I HEARD THAT DISTINCTION IS PRESENT AND IT'S IMPORTANT.
AND I HEARD THAT DISTINCTION BEING MADE IN SEVERAL OF THE CONVERSATIONS TODAY.
AND I WOULD ENCOURAGE PEOPLE WHO ARE WORRIED ABOUT LANGUAGES AND PRESENTATION AND MAKING INFORMATION CULTURALLY AVAILABLE TO CONCENTRATE ON THAT PROBLEM OF CONTENT.
BECAUSE WHILE THE IDN ISSUE IS VERY IMPORTANT, IT'S ULTIMATELY THE SECONDARY ONE.
IF YOU CAN'T GET THE CONTENT IN A FORM IN WHICH PEOPLE CAN READ IT, THEN HAVING IDNS IN THE APPROPRIATE SCRIPT IS REALLY OF VERY SECONDARY IMPORTANCE.
GENERAL COMMENT, I AM DELIGHTED BY THE PRESENTATIONS TODAY.
I AM DELIGHTED THAT THIS WONDERFUL WORKSHOP IS GOING FORWARD WITHOUT MY HAVING TO DO ANYTHING.
AND I THINK I CAN LOOK FORWARD TO RETIRING FROM THIS BUSINESS.
THANK YOU ALL.

>>MOUHAMET DIOP: I DON'T THINK THAT THAT MESSAGE HAS BEEN TAKEN WELL.
WE NEED YOU, AND YOU'RE STILL HERE WITH US FOR MANY YEARS AGAIN, AND BECAUSE WE DON'T HAVE THE CHANCE TO GET MANY PEOPLE WHO HAVE GONE THROUGH DECADES AND DECADES OF THE HISTORY AND FOR THE REPRESENTATION OF THIS CHARACTER SET.
>>MICHEL SUIGNARD: MY NAME IS MICHEL SUIGNARD. I AM WITH MICROSOFT.
I WILL ANSWER THE QUESTION OF THE UNICODE FEE STRUCTURE.
LARGE MEMBERS DON'T REALLY GET MORE VOTES.
ONLY GET ONE.
HOW BIG DOESN'T MATTER, BECAUSE YOU ONLY GET ONE VOTE.
IT'S TRUE THAT YOU HAVE TO BE A FULL MEMBER TO GET A FULL VOTE.
BUT IT DOESN'T MATTER HOW BIG YOU ARE FROM THAT POINT.
THERE'S BASICALLY A LOW THRESHOLD.
SO ALL OF THE 15 OR 16 MEMBERS TODAY HAVE THE SAME WEIGHT, AND SOME ARE VERY BIG AND SOME ARE PRETTY SMALL.
SO IT DOESN'T MATTER.
YOU CAN'T BUY UP ABOVE THE SINGLE MEMBER LEVEL.
>>MOUHAMET DIOP: PIERRE, CAN YOU GIVE COMMENT ON THE FIRST?
>>PIERRE OUEDRAOGO: I WANT TO SAY THANK YOU TO MICHEL FOR HELPING THE NEWCOMER.
I THINK THAT THE QUESTION OF (INAUDIBLE) MEANS MAYBE THAT I DIDN'T HAVE ENOUGH TIME TO EXPLAIN EVERYTHING.
IT IS NOT ONLY MONEY.
AND FOR EVERY MONEY ASPECT, I THINK THE ANSWER OF MICHEL IS VERY CLEAR.
YOU HAVE ONE VOTE.
AND, IN FACT, IF YOU GO ON THE WEB SITE RIGHT NOW, YOU WILL SEE THAT THE OFFICIAL VOTE IS $12,000 U.S. FOR A FULL MEMBER AND 6,000 FOR A SMALL BUSINESS.
FOR THE ASSOCIATE, IT IS 2,000 FOR PROFIT AND 1,200 FOR NONPROFIT, AND $120 FOR INDIVIDUALS.
NOW, THIS IS WHAT -- I THINK I HAVE THE RIGHT FIGURE.
AND THIS IS NOT A LOT OF MONEY AT THE LEVEL OF THE INDUSTRY, YOU KNOW.
SO -- BUT, ANYWAY, IT SHOWS THE INTERESTS OF ORGANIZATIONS IN THE WORK OF THE UNICODE.
AND THIS INTEREST IS ALSO SHOWN THROUGH THE NUMBER OF ENGINEERS THEY PUT ON THE PROJECT TO HELP BUILD THE INTEROPERABILITY OF THE SYSTEMS, TO HELP MAINTAIN THE CHARACTER DATABASE.
AND THERE ARE MANY ENGINEERS AT MICROSOFT, IBM, ADOBE, WHO ARE WORKING.
SO THE 12,000 IS NOTHING COMPARED TO WHAT THESE COMPANIES ARE MAKING -- ARE DOING FOR THE UNICODE.
SO IT IS SOMETHING LIKE A VOLUNTEER ACTION.
AND I THINK THAT IN SOME OTHER FIELDS OF THE HUMAN ACTIVITY, IF THERE WAS SUCH COORDINATION, WE ARE GOING NOT TO BE BOTHERED BY SOME, YOU KNOW, LACK OF INTEROPERABILITY.
BUT HERE, WE HAVE A CHANCE THAT THE INDUSTRY LEADERS GET TOGETHER TO WORK TOGETHER AND TO HAVE EVERYTHING TO WORK, YOU KNOW, FINE.
I WOULDN'T SAY THE SAME THING FOR EVERY ASPECT OF THE INDUSTRY.
AND FOR THE NATIVE LANGUAGE, BRUNO AND KLENSIN ARE, I THINK, RIGHT.
BUT YOU SHOULD MENTION THAT IN THE CASE OF AFRICA, WE NEED TO NORMALIZE FIRST.
AND THIS IS NOT DONE YET.
AND THIS -- AND AGAIN AND AGAIN, WE WILL SAY THAT AFRICANS HAVE TO DO THEIR JOB.
THEY HAVE TO DO THEIR OWN WORK.
THEIR OWN WORK IS TO UPDATE THE CHARACTER SET.
THEIR OWN WORK IS TO BUILD UP A CHARACTER SET THAT IS SIMPLE AND ACCEPTED BY ALL AFRICAN PEOPLE WHO USE THE LATIN CHARACTERS TO WRITE THEIR LANGUAGES.
AND THIS IS SOMETHING WE CALL FOR, AND THIS IS SOMETHING THAT IS NECESSARY FOR THE CONTENT DEVELOPMENT.
IT IS ALSO IMPORTANT FOR IDN.
AND THIS IS WHY WE HAVE BROUGHT THIS ONE HERE.
OKAY, THANK YOU.

>>MOUHAMET DIOP: OKAY.
NEXT PERSON ON THE FLOOR.
>>JOHN KLENSIN: I JUST WANTED TO RESPOND A LITTLE BIT TO THAT SINCE MY NAME WAS MENTIONED.
I AGREE COMPLETELY.
THE CHARACTER SET UPDATING WORK IS VERY IMPORTANT.
AND NONE OF US FROM THE OUTSIDE CAN DO IT FOR YOU.
IF WE TRIED, WE WOULD MAKE A TERRIBLE MESS, I'M SURE.
WE DO HAVE AN ADDITIONAL DIFFICULTY WITH IDN, WHICH IS THAT AS A CONSEQUENCE OF SOME ISSUES ABOUT THE WAY IN WHICH THE UNICODE STANDARD IS UPDATED, IF WE COME IN AND ADD CHARACTERS NOW TO THE -- TO SOME FUTURE VERSION OF THE UNICODE STANDARD, IT ISN'T COMPLETELY CLEAR WHAT THE PROCESS IS GOING TO BE OF MAKING THAT ACCEPTABLE TO IDNS, BECAUSE THE IDN, IDNA STANDARD RIGHT NOW IS FROZEN AT UNICODE STANDARD 3.2.
I'M SURE IF ADDITIONAL WORK IS DONE THAT WE WILL FIGURE OUT HOW TO FIX THAT PROBLEM.
BUT IT'S NOT EASY AND STRAIGHTFORWARD.
BY CONTRAST, IN THE E-MAIL AND THE WEB PAGES AND THE CONTENT PRESENTATION, HANDLING NEW CHARACTERS WHEN THEY'RE DEFINED IS QUITE EASY.
SO, AGAIN, LET'S THINK ABOUT THE CONTENT ISSUES AND THE IDN ISSUES SEPARATELY.
THEY BOTH NEED TO BE SOLVED.
BUT THE CONTENT ISSUES ARE, FORTUNATELY, BOTH THE MORE IMPORTANT AND THE EASIER.
AND I LOOK FORWARD TO THE RESULTS OF YOUR WORK, AND I HOPE I CAN UNDERSTAND IT WHEN YOU PRODUCE IT.

>>MOUHAMET DIOP: THANK YOU, JOHN.
OKAY, NEXT SPEAKER.
>> VICTOR WILSON: GOOD AFTERNOON.
VICTOR WILSON.
IN THIS STANDARDIZATION OF CHARACTER SETS FOR LANGUAGES THAT PREVIOUSLY HAVEN'T BEEN WRITTEN, I WOULD HOPE THAT THOSE INVOLVED USE STANDARD 7-BIT REPRESENTABLE CHARACTERS WHERE POSSIBLE.
FOR EXAMPLE, IN THE 11 SOUTH AFRICAN LANGUAGES, I KNOW OF ONE CHARACTER USED, AN S WITH AN INVERTED CARET, WHICH I WONDER WHY ANYBODY WROTE THAT, BECAUSE IT'S PRONOUNCED JUST LIKE AN SH.
YOU CAN'T EXPECT THE GERMANS TO STOP USING UMLAUTS AND WRITE UE.
BUT IN NEWLY WRITTEN LANGUAGES, WE CAN AVOID USING SPECIAL CHARACTERS.

>>MOUHAMET DIOP: THANK YOU VERY MUCH.
IT'S REALLY IMPRESSIVE WHEN WE ARE TALKING ABOUT MULTILINGUALISM, WE ALWAYS MAKE THAT VERY EASILY WE JUMP FROM THESE TWO ISSUES, BECAUSE WHEN YOU ARE TALKING ABOUT IDN, IT'S A SPECIFICATION.
BUT THE PROBLEM OF THE CONTENT IS MUCH BROADER THAN WHAT WE'RE EXPECTING WHEN WE'RE TALKING ABOUT THE INTERACTION BETWEEN THAT ITEM AND ICANN.
IT'S -- THAT'S REALLY THE ISSUE NOW.
WHAT JOHN TRIED TO RAISE IS, WE NEED TO FIGURE OUT AT WHAT LEVEL WE ARE TALKING.
IF IT IS ONLY PURE IDN IMPLEMENTATION, WE NEED TO UNDERSTAND WHAT THE ISSUE WE ARE RAISING.
AND IF WE ARE TALKING ABOUT MULTILINGUAL.
SO IT'S -- IT'S MUCH BROADER.
JOHN, YES.
>>JOHN KLENSIN: I JUST WANTED TO RESPOND TO A COMMENT WHICH WAS MADE FROM THE FLOOR A MOMENT AGO.
IF YOU'LL INDULGE ME, THIS IS CLEARLY NOT AN ICANN PROBLEM.
BUT HOW WE'VE GOTTEN HERE WITH ALPHABETIC AND NEARLY ALPHABETIC LANGUAGES OVER A FEW THOUSAND YEARS OF HISTORY IS THAT SOMEWHERE ALONG THE LINE, SOMEBODY INVENTED A NOTION OF WHAT WE NOW CALL ALPHABETIC LANGUAGES WHICH USE THESE CHARACTER THINGS TO REPRESENT SOUNDS OR COMBINATIONS OF CHARACTERS TO REPRESENT SOUNDS.
AND THE RANGE OF SOUNDS IN DIFFERENT LANGUAGES HAS RESULTED IN ALPHABETS OR SCRIPTS WITH, TYPICALLY, SOMEWHERE AROUND 20 TO 50 CHARACTERS IN THEM.
BECAUSE MOST LANGUAGES HAVE SOMEWHERE BETWEEN 20 TO 50 SOUNDS IN THEM.
THE FUNNY MARKS AND CHARACTERS, ULTIMATELY, THE DIFFERENCE BETWEEN THE WAY ARABIC LOOKS AND THE WAY ROMAN CHARACTERS LOOK ARE ALL EVOLUTION OVER TIME AND ADAPTATIONS OF DIFFERENT CHARACTER AND SOUND SYMBOLS.
AND TO SAY LET'S NOT USE ANY OF THESE ACCENTED CHARACTERS OR DIACRITICAL CHARACTERS IS ULTIMATELY TO SAY LET'S NOT REPRESENT THAT LANGUAGE AND ITS SOUNDS CORRECTLY EXCEPT BY FUNNY CONVENTIONS AND COMBINATIONS, AND THOSE ARE TYPICALLY MORE PROBLEMATIC THAN SPECIAL CHARACTERS OR SPECIAL VARIANT CHARACTERS.
WE'RE STUCK WITH THIS STUFF BECAUSE OF THE WAY WE SPEAK, AND BECAUSE OF THE WAY WE LISTEN TO EACH OTHER.
AND THAT'S FAIRLY IMPORTANT CULTURAL STUFF.

>>MOUHAMET DIOP: I WANT TO ASK A QUESTION TO THE IRANIAN GROUP.
I WAS REALLY INTERESTED --
>> (INAUDIBLE).
>>MOUHAMET DIOP: YEAH, YOU ARE ONE OF THE SPEAKERS.
>>ROOZBEH POURNADER: SO I RELUCTANTLY AGREE WITH WHAT JOHN MENTIONED.
BUT THERE IS SOME POINT IN THE FIRST -- WHAT THE FIRST GENTLEMAN ASKING THE QUESTION MENTIONED.
THE POINT THERE REALLY IS -- I DON'T WANT TO MAKE HIS WHOLE POINT.
BUT THE THING IS THAT IF YOU'RE WRITING NEW LANGUAGES AND IF YOU'RE USING NEW CHARACTERS THAT NO OTHER LANGUAGE HAS ALREADY USED, NEW SHAPES FOR THAT, THIS IS NOT NECESSARILY A GOOD THING.
SO IF IT'S NOT REALLY THAT IMPORTANT FOR YOU TO USE AN "S" WITH A CERTAIN MARK OVER IT OR AN "S" WITH ANOTHER KIND OF MARK UNDER IT, THE "S" WITH THE MARK UNDER IT IS A NEW CHARACTER THAT NOBODY HAS EVER USED IN ANY OTHER KIND OF LANGUAGE, IT'S POSSIBLY A LITTLE HARDER FOR YOU TO GO AND ENCODE THAT INTO UNICODE.
IT'S POSSIBLY A LITTLE EASIER FOR YOU TO USE THE ALREADY-EXISTING CHARACTER WHICH IS USED IN ANOTHER LANGUAGE, POSSIBLY ANOTHER AFRICAN LANGUAGE OR EUROPEAN LANGUAGE OR SOMETHING LIKE THAT.
SO IT HELPS THE PROCESS.
OTHERWISE, YOU WILL HAVE SOMETHING LIKE AT LEAST TWO OR THREE YEARS BEFORE YOU WILL BE ABLE TO USE THAT CHARACTER IN STANDARD COMPUTER EQUIPMENT.

>>MOUHAMET DIOP: FIRST I WOULD ASK YOU TO JOIN THE PANEL, PLEASE.
PLEASE, I WOULD ASK YOU TO JOIN THE PANEL.
BECAUSE I THINK THAT THERE WERE SOME OTHER QUESTIONS COMING SOON FOR YOU.
>>JOHN KLENSIN: YEAH, FIRST OF ALL, AS WE SEE IN THE IETF, WE ARE IN VIOLENT AGREEMENT.
WHAT I AM ABOUT TO SAY DOES NOT DISAGREE WITH WHAT YOU SAID AT ALL.
AT THE SAME TIME, AND IT'S ALWAYS A BALANCE, THERE ARE TWO OTHER WAYS OF DEALING WITH THE PROBLEM, ONE OF WHICH IS THAT UNICODE HAS A NOTION OF COMBINING CHARACTERS.
AND IF YOU ABSOLUTELY HAVE TO HAVE A -- AN S-LIKE SYMBOL WITH A SQUIGGLE UNDERNEATH IT, YOU CAN PROBABLY FIGURE OUT A WAY TO DO THAT.
ALTHOUGH COMBINING CHARACTERS CAUSE THEIR OWN SETS OF PROBLEMS IN OTHER AREAS.
THE OTHER OBSERVATION IS THAT IF YOU REALLY HAVE A UNIQUE SOUND THERE, YOU MAY BE BETTER OFF WITH A UNIQUE CHARACTER THAN WITH A FUNNY SOUND.
AND THE THIRD PIECE OF THIS OBSERVATION IS, THIS IS ALL ABOUT CONVENTIONS WITHIN THE ALPHABETIC LANGUAGES.
WE HAVE A LARGE NUMBER OF LANGUAGES IN THE WORLD WHICH CAN BE WRITTEN EQUALLY CORRECTLY WITH MORE THAN ONE SCRIPT.
WE'VE GOT LANGUAGES WITH ARABIC REPRESENTATIONS AND ROMAN REPRESENTATIONS.
WE'VE GOT LANGUAGES WITH CYRILLIC REPRESENTATIONS AND ROMAN REPRESENTATIONS.
WE'VE GOT A FEW LANGUAGES IN WHICH WE HAVE ARABIC AND ROMAN AND CYRILLIC REPRESENTATIONS.
AND THE FACT THAT THESE DIFFERENT SCRIPTS WITH SMALL VARIATIONS IN EACH CASE CAN BE FLIPPED BACK AND FORTH TO REPRESENT THE SAME LANGUAGE IS, ON THE ONE HAND, ONE OF THE BEAUTIES OF THIS SITUATION; AND ON THE OTHER HAND, YET ANOTHER OPPORTUNITY FOR VERY SERIOUS CONFUSION IN AN IDN AREA THAT NOBODY HAS REALLY COMPLETELY SORTED OUT YET FOR ANY LANGUAGES THAT I KNOW OF.
WE HAVE A LOT OF CHALLENGES AHEAD OF US.
BUT WE'VE GOT REAL OPPORTUNITIES TO MAKE THIS STUFF CULTURALLY MORE RELEVANT AND ACCESSIBLE, AND WE NEED TO PUSH OUR WAY THROUGH THE CHALLENGES.

>>MOUHAMET DIOP: PRESIDENT.
>>ADAMA SAMASSEKOU: THANK YOU, MOUHAMET.
I WOULD LIKE TO SAY SOMETHING HERE.
I THINK THAT HERE THIS IS A TECHNICAL DEBATE WHICH MAY MAKE THE QUESTION DIFFICULT.
WE NEED TO SIMPLIFY THE PROBLEM AT THE LEVEL OF THE AFRICAN SITUATION, THE CHANCE THAT WE HAVE IN OUR LANGUAGES ARE NOT TALKING ABOUT THE LANGUAGES WHICH HAVE THE WRITTEN TRADITION IN AFRICA.
MOST OF THE LANGUAGES, AS SOMEONE RECALLED, WHO -- PIERRE RECALLED, IN -- SINCE 1928, WE HAD A NORMALIZED ALPHABET ON THE BASIS, BUT TODAY WE HAVE TO UPDATE IT LOOKING AT THE DEVELOPMENT.
BUT WHAT WE CAN SAY IS THAT THE CHANCE TO DEVELOP THE LANGUAGES HAVE -- THOSE WHICH DON'T HAVE THEIR OWN TRADITION IN AFRICA, WE HAVE TO REALLY SIMPLIFY THE SOUNDS OF THE LANGUAGES.
WRITING IS JUST (INAUDIBLE).
SO WE HAVE TO LOOK AT THE REPRESENTATION, WHICH A MAJORITY OF SOUTH AFRICANS, THEY HAVE BEFORE 1978, WHICH WAS FORMALIZED IN 1978.
AND I WOULD LIKE TO SAY THAT THE QUESTION WHICH IS IMPORTANT TODAY FOR US IS -- BECAUSE IT DEPENDS ON US -- IT'S AROUND THE AFRICAN ACADEMY OF LANGUAGES, WE HAVE TO MEET BETWEEN LINGUISTS AND THE COMPUTER SCIENTISTS AND THE PEOPLE WHO ARE WORKING ON THE CHARACTER OF THE LANGUAGE, BECAUSE THEN HERE WE WILL BE ABLE TO AVOID CONFUSION IN THE WRITING, BECAUSE IT'S A LANGUAGE SPOKEN, WE HAVE TO TRANSCRIBE THE GRAMMAR OF THE LANGUAGE.
I THINK THIS WORK WILL BE DONE VERY QUICKLY, AND THERE WON'T BE ANY PROBLEM AT ALL.
THE QUESTION WHICH IS DIFFICULT IS THE QUARREL OF LINGUISTS THAT WE TALKED ABOUT HERE.
AND I AGREE HERE, IT'S ABOUT 20 TO 30 YEARS WE HAVE BEEN TALKING ABOUT THIS, AND LINGUISTS HAVEN'T BEEN ABLE TO AGREE.
AND WE HAVE TO SIMPLIFY THE WHOLE THING.
EVERYTHING IS POSSIBLE TODAY.
WE HAVE TO ORGANIZE AND TO FOLLOW SOME METHODS.
AND WITH THE ACADEMY TOMORROW WE COULD HAVE A SHARED VISION.
>>MOUHAMET DIOP: I WILL TAKE THE LAST -- WE'LL GET TWO COMMENTS AND THEN WE'RE RUNNING OUT OF TIME.
BECAUSE WE WERE SUPPOSED TO STOP FOR THE BREAK AROUND 4:15.
SO I WILL TAKE THE LAST QUESTION FROM SEBASTIEN.
>>SEBASTIEN BACHOLLET: I'LL HAVE TO SPEAK FRENCH.
IT'S INTERESTING, THE PRESENTATION THAT WE HAVE HAD THIS AFTERNOON.
AND WE'VE GOT TO REALLY VALUE BOTH SUBJECTS, BOTH IN CONTENT AND THE DISCUSSION ON THE DOMAIN.
ON THE CONTENT, WE CAN USE IT ON DIFFERENT LANGUAGES, AND I COULD SAY IT WILL APPLY ON ALL THE LANGUAGES EXCEPT ENGLISH.
HOW ARE WE GOING TO DEFEND OUR CULTURES AND OUR LANGUAGES, ON THE DIFFERENCE OF ENGLISH.
DO WE HAVE TO DEFEND THIS IN TERMS OF CONTENT AND CULTURE?
THE QUESTION OF DOMAIN IS A QUESTION THAT HAS COME A BIT LATE.
I DIDN'T UNDERSTAND ANYTHING, THE TECHNICIANS WHO ARE HERE WHO HAVE GOT TO SOLVE TECHNICAL ISSUES HAVE GOT SOMETHING TO SAY ABOUT IT.
PERHAPS THEY ASK THE QUESTION DIFFERENTLY.
WE NEED TO GO -- TO ALLOW SOMEONE TO HAVE SOME INFORMATION AND COMMUNICATE WITH SOMEONE ELSE.
AND IMAGINE TWO SECONDS WE ARE ON THE WAY, AND TAKING THE EXAMPLE OF EUROPE, AND THE PANEL STOPS, AND YOU ARE -- IT'S DIFFERENT FROM ONE COUNTRY TO ANOTHER.
IN THE COUNTRY WHICH I GO ACROSS, WHAT WILL HAPPEN, I WILL HAVE TO -- I HAD TO LEARN THE CODES, YOU KNOW, OF EACH OF THESE COUNTRIES.
I THINK THAT THERE ARE PEOPLE WHO DRIVE ON THE LEFT, AND THERE ARE OTHERS ON THE RIGHT.
PERHAPS WE CAN HAVE -- ASSIGN PANELS THAT COULD BE -- YOU KNOW, GIVE US A THOUGHT.
WE HAVE TO HAVE THE MINIMUM ASSIGNED THAT WE SHOULD HAVE AT THE INTERNATIONAL LEVEL TO USE ALL THE CODES, ROAD SIGN TO COME FROM ENGLISH AND GO TO SOMETHING THAT WILL BE INTERNATIONALLY ACCEPTABLE TO AFFORD SOMEONE BEING -- YOU KNOW, DOING THINGS IN ISOLATION.
WE HAVE TO BE WORKING TOGETHER.
THANK YOU VERY MUCH.
>>MOUHAMET DIOP: THANK YOU, SEBASTIEN.
I THINK IT'S MUCH MORE CONTRIBUTION THAN A QUESTION.
I DON'T THINK IT AS A QUESTION.
NEXT SPEAKER.
MM-HMM.
>>EDMON CHUNG: I JUST WANT TO ADD A COMMENT THAT AGAIN, IT'S NOT VERY TIGHTLY RELATED WITH IDN.
BUT I WOULD DISAGREE WITH A GENTLEMAN THERE SAYING THAT TO AVOID CREATING NEW CHARACTERS, BECAUSE -- JUST FOR ONE PARTICULAR EXAMPLE, CHINESE CHARACTERS ALREADY ARE CONSUMING A LARGE AMOUNT OF CHARACTER POINTS, ARE STILL CREATING NEW CHARACTERS TODAY.
SO WE CAN'T ASSUME THAT AS WE GO FORWARD OR CREATE NEW, YOU KNOW, LANGUAGES.
>>MOUHAMET DIOP: THANK YOU VERY MUCH.
I THINK THAT THIS FIRST PART OF THE SESSION ON IDN WILL HAVE TWO OTHER SESSIONS AFTER THE BREAK.
THE NEXT ONE WILL BE ON IDN APPLICATION DEVELOPMENT, AND THE LAST ONE WILL BE ON IDN POLICY ISSUE AND PROCESSES.
SO I JUST WANT TO THANK ALL OF THESE SPEAKERS HERE.
THEY HAVE TAKEN THEIR TIME TRYING TO BE REALLY EXPLICIT.
AND I THANK ALSO ALL THE (INAUDIBLE) FOR THE FIRST PART ON IDN AND WE'LL HAVE A BREAK FOR 15 MINUTES AND THEN COME BACK.
THANK YOU.
SESSION ADJOURNED.
(APPLAUSE.)
(BREAK.)

SESSION II
PANEL DISCUSSION, IDN APPLICATION DEVELOPMENT

>>JAMES SENG: GOOD AFTERNOON, EVERYONE. WE'RE GOING TO START OUR SESSION SOON. COULD YOU PLEASE JUST SIT DOWN?
OKAY. WE ARE ACTUALLY BEHIND SCHEDULE, SO I WILL TRY, BUT UNFORTUNATELY WE DO HAVE A LOT OF PRESENTERS, AND EVERYONE SEEMS TO HAVE A LOT OF THINGS TO SAY TODAY.
WE HAD A VERY INTERESTING DISCUSSION THE LAST FEW DAYS SO I'LL TRY MY BEST TO CATCH UP WITH TIME BUT IT'S NOT LIKELY TO DO SO. WE'LL BE RUNNING A BIT BEHIND SCHEDULE.
SO INSTEAD OF TRYING TO GET THE FLOOR FOR OPEN DISCUSSION, WHAT I'M GOING TO DO INSTEAD IS PLEASE BEAR WITH YOUR QUESTIONS. LET THE PANELISTS FINISH THEIR PRESENTATION BEFORE YOU ASK A QUESTION. I'LL OPEN UP TWO BLOCKS OF DISCUSSION, OPEN UP SOME TIME FOR DISCUSSION AT THE END OF EACH LOGICAL BLOCK OF TOPICS.
THE FIRST BLOCK OF TOPICS WE'LL DISCUSS WILL BE ON IDN APPLICATION. THEN THE NEXT BLOCK WE WILL TALK ABOUT REGISTRATION -- IDN REGISTRATION POLICY ISSUES.
OKAY? SO WITH THE FIRST PRESENTER, WE HAVE MICHEL FROM MICROSOFT TO TALK ABOUT IDN ON MICROSOFT PLATFORM.
>>MICHEL SUIGNARD:THANKS FOR INVITING ME TO MAKE THIS PRESENTATION.
I SHOULD PROBABLY ALSO SAY WHAT I DO BEYOND BEING PART OF MICROSOFT, I ALSO WORK ON THE UNICODE CONSORTIUM WHERE I AM IN FACT ONE OF THE OFFICERS AS THE TECHNICAL DIRECTOR OF THE CONSORTIUM. SO IF YOU HAVE ANY QUESTIONS ABOUT CHARACTER ENCODING OR HOW TO CONTRIBUTE OR EVEN BECOMING A MEMBER, WE ALWAYS LIKE NEW MEMBERS.
I'M ALSO PART OF 10606 EFFORT. SO WE HAVE A NICE CONNECTION BETWEEN THE TWO EFFORTS. AND I ALSO WORK WITH IETF ON IRI WHICH I'M GOING TO COVER IN MY PRESENTATION.
SO WITH THAT I'LL START THE PRESENTATION. I'VE SAID IT MANY TIMES BUT IT'S WORTH REPEATING BUT IDN IS THE FIRST STEP. WE SEE IDN AS PART OF A PROBLEM WHICH IS REALLY MAKING RESOURCE IDENTIFIERS GLOBAL. JUST GOING BEYOND THE ASCII REPRESENTATION RESOURCES.
BASICALLY, THAT'S A SUMMARY OF WHAT I'M GOING TO DISCUSS DURING THOSE MINUTES.
SO IDN OR DNS, IF YOU WANT, IS REALLY PART, VERY OFTEN, FOR USERS OF WHAT IS KNOWN AS URI. SOMETIME WE USE URL OR URN. URI IS BASICALLY THE UNION OF THE TWO CONCEPT, IT'S A UNIVERSAL RESOURCE AND NAME RESOURCE. IT'S A CONCEPT OF HAVING JUST A NAME SPACE, FOR EXAMPLE IN XML, TYPICALLY WE USE NAME SPACE, URI IS VERY CONVENIENT TO THAT BECAUSE IT'S UNIQUE. UNIQUENESS, WE SEE A BIG ISSUE FOR NAME SPACE.
THE URI IS ASCII ONLY. IF YOU LOOK AT DEFINING THEM WHICH IS RFC 2926, IT'S ALL ASCII. YOU DO HAVE SOME ESCAPING MECHANISM TO REPRESENT NOT ONLY THE RESERVE CHARACTERS, WHICH WOULD CREATE HAVOC IF YOU DIDN'T ESCAPE THEM, BUT YOU CAN ALSO ESCAPE NON-ASCII CHARACTERS, WHICH IS GOOD BUT THE PROBLEM IS THAT THE RFC DOES NOT REALLY GIVE GUIDELINES ABOUT WHICH CONCEPT YOU ARE USING. SO THAT'S ANNOYING, OBVIOUSLY, BECAUSE YOU COULD BE USING ANY CHARACTER SET ENCODING IN THAT ESCAPING.
I MEAN, AS A SOURCE FOR THE ESCAPING. SO IT WOULD BE UTF-8, WHICH IS PREFERRED, BUT IT COULD ALSO BE ONE OF THE EAST ASIAN ENCODING OR ANY OTHERS, BASICALLY.
SO YOU DON'T HAVE A WAY IN URI TO REPRESENT NON-ASCII ENTITIES.
SO PEOPLE DIDN'T REALLY WAIT FOR IETF TO COME WITH A SOLUTION ON THAT. MANY PEOPLE HAVE BEEN HAPPY CREATING SOLUTIONS DE FACTO, WHICH ARE NOT IDN CONFORMANT. THEY DO PUT STUFF IN THE OS NAME WHICH IS NOT IDN VALUE OR DOES NOT CONFORM TO WHAT IS IDN EVEN PRESENTED IN NATIVE UNICODE.
THE SAME THING WITH BIDIRECTIONAL, WHICH IS REALLY A WAY OF EXPRESSING ARABIC ON EBU SCRIPT IS NOT FINALIZED, THERE ARE NO GUIDELINES OR RULES FOR CONVERSION BETWEEN ASCII. THAT'S WHAT IRI DOES BY THE WAY. FOR US AS IMPLEMENTERS OF IDN, IT DOESN'T MAKE SENSE TO IMPLEMENT IDN IF WE DON'T HAVE A CLEAR SENSE OF WHAT IRI IS BECAUSE FOR MOST CUSTOMERS, IDN IS ALWAYS INCLUDED IN IRI BECAUSE IF YOU SEE AN ADDRESS BAR IN A BROWSER, THE IDN IS THE LEFT PART OF THE LINK, AND WHAT IS ON THE RIGHT SIDE IS AS IMPORTANT FOR THE USER. IT'S TRUE THAT IN MANY CASE, YOU DON'T RECEIVE VERY COMPLEX URI, ESPECIALLY IF YOU'RE TRYING TO ENTER YOURSELF, YOU'RE NOT GOING TO ENTER A VERY LONG URI. BUT NEVERTHELESS, IT'S VERY TYPICAL THAT EVEN IN ADVERTISING OR IN -- ON THE SIDE OF THE BUS OR WHATEVER, THE MOVIE CREDIT, ANYTHING WHERE YOU WOULD SEE AN IRI, YOU WOULD HAVE SOME WHAT WE CALL THE PATH. AS AN EXAMPLE, IN FACT, ON THE SCREEN HERE, THERE'S SOME PROJECT, SOME CHARACTERS DON'T REALL!
Y SHOW SO YOU HAVE TO BEAR WITH ME ON THAT. FOR EXAMPLE, THE EXAMPLE DOWN, YOU SEE AN EXAMPLE WITH THE JAPANESE NEWSPAPER, IF I COULD USE THAT. IN FACT, THE SLASH ARE MISSING ON THE PICTURE IN FRONT OF ME SO I GUESS IT WAS MISSING ON THE PROJECTOR, SO YOU HAVE TO IMAGINE THAT BETWEEN THE SEMICOLON OF HTTP AND THE JAPANESE NAME YOU HAVE TWO SLASH, AND AGAIN AFTER THE JP THERE'S ANOTHER SLASH.
SO I DON'T KNOW; THEY DISAPPEARED IN THE VOID.
WE DO COVER -- SO WHAT IRI IS, REALLY, IS FIRST DEFINING NEW PROTOCOL ELEMENTS. I'LL EXPLAIN IN A BIT WHAT THAT IS. WE ALSO COVER TWO SCENARIOS OF ENCODED OR UNENCODED. ENCODED IS SOMETHING YOU WOULD SEE IN A WEB PAGE OR E-MAIL OR SOFTWARE ABLE TO LINK TO ANOTHER RESOURCE. BUT WE ALSO WANT TO REPRESENT THE CASE WHERE IT DOESN'T EXIST; JUST ON A PIECE OF PAPER, ON A CAR, ON A WALL, ANYWHERE. OR EVEN SPOKEN TO YOU. YOU MAY HEAR THE LINK ON THE RADIO.
THIS IS FULLY SPECIFIED MAPPING TO URI, SUPPORTS BIDIRECTIONAL SCENARIOS, ON THIS UNICODE 4.0. THAT'S INTERESTING THAT IN IRI WE ARE MUCH LESS DEPENDENT ON UNICODE VERSIONS. SO IT'S MUCH EASIER FOR US TO EVOLVE TO A NEW VERSION, WHICH STILL MEANS THAT FOR THE IDN FRACTION OF AN IRI, YOU STILL HAVE TO OBEY OVERSEA RULES. WE ARE STILL STUCK IN 3.2 FOR THE RULES.
IT'S ALSO VERY IMPORTANT THAT NORMALIZATION IS IMPORTANT. IRI, YOU WANT TO REDUCE DUPLICATES TO SIMPLIFY MATCHING RULES TO BE NORMALIZED.
ON THE (INAUDIBLE) IS RELATED TO IDN FOR THE AUTHORITY COMPONENT. AND WE HAVE ON THE BOTTOM FOR WHAT TO LOOK LIKE IRI.
IRI USAGE, IT'S VERY IMPORTANT TO UNDERSTAND THAT BECAUSE IT'S REALLY A NEW PROTOCOL ELEMENT, EXISTING SCHEME LIKE HTTP, FTP, MAILTO, THOSE ARE THE ONES THAT EXIST TODAY, YOU SHOULD NOT USE IT DIRECTLY. DIRECTLY, IT'S KIND OF A SUBTLE DISTINCTION. YOU CAN USE IT AS A PRESENTATION ELEMENT FOR AN EXISTING PROTOCOL. THAT'S WHAT, IN FACT, BROWSERS THAT SUPPORT IDN TODAY DO. THEY DO REPRESENT, IF YOU LOOK AT THE ADDRESS BAR, YOU SEE HTTP VALUE WITH A NON-ASCII VALUE. IF YOU LOOK AT THE RFC YOU'RE NOT ALLOWED TO DO THAT. WE CREATED THIS THING OF WHAT IS HTTP. SO YOU HAVE THE SPACE WHERE YOU ALLOW UNICODE VALUES AND YOU HAVE A PROTOCOL LEVELS WHERE THE HTTP IS ON THE ASCII.
SO IT'S ALREADY DE FACTO USE IN BROWSERS. THAT'S KIND OF INTERESTING. SO IRI WORK WAS, TO SOME DEGREE, A WAY OF FORMALIZING WHAT WAS ALREADY IN USE.
THIS IS, FOR EXAMPLE, AN EXAMPLE HOW THIS IS USED. SO ON THE VERY TOP YOU HAVE A PRESENTATION LAYER, WHICH IS BASICALLY THE ONLY THING THAT A CUSTOMER IS SUPPOSED TO SEE ON THE USER INTERFACE. YOU'RE NOT SUPPOSED TO SEE ANYTHING BELOW THAT.
SO THE NEXT LEVEL IS -- IN FACT WAS NOT ALLOWED BY THE URI SPECIFICATION NOW, BUT IT'S BEING, IN FACT, ALLOWED AS I SPEAK, MORE OR LESS, BECAUSE RFC 2396 IS BEING REVISED. IN FACT, IT SHOULD BE APPROVED, I WOULD SAY ANY DAY AT THIS POINT.
ON THE THIRD LEVEL -- OWE THAT USE UTF-8 ESCAPING ON THE HOST PART. THAT'S THE NEW THING. EVERYTHING ELSE IS BASICALLY USING UTF-8 ESCAPING FOR THE -- ALL THE STRING BEHIND.
THE NEXT LEVEL IS WHAT YOU WOULD EXPECT TYPICALLY WHERE THE IDN IS TRANSFORMED IN THE PUNYCODE VALUE. ALL THE REST STAYS IN UTF-8. WHICH IS KIND OF INTERESTING. YOU HAVE TO BE -- WHEN YOU'RE CONVERTING SUCH AN IRI TO URI, YOU HAVE TO BE AWARE YOU HAVE TO DO A DIFFERENT SUMMATION BETWEEN THE HOST PART AND EVERYTHING ELSE.
AND THEN YOU GO TO A DNS RESOLVER WHERE YOU GET THE HOST NAME TO BE RESOLVED.
BIDIRECTIONAL IRIS, THOSE ARE KIND OF INTERESTING BEASTS. FIRST OF ALL, THEY USE A LOGICAL ORDER BECAUSE THERE'S ALWAYS THIS CONCEPT OF VISUAL VERSUS LOGICAL IN BIDIRECTIONAL STRINGS.
THEY'RE PRESENTED AS EMBEDDED LEFT TO RIGHT. IT'S KIND OF IMPORTANT FOR PREDICTABILITY OF THE WAY YOU PRESENT THEM.
AND ALSO, YOU AVOID REORDERING INTERACTION WITH CHARACTERS THAT ARE EITHER BEFORE OR AFTER THAT STRING.
BECAUSE OTHERWISE, YOU GET A SITUATION WHERE YOUR IRI WOULD BE SPREAD OVER IN A SINGLE PARAGRAPH. YOU WOULD NOT EVEN KNOW WHERE THE LINK IS UNLESS YOU PROVIDE SOME SORT OF UNDERLINING OF THE RESOURCE NOTE FIRE.
THE RESTRICTION ON HOST NAME IS THE SAME AS AN IDN. RTL MEANS RIGHT TO LEFT AND LTR MEANS LEFT TO RIGHT. SO IT'S BASICALLY, THOSE ARE REALLY COMING FROM THE IDN SPECIFICATION. WE DIDN'T REALLY CREATE ANYTHING.
WE ALSO SAY THAT BASICALLY THE SAME RESTRICTIONS SHOULD REALLY BE APPLIED TO THOSE OTHER IRI COMPONENTS. IT'S NOT A MUST, IN IETF WAY OF SEEING THINGS, BECAUSE YOU MAY HAVE, IN ACT, OPAQUE RICE. YOU MAY HAVE AN IRI THAT IS NEVER VISITED, NEVER DISPLAYED AND IN THAT CASE YOU DON'T CARE. ALSO QUERY COMPONENTS, THE VERY LAST PIECE OF IRI.
SO AN EXAMPLE HERE, I'M SHOWING SOME REASONABLY SIMPLE CASE OF BIDIRECTIONAL IRI. IT INDICATES, THE RIGHT PART IS THE HOST, THE PART PATH IS THE GREEN AND THE QUERY IS IN BLUE. THE FIRST, YOU CAN SEE IS VERY SIMPLE. THE HOST IS LEFT TO RIGHT. THE SECOND EXAMPLE, YOU HAVE THE PATH IS NOW LEFT TO RIGHT. ON THE PRECEDING THE HOST. THE LAST EXAMPLE IS FULLY (INAUDIBLE). THE HOST IS THE VERY END PART OF THE SCREEN, THE PATH IN THE MIDDLE AND QUERY AT THE BEGINNING. YOU CAN SEE THE HOST IS NOT WHERE YOU THINK IT IS.
THE NEXT EXAMPLE IS A BIT MORE PATHOLOGICAL I WOULD SAY. THE VERY FIRST ONE I JUST USE AS AN EXAMPLE TO SHOW AS A CHANGE FROM THE PREVIOUS ONE. SO IF YOU LOOK AT THE SECOND EXAMPLE, WHAT I DID IN THAT ONE, I INSERTED ABC INSIDE OF THE HOST. THE INTERESTING THING, BY DOING SO, MY HOST GOT SPLIT IN TWO. IN FACT, IF YOU LOOK AT THE RED PART, MY HOST START WITH THE FIRST RED PIECE, GOES TO THE NEXT ONE IS ABC, AND THE THIRD PART IS COMPLETELY ON THE RIGHT. AND THEN YOU JUMP -- AND THEN YOU GO -- THE ARROWS BASICALLY GIVE YOU SOME HINT TO READ THESE THINGS. IT'S NOT EXACTLY CORRECT. I DIDN'T REALLY DECOMPOSE. THE ARABIC DIGIT GOES SLIGHTLY DIFFERENT BUT I DIDN'T WANT TO -- I DID SOME SIMPLIFICATION ON THE DISPLAY.
AND THE LAST ONE IS BASICALLY A SPLIT OF THE PATH. SO IN THAT CASE, MY PATH IS, IN FACT, AROUND MY HOST NAME. SO YOU HAVE PIECES OF PATH PART IS AFTER THE HOST AND THE OTHER PART IS IN FRONT OF IT.
SO IF YOU THINK ABOUT IT, NOW, EVEN KNOWING WHERE THE HOST IS ON THIS THING, IT'S KIND OF COMPLICATED. IN THE FIRST CASE IT'S NICE TO HAVE IT AT THE END, BUT IN THE SECOND EXAMPLE, IT'S SPLIT BETWEEN THE END AND THE BEGINNING. AND ON THE THIRD EXAMPLE, IT'S BASICALLY EMBEDDED INSIDE THE PATH.
ON ALL THOSE IN FACT, THEY ARE EMBEDDED.
SO SKIPPING A BIT BECAUSE I'M RUNNING SHORT ON TIME. RESOURCE NAMING POLICIES. WE'RE KIND OF SAYING THAT IT WOULD BE GOOD THAT THE FIRST NAME SHOULD BE SOME SORT OF LANGUAGE NAME POLICY. WE DON'T REALLY EXPECT A BRITISH NAME TO CONTAIN ARABIC CHARACTERS. IT WOULD BE NICE IF THEY WERE ENFORCED BY THE NICS. WE'VE SEEN THAT ON MUSEUM. WE KIND OF LIKE THAT IDEA. BUT IT'S PROBABLY NOT GOING TO BE ENFORCED BY SOFTWARE. MEANING THAT IF SOMEBODY CREATE SOMETHING WEIRD, WE HAVE TO ACCEPT IT. AT LEAST FROM OUR POINT OF VIEW. IT'S REALLY THE JOB OF REGISTRY TO DO SOME WORK ON POLICING THAT PART.
WE CAN DO SOME CHECK, BUT IT'S VERY DIFFICULT FOR US TO BE TOO -- TOO LIMITING ON THAT ASPECT. WE'RE GLAD TO SEE SOME RULES ON CJK THROUGH THAT RFC 3743.
INTERESTING POINT, THAT MAY NOT BE ENFORCEABLE OR ENFORCED BY THE SUBZONES. IDN OR ICANN IS VERY GOOD AT CREATING GUIDELINES FOR THE MAIN, THE TLDS, BUT WHAT HAPPENED ON THE SUBZONES THAT ARE DELEGATED BY THE TLDS? IT GETS MURKY THE MORE AND MORE YOU GO ON THE LEFT SIDE, IF I TALK IN LEFT TO RIGHT SCRIPT. SO IT'S NOT CLEAR HOW MUCH ENFORCEMENT YOU'RE PUTTING NONE THE DELEGATION. IT'S NOT CLEAR WHAT WE DO, WHEN YOU HAVE THE SUBZONE CREATION THAT'S UNUSUAL.
MULTISCRIPT IS NOT REALLY SOMETHING WE LIKE, BECAUSE TYPICALLY THAT SHOWS TINT TO SPOOF, AS I WILL SHOW IN SOME OF THE EXAMPLE.
LEGACY. I'LL GO QUICKLY ON THAT BECAUSE THAT'S PROBABLY NOT AN ISSUE IN ICANN. THERE IS FRAGMENTATION OUT THERE OF NON-IDN. FOR US, LEGACY IS A BIG DEAL SO WE HAVE TO DEAL WITH IT. ICANN CAN GO ON MORE THAT IF YOU HAVE QUESTION LATER.
SECURITY ISSUE, THAT KIND OF JOINS THE ISSUE THAT WAS MENTIONED BY VALENTIN BEFORE. YOU HAVE A KOREAN NAME, THE PUNYCODE NAMES. AND THEN I HAVE AN EXAMPLE UNDERNEATH IN RED WHERE I DID USE, I THINK IT'S VENDA, WHICH IS ONE OF THE SOUTH AFRICAN LANGUAGES. ON THE LAST ONE IS PUNYCODE. THEY ALL IN FACT POINT TO THE SAME RESOURCE. WHICH ONE DO YOU TRUST?
WE SEE A SIMPLE CASE.
AND SO I WOULD SAY, IN FACT, IN MY OPINION IT'S GOING TO BE ALMOST NECESSARY TO USE CERTIFICATES WITH IDN, BECAUSE HOW AM I GOING TO TRUST YOU OTHERWISE? IT'S VERY DIFFICULT TO TRUST A SITE.
THAT'S NAME SPOOFING HERE. IN ASCII, YOU HAVE EASY SPOOFING, JUST L IS ONE, AND SOME OTHER THINGS, BUT IF YOU TAKE UNICODE, I JUST DID A QUICK CHECK ON THE LATIN, GREEK, AND CHEROKEE, I USED CYRILLIC, GREEK AND CHEROKEE AND COMPARED IT TO THE ORIGINAL LATIN REPERTOIRE, AND I FOUND MANY, MANY DUPLICATES, CHEROKEE BEING OBVIOUSLY THE WORST.
SO YOU CAN ALMOST WRITE ANYTHING THAT LOOKS LIKE LATIN, ESPECIALLY IF YOU USE UPPER CASE AND DON'T CHANGE TO LOWER CASE TO THE TRUE IDN VALUES. BUT FOR SURE IN IRI WHERE I DON'T NEED TO DO THESE KINDS OF THINGS, IT COULD BE VERY CONFUSING.
THE EXAMPLE ON THE BOTTOM IS USING THE VERY WELL-KNOWN EXAMPLE THAT WE USE EVERYWHERE BECAUSE WE GUARANTEE THAT NOBODY TOOK IT, BUT I USE IN FACT MANY CHARACTERS FROM THE CYRILLIC BLOCK. ALL THE ONES IN GREEN WHICH IS EXAPE, WHICH ARE ONLY NOT SAID THAT WAY IN CYRILLIC, AND THEN YOU CAN SEE THE UNICODE. AND IT'S SOMETHING COMPLETELY DIFFERENT. I HAVE NO WAY, VISUALLY TO MAKE THE DIFFERENCE. ABSOLUTELY NO WAY.
ALSO WHAT I MENTIONED BEFORE, THE USER INTERFACE LIMITATION. THE ISSUE BETWEEN BEING LOCALIZED AND BETWEEN BEING GLOBAL. IF I'M LOCALIZED, I GET VERY GOOD REACH TO MY LOCAL MARKET, BUT KIND OF DECREASING THE INTEROPERABILITY. IT MAKES IT A BIT MORE DIFFICULT TO COMMUNICATE WITH OTHER MARKETS OR OTHER CUSTOMERS OR OTHER USERS.
SO IT'S A DIFFICULTY TO ENTER CHARACTERS OR DISPLAY THEM.
OBVIOUSLY, AS VENDOR, WE'RE DOING A LOT OF WORK TO MAKE THAT BETTER. MY PRESENTATION IS PROOF OF THAT. I CAN BASICALLY DISPLAY ANY CHARACTERS I PLEASE, MORE OR LESS, FROM UNICODE. THAT'S PRETTY GOOD.
BUT WE SEE FROM AN INPUT POINT OF VIEW, THAT'S A COMPLETELY DIFFERENT MATTER. IT'S PRETTY EASY TO DISPLAY; IT'S MUCH MORE DIFFICULT TO ENTER BECAUSE YOU NEED SOME SORT OF KEYBOARD MECHANISM TO ENTER THOSE THINGS.
SO WHAT APPLICATIONS, MIDDLEWARE CAN DO? IMPLEMENTING IRI. IT IS, IN FACT, POSSIBLE BECAUSE IRI IS GETTING STANDARDIZED AS WE SPEAK.
VALIDATE IRI AND PUNYCODE NAMES, I MEAN, PRETTY MUCH RIGHT AWAY. IT'S A REALLY BIG ISSUE TO MAKE SURE THAT THE PUNYCODE NAMES YOU GET ARE VALID NAMES, AND SO WE NEED TO VALIDATE THOSE THINGS. HAVE CONSISTENT RULES ABOUT WHAT YOU DISPLAY AND WHAT YOU PROCESS, CREATE LAYERS SO YOU HAVE A PRESENTATION LAYER WHERE YOU DO SOMETHING, LIKE UNDERWIRE LAYER WHERE YOU DO USER ENCODING. DISPLAY CAPABILITY FOR THE IDNA REPERTOIRE. THE BIDIRECTIONAL RULES I JUST TALKED ABOUT, TALKING TO THE USERS ABOUT LANGUAGE AND SCRIPT FILTERS AND MAKING SURE IF SOMETHING LOOKS LIKE SPOOFING WE SHOULD PROBABLY GIVE SOME HINT TO THE USER THAT THAT VALUE THAT IS USING THREE OR FOUR DIFFERENT SCRIPT TO LOOK LIKE LATIN IS PROBABLY NOT SOMETHING THAT THEY SHOULD BE TRUSTING.
SO WHERE WE ARE? IN FACT, WE HAVE IN FACT DONE QUITE A BIT OF WORK ON THAT ALREADY. WE HAVE IMPLEMENTATION OF THE BASIC IDN FUNCTION, IN FACT, IN THE FRAMEWORK. IN FACT, IF YOU GO ON THE MICROSOFT WEB SITE IN THE SDK YOU COULD FIND THEM. WE ALSO HAVE IMPLEMENTED WHAT WE CALL A MANAGED VERSION WHICH IS IN THE NET FRAMEWORK, AND ALSO NATIVE, BASICALLY WIN 32.
INTERNET EXPLORER STATUS, WE USE IRI, BY THE WAY, ON THE ADDRESS BAR. WE JUST KIND OF HAVE TO FIX IT TO BE EXACTLY COMPATIBLE WITH IRI SPECIFICATION. WE NEED TO DO ADDITIONAL WORK TO DO SOME MAPPING BETWEEN IDN, NATIVE UNICODE AND IDN.
SO THAT'S THE END OF IT.
I THINK I HAD -- I'M TRYING TO REMEMBER. YEAH. OKAY.
THIS IS REALLY, SECURITY, YOU KNOW, IS FOR US -- IF I WANT TO CONCLUDE ON SOMETHING, IDN WILL NOT BE IMPLEMENTED BY A LARGE COMPANY WITHOUT HAVING REALLY ADDRESSED SECURITY. THAT'S PROBABLY THE MORE IMPORTANT PART.
WE HAVE BEEN BITTEN BIG TIME ON SECURITY. ON IDN, IT'S JUST A MINE FIELD FOR HACKERS. IT'S REALLY SCARING A LOT OF VENDORS BECAUSE OF THAT.
WE HAVE REALLY TRIED TO ADDRESS SPOOFING ISSUE, ON TRUST, ON WHO YOU ARE, TO BE ABLE TO DEPLOY THAT. IT'S NOT GOING TO BE SUCCESSFUL IF END USERS AND CUSTOMERS OF WEB SITES OR RESOURCES DO NOT TRUST, YOU KNOW, A GIVEN LINK.
SO IT'S IMPORTANT, WHAT WE HAVE TODAY, IDN MAKES THE SITUATION MUCH MORE COMPLICATED. WE HAVE TO BE REALLY SURE THAT WE GET A TRUSTWORTHY SITUATION IN IDN.
OKAY. WITH THAT, I WILL CONCLUDE. I WILL TAKE QUESTIONS LATER, DURING THE QUESTION PART. THANK YOU.
>>JAMES SENG: THANK YOU, MICHEL.
WE WILL HAVE TIME FOR QUESTION AND ANSWER, SO I'M SURE YOU HAVE QUESTIONS. WRITE THEM DOWN SO YOU DON'T FORGET AND PLEASE FORM THE LINE WHEN I OPEN THE FLOOR.
THE NEXT PRESENTER WILL BE SABINE, SHE'S FROM GERMAN DOT DE, SHE'S TALKING ABOUT SOME OF THE STATUS IN GERMANY AND WHY IDN IS IMPORTANT.
>>SABINE DOLDERER: IT WORKS? OKAY. THANK YOU, JAMES. THANK YOU FOR INVITING ME.
I ACTUALLY -- MY PRESENTATION IS BASICALLY MORE ABOUT NOT THE APPLICATION LEVEL BUT THE TECHNICAL LEVEL IN THE REGISTRY AND HOW WE HAVE IMPLEMENTED IDN IN GERMANY. SO A LOT OF THE THOUGHTS WHICH MICHEL GAVE TO SECURITY, A LOT OF THEM SHOULD BE DOWN IN THE APPLICATION LEVEL BECAUSE A LOT OF THEM CAN'T BE DONE BY A REGISTRY.
SO WHAT HAS DENIC DONE? WE HAVE HAD A LOT OF REQUESTS ACCORDING TO IDN, INTRODUCTION IN GERMANY WHEN THE STANDARD WAS PROPOSED BY THE IETF.
WE HAVE STARTED A CONSULTATION AMONGST THE GERMAN SPEAKING REGISTRIES WHICH WAS AUSTRIA, IT WAS SWITZERLAND, GERMANY, AND WE ALL THOUGHT WE SHOULD GO AHEAD WITH THE IDN, THOUGHTFUL, BUT WE SHOULD COME UP WITH A COMMON SOLUTION HOW WE SHOULD IMPLEMENT IT. OUR MAIN GOAL IS WE DON'T WANT TO MAKE CHANGES ON OUR CURRENT POLICY BECAUSE WE THOUGHT WE ALL ARE VERY, VERY KEEN ABOUT WE HAVE SEEN THAT WE HAVE DONE SUCCESSFUL REGISTRATIONS IN THE PAST, AND PEOPLE ARE USED TO IT. PEOPLE DON'T WANT TO SEE REALLY MAJOR CHANGE ON THAT. BUT ON THE OTHER HAND WE WANT TO SUPPORT IDN, SN IDN SO NOT ONLY THE XN DASH DASH REGISTRY REPRESENTATION BUT IF WE START TO REGISTER UMLAUT DOMAINS, THEN THESE DOMAINS ACTUALLY HAVE TO BE THE DOMAINS WHICH WE REGISTER WHICH WE SHOW THE CUSTOMER AS THE ONE -- AS OUR REGISTER BECAUSE WE THINK THE XN REPRESENTATION ON THE LONG RUN WILL BE INVISIBLE, WILL START TO GET INVISIBLE TO THE CUSTOMER.
SO WHAT ARE OUR BASIC GOALS? FIRST OF ALL, THE SUPPORT FOR THE IDN STANDARDS AS THEY WENT OUT. WE HAVE GONE -- WE SAID WE DON'T WANT TO MAKE A HUGE APPROACH, WE WANT TO GO FOR A POSITIVLIST ON THE SUPPORTED CHARACTERS. WE THEN DISCUSSED WHICH CHARACTERS WE SHOULD SUPPORT AND WE ACTUALLY END UP VERY -- SO THE VERY OBVIOUS WOULD BE THE THREE UMLAUT DOMAINS BUT ON THE OTHER SIDE, AS WE LIVE IN THE MIDDLE OF EUROPE, ALL OF US, WE'VE SEEN A LOT OF CHARACTERS ACTUALLY COMING IN OUR LANGUAGE, COMING IN OUR NAMES REGISTRIES IN OUR COUNTRIES. SO IN GERMANY IT'S VERY USUAL TO HAVE AN E WITH AN (INAUDIBLE) E AS A SURNAME. IT'S VERY COMMON THAT YOU HAVE DIFFERENT XNS LEVEL BECAUSE WE HAVE BORDERS, POLAND, WE HAVE LANGUAGE MINORITIES IN GERMANY. YOU MIGHT BE AWARE OF THAT. WE HAVE CERTAIN LANGUAGE MINORITY USING POLISH SCRIPT, DANISH LANGUAGE MINORITY USING DANISH SCRIPTS AND WE HAVE WITH OUR AWARD TO FRANCE QUITE A HUGE AMOUNT OF THE OTHER CHARACTERS COMING IN THE GERMAN AL!
PHABET.
SIMILAR HAPPENED WITH AUSTRIA WITH THE (INAUDIBLE), AND EVEN COMPLICATED THE SITUATION IN SWITZ WHERE THEY HAVE FOUR DIFFERENT LANGUAGES. SO WE ACTUALLY, WE STARTED TO MAKE SOME SERVICE WITHIN OUR COMMUNITIES ASKING IF THERE ARE SORT OF AN OFFICIAL LANGUAGE TABLE COMING FROM THE GOVERNMENT, BUT THERE ARE DE FACTO ALSO NONE.
SO WE START -- WE SAID WE WILL START TO SUPPORT THE MAIN LATIN CHARACTERS, WHICH ARE IN THESE REGIONS WHERE WE ARE REPRESENTED WHICH IS BASICALLY LATIN ONE SUPPLEMENT AND LATIN EXTENDED A.
WE GO FOR ONLY LETTERS, NO SPECIAL CHARACTERS LIKE EURO WHICH PEOPLE ARE ALSO ASKING FOR AND ALSO DOMAIN NAME APPLICATIONS FOR AFTER APPLICATION OF NAME PREP.
WE DO NO VARIANTS OR BUNDLING BECAUSE THERE IS NOT A REAL TRANSLATION BETWEEN ALL THESE CHARACTER -- DIFFERENT CHARACTERS.
THE CONTRACT, AS I TOLD EARLIER, IS FOR THE IDN, SO I PUT TANGENTIALLY TWO DOMAINS AS AN EXAMPLE, CITREON AND NESTLE, WHICH ARE REGISTERED FOR THESE COMPANIES YOU WOULD EXPECT.
WE ENDED UP WITH 92 ADDITIONAL CHARACTERS WHICH WE NOW SUPPORT.
AND WE STARTED AT THE 1ST OF MARCH WITH AN INITIAL REGISTRATION.
WE SAID WE WANT TO GO FOR FIRST COME, FIRST SERVED.
WHAT WE HAVE TO DO -- TO SET UP ALL THESE ISSUES WERE WE WENT ACTUALLY TO EACH AND EVERY APPLICATION TO MAKE IT IDN-READY, WHICH IS OUR WHOIS SERVICE, WHICH IS OUR REGISTRATION SERVICE, WHICH IS UTF-8.
IN THE SAME TIME WE DID THIS, WE ACTUALLY ALLOWED ALSO THE DOMAIN NAME HOLDER TO USE MULTILINGUAL DOMAIN -- SORRY.
I -- I AM -- I TRY TO COVER -- WE TRIED ALSO THE DOMAIN NAME HOLDER TO USE MULTILINGUAL CHARACTERS.
AND WE ACTUALLY BASICALLY WENT THROUGH ALL OUR APPLICATIONS, WHOIS WEB AND SO ON.
WE HAVE RECESSED BETWEEN THE 1ST OF MARCH BETWEEN 10:00 AND 12:00, WE RECEIVED MORE THAN 600,000 APPLICATIONS.
24% WERE SUCCESSFUL.
WE HAVE REGISTERED OVER A QUARTER MILLION IDN DOMAIN NAMES IN GERMANY, AND IT'S STILL INCREASING WE HAVE SEEN, DESPITE INTENSIVE WARNINGS WHICH WE MADE AT THE BEGINNING THAT THERE WERE INFRINGEMENT OF TRADEMARKS, WE ARE IN A LUCKY SITUATION THAT THE GERMAN LAW IS VERY, VERY CLEAR ABOUT INFRINGEMENT OF TRADEMARKS INFRINGEMENT OF NAMING RIGHTS.
AND IT'S VERY EXPENSIVE IF YOU DO SO, SO THESE PROBLEMS ARE SOLVED VERY EARLY, ACTUALLY IN THE FIRST WEEKS.
THE PRESS RESULTS ARE MAINLY POSITIVE.
SO BESIDES THE MISSING IE SUPPORT FOR THE IDNS.
BUT IN TOTAL, WE GOT OVER A THOUSAND PRESS REPORTS, 20 RADIO INTERVIEWS, A LOT OF T.V. STATIONS COG THE REPORT.
SO IDN IS NOW ON THE WAY TO BECOME A COMMON STANDARD.
YEP.
THAT WAS BASICALLY THE PRESENTATION.
FIVE MINUTES?
I GOT IT ALL.
THANKS.
QUESTIONS, PLEASE, LATER ON.
>>JAMES SENG: YEAH, LATER FOR QUESTIONS.
SABINE, THANKS.
THE NEXT PRESENTER IS ACTUALLY DARIN FROM MOZILLA.
BUT WILLIAM WILL PRESENT.
HE HAS KINDLY OFFERED TO DO DARIN'S PRESENTATION ON HIS BEHALF.
SO WE WILL HAVE TO TALK ABOUT THE ROADMAP -- THE IDN IMPLEMENTATION IN MOZILLA AND (INAUDIBLE).

>>WILLIAM TAN: OKAY.
HI, EVERYBODY.
I AM PRESENTING MOZILLA'S IDN IMPLEMENTATION ON BEHALF OF DARIN FISHER.
DARIN AND I HAVE WORKED TOGETHER FOR IDN SUPPORT IN MOZILLA SOME TIME BACK.
BUT HE HAS CONTINUED WORKING ON IT FOR A LONG TIME AFTER I'VE STOPPED.
AND HE'S ACTUALLY THE NETWORK GUY BEHIND MOZILLA.
TO BEGIN WITH, MOZILLA SUPPORT FOR IDN IS BASED ON A CODE FROM JPNIC.
THE CODE WAS INTEGRATED INTO MOZILLA'S SOURCE CODE REPOSITORY.
AND IT'S BASICALLY THE STANDARDS -- IT'S STANDARDS-COMPLIANT.
IT'S BASED ON THE IDNA SPECIFICATION FOR NAMEPREP AND ALSO PUNYCODE.
AND THE HISTORY OF IT WAS THAT MOZILLA ALREADY HAD SUPPORT FOR IDN EVER SINCE 1.4, WHICH WAS RERENDERED AS NETSCAPE 7.1.
AND AFTER WHICH THERE HAVE BEEN NUMEROUS BUG REPORTS AND ENHANCEMENTS, UP UNTIL FIREFOX 1, NETSCAPE 7.2, I WOULD SAY THE USER FEEDBACK, AND WE'VE RECEIVED A LOT OF FEEDBACK AND POSITIVE FEEDBACK FROM THE COMMUNITY.
AND CURRENTLY, MOZILLA TRUNK ALSO HAS A LOT OF IMPROVEMENTS IN TERMS OF IDN SUPPORT.
NOW, LET'S GO INTO HOW MOZILLA HANDLES THE IDN DURING URL PASSING.
THAT IS BASICALLY THE CORE OF IT.
BECAUSE MOZILLA ACTUALLY USES -- HAS A FAIRLY COMPLEX BUT YET ROBUST URL PASSER.
IT -- WHEN WE RECEIVE AN INPUT, THE URL PASSER WILL TAKE THE URL IN THE FORM OF A UNICODE OR ACE STRING.
AS OF MOZILLA 1.8, ACE WILL BE CONVERTED TO UNICODE, THAT IS THE TWO UNICODE OPERATION AS OF THEN.
AND URL OBJECTS A CODE AS UNICODE AND ACE VERSIONS.
SO PRESUMABLY, I THINK THE UNICODE VERSION WOULD BE USED FOR DISPLAY AND THE ACE VERSION WOULD BE USED FOR RESOLUTION.
USER INTERFACE CODE REQUESTS AND UNICODE -- THE -- YEAH, SO THE USER INTERFACE DISPLAY AND RENDERING WOULD USE THE UNICODE VERSION OF THE URL AND THE OTHER USES THE ACE.
AND IT'S AS OF MOZILLA 1.7.
HERE'S A SCREEN SHOT OF MOZILLA USING IDN IN ACTION.
AS YOU CAN SEE, ON THE URL BAR, IT'S INPUT BY THE KEYBOARD.
AND IT'S DISPLAYED PROPERLY.
AND ON THE WEB PAGE ITSELF, YOU'VE GOT TWO LINKS.
THEY'RE A BIT SMALL, I'M SORRY, BUT THEY'RE ACTUALLY EQUIVALENT.
THE FIRST LINK IS THE IDN, AND IT'S IN UNICODE.
AND THE SECOND LINK IS THE PUNYCODE EQUIVALENT.
AND THEY ARE TREATED THE SAME.
AND AS YOU CAN SEE IN THE HISTORY WINDOW UNDER THE LOCATION COLUMN, THE IDN IS ALSO DISPLAYED CORRECTLY.
NOW, FUTURE PLANS.
THERE WILL BE A LOT MORE POLISHING UP TO DO.
AS YOU CAN SEE IN THE PREVIOUS SCREEN, ALL OF THE IDNS ARE DISPLAYED CORRECTLY, BUT THERE ARE LOTS MORE PLACES WHERE HOST NAMES AND URLS ARE DISPLAYED THAT SOME OF THE CASES ARE NOT MAKING USE OF THE UNICODE VERSION OF THE HOST NAME YET.
SO THERE WILL BE MORE WORK TO BE DONE TO POLISH UP SO THAT USERS DON'T GET SURPRISED BY THE XN DASH DASH.
WE -- THEY ALSO NEED TO ENSURE THAT URLS ARE ALWAYS COMPARED IN THE NORMALIZED FORM.
SO THERE IS MORE WORK TO BE DONE THERE.
AND THERE ARE PLANS TO IMPROVE THE SUPPORT FOR IRIS, WHICH MICHEL HAVE TALKED ABOUT.
SO THAT WE CAN LEVERAGE THE ABSTRACTION LAYER INTRODUCED FOR IDN.
AND THAT'S IT.
END OF SLIDE.

>>JAMES SENG: THANK YOU.
WE HAVE ANOTHER OPEN SOURCE PRESENTATION.
THIS TIME IS FROM WILLIAM HIMSELF.
WILLIAM IS ACTUALLY THE DEVELOPER OF IDN OPEN SOURCE BUT IS HERE ON BEHALF OF NEULEVEL.
SO WILLIAM.

>>WILLIAM TAN: OKAY.
YES, SO I'M -- I AM CURRENTLY CONSULTING WITH NEULEVEL, AND I'VE BEEN WORKING FOR THE IDN OPEN SOURCE PROJECT FOR ABOUT A YEAR-PLUS NOW.
NEULEVEL RECOGNIZED THAT THERE IS A NEED FOR APPLICATION SUPPORT IN ORDER FOR IDN TO TAKE OFF.
WE ALL KNOW THAT THE IDN STANDARD, IDNA, MEANS IDN IN APPLICATIONS.
AND IF APPLICATIONS DON'T SUPPORT IDNS, THERE'S JUST NO WAY FOR IT TO GO ANYWHERE, YOU KNOW, FAR ENOUGH.
SO, NATURALLY, WE WOULD LOOK AT WEB BROWSERS, E-MAIL CLIENTS, INSTANT MESSAGING CLIENTS AS WELL.
AND WE NEED A PROJECT TO BASICALLY IMPLEMENT THESE PLUG-INS THAT ARE STANDARDS-COMPLIANT, AND MOST IMPORTANTLY, THEY SHOULD BE OPEN SOURCE, THEY SHOULD BE NONPROPRIETARY, AND IDEALLY, COMMUNITY-DRIVEN AND DEVELOPED.
SO THAT'S HOW IDN-OSS IDEA WAS CONCEIVED.
IT WAS ACTUALLY FIRST CONCEIVED IN ICANN -- IN THE ICANN MEETING IN MONTREAL AND WAS ACTUALLY OVER A BREAKFAST DISCUSSION BETWEEN RICHARD TINDAL AND JAMES.
AND AFTERWARDS, JAMES WROTE THE BUSINESS PLAN, CANVASSED THE RESOURCES AND KICK-STARTED THE PROJECT.
AND WE'RE VERY LUCKY TO HAVE OUR -- ON OUR ADVISORY COUNCIL VINT CERF, MACK DAVIS, MARTIN DÜRST, JOHN KLENSIN, AND PAUL HOFFMAN.
PROJECT IS CURRENTLY HOSTED BY ISC.
AND THE GOAL OF THE PROJECT IS, AS I SAID, TO DEVELOP OPEN SOURCE, STANDARDS-COMPLIANT SOFTWARE FOR IDN IN APPLICATIONS.
WE -- INITIALLY, WE WOULD LIKE TO TARGET WEB BROWSERS, BECAUSE THAT MAKES SENSE, BECAUSE IT IS THE MOST COMMONLY USED APPLICATION FOR THE INTERNET.
AND INTERNET EXPLORER IS CURRENTLY (INAUDIBLE) THAT IT'S LACKING.
SO WE EXPECT TO PROVIDE THIS BRIDGE WHILE IDN FUNCTIONALITY IS COMING UP AS NATIVE SUPPORT IN APPLICATIONS.
SO THE TIME LINE FOR THE PROJECT IS, WE HAVE SUMMER 2003, THE PROJECT WAS FIRST STARTED.
AND THEN IN FALL, ISC HOSTED -- STARTED HOSTING THE PROJECT.
AND IN SPRING 2004, WE RELEASED THE FIRST IE PLUG-IN.
AND IN SUMMER, WE HAVE BEEN WORKING ON VARIOUS CODE IMPROVEMENTS AND EXPECT TO ACTUALLY -- IT'S CURRENT IN VERY ACTIVE DEVELOPMENT.
AND IDN-OSS, THE PROJECT HAS RELEASED SO FAR TWO APPLICATIONS, THE FIRST OF WHICH IS A CONVERSION TOOL THAT ALLOWS DEVELOPERS, DOMAIN OWNERS, AND VARIOUS PEOPLE TO USE IT AND CONVERT IDNS INTO PUNYCODE AND VICE VERSA.
IT ALSO USES THE JPNIC IDN KIT LIBRARY FOR THE IDNA OPERATIONS.
AND THE OTHER ONE IS OUR FLAGSHIP PRODUCT WHICH IS THE PLUG-IN FOR INTERNET EXPLORER.
WHAT IT DOES IS ALLOWS YOU TO USE IDN IN IE BY EITHER TYPING INTO THE URL BAR OR BY LINKS, CLICKING ON LINKS EMBEDDED IN WEB PAGES.
AND WE HAVE A NEW NAME FOR IT, WHICH IS ECHIDNA.
AND BOTH PRODUCTS ARE RELEASED UNDER LICENSE.
THE STATUS OF THE PLUG-IN IS BASICALLY 75% COMPLETE.
BASIC FUNCTIONING WORKS, BUT WE HAVE A LOT MORE TO DO.
SOME IS THAT WE NEED LESS LEAKAGE OF XN DASH DASH STRINGS ALL OVER THE PLACE.
WE NEED TO IMPROVE ON THE USER INTERFACE OF THE PLUG-IN ITSELF.
WE'RE ALSO LOOKING INTO SUPPORTING E-MAIL CLIENTS LIKE OUTLOOK, OUTLOOK EXPRESS.
WE'RE ALSO LOOKING AT SUPPORTING PROGRAMS TO ACTUALLY EMBED THE IE BROWSER ENGINE, LIKE WINAMP AND STUFF.
IT WILL BE ALSO NICE IF THE SOFTWARE CAN AUTOMATICALLY UPDATE ITSELF.
WHAT YOU CAN EXPECT FROM THE PROJECT IN THE SHORT FUTURE IS THAT THERE WILL BE A MORE STABLE VERSION OF ECHIDNA.
WE'RE SETTING UP A CVS REPOSITORY TO MANAGE THE SOURCE CODE SO THAT WE CAN HAVE MORE COMMUNITY INVOLVEMENT.
CURRENTLY, WE DO RELEASE THE SOURCE CODE.
BUT IT'S JUST NOT MANAGED BY ITS SOURCE CONTROL RIGHT NOW.
WE ALSO EXPECT TO HAVE A SHORTER RELEASE CYCLE AND ALSO POSSIBLY THE OUTLOOK AND OUTLOOK EXPRESS PLUG-IN IS ON THE WAY.
WE WOULD ALSO LIKE TO HAVE A NEW WEB SITE SO THAT IT'S MORE COMFORTABLE AND USER FRIENDLY FOR PEOPLE.
SO WE DO NEED, AS IT IS A COMMUNITY PROJECT, WE NEED TO HAVE THE SUPPORT OF THE COMMUNITY.
AND WE WOULD LIKE TO SORT OF CANVASS FOR MORE RESOURCES, SUCH AS FUNDS, USER -- HUMAN RESOURCES FOR TESTING AND PATCHING THE SOFTWARE, WRITING DOCUMENTATION, TRANSLATIONS, AND FOR ANY FEEDBACK SUGGESTIONS, AND JUST GENERAL MARKETING FOR SPREADING THE WORD.
YOU KNOW, AND WE ALSO NEED A LOGO FOR THE PLUG-IN.
AND HERE ARE THE CONTACT DETAILS FOR THE PROJECT, PROJECT WEB SITE IS IDN.ISC.ORG.
JAMES SENG IS THE PROJECT CONVENER.
AND FOR FINANCIAL AND LEGAL ISSUES, PLEASE CONTACT PAUL VIXIE, AND DIRECT THE TECHNICAL STUFF TO ME.
AND THAT'S IT.
THANK YOU VERY MUCH.

(APPLAUSE.)
>>JAMES SENG: OKAY.
THE NEXT SPEAKER IS MAO FROM CNNIC.
HE IS GOING TO TALK ABOUT THE CNNIC CLIENT FOR IDN SUPPORT.

>>MAO WEI: OKAY.
I WILL GIVE A VERY BRIEF INTRODUCTION ABOUT OUR CDN RESOLUTION AND OUR OPINION ABOUT THE IDN E-MAIL ADDRESS.
WE HAVE TWO SOLUTIONS FOR IDN RESOLUTION.
ONE IS IDN PROXY SOLUTION.
USERS DON'T NEED TO DOWNLOAD CLIENT SOFTWARE.
IT'S A REGISTRY SERVER, SET UP A PROXY SERVER TO PROVIDE A RESOLUTION SERVICE.
IT'S A TEMPORARY SOLUTION.
ANOTHER IS A CLIENT SOFTWARE SOLUTION.
FOR IDN PROXY, USER TYPE IDN IN THE I E ADDRESS BAR.
NEXT IS HTTP, CHINESE, OR (INAUDIBLE).CN.
IT GOES THROUGH OUR PROXY SERVER TO OUR -- TO OUR CDN PAGE.
IN THIS PAGE, IT WORKS FOR CHINESE DOMAIN NAME, USER DOWNLOAD OUR CLIENT SOFTWARE.
AFTER USER MAKE A CHOICE WHICH SIDE HE WANTS TO GO, IT WILL DISPLAY THE WEB SITE.
BUT IN THE ADDRESS BAR, IT DISPLAYS THE PUNYCODE AND NOT CHINESE.
IF USER SYSTEM IS NOT CHINESE VERSION PLATFORM, SOMETIMES THE SOLUTION IS DIFFERENT.
CNNIC ALSO PROVIDE THE CLIENT SOFTWARE.
IT WAS LAUNCHED IN 2000.
IN 2003, WE UPGRADE THIS KIND OF SOFTWARE, MAKING IT COMPATIBLE WITH IETF STANDARD.
THIS KIND OF SOFTWARE SUPPORT CONVERTING FROM CHINESE CHARACTER TO PUNYCODE, ALSO CAN SUPPORT DISPLAY OF CHINESE DOMAIN NAME IE ADDRESS BAR.
AND THE SUPPORT IS DIRECTLY INPUT OF CHINESE DOMAIN NAME, I MEAN, DON'T NEED TYPE HTTP OR TRIPLE W.
IT'S MORE CONVENIENT FOR USER.
ALSO CAN SUPPORT IDN E-MAIL ADDRESS AND PING, TELNET.
I INTRODUCED OUR CLIENT/SERVER DETAIL.
BASICALLY, NOT (INAUDIBLE) I WANT TO SAY, OUR CLIENT SOFTWARE IS VERY POWERFUL.
ACTUALLY, MICROSOFT, HOPE MICROSOFT CAN SUPPORT THIS FUNCTION.
I AM VERY GLAD THAT YOU ARE HERE.
I AM SURE THAT YOU WILL GIVE A NEW VERSION OF IE BROWSER.
YEAH.
HERE I GIVE YOU AN EXAMPLE.
USER HAVE AN IDN CLIENT SOFTWARE.
YOU CAN TYPE THE WEATHER, CHINESE WEATHER AND GO TO THIS WEB SITE.
YOU CAN FIND THE ADDRESS BAR DISPLAY CHINESE CHARACTER.
THIS WEB SITE BELONGS TO CHINA CENTRAL TELEVISION.
ANOTHER THING I WANT TO SAY IS E-MAIL ADDRESS.
ANOTHER PROBLEM.
DOMAIN PROTOCOL IS FUNDAMENTAL PROTOCOL MOST AVAILABLE INTERNET PROTOCOL BASED ON THAT.
SO WE HAVE A NEW DOMAIN NAME PROTOCOL.
SO WE ALSO SHOULD MODIFY THIS APPLICATION PROTOCOL.
THE MOST POPULAR APPLICATION IS E-MAIL.
CURRENTLY, THERE'S A PROBLEM.
THERE ARE TWO PARTS BETWEEN AT, THE RIGHT-SIDE PART OF THE ADDRESS HAS BEEN SOLVED BY IDN STANDARD.
BUT THE LEFT SIDE HAS STILL NOT STANDARD TO SOLVE IT.
DIFFERENT COMPANY HAVE THEIR OWN SOLUTION.
SOME OF THAT MAY DISTURB THE STABILITY AND SECURITY OF THE INTERNET.
HERE IS AN EXAMPLE SOME COMPANY PROVIDES.
CURRENTLY, TECHNICAL PEOPLE ARE DISCUSSING THE SOLUTION.
THEY CAME UP WITH THREE PROPOSALS.
I DON'T WANT TO GIVE A DETAILED INTRODUCTION.
FOLLOWING THIS PROPOSAL, CNNIC HAVE THE SOLUTION TO DO THAT.
IF YOU DOWNLOAD THE SOFTWARE, YOU CAN USE THE CHINESE NAME E-MAIL ADDRESS.
NO MATTER WHAT KIND OF SOLUTION, WE DID OUR STANDARD, IT IS VERY USEFUL FOR THE MOST POPULAR APPLICATION, E-MAIL.
THANKS.

>>JAMES SENG: THANK YOU, MAO.
WE HAVE THE NEXT PRESENTER, FROM PRR, KENNY HUANG.
HE IS GOING TO TALK ABOUT SOMETHING SLIGHTLY ISSUE, WE SEE MULTIPLE CLIENTS, WE HAVE INET, -- ALL DIFFERENT ONES, HOW DO THEY INTEROP, WE HAVE MOZILLA, OPERA, CONQUERER, WHICH ALSO SUPPORTS IDN TODAY.
SO, KENNY.
>>KENNY HUANG: OKAY.
THANK YOU.
I'M TRYING TO ELABORATE.
SOME ISSUE HAPPENED IN IDN INTEROPERABILITY.
AND THE ISSUE PROBABLY MAY HAPPEN OR ALREADY HAPPENED IN DIFFERENT KIND OF IDN IMPLEMENTATION.
SO I TRY TO BE QUITE BRIEF BECAUSE WE STILL HAVE SOME SPEAKERS.
BASIC ARCHITECTURE, SOFTWARE ARCHITECTURE IS A BRIDGE BETWEEN REQUIREMENT AND IMPLEMENTATION.
AND ALTHOUGH WE HAVE SET OF IDN STANDARD NOT MENTIONED AND SPECIFIED IN IDN WHAT YOU NEED TO COMPLY WITH IDN STANDARD.
BUT IDN STANDARD DOESN'T REGULATE APPLICATION BEHAVIOR.
SO BASED ON THE SOFTWARE ARCHITECTURE, STILL HAVE DIFFERENT KIND OF IMPLEMENTATION COULD HAPPEN THAT COULD AFFECT THE INTEROPERABILITY AMONG DIFFERENT SOFTWARE.
SO, BASICALLY, TRY TO MENTION SOME PRINCIPLES REGARDING TO INTEROPERABILITY.
AND FOR DEFINITION ABOUT INTEROPERABILITY, WHICH MEANS AN ABILITY FOR MULTIPLE SOFTWARE COMPONENTS WRITTEN IN DIFFERENT PROGRAMMING LANGUAGE TO COMMUNICATE AND INTERACT WITH ONE ANOTHER.
SO THAT'S THE BASIC DEFINITION OF THE INTEROPERABILITY.
AND SOME KEY ASPECT REGARDING TO INTEROPERABILITY IS THE SEAMLESSNESS.
AND THE DEVELOPER OF MULTILINGUAL SOFTWARE SYSTEM NEED TO BE AWARE OF LANGUAGE DIFFERENCES BETWEEN INTEROPERATING SOFTWARE COMPONENTS AND EXAMPLE OF THE SEAM INCLUDE THE APPEARANCE OF FOREIGN FUNCTION INTERFACE CALLS IN THE SOURCE CODE.
AND SOME OPEN SOURCE SOFTWARE ALREADY PROVIDES SOME SORT OF FEATURE TO ALLOW PEOPLE TO DIRECT APPLY THE FUNCTION CODE TO ACTIVATE AN IDN CAPABILITY.
AND I ALSO TRY TO BRING SOME SCENARIO THAT HAPPENED IN CURRENT PRACTICE IN IDN APPLICATION.
AND THAT'S VERY COMMON SCENARIO TO HAPPEN.
FOR EXAMPLE, WE HAVE DIFFERENT KIND OF IDN CLIENT SIDE.
SORRY, IDN CLIENT PLUG-IN SOFTWARE, NOT DEVELOPED, BUT DEVELOPED BY DIFFERENT SOFTWARE DEVELOPERS.
AND WE ALSO HAVE IDN SERVER.
AND ALTHOUGH THE IDN CLIENT MAY BE ABLE TO COMMUNICATE WITH THE IDN SERVER, BUT THEY STILL HAVE DIFFERENCE, LIKE, FOR EXAMPLE, THEIR CAPABILITY TO ALLOW USER TO INPUT THEIR NATIVE ENCODING OR THEIR -- THEIR ABILITY TO PRESENT, LIKE, A RESOURCE LOCATOR THAT CAN PRESENT LATER ENCODING OR LATER (INAUDIBLE) OF PUNYCODE ENCODING.
AND FOR THIS KIND OF IMPLEMENTATION, PROBABLY VERY DIFFERENT BETWEEN DIFFERENT IDN CLIENT PLUG-IN.
THAT'S ANOTHER ISSUE FOR INTEROPERABILITY WITH MULTIPLE IDN CLIENT PLUG-IN.
FOR EXAMPLE, PROBABLY USER COULD DOWNLOAD A DIFFERENT IDN PLUG-IN SOFTWARE FROM DIFFERENT PROVIDER.
AND, BASICALLY, THE IDN CLIENT PROVIDER, THEY PROBABLY HAVE ACTIVATION FUNCTION TO ALLOW USER TO ACTIVATE IDN FEATURE.
BUT MAYBE SOME DON'T HAVE THE ACTIVATION FUNCTION.
SO ONCE THAT HAPPENED, THAT COULD CAUSE SOME PROBLEM, FOR EXAMPLE, LIKE A SYSTEM COULD BE CRASHED, BECAUSE THEY HAVE MULTIPLE IDN PLUG-IN SOFTWARE.
AND THIS ISSUE ALREADY MENTIONED BY JOHN KLENSIN IN PREVIOUS SESSION REGARDING COMPATABILITY AND REVISION, ESPECIALLY FROM UNICODE RESCISSION.
CURRENTLY, IDNA USES UNICODE 3.2.
AND THAT CAUSES SOME COMPARABILITY IF UNICODE CHANGED IN A FUTURE REVISION.
AND WE PROBABLY NEED TO EXPLORE THIS, BECAUSE WE USE IDN IN THE CURRENT UNICODE VERSION.
AND THE OTHER ISSUE ALSO MENTIONED BY MAO IS APPLICATION RELATED TO INTEROPERABILITY.
WE HAVE IDN STANDARD, BUT WE STILL NEED A LOT OF APPLICATION TO SUPPORT THE STANDARD TO MAKE IT HAPPEN.
WE ALSO NEED E-MAIL HAPPEN RUNNING OVER THE IDN.
AND WE ALREADY HAVE BROWSER AND -- OLDER APPLICATION IS A BASIC REQUIREMENT FOR USER TO ENABLE THE IDN ENVIRONMENT.
OKAY.
THANK YOU.

>>JAMES SENG: THANK YOU, KENNY.
WE HAVE A LAST PRESENTER FOR THIS BLOCK, HIRO HOTTA FROM JPRS WHO IS GOING TO TALK ABOUT SOME OTHER INTERESTING APPLICATION OF INTERNATIONALIZED DOMAIN NAMES OTHER THAN USING IT FOR WEB SURFING.

>>HIRO HOTTA: THANK YOU, JAMES.
I'M HIRO HOTTA FROM JPRS, THE DOT JP REGISTRY.
AND I'D LIKE TO TALK ABOUT THE IDN ENVIRONMENT BEING BOOSTED, ESPECIALLY IN THE JAPANESE MARKET.
SO, FIRST OF ALL, THE TECHNICAL STANDARD IS ALREADY SET THERE.
AND THERE SEEMS TO BE REAL MARKET NEEDS IN JAPAN.
FOR EXAMPLE, FAMOUS PERSONS' WEB SITES HAS IDNS, THEY ARE LABELED.
THIS IS A (SPEAKING JAPANESE), WHO'S A KIND OF FAMOUS SOCCER PLAYER'S WEB SITE.
AND THE SECOND ONE IS PRODUCT CAMPAIGNS.
THIS IS A SOFT DRINK CALLED NAMACHA.
AND THAT'S USED.
AND SOME GOVERNMENTAL PROCUREMENT REQUIRES IDN-AWARE BROWSERS HAS TO BE ALREADY BUILT IN WHEN PCS ARE DELIVERED.
IN THIS CASE, THE PATENT OFFICE OF JAPAN REQUIRES -- ACQUIRES 5,000 PERSONAL COMPUTERS WITH IDN-AWARE BROWSERS.
SO AT LEAST IN JAPANESE MARKET, THERE'S SEVERAL, OR MANY, APPLICATIONS WHICH ARE IDN-AWARE.
FOR PERSONAL COMPUTERS, AS YOU ALL KNOW, BROWSERS SUCH AS NETSCAPE, OPERA, FIREFOX, SAFARI ARE ALREADY IDN-AWARE, AND SOME IDN PLUG-INS, SUCH AS I-NAV, ALREADY EXIST.
AND SEARCH ENGINES, GOOGLE IS IDN-AWARE.
I WILL SHOW THE EXAMPLE LATER.
AND, OF COURSE, SOME E-MAIL SOFTWARE, IDN.
AND FTPS AND SOMETHING.
AND FOR MOBILE PHONE THAT HAS INTERNET ABILITY HAS ALREADY IDN FUNCTIONS THERE.
THE BROWSERS ARE MOBILE INTERNET FUNCTIONS SUCH AS OPERA, NETFRONT, JIG.JP OR JAJP.JP.
THOSE BROWSERS ARE AT LEAST JAPANESE DOMAIN NAME AWARE.
AND ALSO, POCKET PC CAR NAVIGATION SYSTEM, WHICH CAN BROWSE THE WEB SITE.
THERE'S SOME APPLICATIONS OF BROWSERS ON THOSE DEVICES WHICH ARE IDN-AWARE ALSO.
SO THIS IS -- THERE ARE SOME EXAMPLES THERE.
THIS IS AN EXAMPLE OF OUTPUT OF GOOGLE.
THIS IS A VERY FAMOUS JAPANESE DOCTOR WHO IS A KIND OF -- ONE OF THE INTERNET FATHERS.
HE HAS A JAPANESE DOMAIN NAME, WEB PAGE, AND IN THE GOOGLE, AS YOU SEE, I HOPE YOU CAN SEE IT, THE JAPANESE CHARACTER DOMAIN NAME DOT JP CAN BE SHOWN AS A RESULT OF THE GOOGLE.
AND IT IS THE HIGHEST RANK -- OF THE HIGHEST RANK OF THE RESULT, -- SORRY, I CLICKED IT, AND IT GOES TO KOTO-SAN'S SPACE. YES. AND THIS IS A SOFT DRINK CAMPAIGN. THIS IS A VENDOR MACHINE WHICH HAS A BANNER OF THE CAMPAIGN, AND AS YOU CAN SEE, THIS IS JAPANESE DOMAIN NAME, AND OF COURSE FOR JAPANESE, THE JAPANESE CHARACTER STRING IS VERY -- IT'S A VERY GOOD ONE TO REMEMBER.
SO THEY INTENTIONALLY USE THE JAPANESE DOMAIN NAME FOR A CAMPAIGN. AND OF COURSE IF I CLICK THIS, IT WILL GO TO NAMACHA PAGE. IT WILL TAKE LONG, SO I WILL SKIP IT.
YES. AND BROWSING WITH MOBILE PHONES, THERE ARE MANY WEB SITE SERVICES FOR MOBILE PHONES IN JAPAN, AND THIS IS ONE OF THEM.
THIS IS TRAIN TRANSFER GUIDANCE SERVICE CALLED NORIKAI (INAUDIBLE), AND WHEN I TYPE IN NORIKAI (INAUDIBLE) DOT JP TO A MOBILE PHONE AND CLICK A BUTTON, THEN THE PAGE IS HERE.
SO THIS IS ALSO A SERVICE NAME HAD BETTER BE IN JAPANESE STREAM WHEN THE SERVICE IS PROVIDED IN JAPANESE MARKET.
SO IT IS -- I THINK THIS IS A GOOD EXAMPLE.
AND ANOTHER ONE, MY BUSINESS CARD IS LIKE THIS.
SO I THINK YOU EXPERIENCE A LOT OF TIME THAT, TOO. WHEN YOU SEE SOME BUSINESS CARD OF SOMEONE, WHETHER HE OR SHE IS STILL IN THE SECTION OR STILL HAS THE SAME TITLE AS SIX YEARS -- SIX MONTHS AGO. SO MY BUSINESS CARD IS LIKE THIS. AND IF I TOOK A PAGE OF MY BUSINESS TITLE, IT SHOULD BE THERE, YES. LIKE THIS.
SO JAPAN DOMAIN NAME IS THE SAME ONE AS MY NAME CAN BE USED LIKE A BUSINESS CARD.
OKAY.
BUT AS YOU SEE, THE OTHER SPEAKERS SAID, THE INTERNET EXPLORERS STILL DON'T HANDLE IDNS, SO THEIR QUESTIONS OR REQUESTS TO MICROSOFT, SOME OF THEM WERE ALREADY SAID BY MAO WEI, BUT I'D LIKE TO SAY WHEN WILL IDN BE SUPPORTED BY IE? I'D LIKE TO KNOW. OR THE PEOPLE LIKE TO KNOW THE ROADMAP.
AND THE SECOND, INTERNET EXPLORER ADD-ON BEFORE FULL WINDOWS IMPLEMENTATION FOR IDNS?
AS MICHEL SAID, THERE ARE MANY ISSUES THAT HAVE TO BE SOLVED TO SUPPORT IDNS IN FULL, BUT THERE MAY BE A LIGHTER WAY TO GIVE THE INTERNET EXPLORER THE IDN FUNCTION.
SO THE IDN IMPLEMENTATION BY MICROSOFT OR IDN ADD-ON BY EXISTING PLUG-INS SUCH AS I-NAV, ARE THEY GOING TO RECOMMEND IT BY MICROSOFT? EXPOSURE OF ADD-ONS IN WINDOWS MARKETPLACE OR AUTO UPDATE WITH IDN ADD-ON. ADVISING IDN ADD-ON DOWNLOAD WHEN IDN IS TRIED TO BE USED. AND ARE THERE SOME OTHER MEASURES? I'D LIKE TO ASK THEM TO MICROSOFT.
THANKS.
>>JAMES SENG: THANK YOU, HOTTA-SAN. WE ACTUALLY HAVE TWO MORE PRESENTATIONS, BUT WE'LL HOLD THAT FOR A LATER TIME AND I'M GOING TO OPEN THE FLOOR FOR QUESTIONS. YOU MAY LINE UP IF YOU HAVE QUESTIONS FOR PANELISTS. BUT IN THE MEANTIME, MICHEL, YOU HAVE A QUESTION ON THE BIG SCREEN FOR MICROSOFT SO WOULD YOU LIKE TO RESPOND TO THAT?
>>MICHEL SUIGNARD: WHEN CONSIDERING THE ROADMAPS FOR SUPPORTS, I GUESS I COULD SAY TODAY WE'RE FAIRLY COMMITTED TO PROVIDE THE IDN SOLUTION ON THE NEXT MAJOR VERSION OF WINDOWS, WHICH IS KNOWN AS (INAUDIBLE), I GUESS, OUTSIDE.
SO THAT'S, YOU KNOW, EVEN IF YOU'RE NOT COMPLETELY PUBLIC ABOUT IT, IT'S FAIR TO SAY IT'S ON THE PLAN TO DO. THERE'S ENOUGH PRESSURE OUTSIDE TO DO SOME WORK ON IT.
THE REASON WE'RE TAKING OUR TIME TO SOME DEGREE, TO SOME DEGREE WE HAVE A HIGHER VOW TO ADDRESS TO DO THEM, AND WE CAN'T BE 75% DONE WITH SOME OTHER PLUG-IN WE'RE THINKING ABOUT. SO WE HAVE TO BE 100% DONE, AT LEAST CLOSE TO IT. LIKE THE QUESTION OF PUNYCODE LEAKING, IT'S AN EXTREMELY HIGH ISSUE AND ONE OF SECURITY. SO WITH THAT IN MIND, WE STILL THINK WE HAVE ENOUGH ELEMENTS TO DO THAT. WE REALLY WANT TO HAVE IRI DONE. I'M STILL CONVINCED YOU CAN'T DO IDN IN A BROWSER IF YOU DON'T GET IRI CORRECTLY DONE. WE CANNOT AFFORD TO CHANGE BASICALLY THE WAY WE HANDLE INTERNATIONALIZED RESOURCES NOTIFIER, HAVING A VERSION AND CHANGING LATER. THAT'S NOT ACCEPTABLE. SO WITH THE IRI, IT'S A QUESTION OF WHEN NOW. SO I'M CONFIDENT WITH IRI DONE, WE CAN DO A GOOD JOB.
SO THAT DOES SOLVE THE ISSUE ON THE NEXT VERSION OF WINDOWS. THE OTHER THING IS WE NEED PLUG-INS OR ADD-ONS OR WHATEVER NAME YOU WANT TO USE FOR THOSE THINGS. FOR THAT IT'S KIND OF A MARKET PRESSURE THING.
WE HAVE DONE SOMETIMES FEATURE WORK ON SPS, KNOWN OTHERWISE AS SERVICE PACK. SO IT'S ALWAYS POSSIBLE THAT WE COULD PROVIDE SERVICE PACK, THROUGH THE SERVICE PACK, IDN FUNCTIONALITY. I DON'T REALLY HAVE -- IT'S NOT AS CLEAR THAT WE'LL DO SOMETHING THERE. IT REALLY DEPENDS. MARKET PRESSURE WILL DICTATE WHAT WE DO.
IN FACT, CONCERNING PLUG-INS, I'M KIND OF STARTING TO BE CONCERNED. WITH THE SCENARIO WHERE YOU HAVE AN UPDATED BROWSER AND YOU HAVE A PLUG-IN ON TOP OF IT, THAT COULD CREATE SOME INTERESTING SITUATION BECAUSE I WOULD SAY WE ALMOST HAVE TO DETECT THE PLUG-INS INSTALLED AND THEN -- IT'S BAD, BECAUSE IN SOME DEGREE THE PLUG-IN IS GOING TO STEAL OUR INTERFACE, AND SO WE HAVE VERY LITTLE CONTROL OF THE PLUG-IN HERE BECAUSE IT'S REALLY KIND OF INTERCEPTING THE INPUT/OUTPUT BETWEEN THE BROWSER AND THE OTHER LAYERS THAT THE BROWSER IS USING.
SO IT'S GOING TO BE A BIT CHALLENGING WHEN WE HAVE A SITUATION WHERE SOMEBODY'S UPDATING HIS BROWSER AND IS STILL TRYING TO MAINTAIN PLUG-INS INSTALLED WHICH SHOULD NOT BE THERE ANYMORE BECAUSE THE FUNCTIONALITY IS DONE BY THE BROWSER. BUT WE MAY GET REALLY SOME NASTY SITUATION IN THOSE CASE.
SO THAT'S GOING TO BE INTERESTING.
I GUESS WE SHOULD BE ABLE TO DETECT THAT THE PLUG-IN IS INSTALLED BY AT LEAST COME SOME SORT OF INVENTORY WORK ON FINDING THE PLUG-IN WAS INSTALLED, TRY TO DO SOMETHING ABOUT IT, BUT I'M NOT REALLY SURE WE CAN REMOVE IT. THERE'S SOME LICENSING AND OTHER ISSUES ABOUT BYPASSING AND SO FORTH ONCE IT IS INSTALLED.
THAT WILL BE INTERESTING. I DIDN'T REALLY THINK ABOUT THAT TOO MUCH UNTIL TODAY BUT WHEN I SEE ALL THOSE PLUG-INS, THAT'S GOING TO BE CHALLENGING.
ANYWAY. SO I HOPE -- WE'RE PRETTY CLEAR WE'RE GOING TO DO SOMETHING PRETTY QUICKLY NOW ON THE IDN FRONT.
THAT'S THE END OF IT, I GUESS.
>>JAMES SENG:HOTTA-SAN, IS THAT GOOD ENOUGH FOR YOU? IT WILL NEVER BE ENOUGH.
>>HIRO HOTTA: I WILL JUST SAY THAT THE COMPLETE STANDARDIZATION OF IRI WILL BE THE KEY FOR YOU TO -- FOR YOU TO COMPLETE YOUR IDN FUNCTIONS. WHEN WILL IT COME, WILL BE THE SECOND QUESTION. AND ARE YOU TRYING TO STANDARDIZE IRIS IN IETF?
>>MICHEL SUIGNARD: WELL, THE FACT THAT WE ARE, IN FACT, ONE OF THE EDITOR OF THE IRI SPECIFICATION SHOULD AT LEAST SHOW THAT WE HAVE SOME INTEREST IN GETTING THIS DONE. IN FACT, IRI HAS BEEN WORKED ON KIND OF A SPECIFIC STATUS. WE DIDN'T REALLY USE IETF WORKING GROUP. IT WAS DONE AS AN INDIVIDUAL CONTRIBUTION FROM MARTIN DÜRST, THAT MANY OF YOU PROBABLY KNOW. IT WAS ALSO DONE INSIDE THE W3C WORKING GROUP. TWO MAJOR CONTRIBUTORS TO THAT WERE MARTIN AND MYSELF.
WE'RE PIGGYBACKING ON THE URI REVISION ITSELF WHICH WAS BEING REVISED AT THE SAME TIME. THAT HAS ASSISTED US TO SOME DEGREE BECAUSE WE WERE COMPLETELY DEPENDENT ON THE URI BEING STABLE. WITH IT BEING UPDATED WE HAD TO FOLLOW THE PROCESS OF URI ITSELF.
THAT BEING SAID, WE ARE, IN FACT -- WE GOT A LAST CODE ISSUED I THINK IN SEPTEMBER. THAT WAS SUCCESSFUL. WE WENT TO THE IESG WHICH WAS -- THE LAST STEP IN THE PROCESS BEFORE GET TO GO AN RFC.
WE HAVE ENOUGH VOTES IN THE IESG TO BE APPROVED. IT GOES -- WE HAVE TO HAVE NO OBJECTION. YOU WE HAVE ONE DISCUSS LEFT BUT WE HAVE ENOUGH TO PASS. SO THAT'S WHY I THINK IT'S A MATTER OF DAYS. BECAUSE IN FACT THE LAST DISCUSS WAS THE FACT, A VERY SERIOUS ISSUE ABOUT THE IRI MATCHING. HOW DO YOU MAKE TWO IRI EQUIVALENT? HOW DO YOU MAKE SURE THEY'RE THE SAME THING. THERE WAS A PRETTY BIG ISSUE ABOUT THAT. I THINK WE DID THE RIGHT THING BETWEEN MARTIN AND I TO ADDRESS THE ISSUE.
SO IN MY OPINION IT'S DONE. FROM A TECHNICAL POINT OF VIEW, IT'S STABLE. WE DIDN'T CHANGE ANYTHING FOR THE PAST I WOULD SAY SIX MONTHS. IT'S BEEN MORE LIKE BRINGING CLARIFICATION TO THE DOCUMENT.
I WOULD EXPECT THE IRI TO BE A RFC IN THE NEXT -- THIS IS A STEP WE DON'T CONTROL SO I WAS GOING TO SAY IT'S GOING TO BE DONE IN A FEW DAYS, A FEW WEEKS, IN MAXIMUM. THEN IT GOES INTO THE FAMOUS RFC EDITING QUEUE, WHERE IT STAYS FOR A FEW MONTHS. IT DEPENDS HOW FAST THE RFC GETS PROCESSED BY THE RFC EDITOR AND THAT IS COMPLETELY BEYOND OUR CONTROL, AT LEAST FROM THE EDITOR POINT OF VIEW.
WE DON'T HAVE TO WAIT FOR THAT. FOR ME, FROM A VENDOR POINT OF VIEW, AS SOON AS WE ENTER THE RFC EDITING QUEUE, WE'RE FINE. WE KNOW THIS IS NOT GOING TO CHANGE ANYMORE AND WE CAN IMPLEMENT IT.
>>JAMES SENG: QUESTIONS FROM THE AUDIENCE, ANYONE? OR FOR THE PANELIST?
PLEASE COME TO THE MIKE AND INTRODUCE YOURSELF FOR THE SCRIBE.
>>PATRICIO POBLETE: HI, I'M PATRICIO POBLETE FROM CHILE, AND I'M GOING TO ASK A QUESTION OF SABINE.
YOU'RE CONSIDERING ACTUALLY DOING AN IDN. THE PROBLEM WE HAVE IS PROBABLY SOMEHOW SIMILAR TO WHAT YOU HAVE IN THAT OUR ALPHABET IS BASICALLY THE PLAIN ASCII PLUS THE THING THAT YOU PUT -- YOU CAN PUT ACUTE ACCENTS OVER VARIOUS VOWELS, FOR INSTANCE, AND THINGS LIKE THAT.
AND UP TILL NOW, PEOPLE WHOSE NAMES HAVE THOSE ACCENTS HAD TO REGISTER WITHOUT THEM. AND PEOPLE GOT USED TO OMITTING THOSE DECORATIONS ON LETTERS.
SO WHEN WE THINK OF DOING IDN IT'S KIND OF NATURAL TO CONSIDER THOSE AS VARIANTS, A LETTER WITH OR WITHOUT THE ACUTE ACCENT, THOSE A VARIANT. YOU CAN PUT THEM OR NOT PUT THEM AND IT'S ABOUT THE SAME.
IT IS NOT REALLY THE SAME, BUT MOST PEOPLE WOULD EXPECT BOTH TO WORK.
AND SO USING THOSE VARIANTS IS KIND OF A BASIC ASSUMPTION FOR US.
AND I SEE THAT YOU DIDN'T DO THAT. IS IT BECAUSE GERMAN IS VERY DIFFERENT FROM SPANISH IN THAT RESPECT OR IS THERE ANOTHER REASON THAT YOU HAVE.
>>SABINE DOLDERER: IT'S A DIFFERENT CHARACTER AND THEREFORE WE TREAT IT DIFFERENT. AND IT'S NOT ALWAYS THE SAME. IT CAN BE THE SAME, BUT IT'S NOT ALWAYS THE SAME. AND WE WOULD THINK THAT THE REGISTRANTS SHOULD HAVE TO HAVE FREEDOM TO CHOOSE.
SO IT'S BASICALLY A FREEDOM OF CHOICE WE GIVE TO THE REGISTRANTS. AND BASICALLY, WE DO THE SAME THAT THE GERMAN TRADEMARK OFFICE DOES F YOU REGISTER A TRADEMARK IN GERMAN WITHOUT THE ACCENT YOU DON'T GET THE TRADEMARK WITH THE ACCENT ADDITIONAL. SO YOU HAVE TO THINK ABOUT WHAT YOU WANT TO REGISTER. AND WE DON'T WANT TO MAKE A PERCEPTION OF WHAT PEOPLE MAY THINK OR WHAT THEY MAY WANT TO HAVE. BECAUSE WE DON'T THINK THAT'S SOMETHING THE REGISTRY SHOULD DO.
AND ADDITIONALLY, IF YOU TALK ABOUT VARIANTS, IF YOU LOOK AT THE UMLAUT, IN GERMAN IT'S TRANSCRIBED TO UE. IF YOU HAVE AN E UMLAUT COMING FROM HUNGARIA IT'S TRANSCRIBED TO U. IF YOU LOOK AT HUNGARIAN URLS, WHERE THE PEOPLE HAVE A U AND IT'S TRANSCRIBED TO UMLAUT, IF YOU HAVE GERMAN U E IS TRANSCRIBED TO UMLAUT, IT'S ALSO DIFFERENT.
WE HAVE ABOUT 70 TO 80% OF OUR REGISTRATIONS ARE PERSONAL NAMES. SO THERE ARE NAMES FROM ALL OVER EUROPE. A LOT OF PEOPLE IN GERMAN MIGHT (INAUDIBLE). SO WE HAVE BASICALLY ALL THESE VARIANTS, WE HAVE THAT BUT WE DON'T THINK IT'S ON THE REGISTRY TO DECIDE WHICH ONE IS THE DOMAIN NAME TO BE USED.
>>PATRICIO POBLETE: AND IN YOUR EXPERIENCE OF HANDLING THOUSANDS OF DOMAIN NAME REGISTRATIONS, IN ACTUAL PRACTICE DID IT HAPPEN THAT THIS CREATED AN OPPORTUNITY FOR A LOT OF TYPE (INAUDIBLE), ERASING BY ACCIDENT FROM BALLOT NAMES?
>>SABINE DOLDERER: IT'S -- I DON'T THINK IT'S -- IT'S NOT CREATING MORE TYPE OF TRAPPIST PROBLEMS THAN THE USUAL TYPE OF TRAPPIST PROBLEMS.
THERE ARE PEOPLE CONCERNS THAT THERE ARE SORT OF SIMILAR LOOKING DOMAIN NAME AS THEY ARE CONCERNED THAT THEY HAVE SIMILAR LOOKING TRADEMARKS, WHICH IS SOMETHING -- WHICH IS A REAL LIVE PROBLEM AND WHICH PEOPLE DEALT WITH FOR YEARS.
SO WE DON'T THINK THAT IT'S REALLY -- WE SHOULD GO FOR A SINGLE-TYPE SOLUTION IN THE DOMAIN NAME WHERE THERE IS NO SINGLE SOLUTION, EVEN IN THE TRADEMARK SPACE WHERE THESE PROBLEMS ARE ACTUALLY WELL-KNOWN SINCE YEARS, AND WHERE THE RESULTS ARE NO EASY SOLUTION ON THAT.
SO WE DON'T THINK WE SHOULD INVENT A WHEEL WHERE THERE IS NO WHEEL POSSIBLE.
>>>: OKAY. THANKS.
>>JAMES SENG: QUESTION FROM THE AUDIENCE? I'D REALLY LIKE TO HAVE MORE PARTICIPATION.
HMM. I HAVE A QUESTION FOR MICHEL. YOU SAID SOMETHING ABOUT NEXT VERSION OF WINDOWS WHICH I PRESUME YOU MEAN LONG HORN, AND LONG HORN IS -- WHAT? SCHEDULED IN 2006. SO ARE YOU SAYING TO THE AUDIENCE YOU EXPECT SUPPORT FOR WINDOWS ONLY IN 2006?
>>MICHEL SUIGNARD: WELL, I'M NOT GOING TO COMMENT ON WHEN THE THING IS SHIPPING. I'M NOT THE RIGHT PERSON TO ANSWER THAT. I THINK IT'S FAIR TO SAY THAT YOU'RE RIGHT.
BUT OBVIOUSLY THERE WILL BE -- EVEN IF THAT'S DONE IN 2006, THERE WILL BE A BETTER VERSION AVAILABLE BEFORE THAT. SO MANY PEOPLE WILL BE ABLE TO VALIDATE, YOU KNOW, THE SUPPORT AND BASICALLY MAKE SURE IT DOES WHAT THEY EXPECT.
IT TAKES A WHILE.
LIKE I SAID, WE HAVE A PLAN TO -- WE'RE OBVIOUSLY LOOKING AT THE EXISTING VERSION OF WINDOWS AND POSSIBLY DOING SP, SERVICE PACK, UPDATES. BUT IT'S ONLY FAIR I'M PRETTY SURE WE'LL DO IT FOR LONG HORN. YES, THAT MAY BE IN 2006, I UNDERSTAND.
BUT IN THE MEANTIME, YOU HAVE TO REMEMBER THAT THERE'S A LOT OF PLUG-INS OUT THERE. WE SEE THE COMPETITION IS NOT WAITING FOR US. SO THERE ARE SOLUTIONS OUT THERE, I THINK. EVEN IF NOT AS CONVENIENT AS BEING DIRECTLY SUPPORTED IN THE OPERATING SYSTEM.
>>JAMES SENG: ANY QUESTION FROM THE PANELISTS?
WILLIAM. FOR YOUR OPEN SOURCE, HOW DO YOU THINK -- YOU HAVE DONE A PRESENTATION TODAY. I KNOW THERE'S A (INAUDIBLE) OF INTEREST, SO I LET ME NOTE I DID HELP TO START THE IDN OPEN SOURCE HERE, BUT SINCE YOU HAVE BEEN DRIVING THE IDN OPEN SOURCE FOR THE LAST YEAR WITHOUT MUCH INTERACTION FROM ME, WHAT CAN WE DO TO HELP MORE THE IDN OPEN SOURCE DEVELOPMENT, IN YOUR OPINION?
>>WILLIAM TAN: I WOULD SAY THAT JUST BY USING IT, BY TESTING IT, BY ALSO GETTING INVOLVED IN WAYS SUCH AS SUPPORTING OTHER USERS. WE'RE GOING TO SET UP A FORUM AS WELL FOR SUPPORT SO WE JUST GET THE WHOLE THING GOING AGAIN.
THERE HAVE BEEN SOME LULL PERIODS IN THE PROJECT, BUT REST ASSURED THAT WE'RE GETTING IT UP AND RUNNING AGAIN.
WE ARE LOOKING AT GETTING A VERSION OUT SOMETIME WITHIN THE NEXT MONTH.
SO YEAH, JUST BASICALLY AS LONG AS PEOPLE START USING IT AND PROMOTING IT, I'M SURE IT WILL JUST GAIN MOMENTUM AND TAKE OFF.
>>JAMES SENG: LET ME GIVE SOME AIR TIME TO EDMON. EDMON IS FROM AFILIAS. DO YOU HAVE ANY CLIENT APPLICATIONS OR DO YOU WANT TO TALK ABOUT IT?
>>EDMON CHUNG: NO. AFILIAS ITSELF IS NOT WORKING ON A SPECIFIC IDN CLIENT FOR ANY BROWSERS.
>>JAMES SENG: MAO WEI, YOU MENTIONED ABOUT IDN IS E-MAIL CLIENTS, DO YOU SEE -- I UNDERSTAND YOU ACTUALLY HAVE A SERVICE, YOU HAVE OXYGEN.CN.CN WHICH IS CHINESE E-MAIL SERVICE. DO YOU MIND TELLING ME IS THERE A LOT OF DEMAND ON E-MAIL SERVICES? HOW IS THE DEMAND ON CHINESE E-MAIL? CHINESE E-MAIL, WHAT IS THE DEMAND ON THE CHINESE E-MAIL SERVICES?
>>MAO WEI: WE HAVE OUR SOLUTION FOR CHINESE E-MAIL BUT IT'S NOT A STANDARD.
SO THE E-MAIL SERVICE PROVIDER DON'T -- A LOT OF THEM DON'T TAKE THIS SOLUTION.
SO WE CONTACT OUR REGISTRAR AND, OURSELVES, WE PROVIDED THE SERVICE. SO IT'S LIMITED. NOT TOO MANY PEOPLE USE IT.
>>JAMES SENG: ANY MORE QUESTIONS FROM THE AUDIENCE? IF NOT, WE HAVE TWO MORE PRESENTATION. OKAY. GREAT.
YES, PLEASE.
>>ROOZBEH POURNADER: MOSTLY THIS IS SOMETHING THAT HAS BEEN GOING ON WITH IT. WE ALREADY HAD A TALK ABOUT THE GERMAN POSSIBLE CONFLICT ISSUES AND THE TRADEMARK PROBLEMS. I HAVE EXAMPLES OF THAT HERE. WOULD IT BE POSSIBLE FOR ME TO PLUG IN SOMEWHERE AND SHOW IT TO YOU?
THERE WAS SOME LINE THERE.
>>JAMES SENG: GO TO THE LECTERN.
>>ROOZBEH POURNADER: SO SOMEBODY IS -- OH, OKAY.
SO AS YOU SEE HERE, IT'S REALLY -- THE FIRST COLUMN, YOU SEE A SMALL LATIN CHARACTER. IN THE SECOND COLUMN, YOU SEE ITS CAPITAL VERSION. AND THE THIRD IS THE UNICODE NAME. ALL OF THESE THREE CHARACTERS ARE LATIN, AND YOU ACTUALLY ARE ALLOWING THE FIRST TWO TO BE REGISTERED.
I REALLY HONESTLY WONDERED ABOUT THE SITUATION, THAT PEOPLE WILL DO -- THEY MAY TRY TO DO SOME SCRIBING SCRIPTS WITH THE CAPITAL D'S THAT LOOK EXACTLY THE SAME IN EVERY CASE. AND THAT'S SOMETHING, I HAVE SOME ANSWERS FOR YOU AND WHILE YOU'RE STILL ALLOWING BOTH OF THESE IN THE DOT DE DOMAIN NAMES. BUT I LEARNED SOMETHING FROM YOU, A DIFFERENT ISSUE.
AUTOMATICALLY, WHEN PEOPLE LIKE US TRY TO DO THAT, WE SAY OKAY, THEN YOU MUST HAVE BUNDLED, IF YOU WANT TO ALLOW BOTH OF THESE CHARACTERS. BECAUSE IT WILL CREATE A PROBLEM.
BUT YOU HAD BETTER IDEAS. I MEAN YOU HAD SOME INTERESTING POINTS ABOUT NOT USING BUNDLES.
SO WHAT I WOULD REALLY LIKE TO ASK YOU IS WOULD YOU PLEASE EXPLAIN THEM TO THE WHOLE AUDIENCE.
>>SABINE DOLDERER: ACTUALLY, THEY'RE STILL DIFFERENT -- SORRY.
ACTUALLY, BOTH OF THEM ARE STILL A DIFFERENT CHARACTER. AND IF YOU -- I THINK MICHEL HAS DONE SIMILAR EXAMPLES ABOUT THAT, THAT THERE IS ALWAYS A -- WE USE BASICALLY ONLY LATIN CHARACTER SET AND WE DON'T MIX-UP DIFFERENT SCRIPTS. SO I SHOULD BE CLEAR ABOUT THAT. AND I MAY -- IT'S MAYBE POSSIBLE THAT IF WE START TO REGISTER ARABIC SCRIPTS, WHICH I HARDLY WOULD EXPECT THAT WE DO SO, THAT WE DON'T MIX THEM UP.
BUT I THINK THERE IS ALSO THE POSSIBILITY THAT PEOPLE ACTUALLY ARE -- YOU HAVE TO LOOK -- IF YOU ARE LOOKING ON THESE TWO CHARACTERS AND SAY PEOPLE CAN ACTUALLY GET CONFUSED ABOUT THAT. I UNDERSTAND THAT YOU ARE CONFUSED ABOUT THAT BECAUSE THAT'S SOMETHING WHICH EVEN YOU AS A TECHIE CAN'T REALLY PREVENT TO UNDERSTAND WHAT'S CLEARLY BEHIND THE SCENE.
BUT USUALLY PEOPLE ARE, MUCH EARLIER, THEY ARE CONFUSED. IF SOMEBODY HAS A DEUTSCHE BANK AND ANOTHER ONE HAS DEUTSCHE-BANK WITH A DASH IN BETWEEN, HE IS ALSO CONFUSED IF HE DOESN'T GET ON THE SAME WEB SITE. SO WE ACTUALLY HAVE TO EDUCATE THE PEOPLE. WE HAVE TO EDUCATE THE PEOPLE THAT REALLY A DOMAIN NAME IS NOTHING WHICH IS -- BEHIND THE DOMAIN NAME IS NOT ALWAYS WHAT YOU EXPECT THAT IS BEHIND IT. AND THAT'S NOTHING TO DO WITH IDN OR NOT IDN, BUT IT HAS SOMETHING TO DO WITH CONTENT OR WITH MEANING AND WITH EDUCATION OF THE CUSTOMERS.
AND ALSO, SOMETHING VERY CLEAR, WHAT MICHEL SAID, THAT WE CAN ONLY CONTROL THE SECOND LEVEL. WE HAVE NO CONTROL AT ALL ABOUT THE THIRD LEVEL.
SO IF SOMEBODY STARTS TO MIX ALL THESE SCRIPTS UP ON THE THIRD LEVEL, THERE IS NO INFLUENCE WE HAVE ON THAT.
SO ACTUALLY WHAT WE HAVE TO DO AND WHAT WE HAVE TO LOOK FORWARD IS THAT WE PUT TRUST AND SECURITY AND AUTHENTICATION IN THE APPLICATIONS, BUT I DON'T THINK THAT IT'S VERY USEFUL TO PUT IT ON THE PROTOCOL LEVEL, WHICH IS SIMPLY THERE TO ADDRESS HOSTS OR SERVICES. BECAUSE I THINK TO SOLVE THERE ALL THE PROBLEMS WHICH ARE REAL PROBLEMS IS HARDLY IMPOSSIBLE.
>>JAMES SENG: OKAY.
ANY MORE QUESTION FROM THE AUDIENCE?
>>MICHEL SUIGNARD: DO YOU WANT ME TO ANSWER?
>>JAMES SENG: PLEASE.
YOU ARE THE PANEL, SO FEEL FREE TO CHIP IN.
>>MICHEL SUIGNARD: IT'S VERY IMPORTANT THAT IF YOU WANT TO BE SECURE ON THAT DOMAIN, I WOULD SAY NO WAY AROUND USING CERTIFICATES.
IT'S LIKE WHAT VALENTIN WAS SHOWING BEFORE.
IF YOU REALLY WANT TO BE TRUSTED, YOU WILL HAVE TO USE, YOU KNOW, SOME SSL OR SOMETHING TO REALLY PHYSICALLY CERTIFY THAT YOU ARE REALLY WHAT YOU PRETEND TO BE.
THERE'S REALLY NO -- IT MAKES IN FACT THE PRIME MORE ACUTE FOR IDN THAN IT WAS BEFORE.
BUT I THINK WE WILL SEE MUCH MORE SECURE AUTHENTICATED WEB SITE WITH IDN, BECAUSE THAT'S GOING TO BE A NEED PRETTY QUICKLY.
BECAUSE ANY -- ANY MAJOR BRAND TODAY CAN BE PRETTY MUCH SPOOFED RIGHT AWAY IN IDN.
WE OBVIOUSLY EXPECT MOST OF THE, YOU KNOW, REGISTRY TO LOOK FOR IT SO THEY DON'T, YOU KNOW -- LIKE THE MAJOR ONE, LIKE WE SEE FROM DOT-COM, BUT WE SEE ALSO THE MAJORCCTLD TO AT LEAST DO SOMETHING AT THE FIRST LEVEL.
OTHERWISE, IT'S GOING TO BE A ZOO OUT THERE, YOU KNOW.

>>JAMES SENG: THAT'S VERY INTERESTING.
BECAUSE I HEAR A VERY DIFFERENT ANSWER FROM SABINE.
>>SABINE DOLDERER: NO, NOT REALLY.
>>JAMES SENG: NOT REALLY.
FOR SABINE, DO YOU FEEL THAT CCTLDS SHOULD NOT TRY --
>>SABINE DOLDERER: ACTUALLY, WE DON'T DO TRADEMARK PROTECTION.
BUT WE ASSIST TRADEMARK HOLDERS TO HAVE A SORT OF VERY CLEAR UNDERSTANDING WHO HAS REGISTERED WHICH DOMAIN NAME, AND WE ACTUALLY ASK -- AT THE MOMENT, WE DON'T MIX SCRIPTS, DIFFERENT SCRIPTS.
AND I WOULD -- I UNDERSTAND THAT THAT MIGHT BE REALLY NOT THE IDEAL IDEA TO GO AHEAD WITH THAT.
BUT I THINK STILL THEN THERE'S A LOT OF POSSIBILITIES TO CONFUSE CUSTOMERS AND TO SHOW THEM THINGS WHICH ARE NOT REAL.
AND YOU HAVE -- YOU REALLY HAVE TO TEACH THEM.
IF SOMEBODY'S SELLING SHOES WITH A TRADEMARK LOGO ON THE SHOE, THERE IS NO REGISTRY WHO COULD CERTIFY THAT THE TRADEMARK LOGO ON THE SHOE IS NOT FROM THE FAMOUS VENDOR.
AND ONLY BECAUSE WE HAVE IN THE DNS A CENTRAL POINT WHERE ALL THESE REGISTRIES ARE, YOU HAVE TO UNDERSTAND THAT STILL A LOT OF THE PROBLEMS ARE REAL PROBLEMS WHICH HAVE TO BE DONE, TO SOME EXTENT, BY THE ONE WHO IS PROTECTING HIS TRADEMARK.

>>WILLIAM TAN: I THINK THAT THERE WAS A VERY VALID POINT THERE.
I WOULD SAY WHY DON'T YOU, SAY, HAVE -- SAY, RECOMMEND TO THE REGISTRANTS THAT THEIR DOMAIN NAME MAY HAVE POSSIBLE CONFUSIONS AND LIST THE VARIANTS, EVEN IF YOU DON'T DECIDE TO BUNDLE THEM.
LIKE I THINK MUSEUM DOES THAT PROBABLY.
JUST LIST THE VARIANTS INSTEAD OF BUNDLING THEM IF YOU DECIDE NOT TO DO, SO --
>>SABINE DOLDERER: IF YOU'LL LOOK AT THE POSSIBLE -- THE PROBLEM IS I CAN'T SEND TO EACH AND EVERYBODY REGISTERING A DOMAIN WITH UE THAT IT'S POSSIBLY -- THAT THERE IS A POSSIBLE CONFLICT.
BECAUSE IF I LOOK AT THE GERMAN WORDINGS, THERE IS BAUER, MAUER, WHERE THERE IS NO CLEAR UNDERSTANDING ABOUT THAT.
THERE IS NO U UMLAUT POSSIBLE CONFUSING BECAUSE IT'S SPELLED COMPLETELY DIFFERENT.
AND NOBODY WOULD EXPECT IT'S SIMILAR.
SO, ACTUALLY, IF YOU LOOK AT THE UES IN THE GERMAN DICTIONARY, I WOULD EXPECT IF YOU -- I WOULD EXPECT YOU'D FIND A LOT OF THEM AND THEY ARE NATURALLY NOT SUBSTITUTING SOMETHING ELSE.
>>WILLIAM TAN: YEAH, I UNDERSTAND THAT.
BUT I BELIEVE, LIKE, FORGERMAN CHARACTERS AND THE GERMANS, YOUR GERMAN REGISTRANTS PROBABLY DO UNDERSTAND THE -- LIKE, THE AUTHENTICATIONS OF REGISTRATION AND REGISTERING THEIR IDN NAMES.
BUT ON EXAMPLES JUST SUCH AS SHOWN JUST NOW, PROBABLY, YOU KNOW, IT MIGHT BE USEFUL FOR PEOPLE TO KNOW THAT THERE ARE POSSIBLE CONFUSIONS.
>>SABINE DOLDERER: ACTUALLY, I'M SURE THAT NO ONE EVER HAVE ACTUALLY -- THE PROBLEM IS, WE ARE OFFERING THESE LATIN SCRIPTS.
BUT I WOULD HARDLY EXPECT THAT A LOT OF PEOPLE ARE REALLY REGISTERING.
BUT WE ACTUALLY DON'T WANT TO DRAW A LINE WHERE WE SAY A CERTAIN CHARACTER WHICH IS USED 20 TIMES IS NOT TO BE REGISTERED.
BUT, BASICALLY, IF YOU -- I CAN SHOW YOU TABLES OF STATISTICS, THE MOST-USED CHARACTERS ARE DEFINITELY A U AND O UMLAUT.
BUT, ADDITIONALLY, THE E WITH THE (INAUDIBLE) IS MORE OFTEN USED THAN I THINK THE Y AND X.
AND ALSO THE SPANISH CHARACTER, I WITH A SORT OF -- WHATEVER ON IT.
WHICH IS ALSO MORE OFTEN USED THAN SOME OF THE -- NOW -- STILL -- NOW, MORE OFTEN USED THAN THE X AND THE Y FROM THE ASCII ALPHABET.
SO THERE IS A NEED FOR ADDITIONAL CHARACTERS BESIDES THE GERMAN UMLAUTS.
BUT I'M NOT SURE IF WE WILL SEE ALL OF THEM WORKING IN THE -- WORKING COMING UP AND POPULATED.
ON THE OTHER HAND SIDE, WHERE YOU SEE THIS CONFUSION AND THESE CONFLICTS, THEY ARE VERY, VERY STRANGE CHARACTERS.
I WOULDN'T EXPECT A MAJOR TRADEMARK TO HAVE THESE CHARACTERS IN THEM, AND I WOULDN'T EXPECT SOMEBODY TO BUY ON THIS WEB SITE.
IT'S SOMEBODY LOOKING FOR A PHISHING TARGET.
IT'S A VERY EXOTIC PROBLEM WITH AN EXOTIC USE.
>>JAMES SENG: I THINK THE ISSUE IS NOT JUST A TRADEMARK PROBLEM.
A TRADEMARK IS OF A LESSER CONCERN COMPARED TO ACTUAL SECURITY WHERE ONE SITE COULD POTENTIALLY PRETEND OR LOOKS LIKE ANOTHER SITE, WHICH I THINK IS WHAT MICHEL IS TRYING TO BRING ACROSS, WHICH IS THE BIG SECURITY CONCERN FOR SOME OF THEIR BIGGER CUSTOMERS.
>>SABINE DOLDERER: YEAH, BUT AN O AND A ZERO CAN ALSO LOOK SIMILAR.
AND I'M NOT SURE THAT MICROSOFT PROBLEMS WILL POP UP EVERY TIME WHEN AN O IS PUT IN THAT, HERE, SHOW ME, THERE CAN BE A ZERO.
SO THERE IS A CERTAIN AMOUNT OF PEOPLE HAVE TO KNOW WHAT THEY ARE TYPING IN.
THEY SHOULD BE AWARE THAT NOT EVERY WEB SITE IS SAFE. AND JUST COME BACK TO REAL LIFE, IF I GO TO SOMEBODY ON THE STREET AND HE SAYS, "I SELL YOU SOMETHING, A HOUSE OR SO," I ALSO DON'T TRUST HIM, BUT ASK FOR CERTIFICATES, ASK FOR FURTHER ASSURANCE.
AND I THINK THAT'S THE MUCH MORE IMPORTANT ISSUE, TO EDUCATE CUSTOMERS, TO GIVE THEM POSSIBILITIES TO BE SURE, INSTEAD OF COME UP WITH A DNS WHERE PEOPLE SAY IT MIGHT BE SURE, BUT THERE IS STILL A LOT OF POTENTIAL CONFLICTS.
>>MICHEL SUIGNARD: WELL, YOU KNOW, YOU SHOULD NEVER CONDUCT A TRANSACTION, A FINANCIAL TRANSACTION ON A NONSECURED, YOU KNOW, WEB SITE ANYWAY.
SO IF YOU'RE DOING IT, MAYBE THAT'S A QUESTION OF EDUCATING THE -- YOU KNOW, THE CUSTOMERS, THAT YOU NEVER, YOU KNOW, NEVER, NEVER TRUST A SITE THAT IS NOT CERTIFIED FOR FINANCIAL TRANSACTION.
OTHERWISE, YOU'RE JUST ASKING FOR TROUBLE.
ALREADY TODAY, ON -- IDN IS MAKING THE ISSUE MORE IMPORTANT.
BUT IT ALREADY EXISTS TODAY.
SO -- YEAH, I MEAN, CERTIFICATION IS A BIG ISSUE HERE.
AND YOU HAVE TO BE AWARE OF IT.
BUT IT'S NOT REALLY ANYTHING NEW.
IT'S JUST MAKING IT MORE ACUTE.

>>JAMES SENG: THANK YOU.
OKAY.
>>CARY KARP:CERTAINLY IN THE GERMAN SITUATION, YOU'RE AWARE JUST HOW DYNAMIC ORTHOGRAPHY IS.
CURRENT PRACTICE IS CURRENT RIGHT NOW.
IT'S NOT NECESSARILY GOING TO BE CURRENT TOMORROW.
BUT THE CCTLD REGISTRIES ARE IN A UNIQUE POSITION TO TRACK THESE CHANGES.
AND IF THERE WAS SOME SORT OF A COORDINATED EFFORT NOT TO ELIMINATE ALL CONFUSION, BUT AT LEAST TO EXPLAIN THE POTENTIAL SOURCES OF CONFUSION, WHAT ORTHOGRAPHIC EQUIVALENCES ARE THERE BETWEEN ADJACENT LANGUAGES?
WHAT POSSIBILITY IS THERE FOR -- WELL, I SUPPOSE DELIBERATE MISUSAGE.
IF YOU'RE EDUCATING THE WELL-INTENDED USERS, YOU'RE ALSO EDUCATING THE PEOPLE WHO WOULD BE DELIBERATELY DECEPTIVE.
BUT I THINK IN ALL OF THIS, THERE'S SOMETHING MISSING, AND THAT IS SOME COORDINATED METHOD OF EXPRESSING THE REQUIREMENTS OF INDIVIDUAL LANGUAGES.
THERE'S THE CHARACTER REALM, THE ALPHABETIC REALM.
AND EVERYBODY IS WORRIED ABOUT THEIR OWN REGISTRY AND THEIR OWN CLIENT BASE.
BUT I WONDER IF OUR JOINT PURPOSES MIGHT NOT BE SERVED IF THERE WERE, AGAIN, SOME REPOSITORY SOMEWHERE WHERE CURRENT INFORMATION RELEVANT TO THE FORMATION AND REGISTRATION OF IDN MIGHT NOT BE AVAILABLE.
AND YOU KNOW ALL YOU COULD POSSIBLY -- ALL THERE IS TO KNOW ABOUT GERMAN.
THERE ARE PROBABLY PEOPLE IN THIS ROOM WHO WOULD REALLY BENEFIT FROM AT LEAST SOME PORTION OF THAT KNOWLEDGE.
AND THE SAME ATTACHES TO EVERY SINGLE LANGUAGE REPRESENTED HERE AND EVERY SINGLE LANGUAGE THAT'S GOING TO BE APPEARING IN IDN SPACE.

>>SABINE DOLDERER: I THINK WE DO TRY AN EFFORT.
SO WHAT I HAVE PRESENTED WAS NOT ONLY THE GERMAN WAY, BUT IT WAS ACTUALLY THE WAY THE SWISS, THE AUSTRIAN, AND OUR REGISTRY MOVED FORWARD.
AND WE -- THERE'S ALSO A CENTRAL WEB SITE WHERE WE TRY TO PUT A LOT OF THESE ISSUES TOGETHER.
BUT I HAVE TO ADMIT, WE ARE NO LINGUISTS.
SO WE ARE AWARE OF A LOT OF CURIOSITIES IN LANGUAGES, BUT WE ARE NO LINGUISTS.
SO WE WILL NOT BECOME SOME.
SO WHAT WE CAN -- WHAT WE CAN -- WE EDUCATE PEOPLE.
SURELY WE DO THAT.
WE GIVE A LOT OF INFORMATION TO OUR CUSTOMERS, DEFINITELY.
WE -- IF YOU GO ON OUR WEB SITE, YOU SEE THAT THERE IS A LOT OF INFORMATION AVAILABLE ABOUT THE DOMAIN NAME AND THE FRAMEWORK.
BUT IT HAS TO EVALUATE -- I DON'T THINK WE CAN REALLY -- THERE IS NO FINAL KNOWLEDGE BASE YET.
>>CARY KARP: WHAT I'M SUGGEST SOMETHING THAT THERE MIGHT BE SOME AGENCY, EITHER CURRENTLY ACTIVE WITHIN THE IDN MOVEMENT, TO COIN A TERM, I SUPPOSE, OR PERIPHERAL TO IT, THAT FRUITFULLY CAN COORDINATE THIS.
WE'VE HEARD REFERENCE TO PROFESSIONAL LINGUISTS DURING NUMEROUS PRESENTATIONS TODAY.
THIS IS EXACTLY THE SORT OF THING YOU ARE TALKING ABOUT.
THE RAW KNOWLEDGE IS ALL IN OUR POSSESSION.
BUT HOW TO EXPRESS THIS IN A GENERALIZED MANNER AND WHERE TO EXPRESS IT IS ONE OF THE INGREDIENTS THAT'S STILL MISSING.
SO I SUSPECT THAT THERE MAY BE SOME, AGAIN, AGENCY THAT HAS YET TO MANIFEST A CLEAR PRESENCE IN THIS PROCESS THAT COULD VERY FRUITFULLY ASSIST US ALL AND THE USER COMMUNITY BY HELPING US ENCODE THIS INFORMATION IN A MANNER THAT IS GLOBALLY USEFUL.
>>SABINE DOLDERER: IT CREATES A LOT OF WORK.
WHEN WE TALKED TO THE GERMAN BIBLIOTEK ON THAT AND ON DIFFERENT ISSUES, THEY SAY THAT'S A GREAT QUESTION.
IT'S AT THE MOMENT STANDARDIZATION, DO YOU WANT TO HELP US ASSIST WITH THAT.
>>JOHN KLENSIN: LET ME MAKE A PERSONAL OBSERVATION ABOUT THIS, WITH THE UNDERSTANDING THAT IT'S REALLY JUST A PERSONAL OBSERVATION.
IT SEEMS TO ME THAT IN EACH OF THESE CASES, WHAT ONE IS DEALING WITH IS A TRADEOFF BETWEEN ELIMINATING SOME PROBLEMATIC STUFF AND INTRODUCING SOME ADDITIONAL COMPLEXITY WITH THE VARIANT SYSTEM AND THROWING MORE WEIGHT ON THE USERS AND THE TRADEMARK ENFORCERS WITHOUT THOSE KINDS OF SYSTEMS.
AND WHILE, FOR EXAMPLE, I THINK WITH SOME CONSTRUCTIVE DIFFERENCES ON HOW VARIANTS WORK, I THINK ONE COULD HANDLE THE UMLAUT VERSUS UE WITH THE VARIANT SYSTEM.
THAT DOESN'T MEAN ONE SHOULD.
AND IT SEEMS TO ME THAT MUCH AS I'D LOVE TO HAVE A PERFECT SOLUTION DICTATED BY SOMEBODY WITH PERFECT KNOWLEDGE OF SCRIPTS, THAT THE REALITY HERE IS GOING TO HAVE TO BE REGISTRY-BY-REGISTRY DECISIONS ABOUT WHAT THE RIGHT THING IS TO DO, GIVEN REGISTRY EXPERIENCE AND REGISTRY KNOWLEDGE ABOUT WHAT'S GOING TO WORK IN THAT PARTICULAR ENVIRONMENT.
AND, YOU KNOW, I WISH IT WERE EASIER.
BUT I CAN'T -- MY GUT TELLS ME THAT IF I WERE DOING GERMAN, I'D LIKE TO SEE VARIANTS.
BUT I HAVE TO ASSUME THAT THE DENIC DECISION IS RIGHT FOR DENIC.
>>JAMES SENG: TOTALLY AGREE WITH JOHN.
I DON'T LIKE WHAT'S GOING ON, BUT I JUST PRESUME YOU HAVE DONE YOUR DUE DILIGENCE.
OKAY.
I THINK WE NEED TO MOVE ON.
WE HAVE TWO MORE PRESENTATIONS.
SO, EDMON, ARE YOU READY?
EDMON IS GOING TO TALK ABOUT IDN OVER EPP.
APPARENTLY HE HAS SOME ISSUES ABOUT IDN-OVER-EPP, AND HE WOULD LIKE TO HAVE SOME DISCUSSION.
>>EDMON CHUNG: GOOD AFTERNOON, EVERYONE.
SO MY NAME IS EDMON CHUNG FROM AFILIAS.
I'D LIKE TO SHIFT GEAR A LITTLE BIT, TALK ABOUT ANOTHER TYPE OF APPLICATION, REALLY, FOR IDNS.
AND THAT HAS TO DO WITH EPP.
AND EPP IS AN ACRONYM FOR EXTENSIBLE PROVISIONING PROTOCOL, PROBABLY MOST -- A LOT OF PEOPLE AROUND HERE WOULD KNOW ABOUT IT.
I THINK IT'S SLIGHTLY DIFFERENT THAN RESOLUTION OF DOMAIN, BUT IT IS REALLY PERHAPS EVEN MORE RELEVANT, DIRECTLY RELEVANT, TO THE COMMUNITY HERE.
AND ONE OF THE THINGS THAT IS IMPORTANT IS THAT IT IS BEING ADOPTED BY ICANN AND ADOPTED BY A LOT OF THE GTLDS, ESPECIALLY THE NEW ONES.
AND WHEN WE TALK ABOUT IDNS AND THERE ARE CERTAIN ELEMENTS, FOR EXAMPLE, WE'VE TALKED ABOUT THE VARIANTS ISSUES AND LANGUAGE ISSUES THAT MAY REQUIRE ADDITIONAL INFORMATION TO BE PASSED BETWEEN REGISTRARS AND REGISTRIES, SO THIS IS BASICALLY THE TOPIC.
SOME KEY RELEVANT, I'LL ALSO GO QUICKLY.
THE PUBLISHING OF THE PROPOSED STANDARD RFCS, THERE'S THE IESG IDN STATEMENT, FOLLOWED BY THE ICANN IDN IMPLEMENTATION GUIDELINES, AS WELL AS THE ESTABLISHMENT OF THE IANA IDN LANGUAGE-TABLE REGISTRY.
AND RECENTLY, IN AN IETF MEETING, THERE WAS AN IDN-OVER-EPP BOF THAT WAS HELD TO TALK ABOUT THIS SPECIFIC TOPIC.
TWO PARTICULAR RESULTS CAME OUT FROM THE BOF SESSION.
THE FIRST ONE, THERE WAS A CONCERN THAT THERE WAS NO CLEAR DEMAND FROM THE INDUSTRY, THAT IS, THE REGISTRIES AND REGISTRARS, FOR A STANDARDIZED FRAMEWORK FOR IDN-OVER-EPP.
AT THE SESSION, SPECIFICALLY, THERE WASN'T REALLY THAT SENSE.
AND THE OTHER ITEM THAT CAME OUT WAS THAT IT WAS NOT CLEAR WHETHER THE ICANN IDN GUIDELINES HAVE ALREADY GIVEN ENOUGH DIRECTION FOR A COMMON SET OF ELEMENTS TO ESTABLISH A STANDARDIZED FRAMEWORK.
MEANING THAT WHETHER THERE ARE ENOUGH MEANINGFUL SET OF ELEMENTS TO ADD AS A STANDARDIZED SET, BECAUSE EACH REGISTRY ITSELF MAY HAVE A SLIGHTLY DIFFERENT POLICY, IS IT MEANINGFUL AT ALL TO CREATE A STANDARDIZED FRAMEWORK.
TWO PARTICULAR OBSERVATIONS, REALLY, IS THAT I BELIEVE THERE IS A RATIONAL, REASONABLE, REALLY, DEMAND FOR HAVING A STANDARDIZED FRAMEWORK, AND THEN THAT THE ICANN IDN GUIDELINES CURRENTLY ALREADY PROVIDE CERTAIN COMMON ELEMENTS FOR A MEANINGFUL FRAMEWORK.
TALKING ABOUT THE -- THE RATIONALE BEHIND IT, IDN IS LIKELY GOING TO BECOME ONE OF THE CORE OFFERING FOR MOST DOMAIN REGISTRIES.
AND EPP WAS DEVELOPED SPECIFICALLY TO ALLOW FOR A STANDARDIZED FRAMEWORK BETWEEN REGISTRARS AND REGISTRIES TO PROMOTE INTEROPERABILITY.
AND IT'S BEING ADOPTED BY THE ICANN COMMUNITY.
AND ONE OF THE CURRENT SITUATION IS THAT THERE IS A DIVERGING SITUATION WHERE THE DIFFERENT REGISTRIES ARE CREATING DIFFERENT IDN EXTENSIONS TO EPP TO ALLOW FOR DOMAIN REGISTRATION.
FOR EXAMPLE, AFILIAS, OURSELVES, HAVE A CERTAIN APPROACH.
THERE ARE DIFFERENT APPROACHES, I'M SURE, FOR WHEN NEUSTAR STAR LOOKS AT IT AND FOR DOT MUSEUM.
THE CURRENT IDN -- ICANN IDN GUIDELINES DOES PROVIDE SOME SORT OF COMMON FRAMEWORK ALREADY AS FAR AS THE LANGUAGE GOES THAT I CAN OBSERVE.
ONE, IT SPECIFICALLY SAYS THAT, YOU KNOW, TO RECOMMEND THE ASSOCIATION OF EACH REGISTERED IDN WITH ONE LANGUAGE OR A SET OF LANGUAGES.
THE OTHER BEING THAT EMPLOY LANGUAGE-SPECIFIC REGISTRATION AND ADMINISTRATION RULES SUCH AS THE RESERVATION OF DOMAINS WITH EQUIVALENT CHARACTER VARIANTS, THE VARIANT ISSUE OF BUNDLING THAT WAS MENTIONED.
SO DERIVING FROM THAT, THERE ARE REALLY TWO SETS OF COMMON ELEMENTS THAT CAN MAKE A MEANINGFUL FRAMEWORK FOR STANDARDIZED IDN EXTENSIONS, THAT IS, THE PROVISIONING OF LANGUAGE TAGS AS WELL AS THE PROVISIONING OF VARIANTS FOR EACH PARTICULAR DOMAIN.
SOME IMPORTANT CONSIDERATIONS, IDN OVER EPP, IN FACT, IN MY VIEW, PROMOTES THE ADOPTION OF IDN AS WELL, BECAUSE IT ENCOURAGES INTEROPERABILITY BETWEEN REGISTRARS AND REGISTRIES AND PROMOTES SALES, BASICALLY, OF IDNS, WHICH IN TURN CREATES THE MARKET PRESSURE WHICH MICHEL TALKED ABOUT, HAVING THE MORE REGISTERED IDNS THERE ARE, THE BIGGER THE MARKET PRESSURE THERE WOULD BE FOR APPLICATION PROVIDERS TO INCORPORATE IDN FUNCTIONALITIES.
THERE CURRENTLY IS A DIVERGENT TREND.
THERE ARE DIFFERENT APPROACHES TO CREATE IDN EXTENSIONS TO EPP.
AND IT MAY BASICALLY, YOU KNOW, DEFEAT THE PURPOSE, THE ORIGINAL PURPOSE OF HAVING EPP AS AN INTEROPERABLE FRAMEWORK FOR TRANSACTIONS BETWEEN REGISTRARS AND REGISTRIES.
SO THIS PRESENTATION REALLY IS A CALL FOR PARTICIPATION AND SUPPORT FROM THE COMMUNITY TO WORK ON LOOKING AT WHAT ARE THE COMMON ELEMENTS AND WHAT IS THE FRAMEWORK THAT IS -- THAT WOULD BE GOOD FOR IDN-OVER-EPP.
AND SO I HOPE -- I WILL BE TRYING TO CREATE AN IDN-OVER-EPP MAILING LIST.
SO PLEASE JOIN THAT.
AND IF YOU THINK THAT WHAT I WAS TALKING ABOUT MAKE SENSE, EXPRESS YOUR SUPPORT TO IETF OR -- THAT A STANDARDIZED FRAMEWORK DOES MAKE SENSE SO WE CAN MOVE FORWARD OR TAKE A LOOK AT THE CORE REQUIREMENTS FOR THE STANDARDIZED FRAMEWORK FOR IDN-OVER-EPP.
THANK YOU.
>>JAMES SENG: WE HAVE ONE MORE PRESENTATION BEFORE QUESTIONS. MAO WEI, YOU ARE GOING TO TALK ABOUT THE CDN REGISTRATION.
>>MAO WEI: I WILL GIVE A BRIEF INTRODUCTION ABOUT OUR EXPERIENCE FOR CDN REGISTRATION.
I WORKED TOGETHER WITH (INAUDIBLE) FOR MANY YEARS TO FIND THIS SOLUTION.
CHINESE HAS A SPECIAL PROBLEM THAT IS SIMPLIFIED CHINESE AND TRADITIONAL CHINESE EQUIVALENCE. THESE TWO CHARACTERS ARE COUNTRY, AND THIS IS SIMPLIFIED AND THIS IS TRADITIONAL. AND AT THE SAME TIME THERE ARE LOT OF VARIANT CHARACTER, SAME MEANING, FOR COUNTRY.

BASED ON THE RFC 3743, WE HAVE OUR SOLUTION TO DO THAT. THIS RFC S WAS IN APRIL OF THIS YEAR.
OUR REGISTRATION BUNDLING POLICY, A USER REGISTER CHINESE DOMAIN NAME IN SIMPLIFIED CHINESE ONLY. HE CAN GET ANOTHER FREE EQUIVALENT DOMAIN NAME IN TRADITIONAL CHINESE CHARACTER ONLY. FOR EXAMPLE, THE TSINGHUA UNIVERSITY AND OTHER DOMAIN NAME.
SOMETIME A USER WILL TYPE A MIXTURE CHARACTER DOMAIN NAME. SO WE CAN GIVE THIS ORIGINAL FORM TO HIM. AT THE SAME TIME, ANOTHER TWO DOMAIN NAME, TC ONLY AND SC ONLY DOMAIN NAME TO HIM.
WE PROVIDE THE RESOLUTION SERVICE FOR THE PACKAGING. THE ORIGINAL, THE SIMPLIFIED ONLY, AND THE TRADITIONAL ONLY. AND THE RESERVED RESOLVED ALL THE VARIANT VERSION DOMAIN NAME.
WHEN YOU CREATE AND TRANSFER THE APPLET, YOU MUST CHANGE ALL THE PACKAGES DOMAIN NAME. IT'S A LITTLE BIT COMPLEX, BUT IT'S TRANSPARENT FOR REGISTRAR. THIS IS AN EXAMPLE. TYPE ANY CHARACTER, ANY THREE DOMAIN NAME, AT TSINGHUA UNIVERSITY, YOU CAN GO TO THE SAME WEB SITE.
OKAY. THAT'S ALL. THANK YOU.
>>JAMES SENG: OKAY. WE OPEN THE FLOOR FOR QUESTIONS.
YES.
>>MICHEL SUIGNARD: YEAH, I JUST WANTED TO MAKE SURE THAT, SO WHICH VERSION ARE YOU RESOLVING? AND YOU'RE RESERVING THEM, BUT WHICH ONE? ARE YOU RESERVING BASICALLY THE ONE THE USER DID ASK FOR OR DO YOU ALSO RESERVE THE OTHER ONES? OR WHICH ONE DO YOU RESERVE?
>>MAO WEI: RESERVE?
>>MICHEL SUIGNARD: BASICALLY IF I PUT THE VARIANTS THAT YOU HAVE RESERVED, IS THAT ALSO BEING RESOLVED?
>>MAO WEI: RESOLVE ALL THE PACKAGING, BUT I DON'T PROVIDE A RESOLUTION SERVICE. ONLY PROVIDE THREE DOMAIN NAME RESOLUTION SERVICE.
>>JAMES SENG: MAYBE I CAN HELP OUT HERE BECAUSE I HELPED TO WRITE THE GUIDELINE. THE RESERVE VARIANTS DO NOT GO INTO THE DOMAIN FILE. SO NO ONE ELSE CAN REGISTER THE SAME NAME. ONLY THE ACTIVATED VARIANTS GO INTO THE DOMAIN FILE. THOSE WOULD BE RESERVABLE.
OKAY. THEN I ACTUALLY HAVE A QUESTION. WE HAVE SEEN THE PRESENTATION FROM EDMON WHICH TALK ABOUT IDN-OVER-EPP. SO MAO WEI, IN YOUR EXPERIENCE, DO YOU THINK IDN-OVER-EPP WILL BE USEFUL IN ORDER TO HELP THE IMPLEMENTATION?
>>MAO WEI: YES, IF WE HAVE THIS STANDARD IT'S CONVENIENT FOR THE REGISTRAR. CURRENTLY, WE PROVIDE TOOLKIT TO THE REGISTRAR, THE TRANSFER LOCAL CODE TO THE PUNYCODE AND THE APPLICATION QUERY TO OUR DATABASE SO WE CAN DO THAT CURRENTLY. BUT IF WE HAVE A STANDARD, WE CAN DIRECT THE REGISTRY TO USE LOCAL CODE.
>>JAMES SENG: HOW ABOUT GERMANY?
>>SABINE DOLDERER: ACTUALLY, IT'S ESPECIALLY THEN STARTING TO GET USEFUL WHEN YOU REGISTER NOT ONLY THE IDN BUT ALSO HAVE ALL THE OTHER DATA IN OTHER CHARACTER SETS, WHICH IS -- SO IF YOU HAVE THE NAMES MULLER THEY'RE ALSO IN THE REGISTRY AS MULLER AND NOT THE SORT OF VARIANT WHICH IS RELATED TO THE ASCII VERSION.
SO I THINK IT'S USEFUL AND I THINK PEOPLE SHOULD GO AHEAD WITH IT.
>>JAMES SENG: BUT YOU'RE NOT DOING VARIANTS.
>>SABINE DOLDERER: PARDON ME?
>>JAMES SENG: BUT YOUR NOT DOING VARIANTS.
>>SABINE DOLDERER: NO, WE DON'T DO VARIANTS BUT WHAT WE ARE DOING IS WE REGISTER THE PEOPLE'S NAME AS THE NAMER, SO WE ACTUALLY REGISTER AT THE MOMENT THE WHOLE CHARACTER SET IN TERMS OF DOMAIN NAME, PEOPLE'S NAME AND SO ON. BECAUSE PEOPLE ARE NOT LIMITED TO THE ASCII SPACE TO DO SO. AND THEREFORE I THINK YOU NEED AN UTF-8 EPP STANDARD.
KENNY, IS (INAUDIBLE) DOING ANYTHING?
>>KENNY HUANG: IT'S VERY INTERESTING AND PROBABLY I'M NOT IN A POSITION TO ANSWER RIGHT AWAY, BECAUSE WE ARE AN OPERATION PROVIDER AND PROVIDE TECHNOLOGY DUE DILIGENCE. SO IF THE PROPOSAL IS PROVIDED, FROM MY POINT OF VIEW, THAT WOULD BE A VERY POSITIVE MOVE FROM PIR POINT OF VIEW. THANK YOU.
>>JAMES SENG: ANY QUESTION FROM THE AUDIENCE?
YES.
>>ED LEWIS: MY NAME IS ED LEWIS FROM NEULEVEL.
THIS IS REGARDING EDMON CHUNG'S PRESENTATION ABOUT THE EPP BOF AND THE IETF. THE OTHER THING I WANTED TO SAY ABOUT THE SAME TOPIC, AND MOSTLY IN AGREEMENT, IS THAT AT THE BOF, THE WAY THE IETF LOOKED AT THIS IS THAT WHAT THEY WOULD LIKE AT THE END IS TO HAVE ONE WAY OF DOING THINGS, SPECIFICATION AND PROTOCOL. BUT IT'S NOT CLEAR THAT WE CAN GET TO THAT UNLESS WE HAVE ONE SET OF REQUIREMENTS THAT WOULD LEAD US TO THAT ONE SPECIFICATION, BECAUSE IF WE DON'T HAVE -- IF WE HAVE MULTIPLE SETS OF REQUIREMENTS THAT ARE IN CONFLICT WITH EACH OTHER, WE'LL SPEND A LOT OF TIME TRYING TO GET TO THIS PROTOCOL. BUT THE COMMENT WHAT WE NEED TO DO IS WE HAVE TO UNDERSTAND THE PROBLEM SPACE. AND RIGHT NOW IT SEEMS THERE WERE MULTIPLE UNDERSTANDINGS OF THE PROBLEM SPACE. I THINK IT WOULD BE GOOD IF THERE WAS ONE THAT CAME OUT OF ICANN, HAVING INPUT FROM THE ICANN PROCESS, HAVING SOME WAY OF HOW EVERYONE WANTS TO SEE IDN. BUT ALSO HAVE AN APPROACH FOR PEOPLE WHO DON'T COME TO THE I!
CANN MEETINGS, DON'T GO THROUGH THIS PROCESS, AND SEE IF THAT IS ONE GOOD SET OF REQUIREMENTS BEFORE IETF SPENDS TIME.
EPP CAN SUPPORT MULTIPLE APPROACHES TO SOMETHING BUT ON THE OTHER HAND I THINK IT WE WOULD LIKE TO HAVE ONE HARMONIOUS WAY TO HAVE EVERYBODY GO THROUGH THIS PHASE BECAUSE REGISTRARS AND REGISTRIES LIKE TO HAVE THE SAME INTERFACE, THAT'S OBVIOUS. RIGHT NOW, I THINK AT THE IETF WHAT WAS A CONCERN IS CAN WE GET THERE WITHOUT A REAL GOOD SET OF REQUIREMENTS THAT COVERS EVERYBODY'S NEEDS AND THAT'S WHAT WE NEED TO GET TO BEFORE WE GO TO THE IETF AGAIN, I THINK.
>>EDMON CHUNG: RIGHT. I COMPLETELY AGREE AND ONE OF THE CALL FOR PARTICIPATION IS REALLY TOWARDS THAT. AND BASICALLY, ACTUALLY, OUT OF THE IETF BOF AS WELL, WE'RE THINKING TO CREATE A MAILING LIST TO START TO TALK ABOUT THAT, AND DISCUSS THE REQUIREMENTS.
WHAT I SORT OF POINTED OUT IS THAT THERE IS CERTAIN DIRECTIONS THAT THE ICANN IDN GUIDELINES ALREADY PROVIDED, SO WORKING ON THAT, AND MOST PROBABLY THE IANA LANGUAGE TABLE REGISTRY, WE HAVE A GOOD SET OF SUBSTANCE TO WORK WITH, INCLUDING THE EXPERIENCE FOR CHINESE, WE HAVE A GOOD SET OF ELEMENTS TO WORK WITH TO HAVE THOSE REQUIREMENTS DONE. BUT YOU'RE COMPLETELY RIGHT; HAVING THAT BEFORE HAVING -- SORT OF SQUEEZING THE BRAIN JUICE OUT OF IETF WOULD BE VERY USEFUL, YEAH.
>>JAMES SENG: THANK YOU. AT THIS MOMENT IN TIME, I DO NEED TO CUT THE PANEL SHORT. WE ARE RUNNING OUT OF TIME. IN FACT, WE'RE WAY OVER TIME. BUT I THINK WE HAD A VERY INTERESTING DISCUSSION, OR AT LEAST I HOPE WE HAD AN INTERESTING DISCUSSION. ABOUT THE IDN OVER EPP I BELIEVE EDMON IS GOING TO DO THE PRESENTATION AT THE CCTLD FORUM.
SO THAT WOULD BE A GOOD PLACE TO CONTINUE THE DISCUSSION.
CAN I ASK EVERYONE TO HAVE A ROUND OF APPLAUSE FOR THE PANELISTS AND EVERYONE WHO HAS ATTENDED HERE TODAY
(APPLAUSE.)
>>JAMES SENG: THE SESSION IS CLOSED. WE'RE GOING TO HAVE A BREAK.
>>TINA DAM: THANKS. WE'RE JUST GOING TO TAKE A QUICK MINUTE TO CHANGE PANELISTS PARTICIPANTS TO THE NEXT PANEL DISCUSSION. I REALIZE IT'S STARTING TO BE A LONG DAY BUT WE DO HAVE SOME ADDITIONAL TOPICS ON THE PROGRAM. I THINK THIS PANEL MANAGED TO COVER A LARGE AMOUNT OF TOPICS, EVEN IN A SHORT PERIOD OF TIME FOR THE TOPICS THAT WAS UP FOR DISCUSSION, AND I HOPE THAT DISCUSSIONS ARE GOING TO CONTINUE WHEN WE'RE DONE FOR THE DAY OUT IN THE HALLWAY WHERE THERE WILL BE REFRESHMENTS AFTER THE LAST PANEL DISCUSSION.
SO JUST A QUICK ONE-MINUTE BREAK.






SESSION III: PANEL DISCUSSION.
IDN POLICY ISSUES AND PROCESSES.




>>RAM MOHAN: GOOD EVENING, WE'RE GOING TO GET KICKED OFF WITH THIS IDN PANEL, NEXT PANEL SESSION RIGHT AWAY.
I KNOW IT'S RATHER LATE IN THE EVENING AND I KNOW THAT WE'RE STARTING LATE, BUT THE TOPICS THAT WE'RE DISCUSSING TODAY ARE IMPORTANT TOPICS, AND ONES THAT DESERVE BOTH YOUR INPUT AND YOUR ATTENTION.
TODAY'S TOPICS OF DISCUSSION, WE INTEND TO TALK ABOUT UDRP, IP ISSUES AND IDN; THE REPRESENTATION OF IDN IN WHOIS, WHICH IS FRANKLY A TOPIC WE'VE COVERED MULTIPLE TIMES BEFORE, SO THE FOCUS TODAY IS A LITTLE BIT MORE ON THE POLICY AREAS AND THE TECHNICAL AREAS; DEVELOPMENT OF LANGUAGE TABLES, WHICH ARE CRITICAL TO SPARK THE CONTINUED DEVELOPMENT, THE CONTINUED EXPANSION OF IDNS AND DOMAIN NAMES; AND THEN A VERY INTERESTING TOPIC, IDN.IDN. WHAT DO YOU DO WHEN YOU HAVE MULTILINGUAL.MULTILINGUAL TYPE NAMES AND WHERE SHOULD THAT GO? SOME OF THE COMMUNICATION ASPECTS AND SOME OF THE END-USER DEMANDS AND REQUIREMENTS.
JOINING US TODAY IS A DISTINGUISHED PANEL. JUST STARTING FROM HERE WE HAVE CHARLES SHABAN FROM THE ABU-GHAZALEH INTELLECTUAL PROPERTY. WALTER WU FROM CNNIC, JAMES SENG, ELIZABETH LONGWORTH FROM UNESCO, AND HONG XUE FROM ALAC, CHINA.
I WANT TO TAKE A COUPLE OF MINUTES AND TALK ABOUT WHY THIS PARTICULAR PANEL HAS RELEVANCE IN TODAY'S ICANN MEETING AS WELL AS IN WHERE IDNS GO LONGER TERM.
AT THE END OF THE DAY, GOOD TECHNOLOGY CANNOT RESOLVE BAD POLICY. THIS IS A SESSION ABOUT POLICY AND PROCESS.
IDN TECHNOLOGY ITSELF IS NOW MORE MATURE. THERE'S OVER FOUR YEARS OF IMPLEMENTATION EXPERIENCE RIGHT NOW. THERE ARE MORE VARIETIES OF IMPLEMENTATION THAT ARE OUT THERE BETWEEN GTLDS AND CCTLDS, AND DIFFERENT APPROACHES ARE BEING TAKEN.
HOWEVER, MANY POLICY AND TECHNOLOGY AND PROCESS AREAS HAVE NOT CAUGHT UP WITH WHERE TECHNOLOGY HAS GONE.
IN ADDITION, THE METHOD OF INTERACTION OF ORGANIZATIONAL AND BODIES, AND I'VE LISTED A NUMBER OF THEM OUT THERE, IS UNDEFINED.
THERE IS MORE TO THIS. THERE IS NOT CONSENSUS ON WHAT ARE THE ISSUES. NO CONSENSUS ON WHO OWNS THE ISSUES. AND REALLY, NO CONSENSUS ON WHAT PROCESSES SHOULD WE USE TO RESOLVE THE ISSUES.
OUR PLAN TODAY IS TO INFORM AND EDUCATE ON THE REASONS WHY EACH OF THE ISSUES WE'RE TALKING ABOUT ARE IMPORTANT TO IDNS; TO LIST THE PROBLEMS THAT NEED TO BE SOLVED, EVEN IF WE CAN'T AGREE ON THE ENTIRE LIST OF PROBLEMS; TO PROVIDE OUR PERSPECTIVE ON A WAY TO MOVE THE PROBLEM RESOLUTION FORWARD; AND INVITE YOUR COMMENTS ON OUR THOUGHTS AS WELL AS YOUR OWN IDEAS ON THESE TOPICS.
WITH THAT, I WANT TO MOVE TO THE VERY FIRST TOPIC, WHICH IS UDRP, IP ISSUES AND IDN. AND TO DO THAT, WE'VE GOT A VERY DISTINGUISHED PANELIST, CHARLES SHABAN, AND WHILE I PASS THE -- THIS OVER TO YOU, LET ME JUST GIVE YOU A LITTLE BIT OF AN INTRODUCTION TO CHARLES.
CHARLES IS THE EXECUTIVE DIRECTOR OF THE REGIONAL OFFICE OF ABU-GHAZALEH INTELLECTUAL PROPERTY BASED IN AMAN, JORDYN. CHARLES WAS THE CHAIR OF THE IPC'S IDN WORKING GROUP IN 2003.
CHARLES, YOU HAVE A BRIEF PRESENTATION, SO LET'S GO TO YOU.
>>CHARLES SHABAN: THANK YOU VERY MUCH. IT WILL BE VERY SHORT BECAUSE AS YOU SAID, WHILE I WAS SHOWING THE IPC OR WORKGROUP ON THE INTELLECTUAL PROPERTY ON IDN ISSUES WE ISSUED A FULL REPORT, AROUND TEN PAGES, IT'S ON THE IPC WEB SITE SO I WILL NOT REPEAT AGAIN. ANYONE WHO WOULD LIKE TO SEE IT CAN SEE IT ON THE IPC WEB SITE OR HE CAN SEND TO ME AND I WILL DO IT.
SO I JUST HAVE FIVE SMALL SLIDES TO TALK ABOUT THE MAIN USUALS, WHICH IS I THINK THE MOST IMPORTANT. THE UDRP UNIFORM DISTRIBUTION POLICY.
THE UDRP USUALLY CAN BE USED BECAUSE IT'S MAINLY ABOUT TRADEMARKS. AND SO IF IT IS -- EVEN FOR IDN, WE THINK IT WILL BE WORKABLE. BUT STILL, IT'S NOT VERY CLEAR FOR THE IP PEOPLE -- I'M SORRY, I MEAN INTELLECTUAL PROPERTY, BECAUSE SOMETIMES HERE IT MAY BE INTERNET PROTOCOL, MAYBE -- FOR THE INTELLECTUAL PROPERTY PEOPLE, IT'S STILL SOMETIMES NOT CLEAR HOW IT WILL BE UNTIL THERE IS SOME MORE MAYBE IMPLEMENTATIONS. BUT WE UNDERSTAND THERE MAY BE PROBLEMS IN TRANSLATION.
THE OTHER PROBLEM IS THE WIPO II PROCESS WHICH EVERYBODY KNOWS ABOUT. THE WIPO II PROCESS, ALTHOUGH IT MOVED IN SOME FORAS, BUT FOR EXAMPLE, IN THE INTERNATIONAL TRADEMARK ASSOCIATION, THEY HAD ZERO FOR IT. IN THE IDN, AGAIN, I THINK IT'S STILL NOT CLEAR, BUT MAYBE THERE WILL BE SOME -- WE NEED TO STUDY IT WHEN WE START HAVING MAYBE THE FULL IDN AS WELL.
FOR THE CCTLDS, ALTHOUGH IN THE PREVIOUS PRESENTATION I SAW VERY NICE PRESENTATIONS, BUT FOR THE INTELLECTUAL PROPERTY PART I THINK IT WILL BE MUCH EASIER BECAUSE USUALLY IN EACH COUNTRY THERE IS SPECIFIC LAWS FOR THE INTELLECTUAL PROPERTY. SO THEY CAN FOLLOW THE SAME TREND IN THE INTELLECTUAL PROPERTY ISSUES REGARDING THE DOMAIN NAMES.
BUT STILL, WE EXPECT THAT SOME OF THE COUNTRIES, OF COURSE MAINLY THE DEVELOPING, THEY WILL REQUIRE TO UPDATE THEIR LAWS AND MAYBE TO AMEND SOME.
THE WHOIS, WE HAVE ONLY ONE PROBLEM, AND I THINK ONE OF MY COLLEAGUES WILL TALK IN DEPTH. FOR ME I WILL JUST MENTION THAT WHEN THERE IS A CASE OR LITIGATION WE DEPEND ON THE WHOIS TO KNOW WHO IS THE OTHER THAN, WHO REGISTERED THIS NAME OR SOMETHING. SO IF IT IS ONLY IN ONE LANGUAGE, THAT'S MAYBE NOT ACCESSIBLE TO OTHERS. MAYBE WE THINK THAT WE'D LIKE THE WHOIS TO BE TWO LANGUAGES OR SOMETHING LIKE THIS.
THE RFCS, LIKE THE RFCS THAT'S SENT TO THE IETF OR THE OTHER BODIES, I'D LIKE TO MENTION ONE OF THE EXAMPLES. I WAS PART OF THE ARAB DOMAIN NAMES TASK FORCE THAT WORKED ON SOME KIND OF RFC AND IT WAS SENT TO THE IETF.
WELL, WE SAW THAT IT ONLY CONTAINED SOME BASIC ISSUES THAT WILL NOT BE SUFFICIENT. SO WE ARE WORKING ON A SECOND PHASE IN COOPERATION WITH MINC AND THE U.N. ESCWA TO SEE IF WE CAN INCLUDE SOME OF THE INTELLECTUAL PROPERTY ISSUES THAT SHOULD BE INCLUDED OTHER THAN JUST INTELLECTUAL PROPERTY ISSUES IN GENERAL. BUT IT WILL BE LOOKED AT IN A VERY EXISTENCE I HAVE WAY. AND THAT'S IT. I PROMISED YOU ONLY THREE TO FOUR MINUTES. THANK YOU VERY MUCH.
>>RAM MOHAN: CHARLES, THANK YOU. IT WAS THREE TO FOUR MINUTES BUT THERE WAS A LOT OF CONTENT IN THERE. I HAD A FEW QUESTIONS AND I WONDER IF THE AUDIENCE AND PANELISTS HAD SOME THOUGHTS ON THESE.
WHAT IS HAPPENING TODAY -- I DON'T KNOW IT; WE SHOULD PASS IT ON.
CHARLES, WHAT IS HAPPENING TODAY IN TERMS OF IDN DISPUTES? ARE THEY CURRENTLY HANDLED UNDER UDRP? HAVE CASES COME OUT THERE THAT WORK WITH IDN DISPUTES?
>>CHARLES SHABAN: YES, BUT IT IS AN IDN, OF COURSE, DOT COM OR DOT NET, THE USUAL REGISTRIES. THERE IS SOME CASES ALREADY REGISTERED AND YOU CAN CHECK THEM ON THE WIPO WEB SITE, AND THE OTHER DISPUTE RESOLUTION FORUMS THAT DO -- THAT SORT THESE ISSUES, BUT THE WIPO CONTAINS THE MOST.
THEY'RE FOLLOWING OF COURSE THE SAME RULES, TO PROVE THE THREE THINGS. IF YOU PROVE IT WAS REGISTERED IN BAD FAITH AND SO ON, AND YOU HAVE TO SHOW THAT YOU HAVE A REGISTERED TRADEMARK IN THAT LANGUAGE.
>>RAM MOHAN: GIVEN THAT YOU HAD SAID THAT WIPO II HAD ZERO PERCENT SUPPORT FROM INTIF, WHAT'S THE STEP FORWARD? WHERE DOES THIS GO?
>>CHARLES SHABAN: WE THINK THAT THERE SHOULD BE THE OTHER MECHANISMS TO SORT THE OTHER ISSUES LIKE THE COUNTRY NAMES OR IGO NAMES AND SO ON.
THE WIPO II TRIES TO MAKE THE UDRP ITSELF SOLVE IT BECAUSE THE RULES WHEN THE UDRP WAS INITIALLY FORMED, DONE, WAS ENFORCED, IT WAS SPECIFICALLY FOR THE TRADEMARK REGISTERED. REGISTERED TRADEMARK, I MEAN. IF IT'S NOT REGISTERED, YOU CAN'T USE, USUALLY, THE UDRP.
SO HOW YOU CAN SOLVE IT, FOR THE CASE OF COUNTRY NAMES OR IGOS OR SOMETHING LIKE THIS, MAYBE SOME WILL FALL UNDER SOMETHING WHICH IS CALLED COMMON TRADEMARK, BUT IN GENERAL, WE THINK THAT THERE SHOULD BE OTHER PROCESSES MAYBE TO SOLVE THESE LITIGATION ISSUES BETWEEN COUNTRY NAMES OR IGO NAMES AND SO ON.
>>RAM MOHAN: PANELISTS, ANY QUESTIONS?
I CAUGHT YOU PREPARING ON YOUR PRESENTATIONS.
ANY QUESTIONS ON THIS TOPIC?
>>HONG XUE: THANKS, CHAIR.
IT'S NOT REALLY A QUESTION. IT'S A COMMENT.
AT PRESENT, MOST IDN REGISTERED AT SECOND LEVEL, SO IT'S IDN.GTLD. SO THEY CAN BE RESOLVED IN THE UDRP. THERE'S NOTHING WRONG WITH THAT. I MYSELF WAS APPOINTED NEUTRAL BY WIPO SO I DECIDED A CASE ON IDN.GTLD. BUT IN THE FUTURE WE HAVE TO REALIZE THAT PROBABLY IN THE FUTURE WE INTRODUCE THE IDN GTLD. IN THAT CASE WE HAVE TO CONSIDER WHETHER WE NEED TO SUBSTANTIALLY CHANGE UDRP.
THANK YOU.
>>RAM MOHAN:
>>CHARLES SHABAN: THAT'S EXACTLY WHAT I MEANT IN MY SLIDE. THANK YOU VERY MUCH.
>>RAM MOHAN: A FOLLOW-ON QUESTION. IN SUCH A CASE, DO PANELISTS NEED EXTRA TRAINING OR EDUCATION?
>>HONG XUE: VERY GOOD QUESTION. TO BE READY FOR THOSE IDN CASES, THE PANELISTS SHOULD HAVE THE LANGUAGE CAPACITY TO REVIEW THE CASE MATERIAL IN SPECIFIC LANGUAGE; RIGHT, YES. BUT I CAN'T SEE ANY SPECIFIC TRAINING PROGRAM FOR PANELISTS.
THANKS.
>>RAM MOHAN: THANK YOU. QUESTIONS FROM THE AUDIENCE?
OKAY.
THE NEXT TOPIC WE HAVE IS WHOIS REPRESENTATION ISSUES. I WANTED TO ENGAGE ALL OF OUR PANEL HERE ON THIS TOPIC RATHER THAN GO THROUGH A CANNED PRESENTATION.
I HAD A FEW THOUGHTS, AND I JUST WANTED TO KICK IT OFF WITH THOSE THOUGHTS AND THEN HAVE THE REST OF THE PANEL CHAT ABOUT THIS FOR A WHILE AND THEN INVITE COMMENTS FROM THE AUDIENCE.
AS YOU ALL KNOW, THE WHOIS IS A VENERABLE OLD APPLICATION. IT'S STILL WIDELY USED AND IS DEPENDENT UPON FOR DIRECTORY LOOK-UPS. IT'S GENERALLY INVOLVED BOTH ON A WEB INTERFACE AND ON THE USING THE PORT 43 PROTOCOL.
TO MY WAY OF THINKING, THERE ARE FEW AREAS THAT WE NEED TO TALK ABOUT WHEN IT COMES TO IDNS AND WHOIS. PORT 43, FOR INSTANCE, TODAY SUPPORTS ONLY ASCII CHARACTER DISPLAY. HOW DO YOU DISPLAY VARIANTS AND PACKAGES WHEN IT COMES TO WHOIS.
WHEN YOU LOOK AT IDN PACKAGES IT'S A PRETTY INVOLVED ISSUE WHEN IT COMES TO DATABASES.
JOHN KLENSIN WAS HELPFUL IN DEVELOPING THIS THOUGHT FOR ME. THE DISPLAY OF RESERVED BUT NOT REGISTERED -- IN OTHER WORDS, NOT ACTIVATED -- DOMAIN NAMES THAT ARE NOT REGISTERED, WHAT SHOULD WHOIS DO IN THESE AREAS?
AND THERE IS AN IETF WORKING GROUP, THE CRISP WORKING GROUP THAT IS IN THE FINAL STAGES OF DEVELOPING A PROTOCOL, IRIS. DO YOU THE PANELIST AND THE AUDIENCE THINK THAT IRIS COULD BE A GOOD REPLACEMENT FOR PORT 43?
THAT'S A QUESTION FOR YOU FOR SOME THOUGHTS. I'LL OPEN IT UP TO THE PANEL FIRST AND THE AUDIENCE. PLEASE COME UP TO THE MIKE IF YOU HAVE QUESTIONS.
ANY THOUGHTS?
>>JAMES SENG: LET ME HELP, RAM, SINCE NO ONE WILL TAKE THE QUESTION.
WELL, (INAUDIBLE) BACK IN THE DAYS WHERE EVERYTHING IS DOING ASCII. BUT ACTUALLY IF YOU LOOK TODAY, I DOUBT ANYONE STILL USE A TERMINAL OR TELNET SESSION TO DO TELNET TO POP FOR THE RETRIEVE TO LOOK FOR WHOIS. THAT DOESN'T HAPPEN ANYMORE. MOST OF US WILL GO THROUGH -- EXCEPT FOR JOHN WHO IS WAVING HIS HAND. BUT MOST HUMAN BEINGS WILL GO TO A PROPER WEB SITE. SO LET'S NOT TALK ABOUT NON-HUMAN BEINGS.
THEY WILL GO TO A WEB SITE. AND IF YOU TALK ABOUT LOOKING AT A POP RETRIEVE TO DO A QUERY DOES NOT MAKE SENSE ANYMORE. IT'S BETTER TO HAVE A MORE INTELLIGENT MACHINE-TO-MACHINE COMMUNICATION WITH THE REGISTRIES THAT DO CONTAIN THE WHOIS.
AND IN THAT CASE, IRIS BEING THE CONSENSUS WITHIN THE IETF IS A DEFINITELY A GOOD STEP FORWARD.
>>RAM MOHAN: WHAT IS DOT CN DOING WHEN IT COMES TO DISPLAYING WHOIS FOR CHINESE CHARACTERS?
>>WALTER WU: ACTUALLY, I CAN JUST GIVE SOME INTRODUCTION ABOUT THE DOT CN IMPLEMENTATION ON THE WHOIS DISPLAY.
ACTUALLY, AT THIS POINT, I THINK THE WHOIS DISPLAYS HAVE A CLOSE RELATIONSHIP WITH REGISTRATION POLICIES, BECAUSE CHINESE LANGUAGE, THE PROBLEM IS CURRENTLY WHEN WE OFFER REGISTRATIONS FOR CHINESE DOT CN, THE BUSINESS RULE IS PEOPLE CAN REGISTER A PURE SIMPLIFIED FORM OR PURE TRADITIONAL FORMS, AND THEN THEY CAN GET THE VARIANTS AT THE SAME TIME.
FOR EXAMPLE, THEY REGISTER SIMPLIFIED FORM THEY WILL GET TRADITIONAL FORM. ONLY ONE CASE, THE REGISTRANT WILL GET THREE FORMS BECAUSE IF THEY SUBMIT A MIXED DOMAIN NAMES WITH SIMPLIFIED AND TRADITIONAL CHARACTERS, THEN THEY WILL GET THREE VARIANTS.
BUT IN WHOIS DISPLAY, IT DISPLAYS ALL OF THE VARIANTS THAT THE REGISTRANT GOT FROM THE REGISTRATION.
>>JOHN KLENSIN: JAMES, THERE MAY BE AN IMPORTANT ISSUE HERE WHICH I THINK, IN US TEASING EACH OTHER, WE MAY BE SKIPPING AND I THINK IT MAY BE IMPORTANT.
TO TALK ABOUT TELNETING TO PORT 43 IS ONE THING. AND WHETHER I ACTUALLY DO IT OR NOT IS IRRELEVANT.
BUT TO HAVE TO GO TO A WEB PAGE FOR THAT INFORMATION IS AN ENTIRELY DIFFERENT STORY. AND I THINK YOU'LL FIND IF YOU LOOK AT THE OPERATIONS COMMUNITY THAT MOST THE OPERATIONS COMMUNITY, WHEN THEY USE WHOIS TO RETRIEVE MATERIAL, ARE GOING TO PORT 43 WITH A WHOIS CLIENT BECAUSE IT IS QUICK AND IT PRODUCES OUTPUT IN A FORM THAT IS EXTREMELY EASY TO PUT INTO A FILE.
TO DO THAT FROM A WEB PAGE IF YOU'RE AN OPERATOR TRYING TO RETRIEVE INFORMATION QUICKLY OR IF YOU'RE SOMEBODY IN A DEVELOPING COUNTRY TRYING TO RETRIEVE INFORMATION OVER A REALLY BAD DIAL-UP LINK WHICH DOES HAPPEN TO SOME OF US SOMETIMES, THEN THE NOTION OF GOING TO A WEB PAGE WHOSE INTERFACE AND LOCATION WHERE THE INFORMATION GETS TYPED IN IS DIFFERENT FOR EVERY SINGLE REGISTRY, AND OFTEN EVERY SINGLE REGISTRAR IN THE THIN REGISTRIES, AND IT IS OFTEN SURROUNDED BY LOTS OF FUNNY GRAPHICS AND DANCING GIRLS WHICH TAKE ABSOLUTELY FOREVER TO DOWNLOAD, AND THEN THE INFORMATION IS DELIVERED IN AN INCONVENIENT FORM.
SO LET'S NOT LOSE THAT IN THE HOPES OF ASSUMING THAT EVERYTHING IS LOVELY GRAPHICAL WEB INTERFACES.
>>JAMES SENG: YES. SORRY IF YOU MISUNDERSTOOD ME. I NEVER SUGGESTED WE WOULD USE WEB INTERFACE AS INTERFACE FOR WHOIS. I DON'T THINK THAT IS THE PURPOSE OF CRISP OR IRIS. IRIS IS REALLY AN INFORMATION SYSTEM THAT TALKS ABOUT MACHINE-TO-MACHINE COMMUNICATION. IT DOES ENABLE SOMEONE TO ACTUALLY DEVELOP AN IRIS CLIENT THAT RETRIEVES INFORMATION FROM THE REGISTRAR AND THEN PROCESS IT ACCORDINGLY. YOU DON'T NEED TO GO TO THE WEB PAGE. BUT WHAT I'M TRYING TO SAY IS THAT THE LEGACY OF 43 -- I MEAN THE MAINTENANCE OF PORT 43 IS NO LONGER A VERY CRITICAL PART OF REGISTRY OPERATION SO LONG AS THERE IS AN EQUIVALENT MACHINE-TO-MACHINE COMMUNICATION PROVIDED FOR THE REGISTRANT.
>>JOHN KLENSIN: WE'RE IN COMPLETE AGREEMENT ABOUT THAT.
>>EDMON CHUNG: HI. IT'S EDMON CHUNG FROM AFILIAS.
I DON'T SEE ANYBODY FROM TAIWAN HERE BUT I CAN PROBABLY GIVE AN -- ADD TO THE DISCUSSION HERE.
WHAT I -- BASED ON MY UNDERSTANDING, WHAT THE DOT TW DOES IN TERMS OF PUSHING THE VARIANTS INTO THE WHOIS, BECAUSE IT MAY GENERATE A LONG LIST, WHAT THEY ACTUALLY DO IS ONLY PRESENT THE VARIANT -- THE PARTICULAR VARIANT CHARACTERS. EACH CHARACTER THAT IS SUBJECTED TO A VARIANT, PRESENT THAT, AND THEN WITH THE POLICIES THAT THEY PRESENT AND THE WEB PAGE TOGETHER, YOU SUPPOSEDLY COULD CREATE THE PERMUTATIONS OF THE DIFFERENT VARIANTS.
>>RAM MOHAN: IT ALMOST SOUNDS LIKE EACH CCTLD, AND POTENTIALLY EACH GTLD COULD GO ABOUT, DEVELOP ITS OWN POLICIES ON HOW TO DISPLAY THIS STUFF. AND EFFECTIVELY LOSE THE UNIFORMITY, TO WHATEVER LEVEL THAT IT EXISTS TODAY, TO LOSE THAT BY SOME OF THESE APPROACHES THAT YOU'RE JUST DESCRIBING.
>>EDMON CHUNG: I THINK MAYBE A LOT OF PEOPLE WOULD AGREE WITH ME THAT THE UNIFORMITY OF WHOIS DON'T REALLY EXIST THAT MUCH TODAY. AND BY DEFAULT, IT'S JUST A TEXT OUTPUT.
>>RAM MOHAN: I WAS REALLY TALKING ABOUT NOT THE UNIFORMITY OF WHOIS IN DISPLAY BUT UNIFORMITY OF WHOIS IN ACCESS METHODS. IT SOUNDS LIKE THE METHOD YOU'RE DESCRIBING HERE IS, IF THIS, THEN SELECT MORE AND GET MORE. AND THAT DOESN'T SEEM TO BE WHAT THE -- THE NETWORK OPERATOR WHO HAS A SCRIPT THAT GOES AND JUST PULSE SOMETHING QUICKLY, THEY CAN JUST USE THE SCRIPT TODAY THAT THEY GO TO A NORMAL PORT 43 AND JUST PICK THE INFORMATION. THAT'S WHAT I MEAN BY UNIFORMITY.
>>EDMON CHUNG: I THINK YOU MAY MISUNDERSTAND, THEN. REALLY WHAT I'M TALKING ABOUT IS I THINK DOT TW, WHAT THEY DO IN THE WHOIS OUTPUT IS AT THE END THEY ALSO PROVIDE THE SPECIFIC CHARACTERS AND THEIR VARIANTS. SO THAT INSTEAD OF PROVIDING THE WHOLE LIST, BECAUSE THAT COULD BE, LIKE, 10,000 DOMAINS, THEN YOU WOULD USE THOSE CHARACTERS AS PERMUTATIONS; NOT THAT IT USES A DIFFERENT FORMAT IN TERMS OF ACCESS.
>>RAM MOHAN: OKAY. THANKS FOR CLARIFICATION. THAT IS HELPFUL.
ANY OTHER QUESTIONS ON THE TOPIC OF WHOIS AND IDNS? I KNOW IT'S LATE IN THE DAY, BUT IT'S A TOPIC THAT'S REALLY NOT GOING TO GO AWAY.
ALL RIGHT. WE TALKED ABOUT SIX TOPICS TO BE DISCUSSED TODAY. WE'RE DONE WITH TWO. WE'RE DONE WITH UDRP, WE'RE DONE WITH REPRESENTATION OF IDN AND WHOIS, AT LEAST FOR TODAY'S SESSION.
THERE ARE FOUR OTHER TOPICS LEFT: THE DEVELOPMENT OF LANGUAGE TABLES, IDN.IDN, CULTURAL COMMUNICATION ASPECTS, AND END-USER DEMANDS AND REQUIREMENTS.
MY THOUGHT WAS THAT SINCE WE'VE TALKED ABOUT TWO RELATIVELY TECHNICAL TOPICS -- ACTUALLY, IDN AND WHOIS IS MORE TECHNICAL. THE UDRP IS NOT NEARLY THAT TECHNICAL. PERHAPS WE TAKE ONE MORE SOMEWHAT TECHNICAL TOPIC AND THEN MOVE ON TO ELIZABETH WITH THE CULTURAL COMMUNICATION ASPECTS, PERHAPS, AND THEN GO FROM THERE.
SO WE ARE GOING TO GO TO YOU, TO CNNIC, ON SUGGESTIONS THAT THEY HAVE ON WHAT I CALL IDN.IDN, BUT THIS IS -- LET'S TALK ABOUT IDN.IDN AND THEN LET'S THEN TALK ABOUT CULTURAL COMMUNICATION ASPECTS, STOP AND TAKE SOME QUESTIONS, THEN COME BACK TO DEVELOPMENT OF LANGUAGE TABLES AND END-USER DEMANDS AND REQUIREMENTS.
TO YOU.
>>WALTER WU: THANK YOU. OKAY. THANK YOU. ACTUALLY, I WILL GIVE A VERY QUICK PRESENTATION ON THAT. AND AFTER RESEARCHING THE INTERNATIONALIZED DOMAIN NAME FOR A VERY LONG PERIOD OF TIME, CNNIC RECOGNIZED THAT THE NAMES ARE DIFFICULT BUT ARDUOUS WORK, AND (INAUDIBLE) CONCERN OF CCTLDS. TO PUSH FORWARD SUCH A COMPLEX TASK. THEREFORE, CNNIC NOW WILL MEET FOLLOWING A PROPOSAL, WE ARE LOOKING FOR COMMENTS EVEN IN THIS PANEL OR FROM THE COMMUNITY.
IN ADDITION TO INTERNATIONALIZED IDN NAMES, IT'S NECESSARY FOR IDN USERS AND SHOULD BE DONE.
IT IS THE ONLY WAY TO POPULAR RISE THE USE OF INTERNET IN NON-ENGLISH SPEAKING COMMUNITIES. IT IS A REQUIREMENT FOR RESPECTING AND PRESERVING THE CULTURE AND SPECIAL INTERESTS OF PEOPLE IN DIFFERENT GEOGRAPHIC REGIONS.
AND (INAUDIBLE) INCLUDING RFC 3490, 3491, 3454 AND 3743, MAKE IT VERY CLEAR THAT INTERNATIONALIZING IDN CCTLDS IS FEASIBLE. (INAUDIBLE) HAS PROVEN THAT IT'S THE RIGHT TIME TO DO THE IDN NAMES IN TERMS OF TECHNICAL AND MARKETING IMPLEMENTATION. AND IN THE INITIAL PHASE, WE SUGGEST THAT EACH CCTLD WILL HAVE ONLY ONE SUIT OF CORRESPONDING NAMES IN IDN FORM. FOR EXAMPLE, SUCH AS CHINESE, DOT CHINA, AND BASED ON BEST PRACTICES AND ACCUMULATED EXPERIENCE, SUCH NAMES COULD LATER BE MODIFIED OR ADDED.
THE CCTLD MANAGER ACTUALLY HAVE A RIGHT TO CHOOSE THEIR OWN LANGUAGES, OR ONE OF THEIR CHARACTER SETS, TO INDICATE THEIR SPECIFIC CCTLD NAMES.
AND THEN, REGISTER THESE NAMES TO ICANN'S ROOT IN PUNYCODE FORM.
WHEN THE REGISTRY IS READY FOR LAUNCHING THE REGISTRATION SERVICE, IT MAY SUBMIT THE APPLICATION TO ICANN.
THE REGISTRIES MUST BE WELL PREPARED BEFORE INITIATING THE APPLICATION PROCEDURES.
SO CURRENTLY, CNNIC HAS ACCOMPLISHED DOT (INAUDIBLE) TESTING AND ALSO HAS EXPERIENCE IN TESTING THE IPV6 ROOT SERVER. OUR RESULTS WERE SUCCESSFUL AND PROVE THAT THE ADDITIONAL FOR INTERNATIONALIZED CCTLD NAMES IN THE ROOT IS FEASIBLE.
SO CONTRARILY, POTENTIALLY CCTLD USERS MAY FIND THE LESS TECHNICAL SOLUTIONS FOR THEIR NEEDS, WHICH WILL BRING ABOUT A DISADVANTAGE TO THE CONSOLIDATION OF INTERNET STANDARDS AND THE OPERATIONAL STABILITY OF THE INTERNET.
OKAY. THAT IS MY PRESENT. THANK YOU.
>>RAM MOHAN: ANY QUICK QUESTIONS?
WE HAVE JAMES SENG WITH SOME FOLLOW-ON THOUGHTS ON IDN.IDN.
AND THEN I HAD, ACTUALLY, AN ARTICLE THAT I WANTED TO SHOW YOU ALL BEFORE WE ACTUALLY JUMP INTO THE DISCUSSION OF IDN.IDN.
JAMES.
>>JAMES SENG: PERHAPS IT'S QUITE SIMILAR TOPIC.
SO PERHAPS WE SHOULD DO THE PRESENTATION TOGETHER.
FIRST, A DISCLAIMER.
I KNOW I'M HERE -- MY TITLE HERE SAYS IDN.
BUT I DO NOT SPEAK FOR IDN HERE.
I AM SPEAKING IN MY OWN PERSONAL CAPACITY.
SO I AM BEING WARNED SEVERAL TIMES I MUST HAVE DISCLAIMER IN ANY PRESENTATION I DO ON MY PERSONAL BEHALF.
I WOULD LIKE TO TALK ABOUT A PERFECT SOLUTION.
ONE OF THE THINGS THAT ICANN AND THE INTERNET COMMUNITY HAS BEEN WORKING ABOUT IS REALLY LOOKING AT HOW DO WE SOLVE THE IDN.IDN PROBLEM, IS THERE A PERFECT SOLUTION.
AND IN THE IDEAL WORLD, THERE'S A PERFECT SOLUTION AND IT'S 100% FOOLPROOF TO ALL PROBLEMS.
BUT WE DON'T LIVE IN A PERFECT WORLD.
WE DO THE BEST WE CAN.
EVERY SOLUTION WILL HAVE SOME PROBLEM OR OTHER.
THE BEST WE CAN DO IS FIND A SOLUTION WITH THE LEAST AMOUNT OF PAIN, THE SOLUTION THAT ALLOWS US TO MOVE FORWARD WITH THE LEAST NUMBER OF PROBLEMS, SO LONG AS IT WORKS, CAN SCALE, AND CAN BE IMPLEMENTED REASONABLY.
THERE'S OBVIOUSLY EDGE CASES IN EVERY PROBLEM.
EVERY PROPOSAL THAT HAS BEEN BROUGHT OUT, THERE IS ALWAYS PROBLEM.
CNNIC JUST MADE A PROPOSAL.
I CAN THINK OF AT LEAST TWO OR THREE REASONS WHY WE SHOULD NOT DO THAT.
THE -- THERE IS EDGE CASE PROBLEMS.
BUT YOU DON'T THROW OUT THE WHOLE PAPER.
IT'S NOT SO BIG THAT YOU THROW OUT THE WHOLE THING.
SO LONG AS IT CAN BE REASONABLY DEALT WITH, YOU DEAL WITH THE EDGE PROBLEMS SEPARATELY IN EXCEPTION CASE.
THE EASY WAY TO DO THAT TODAY IF YOU COME IN A POSITION IS DON'T DO ANYTHING ABOUT IT.
JUST DON'T.
IT'S BETTER THAN NOT DOING ANYTHING FOR ALL IDN TLD TODAY.
AT LEAST WE GET SOME IDN TLD TO MOVE FORWARD.
THE THREE MAIN REASON WHY IS WE HAVE A LOT OF DIFFICULTIES WITHIN THE IDN TLD -- PERHAPS THERE'S MORE WHICH I AM NOT AWARE OF.
BUT THE THREE THAT I CAN SEE, THAT ICANN IS DEFINITELY VERY PRUDENT ABOUT RELEASING TLDS.
OVER THE LAST FIVE YEARS IN ICANN, I THINK WE RELEASED LESS THAN 12, SORRY, LESS THAN TEN IN TOTAL.
THE FACT THAT THE TLDS, CCTLDS, MULTIPLY THAT BY X NUMBER THAT YOU CAN THINK OF.
YOU ARE TALKING ABOUT HUNDREDS OF THOUSANDS OF TLDS IS REALLY SCARING THE SHIT OUT OF EVERYONE.
SORRY.
I SHOULDN'T USE THAT WORD.
YOU KNOW MY POINT.
IT'S VERY SCARY.
AND IT'S -- AND ICANN, WHICH IS CONCERNED ABOUT THE STABILITY OF INTERNET, IS OBVIOUSLY CONCERNED WHAT THE IDN TLDS IS GOING TO INTRODUCE.
THE OTHER PROBLEMS THAT WE HAVE TO DEAL WITH IS THERE ARE ACTUALLY THREE KINDS OF TLD: THE GTLD, THE STLD, AND THE CCTLDS.
EACH HAS DIFFERENT POLICIES, DIFFERENT SYSTEM AND YOU CAN'T HAVE UNIFIED RULES FOR ALL OF THESE.
USUALLY THEY HAVE TO BE DEALT WITH SEPARATELY AND THE POLICIES MIGHT BE VERY DIFFERENT.
WHICH IS THE BEST ONE TO GO, WITH GTLD FIRST, STLD, ORCCTLD, IF WE GO WITH S, CCTLD,ED TO WE HAVE A NICE LIST THAT SAYS WHO AND WHAT, BUT FOR IDN CCTLD, WE DON'T.
THERE'S NO SUCH LIST AVAILABLE TODAY.

I'M GOING TO MAKE SOME PROPOSAL.
THIS IS BASED ON SOME DISCUSSION WE HAVE IN THE LAST ONE -- ACTUALLY, THIS MORNING, WHEN WE HAVE A PANEL DISCUSSION, PANEL PREVIOUS MEETING IN THE ROOM, FOR THOSE WHO ACTUALLY WORKED PAST LEVEL 2.66, YOU MIGHT ACTUALLY HEARING A LOT OF YELLING AND SHOUTING AND DEBATING IN THE ROOM.
WE WERE HAVING A DISCUSSION, SO I WAS ASKED TO COME AND GIVE A ROUGH CONSENSUS PRESENTATION OF WHAT WE THINK A GOOD PROPOSAL MIGHT BE.
THIS IS IN NO WAY FINAL, BUT IT'S ONLY A SMALL GROUP OF US SITTING AROUND, JUST TALKING ABOUT THE ISSUES AND COMING UP WITH SOME IDEAS.
BUT I THOUGHT IT WOULD BE GOOD TO SHARE.
FIRST OF ALL, ON GTLD, STLD,CCTLD, WE ASKED WHICH DO WE GO FIRST.
AND THE GROUP CONSENSUS IS WE SHOULD MOVE WITHCCTLD FIRST.
THE REASON IS SIMPLE: THERE'S NO DISPUTE WHO OWNSCCTLD.
AND WE ARE NOT CREATING ANY NEW STAFF AUTHORITY.
IT'S THE SAME PEOPLE WHO RUN THECCTLD WHO GET THECCTLD.
THECCTLD ALSO KNOWS WHAT'S BEST FOR THE COMMUNITY, WHAT LANGUAGES TO DEPLOY, WHAT IS MAJOR, WHAT HAVE THE MOST VISIBILITY IN DOING IT, VERSUS SAY A GTLD OR SPECIAL INTEREST GROUP.
AND EACHCCTLD MAY NOT DEAL WITH A LOT OF LANGUAGES.
THEY ONLY HANDLE ONE OR TWO.
SOME COUNTRIES MAY HAVE -- INDIA HAS 14 DIFFERENT LANGUAGES.
WE MAY POTENTIALLY LOOK AT THAT.
BUT I'LL GET TO THE SOLUTION TO THAT PROBLEM LATER ON, IN THE NEXT FEW SLIDES.
THE IMPORTANT THING IS NOT EVERYCCTLD REALLY WANTS TO DO IDN TLD.
LET'S PUT IT THIS WAY.
HOW MANY CCTLDS DO HAVE IDN CAPABILITY?
I THINK BY TODAY'S COUNT, IT'S APPROXIMATELY 125 TO 130CCTLD WHICH HAS CAPABILITY TO LAUNCH IDN CCTLD AND IDN -- AND EVEN OF THAT AMOUNT, 30 OF THEM, HALF OF THEM HAVE NO INTEREST IN IDN CCTLD.
GERMANY, FRANCECCTLD.
IT MAKES NO DIFFERENCE, IT'S ONLY REALLY AT THE END OF THE DAY PERHAPS FIVE TO TEN WHO REALLY WILL TAKE UP THIS OFFER IF YOU GO BY THECCTLD ROUTE, INITIAL FIRST ROUND.
WE STILL HAVE TO WORK ON GTLD AND STLD.
IT HAS TO BE DEALT WITH SEPARATELY.
CCTLD MAY NOT BE ABLE TO HANDLE IT EITHER.
IF A COUNTRY LIKE INDIA WHICH HAS 14, MAY NOT DO WITH ALL, BUT JUST WITH MAJOR ONES LIKE HINDI AND A COUPLE OTHERS.
WE SHOULD LOOK AT HOW TO USE WITH THOSE LANGUAGES, ESPECIALLY THOSE THAT DO NOT HAVE A COUNTRY REPRESENTATION, FOR EXAMPLE, LIKE TAMIL THAT COULD BE DONE, FOR EXAMPLE, THROUGH STLD WITH A SPECIAL INTEREST GROUP WORKING TOGETHER.
BUT I AM NOT MAKING THAT PROPOSAL TODAY.
THE PROPOSAL IS LET'S START WITHCCTLD BECAUSE IT'S THE EASY WAY TO START.
CCTLD DO NOT HAVE ISO3166, WHICH WE SAID.
IF YOU GO WITH THAT, DON'T EXPECT IT WILL CREATE AN IDN CCTLD EQUIVALENT.
IT'S NOT GOING TO HAPPEN.
THEY ARE NOD -- 3166 IS DESIGNED FOR POSTAL MAIL, CODE IS GOOD ENOUGH FOR ANYONE AND THERE'S NO INCENTIVE FOR ISO TO DO THE TRANSLATION AND GO THROUGH THE POLITICAL HURDLES TO GET CONSENSUS AND DO THE TRANSLATION.
IF YOU WAIT FOR THAT TO DEFINE THECCTLD, IT'S GOING TO BE DELAYED IDN DEPLOYMENT INDEFINITELY.
AND IT WILL BE A MINEFIELD FOR ANYONE WHO ATTEMPTS TO DO SO.
SO THINK ABOUT OUT OF THE BOX.
YOU DON'T LOOK AT THE LIST REFERENCE, BUT WHY NOT LIKE AT RULES.
STOP LOOKING AT LIST.
THINK OF THE RULES.
LOOK AT WHAT ARE THE PROCEDURES TO ACTUALLY REGISTER AN IDN CCTLD.
ICANN CAN COME UP WITH SOME RULES THAT SAY STEP ONE, STEP TWO, STEP THREE.
IF YOU FULFILL THE REQUIREMENTS, ONE, TWO, THREE, AND YOU GO THROUGH THE PROCESS, AND AT THE END OF THE PROCESS, YOU GET IT.
SOMETIMES YOU MAY NOT GO THROUGH A PROCESS, SOMETIMES YOU MAY, DEPENDING HOW IT GOES.
AND THE PROPOSED PROCEDURE, AS I WOULD SAY, AND REQUIREMENTS, IS, FIRST OF ALL, APPLICANTS MUST BE EXISTINGCCTLD.
AT THE LAST MINUTE, I THINK IT'S IMPORTANT THAT THECCTLD MUST HAVE DEMONSTRATED CAPABILITY TO HANDLE IDN IN THE FIRST PLACE.
SO YOU MUST DO IDN IN THE SECOND-LEVEL DOMAIN AND DEMONSTRATE YOU HAVE THAT CAPABILITY BEFORE YOU ACTUALLY QUALIFY.
YOU DON'T ALLOW ANY LANGUAGES FOR THE CCTLD -- THE LANGUAGE THAT YOU TRANSLATE, YOUR IDN CCTLD MUST BE A MAJOR LANGUAGE FOR THE REGION, MAJOR CAN BE DEFINED IN A COUPLE WAYS.
COULD BE DEFINED AS OFFICIAL LANGUAGE OR COULD BE DEFINED BY THE PERCENTAGE OF THE SPEAKERS FROM THAT REGION.
LIKE I SAY, MAY BE SOME FALL OFF THE RADAR, BUT THAT COULD BE HANDLED AS TIME GOES BY.
BUT THEREAFTER, WE DO NEED TO PUBLISH THE APPLICATIONS FOR THOSE IDN CCTLDS ON ICANN OR SOME OTHER WEB SITE FOR X NUMBER OF DAYS.
THAT COULD BE 30, 60, 90 DAYS, WHICHEVER.
THE REASON FOR THAT PUBLICATION, IT'S READY FOR PUBLIC COMMENTS AND FOR OFFICIAL OBJECTION.
THE REASON IS, WE REALIZE THERE WILL BE A FEW COUNTRIES IN DIFFERENT LANGUAGES THAT TRANSLATE INTO THEIR OWN NATIVE NAMES WHO END UP HAVING THE SAME STRING OF -- SAME STRING THAT YOU WOULD WANT TO USE FOR THE COUNTRY CODE AND THAT'S POTENTIAL CONFLICT.
IT IS IMPORTANT THAT WE PUT IT -- END PROPOSAL GOES THROUGH A PUBLIC COMMENT AND OFFICIAL OBJECTION.
AND HOW DO WE DO OBJECTION?
IF WE DO GET OBJECTION, I AM SO SORRY, YOU WILL NOT GO THROUGH.
LEAVE YOU ASIDE, AND THEN WE -- YOU LIKE TO PARTICIPATE IN CONFLICT TO WORK OUT ISSUES BEFORE WE ACTUALLY PROCEED.
THE PROS AND CONS OF THIS PROPOSAL THAT WE NEED TO DISCUSS IS IT'S REALLY -- IT'S A SMALL STEP.
IT ALLOWS NONCONTROVERSIAL IDN CCTLD TO PROCEED.
NOT EVERYONE GETS IT.
IT'S NOT A PERFECT SOLUTION.
BUT IT ALLOWS LIMITED EXPERIMENTATION.
IF ICANN OR THE COMMITTEE DO NOT FEEL ALREADY TO GO THAT FAR, THEN WE CAN ACTUALLY SAY THAT THIS IS A TRIAL, WE RUN IT FOR SIX OR 12 MONTHS BEFORE WE ACTUALLY PUT IT INTO LIVE OPERATIONS.
AT THE END OF THE DAY, I MENTIONED WE ARE NOT GOING TO CREATE HUNDREDS OF IDN CCTLDS THROUGH THIS PROCESS.
THERE IS ONLY THAT HANDFUL OF CCTLD WHO WILL BE INTERESTED TO CREATE THEIR OWN IDN CCTLD, MAYBE FROM THE MIDDLE EAST AND FROM THE CJK REGION.
IN THE MEANTIME, WHILE WE ACTUALLY START WITH THIS TEST AND TRIAL, WE CAN CONTINUE TO WORK ON SOME OF THE ISSUES LIKE THE (INAUDIBLE) ISSUES, THE OTHER ISSUE, GTLD AND STLD IN IDN.
I HAVE A RESPONSE TO JOHN'S PROPOSAL.
I KNOW JOHN MADE A PROPOSAL ABOUT THE IDN TLD.
I PUBLISHED IT ON (INAUDIBLE).
I WILL NOT GO TO IT TODAY.
I'M DONE.

>>RAM MOHAN: I HAVE SOMETHING TO SHOW TO THE GROUP HERE.
SO LET ME JUST TAKE THAT FOR A MOMENT.
IT'S ALL IN THE WRIST.
THANK YOU.
THIS IS FROM THE GULF DAILY NEWS FROM THE 28TH OF NOVEMBER.
THE ITU HAS A QUOTE THAT SAYS "INTERNET DOMAIN NAMES COULD BE MADE AVAILABLE IN ARABIC FOR THE FIRST TIME," AND EXPERTS ARE CONSIDERING HOW TO IMPLEMENT SUCH A MOVE, AND THE ITU IS WORKING ON HAVING MULTILANGUAGE DOMAIN NAMES.
FIRST TIME I HEARD THAT THE ITU IS ACTUALLY RUNNING DOMAIN NAMES.
BUT THE -- THERE ARE A NUMBER OF ISSUES THAT COME UP.
LET'S TALK ABOUT THE TWO PRESENTATIONS THAT WERE JUST MADE ON IDN.IDN.
AND I HAVE -- LET ME JUST LEAD WITH A FEW QUESTIONS.
SHOULD IDN -- I MEAN, JAMES, YOU'RE SAYING AND BOTH OF YOU ARE SAYING IDN.IDN, JAMES, YOU'RE SAYING CCTLDS SHOULD BE ALLOWED INTO THE ROOT.
WHAT ARE THE IMPLICATIONS?
WHAT'S IT MEAN?
HOW DOES THAT GET REPRESENTED?
WHAT ABOUT USER INTERFACE ISSUES?
AND FOR THIS ARTICLE THAT'S OUT THERE, MY OWN PERSONAL OPINION IS, I THINK IT'S DANGEROUS AND FOOLISH TO PLAN A LAUNCH SOMETHING LIKE THIS IN POTENTIALLY SOME SORT OF AN ALTERNATE ROOT.
AT THE SAME TIME, THERE IS CLEARLY A COMMUNITY THAT IS ASKING FOR IDN.IDN.
AND I THINK ONE OF THE ROLES HERE IS TO FIGURE OUT HOW TO SUPPORT THE COMMUNITY WHILE MAINTAINING STABILITY.
AND THAT'S FOOD FOR THOUGHT BOTH FOR THE PANELISTS HERE AND FOR THE AUDIENCE.

>>JAMES SENG: MAYBE YOU'LL LET ME ANSWER.
FIRST OF ALL, AS YOU CAN SEE FROM THIS ARTICLE, THAT IF -- THAT THERE ARE PEOPLE WHO ARE DEMANDING OR WHO ARE ASKING FOR IDN.IDN, IDN TLDS.
AND IF THE COMMUNITY IS NOT ABLE TO RESPOND IN AN APPROPRIATE TIME, IN A REASONABLE AMOUNT OF TIME, THEN THEY WILL FIND ALTERNATIVE AVENUES AND SOLUTIONS TO IMPLEMENT WHATEVER THEY DO.
I UNDERSTAND YOU HAVE YOUR OWN IDN IMPLEMENTATION IN YOUR OWN TEST BED.
AND THERE ARE OTHER EXAMPLES.
I WOULD NOT BE TOO ALARMED ABOUT THIS.
BUT THE FACT IS IT IS GROWING.
THERE WILL BE MANY OF THEM.
AS FAR AS THE REPRESENTATION OF THE ISSUES OF WHETHER WE CAN DO IDN.IDN IN THE ROOT ZONE FILE, FUNDAMENTALLY, IDN TLDS IS JUST A DOMAIN NAME.
THE REPRESENTING -- WHAT GOES INTO A ROOT IS NOT SOME A, B STRINGS, IT'S PUNYCODE, IT'S ASCII.
IT'S LIKE ADDING ANY TLDS.
THE PROBLEM IS REALLY THE NUMBERS THAT MAY POTENTIALLY GO INTO THE ROOT.
THAT WE DON'T REALLY KNOW THE EFFECT.
MAYBE SOME STUDIES NEED TO BE MADE IN GREATER DETAIL.
BUT THAT SHOULD NOT STOP SOME OF THE HANDFUL OF PEOPLE WHO IS ASKING FOR THE IDN.IDN TO GO AHEAD IN SOME LIMITED WAYS FOR ICANN TO EXPERIMENT, WHETHER THERE IS ANY REAL ISSUES FOR THE PUNYCODE TO GO INTO (INAUDIBLE), THAT'S ONE.
ON THE SECOND QUESTION OF REPRESENTATION AND WHETHER IT CAN ACTUALLY WORK.
VERIFY IT WORKS FOR ALL, WHETHER A SECOND-LEVEL DOMAIN OR A TOP-LEVEL DOMAIN.
SO THERE IS NO REASON A WELL-WRITTEN IDN A IMPLEMENTATION OF A CLIENT SHOULD RENDER ONLY THE SECOND AND THIRD LEVEL AND NOT THE TOP LEVEL.
UNDERSTAND THAT'S WHAT I-NAV DOES.
BUT THAT'S I-NAV.
IF YOU TAKE ANY WELL IMPLEMENTATION FROM MOZILLA TO IDN OPEN SOURCE IMPLEMENTATION TO OPERA AND TO SAFARI, IT ALL WORKS.

>>CHARLES SHABAN: JUST THIS ONE COMMENT ON THE ARTICLE YOU SHOWED.
WE HAD TWO YEARS BACK ARABIC.ARABIC.
ALMOST 40,000 DOMAIN NAMES WERE REGISTERED BY THE USUAL PEOPLE WHO WERE TRICKED AND SOME OF THE COMPANIES IS BANKRUPT.
SO IT WAS SOME KIND OF MAPPING TABLES AND SO ON, WHICH WASN'T CORRECT.
SO I JUST WANT TO CONFIRM JAMES' IDEA THAT WE SHOULD PUSH FOR IT AND TO BE AVAILABLE AS SOON AS POSSIBLE ON A STANDARD BASIC LIKE THE ASCII SO AS NOT TO HAVE THE SAME PROBLEM AGAIN.
AND I'M SURE IT HAPPENED IN THE OTHER LANGUAGES, NOT ONLY ARABIC.
THANK YOU.

>>RAM MOHAN: DO YOU THINK ICANN HAS A ROLE HERE?

>>CHARLES SHABAN: THE ROOT SERVER, IF IT DOESN'T CONTAIN THE ARABIC, THEY WILL START USING IT SOMEWHERE ELSE, LIKE TWO YEARS AGO.
SO I THINK, SURE, IT SHOULD HAVE A ROLE IN IT.

>>TINA DAM: THANKS.
I JUST WANTED TO GIVE A QUICK COMMENT SINCE ICANN IS MENTIONED SO MUCH.
RIGHT NOW, ICANN IS IN THE PROCESS OF DEVELOPING A PROCEDURE FOR IMPLEMENTATION OF NEW GTLDS, AND THERE'S A RANGE OF TOPICS THAT NEEDS TO BE DISCUSSED AND DECIDED UPON BEFORE THAT ESSENTIALLY CAN HAPPEN.
AND I IMAGINE THAT ADDITIONAL TOPICS WILL OCCUR AT A LATER STAGE AS WELL, SO IT'S LIKE AN ONGOING PROCESS.
ONE IMPORTANT TOPIC IS, OBVIOUSLY, IDN.IDN.
SO I JUST WANTED TO SAY THAT THE PROCESS IS GOING TO BE IN PLACE.
IDN.IDN IS SOMETHING WE VERY MUCH WANT TO HAVE THE OPPORTUNITY TO CONTINUE TALKING ABOUT AND GETTING THE DIFFERENT OPINIONS ABOUT WHAT THE PROBLEM IS AND HAVING SOME DISCUSSIONS ON IT, JUST LIKE YOU'RE DOING TODAY AND LIKE YOU DID EARLIER TODAY IS VERY VALUABLE INPUT INTO THAT PROCEDURE SO DECISIONS CAN BE MADE ON IT.

>>RAM MOHAN: THANK YOU.
I THINK ONE OF THE THINGS YOU ARE HEARING FROM THE PANEL, AT LEAST SOME OF THE PANELISTS HERE, IS THAT ICANN HAS A CLEAR ROLE AND ICANN CAN PLAY A ROLE IN MAKING SURE THAT WE DON'T HAVE ROGUE IMPLEMENTATIONS.
>>TINA DAM: YEAH.
AND I THINK THAT'S UNDERSTOOD.
AND ALL THAT INFORMATION, TOGETHER WITH THE SUGGESTION THAT JAMES WAS MAKING WAS CERTAINLY SOMETHING THAT'S GOING TO BE BROUGHT BACK AND MAYBE CAN FACILITATE AND LAUNCH A LARGER DISCUSSION ON THE TOPIC.
>>JAMES SENG: JUST A QUICK CLARIFICATION.
THE PROPOSED PROCEDURE THAT I MADE WOULD WORK FORCCTLD BUT PROBABLY WOULD NOT WORK FOR GTLD.
>>TINA DAM: NO, I UNDERSTAND.

>>RAM MOHAN: ANY OTHER COMMENTS FROM THE AUDIENCE?
OKAY.
TIME TO MOVE ON TO ELIZABETH.
>>ELIZABETH LONGWORTH: THANKS VERY MUCH.
MY NAME'S LIZ LONG WORTH, AND I'M FROM -- I'M NOT GOING TO DO A PRESENTATION -- FROM UNESCO, WHICH, FOR THOSE OF WHO YOU ARE NOT AWARE, THAT'S THE UNITED NATIONS AGENCY THAT SPECIALIZES IN EDUCATION, SCIENCES, COMMUNICATION, AND CULTURE.
AND I AM THE DIRECTOR OF THE INFORMATION SOCIETY DIVISION IN THE COMMUNICATION AND INFORMATION SECTOR.
SO OUR MANDATE IS ONE WHICH THE FOUNDING MEMBERS OF UNESCO, AFTER WORLD WAR II IN THE 1940S WERE VERY FARSIGHTED.
BECAUSE THEY SAID THE MANDATE HAS TO BE ABOUT THE FREE FLOW OF INFORMATION, IDEAS, WHETHER BY WORD OR BY IMAGE.
AND, OF COURSE, THAT PREDATES ANY DREAMS OF THE INTERNET.
SO THE ISSUES THAT WE'RE TALKING ABOUT TODAY ARE EXTREMELY TOPICAL FOR US.
AND WHAT I WANT TO DO IS I WANT TO POSE A QUESTION TO YOU, BECAUSE THERE ARE SO MANY IDEAS AND THERE ARE SO MANY ISSUES THAT NEED TO BE WORKED ON FROM A POLICY PERSPECTIVE THAN I WANT TO HEAR FROM YOU WHAT TYPES OF ENGAGEMENT ARE NEEDED AND BY WHOM.
BECAUSE THERE ARE TWO OPPORTUNITIES COME UP THAT I WILL TELL YOU ABOUT A LITTLE LATER THAT WILL HAVE AN INTERNATIONAL AUDIENCE.
AND THEY MIGHT BE THE OPPORTUNITY TO GATHER UP SOME OF THESE IDEAS AND BE VERY CONSTRUCTIVE IN TRYING TO CHANNEL THEM INTO THE RIGHT AVENUES.
BUT FIRST I WANT TO PICK UP ON THE THEME THAT WAS SO ELOQUENTLY INTRODUCED BY MR. SAMASSEKOU AND MR. SOMÉ TODAY.
AND I WANT TO PUT THE POLITICAL DEBATE INTO ITS SOCIAL, AND PERHAPS EVEN ETHICAL, CONTEXT.
SO I WANT TO TAKE US UP FROM THE LEVEL OF DETAIL THAT WE HAVE BEEN TALKING AND SHARE WITH YOU WHY WHAT YOU DO IS SO IMPORTANT FOR SO MANY PEOPLE IN THE WORLD AND WHY WE NEED TO MAKE SURE THAT WE ARE BROADENING OUT THIS DISCUSSION AND ARE ALWAYS MINDFUL OF WHERE THIS DISCUSSION SITS IN THE WIDER POLICY FRAMEWORK.
THE REASON THERE'S SO MUCH SCRUTINY AROUND IDN AT PRESENT IS THE GROWING REALIZATION OF ITS SIGNIFICANCE FOR CULTURE AND COMMUNICATION, FOR PARTICIPATION, AND FOR DEVELOPMENT.
AND PERHAPS IF I REFER BACK TO SOME OF OUR STRATEGIC OBJECTIVES IN UNESCO TO ILLUSTRATE THIS.
WE IN OUR PLANS TALK ABOUT A GOAL CALLED UNIVERSAL ACCESS.
AND THE IDEA IS THAT WE WANT TO INTRODUCE THE UNIVERSALITY OF ACCESS TO THE TECHNOLOGY AND ACCESS, MOST IMPORTANTLY, TO THE KNOWLEDGE THAT SITS BEHIND THAT TECHNOLOGY, WHERE THE ICTS ARE A MEDIA.
THE OTHER THEME THAT RUNS THROUGH OUR MANDATE IS THE NEED TO RESPECT AND PROMOTE CULTURAL DIVERSITY.
AND THERE ARE ALL DIFFERENT ELEMENTS OF THAT.
THERE'S THE DIVERSITY OF CULTURES, SOCIETIES, KNOWLEDGE SYSTEMS, AND THE ENRICHING EFFECT IF WE CAN OPEN UP THE WINDOW AND IF WE CAN USE THE TECHNOLOGY AS A WINDOW ON THESE DIFFERENT CULTURES.
BUT THE END GOAL IS ALWAYS TO ACHIEVE AN ACCELERATED SOCIAL AND ECONOMIC DEVELOPMENT, BUT A DEVELOPMENT WHICH IS INCLUSIVE AND THAT EMPOWERS PEOPLE WHERE PEOPLE CAN GENERATE THEIR OWN CONTENT, WHERE THEY CAN ACCESS LOCALIZED CONTENT, WHERE THEY CAN MAKE THEIR OWN DECISIONS, WHERE THEY CAN REALLY BE FULL PARTICIPANTS IN WHAT UNESCO CALLS KNOWLEDGE SOCIETIES, BROADENING IT OUT FROM THE TECHNOLOGICAL DEBATES.
AND THAT WAS THE MESSAGE THAT WE TOOK TO THE WORLD SUMMIT.
AND WE TRIED TO PUT THE ISSUES THAT YOU'RE TALKING ABOUT INTO THAT BROADER FRAMEWORK.
BECAUSE IF YOU THINK ABOUT LANGUAGE AND THE WAY THAT GETS REPRESENTED IN THE IDN SYSTEM, FOR EXAMPLE, LANGUAGE IS AN ISSUE OF ACCESS, IT'S AN ISSUE OF CULTURE, OF LITERACY, HERITAGE, EMPOWERMENT.
IT'S WHAT ENABLES THE DIALOGUE, AND, HOPEFULLY, THE PEACEFUL -- THE HARMONIOUS COEXISTENCE.
AND IF OUR TECHNICAL SYSTEMS, WHETHER IT BE THE IDN, THE PLATFORMS THAT WE'RE USING, THE GATEWAYS TO THE WEB, WHATEVER, IF THAT ARCHITECTURE DOES NOT RESOLVE THESE LANGUAGE CONCERNS, THEN, OF COURSE, YOU HAVE A GROWING NUMBER OF PEOPLE IN THE WORLD WHO SEE THAT AS A BARRIER, A BARRIER TO ACCESS, BARRIER TO RECOGNIZING THEIR CULTURES, TO REPRESENTING THEIR CULTURES.
SO FOR UNESCO, WE ARE HUGELY INTERESTED IN THE ISSUE OF LANGUAGE AND THE IMPLICATIONS IT HAS FOR THE INTERNET.
WE SHOW THAT INTEREST IN A RANGE OF WAYS.
WE HAVE THE ANGLE OF PRESERVATION OF LANGUAGES, BECAUSE AT THE CURRENT TIME, WE ESTIMATE ABOUT [??] % OF THE WORLD'S 6,000 LANGUAGES ARE NOT REPRESENTED ON THE INTERNET.
WE HAVE A PROGRAM CALLED INITIATIVE BABEL, WHICH IS AIMED AT THE PRESERVATION OF LANGUAGES AND INDIGENOUS KNOWLEDGE SYSTEMS.
WE HAVE OTHER WORK ON SCRIPT ENCODING WITH UNICODE PROPOSALS.
BUT OUR INTEREST GOES WAY BEYOND THE PRESERVATION.
AND AS I MENTIONED, THAT WE'RE NOT JUST TALKING ABOUT SURVIVAL OF LANGUAGES IN A DIGITAL WORLD, BUT WE'RE TALKING ABOUT DIVERSITY AND PARTICIPATION VIA THE INTERNET.
SO YOU MAY BE INTERESTED TO KNOW THAT THIS NOVEMBER, WE OBTAINED THE RESULTS FROM SOME RESEARCH THAT UNESCO HAS SPONSORED THAT'S SHOWING A CHANGE IN THE TRENDS AWAY FROM THE GLOBAL USE OF ENGLISH.
OUR RESEARCH CONFIRMS THAT THE WORLDWIDE WEB IS DEFINITELY FRACTURING INTO LANGUAGE COMMUNITIES AND THAT THOSE COMMUNITIES ARE BROADENING OUT BEYOND THE STANDARD EUROPEAN LANGUAGES.
AND AS THE TECHNICAL PROBLEMS WITH NON-ROMAN ALPHABET ARE BEING OVERCOME, THEN YOU SEE, FOR EXAMPLE, IN THE ARAB COUNTRIES, CHINESE AND JAPANESE WEB SITES.
THOSE AREAS, OF COURSE, AND THOSE LANGUAGES, THE NUMBER OF WEB SITES ARE INCREASING EXPONENTIALLY.
AND THE USE OF REGIONAL AND MINORITY LANGUAGES IN A SYMBOLIC WAY IS ALSO A LOT MORE WIDESPREAD NOW.
AND I SAY SYMBOLIC BECAUSE A NUMBER OF THOSE LANGUAGES ARE PROMOTED BY BEING GIVEN OFFICIAL LANGUAGE STATUS.
SO WE STILL HAVE THE IMPEDIMENT, OF COURSE, TO THE DIVERSITY IN THE FORM OF THE VERTICAL PENETRATION, AND TO SMALLER AND LESSER-USED LANGUAGE COMMUNITIES DUE TO THE HARDWARE OWNERSHIP, THE LEVELS OF IT LITERACY AND INTERNET CONNECTIVITY OR ACCESS, TECHNICAL ACCESS.
SO THERE'S A LONG WAY TO GO.
BUT THERE'S ALSO SOME VERY INTERESTING TRENDS.
FOR THOSE OF YOU THAT ARE FEELING THE PRESSURE OF THE SCRUTINY ON THE VARIOUS SYSTEMS THAT ARE IN PLACE, AND PROCESSES, IF I COME BACK TO THE ISSUE OF DOMAIN NAMES, YOU CAN EXPECT THERE TO BE INCREASING SCRUTINY, BECAUSE IT IS FUNDAMENTALLY PERCEIVED AS AN ISSUE OF ACCESS AND PARTICIPATION.
KNOWLEDGE AND THE HUMAN CAPACITIES THAT WE HAVE TO APPLY KNOWLEDGE ARE THE KEYS TO ACCELERATING DEVELOPMENT, AND IT'S THE KEY TO FACILITATING INTERCULTURAL DIALOGUE.
SO BUILDING ON OUR EXPERIENCE WORKING WITH DIFFERENT LANGUAGE COMMUNITIES THAT DO NOT SHARE LATIN SCRIPT -- AND I MIGHT ADD THAT OUR WEB SITE, WEB WORLD, HAS LOTS OF EXAMPLES OF DIFFERENT PROJECTS THAT WE'VE BEEN INVOLVED WITH -- WE HAVE TWO PRACTICAL AREAS WHERE WE HOPE TO ENGAGE IN THE NEXT FEW MONTHS.
SHORTLY, WE HOPE TO OFFER A MUTUAL PLATFORM BY WHICH THE ISSUES OF THE ARABIZATION OF DOMAIN NAMES CAN BE ADDRESSED. AND THE IMPORTANCE OF THIS IS ILLUSTRATED IF YOU REFLECT THAT THE INTERNET PENETRATION ON THE ARAB WORLD IS ONLY 1.5%, AND THAT'S AN OLD FIGURE.
NOW, WE ARE HOPING TO COLLABORATE WITH A WIDE RANGE OF STAKEHOLDERS IN THE ARAB REGION SO THAT WE CAN AT UNESCO HOST IN THE REGION FACE-TO-FACE EXPERT MEETINGS ON THE TECHNICAL ISSUES.
AND WE HOPE TO DRAW ON OUR EXPERIENCE WORKING WITH OUR LANGUAGE EXPERTS IN ASIA WHO FACE THE SAME CHALLENGES.
AND, OF COURSE, ALSO HOST THE MEETINGS ON THE POLICY ASPECTS, BECAUSE AS YOUR LITTLE CLIP SHOWS, THERE ARE MANY, MANY INITIATIVES UNDERWAY, AND CHARLES WOULD BE ABLE TO SPEAK TO THAT AS WELL.
AND THE LOST OPPORTUNITY IS THAT WE NEED A FORUM WHERE THESE PEOPLE CAN ACTUALLY SHARE THOSE IDEAS, WHICH LEADS ME TO THE SECOND PRACTICAL ENGAGEMENT.
UNESCO HAS ESTABLISHED SOMETHING WE NICKNAMED KNOWLEDGE COMMUNITIES. BUT IT'S BASICALLY AN ONLINE COMMUNITY OF PRACTICE ON MULTILINGUALISM IN CYBERSPACE.
AND WE ARE HOPEFUL THAT THE SUBCOMMUNITIES WITHIN THAT COULD USE THAT AS A PLATFORM TO SHARE IDEAS, MORE THAN IDEAS, PROPOSALS, TECHNICAL SOLUTIONS.
THERE SEEMS TO BE A NEED FOR A FACILITY WHERE YOU CAN HAVE THE COLLECTIVE DIALOGUE.
SO STAYING WITH OUR ROLE AS THE FACILITATOR, MR. ADAMA SAMASSEKOU MENTIONED AN EVENT THAT WE'RE PLANNING IN THE CONTEXT OF THE WORLD SUMMIT IN MAY NEXT YEAR WHICH IS A MEETING ON MULTILINGUALISM IN CYBERSPACE WHICH WILL BE HOSTED IN MALI. BUT AGAIN WHAT I WANT TO SAY TO YOU IS THIS IS AN OPPORTUNITY TO BRING INTERESTED PARTIES TOGETHER TO SEE WHAT'S BEING DONE, BUT MORE IMPORTANTLY TO SEE WHERE SHOULD WE MOVE TO TO TRY AND PULL ALL THESE ISSUES TOGETHER, AND NAVIGATE A PATH TO SOME RESOLUTION.
HOPEFULLY, AT THAT MEETING WE COULD ELEVATE THE POLICY SIGNIFICANCE OF THE ISSUES THAT WE'VE BEEN TALKING ABOUT, BUT WHAT'S NEEDED I THINK AT THE INTERNATIONAL LEVEL IS YOU'VE GOT REGIONS WORKING ON THESE ISSUES BUT REALLY APART FROM MEETINGS SUCH AS THIS, VERY LITTLE OPPORTUNITY TO SHARE WHAT THEY ARE DOING AND TO SHARE THE SOLUTIONS.
AND THIS ISSUE CAN BE EXTREMELY DIVISIVE IF YOU ALLOW IT TO BE, AND WHAT UNESCO HOPES IS THAT THE ENGAGEMENT CAN BE CHANNELED IN A WAY THAT IS EXTREMELY CONSTRUCTIVE, BECAUSE THERE'S SO MUCH AT STAKE.
THERE'S THE PEACEFUL AND RESPECTFUL INTERCULTURAL DIALOGUE THAT'S AT STAKE; THE ACCESS, WHICH LEADS TO THE PARTICIPATION, WHICH LEADS TO THE EMPOWERMENT OF PEOPLE WHO ARE NOT REPRESENTED HERE, AND OF COURSE THE NEED TO SAFEGUARD A VITAL GLOBAL FACILITY, WHICH IS INTRINSICALLY LINKED TO THE SOCIAL AND ECONOMIC PROGRESS OF EVERYBODY.
AND I HAVE A FEELING THAT WE'RE ONLY AT THE BEGINNING OF THE JOURNEY. AND WE HOPE THAT IN OUR COLLECTIVE DIALOGUE, WHETHER IT BE IN MALLI OR WHETHER IT BE IN AN ONLINE COMMUNITY OF PRACTICE, OR WHETHER IT BE IN THE CONTEXT OF THE DIFFERENT PROJECTS THAT I MENTIONED, THAT ALL OF THIS EFFORT WILL HELP US TO ADVANCE WHAT I'M SURE IS A SHARED VERSION OF THE INTERNET THAT CAN BENEFIT ALL OF HUMANITY AND I THINK IT IS TIME TO ACTUALLY TRY TO PULL TOGETHER SOME OF THE DIFFERENT IDEAS. A LOT OF PEOPLE ARE WORKING IN THAT SPACE AND TRYING TO BE VERY CONSTRUCTIVE. BUT I WANT TO HEAR YOUR IDEAS ONCOMING BACK TO THAT FIRST QUESTION: WHAT SHOULD BE THE NATURE OF ENGAGEMENT HERE? HOW CAN WE SHARE THESE IDEAS AND ELEVATE THE IDEAS THAT LOOK THE MOST PROMISING, LOOK LIKE THEY COULD PROVIDE A SOLUTION.
SO I HAND BACK TO YOU. JUST SOME PRELIMINARY COMMENTS ON THE BROADER CONTEXT.
>>RAM MOHAN: VERY THOUGHTFUL PRESENTATION.
I HAVE A COUPLE OF COMMENTS, BUT BEFORE I GET THERE, I WANT TO OPEN IT UP TO THE FLOOR, TO THE -- IF THERE ARE ANY COMMENTS FROM THE FLOOR, AS WELL AS FROM THE PANELISTS HERE FOR ELIZABETH'S PRESENTATION.
HOW ABOUT THE PANELISTS? DO YOU HAVE SOME THOUGHTS ON ELIZABETH'S FIRST QUESTION, WHICH IS WHAT -- IT'S A "WHAT" QUESTION; RIGHT?
>>HONG XUE: I HAVE A QUICK COMMENT. WE'VE BEEN TALKING ABOUT THE CULTURAL DIVERSITY AND THE INTERNET ACCESS. SINCE WE ARE HAVING THE MEETING IN AFRICA, IT'S REALLY SAD WHEN WE LEARN THAT IN THIS CONTINENT, THE INTERNET PENETRATION IS ONLY 2%.
SO HOW TO BRING THE REST 98% USERS -- THE PEOPLE TO THE INTERNET? THIS IS REALLY AN ISSUE WE SHOULD LOOK AT.
LANGUAGE IS A SUBSTANTIAL BARRIER TO PREVENT THOSE PEOPLE, ORDINARY PEOPLE, TO ACCESS THE INTERNET. SO THAT AS ELIZABETH PRESENTS, IDN COULD BE A SOLUTION TO FACILITATE THE INTERNET ACCESS OF THOSE PEOPLE.
THANK YOU.
>>RAM MOHAN: JOHN.
>>JOHN KLENSIN: I WANT TO APOLOGIZE. IT'S BEEN A VERY LONG DAY AND I'M EXHAUSTED AND THOSE OF YOU SITTING ON THE PLATFORM MUST BE AS WELL. SO I'M NOT CERTAIN THAT WHAT I'M ABOUT TO SAY IS COHERENT.
ELIZABETH, MY REACTION IS THAT YOU'RE RIGHT, AND MY REACTION IS FURTHER, AND HAS BEEN, THAT MOST OF THESE CULTURAL ISSUES AND THE LANGUAGE ISSUES, THE SCRIPT ISSUES HAD BETTER BE IN YOUR HANDS OR HANDS OF ORGANIZATIONS MUCH LIKE YOURS, AND I DON'T KNOW OF ANY OTHERS, AND NOT IN THE HANDS OF THE ICANNS AND IETFS OF THE WORLD BECAUSE WE WILL MESS IT UP.
DRAWING THAT BOUNDARY PROPERLY IS A TRICKY MATTER. IT INVOLVES BEING VERY CLEAR ABOUT THE DISTINCTIONS BETWEEN CONTENT AND NAMES AND LABELS AND IDNS, WHICH WE WERE TALKING ABOUT EARLIER.
IT REQUIRES NOT BEING FACILE ABOUT HOW FAST WE CAN DEPLOY IDNS SOMEWHERE, SOMEHOW, SOME WAY, BUT RATHER SPENDING THE TIME TO ACTUALLY GET IT RIGHT BECAUSE ONE OF THE PROPERTIES OF DOING MANY OF THESE THINGS ON THE INTERNET IS THAT IF YOU DO THEM IN A WAY WHICH IT IS WRONG BUT NOT FATAL, YOU NEVER GET RID OF IT.
AND WRONG BUT NOT FATAL IS ONE OF THE WORST FORMS OF IT LOOKS RIGHT TO PEOPLE WHO ARE NOT FAMILIAR WITH THE NEEDS OF A PARTICULAR CULTURE BUT IT'S WRONG WITHIN THAT CULTURE.
SO I AT LEAST LOOK FORWARD AND HOPE THAT WE CAN FIND A WAY TO WORK THOSE BOUNDARIES TO TRY TO UNDERSTAND WHAT ISSUES ARE PROPERLY HANDLED IN PLACES LIKE UNESCO AND WHAT ISSUES ARE PROPERLY HANDLED IN PLACES LIKE ICANN, AND TRY TO MAKE CERTAIN THAT WE MAKE THOSE ALLOCATIONS CAREFULLY, KEEP TALKING WITH EACH OTHER, BUT TRY NOT TO DO THINGS IN WHICH WE ARE INCOMPETENT OR THINGS IN WHICH UNESCO DOESN'T HAVE SPECIAL EXPERTISE.
AND I THINK IF IT WERE THREE HOURS EARLIER, I MIGHT HAVE SOMETHING MORE COHERENT TO SAY, BUT AT THIS POINT I DON'T.
>>ELIZABETH LONGWORTH: I THOUGHT YOU WERE PERFECTLY COHERENT.
YES. WELL, WE DO HAVE A MANDATE IN THE THE AREA. VERY MUCH SO, BASED AROUND THE COMMENT THAT YOU WERE MAKING ABOUT THE IMPORTANCE OF ACCESS, WHICH OUR AGENCY AND MANY OTHER AGENCIES ARE WORKING SO HARD TO TRY AND DO SOMETHING ABOUT.
BUT ALSO, IN TERMS OF THE THEME OF CULTURAL DIVERSITY.
AND WE HAVE FOLLOWED THROUGH ON THE MANDATE IN THAT WE HAVE MANY PROJECTS IN THIS AREA, AND WE'RE WORKING IN THE SPACE. BUT I SENSE THAT THERE'S SOMETHING MUCH BIGGER. THERE'S SOMETHING MUCH BIGGER SITTING IN BEHIND ALL THIS WORK. AND BY THAT I MEAN YOU HAVE, UP UNTIL NOW, HAD SO MUCH ACTIVITY IN DIFFERENT PARTS OF THE WORLD BY DIFFERENT LEVELS OF PEOPLE, FROM TECHNICAL THROUGH TO THOSE THAT HAVE BEEN WORKING SO HARD WITHIN THE VARIOUS COMMITTEES AND ICANN. BUT HOW DO YOU MOVE IT FROM THERE? YOU KNOW, WE CAN EXPLORE WITH PLATFORMS, ELECTRONIC PLATFORMS, TO PUT PEOPLE IN TOUCH AND TO LET THEM -- TO FACILITATE THEM SHARING SOLUTIONS AND IDEAS, AND OF COURSE THE ICANN PROCESS HAS ITS OWN PLATFORMS AS WELL. THERE'S THE POLITICAL ACTIVITY THAT CAN HAPPEN. UNESCO HAS AN INTERNATIONAL INSTRUMENT IN THIS AREA, AND IT'S CALLED THE RECOMMENDATION ON MULTILINGUALISM AND UNIVERSAL ACCESS IN CYBERSPACE. AND IT WAS RATIFIED BY 189 COUNTRIES LAST YEAR AT OUR GENERAL CONFERENC!
E.
AND THE MEMBER STATES HAVE AN OBLIGATION TO FOLLOW THROUGH ON THAT RECOMMENDATION.
THE QUESTION IS WHETHER THERE IS SOMETHING ELSE IN THE AREA OF MULTILINGUALISM THAT'S NEEDED.
AND THEN COMING BACK TO YOU, JOHN, WHAT STRUCTURES, WHAT PROCESSES, OR DO YOU HAVE JUST A VERY LOOSE FEDERATION OF ISSUES? IT'S VERY -- IT'S VERY DIFFICULT.
WHAT I'M POSING -- WHAT I'M PUTTING TO YOU IS I'M SAYING THERE IS AN OPPORTUNITY TO BRING A LOT OF PEOPLE TOGETHER IN MALI NEXT YEAR, AND WE ARE GOING TO TALK ABOUT POLICY ISSUES. WE'RE ALSO GOING TO TRY AND SHOWCASE SOME OF THE INNOVATIVE ACTIVITIES THAT ARE OUT THERE, SO PEOPLE CAN GET IDEAS FROM THAT. BUT IT IS AN OPPORTUNITY, ALSO, TO PERHAPS LOOK FOR SOMETHING A LITTLE BIT MORE OUT OF THAT MEETING, AND SO I JUST PUT THE QUESTION BACK, IF ANYBODY WANTS TO MAKE SUGGESTIONS.
>>JOHN KLENSIN: TO ANSWER YOUR SPECIFIC SUGGESTION, I'VE GOT QUITE A BIT MORE THAN A VAGUE FRAMEWORK CONCEPT, BUT NOT AT THIS HOUR OF THE AFTERNOON. AND THE ONE COMMENT I WOULD MAKE, AT THAT BOUNDARY WHERE THE FRAMEWORK BECOMES THE VAGUE REALITIES. NEITHER UNESCO OR ICANN OUGHT TO BE REQUIRING THINGS THAT REQUIRE COUNTRIES TO VIOLATE THEIR OWN CONSTITUTIONS IN ORDER TO MAKE SOMETHING TECHNICALLY CONVENIENT.
>>RAM MOHAN: CARY.
>>CARY KARP: THE DOCUMENT TO MULTILINGUALISM IN CYBERSPACE WHICH YOU JUST CITED MAKES REFERENCE TO FORGING ALLIANCES WITH MULTILINGUAL DOMAIN NAMES. AND I'M PRETTY SURE THAT THAT SENTENCE WAS VERY CAREFULLY CRAFTED TO HARMONIZE WITH THE IDN INITIATIVES. HOW DID THAT RECOMMENDATION END UP BEING WORDED AS WHAT? ESPECIALLY POINT 14 IN IT WHICH MAKES EXPLICIT REFERENCE TO DOMAIN NAMES.
>>ELIZABETH LONGWORTH: YES, AND I CHAIRED THAT SESSION AND I CAN'T ANSWER YOUR QUESTION. BUT I PROMISE YOU IT WAS CAREFULLY CRAFTED. IN FACT, WE AGONIZED OVER THAT TEXT AND IT WENT THROUGH ITERATION AFTER ITERATION.
BUT I HAVE TO SAY THAT WHEN THE WORDING WAS FIRST PROPOSED IN THAT DOCUMENT, IT PREDATED THE CURRENT SCRUTINY, PREDATED THE LANGUAGE INTERNET GOVERNANCE.
SO I DON'T KNOW THAT WE SHOULD READ MORE INTO IT THAN WHAT -- YOU KNOW, THAN WHAT'S IN THE BLACK-AND-WHITE. BUT IT DOES PROVIDE A VERY USEFUL FRAMEWORK, THAT DOCUMENT, FOR COUNTRIES, MEMBER STATES, TO ACTUALLY THINK ABOUT HOW THEY'RE GOING TO REFLECT THESE ISSUES AND MOVE ON THEM.
OF COURSE, WHAT WE'RE DOING HERE TODAY IS WE'RE TALKING ABOUT INTERNATIONAL ALLIANCES AND INTERNATIONAL STRUCTURES THAT HAVE BEEN SET UP, AND WHAT IS NEEDED.
JOHN.
>>JOHN KLENSIN: I WAS JUST GOING TO MAKE THE OBSERVATION, AND IT'S ONE OF THE REASONS WHY WE NEED TO WORK CAREFULLY ON THIS, AND TOGETHER, ONE OF THE INTERESTING THINGS ABOUT THAT STATEMENT, WHICH I ALSO FOUND VERY HELPFUL AND INTERESTING, IS THAT IT TURNS OUT THAT IN SPITE OF THE FACT THAT THE TERM HAS GOTTEN INTO THE VOCABULARY, NOBODY WANTS MULTILINGUAL DOMAIN NAMES. DOMAIN NAMES IN MANY LANGUAGES, YES, BUT MULTILINGUAL DOMAIN NAMES ARE PROBABLY INSANITY.
>>RAM MOHAN: I HAD A COUPLE OF QUESTIONS AND COMMENTS AS WELL.
AS YOU WERE MENTIONING TODAY, IDNS, BOTH THE POLICY AND THE TECHNOLOGY COMPONENTS, ARE BEING DISCUSSED IN WHAT TO ME SEEMS A LOT OF DISCUSSION HAPPENING IN LOCAL FORA. AND THEN SOME OF THE LARGER ISSUES BUBBLING UP TO GLOBAL AREAS.
SO A CCTLD, A LOCAL NIC, DECIDES THIS IS THE -- BEGINS WITH TECHNOLOGY. IT SAYS HERE IS THE TECHNOLOGY THAT I WANT TO IMPLEMENT. I WANT TO GET AN IDN NAME OUT THERE. AND THEN IT PUTS OUT A TABLE. IT THEN JUMPS INTO POLICY. THERE'S A GRAY -- THERE'S A LINE THERE THAT GETS CROSSED. AND THEN, BEFORE LONG, YOU HAVE A TLD, A GTLD, A CCTLD, SOMEBODY, SAYING WE HAVE PUBLISHED THE DEFINITIVE GUIDE TO HOW TO REPRESENT THIS LANGUAGE. NO ONE ELSE SHALL. AND IF ANYONE ELSE DOES, THEN THERE IS A PROBLEM.
HOW DO YOU THINK THE PROCESS THAT YOU'RE SUGGESTING, HOW DO YOU THINK THAT COULD HELP OR ADDRESS THE SITUATION? BECAUSE WE'VE JUST SEEN THE START OF IT. I FEAR WE'RE GOING TO HAVE LOTS MORE OF IT.
>>ELIZABETH LONGWORTH: WELL, THAT'S WHAT WE'RE LOOKING FOR, WHAT SHOULD THE PROCESS BE.
AND IT'S VERY DIFFICULT, BECAUSE YOU NEED TO HAVE FACILITATORS WHO CAN WORK WITH DIFFERENT LANGUAGE GROUPS AND REALLY THE PEOPLE NEED TO BE INVOLVED AND THEY NEED TO RESOLVE THE VARIATION AROUND THE CHARACTER SETS. BUT HOW DO YOU DO THAT IN A WAY THAT DOESN'T TRIGGER THE SORT OF CONCERN THAT JOHN WAS GETTING AT OVER PROTECTING THE UNIQUE IDENTIFIER SYSTEM THAT WE HAVE. AND IT'S A VERY, VERY DELICATE TASK.
AND THE OTHER DIFFICULTY IS THAT WHEN YOU LOOK AROUND AT WHO ARE THE STAKEHOLDERS INTERNATIONALLY, AND WE HAVE NATIONAL LANGUAGE INSTITUTES IN DIFFERENT PARTS OF THE WORLD WHO HAVE A VITAL ROLE TO PLAY, BUT WE HAVE TO REMEMBER THAT LANGUAGE RUNS ACROSS NATIONAL BOUNDARIES.
SO I THINK IT'S INCREDIBLY HEALTHY THAT IT'S BEING DEBATED. I THINK THAT'S ACTUALLY A REALLY, REALLY GOOD SIGN, AND THAT GRADUALLY, IN THAT DEBATE, PEOPLE BECOME MORE INFORMED. THAT'S WHY I SUSPECT THAT IN THE LANGUAGE THAT JOHN WAS QUESTIONING, OR IT MIGHT HAVE BEEN CARY, IN THE RECOMMENDATION THAT NO ONE IS SUGGESTING THAT LANGUAGE DROP DOWN INTO THE INTRICACIES OF THE UNIQUE IDENTIFIER SYSTEM AND WHAT MIGHT LIE BEHIND THAT LANGUAGE. WHAT THEY WERE TRYING TO DO IS SEND A POLITICAL SIGNAL THAT WE HAVE TO TALK ABOUT THIS AND WE HAVE TO ELEVATE THIS ISSUE, AND THAT WAS THE PURPOSE BEHIND THAT RECOMMENDATION. AND THAT'S INCREDIBLY HELPFUL. BUT NOW IT NEEDS THE TECHNICAL PEOPLE WHO HAVE BEEN DEBATING IN FORA LIKE THIS TO COME TOGETHER WITH THOSE WHO SHARE THAT WIDER POLICY CONCERN AND TRY TO GET AN ACCOMMODATION.
AND SO I HAVEN'T GOT A PROCESS TO SUGGEST. I'M ACTUALLY LOOKING TO SEE WHAT CAN BE DONE CONSTRUCTIVELY HERE. AND JUST HAVING A SESSION LIKE THIS IS A VERY POSITIVE THING. AND MAYBE CHARLES HAS GOT A COMMENT TO MAKE IN THIS AREA.
>>CHARLES SHABAN: WELL, I'M WITH YOU. I THINK THE MOST IMPORTANT THING IS TO COME FROM THE LOCAL PEOPLE, FOR THE LINGUISTIC. THE TECHNICAL WILL NEED GUIDANCE OF COURSE FROM THE INTERNATIONAL FORUMS SO EVERYBODY WILL BE, BUT TO BE SURE THAT NOBODY WILL SAY, AS I THINK EVERYBODY WITNESSED SOME OF THE EXAMPLES, HOW COME OTHER PEOPLE DEFINE, I WILL NOT ACCEPT THE CHINESE TO DEFINE THE ARABIC LANGUAGE. AT THE SAME TIME, I'M SURE THEY WOULD NOT ACCEPT THE ARABIC PEOPLE TO DEFINE THE CHINESE LANGUAGE BECAUSE THEY KNOW MORE ABOUT IT.
SO I THINK THERE SHOULD BE SOME KIND OF COORDINATION BETWEEN THE DIFFERENT TECHNICAL BODIES AND LINGUISTIC BODIES AND SOME KIND OF INTERNATIONAL FACILITATOR TO MOVE, FOR EXAMPLE, WHEN WE HAVE FINALLY A STANDARD IN ARABIC, THAT EVERYBODY CAN USE, MAYBE, WHEN THEY WANT TO IMPLEMENT ARABIC IN THEIR CCTLD OR GTLD OR SO ON.
THAT'S IT.
>>RAM MOHAN: CARY.
>>CARY KARP: THERE HAVE BEEN, SINCE THE OUTSET OF THIS DISCUSSION, IT'S BEEN BIPOLAR. ONE HAS HAD THE AGENCY REPRESENTING THE TECHNICAL ASPECTS OF ALL OF THIS, AND ONE HAS HAD AN AGENCY REPRESENTING THE POLICY ASPECTS OF ALL OF THIS. AND INTRICATE ISSUES GET BOUNCED BACK AND FORTH. AND THERE'S A HIGHER LEVEL, IF NEITHER OF US WANTS TO DO IT THEN IT'S AN APPLICATIONS ISSUE. AND THE TERM APPLICATIONS HAS BEEN USED, THUS FAR IN THE DISCUSSION, TO DESIGNATE THE NEED FOR APPLICATION SOFTWARE. SOMEONE HAS TO DO SOME CODING; THAT SOLVES A PROBLEM THAT NEITHER THE ARCHITECTS NOR THE POLICY PEOPLE HAVE BEEN ABLE TO SOLVE.
BUT IF WE TALK ABOUT APPLICATION NOW IN THE SOCIETAL CONTEXT, PERHAPS WE SHOULD MAKE THIS TRIPOLAR. AND I'M NOT SURE WHAT THE LABELS SHOULD BE. WE HAVE THE TECHNICAL CONCERNS, WE HAVE THE POLICY CONCERNS, AND WE HAVE THE PURPOSE CONCERNS.
AND SOME AGENCY HAS TO STEP FORWARD AND COORDINATE THAT. WITHOUT WISHING TO BE OBSEQUIOUS, I THINK UNESCO WOULD BE ON THE LIST OF CANDIDATES FOR SERVING THAT ROLE. AND WE'RE TALKING NOW ABOUT THE REASONS WHY PEOPLE WANT TO BE ABLE TO DO THIS. AND IF THE INITIAL NOTION OF THE DOMAIN NAMES BEING DEVOID OF SEMANTIC VALUE HAS ANY SIGNIFICANT MERIT TO IT THEN ALL WE'RE DOING WITH IDN IS CREATING THIS BROAD REPERTOIRE WITH WHICH WE CAN REPRESENT NOTHINGNESS RATHER THAN A FOCUSED REPERTOIRE.
SO VERY OBVIOUSLY, THERE IS THE -- IT PROBABLY RELATES TO THE REASONS WHY LANGUAGES DEVELOP AND WHY THEY DIVERGE. NONETHELESS, THERE'S A SOCIETAL COMPONENT TO THIS THAT NEEDS TO BE SOMEHOW BROUGHT TO THE FOREFRONT AND EVERYTHING WE'VE BEEN TALKING ABOUT ON THE TECHNICAL AND POLICY SIDE NEEDS TO MAP INTO THAT. SO WHO COORDINATES THAT?
>>RAM MOHAN: THAT'S ANOTHER GOOD QUESTION.
COULD YOU MAKE SURE THAT YOU -- IF THERE IS A URL FOR, YOU KNOW, FOR THE CONFERENCE COMING UP NEXT YEAR, IF THAT COULD GET FORWARDED SO IT COULD GET POSTED ON THE ICANN SITE. AND AS WELL AS TO THE DIFFERENT REGIONAL BODIES THAT ARE CURRENTLY WORKING, INCLUDING THE CCNSO AND OTHER BODIES. I THINK THERE IS VALUE IN PROPAGATING THAT.
>>ELIZABETH LONGWORTH: YES, I WILL. YES, DEFINITELY.
JUST THINKING ABOUT YOUR COMMENT THERE AND WHO COULD COORDINATE, HANDING ISSUES ACROSS TO ONE OR MORE AGENCIES DOES NOT NECESSARILY SOLVE ISSUES IN THE MODERN WORLD, WE FIND, BECAUSE SO MUCH DEPENDS ON HOW YOU DEFINE THEIR ROLE, HOW DO THEY WORK, IS IT PURELY COORDINATING, IS IT FACILITATING, AND SO ON.
THERE ARE QUITE A FEW ISSUES IN THERE.
AND THE OTHER THING -- THE ONLY COMMENTS -- I'M NOT REJECTING THE IDEA, I'M JUST MAKING AN OBSERVATION THAT THE INTERNET HAS SHOWN US SOMETHING ABOUT HOW CONSENSUS WORKS AS WELL. AND I THINK IT'S A VERY -- IT'S A VERY CIVILIZED WAY TO OPERATE WHEN YOU CAN SHARE IDEAS WITH A WIDE RANGE OF PEOPLE, THAT YOU CAN GET WHAT I CALL BOTTOM-UP COMMENT, AND YOU CAN THEN REALLY GET A LOT MORE PEOPLE TO BUY INTO AN IDEA. AND I'M THINKING THAT IN THE AREA OF LANGUAGE, THIS IS AN AREA WHERE WE REALLY NEED TO LEARN FROM OUR DISTRIBUTED WAY OF DOING THINGS AND MAKE SURE THAT IF ONE OR MORE AGENCIES PLAY A ROLE, WE ALSO USE THE POWER OF THE TECHNOLOGY WE HAVE TO MAKE SURE THAT INDIVIDUALS -- BECAUSE THE SUCCESS OF RESOLVING THESE SORTS OF ISSUES WILL DEPEND VERY MUCH ON WHETHER GROUPS FEEL EMPOWERED BY IT, WHETHER OR NOT THEY RECOGNIZE THEIR OWN SOLUTIONS AND INPUTS INTO THE RECOMMENDATIONS AND WE'RE GETTING MORE AND MORE WAYS TO DO THAT AND YOU DEFINITELY NEED MECHANISMS TO DRIVE PRO!
GRESS AND TO GET PEOPLE TO FOCUS.
BUT I THINK WE NEED TO -- WE NEED TO THINK CAREFULLY ABOUT STRUCTURE AND PROCESSES AND MAYBE ALLIANCES AND HOW YOU FEDERATE ISSUES.
SO IT'S JUST AN OBSERVATION. AND THAT'S WHY THE TASK IS BOTH EXCITING, BUT ALSO VERY CHALLENGING.
>>RAM MOHAN: CHARLES, YOU HAD A COMMENT, AND THAT WILL PROBABLY BE THE LAST COMMENT ON THIS TOPIC BEFORE WE GO TO THE NEXT ONE.
>>CHARLES SHABAN:THANK YOU. I WANTED TO LINK IT TO THE WORKSHOP DONE IN THE MORNING, THE WORLD SUMMIT ON THE INFORMATION SOCIETY.
SOMETHING LIKE A PROPOSAL, SOME OF THE PROPOSALS, AND ONE OF THEM I DISCUSSED IN THE MORNING. MAYBE WE NEED SOME KIND OF FORUM THAT INCLUDES ALL THE ORGANIZATIONS THAT WORK ON THE IDN FIELDS: ICANN, MINC, UNESCO, THE LOCAL GROUP IN EACH REGION, LET'S SAY.
SO THIS WAY I THINK THERE WILL NOT BE THAT -- THIS IS THE ONLY FOR MINC TO DO OR UNESCO AND SO ON. SO ONLY AGREED. IF I WERE IN MY WORKGROUP ON INTERNET GOVERNANCE, I WILL MENTION THIS SUBJECT IN OUR MEETINGS, I THINK, AND WE SHOULD CONSIDER IT, MAYBE.
THANK YOU.
>>JAMES SENG: I THINK THE SOLUTION TO A PROBLEM IS NOT BY CREATING MORE COMMITTEES AND FORUMS.
>>RAM MOHAN: I ACTUALLY HAVE A DIFFERENT POINT OF VIEW. WHILE YET ANOTHER COMMITTEE DOESN'T SOLVE IT; HOWEVER, I THINK MULTILATERALISM IS THE WAY TO GO HERE. AND THAT'S THE WAY -- I THINK THAT'S WHAT CHARLES IS REALLY SUGGESTING AND THAT'S WHAT ELIZABETH IS SUGGESTING. I THINK THAT'S THE WAY TO GO.
WITH THAT, LET'S MOVE ON TO THE NEXT PRESENTATION, WHICH SHOULD REALLY BE THE LAST. THERE WAS -- THERE IS -- THERE WERE REALLY TWO PRESENTATIONS LEFT OR TOPICS LEFT. ONE WAS DEVELOPMENT OF LANGUAGE TABLES, AND THE OTHER IS ON END-USER DEMANDS AND REQUIREMENTS. I WAS SCHEDULED TO TALK ABOUT THE DEVELOPMENT OF LANGUAGE TABLES. WHAT I WANTED TO DO WAS ASK FOR FIVE MINUTES OF YOUR TIME NOW, SO I'LL SIMPLY LIST SOME OF THE KEY ISSUES RATHER THAN ACTUALLY GO THROUGH A FULL-ON PRESENTATION HERE, IN THE INTEREST OF TIME. AND LIST SOME OF THE IMPORTANT THINGS THAT NEED TO BE FOCUSED ON AND THEN PASS IT BACK ON TO YOU.
WHY ARE LANGUAGE TABLES IMPORTANT? WHEN A REGISTRY, A CCTLD OR GTLD REGISTRY IMPLEMENTS A NEW LANGUAGE, IF YOU WILL, IN THEIR DOMAIN, THEY NEED TO ENSURE CLARITY OF WHAT CHARACTERS IN THAT LANGUAGE THE REGISTRY HAS SUPPORTED AND WHAT IT HAS IMPLEMENTED.
ALSO, CLARITY IN UNDERSTANDING WHAT CHARACTER VARIANTS HAVE BEEN IMPLEMENTED, AND SPECIFY ANY OTHER LANGUAGE-SPECIFIC POLICIES OR ISSUES THAT APPLIES IN THE DOMAIN THAT IT'S SUPPORTING.
ICANN HAS CREATED SOME GUIDELINES FOR IDNS, AND THE GUIDELINES BROADLY ASK FOR COOPERATION, COORDINATION, AND WORKING IN SUCH A WAY THAT REGISTRATIONS IN THE LANGUAGE ARE PERFORMED ONLY WHEN AN APPROPRIATE TABLE IS AVAILABLE.
PARTICULARLY WHEN THE REGISTRY FINDS THAT REGISTRATION AND ADMINISTRATION RULES WOULD REALLY HELP BY PUBLISHING SUCH A TABLE.
SO THERE IS A MANDATE TO WORK COLLABORATIVELY WITH RELEVANT AND INTERESTED STAKEHOLDERS. THERE'S NO IDENTIFICATION OF WHO THESE STAKEHOLDERS ARE. THERE IS A REQUEST OR A GUIDELINE THAT SAYS DEVELOP LANGUAGE-SPECIFIC REGISTRATION POLICIES, BUT THERE IS NOT A CLEAR DEFINITION OF IF YOU DEVELOP IT, IS IT REALLY RIGHT? WHO RATIFIES IT? SHOULD IT BE RATIFIED? UNDER WHAT AUTHORITY SHOULD IT BE RATIFIED?
THERE IS SOMETHING EVEN MORE FUNDAMENTAL THAT IS NOT FULLY DEFINED. HOW DO WE KNOW, HOW DOES ONE KNOW THAT A TABLE, A LANGUAGE TABLE THAT'S BEEN CREATED, IS THE RIGHT ONE?
WHAT ARE THE RIGHT GUIDELINES TO HELP CREATE TABLES? THIS DOESN'T EXIST. THIS IS SOMETHING THAT ICANN CAN DEFINITELY HELP ON.
BUT IT'S NOT JUST ICANN. IT'S ICANN, IT'S CCTLDS. THERE ARE VARIOUS PARTIES. THEY HAVEN'T BEEN IDENTIFIED YET. WHO ARE THE APPROPRIATE PARTIES WHO SHOULD COME TOGETHER TO CREATE TABLES?
AND SOME FOLKS HAVE MADE THE WRONG ASSUMPTION THAT JUST BECAUSE IANA PROVIDES THE ABILITY TO PUBLISH TABLES FROM REGISTRIES, AND IT PUBLISHES IT, SOME FOLKS THINK THAT MEANS IANA HAS RATIFIED IT. AND IANA, THERE'S NO RATIFICATION IMPLIED OR INTENDED. IT'S SIMPLY A PLACE TO STORE AND DISPLAY.
SO THE CLOSING COMMENTS ON THIS PARTICULAR AREA OF LANGUAGE TABLES ARE WE NEED TO FIGURE OUT WHICH ARE TO CURRENT TO ICANN GUIDELINES, AND IF THEY ARE NEEDED WE HAVE TO FIGURE OUT HOW TO GET THERE AND ALSO WHO SHOULD BE CHARGED TO GO AND DEVELOP THE CONTINUATION AND ENHANCEMENT OF THE IDN GUIDELINES.
WE ALSO NEED SOME IDENTIFICATION OF WHAT IS NEEDED FROM ICANN ITSELF IN THIS AREA THAT IS NOT BEING DONE TODAY AND FINALLY HOW CAN COOPERATION BETWEEN INTERESTED GLOBAL BODIES BE ENSURED? THOSE ARE THREE QUESTIONS THAT ARE HANGING FIRE.
WITHOUT THESE BEING DONE, FOR A REGISTRY THAT WANTS TO GO AND IMPLEMENT LANGUAGES OR SCRIPTS, AT THE VERY BASIC LEVEL, WE'RE NOT TALKING POLICY HERE BUT AT THE VERY BASIC LEVEL OF IMPLEMENTATION, YOU'VE GOT AMBIGUITY, AND IN THE DNS, AS WE KNOW, AMBIGUITY IS NOT SUCH A GOOD THING.
>>JAMES SENG: I HAVE A QUESTION BACK TO YOU. YOU MENTIONED THREE THINGS. ONE OF THEM IS RECOMMENDATION OR CHANGES TO THE IDN GUIDELINES FROM THE ICANN IDN GUIDELINES. IN WHAT SPECIFIC INSTANCES ARE YOU PROPOSING TO CHANGE? THAT'S ONE.
AND SECOND, WHAT IS THE VALUE OF DEFINING WHICH GROUP FOR THE PARTICULAR LANGUAGE, IF WE EVER ACTUALLY DEFINE THAT? WHAT'S THE VALUE OF DOING SO?
>>RAM MOHAN: OH, LET ME GIVE YOU AN EXAMPLE OF THE VALUE.
TAKE A LOOK AT FRENCH. YOU'VE GOT FRENCH AS IT IS REPRESENTED AND SPOKEN IN FRANCE AND AS IT IS REPRESENTED AND SPOKEN IN QUEBEC.
AND THE REPRESENTATION OF IT AND THE USE OF IT IS VALID IN BOTH PLACES.
IF YOU'RE A REGISTRY, IF YOU'RE DOT CA YOU MAY CHOOSE A SET OF RULES ON HOW TO IMPLEMENT FRENCH THAT MAY BE DIFFERENT THAN IF YOU ARE DOT FR AND WHICH MAY BE DIFFERENT THAN IF YOU ARE DOT ORG OR DOT COM.
IN A CASE LIKE THAT, THERE IS GREAT VALUE IN IDENTIFYING WHICH GROUP, WHICH GROUPS SHOULD COME TOGETHER BECAUSE OTHERWISE YOU'RE GOING TO END UP HAVING FOOD FIGHTS OVER WHO OWNS WHICH TABLE.
WHAT WAS YOUR FIRST QUESTION?
>>JAMES SENG: LET'S DISCUSS THIS FIRST. IT'S QUITE INTERESTING.
LET ME PUT IT IN BLUNT WAY. IT'S NOT A PROBLEM FOR CCTLD BECAUSE A CCTLD KNOWS EXACTLY WHO TO WORK WITH. DOT CA WOULD PROBABLY WORK WITH QUEBEC PEOPLE, THE LINGUISTS IN QUEBEC AND THEY PROBABLY WOULDN'T CARE WHAT'S HAPPENING IN FRANCE. LIKEWISE AND VICE VERSA. IF THERE'S ANY COORDINATION REQUIRED, THEN THEY PROBABLY WOULD DO IT AMONGST THEMSELVES BUT PRIMARILY DOSSIERS OF THE CANADIAN COMMUNITY WOULD WORK WITH THE (INAUDIBLE).
AND THE PROBLEM ONLY OH FOR GTLD OR STLD WHICH IS NOT TITLE COUNTRY CODE WHERE YOU NEED TO PROVIDE FRENCH DOMAIN NAMES FOR BOTH FRENCH SPEAKING PEOPLE FROM QUEBEC, FRANCE AND ANYWHERE ELSE. MICHEL SPEAKS FRENCH, WHO IS ORIGINALLY FROM FRANCE.
SO IN A BLUNT WAY WHAT YOU'RE TRYING TO SAY IS THAT MAYBE ICANN SHOULD DO THIS DIFFICULT JOB OF IDENTIFYING LINGUIST GROUP FOR THE GTLD CONSTITUENCY.
>>RAM MOHAN: I DON'T THINK I'M SUGGESTING THAT. I THINK I'M PRESENTING A PROBLEM. I DON'T KNOW IF I HAVE THE SOLUTION HERE. BUT LET ME ACTUALLY TAKE ISSUE WITH WHAT YOU'RE SAYING.
FRENCH IS SPOKEN HERE IN AFRICA.
AND A DOT FR SITE AND THE WAY IT WORKS, THE WAY IT'S REPRESENTED BY THE FRENCHNIC IN FRANCE DOESN'T NECESSARILY WORK THE RIGHT WAY OR DOESN'T NECESSARILY TRANSLATE EXACTLY RIGHT HERE IN AFRICA.
YOU CAN'T JUST SAY THE FRENCH NIC CAN JUST DECIDE WHAT THE FRENCH LANGUAGE TABLE IS.
AND THAT WHICHEVER NIC LOCAL HERE THAT DOES FRENCH DOES IT IN A COMPLETELY DIFFERENT WAY.
>>JAMES SENG: PRECISELY, WHICH IS WHY IF YOU LOOK AT THE WAY WE DEFINE LANGUAGE TABLES, IT'S NOT DEFINED AS "THE" LANGUAGE TABLE FOR ALL THE REGION.
IT'S DEFINED THE REGION, THE LANGUAGE TABLE PER CCTLD OR PER TLD.
>>RAM MOHAN: JAMES, I REALIZE THAT.
BUT DO YOU NOT SEE THAT IF YOU JUST LET THIS GO FORWARD AND YOU DO NOT HAVE SOME WAY OF COORDINATION, IT DOESN'T HAVE TO BE AN ICANN.
BUT IF YOU DON'T HAVE SOME WAY OF IDENTIFYING WHAT ARE COMMONALITIES AND NOT COMMONALITIES, YOU'RE GOING TO GET TO THE POINT WHERE EACH REGION OR EACH COUNTRY WITHIN EACH REGION IS GOING TO DEFINE ITS OWN SPECIAL IMPLEMENTATION.
MAYBE THAT'S RIGHT.
BUT I THINK IT NEEDS TO BE DEBATED.
I DON'T THINK WE SHOULD JUST --
>>JAMES SENG: IT'S NOT THAT I DISAGREE WITH YOU.
I -- WELL, (INAUDIBLE) DEVELOPED A CJK GUIDELINE.
WHY DO WE DEVELOP A CJK GUIDELINE?
BECAUSE WE ARE TRYING TO HARMONIZE THE ISSUES EXACTLY THAT YOU BROUGHT UP.
HOWEVER, WHAT I DO REALIZE IS THAT THE CIRCUMSTANCES, THE EXPERIENCE THAT I HAVE IN DEVELOPING THE CJK GUIDELINE, GETTING THE CHINESE AND THE TAIWANESE AND JAPANESE AND THE KOREANS TO WORK TOGETHER TO DEVELOP THE CJK GUIDELINES IS GOING TO BE A VERY DIFFERENT EXPERIENCE FROM SOMEONE WHO IS TRYING TO UNIFY FRENCH-SPEAKING PEOPLE IN FRANCE, QUEBEC, OR ANYWHERE ELSE THAT IS SPEAKING FRANCE, WHICH IS GOING TO BE DIFFERENT FROM PEOPLE WHO -- TO TALK ABOUT THE HARMONIZATION OF ARABIC SCRIPTS IN THE MIDDLE EAST.

>>RAM MOHAN: AND THAT'S THE VALUE OF IDENTIFYING THE GROUPS, BACK TO YOUR EARLIER QUESTION.
>>JAMES SENG: I CANNOT EVEN DEFINE THE GROUP.
EVENING UNESCO, I DON'T THINK THEY CAN POINT TO A GROUP THAT WOULD REPRESENT THE FRENCH COMMUNITY IN THE WHOLE WORLD.
WOULD YOU?
>>ELIZABETH LONGWORTH: YOU'RE MAKING THE POINT VERY ELOQUENTLY.
BUT I'M NOT SURE THAT YOU HAVE THE SOLUTION YET.

>>JAMES SENG: I AM NOT PROPOSING A SOLUTION.
I AM SAYING THAT THE PROPOSAL IS -- IS -- THE PROPOSAL THAT YOU MADE TO IDENTIFY GROUPS IS A VERY DANGEROUS, SLIPPERY SLOPE TO GO DOWN, BECAUSE YOU JUST CAN'T SAY, OH, A LANGUAGE GROUP IS BETTER THAN B LANGUAGE GROUP OR WE LISTEN TO A AS AUTHORITY VERSUS B.
YOU ARE JUST GETTING (INAUDIBLE) IT'S BETTER TO GET A AND B TOGETHER IF THEY REALIZE THAT THERE IS A PROBLEM AND COME UP WITH SOLUTION FOR YOU.
>>RAM MOHAN: I DON'T KNOW IF YOU AND I ARE SO FAR APART.
WHAT I'M SAYING IS THAT A AND B TODAY DO NOT NECESSARILY WORK TOGETHER.
AND WHAT YOU'RE SUGGESTING IS PERHAPS A DIFFERENT SIDE OF THAT SAME SLIPPERY SLOPE.
YOU HAD A COMMENT.
>>MICHEL SUIGNARD: YEAH.
I HAVE BEEN WORKING ON, I WOULD SAY, CHARACTER LANGUAGE ISSUE FOR OVER 15 YEARS NOW.
AND I HAVE SEEN MANY, MANY, IN FACT, AND I WOULD -- CAN'T DISAGREE WITH WHAT YOU HAVE SAID EARLIER.
I HAVE SEEN MANY, MANY LANGUAGE TABLES, MANY, ESPECIALLY IN EUROPE, WHOEVER DID THE WORK CREATING THOSE LANGUAGE TABLES.
AND ONE THING I FOUND, THAT ALL OF THEM ARE INCORRECT, ALL OF THEM.
BECAUSE IT'S A VERY DYNAMIC, YOU KNOW, CONCEPT TO THINK WHAT IS A LANGUAGE.
IT'S CHANGE ALL THE TIME.
SO TO SOME DEGREE I DON'T HAVE AN ISSUE AS LONG AS YOU KEEP IT AS A GUIDELINE.
BUT THE PROBLEM IS, HERE IS THE WORD "GUIDELINE."
BUT AT THE SAME TIME I HEAR THAT YOU ARE TRYING TO ENFORCE IT.
ENFORCING A GUIDELINE CANNOT GIVE ME THE FREEDOM.
IN FACT, I WOULD ARGUING THAT CCTLD IS EVEN ENFORCING TOO NARROWLY THE GUIDELINE, THEY ARE GOING TO BE NOT RELEVANT, BECAUSE THEY WILL GO SOMEWHERE ELSE TO BE, YOU KNOW, KNOWN OR TO BE REGISTERED.
THEN YOU WILL SEE PEOPLE GOING TO, LET'S SAY, MAYBE DOT-COM OR SOMEWHERE ELSE WHERE THEY WILL GET WHAT THEY WANT.
BECAUSE IN MANY ASPECTS, THERE IS A LOT OF CASES WHERE YOU WANT TO BE ABLE TO EXPRESS A LOT IN WHAT YOU WOULD CALL A LANGUAGE, BECAUSE, YOU KNOW, WORDS GET BORROWED IN ENGLISH.
I MEAN, ENGLISH IS SUPPOSED TO BE USE OF A TO Z.
IN FACT, IF YOU READ ANY NEWS MAGAZINE EVEN IN THE U.S., THEY ARE FULL OF FRENCH AND GERMAN AND WHATEVER.
I DON'T WANT TO BE LIMITED TO EXPRESS MYSELF IN THE A TO Z CONCEPT WHEN I AM WRITING IN ENGLISH.
THAT'S A VERY SIMPLISTIC VIEW OF THAT.
BUT IT EXISTS IN EVERY LANGUAGE.
SO PLEASE SAY GUIDELINE, BUT THEN DON'T ENFORCE GUIDELINES, BECAUSE THEN, YOU KNOW, I DON'T THINK IT WOULD BE RELEVANT.
SO -- BECAUSE THEY'RE ALL INCORRECT.
>>ROOZBEH POURNADER: I WOULD REALLY LIKE TO OBJECT TO THAT, THE IDEA OF TRYING TO REALLOCATE NEW CONSORTIUMS FROM A TOP TO BOTH OF THEM LEVEL INSTEAD OF BOTTOM TO TOP.
SO WHAT WE NEED HERE IS SOMETHING LIKE, SAY, THEY GOT TOGETHER, WORKED OUT SOMETHING TOGETHER, AND THEY USED THIS FOR THEMSELVES, NOT FOR ANYBODY ELSE, NOT RECOMMENDED FOR ANYBODY ELSE.
SO IT'S REALLY OKAY.
SO IF THE PEOPLE WITH COMMON INTERESTS COULD MEET AND COULD DESIGN THEMSELVES SOMETHING.
BUT THE MAIN IDEA IS, LET THEM DECIDE FOR THEMSELVES AFTER ALL.
SO LET'S SAY IF IN A CERTAIN ARAB COUNTRY LET'S SAY IN NORTHERN AFRICA, WHEN THEY HAVE SPECIAL ADDITIONAL LETTERS FOR ACTUALLY TRANSCRIBING ENGLISH OR EUROPEAN NAMES INTO THEIR LANGUAGE, WHILE IN SOME OTHER ARAB COUNTRIES THEY USE DIFFERENT LETTERS FOR THAT THING, AND THEY CAN'T REALLY UNIFY ON THAT -- I WON'T GO INTO DETAILS, OF COURSE.
THE MAIN ISSUE HERE IS, LET EVERYBODY HAVE HIS OWN VERSION.
IF THEY WANT TO WORK TOGETHER, LET THEM WORK TOGETHER.
OTHERWISE, THERE WILL BE USER OBJECTIONS, THERE WILL BE -- I MEAN, EVERYBODY CAN HANDLE THAT CASE BY HIMSELF.
IF YOU WANT THEM UNIFIED, THERE WILL ALWAYS BE THE ISSUE OF MAJORITY VERSUS MINORITY.
I HAVE HAD THAT EXPERIENCE WHEN WORKING WITH THE MANY ARAB COUNTRIES WHOSE MAIN LANGUAGE IS ARABIC WHILE OUR LANGUAGE WAS NOT ARABIC, ONLY OUR SCRIPT WAS ARABIC.
YOU ARE ALWAYS MARGINALIZED.
SO THE THING IS LET EVERYBODY WORK HIS OWN, EVEN ARABIC LANGUAGE IS NOT A VERY STANDARD THING, ALTHOUGH IT'S USED WIDELY AND CONSIDERED LIKE THAT.
THAT'S MY WHOLE THING.

>>CHARLES SHABAN: THANK YOU.
I NEVER MEANT TO HAVE A NEW COMMUNITY OR NEW FORUM OR SOMETHING WHICH IS -- AS A BODY, JUST A PLACE FOR EVERYONE TO BE THERE.
SO YOU'RE RIGHT.
LET THEM BOTH BE THERE.
AND IF THEY AGREE, OKAY.
IF NOT, EVERYONE CAN SUPPORT SOME, BUT AT LEAST SOME PLACE THAT EVERYBODY CAN DISCUSS THE ISSUE, AS RAM EXPLAINED.
NOT TO CREATE ANY NEW BODY, FOR SURE.
BECAUSE YOU DON'T WANT ANY TOP-DOWN -- TOP-BOTTOM PROCESS OR SOMETHING.
THANK YOU.

>>RAM MOHAN: OKAY.
TO THE FINAL PRESENTATION.
>>HONG XUE: THANKS.
SO USERS' DEMANDS AND REQUIREMENTS IS THE LAST TOPIC.
AND HOPEFULLY THAT MEANS THE USERS INTERESTS ARE NOT THE LEAST IMPORTANT IN THE IDN PROCESS.
I AM A MEMBER OF ICANN.
WILL YOU PLEASE REACH THIS -- A MEMBER OF ICANN AT-LARGE ADVISORY COMMITTEE, BUT I'M NOT SPEAKING FOR THE COMMITTEE, ALTHOUGH I APPRECIATE THE GREAT CONTRIBUTION -- I APPRECIATE THE CONTRIBUTION MADE BY THE -- VITTORIO, THE CHAIR, AND OTHER COLLEAGUES IN ALAC.
AND WE HAVE TO GO TO THE FUNDAMENTAL PURPOSE OF IDN.
ACTUALLY, THE USERS' DEMAND IS THE STARTING POINT OF IDN, BECAUSE PEOPLE NEED IT. AND THAT IS THE REASON WE ARE GOING TO DEVELOP IDN.
AND THIS HAS BEEN ACKNOWLEDGED BY THE ICANN DOCUMENT.
WE CAN SEE ICANN'S RESOLUTION.
THIS IS IMPORTANT, THAT INTERNET EVOLVE TO BE MORE ACCESSIBLE TO THOSE WHO DO NOT USE THE ASCII CHARACTER SETS.
WHEN WE LOOK AT THE POLICY-MAKING, WE WILL SEE THAT ICANN HAS ACKNOWLEDGED THAT IT'S IMPORTANT TO IMPLEMENT LANGUAGE-SPECIFIC POLICIES.
OF COURSE, MANY TECHNICAL EXPERTS ARE VERY CAUTIOUS TO USE THE WORD "LANGUAGE."
THEY ARE MORE COMFORTABLE TO USE THE WORD "SCRIPTS" OR "CHARACTERS."
HOWEVER, IDN ARE EVENTUALLY SERVING THE INTEREST, SERVING THE DEMAND OF THE USERS.
SO THAT IS QUITE IMPORTANT TO HAVE A USER FRIENDLY INTERFACE.
WHEN WE'RE TALKING ABOUT A LANGUAGE-SPECIFIC POLICY, WE HAVE TO GO OUT TO CONSULT WITH THE LANGUAGE COMMUNITY, THAT IS, THE USERS' COMMUNITY OF A SPECIFIC LANGUAGE OR A GROUP OF LANGUAGES.
LET'S LOOK AT THE DOCUMENT THAT CAN HARDLY BE REMEMBERED, NEARLY ALMOST FORGOTTEN.
THE FINAL REPORT OF THE ICANN IDN COMMITTEE PUBLISHED JUNE 27TH, 2002.
ONE OF THE KEY CRITERIA PROPOSED FOR KEY NON-ASCII REGISTRY IS LOOK AT BULLETIN 2, SUPPORT FROM RELEVANT COMMUNITY OF INTEREST.
THAT'S VERY IMPORTANT.
AND FOR ANOTHER PERSPECTIVE THAT HAS BEEN PRESENTED BY OUR MODERATOR, I AM NOT GOING TO REPEAT THAT.
WHAT I AM GOING TO SAY IS THAT ICANN HAS FACILITATED THESE CHARACTER VARIANTS TABLES AND SHOULD MAKE SURE THESE TABLES ARE IMPLEMENTED BY THOSE TLD REGISTRIES UNDER THE AGREEMENT WITH ICANN.
WE LOOK AT THE IDN SERVICES.
THE MOST IMPORTANT POINT IS THAT ALL THESE SERVICE PROVIDERS SHOULD HAVE MULTILINGUAL CAPACITY.
THIS IS A REQUIREMENT; IT'S NOT A SUGGESTION.
AND WHEN WE LOOK AT ICANN IDN GUIDELINE, WE CAN SEE THAT THE TLD REGISTRIES SHOULD PROVIDE INFORMATION REAL RESOURCES AND SERVICES IN ALL LANGUAGES FOR WHICH THEY OFFER INTERNATIONALIZED DOMAIN NAME REGISTRATIONS.
AND WHEN WE LOOK AT REGISTRATION FOR EQUIVALENT VARIANTS, THIS IS NOT ONLY AN ISSUE OF THE VARIANTS TABLE, IT'S ALSO ABOUT REGISTRATION, THE TRUE REGISTRATION.
WE BELIEVE THE EQUIVALENCE SHOULD BE DEFINED FOR THE BENEFIT OF THE END USER.
WE HAVE BEEN TALKING ABOUT THE IMPORTANCE TO PREVENT TRADEMARK CONFUSION.
TRADEMARK OWNERS ARE ONLY A SMALL SECTOR OF USERS.
WE NEED TO LOOK AT THIS ISSUE IN A BROADER CONTEXT TO PROTECT THAT ALL THE END USERS, NOT ONLY THE TRADEMARK OWNERS, AND ALL THE EQUIVALENT RELEVANT OF A DOMAIN NAME SHOULD BE SOLD AND REGISTERED IN BUNDLE.
SO WE ARE PROPOSING A BUNDLE SOLUTION, BUT WHAT'S MORE RADICAL IS THAT SUCH A BUNDLE SHOULD ONLY BE CHARGED AS A SINGLE NAME, BECAUSE INDIVIDUAL USERS ARE VERY PRICE SENSITIVE.
IMPLEMENTATION OF IDN SHOULDN'T IMPOSE EXTRA FINANCIAL BURDEN ON INDIVIDUAL USERS.
AND THE REGISTRY'S OBLIGATIONS.
PROBABLY THIS DOCUMENT, PUBLISHED BY ICANN IN 2000 HAS DISAPPEARED IN OUR VISION.
BUT WHEN WE READ IT, IT'S STILL INTERESTING.
IT SAYS THAT ANY DEPLOYMENT OF AN IDN SOLUTION PRIOR TO DECISION OF IETF MUST REMAIN AN EXPERIMENT.
IT SAYS THAT, IDN SERVICE IF IT IS AN EXPERIMENT, IT SHOULD BE LABELED AS AN EXPERIMENT, AS A TEST BED.
IT SAYS THAT THE REGISTRY HAS OBLIGATION TO DISCLOSE THIS TRANSIENT AND TEMPORARY NATURE OF IDN REGISTRATION.
OTHERWISE, IT WILL BE SUSCEPTIBLE OF DECEPTIVE PRACTICES.
BUT AT PRESENT, IT BECOMES A PROBLEM, BECAUSE IETF IS ALREADY THREE OR A COUPLE OF RFCS PUBLISHED, AND ICANN HAS PUBLISHED THE GUIDELINES.
SO IS THERE STILL SUCH AN OBLIGATION IMPOSED ON REGISTRIES?
THAT IS AN OPEN QUESTION.
AND THE SECOND POINT, MANY DOMAIN NAME HAS ALREADY BEEN REGISTERED.
IDN HAS BEEN REGISTERED.
IF WE NOW ARE GOING TO FULLY IMPLEMENT THIS IDN PROCESS, SAY, FOR EXAMPLE, IN THE FUTURE, SOME IDN TLD WILL BE DELEGATED BY ICANN, SHOULD THE TRANSITIONAL MEASURE BE TAKEN TO REMEDY IN THOSE TEST BED PERIOD.
I REGISTERED A DOMAIN NAME, BUT NOW IT'S TRANSFERRED TO ANOTHER SYSTEM.
SO SHOULD THOSE TEST BED REGISTRATION BE ALLOWED TO TRANSFER AUTOMATICALLY TO A FORMER REGISTRATION AS ONE SOLUTION, OR THE USER SHOULD BE REFUNDED OR COMPENSATED FOR THE CHANGE; RIGHT?
YES, THIS -- AND THANK YOU VERY MUCH.
OKAY.
THANKS.

>>RAM MOHAN: QUESTIONS?
PANELISTS, ANY QUESTIONS?
AUDIENCE?

OKAY.
THAT CONCLUDES THIS SESSION THAT HAS RUN -- CONTINUED TO KEEP THE SAME TIME BEHIND IT WAS WHEN IT STARTED.
SO THANK YOU FOR ALL OF WHO YOU HAVE STAYED BEHIND AND WHO HAVE LISTENED AND FOR ALL OF THE PANELISTS HERE.
I APPRECIATE YOUR COMING HERE AND FOR SHARING YOUR THOUGHTS, IMPORTANT TOPICS THAT DOUBTLESS WILL GET DISCUSSED FURTHER AS WE MOVE FORWARD.
THANK YOU ALL.
THANK YOU VERY MUCH.
(APPLAUSE.)
>>TINA DAM: THANK YOU, RAM.
I JUST WANTED TO MENTION, WE HAD REFRESHMENTS OUT AN HOUR AGO, AND SINCE WE HAVE GONE OVER TIME, IT APPEARS THEY HAVE ALREADY BEEN EATEN.
IT'S THE FIRST TIME I AM GLAD TO SEE THAT THERE AREN'T VERY MANY PEOPLE IN THE AUDIENCE BECAUSE IT MEANS I CAN ASK YOU TO COME OVER TO THE HOTEL AND HAVE A DRINK OR COFFEE OR WHATEVER YOU FEEL LIKE.
SO WE'LL FINISH UP HERE AND PLEASE FEEL FREE TO JOIN ME.
THANKS.
(8:45 P.M.)

© Internet Corporation for Assigned Names and Numbers