ICANN|ARCHIVES

Internet Corporation for Assigned Names and Numbers

>>VINT CERF: THIS IS TRULY DISTURBING TO THINK THAT THE WAY TO CLEAR THE ROOM IS FOR ME TO GET UP IN FRONT OF THE MICROPHONE AND EVERYONE LEAVES.
THERE HE GOES.
I'VE BEEN ASKED TO OPEN THIS MEETING WITH A VERY BRIEF STATEMENT AND THEN ALLOW CARY KARP AND THE OTHERS TO LITERALLY CARRY ON, IF YOU'LL PARDON THE PUN.
THE SUBJECT OF IDNS IS ABSOLUTELY CRITICAL TO THE FUTURE EVOLUTION OF THE INTERNET, BECAUSE WE HAVE A GROWING COMMUNITY OF PEOPLE FOR WHOM ENGLISH IS NOT THEIR FIRST LANGUAGE.
THIS IS A NONTRIVIAL PROBLEM.
THERE ARE MANY TECHNICAL ISSUES THAT HAVE TO BE RESOLVED.
AND THIS COMMITTEE, THIS PANEL, RATHER, I HOPE, WILL BE ABLE TO TOUCH ON SOME OF THOSE PROBLEMS, AND I HOPE THROW SOME LIGHT ON POSSIBLE SOLUTIONS.
IN ANY EVENT, NO MATTER HOW HARD THIS IS, WE HAVE TO MAKE PROGRESS ON IT.
CARY, IT'S ALL YOURS.
>>CARY KARP: THANK YOU VERY MUCH, VINT.
THE INTERNATIONALIZATION OF THE DOMAIN NAME SPACE IS A VERY OBVIOUS COMPONENT OF THE INTERNATIONALIZATION OF THE NET ITSELF.
AS THE NUMBERS OF LANGUAGES THAT APPEAR AND THE CONTENT ON THE NET INCREASE, IT IS REASONABLE FOR USERS TO EXPECT TO SEE THE AUTHORS OF AND THE CONSUMERS OF THOSE DOCUMENTS TO BE ABLE TO SEE THEM IDENTIFIED IN THE SAME LANGUAGES, USING THE SAME RANGE OF SCRIPTS.
AND, FORTUNATELY, PERHAPS SOMEWHAT UNFORTUNATELY AS WELL, THERE ARE A NUMBER OF SPECIFIC PROBLEMS THAT ATTACH TO THE DOMAIN NAME COMPONENT OF THIS.
AND THESE HAVE BEEN ADDRESSED BY SEVERAL BODIES ENGAGED IN A SHARED NORMATIVE EFFORT FOR QUITE SOME TIME NOW.
MOST OF THEM ARE ON THE PANEL TODAY.
AND A FEW OF THE KEY ONES WILL BE MAKING PRESENTATIONS ABOUT THEIR PERSPECTIVE ON THE UNDERLYING ISSUES.
IT IS OUR INTENTION TO SPEND MOST OF THE TIME AT OUR DISPOSAL IN DISCUSSION.
SO I'M ABOUT TO ASK EVERYONE HERE TO INTRODUCE THEMSELVES VERY BRIEFLY.
AND THEN THERE WILL BE THREE TEN-MINUTE PRESENTATIONS FROM THREE OF THE PANELISTS.
AND THEN I AM GOING TO OPEN THE MICROPHONE TO QUESTIONS FROM YOU ABOUT ALL OF THIS.
AND WE ARE PREPARED TO DISCUSS SEVERAL KEY ISSUES OURSELVES.
JUST TO PUT YOU IN TUNE WITH ONE OF THE REASONS WHY WE ARE HERE, AS ICANN RECOGNIZED ITS RESPONSIBILITY FOR ALL OF THIS, ONE OF THE FIRST THINGS THAT IT DID WAS DRAFT A SET OF GUIDELINES TO BE IMPLEMENTED BY THE TLD REGISTRIES.
AND, IN FACT, THERE WAS SOME CONTINGENCY HERE.
IF A GTLD REGISTRY WISHED TO BE ABLE TO ACCEPT THE REGISTRATION OF WHAT ARE CALLED IDNS, INTERNATIONALIZED DOMAIN NAMES, WHICH IS NOT AN ENTIRELY APPROPRIATE TERM FOR THEM, BUT NONETHELESS WE'LL USE THE ACRONYM, IF YOU WISH TO REGISTER IDNS, YOU NEED TO MAKE A STATEMENT OF WILLINGNESS BOTH TO ADHERE TO THESE GUIDELINES AND TO CONTRIBUTE TO THEIR ONGOING DEVELOPMENT AS WE GATHER ACTUAL EXPERIENCE IN THE USE OF IDNS.
AND WE ARE NOW AT A JUNCTURE WHERE THAT REVISION IS BOTH NECESSARY, IF NOT OUTRIGHT LONG OVERDUE.
THE PARTICULAR ISSUE THAT GENERATES THIS SENSE OF URGENCY IS THAT THE RANGE OF CHARACTERS THAT ARE AVAILABLE FOR INCLUSION IN AN IDN AT THE MOMENT IS FAR LARGER THAN ANY REASONABLE PURPOSE OF IDENTIFYING A DOMAIN MIGHT REQUIRE.
AND, SADLY, THOSE WHO WOULD MISUSE ANYTHING THAT THEY CAN TO LESS LAUDABLE END HAVE DISCOVERED THAT AS THE RANGE OF AVAILABLE CHARACTERS INCREASES, THE ABILITY TO DECEIVE POOR USERS INCREASES EXPONENTIALLY.
AND WE'RE IN A SITUATION NOW WHERE WE REALIZE THAT THE UNIVERSE HAS TO BE SHRUNK.
AND THE BASIS FOR THAT SHRINKAGE IS THE UNICODE CONSORTIUM STATEMENT OF -- ACTUALLY, THE UNICODE CODE CHARTS SAY WHAT EXISTS.
DURING THE NORMATIVE PROCESS, WE HAVE AGREED ON SOME SMALLER AMOUNT OF THAT TOTALITY THAT IS AVAILABLE FOR IDN, AND WE NOW NEED TO AGREE BOTH ON A SMALLER STILL SET OF CHARACTERS AND POLICIES THAT ARE MORE CLEARLY SHARABLE AMONG REGISTRIES, AND NOT JUST THE GTLD REGISTRIES, BUT ANY REGISTRY ON ANY LEVEL THAT WISHES TO SUPPORT IDN NEEDS TO HAVE SOME SUPPORT DOCUMENT, WHICH IS IN ITSELF COMPELLING.
"THIS IS WHAT YOU NEED TO DO, BECAUSE IT IS WHAT YOU TRULY WANT TO DO.
AND IT IS NOT INTUITIVE, WE WILL HELP YOU."
SO THE INITIAL PURPOSE OF THIS ONE OF THE TWO SESSIONS IN THIS WORKSHOP IS TO GET THE PROCESS OF REVISING THE IDN GUIDELINES UNDERWAY TOWARD THE NEXT-TERM GOAL OF THIS BEING ABLE TO SERVE AS A GOOD PRACTICE STATEMENT, COMPELLING IN ITSELF, WHICH WILL HELP THOSE WHO WOULD USE IDN TO NOBLE PURPOSE TO BE ABLE TO DO SO, AND AT THE SAME TIME WILL BE DISCUSSING THE NORMATIVE CONSTRAINTS THAT WOULD MAKE IT LESS POSSIBLE FOR THOSE WHO WOULD USE IDN FOR NEFARIOUS PURPOSE TO DO SO.
WITH THAT, IF I CAN ASK EACH OF YOU BRIEFLY TO SAY WHO ARE YOU AND WHY YOU BELIEVE YOU ARE HERE, WE CAN TAKE IT FROM THERE.

>>MICHAEL EVERSON: HELLO.
MY NAME IS MICHAEL EVERSON, EVERTYPE.COM.
I AM AN EXPERT IN THE WORLD'S WRITING SYSTEMS AND A LINGUIST.
AND I HAVE FOR MORE THAN A DECADE BEEN ONE OF THE PRINCIPAL PEOPLE INVOLVED IN ADDING MINORITY LANGUAGES, MINORITY SCRIPTS AND CHARACTERS, TO THE UNIVERSAL CHARACTER SET OR UNICODE, AND ISO IEC 10646.
AND I BELIEVE THAT I'M HERE TO GIVE THE LINGUISTIC VIEWPOINT AS TO HOW CHARACTERS AND SCRIPTS RELATE TO THE REALITIES OF WHAT PEOPLE ARE TRYING TO ENCODE IN IDN.
>>RAM MOHAN: I'M RAM MOHAN.
I'M WITH AFILIAS.
WE ARE THE REGISTRY FOR DOT INFO AND PROVIDE BACK-END REGISTRY SERVICES FOR DOT ORG ON THE GTLD LEVEL AS WELL AS FOR A NUMBER OF CCTLDS.
I'VE BEEN INVOLVED WITH IDNS FOR QUITE A WHILE NOW, AND I'M MULTILINGUAL MYSELF, AND WAS PART OF THE ORIGINAL COMMITTEE THAT HELPED CREATE THE ICANN GUIDELINES.
AND I'M HERE REPRESENTING THE REGISTRIES, THE GTLD REGISTRIES CONSTITUENCY.
BUT IT'S NOT REALLY ONLY IN A CONSTITUENCY ROLE, BUT TO ALSO CONTRIBUTE TO THE DISCUSSION ON IDNS, AND TAKE THE GUIDELINES THAT WE HAVE AND MOVE THEM INTO SOMETHING MORE CONCRETE.
>>CLAUDIO MENEZES: MY NAME IS CLAUDIO MENEZES.
I WORK FOR UNESCO INFORMATION SOCIETY DIVISION, WHERE I AM IN CHARGE OF THREE MAJOR ISSUES.
ONE IS PROMOTING ICT FOR PEOPLE WITH SPECIAL NEEDS.
NUMBER TWO, PROMOTING ACCESS TO SCIENTIFIC ADDITIONAL INFORMATION.
AND THE THIRD ONE, WHICH IS THE REASON WHY I'M HERE, I AM COORDINATING UNESCO PROGRAM ON MULTILINGUALISM IN THE DIGITAL WORLD.
THANK YOU.
>> PAT KANE: GOOD MORNING, MY NAME IS PAT KANE.
I AM PRODUCT MANAGER AT VERISIGN FOR COM, NET, AND IDN.
I'VE BEEN WORKING WITH IDN SINCE THE LAUNCH OF OUR TEST BED IN NOVEMBER OF 2000.
AND THE PRIMARY REASON I'M HERE, I THINK, IS WHY A LOT OF US ARE HERE, WHICH IS TO THROW ME UNDER THE BUS.

(LAUGHTER.)
>>MICHEL SUIGNARD: MY NAME IS MICHEL SUIGNARD.
I WORK FOR MICROSOFT ON IDN ISSUES FOR MICROSOFT.
I AM ALSO PART OF THE -- ONE OF THE CONSORTIUM ON UTF 46, THAT PROBABLY WILL BE MENTIONED WHICH IS ABOUT UNICODE SECURITY (INAUDIBLE) THAT INCLUDED A LOT OF IDN CONCERNS.
IN ADDITION, I'M ALSO THE ISO PROJECT EDITOR FOR 10606, SO THERE MAY BE SOME CONNECTION WITH THE DISCUSSION TODAY ABOUT THAT.
.
>>CARY KARP: OKAY. AND WITH THAT, I THINK WE CAN BEGIN WITH THE PREPARED PRESENTATIONS.
CLAUDIO, ARE YOU HOOKED UP TO BE ABLE TO DO WHAT YOU WANT?
ARE WE ROLLING ON THE WEB CAST YET?
OKAY.
THANK YOU.
JUST FOR YOUR REFERENCE, BECAUSE WE'RE RUNNING UNDER SUCH TIGHT CONSTRAINTS OF TIME, UP UNTIL THE HIGH SIGN THAT I JUST GOT, THE SCRIBAL RECORD IS SECURE, BUT WE HAVE NOT BEEN VISIBLE ON THE NET.
SO THOSE OF YOU WHO ARE FOLLOWING THE WEBCAST, WELCOME.
YOU HAVE MISSED A FEW INTRODUCTORY REMARKS.
AND I'M SURE ALL OF THE SUBSTANCE OF WHAT HAS BEEN SAID SO FAR WILL BE REPEATED DURING THE WEBCAST ITSELF.

>>CLAUDIO MENEZES: OKAY.
GOOD MORNING.
I WILL SPEAK ABOUT MULTILINGUALISM EDUCATION AND THE CULTURE OF DIVERSITY IN THE DIGITAL WORLD.
MY TALK WILL HAVE THREE TOPICS: INTRODUCTION, MULTILINGUALISM EDUCATION AND CULTURE DIVERSITY, AND SOME COMMENTS ON IDN, INTERNATIONALIZED DOMAIN NAMES, MULTILINGUALISM, AND CULTURAL DIVERSITY, AND ASCERTAIN COMMENT ON FORTHCOMING ACTIVITIES AND CONCLUSIONS.
I THANK YOU VERY MUCH FOR HAVING INVITED UNESCO TO PARTICIPATE IN THIS IDN WORKSHOP.
IT'S THEREFORE MY PLEASURE TO CONGRATULATE ICANN, THE ORGANIZERS AND THE SPONSORS OF THIS WORKSHOP, FOR HAVING LAUNCHED THE DISCUSSION ON INTERNATIONALIZED DOMAIN NAMES AND ITS RELATION TO THE PROMOTION AND USE OF MULTILINGUALISM IN THE DIGITAL WORLD.
AS IT MIGHT BE KNOWN BY SOME OF YOU, UNESCO OPERATES IN CONFORMITY WITH THE WILL AND THE VOTE OF ITS 192 MEMBER STATES THROUGH ITS DECISION BODIES, NAMELY, THE GENERAL CONFERENCE AND THE EXECUTIVE BOARD.
THE 33RD GENERAL CONFERENCE TAKES PLACE THIS YEAR FROM 21ST OF OCTOBER IN PARIS, AND WE APPROVE THE PROGRAM AND THE BUDGET FOR 2006-2007.
THERE ARE MANY PROGRAMS, PROGRAMS, TRANSVERSAL AREAS AND OPERATIONAL MODALITIES, INCLUDING THE INFORMATION FOR ALL PROGRAMS TO WORK WITH UNESCO.
HOWEVER, DUE TO TIME CONSTRAINTS, I WILL NOT TO BE ABLE TO PRESENT THE DETAILS OF ALL THESE PROGRAMS, THEIR POLICIES, DECISIONS, AND ADVISORY BODIES DURING THIS TALK.
I INVITE YOU ALL TO VISIT OUR WEB SITES AND ALSO RAISE SPECIFIC QUESTIONS DURING THE DEBATES.
I WILL BE VERY MUCH PLEASED TO REPLY TO YOUR QUESTIONS.
SINCE IDN IS PART OF MULTILINGUALISM AND CULTURAL DIVERSITY IN THE DIGITAL WORLD, I DRAW YOUR ATTENTION TO THREE MAJOR RELEVANT DOCUMENTS DEALING WITH THE PROMOTION OF MULTILINGUALISM IN UNESCO.
THE UNIVERSAL DECLARATION ON CULTURAL DIVERSITY, THE CONVENTION ON THE PRESERVATION OF THE INTANGIBLE HERITAGE, AND THE RECOMMENDATION ON THE PROMOTION OF MULTILINGUALISM AND ACCESS, AND UNIVERSAL ACCESS TO THE CYBERSPACE, BOTH APPROVED BY GENERAL CONFERENCES OF UNESCO.
I AM WELL AWARE THAT THE MAIN OBJECTIVE OF THIS WORKSHOP IS THE REVIEW OF THE GUIDELINES FOR THE IMPLEMENTATION OF INTERNATIONALIZED DOMAIN NAMES.
I WOULD UNDERLINE ALSO THE CONCLUSIONS OF RECENT WSIS MEETINGS, MULTILINGUALISM, FOR CULTURAL DIVERSITY, AND THE PARTICIPATION OF ALL IN THE CYBERSPACE ORGANIZED BY UNESCO IN MALI ON THE 6TH AND 7TH OF MAY, 2005, IN WHICH SOME OF THE PEOPLE IN THIS PANEL ADDRESSED SOME OF THE ISSUES RELATED TO IDN.
THE FRAMEWORK OF UNESCO ACTION ON MULTILINGUALISM, CULTURAL DIVERSITY, AND THE PRESERVATION OF INTANGIBLE HERITAGE HAS BEEN DISCUSSED BY THE MEMBER STATES AND HAS BEEN PUT FORWARD THROUGH THE EARLIER MENTIONED RECOMMENDATIONS.
THE MOST RECENT DEVELOPMENTS TO PROMOTE LINGUISTIC AND CULTURAL DIVERSITY AT UNESCO ARE MULTILINGUALISM IN THE DIGITAL WORMED, INCLUDING (INAUDIBLE) OF WHICH YOU HAVE A BROCHURE HERE THAT CAN GIVE MORE DETAILS.
OF COURSE, THE WSIS, THEIR MEETINGS.
AND THE FORTHCOMING UNESCO PROGRAM FOR 2006-2007, INCLUDING INTERSECTORAL PROGRAM FOR LANGUAGES AT UNESCO, THE DOCUMENT 171EX-20.
THE WSIS ACTION PLAN HAS SOME AREAS IN WHICH MULTILINGUALISM IS ADDRESSED, C3, ACCESS TO INFORMATION AND KNOWLEDGE, AND THE CHAPTER C8, CULTURAL DIVERSITY AND IDENTITY, LINGUISTIC DIVERSITY, AND LOCAL CONTENT.
THESE ARE THE TRUE MAJOR CHAPTERS ADDRESSING MULTILINGUALISM.
THE LANGUAGE IS AN ENABLING COMPONENT OF THE INTANGIBLE HERITAGE, AND THIS IS A MAJOR REASON TO SEE OUR ORGANIZATION AT WORK IN PROMOTING AND PRESERVING LINGUISTIC AND CULTURAL DIVERSITY AND EXPRESSION AND SUPPORTING ACCESS TO INFORMATION IN ALL LANGUAGES, BOTH ONLINE AND OFFLINE.
THE DISSEMINATION OF KNOWLEDGE IS ESTABLISHED BY UNESCO CONSTITUTION, STATING -- UNESCO CONSTITUTION IN WHICH IT'S STATED THAT FULL AND EQUAL OPPORTUNITIES FOR EDUCATION FOR ALL IS -- AND THE FREE EXCHANGE OF IDEAS AND KNOWLEDGE ARE KEY POINTS.
GIVEN THE INSTITUTIONAL AND GENERAL FRAMEWORK OF UNESCO ACTIONS, BOTH RELATED TO WSIS AND ITS OWN PROGRAM, I WISH TO DRAW YOUR ATTENTION TO SPECIFIC ACTIONS OF MULTILINGUALISM IN OUR FIELDS OF COMPETENCE.
PERHAPS WE WILL NOT HAVE TIME TO PRESENT ALL THESE INITIATIVES THAT UNESCO HAS PROMOTED, BUT ONLY TO HAVE A VERY SHORT HIGHLIGHT, WE HAVE HAD ACTIONS ON EDUCATION, ON SCIENCE, ON CULTURE, AND ON COMMUNICATION AREA.
SOME EXAMPLES ARE SHOWN ON THE -- LET ME SEE, YEAH, ON EDUCATION AREA, YOU HAVE THE PRINCIPLES THAT UNESCO HAS ADOPTED IN IMPLEMENTING ITS PROGRAM: MOTHER TONGUE INSTRUCTION AS A MEANS OF IMPROVING EDUCATIONAL QUALITY.
BILINGUAL AND MULTILINGUAL EDUCATION AT ALL LEVELS AS A KEY ELEMENT OF INTERCULTURAL EDUCATION AND OF LINGUISTICALLY DIVERSE SOCIETIES.
RESPECT FOR THE EDUCATIONAL RIGHTS OF PERSONS BELONGING TO MINORITIES AS WELL AS INDIGENOUS PEOPLES, IN ORDER TO FOSTER FUNDAMENTAL FREEDOMS, LINGUISTIC DIVERSITY, AND UNDERSTANDINGS BETWEEN POPULATION GROUPS.
WE PRODUCED A POLICY PAPER: EDUCATION IN A MULTILINGUAL WORLD.
UNESCO PROMOTES THE INTERNATIONAL MOTHER LANGUAGE DAY, THE 21ST OF FEBRUARY OF EACH YEAR.
THESE ARE IN THE EDUCATION AREA.
IN CULTURE, UNESCO HAS PROMOTED CERTAIN NORMATIVE ACTIONS, UNIVERSAL DECLARATION OF CULTURAL DIVERSITY, THE CONVENTION FOR THE SAFEGUARDING OF THE INTANGIBLE CULTURAL HERITAGE.
RESEARCH AND INFORMATION, A CERTAIN NUMBER OF INITIATIVES HAVE BEEN UNDERTAKEN.
CULTURAL POLICIES, MEETINGS TOP ADVISE OUR MEMBER STATES ON ACTION PLANS ON MULTILINGUALISM, SUCH INTERNATIONAL ACTION PLAN WITH THE OBJECTIVE OF ADVISING MEMBER STATES OWN MEASURES TO SAFEGUARD ENDANGERED LANGUAGES, ESTABLISHING AN ADMINISTRATIVE AND FINANCIAL MECHANISM FOR SAFEGUARDING ENDANGERED LANGUAGES, AND ENHANCING UNESCO'S FUNCTION ON LANGUAGE DIVERSITY AND ENDANGERMENT ARE EXAMPLES OF THESE ACTIONS.
IN MY SECTOR, IN COMMUNICATION AND INFORMATION SECTOR, WE HAVE A NUMBER OF INITIATIVES, YOU HAVE ONLY SOME HEADLINES, LIKE INFORMATION FOR ALL PROGRAM, LIKE THE INTERNATIONAL PROGRAM FOR DEVELOPMENT OF COMMUNICATION, MEMORY OF THE WORLD, REGULAR PROGRAM OF UNESCO, THE PARTICIPATION PROGRAM, AND SOME ACTIONS RELATED TO THE WORLD SUMMIT ON THE INFORMATION SOCIETY.
THERE IS A DIRECTORY AVAILABLE FROM OUR WEB PAGE IN WHICH YOU HAVE ALL THE ACTIONS IN WHICH UNESCO IS INVOLVED AS RELATED TO THE WORLD SUMMIT ON THE INFORMATION SOCIETY.
A MAJOR POLICY DOCUMENT THAT WAS PREPARED BY OUR SECTOR, THE RECOMMENDATION ON THE PROMOTION AND USE OF MULTILINGUALISM AND UNIVERSAL ACCESS TO CYBERSPACE.
THIS DOCUMENT HAS FOUR MAJOR TOPICS: THE DEVELOPMENT AND PROMOTION OF MULTILINGUAL CONTENT AND SYSTEMS.
I TAKE THIS OPPORTUNITY TO SPEAK VERY BRIEFLY THAT IT'S VERY (INAUDIBLE) TO SAY, WELL, WE HAVE TO PROMOTE LOCAL CONTENT, YES.
BUT, UNFORTUNATELY, LOCAL CONTENT IS THE FIFTH ELEMENT OF A CHAIN.
IF YOU WISH TO PROMOTE LOCAL CONTENT IN CYBERSPACE, YOU HAVE TO DEVELOP SOFTWARE.
IF YOU HAVE TO DEVELOP SOFTWARE, YOU HAVE TO HAVE A CODING, TO HAVE THE CODE, YOU HAVE TO HAVE THE SCRIPTS.
AND THEN YOU CAN GO TO LOCAL CONTENT.
OTHERWISE, YOU PROMOTE LOCAL CONTENT, USE THE TRADITIONAL POINTS.
WELL, I HAVE VERY -- FOUR MINUTES TO FINISH.
SO I WILL SIMPLY JUMP A LITTLE BIT TO THE -- SOME TOPICS THAT ARE RELATED TO THE OBJECTIVE OF THIS WORKSHOP.
PART OF IT ARE INCLUDED IN ONE OF UNESCO'S INITIATIVES, B@BEL.
YOU HAVE THE BROCHURE, YOU HAVE A LOT OF DETAILS ON THE INITIATIVES AND ON THE ACTIVITIES PROMOTED THROUGH B@BEL INITIATIVE, WHICH IS ONE OF UNESCO'S INITIATIVES ON MULTILINGUALISM.
YOU HAVE ON THE SCREEN A MORE OR LESS SELF-EXPLANATORY REASON WHY UNESCO HAS BEEN PROMOTING MULTILINGUALISM.
THERE IS A STRATEGY OF B@BEL.
THERE ARE CURRENT PARTNERS BOTH IN UNESCO AND EXTERNAL PARTNERS LIKE UNIVERSITY OF BERKELEY, INFOTERM, AND OTHERS. AND IT'S STILL AN ONGOING ACTIVITY.
WE ARE VERY MUCH WORRIED ABOUT THE RISK OF DISAPPEARANCE OF LANGUAGES. AND IT'S ONE OF THE REASONS WHY UNESCO HAS BEEN WORKING ON MULTILINGUALISM.
IDN, MULTILINGUALISM, EDUCATION AND CULTURAL DIVERSITY.
WE DO NOT HAVE A SPECIFIC MANDATE ON IDN IN INTERNATIONALIZED DOMAIN NAMES.
BUT UNESCO HAS THE GENERAL PRINCIPLES THAT ARE, BY THE WAY, REFLECTED IN THE DOCUMENT ON INTERNET GOVERNANCE THAT UNESCO ISSUED ABOUT THESE PRINCIPLES, WHICH ARE MORE OR LESS COHERENT WITH OUR MANDATE, WHICH ARE FREEDOM OF EXPRESSION, CULTURAL AND LINGUISTIC DIVERSITY, AND OPENNESS.
AND WE, OF COURSE, HOPE THAT THE FRAMEWORK FOR INTERNATIONALIZED DOMAIN NAMES WILL ADHERE TO THESE PRINCIPLES.
WE -- OF COURSE, WE HOPE THAT IDN WILL BE A MECHANISM ENABLING GREATER USE OF INTERNET, GREATER PARTICIPATION OF THE INFORMATION -- IN THE INFORMATION WORLD, INCREASING NUMBER OF USERS FROM DIVERSE LINGUISTIC AND CULTURAL BACKGROUNDS.
AND WE ARE, OF COURSE, WILLING TO INTERACT WITH OTHER INSTITUTIONS TO IMPLEMENT THESE PRINCIPLES.
I KNOW THAT I -- MY TIME IS ABOUT TO FINISH, AND I WOULD ONLY COMPLETE MY PRESENTATION BY SAYING THAT WE ARE PROMOTING FOLLOW-UP ON THESE ACTIVITIES ON THE MULTILINGUALISM FOR CULTURAL DIVERSITY IN CYBERSPACE, AND ALSO THE OTHER INITIATIVE, WHICH WAS HELD IN ST. PETERSBURG ON CULTURAL DIVERSITY AND SOCIETIES.
AND ON THE DOCUMENTS THAT I COMMENTED BEFORE, RECOMMENDATION ON MULTILINGUALISM IN THE UNESCO ACCESS TO CYBERSPACE, AND AT THE VERY END, AS A SUMMARY, THE CONCEPT THAT UNESCO IS PROMOTING BOTH FOR INTERNET AND BEYOND INTERNET IS THAT WE HAVE TO PROMOTE KNOWLEDGE SOCIETIES BASED IN THESE FOUR COLUMNS, KNOWLEDGE CREATION, KNOWLEDGE PRESERVATION, KNOWLEDGE DISSEMINATION, KNOWLEDGE UTILIZATION.
YOU HAVE MANY OTHER DOCUMENTS OF UNESCO TALKING ABOUT THIS.
BUT I'M SORRY THAT THE TIME IS VERY SHORT AND I CANNOT GO ABOUT IT.
BUT I'M HERE READY TO TALK TO YOU IN THE QUESTIONS.
AND THANK YOU VERY MUCH AGAIN.
BYE BYE.
(APPLAUSE.)
>>CARY KARP: WE'RE GOING TO HOLD QUESTIONS UNTIL THE THREE PRESENTATIONS HAVE BEEN MADE. SO MICHEL, ARE YOU IN A POSITION TO CONTINUE NOW?
ALL OF THE SPEAKERS ARE HERE TO PARTICIPATE TO THE FULL EXTENT OF THEIR KNOWLEDGE AND ABILITY IN THE DISCUSSION WE WILL BE OPENING. SO THERE'S NOTHING THAT ANY OF THE PANELISTS THAT THEY MIGHT WISH TO SAY THAT THEY WON'T HAVE THE OPPORTUNITY TO HAVE SAID DURING THE COURSE OF THE SESSION.
>>MICHEL SUIGNARD: SO, THE FIRST THING I WOULD LIKE TO DO AS A REMARK FROM THE PRESENTATION FROM MY COLLEAGUE, THAT IDN IS NOT REALLY ABOUT REPRESENTATION OF ALL CONTENT. IT'S REALLY ABOUT DEFINING RESOURCES.
SO IT'S A BIT DIFFERENT. THE MANDATE OF IDN IS QUITE A BIT MORE RESTRICTED THAN TRYING TO ADDRESS, YOU KNOW, A LANGUAGE NEED OF ALL THE WORLD COMMUNITY.
SO WE REPRESENT A LARGER -- AS LARGE -- I MEAN AS LARGE AS POSSIBLE, BUT WE HAVE SOME RESTRICTION THAT WE WILL SEE LATER DURING THIS TALK AND ALSO, I GUESS, DURING THE OTHER PRESENTATIONS.
THE FIRST THING I WOULD LIKE TO SAY, NOW WE SEE THAT IDN IS A STANDARD, WE ARE FACED WITH IMPLEMENTING IT. WHEN WE LOOK AT THAT, WE HAVE FOUND ISSUES THAT I WOULD LIKE TO DESCRIBE HERE.
THE FIRST THING WE SEE, IT'S A VERY LARGE REPERTOIRE. IT'S A MOVABLE REPERTOIRE. IT'S NOT FIXED. AND THERE'S A LOT OF THINGS IN IT THAT ARE NOT VERY APPROPRIATE FOR DEFINING RESOURCES, AGAIN. DO WE REALLY NEED SYMBOLS? DO WE NEED, LIKE, CHARACTERS THAT ARE NOT EVEN USED IN MODERN USE. THEY'RE COMPLETELY EXTINCT. WE'RE NOT TALKING ABOUT DYING LANGUAGES. WE'RE TALKING ABOUT KUNIER FORM OR URI GLYPH, ALL ITALIC. THERE ARE A LOT OF THINGS THAT COULD BE IN IDN IF WE KEEP THE (INAUDIBLE) PROCESS. MANY PEOPLE THINK THEY ARE PROBABLY NOT APPROPRIATE FOR THAT.
SO YOU'VE GOT CONSORTIUM CREATED, GUIDELINE TO CREATE NOTIFIERS BY TAKING A SUBSET OF THE WORLD UNICODE REPERTOIRE, THAT IS CALLED UAX31. THAT CAN BE FOUND ON THE (INAUDIBLE) WEBSITE. THE REPERTOIRE FOR IDN IS NOT ALIGNED WITH THAT. THERE ARE CHARACTERS THAT (INAUDIBLE) SHOULD NOT BE NOTIFIERS THAT ARE TODAY NOT ALLOWED IN IDN.
SO NOW COMING TO THE IDN GUIDELINES. THE CURRENT GUIDELINES ARE LANGUAGE BASED. THE WORD LANGUAGE IS REPEATED TEN OR 15 TIMES IN THE TWO PAGES OF GUIDELINES. THE SCRIPT CONCEPT IS NEVER MENTIONED. AND WE REALLY THINK THAT TO SOME DEGREE, THE LANGUAGE CONCEPT IS OVER USED. I WILL COME TO THAT POINT LATER.
CONFUSABLE CHARACTERS IS NOT ADDRESSED. AGAIN, I WILL COME TO EXPLAIN WHAT I MEAN BY CONFUSABLE ADDRESS CHARACTERS. THERE'S A SAFETY ISSUE, AS HAS BEEN SEEN EARLIER THIS YEAR WITH THE FAMOUS PAPAL EXAMPLE THAT MANY OF YOU ARE AWARE. AND THEN UNICODE IS STUCK AT 3.2, MEANING THERE'S NOT A COMMUNITY REPRESENTED IN UNICODE NOT IN LINE WITH 3.2 WITH IDN TODAY IS OUT OF LUCK. THAT MEANS REPERTOIRE FROM N'KO IN AFRICA OR TIFINAGH IN NORTH AFRICA CANNOT BE REPRESENTED IN IDN. ALSO WE HAVE DONE AN UPDATE TO EXISTING REPERTOIRES, ESPECIALLY ON THE INDIAN CONTINENT FOR THE MANY INDIAN LANGUAGES. THOSE CANNOT BE CORRECTLY PRESENTED IN 3.2. THE LAST POINT IS A MINOR POINT BUT IT'S VERY IMPORTANT BY MANY THAT THE NORMALIZATION SCHEMED USED IN 3.2, THERE WERE ERRORS FIXED LATER, BUT BY THE WAY IT IS ALLOWED, WE'RE NOT ALLOWED TO REFER TO IT.
SO THE DEFINITIONS OF CONFUSABILITY. YOU HAVE MANY WAYS TO CONFUSE CHARACTERS. THEY CAN BE SINGLE SCRIPT, WHICH IS PROBABLY THE MORE DANGEROUS ONE, BECAUSE PEOPLE DON'T REALLY -- BASICALLY WE DON'T PUT PROTECT AGAINST SINGLE SCRIPT SPOOFING, TYPICALLY. IT'S KIND OF DIFFICULT. HERE IS AN EXAMPLE WHERE I USE THE FACT THAT YOU CAN RIGHT THE O SLASH AS A CHARACTER OR AS AN O FOLLOWED BY THE COMBINING SLASH. THOSE ARE NOT NORMALIZED BY IDN. AS YOU CAN SEE ON THE SCREEN, THEY LOOK PRETTY SIMILAR. IN FACT, THEY'RE COMPLETELY DIFFERENT CHARACTERS. AND THEN IF A REGISTRY ALLOWS BOTH TO BE ENTERED THEY WILL LOOK THE SAME OR THEY WILL BE VERY, VERY DIFFERENT.
MIXED SCRIPT CONFUSABLE IS WHEN YOU USE ONE CHARACTER FROM ONE SCRIPT IN ANOTHER SEQUENCE. ON THAT THERE IS THE FAMOUS PAPER EXAMPLE WHERE "A" IN RED IS, IN FACT, A CYRILLIC CHARACTER. I DID USE ON MY MACHINE THE CYRILLIC "A," AND YOU CAN'T SEE A DIFFERENCE.
AND THEN THERE'S THE LAST POINT PROBABLY NOT AS COME, IF THE WHOLE SCRIPT CONFUSABLE WHERE YOU CAN REPRESENT A SEQUENCE OF CHARACTER FROM A COMPLETELY DIFFERENT SCRIPT THAN WHAT YOU THOUGHT IT WAS. LIKE SCOPE CAN TODAY BE REPRESENTED TODAY BOTH IN LATIN AND CYRILLIC.
SO UNICODE RECOMMENDATIONS. THOSE ARE IN FACT EXPRESSED ON TR 36 OF THE -- ACCESSIBLE ON THE UNICODE WEBSITE. FIRST OF ALL, IT'S REALLY A GUIDANCE TO ALL CONSTITUENCIES. IT DOESN'T RESTRICT TO PROGRAMMERS. YOU REALLY COVER CUSTOMER AS END USERS, REGISTRIES, REGISTRARS, USER AGENTS, BROWSERS. DEFINE RESTRICTION LEVELS, BY THE FACT THAT WE SAY AS IDN IS DEFINED TODAY, WE CANNOT PREVENT PEOPLE TO USE IDN ALL EXTENT. SO IF PEOPLE WANT TO BE UNSAFE, IF YOU WANT, WE CANNOT BLOCK THAT AT THIS POINT BECAUSE THAT'S WHAT IDN SAYS YOU HAVE TO DO.
SO YOU CAN GO FROM THE COMPLETE RESTRICTED LEVEL WHERE YOU ONLY ALLOW ASCII AS IF YOU HAD NO IDN TO BASICALLY ALLOWING SINGLE SCRIPT, MULTIPLE SCRIPT, OR UNRESTRICTED.
ONE THING THAT IT DOES DEFINE IS A NEW UNICODE COLLECTION, IS BASICALLY A SEQUENCE OF CHARACTER CODES, WHICH IS A SUBSET OF THE CURRENT IDN REPERTOIRE. IT MAINTAINS THE LDH PRINCIPLE. LDH IS BASICALLY IN DNS SPEAK LATIN CHARACTERS, DIGITS AND HYPHENS.
SO WE MAINTAIN AS MUCH AS WE CAN THE PRINCIPLE ON UNICODE. SO BY REMOVING NOT IN MODERN USE CHARACTERS, REMOVING SYMBOLS, ON RESTRICTING INPUT CHARACTERS, BY BASICALLY INPUT CHARACTERS, AGAIN THAT'S ANOTHER THING, IDN, THAT YOU HAVE A CONCEPT OF INPUT AND OUTPUT. OUTPUT IS WHAT YOU RESERVE BUT IN FACT ON AN OUTPUT THAT IS THE USER INTERFACE YOU ARE ALLOWED MORE CHARACTERS THAT ARE PRESENTED TO THE USER AS VALID IDN CHARACTERS. I DON'T HAVE TIME TO GO INTO DETAIL ON WHAT IT IS BUT WE ARE TRYING TO RESTRICT THE INPUT CHARACTERS TO THE UPPER CASE VARIATION OF WEB BROWSER CHARACTERS THAT CAN BE RESOLVED. AND THEN ON PROCESSING RESTRICTIONS, WE BASICALLY -- IF YOU HAVE A WHITE LIST, BASICALLY WHITE LIST IS YOUR KNOWN LIST OF SAFE USER, YOU CAN SEE WE CAN DO MUCH BETTER ON MAKING SURE THAT YOU GO TO A SAFE SITE.
DON'T ALLOW MULTI-SCRIPT LABELS WITH KNOWN EXCEPTIONS. SO THE EXCEPTIONS, WE KNOW IN JAPANESE AND KOREAN THAT NEEDS THAT BUT DOESN'T WANT, IN MOST CASE, DOESN'T NEED TO MIX SCRIPT IN A SINGLE LEVEL.
DON'T ALLOW A NAME WHICH IS VISUALLY CONFUSABLE WITH A WHITE LIST CONTENT.
ON ALERT, IF A NAME IS WHOLE SCRIPT OR MIXED SCRIPT CONFUSABLE.
SO I THEN GO INTO ICANN GUIDELINES BECAUSE THAT'S THE TOPIC OF THE DISCUSSION.
FIRST WE WOULD LIKE TO SEE A SUBSET OF THE CURRENT IDNA REPERTOIRE, SO LIKE WAS MENTIONED BY CARY AT THE BEGINNING, WE WOULD LIKE TO HAVE A SMALLER REPERTOIRE. I SOON WOULD LIKE TO HAVE A FOCUS ON SCRIPT INSTEAD OF LANGUAGES, BECAUSE FIRST OF ALL, LANGUAGE REPERTOIRE ARE VERY DIFFICULT TO DETERMINE. I MEAN AS YOU CAN SEE TODAY, IN FACT THERE'S A DIFFERENCE IN THE CCTLD BETWEEN PORTUGAL AND BRAZIL. THEY DON'T AGREE EXACTLY ON WHAT PORTUGUESE IS.
NOT MUCH, BUT A SLIGHT DIFFERENCE.
SCRIPT CONTENT IS MUCH EASIER TO DEFINE BECAUSE IT'S BASICALLY AN INTERSECTION BETWEEN THE SCRIPT PROPERTY DEFINED BY THE UNICODE CONSORTIUM ON THE SUBSET I JUST TALKED BEFORE.
I'M BASICALLY ALSO SAYING AVOID THE CREATION OF SCRIPT SUBSETS FOR GTLD AND STLD. CCTLD IS DIFFERENT BECAUSE I CAN UNDERSTAND THEY MAY WANT TO HAVE THEIR OWN SET.
AND THEN WE WOULD LIKE TO SEE A TEMPLATE IN ICANN THAT ALLOW TO DO THAT. TODAY'S TEMPLATE IS ONLY FOR LANGUAGES. THERE'S NO TEMPLATE FOR A GROUP OF LANGUAGES OR FOR SCRIPT.
ALSO, RECOGNIZE THAT SOME CHARACTERS CANNOT BE FIRST. I MEAN, YOU CANNOT PUT THE COMBINING CHARACTERS IN THE FIRST POSITION IN A LABEL. THAT HAS TO BE CLEARLY SAID. WE HAVE TO HAVE A LIST OF THOSE CHARACTERS THAT CANNOT BE CREATED AS A FIRST CHARACTER. AGAIN, THAT'S A SAFETY ISSUE.
AND THEN FORMALIZE MULTISCRIPT RESTRICTIONS ON LABEL. IT WOULD BE NICE IF ICANN WAS RECOGNIZING THAT MULTISCRIPT ON LABELS IS REALLY A BAD IDEA, TYPICALLY.
IDN, INPUT/OUTPUT CHARACTER STATUS I MENTIONED.
I KNOW IN THE NEXT SESSION, I KNOW AS IMPLEMENTERS OF IDN, WE HAVE TO ALLOW THAT ON THE LOCAL NETWORKS, SO WHAT DO YOU SIGN ON IDN, IDN IS IN FACT GOING TON IMPLEMENTED BEFORE BY LOCAL NETWORKS. NETWORKS THAT LIVE BEHIND THE FIREWALL WHICH ARE VERY IMPORTANT IN LARGE CORPORATIONS TODAY.
SHOULD IDN BE REVISED? THAT'S ANOTHER QUESTION I PUT TO THE AUDIENCE OR MEMBERS. DO WE NEED TO REVISE IDN OR DO WE NEED TO LEAVE IDN ON CREATIVE FRAMEWORK -- LIMITING FRAMEWORK WITHIN IDN FOR THE ICANN NEEDS.
JUST A QUICK THING I WILL SHOW. THE EXTENT OF THE PROBLEM WE HAVE TODAY WITH THE NEW SET, BECAUSE PEOPLE HAVE BEEN LOOKING AT WHAT IS THE SUBSET WE JUST CREATED. AND IN FACT THAT'S THE EXTENT OF THE SYMBOLS WE HAVE TODAY IN THE NEW LIST. WE WENT DOWN QUITE A BIT ON THE NUMBER OF SYMBOLS.
BASICALLY THE IDEA WAS TO REMOVE THE SCRIPT WITH NO -- TO REMOVE THE PUNCTUATION FROM THE SCRIPTS, AND THEN TO REDUCE THE (INAUDIBLE) ACCOUNT. AND TO THE CCTLD, ADD IICORE COLLECTION. (INAUDIBLE).
SO THAT'S MY PRESENT. I WILL PASS TO PAT.
>>CARY KARP: WHILE PAT IS BUSY SETTING UP, SOMETHING I PROBABLY SHOULD HAVE SAID DURING THE INTRODUCTORY REMARKS, THE ISSUE OF IDN AS IT APPEARS ON THE TOP-LEVEL IN A DOMAIN NAME IS THE SUBJECT OF THE SECOND SESSION OF THIS WORKSHOP. SO AT PRECISELY 12:30 THIS SESSION WILL TERMINATE AND THE NEXT WILL START. SO ANY ISSUES TO THE WAY DOMAIN NAMES APPEAR IN TOP LEVEL REPRESENTATIONS IS A TOPIC ALL OF ITS OWN.

>>PAT KANE: I'M PAT KANE WITH VERISIGN AND I'M SPEAKING TODAY ON BEHALF OF THE GTLD CONSTITUENCY. EARLIER THIS WEEK WE GOT TOGETHER AND TALKED ABOUT IDNS AND TALKED ABOUT THIS PARTICULAR PANEL SESSION, AND THE TWO THINGS THAT WE CAME AWAY WITH WERE THAT, ONE, THERE CLEARLY IS MARKET DEMAND FOR IDNS. PEOPLE WANT TO USE IDNS. PEOPLE WANT TO REPRESENT THEMSELVES IN THEIR NATIVE CHARACTER SETS. AND THE SECOND THING IS THAT THERE IS VALUE IN THE REGISTRIES, AT LEAST THE GTLD REGISTRIES, HAVING A CONSISTENT APPROACH TO DELIVERING IDNS. AND WE ALSO BELIEVE THERE'S VALUE IN THE CCS ALSO COMING ALONG WITH SOME CONSISTENT APPROACHES TOWARDS IDNS.
SO I WANTED TO PRESENT A FEW STATEMENTS THAT WE HAVE OUT OF THE GTLD REGISTRY CONSTITUENCY.
THE FIRST BEING THAT MEMBERS OF THE GTLD REGISTRY CONSTITUENCY ARE COMMITTED TO SERVING THE NEEDS OF COMMUNITIES DESIRING TO BUILD AN ONLINE IDENTITY USING THE CHARACTERS AND SCRIPT THAT THEY CHOOSE TO REPRESENT THEMSELVES. WE DO RECOGNIZE THERE IS MARKET DEMAND AND WE WANT TO BE ABLE TO MEET THE MARKET DEMAND.
THE NEXT ADDRESS THE ISSUES AROUND THE GUIDELINES THAT WE HAVE TODAY.
AS LONG AS THE CONCEPT OF RELATING LANGUAGE TO A DOMAIN EXISTS, AND TO MICHEL'S POINT, DOES IT MAKE -- DOES IT STILL MAKE SENSE, THE GTLD REGISTRIES WILL FOLLOW ESTABLISHED AUTHORITATIVE LANGUAGE TABLES. THE KEY THERE WILL BE WHAT IS AUTHORITATIVE AND WHO IS AUTHORITATIVE.
THE THIRD PART, WHERE LANGUAGE TABLES DO NOT EXIST, THE GTLD REGISTRIES WILL NOT PERMIT THE COMMINGLING OF SCRIPTS IN A DOMAIN LABEL. THAT IS WE WILL NOT ALLOW TWO SCRIPTS TO BE REPRESENTED WITHIN THE SAME DOMAIN REGISTRATION.
AND WHERE LANGUAGE TABLES ARE DESIRED, WE WILL SEEK FUNDING AND APPROPRIATE AUTHORITATIVE BODY TO ASSIST WITH THE DEVELOPMENT OF SUCH TABLES.
SO THAT'S WHERE THE STATEMENT THAT WE HAVE FROM THE GTLD REGISTRY CONSTITUENCY IN TERMS OF HOW WE'RE GOING TO MOVE FORWARD WITH OUR IDN PROGRAMS.
AS FAR AS THESE BEING THE IDN GUIDELINES THAT WE HAVE FROM -- THAT WERE ESTABLISHED RIGHT BEFORE THE MARCH -- MONTREAL ICANN MEETING IN 2003, THERE ARE TWO ISSUES I CONTEND WITH ON A REGULAR BASIS IN TERMS OF RUNNING THE PROGRAM. THE FIRST COMES FROM THE TECHNICAL COMMUNITY AND THE SECURITY COMMUNITY AROUND THE COMMINGLING OF SCRIPTS.
SO DOES A SINGLE SCRIPT AND A SINGLE DOMAIN WORK? WE'VE ALREADY GIVEN THE EXAMPLE ALREADY TODAY AROUND WHAT HAPPENS IN SOME OF THE ASIAN LANGUAGES, CHINESE, KOREAN, AND JAPANESE AS TO THERE ARE APPROPRIATE REASONS TO COMMINGLE SCRIPTS. THEY HAVE DEVELOPED TABLES THAT WE CAN FOLLOW IN ACCORDANCE WITH THE STATEMENT WE HAVE MADE ALREADY.
JUST SOME OTHER ITEMS. .2% OF THE COM/NET REGISTRATIONS COMBINE SCRIPTS QUESTIONABLY. THAT'S MY DETERMINATION OF WHAT IS QUESTIONABLE AND FROM WHERE WE ARE I TAKE A LOOKING AT BASIC LATIN AS BEING SOMETHING THAT PEOPLE LIKE TO COMBINE WITH OTHER SCRIPTS.
75% OF THE CYRILLIC-LATIN COMBINATIONS ARE NUMBERS AND DASHES TODAY. SO WE WHEN WE SEE A LANGUAGE COME IN TAGGED AS LATIN OR TAGGED AS CYRILLIC, 75% OF THOSE ARE JUST NUMBERS AND DASHES.
SOME OF THE OTHER COMMINGLING ISSUES THAT WE HAVE, THERE ARE SOME CHARACTERS THAT ARE USED ACROSS MULTIPLE SCRIPTS AGAIN. WE SEE COMPANIES THAT ARE INCREASINGLY BRANDING THEMSELVES USING SYMBOLS, AND SOME LANGUAGES HAVE BEEN REPRESENTED IN MULTIPLE SCRIPTS FOR THE LAST 125 YEARS OR EVEN LONGER.
AND THE QUESTION THAT I ALWAYS HAVE IS DO CHARACTERS EVER BLEND IN TERMS OF USAGE, IN TERMS OF REPRESENTATION IN THOSE COMMUNITIES.
AS FAR AS LANGUAGE TABLES, I HAVE A QUESTION THAT COMES IN FROM OUR CUSTOMERS, AND THEIR CUSTOMERS AT LEAST ON A WEEKLY BASIS, AND THAT IS WHAT DOES THE LANGUAGE TABLE MEAN, WHAT DOES THE LANGUAGE TAB MEAN IN TERMS OF WHEN I ENTER MY DOMAIN. IF THEY PUT IN ALL ASCII CHARACTERS THEY DON'T HAVE TO HAVE A TAG BUT IT COULD CERTAINLY BE THE GERMAN LANGUAGE.
SO WHEN WE TAG A LANGUAGE -- OR A REGISTRATION WITH A LANGUAGE, WHAT DO WE MEAN? IS IT THE GERMAN ALPHABET, 29 CHARACTERS; IS IT THE CHARACTERS COMMONLY USED WITHIN THE GERMAN LANGUAGE OR SOMETHING THAT DENIC IS DOING IN TERMS OF A CODE PAGE WITH THAT HANDLES THE LANGUAGES USED IN GERMANY?
AND THEN WHEN WE COME TO AUTHORITY OF WHO CAN SPEAK FOR A LANGUAGE TABLE, ARABIC AS A SCRIPT HAS ONE TIME OR ANOTHER SUPPORTED ALL OF THESE LANGUAGES LISTED THERE. THE ONES THAT ARE BOLDED ARE THE ONES CURRENTLY USING ARABIC TODAY. SO WHO IS GOING TO SPEAK FOR THE ARABIC TABLE AND WHAT IS ALLOWABLE OR NOT ALLOWABLE?
SO MAYBE LANGUAGE TAGS ONLY CONFUSE THE ISSUE. TO MICHEL'S POINT, MAYBE IT IS ABOUT SCRIPTS AND MAYBE IT'S NOT ABOUT LANGUAGE.
AND THE GUIDELINES ADDRESS VARIANT TABLES AND NOT CHARACTER TABLES. SO WHEN WE TALK ABOUT VARIANT TABLES, WE THINK OF THE CHINESE AND SIMPLIFIED AND TRADITIONAL CHINESE TABLES WHERE THEY MAP ONE TO THE OTHER OR SEVERAL TO ANOTHER.
SO WHILE I WAS ABLE I WANTED TO PROVIDE A FEW STATISTICS AROUND IDNS AROUND GTLDS, SPECIFICALLY COM/NET AND INFO AND WORLDWIDE.
TODAY, ONE AND A HALF PERCENT OF ALL DOMAINS ARE IDN'S. THERE ARE 79.1 MILLION DOMAINS REGISTERED WORLDWIDE, AND ABOUT A MILLION OF THEM ARE IDN'S. 37% OF INTERNET USERS ONLY ARE ENGLISH SPEAKING AND THAT NUMBER GETS SMALLER AND SMALLER EVERY DAY. AND TODAY WE HAVE OVER 40 TLDS THAT REGISTER IDNS AND THEY ARE GROWING EVERY DAY. WITHIN COM/NET THE LARGEST LANGUAGES WE HAVE REGISTRATIONS FOR ARE KOREAN, GERMAN AND CHINESE. OF OUR REGISTRARS, WE HAVE 121 REGISTRARS WHO HAVE IDNS UNDER MANAGEMENT TODAY. NOT ALL OF THEM ARE ACTIVELY SELLING IDNS, BUT 121 HAVE A -- ARE MANAGING AT LEAST ONE IDN.
THIS NEXT ONE IS SOMETHING THAT I FOUND VERY INTERESTING IN DECEMBER OF LAST YEAR. OUR RENEWAL RATE ON IDNS APPROACHED 70%. THAT IS VERY BIG FROM THE STANDPOINT OF PEOPLE BELIEVING IN WHAT THEY HAVE IN TERMS OF A REGISTRATION. OUR ASCII RATE IS JUST SLIGHTLY ABOVE THAT. SO THEY ARE ON PAR WITH OUR OTHER DOMAINS.
RESOLUTION TRAFFIC IS GROWING. IN NOVEMBER OF LAST YEAR, I'M SURE EVERYONE IS AWARE, FIRE FOX PUT OUT THEIR 1.0 VERSION OF THEIR BROWSER. WE STARTED TO SEE RESOLUTION TRAFFIC GROWING, GROWING BASED UPON THAT. SO I BELIEVE THAT'S DIRECTLY ATTRIBUTABLE TO FIRE FOX.
WE ALSO HAVE 62% OF OUR COM/NET IDNS RESOLVING TO A LIVE SITE. SO PEOPLE ARE USING THEM AND WORKING WITH THEM. RAM HAS PROVIDED INFORMATION AROUND .INFO AND ON THEY REGISTERED 15,000 ON THE FIRST DAY OF REGISTERING GERMAN NAMES.
SO IF WE HAVE A LOT OF ISSUES THAT PEOPLE HAVE POINTED OUT AROUND IDNS, WHY SHOULD WE CONTINUE. IT'S STILL REALLY ONLY HALF A PRODUCT. WE HAVE ONLY INTERNATIONALIZED THE HOST NAME. WE DON'T HAVE ANYTHING TO THE LEFT OF THE @ AND NO TLD. SO WHY SHOULD WE CONTINUE? WE STILL HAVE LOCALIZED DEMAND. THE ONLINE POPULATION AGAIN IS BECOMING MORE NON-ENGLISH SPEAKING EVERY SINGLE DAY. WE STILL HAVE OVER A MILLION IDNS WORLDWIDE. RESOLUTION NUMBERS INCREASING. RENEWAL RATES ARE INCREASING. CCTLDS CONTINUE TO DEPLOY IDN TECHNOLOGY. IN APRIL OF THIS YEAR, VERISIGN DID A LAUNCH IN SOUTH AMERICA AND WE HAD 34 REPORTERS COME TO LAUNCH, 32 ARTICLES WERE WRITTEN AND THE LEAD ITEM IN MOST OF THOSE ARTICLES IS YOU COULD REGISTER PORTUGUESE DOMAIN NAMES. WE SAW AN IMMEDIATE 30 PERCENT INCREASE IN REGISTRATIONS BASED UPON THAT AND IN THIS IN THREE OR FOUR WEEKS .BR WAS OFFERING IDNS AND THEN CHILE IS OFFERING IDNS AS WELL. SO THERE IS DEMAND. AND WE'RE STILL TALKING ABOUT IDNS AT ANY ICANN MEETING AT DOMAIN EVENTS. SO IT'S STILL A TOPIC, STILL AN ISSUE.
THE THING I WORRY ABOUT IS IF DOMAINS CAN'T PROVIDE THE EXPERIENCE, IF WE CAN'T FIGURE OUT HOW TO PROVIDE A DOMAIN EXPERIENCE FOR THE END USER, OTHER PRODUCTS WILL, AND WE WON'T BE SELLING IDNS AT ALL.
THANK YOU.
(APPLAUSE.)
>>CARY KARP: ALL RIGHT. BEFORE WE ALLOW THE AS-YET SILENT PANELISTS TO CONTRIBUTE THEIR THINKING TO ALL OF THIS, I WOULD LIKE TO OPEN THE MICROPHONE. IF THERE ARE ANY PEOPLE IN THE AUDIENCE WHO HAVE SPECIFIC QUESTIONS TO THE SPEAKERS THUS FAR OR ADDRESSING ANY ASPECT OF THE TOPIC OF THIS SESSION, PLEASE, THE FLOOR IS YOURS.

>>ANDRZEJ BARTOSIEWICZ: ANDRZEJ BARTOSIEWICZ FROM (INAUDIBLE). I WOULD LIKE TO KNOW WHEN MICROSOFT WILL SUPPORT -- WILL START SUPPORTING IDNS. WE ARE DISCUSSING THIS FOR TWO YEARS ALREADY OR MORE. AND WE ARE STILL WAITING FOR THE NEW VERSION OF THE INTERNET EXPLORER SUPPORTING IDNS. WE HAVE FIRE FOX ALREADY.
>>MICHEL SUIGNARD: I'M NOT SURPRISED BY THAT QUESTION, AS YOU CAN IMAGINE. IN FACT, WE'RE HAPPY TO BE A BIT LATE TO THE GAME WHEN WE HAVE SEEN IN FACT THE SECURITY ISSUES THAT THIS CREATED. IN FACT, IT'S KIND OF GOOD THAT WE TOOK OUR TIME TO GET THIS.
WE HAVE BEEN PUBLIC NOW SAYING E 7 WILL SUPPORT IDN. THERE WILL BE MULTIPLE (INAUDIBLE), A BETTER VERSION. IT'S MY UNDERSTANDING THE FIRST VERSION OF BETTER WILL NOT HAVE 87, BUT FOLLOWING VERSION WILL SUPPORT IDN AND THE LAST VERSION, SHIPPING VERSION WILL SUPPORT IDN.
>>CARY KARP: ANYTHING ELSE?
THAT WAS PROBABLY THE MOST ANTICIPATED QUESTION OF ALL THAT COULD POSSIBLY HAVE BEEN ASKED, AND I AM DELIGHTED THAT IT HAS BEEN ASKED AT THE OUTSET OF THE DISCUSSION.
AND HERE IS YET ANOTHER SURPRISE IN PRESENT CONTEXT. JOHN.
>>JOHN KLENSIN: JOHN KLENSIN. MICHEL, YOU MADE A NUMBER OF VERY INTERESTING RECOMMENDATIONS, SOME OF WHICH DIRECTLY CONTRADICTED THE RECOMMENDATIONS THE UTC SEEMED TO BE MAKING WHEN THE IDNA STANDARDS WAS FIRST PUT TOGETHER. I'M GLAD TO SEE WE'RE LEARNING IN THIS PROCESS. BUT SOME OF THOSE RECOMMENDATIONS RAISE TWO OTHER ISSUES THAT I'D LIKE YOU TO ADDRESS IF POSSIBLE.
ONE OF THOSE ISSUES IS TO THE EXTENT TO WHICH THESE CHANGES ARE MADE BY REGISTRATION RESTRICTIONS, RATHER THAN BY CHANGING IN THE STANDARD, THERE IS NO POSSIBLE ENFORCEMENT MECHANISM BELOW ABOUT THE SECOND LEVEL OF THE DNS. AND THE BROWSER VENDORS, FOR EXAMPLE, HAVE BEEN EXTREMELY CONCERNED THAT IF SPOOFING CAN OCCUR AT THE SECOND LEVEL, SPOOFING CAN ALSO OCCUR AT THE THIRD LEVEL OR THE FOURTH LEVEL OR THE FIFTH LEVEL IF THE ONLY PROTECTION AGAINST SPOOFING IS THE SPECIFIC REGISTRATION BEHAVIOR OF THE TOP-LEVEL DOMAIN REGISTRIES.
SO ONE QUESTION IS WHAT DO YOU SEE US DOING ABOUT THAT OR DO WE HAVE TO LIVE WITH THIRD AND FOURTH LEVEL SPOOFING?
THE OTHER QUESTION IS THAT SOME OF THOSE RECOMMENDATIONS SEEM TO IMPLY VERSIONING IN THE DNS. AND AS FAR AS I KNOW, WE DON'T HAVE ANY WAY -- ANY GOOD WAY TO GET A VERSION NUMBER INTO A DNS RESOURCE RECORD UNLESS WE CHANGE THE PREFIX.
IF WE CHANGE THE PREFIX IN ORDER TO DO A VERSION NUMBER, THEN THE REGISTRIES AND ALL OF THE APPLICATIONS WRITERS ARE GOING TO HAVE TO SUPPORT BOTH PREFIXES AND TWO DIFFERENT SETS OF RULES, PROBABLY FOREVER.
AND SOME OF US HAVE FELT THAT THAT PROBABLY IS NOT AN OPTIMAL SOLUTION, BUT AGAIN, I'D BE INTERESTED IN COMMENTS ON IT.
>>MICHEL SUIGNARD: SO ON THE SECOND QUESTION -- I'M TRYING TO REMEMBER THE FIRST QUESTION.
YEAH, WE SEE -- I DON'T REALLY HAVE THE ANSWERS TO A LOT OF THOSE, BECAUSE, YOU KNOW, SHOULD WE REVISE -- THAT'S A QUESTION I PUT TO THE PANEL. WHAT SHOULD WE DO WITH THE CURRENT VERSION OF IDN? UNTIL IDN IS CHANGED, YOU REALLY WRITE -- THE BROWSER COMMUNITY CANNOT ENFORCE A RESTRICTION. WE CAN BE ALERTING THAT YOU'RE GOING BASICALLY ON SAFE TERRITORY, BUT WE CANNOT BE PREVENT PEOPLE FROM USING THE WALL IDN. THERE IS EVEN A LIABILITY ISSUE ON THIS THING. IF SOME PEOPLE ARE CREATING -- OR SOME CCTLDS ARE CREATING CONTENT THAT IS LEGIT BY IDN TERMS, WE CANNOT BLOCK IT.
THE CCTLD HAVE BEEN RELATIVELY SHY IN CREATING CONTENT. IN FACT, PART OF THE REPORT WAS GOING AFTER ALL THE CCTLD. ON GTLD WE COULD FIND ON ALREADY CREATED CONTENT. WE HAVE TODAY A SUPERSET OF ALL OF THAT. SO IF WE CHANGE IT SOON ENOUGH WE MAY NOT HAVE THE ISSUE BECAUSE IT WOULD BE KIND OF DE FACTO LIMITED TO WHAT WE HAVE TODAY.
THE MORE THAT WE WAIT TO DO SOMETHING ABOUT IT, THE MORE IT WOULD BE DIFFICULT TO CHANGE ANYTHING. BECAUSE TODAY YOU COULD ARGUE THAT YOU DON'T HAVE TO CHANGE A PREFIX BECAUSE ANYTHING CREATED WITH A PREFIX AS OF TODAY WOULD, IN FACT, BE STILL LEGIT TOMORROW WITH THE RESTRICTION. BUT THE MORE WE WAIT, THE LESS THAT WOULD BE TRUE.
I'M TRYING TO REMEMBER THE OTHER QUESTION. I THINK I'M NOT ANSWERING.
>>RAM MOHAN: MICHEL, I WANTED TO ADD TO YOUR COMMENT. I THINK IT'S TIME TO STOP PUSSY FOOTING AROUND THIS TOPIC AND I'M PROPOSING WE GET TO A PROCESS WHERE WE GET A BCP, A BEST CURRENT PRACTICE, TRACK GOING AND GET THIS CODIFIED.
IT'S BETTER FOR US TO UNDERSTAND THE ISSUES AND GET TO MUSTS AND SHOULDS AND CANNOTS NOW THAN JUST CREATE GUIDELINES THAT ARE MUSHY AND THAT ALLOW DIFFERENT -- DIFFERENT COMPONENTS OF THE INDUSTRY TO JUST INTERPRET THEM ANY WAY THEY WANT.
AND THAT'S, TO SOME EXTENT, WHAT'S HAPPENING RIGHT NOW.
SO I ENDORSE WHAT YOU'RE SAYING. AND SPECIFICALLY, I THINK A BCP TRACK IS A GOOD TRACK TO MOVE THIS FORWARD.
(APPLAUSE.).
>>CARY KARP: MERELY NOTING THAT, AUDIENCE RESPONSE TO THAT SUGGESTION I THINK IS ACTUALLY QUITE ENCOURAGING, BECAUSE THE EXTENT TO WHICH THERE CAN BE PRESCRIPTIVE DOCUMENTATION AND THE EXTENT THERE CAN BE ALGORITHMIC ENFORCEMENT OF THAT STILL DOESN'T ANY ANYWHERE NEAR THE PURPOSE OF -- HOW SHALL WE SAY IT? AN INSPIRATIONAL DOCUMENT WITH TEETH IN IT CAN DO.
SO WE ARE STILL IMMEDIATELY TALKING ABOUT THE NEXT VERSION OF THE ICANN GUIDELINES. BUT SEEING THAT AS ALSO BEING THE FIRST STEP ON A TRACK TOWARD THE KIND OF DOCUMENT THAT RAM IS REFERRING TO I THINK MIGHT BE A VERY LAUDABLE THING AND WE WILL TAKE THE AUDIENCE APPLAUSE AT THE SUGGESTION AS AFFIRMATIVE ABOUT THE COMMUNITY'S PERCEIVED VALUE OF THAT.
BEFORE ALLOWING THE NEXT QUESTION ON THE FLOOR, I'D ALSO LIKE TO HERE THE FINAL OF OUR PANELISTS COMMENT ON THE WAY ALL OF THIS SEEMS FROM THE LINGUIST'S POINT OF VIEW.
>>MICHAEL EVERSON: FROM THE LINGUIST'S POINT OF VIEW WHAT WE HAVE IS A LITTLE BIT OF A DISCONNECT BETWEEN THAT AND THE COMMERCIAL POINT OF VIEW.
I'M NOT CRITICIZING WHAT YOU HAVE SAID TODAY, ACTUALLY.
WHAT WE NEED TO DO IS WE NEED TO HAVE COPPER FASTENED, REAL HONEST TO GOD RULES THAT PEOPLE ARE NOT ALLOWED TO BREAK. ONE OF THOSE RULES NEEDS TO BE THAT BETWEEN THE TWO DOTS IN ANY PARTICULAR ELEMENT, YOU MAY NOT MIX SCRIPTS. A COMPLETE BAN ON THAT, UNLESS THERE ARE VERY, VERY, VERY SPECIFIC REASONS, GROUNDS FOR DOING OTHERWISE.
FOR INSTANCE, IT IS KNOWN THAT KOREAN MUST MIX HANGUL AND HAN SO THAT WOULD BE A CULTURALLY SPECIFICALLY APPROPRIATE AND NONDANGEROUS IN TERMS OF SECURITY EXCEPTION TO A BAN ON SCRIPT MIXING.
WE ALSO NEED TO MAKE SURE THAT WE DON'T INCLUDE ANYTHING THAT IS FRIVOLOUS. ONE OF THE FAMOUS EXAMPLES ABOUT THIS IS TOYS YA US WHERE IT IS SUGGESTED THAT THE CYRILLIC LETTER YA CAN BE USED IN ANY MEANINGFUL FASHION IN THE TOYS HYPHEN R HYPHEN US BRAND NAME.
NOW, NOT A SINGLE ONE OF THE CUSTOMERS OF TOYS R US PROBABLY KNOWS WHAT THE CYRILLIC LETTER YA IS AND IN FACT IT'S A VIOLATION OF THE CHARACTER GLYPH TO SUGGESTION THAT YA IS A VARIANT OF R. WHAT WE HAVE HERE IS DECORATION, AND THESE TYPES OF LABELS SHALL NOT BE PERMITTED, IF WE'RE SMART, TO ALLOW ANY KIND OF CORPORATE DECORATION. THAT SHOULD BE OUT OF SCOPE.
WE NEED TO MAKE SURE THAT WHAT WE'RE DOING IS ACTUALLY LINGUISTICALLY SOUND STUFF THAT PEOPLE ARE ACTUALLY GOING FOR.
THANK YOU.
>>PETER KOCH: THANK YOU.
MY NAME IS PETER KOCH.
I DID NOT WANT TO LET THE SUMMARY OF THE PREVIOUS POINT GET AWAY UNCOMMENTED, SO I WOULD LIKE TO ADD SOME DISCOURAGEMENT TO THE SUGGESTION OF DOS AND DON'TS AND MUSTS.
FOR THE CCTLDS, I GUESS, SETTING THE REGISTRATION POLICY IS A MATTER OF THAT CCTLD AND MAYBE ITS CONSTITUENCY.
IN ADDITION TO THAT, BCPS, THE ACRONYM OR THE ABBREVIATION IS COMPRISED OF THREE LETTERS, WHERE THE "C" STANDS FOR CURRENT PRACTICE.
AND RECOMMENDING THINGS THAT ARE CURRENTLY PROBABLY NOT PRACTICED MAY NOT BE A GOOD THING.
SO THE RECOMMENDATION IN THE TECHNICAL REPORT 36 OF KIND OF BUNDLING OR BLOCKING, WE, IN PARTICULAR, HAVE SEVERE CONCERNS ABOUT THAT.
AND AS MY ESTEEMED COLLEAGUE JAAP JUST REMINDED ME, COMING TO YOUR POINT, JUDGING THE NAMES, WE HAVE A FAIR AMOUNT OF HISTORY OF KEEPING OUT OF THAT BUSINESS.
SO JUDGING THE LEGITIMACY OF A NAME, WHETHER THE CYRILLIC "YA" IS A TRADEMARK COMPONENT OR THINGS LIKE THAT, AGAIN, I WOULD RAISE CONCERNS TO THAT REGARD.
THANK YOU.
>>CARY KARP: MAY I ASK YOU A QUESTION BEFORE YOU LEAVE.
THE JUSTIFICATION -- THERE ARE A CERTAIN AMOUNT OF -- THERE'S A CERTAIN AMOUNT OF REGULATION THAT IS NECESSARY TO RENDER IDN MAXIMALLY USEFUL IN EVERY CULTURAL AND COMMERCIAL REGARD THAT IT MIGHT SERVE AND MINIMALLY DANGEROUS IN TERMS OF THE ABILITY FOR PEOPLE TO USE IT TO HARM OTHER PEOPLE.
OKAY.
AND ALTHOUGH IT MAY BE INAPPROPRIATE TO PASS JUDGMENT ON THE APPEARANCE OF A BACKWARDS "R" WHAT IS INTENDED TO LOOK LIKE A BACKWARDS "R" USING A CYRILLIC LETTER THAT HAPPENS TO REPRESENT THAT.
WOULD YOU AGREE THAT THE RULE STATING THAT YOU DO NOT MIX DIFFERENT SCRIPTS UNLESS THERE IS VERY CLEAR JUSTIFICATION FOR IT IS A REASONABLE RULE AND THAT WHAT WE WOULD BE ARGUE SOMETHING WHAT CONSTITUTES REASONABLE JUSTIFICATION, WHERE "NOTHING" IS NOT AN ACCEPTABLE ANSWER?
>>PETER KOCH: OKAY.
SO WITH REGARD TO MIXING SCRIPTS, FOR DE, WE DO NOT HAVE THAT PROBLEM AT THE MOMENT, BECAUSE WE JUST ALLOW FOR LETTING ONE IN AND DERIVATIVES.
SO I COULD IMAGINE -- I COULD IMAGINE LEGITIMATE USE OF MIXED SCRIPTS IN LABEL, LIKE PERSONS HAVING, SAY, DOUBLE NAMES ORIGINATING FROM DIFFERENT CULTURE.
SO IF THAT IS ONE OF THOSE JUSTIFICATIONS YOU SUGGESTED AND GIVES A -- OPENS THE ROOM FOR EXEMPTION, THEN I WOULD PROBABLY SAY YES.
>>CARY KARP: ONE FINAL QUESTION.
WE ARE GTLD PEOPLE.
YOU ARE A CCTLD PERSON.
DO YOU PERCEIVE THERE TO BE TRUE BENEFIT TO THE COMMUNITY FOR US TO AGREE UPON WHAT THESE RULES MIGHT BE BEFORE WE THEN GET BACK TO THE BUSINESS OF COMPETING WITH EACH OTHER IN THEIR APPLICATION?
>>PETER KOCH: WELL, WOULD YOU MIND IF I REFRAIN FROM ANSWERING THAT QUESTION WITH "YES" OR "NO"?
THE POINT IS THAT I AM PRETTY MUCH AWARE THAT THE DISCUSSION IS AROUND GTLDS.
BUT THE TECHNICAL REPORT, THE TR-36, DOESN'T MAKE THAT DISTINCTION.
SO WHAT IS IMPORTANT, I THINK, FOR THE CCTLDS AND THEIR CONSTITUENCIES IS THAT THEY CAN DEAL WITH POLICY FOCUSED TO THEIR CONSTITUENCY.
AND THERE MIGHT BE A BENEFIT OF, WELL, TALKING AND AGREEING ON TECHNICAL STANDARDS.
THAT IS, OF COURSE, ALWAYS TRUE.
BUT I'M NOT SURE THAT IMPLYING POLICY THROUGH GTLDS INTO CCTLDS IS A GOOD IDEA.
>>CARY KARP: I'M SUGGESTING, ACTUALLY, THAT WE REFRAIN FROM ANY SUCH DISTINCTION AND DETERMINE WHAT IT IS THAT IS NECESSARY FOR IDN TO BECOME A TRULY USEFUL DEVICE TO THE INTERNET COMMUNITY IN ALL ITS MANIFESTATIONS FOR ALL ITS PURPOSES, EXCEPT THE ONES FOR EVIL.
>>RAM MOHAN: LET ME ADD TO WHAT CARY IS SAYING.
IT'S TIME FOR US TO STOP TALKING ABOUT IDNS IN A GTLD CONTEXT AND IN A CCTLD CONTEXT.
THESE ARE IDNS.
AND PEOPLE EXPECT THEM TO JUST WORK WHEN THEY'RE TYPED IN OR WHEN THEY COME UP IN A SEARCH ENGINE.
AND THEY DON'T -- I MEAN, A LOT OF PEOPLE DON'T SIT THERE AND SAY, "THIS IDN CAME UP IN A CCTLD.
THEREFORE, I EXPECT IT TO WORK DIFFERENTLY THAN WHEN IT CAME UP IN A GTLD."
>>PETER KOCH: NO, SORRY, IDNS DON'T WORK DIFFERENTLY.
IF IT'S REGISTERED AND IT'S -- THE TRANSLATION, THE ALGORITHM THAT IS APPLIED IS ALWAYS THE SAME.
>>RAM MOHAN: YOU KNOW, THAT'S NOT TRUE.
IF YOU DON'T HAVE RULES AND IF YOU HAVE APPLICATION DEVELOPERS WHO DON'T UNDERSTAND WHAT THE RULES ARE, TAKE VAGUE GUIDELINES, AND IMPLEMENT WHATEVER THEY WANT TO DO IT, I DON'T SEE MUCH ABILITY FOR THEM TO WORK CONSISTENTLY AND COHERENTLY.
>>PAT KANE: IF I MAY, FROM A REGISTRATION STANDPOINT, THAT'S WHERE THE GUIDELINES APPLICABLE.
SOME OF THE QUESTIONS I GET IS WHY CAN I DO THIS IN DE AND NOT IN COM FROM. A REGISTRATION STANDPOINT, I'D LIKE TO SEE SOME CONSTITUENCY SO THAT THE END USER, THE REGISTRANT, CAN ACTUALLY ACQUIRE A DOMAIN THEY WANT IN DE AND IN COM AND IN INFO AND IN BIZ AND HAVE A CONSISTENT EXPERIENCE.
THE END USERS ARE GOING TO SEE RESOLUTION WORK THE SAME ALL THE TIME.
>>CARY KARP: I HAVE A SPEAKER'S LIST NOW.
THIS IS WHAT I WAS HOPING I WOULD NEED TO DO.
MICHAEL, CLAUDIO, AND THEN YOU AGAIN.
>>MICHAEL EVERSON: WHAT WE HAVE IS A SECURITY PROBLEM.
AND WHAT NEEDS TO BE MADE CLEAR TO THE REGISTRARS AND REGISTRANTS AND EVERYBODY IS THAT GOOD NEIGHBORSHIP NEEDS TO HAPPEN IN ORDER FOR THE WHOLE DOMAIN OF IDN TO BE USEFUL AND SAFE.
AND THAT MEANS THAT SOME DIFFICULT DECISIONS NEED TO BE TAKEN.
AND IF WE HAVE A RULE THAT YOU CANNOT MIX SCRIPTS WITHIN A PARTICULAR ELEMENT AND SOMEBODY COMES IN WITH A HALF-GREEK, HALF-LATIN NAME, THEN IN ORDER FOR THEM TO GET WHAT THEY WANT, THEY WILL HAVE TO PUT THEM BETWEEN TWO DOTS.
THERE WILL BE SOME COMPLAINT FROM SOME OF THE CUSTOMERS, AND THE RESPONSE MADE TO THE CUSTOMERS IS GOING TO HAVE TO BE, "WELL, YOU CAN'T HAVE WHAT YOU WANT BECAUSE IT'S A SECURITY ISSUE, AND THAT'S THE WAY THE RULES ARE."
I UNDERSTAND THAT THE COUNTRY CODE -- THE NATIONAL ONES -- WILL HAVE THEIR OWN REQUIREMENTS AND MAY DIFFER, PERHAPS, IN SOME RESPECTS FROM THE KINDS OF RECOMMENDATIONS THAT WE'RE HAVING.
I DON'T THINK THAT THAT'S NECESSARILY VERY WISE, BECAUSE IF A DOMAIN IN ANY RESPECT IS CONSIDERED UNSECURE OR UNSAFE, THEN THAT'S JUST GOING TO LEAD TO THE DETRIMENT OF THAT REGISTRATION AUTHORITY.
WE CAN'T IMPOSE ANYTHING ON THE REST OF THE COMMUNITY, BUT WE CAN MAKE RECOMMENDATIONS WHICH, IF ADOPTED BY THE GENERIC TLDS, SHOULD PROBABLY BE ADOPTED BY THE OTHER ONES.
I DON'T -- I THINK THAT -- THERE'S COMPETITION AND THERE'S TURF, BUT AT THE END OF THE DAY, IT'S THE WHOLE COMMUNITY WHICH NEEDS TO DO SOME AGREEMENT.
I'M NOT SURE IF THAT'S COHERENT TO ANSWER YOUR QUESTION.
FOR INSTANCE, TOYS R US, AND THEY ARE OUR FAVORITE THING.
WHEN THEY'RE SAYING TOYS YA US, THEY'RE ACTUALLY SPOOFING THEMSELVES.
THAT'S NOT SERIOUS.
AND WE WANT TO AVOID THAT KIND OF SPOOFING.
AND IF YOU ALLOW IT IN ONE INSTANCE, THEN YOU'RE OPENING IT FOR EVERYTHING.
>>CLAUDIO MENEZES: WELL, YOU KNOW THAT THERE ARE AROUND 6,000 LANGUAGES, OUT OF WHICH AROUND 12 OF THESE LANGUAGES ARE WELL REPRESENTED IN CYBERSPACE.
IN UNESCO, WE UNDERSTAND THAT A DOMAIN NAME IS A SORT OF ENTRY POINT FOR THE KNOWLEDGE SOCIETY.
AND WE ARE READY TO SEE WHAT WILL BE THE IMPACT OF OFFERING INTERNATIONALIZED DOMAIN NAMES UNDER THIS PERSPECTIVE.
BECAUSE YOU KNOW THAT ICANN IS VERY TECHNICAL BODY.
WE HAVE BEEN ENCOUNTERED WITH OTHER SORT OF CONSTITUENCIES, LIKE LIBRARIANS, LIKE ARCHIVES.
AND SOMETIMES WE FEEL THAT SORT OF -- MAYBE I CAN TELL A SORT OF DIDACTIC APPROACH TO FACILITATE FURTHER ALLIANCES.
BECAUSE WHEN WE -- AS I MENTIONED IN MY TALK, WHEN WE TALK OF LOCAL CONTENT, EVERYBODY SAYS "YES."
BUT WHEN WE HAVE TO ADDRESS LOCAL CONTENT AS AN ELEMENT OF A CHAIN WHICH REQUIRES INVESTMENTS, PEOPLE SOMETIMES, "OH, MAYBE WE ARE NOT ABLE TO DO THIS KIND OF INVESTMENT."
MICHAEL EVERSON KNOWS VERY WELL HOW IT GOES ABOUT.
I UNDERSTAND THAT IF WE WISH TO HAVE NATIONAL POLICIES THAT WILL HELP IN BUILDING UP KNOWLEDGE SOCIETIES, I UNDERSTAND THAT, FOR EXAMPLE, FROM UNESCO POINT OF VIEW, WE HAVE TO BE ABLE TO GET SOME SUPPORT FROM NATIONAL GOVERNMENTS AND ALSO INTERNATIONAL ORGANIZATIONS, INCLUDING THOSE DEALING WITH INVESTMENTS, BECAUSE, FOR EXAMPLE, UNESCO CANNOT HANDLE 6,000 LANGUAGES.
IT'S TOO MUCH FOR OUR RESOURCES.
>>PETER KOCH: OKAY.
THANKS.
TWO SHORT RESPONSES.
PAT, I GUESS TLDS ARE DIVERSE, SO IT MAY HAPPEN THAT THE CONSUMER OR THE POTENTIAL REGISTRANT HAS AN EXPERIENCE THAT CAN'T BE FULFILLED IN ONE TLD OR ANOTHER.
EVEN TODAY, I MEAN, PROBABLY PEOPLE ARE GOING TO ALLOW SOMETHING IN DOT COM WHICH THEY ARE NOT ALLOWED TO DO IN DOT DE, LIKE WE DEMAND THAT NAME SERVERS ARE SET UP BEFORE THE DELEGATION IS DONE OR SOMETHING LIKE THAT.
SO THERE MAY BE DIVERSITY IN WHICH IDNS MAY OR MAY NOT BE REGISTERED.
I GUESS PEOPLE WILL HAVE TO LIVE WITH THAT.
THE OTHER POINT IS, I WOULD LIKE TO STATE THAT WE ARE NOT THE NAMES OR THE INTERNET POLICE.
AND ANYTHING THAT WOULD FORCE US OR STRONGLY SUGGEST THAT WE SOMEHOW CHECK HOW LEGITIMATE THE CHOICE OF THE NAME IS IS A PATH THAT I WOULD PREFER NOT TO HAVE TO GO DOWN TO.
>>MICHAEL EVERSON: WHAT I AM SAYING IS THAT IF I AM GOING TO REGISTER EVERSON.DE AND DE HAS A POLICY WHICH ALLOWS SCRIPT MIXING SO THAT "EVERSON" COULD BE REPRESENTED WITH CYRILLIC YA OR CYRILLIC AU, OR CYRILLIC ZA OR GREEK OMICRON, THEN I WOULD NOT TAKE MY BUSINESS TO DE BECAUSE THAT WOULD BE UNSAFE FOR ME AND MY NAME.
THAT'S WHY WE NEED TO HAVE NO MIXING IN SINGLE ELEMENTS BECAUSE IT'S DANGEROUS.
THAT'S WHY WE'RE HERE.
>>PETER KOCH: I DON'T THINK SAFE IS THE CORRECT WORD FOR THAT.
AND TO GO TO THAT PARTICULAR POINT, IT WON'T HAPPEN, BECAUSE, ACTUALLY, AT THE MOMENT, WE ONLY ALLOW LATIN CHARACTERS.
SO -- BUT I GUESS I UNDERSTAND WHAT YOU ARE TRYING TO SAY.
THANKS.
>>CARY KARP: WE HAVE -- PLEASE IDENTIFY YOURSELF, SIR, FOR THE --
>>BRUCE TONKIN: BRUCE TONKIN FROM MELBOURNE I.T.
I'M A DOMAIN NAME REGISTRAR.
I TEND TO -- I THINK MY AGREEMENT IS MORE STRONGLY TOWARDS WHAT RAM HAS SAID AND MICHEL HAS SAID, AND ALSO THAT WE SHOULDN'T BE MIXING SCRIPTS.
AND I THINK ONE WAY THAT WE SHOULD BE THINKING ABOUT IT IS THERE'S A DIFFERENCE BETWEEN A DOMAIN NAME AND WHAT A DOMAIN NAME IS FOR, JUST AS THERE'S AN I.P. ADDRESS.
ONE OF THE BENEFITS OF I.P. ADDRESSES IS THEY'VE BASICALLY GOT NUMBERS IN THEM, WE'VE GOT 1 TO 9 AND A 0.
I CAN TYPE THAT IN THE KEYBOARD.
IT DOESN'T REQUIRE ANYTHING SPECIAL AND THERE'S NOT A LOT OF CONFUSION WITH THOSE NUMBERS.
WHEN I LOOK AT DOMAIN NAMES AS THEY HAVE BEEN TRADITIONALLY WITH DOT COM, IT'S A PRETTY CONSTRAINED SET OF CHARACTERS.
THERE'S -- BECAUSE OF THAT CONSTRAINT, MOST KEYBOARDS, WHETHER YOU'RE IN CHINA OR WHETHER YOU'RE IN AUSTRALIA, THERE'S GENERALLY A WAY TO BE ABLE TO ENTER THOSE THINGS IN ON THE KEYBOARD THAT THE AVERAGE USER IS ABLE TO DO.
SO THINK ABOUT WHAT YOU'RE ACTUALLY DOING IN TYPING THE CHARACTER.
THEN IF WE START MOVING INTO GRAPHICS, THAT'S A VERY DIFFERENT THING.
IF I LOOK AT THE LOGO THERE AND THE LOGO UP THERE FOR LUXEMBOURG.2005, I DON'T EXPECT TO BE ABLE TO GET THAT RED SQUARE WITH THE WHITE "LU" WHEN I'M TYPING IN THE DOMAIN NAME.
I SEE THAT.
THAT'S A GRAPHIC ICON.
AND IF YOU WANT TO HAVE "TOYS" WITH A GRAPHIC "R US," THAT SHOULD JUST BE A GRAPHIC IMAGE IN THE APPLICATION AND YOU'RE CLICKING ON THE GRAPHIC IMAGE, WHICH, THEN, IN TURN MIGHT TRANSLATE TO A DOMAIN NAME, JUST AS A DOMAIN NAME TRANSLATES TO AN I.P. ADDRESS.
SO I WOULD URGE THAT WE KEEP IT SIMPLE.
BECAUSE ONCE WE START MIXING SCRIPTS, WE'RE GETTING INTO A LOT MORE COMPLEXITY AND I DON'T THINK WE'RE GETTING A LOT MORE ADDITIONAL BENEFIT FOR THAT ADDITIONAL COMPLEXITY.
SO LET'S TRY AND KEEP IT SIMPLE.
I THINK IT'S A SENSIBLE STEP TO MOVE FROM JUST THE CONSTRAINTS THAT WE HAVE FOR DOMAIN NAMES TODAY TO, SAY, OKAY, WELL, LET'S START, ACTUALLY, CONSTRAINING SOME SCRIPTS AND MAKING SURE WE HAVE JUST ONE -- ONE SCRIPT AT A TIME, NOT MIX IT IN THE DOMAIN NAME.
BECAUSE HOW WOULD YOU ENTER THAT?
IF I SAW "TOYS R US," I DON'T KNOW HOW TO ENTER THAT CHARACTER ON MY ENGLISH KEYBOARD IN CYRILLIC.
IF I WAS SOMEBODY THAT WAS USING -- IF I WAS WRITING ALL MY E-MAILS AND EVERYTHING IN CYRILLIC AND I HAVE ALL MY KEYBOARD SET UP THAT WAY, SURE.
BUT THEN I'D WANT THE WHOLE DOMAIN NAME TO BE IN CYRILLIC.
THINK ABOUT IT IN TERMS OF WHAT YOU CAN TYPE IN ON THE KEYBOARD JUST -- I'M JUST BASICALLY SAYING KEEP IT SIMPLE.
>>CARY KARP: CHRISTOPHER.
>>CHRISTOPHER WILKINSON: GOOD MORNING.
THIS IS VERY IMPORTANT AND INTERESTING EVENT.
I'M CHRISTOPHER WILKINSON FROM THE GAC SECRETARIAT.
THE IDN WORKING GROUP IN THE GOVERNMENTAL ADVISORY COMMITTEE HAS TAKEN AN INITIATIVE TO PLAY A MUCH MORE ACTIVE ROLE IN THIS WHOLE FIELD.
THE WORKING GROUP IS LED BY INDIA, AND THERE ARE MANY OTHER GOVERNMENTS INVOLVED.
FOR CONVENIENCE, I'VE BEEN ASKED JUST TO GIVE A SHORT PRESENTATION AS TO WHAT THE WORKING GROUP IS DOING AND WHERE WE'RE COMING FROM.
THE MAIN -- JUST IN TERMS OF PARAGRAPH HEADINGS, THE MAIN AREAS OF WORK CONCERN IDN TECHNOLOGY, WHERE WE THINK IT'S VERY IMPORTANT THAT THE GOVERNMENTS CONCERNED UNDERSTAND IN SOME DETAIL THE TECHNOLOGY THAT IETF AND THE OPERATORS ARE DEVELOPING.
SECONDLY, IDN STANDARDIZATION, WHERE THE QUESTION OF WHO IS DOING WHAT AND WHAT THE OUTPUT IS AND HOW IT CAN BE APPLIED IN THE IDN AREA IN VARIOUS LANGUAGES AND COUNTRIES.
THIRDLY, A PARTICULAR EMPHASIS ON CROSS-BORDER LANGUAGE COMMUNITIES.
I BELIEVE THAT MOST GAC MEMBERS, AND CERTAINLY MEMBERS OF THE WORKING GROUP, RECOGNIZE THAT NEARLY ALL IDN LANGUAGE GROUPS INVOLVE MORE THAN ONE GOVERNMENT.
INDEED, THERE ARE VERY FEW WHO DON'T.
AND THIS IS AN AREA WHERE THE GOVERNMENTAL ADVISORY COMMITTEE MEMBERS, THROUGH THEIR ASSOCIATION IN SUCH A WIDE RANGE OF OTHER INTERNET MANAGEMENT POLICIES, HAVE AN ADVANTAGE TO BE ABLE TO FACILITATE CROSS-BORDER COOPERATION.
THERE IS CONSIDERABLE INTEREST IN TECHNICAL IMPLEMENTATION, WHICH HAS ALREADY BEEN REFERRED TO IN ONE OR TWO OF THE COMMENTS THAT HAVE BEEN MADE AND QUESTIONS THAT HAVE BEEN MADE ALREADY TODAY.
WE SEEK TO DEFINE, AND TO SOME EXTENT, CIRCUMSCRIBE, THE ROLE -- THE SPECIFIC ROLE OF GOVERNMENTS, BUT RECOGNIZE THAT INTELLECTUAL PROPERTY PROTECTION, THE RESPECTIVE ROLES OF GTLD AND CCTLD REGISTRIES, ICANN, AND OTHERS ARE SIGNIFICANT, MUST BE UNDERSTOOD, WIDELY UNDERSTOOD.
PARTICULARLY, AS HAS ALREADY BEEN INDICATED IN YOUR CONVERSATIONS, THERE ARE SIGNIFICANT RISKS OF ABUSE IN CERTAIN ASPECTS OF IDN.
AND, FINALLY, DISPUTE RESOLUTION PROCEDURE.
THIS WORK WILL BE CONDUCTED IN AN OPEN ENVIRONMENT.
WE HAD A WORKING GROUP SESSION WHICH WAS VERY WELL ATTENDED BY 26 GOVERNMENTS AND ABOUT ANOTHER 20 OR 30 PRIVATE-SECTOR PARTICIPANTS, ON SATURDAY AFTERNOON.
RAM MOHAN AND CARY KARP WERE THERE, SO THEY'RE FAMILIAR WITH THIS TO DATE.
AND I EXPECT SUBJECT TO THE AGREEMENT OF THE CHAIR AND OTHER MEMBERS OF THE WORKING GROUP, I EXPECT THAT THIS OPEN AND CONSULTATIVE PROCESS WILL CONTINUE DURING THE MONTHS TO COME, WITH A VIEW TO FURTHER PROGRESS IN THIS AREA IN THE VANCOUVER MEETING.
THANK YOU.
>>CARY KARP: I CERTAINLY HAVE NO MORE HIGHER HOPE THAT THAT IS EXACTLY THE PATH UPON WHICH WE ARE NOW PRODUCTIVELY PROCEEDING.
ANYONE ELSE?
>>ANDRZEJ BARTOSIEWICZ: I HAVE ONE MORE QUESTION TO MICROSOFT.
WHAT DO YOU THINK ABOUT THE RECENT IDEA OF FIREFOX DEVELOPERS TO PRESENT TO DISPLAY THE IDNS IN A DIFFERENT WAY DEPENDING ON FROM WHICH CCTLD THIS DOMAIN NAME IS COMING.
SO, FOR EXAMPLE, DOMAIN NAME FROM DOT DE IS DISPLAYED IN A TOTALLY DIFFERENT WAY AS FROM DOT NU.
AND ONE IS DISPLAYED IN IDNA FORM AND THE SECOND ONE IS DISPLAYED IN XN DASH DASH FORM.
>>MICHEL SUIGNARD: IT SOUNDS LIKE DISCRIMINATION TO ME ON WHO WILL -- YOU KNOW, FIREFOX CANNOT BE SUED, SO LIABILITY IS NOT A CONCERN FOR THEM.
BUT, YOU KNOW, I AM NOT GOING TO COMMENT ON WHAT WE'RE GOING TO DO OR WHAT NOT DO UNTIL WE SEE THE PRODUCT ON THAT ASPECT.
BUT I THINK DOING THIS KIND OF DISCRIMINATION IS TYPICALLY PRETTY RISKY, IN MY OPINION.
>>PAT KANE: I DON'T LIKE IT BECAUSE DOT COM IS NOT ON THAT LIST.

(LAUGHTER.)
>>RAM MOHAN: I DON'T MIND, BECAUSE DOT INFO IS.
>>MICHAEL EVERSON: I DON'T LIKE IT BECAUSE IT'S COMPLETE DISSERVICE TO THE END USER, OKAY, LIKE A LANGUAGE GEEK LIKE ME MIGHT BE ABLE TO GO FOR IT, BUT DON'T GIVE THAT TO MY DAD, OKAY?
>>CARY KARP: OKAY.
FOLKS, WE'VE GOT A LITTLE BIT MORE THAN FIVE MINUTES.
AND WE REALLY ARE RUNNING ON A TIGHT SCHEDULE.
AND I THINK THAT I WILL BE ALLOWED, WITHOUT HAVING MY HEAD HANDED TO ME, TO LET THIS RUN OVER FIVE MINUTES.
I WOULD ACTUALLY LIKE TO GO ACROSS THE PANEL JUST FOR A FINAL FEW MOMENTS.
WE NEED, ACTUALLY, TO MAKE SOMETHING THAT IS A TANGIBLE STATEMENT OF WHERE WE INTEND TO GO NEXT, NOT JUST SOMETHING THAT SOUNDS NICE.
SO, MILTON.
>>MILTON MUELLER: MAKE THIS A QUICK QUESTION.
THE ALAC AND THE NCUC WAS TRYING TO COME UP WITH A STATEMENT ON A POLICY ISSUE RELATED TO IDNS REGARDING WHETHER THE ADDITION OF TOP-LEVEL DOMAINS AS IDNS SHOULD BE UNIFIED AND INTEGRATED WITH THE DEFINING A PROCESS FOR ASCII TOP-LEVEL DOMAINS OR WHETHER THEY SHOULD BE SEPARATED.
WE DEVELOPED A POSITION SAYING WE THINK THEY SHOULD BE SEPARATED FOR A VARIETY OF REASONS.
THE WE WANT TO KNOW IF ANY OF YOU REACT IN SHOCK AND HORROR TO THAT.
>>CARY KARP: THAT'S THE TOPIC OF THE SESSION THAT WILL BE STARTING SEVEN MINUTES FROM NOW.
>>MILTON MUELLER: OH, OKAY.
>>CARY KARP: THOSE OF WHO YOU ARE GOING TO BE HERE AND ARE THINKING ABOUT THIS, REMEMBER MILTON'S QUESTION, AND WE WILL LEAVE IT TO THE NEXT PANEL TO ANSWER IT.
ANYTHING ELSE SPECIFIC TO WHAT WE ARE -- THERE ARE VARIOUS DEGREES OF FORMALITY IN WHAT WE ARE TALKING ABOUT.
BUT THIS IS NOT A GTLD ISSUE; THIS IS NOT EVEN A GTLD SESSION.
THIS IS A SESSION ABOUT IDN.
IDN IS RELEVANT TO EVERY REGISTRY ON EVERY LEVEL.
THE IDN GUIDELINES WERE AN INITIAL STATEMENT OF WHAT THE -- THE CONTRACTED ENTITIES OR WHOEVER -- HOWEVER YOU WANT TO DEFINE THEM WITH ICANN, NEEDED TO ADHERE TO AT THE OUTSET.
WE ARE NOW REVISING THEM.
WE WOULD LIKE TO SEE THEM REVISED IN A MANNER THAT IN AND OF ITSELF IS A GENUINELY INFORMATIVE STATEMENT OF WHAT NEEDS TO BE KNOWN AND PROVIDES COMPELLING IMPETUS TO EVERYBODY TO DO THIS BOTH RIGHT AND TOGETHER. AND WITH SOME TREPIDATION, I NOW RECOGNIZE THE NEXT SPEAKER.
>>KHALED FATTAL: THANK YOU, CARY.
MY NAME IS KHALED FATTAL.
I'M THE CHAIRMAN OF MINC, MULTILINGUAL INTERNET NAMES CONSORTIUM.
ACTUALLY, I DON'T HAVE A QUESTION.
I HAVE A GENERAL COMMENT.
I WOULD LIKE TO CONGRATULATE ICANN FOR AGAIN HOSTING THIS IDN WORKSHOP, AND AT LEAST THIS IS AN OPPORTUNITY FOR THOSE WHO ARE STILL LEARNING ABOUT WHAT IS REQUIRED TO RECOGNIZE THE NEEDED STEPS FORWARD.
SO I THINK WE CAN ALL TREAT THIS AS AN EDUCATIONAL PROCESS AND THERE ARE DIFFERENT PEOPLE WHO ARE AT DIFFERENT LEVELS OF THE EDUCATIONAL PROCESS.
SO I THINK THAT'S VERY IMPORTANT.
ONE SECOND POINT IS THE -- AND I'VE HIGHLIGHTED THIS IN THE PAST -- THE FOCUS IS ALWAYS ON IDN, IDN.
AND I THINK PERHAPS WE SHOULD TAKE A STEP BACK TO FOCUS ON WHAT ARE THE OTHER ELEMENTS THAT WOULD MAKE IDN FUNCTIONAL AND USABLE FOR THE CONSUMER OR THE CITIZEN IN THE LOCAL COMMUNITY WHO DOES NOT SPEAK -- I USE THIS JOKINGLY -- WHO DOES NOT SPEAK ASCII OR ENGLISH.
AND THIS IS QUITE RELEVANT.
BECAUSE AT THE END OF THE DAY, IT'S ALREADY BEEN RECOGNIZED THAT MULTILINGUALIZATION OF THE INTERNET IS SOMETHING THAT IS TO HELP HUMANITY MOVE FORWARD.
IT'S BEYOND JUST, YOU KNOW, SELLING A DOMAIN NAME OR ACTUALLY SELLING A PRODUCT.
SO WITH THAT IN MIND, I THINK WE PERHAPS SHOULD -- NO ORGANIZATION KNOWS EVERYTHING.
AND PERHAPS WE DON'T NEED TO REINVENT THE WHEEL OR THE HOT WATER.
WE CAN WORK TOGETHER TO LEARN FROM OTHER PEOPLE'S EXPERIENCES IN THE LOCAL COMMUNITY, ACROSS INSTITUTIONS.
AND THE KEY IS COLLABORATION.
SO, AGAIN, CONGRATULATIONS FOR SETTING UP THE WORKSHOP.
GOOD LUCK.
>>CARY KARP: THANK YOU.
FIVE MINUTES LEFT.
FIVE PANELISTS.
ONE MINUTE EACH, TO THE SUBJECT OF COLLABORATION TOWARD THE END OF MAKING THIS WORK, OKAY?
MICHEL, AND THEN --
>>MICHEL SUIGNARD: FIRST, I HEARD A CONCERN ABOUT TR 36. I WOULD LIKE TO GET THEM IN A FORM THAT WE CAN ACT ON THEM SO WE CAN GET ACTIVE FEEDBACK ON IT. TR 36 IS SEEN BY A LOT OF VENDORS WILLING TO IMPLEMENT IDN AS THE WAY THEY'RE GOING TO BE TOGETHER TO DISCUSS THE ISSUE. SO IF YOU HAVE CONCERN ABOUT THAT TR, PLEASE LET THE COMMUNITY AND THE CONSORTIUM KNOW ABOUT IT ON THE TERMS WE CAN DO SOMETHING ABOUT IT.
I REALLY WANT THERE TO BE A CHANGE ON THE GUIDELINES THAT, AGAIN, MAKE IDN IMPLEMENTABLE. TODAY I'M AFRAID THAT AS IT IS, IT'S IMPLEMENTABLE, BUT THERE ARE SO MANY SAFETY HOLE ON THIS THING THAT THERE ARE GOING TO BE MAJOR CONCERNS FROM MOST OF THE VENDORS. SO UNLESS THIS COMMUNITY ACT ON IT, IDN IS GOING TO BE A VERY TOUGH SELL IN SOME SITUATION.
I MEAN, ESPECIALLY IN GTLD CONTEXT, AND WE DON'T LIKE THAT.
>>PAT KANE: FIRST OF ALL, I DON'T THINK IDNA IS BROKEN. I THINK THE PROTOCOLS IN THE RFCS WORK AND WE SEE THEY RESOLVE RIGHT AND RESOLVE WELL.
THE LOCAL EXPERIENCE IS NOT COMPLETE. WE DON'T HAVE A HOST NAME. WE DON'T HAVE MAIL, DON'T HAVE A TLD, WE DON'T HAVE THE IRI TO THE RIGHT BEYOND THE TLD AS WELL. SO THERE IS STILL WORK TO BE DONE IN TERMS OF WHAT THAT LOCALIZED EXPERIENCE IS.
I THINK WE'RE WISER AND SMARTER TODAY. WE HAVE A LOT OF EXPERIENCE IN THE REGISTRATION AND RESOLUTION OF IDN'S ACROSS G'S AND CC'S. AND WE OUGHT TO LEVERAGE THAT AND MAKE THE GUIDELINES WORK FOR A CONSISTENT EXPERIENCE FOR REGISTRANTS SO THEY ARE NOT CONFUSED AND THAT THEY ARE SECURE.
>>CLAUDIO MENEZES: WELL, OUR PERSPECTIVE FOR IDN IS HOW COULD UNESCO CONTRIBUTE TO FACILITATE THE ACCESS TO KNOWLEDGE, BECAUSE WE HAVE NO INTENTION TO DUPLICATE. AND WE ARE READY TO DISCUSS WITH OTHER COMMUNITIES, LIKE MUSEUMS, LIKE LIBRARIANS, LIKE ARCHIVES, AND OTHER COMMUNITIES THAT ARE ALSO INTERESTED.
BUT FROM A DIFFERENT PERSPECTIVE, BECAUSE TECHNICALLY SPEAKING, YOU UNDERSTAND THAT ICANN HAS THE COMPETENCE AND THE FRAMEWORK TO DISCUSS.
SO OUR PERSPECTIVE IN UNESCO IS WHICH ARE -- WHAT CAN BE OUR CONTRIBUTION, NOT TO IDN ITSELF BUT HOW TO MAKE IDN AS AN ENTRY POINT TO WHAT WE CALL A KNOWLEDGE SOCIETY.
>>CARY KARP: RAM.
>>RAM MOHAN: THANK YOU. CLEARLY THE CURRENT IDN GUIDELINES ARE IN NEED OF REVISION AND WE'RE KICKING OFF A PROCESS. I ENDORSE THE SHIFT FROM LANGUAGE TO SCRIPTS. IT'S TIME FOR US TO SOLIDIFY THE FOUNDATIONS FOR IDNS. I COMMEND A STANDARDIZATION TRACK FOR IDNS SO THAT WE DO NOT END UP WITH ONLY YET MORE GUIDELINES.
WITHOUT STANDARDS, SOFTWARE AND APPLICATIONS BECOME IMPOSSIBLE TO IMPLEMENT CONSISTENTLY, AND AS CLAUDIO HAD MENTIONED AT THE VERY START, IF YOU DO NOT HAVE, IN THAT CONTINUUM -- TO GET CONTENT AND TO GET IDNS ACTUALLY WORKING, YOU NEED THE STANDARDIZATION.
UNESCO I THINK HAS A GREAT OPPORTUNITY TO HELP ENHANCE A MULTICULTURAL FACILITATION, PERHAPS ALONG WITH THE GAC AS WELL. THEY DO KNOW MANY LANGUAGE AUTHORITIES AND CAN POINT IN THE RIGHT DIRECTION. AND TO ICANN I RECOMMEND THAT THEY CONTINUE TO FUND IDN APPROPRIATELY IN THEIR BUDGETS BECAUSE THIS IS A TOPIC THAT ISN'T GOING TO GO AWAY, AND NONE OF US IN THE COMMUNITY CAN AFFORD TO HIDE OUR HEADS IN THE SAND.
>>CARY KARP: AND MICHAEL.
>>MICHAEL EVERSON: I THINK -- IT'S ABOUT COLLABORATION. I THINK IT'S REALLY REMARKABLE THAT -- CLAUDIO WASN'T HERE WHEN WE MET YESTERDAY, BUT THE FIVE OF US -- THE FOUR OF US CAME TOGETHER AND WE ALL HAVE VERY DIFFERENT VIEWPOINTS AND WE ALL SEEM TO HAVE VERY CLEARLY CONVERGED ON A WAY THAT WE BELIEVE, FROM ALL OF OUR VERY DIFFERENT POSITIONS, MAKES A LOT OF SENSE. AND IT HAS TO DO WITH DISCIPLINE IN TERMS OF WHAT CAN BE PERMITTED, IN TERMS OF THE SIZE OF THE CHARACTER SET PERMITTED AND IN TERMS OF THE WAY THAT VARIOUS SUBSETS OF THAT CHARACTER SET CAN BE USED.
THE UNICODE CONSORTIUMS DRAFT UNICODE TECHNICAL REPORT, WHICH IS NOT A TECHNICAL STANDARD, IT'S JUST A REPORT, IT'S GOING ON TOWARDS RECOMMENDATIONS, AND IT WILL I BELIEVE BE REVISED IN ALIGNMENT WITH THE KINDS OF CONSENSUS THAT WE UP HERE SEEM TO HAVE COME TOGETHER WITH TODAY.
AND I WOULD COMMEND ICANN AND ANYONE LISTENING TO THIS WEBCAST TO ACTUALLY TAKE THE TR 36 SERIOUSLY AND HELP MAKE IT A BETTER THING.
ALONG WITH THAT TECHNICAL EXERCISE, WORK NEEDS TO CONTINUE IN MAKING CERTAIN OF THE ACTUAL CHARACTER REQUIREMENTS OF THE LANGUAGES OF THE WORLD THAT WE WANT TO SUPPORT BECAUSE THAT'S, AT THE END OF THE DAY, WHAT THIS IS ALL ABOUT. IT'S ALLOWING PEOPLE TO HAVE ACCESS TO THE INTERNET THROUGH THEIR PREFERRED LANGUAGES AND SCRIPTS.
IN PARTICULAR, THE LANGUAGES OF AFRICA AND THE ALPHABETIC REQUIREMENTS FOR THEM IS -- SHOULD BE A PRIORITY BECAUSE THAT IS VERY, VERY MUCH UNDERREPRESENTED. ON MY OWN WEB SITE I HAVE AN ALPHABETS OF EUROPE CATALOGUE WHICH IS VERY, VERY LARGE. WE NEED THE SAME SORT OF THING FOR AFRICAN LANGUAGES. FUNDING FOR THAT WOULD CERTAINLY NEED TO BE SOUGHT.
>>CARY KARP: I WOULD LIKE TO THANK THE AUDIENCE. I WOULD CERTAINLY LIKE TO THANK THE PANELISTS. I LOOK FORWARD TO WHAT HAPPENS NEXT. DON'T GO AWAY. WHAT HAPPENS NEXT HAPPENS JUST AS SOON AS WE CAN SWITCH ON THE STAGE.
(APPLAUSE.)

>>JOHN KLENSIN: WE HAD BETTER GET STARTED. WE'VE GOT A HARD STOP, BECAUSE THEY NEED TIME TO REARRANGE THIS ROOM A LITTLE BIT BEFORE THE PUBLIC FORUM STARTS AT 1:00.
SO I'D APPRECIATE IT IF EVERYONE WHO INTENDS TO LISTEN TO THIS SESSION WOULD GET THEMSELVES SETTLED IMMEDIATELY, AND EVERYBODY WHO DOESN'T WOULD GET OUT OF THE WAY SOMEHOW.
I'M JOHN KLENSIN. WE'RE GOING TO TALK A LITTLE BIT ABOUT IDNS AT THE TOP LEVEL. THE PURPOSE OF THIS DISCUSSION IS TO GET A DISCUSSION STARTED AND GET SOME OF THE ISSUES ON THE TABLE AND BEGINNING TO GET DISCUSS SEWED THAT WE CAN FOLLOW IT UP WITH A MORE DETAILED SESSION IN VANCOUVER.
I'M GOING TO DO A LITTLE BIT OF AN INTRODUCTION TO SOME OF THE ISSUES ABOUT THE CONTEXT IN WHICH THIS PROBLEM IS EMBEDDED, OR THIS OPPORTUNITY IS EMBEDDED, AND THEN WE'RE GOING TO GET SOME BRIEF COMMENTS FROM MY COLLEAGUES, AT LEAST ONE OF WHOM I DON'T THINK HAS SHOWN UP YET, AND THEN WE'LL PROCEED TO TAKE QUESTIONS AND HAVE DISCUSSION UNTIL THE CLOCK RUNS OUT.
AS I SAY, WE HAVE GOT A VERY HARD STOP BECAUSE WE NEED TO SET UP FOR THE PUBLIC FORUM.
ONE OF THE THINGS WHICH IS IMPORTANT ABOUT THE TOP-LEVEL IDN PROBLEM OR TOP-LEVEL IDN ISSUE IS UNDERSTANDING WHAT THE PROBLEM IS AND WHAT PROBLEM WE'RE TRYING TO SOLVE.
IN THE OPINION OF SOME OF US, THE REASONS WHY THE PREVIOUS SESSION ON IDN WAS NECESSARY, WHY WE'RE NOW GOING BACK AND LOOKING AT WHETHER THE GUIDELINES AND THE STANDARDS AND THE PERMITTED CHARACTER SETS WERE CORRECT, WAS BECAUSE THERE WAS SOME UNCERTAINTY WHEN THAT PARTICULAR PIECE OF WORK WAS DONE ABOUT EXACTLY WHAT PROBLEM WAS BEING SOLVED.
AND BECAUSE OF THE UNCERTAINTIES ABOUT THE PROBLEM, THERE WERE DIFFICULTIES GETTING THINGS DESIGNED CORRECTLY AND WORKING RIGHT.
ANOTHER ISSUE, THEN, WHICH I WOULD ENCOURAGE PEOPLE NOT TO GET INVOLVED WITH TODAY OR SOON IS TRYING TO CREATE SO MUCH PRESSURE ON GETTING THE PROBLEM SOLVED IMMEDIATELY THAT A QUICK FIX IS MORE IMPORTANT THAN A CORRECT FIX.
THERE'S AN OFF THE CITED PRINCIPLE AROUND THE INTERNET AND ELSEWHERE WHICH IS THAT YOU CAN HAVE SOMETHING QUICKLY OR CHEAPLY OR DONE WELL, BUT YOU CAN'T GET MORE THAN TWO OF THE THREE. AND THAT SITUATION HAS VERY MUCH APPLIED TO IDN WORK SO FAR. IT WOULD BE GOOD IF WE COULD AVOID MAKING THAT CHOICE IN THE SAME FASHION WITH THIS TOP-LEVEL PROBLEM AS WITH THE SECOND.
AS PART OF THIS, IT'S IMPORTANT TO NOTE THE USERS TYPICALLY DON'T USE DOMAIN NAMES. THAT SEEMS LIKE A SURPRISING STATEMENT, BUT END USERS USE E-MAIL ADDRESSES AND URLS AND URIS AND MAYBE IRIS AS THEY COME ALONG MORE GENERALLY. THERE ARE A FEW PROTOCOLS THAT TELL THAT, BUT IN GENERAL, THE AVERAGE USER TO THE NETWORK DOESN'T USE IDNS.
OUR REAL GOAL IN THAT REGARD IS TO HAVE USERS ENTER AND SEE AND USE LOCATORS AND IDENTIFIERS IN LOCAL CHARACTERS, BUT ALL OF THOSE LOCATORS AND IDENTIFIERS LOCAL CHARACTERS. AND IDNS ARE PART OF THAT, BUT ONLY PART.
WHETHER THEY BE AT THE TOP LEVEL OR OTHER LEVELS OR ANYWHERE IN BETWEEN. WHEN WE LOOK AT URIS AND IRIS THEY'RE DEFINED WITH CERTAIN KINDS OF ELEMENTS. PROTOCOL NAMES ARE RESERVED SYNTAX WHICH ARE FORCED TO BE IN ASCII OR FORCED TO BE IN WHATEVER CHARACTER SET THEY'RE DEFINED IN, BUT ALL THE PROTOCOLS TODAY ARE IN ASCII. IF WE LOOK AT HTTP://WWW.ICANN.ORG/IDN, INTERNATIONALIZED DOMAIN NAMES CAN IMPACT THE WWW.IDN.ORG PART, BUT THAT'S ALL. THE PROTOCOL NAME MUST BE IN ASCII, THE SYNTAX LIMITERS MUST BE IN ASCII. URIS REQUIRE THAT EVERYTHING BE IN ASCII OR USE NUMERIC CODE ESCAPES. THAT'S ONE OF THE REASONS WHY THE IRI WORK IS IMPORTANT. BUT AGAIN, IT DOESN'T SOLVE THE SYNTAX PROBLEM.
THE ELEMENTS ARE OFTEN LOCAL SCRIPTS TODAY. THEY'RE LOCAL SCRIPTS TODAY BECAUSE OF USER INTERFACE TRANSLATION OR RECODING OR WAYS TO SIMULATE THAT TRANSLATION OR RECODING.
THE INTERNET DOESN'T CARE WHAT THE USER SEES AS LONG AS THE RIGHT THING GOES UP AND DOWN THE WIRE.
AND IDNA ITSELF IS A PRESENTATION FUNCTION. WHAT'S GOING INTO THE DNS IS STILL THE ORIGINAL SET OF HOST NAME CHARACTERS WHICH WE HAVE BEEN PERMITTING FOR YEARS.
IF WE ACTUALLY TRANSLATE URI LOCALLY A LOT OF THINGS ARE POSSIBLE. THEY MAKE THINGS LOOK THE WAY USERS EXPECT. WITH CULTURAL FORMS THAT ARE CULTURALLY CORRECT NOT JUST INTERNATIONALIZED.
.
IT'S NOT THE ONLY PROBLEM, IT MAY NOT EVEN BE THE WORST PROBLEM. BUT IF WE'RE TRANSLATING AND REMAPPING, NOT JUST ALTERING OR ADDING DNS ENTRIES, WE'VE GOT A REALLY BROAD RANGE OF ALTERNATIVES AND WE NEED TO THINK ABOUT THE TOP LEVEL OF INTERNATIONALIZATION PROBLEM IN THE CONTEXT OF ALL THOSE ALTERNATIVES, NOT JUST AS AN ISSUE OF HOW WE MAKE THINGS HAPPEN IN THE DNS.
AT THE SAME TIME WE NEED TO REMEMBER ANY USER INTERFACE ISSUE, ANY CLIENT INTERFACE ISSUE BRINGS A PROBLEM OF SOME SORT AND IDNA IS NO EXCEPTION. THE PROBLEM IS THE USERS NEED TO UNDERSTAND WHAT'S GOING ON, AND USERS OFTEN DON'T LIKE UNDERSTANDING WHAT'S GOING ON.
THERE'S NO ISSUE FOR COMMUNICATION WITHIN A COMMUNITY WHERE EVERYBODY IS USING THE SAME SOFTWARE, BUT AS SOON AS A DOMAIN NAME OR AN IRI IS SENT TO SOMEONE ELSE OR IS USING THEM NORMALLY IN ONE ENVIRONMENT USES A STRANGE MACHINE, THEN THE NAMES NEED TO BE IN A UNIVERSAL FORM. NOT A LOCAL ONE, OR PEOPLE NEED TO UNDERSTAND HOW TO GET BACK AND FORTH. OTHERWISE THE SEMANTICS OF THOSE NAMES CHANGE WITH THE MACHINE. IF WE CAN'T LIVE WITH THAT SITUATION, WE'RE IN SERIOUS TROUBLE. WE REALLY CAN'T EVEN HAVE IDNS AS THEY'RE DEFINED TODAY, MUCH LESS TOP-LEVEL IDNS, AND WE NEED TO RETHINK THE PROBABLY AND UNDERSTAND WE'RE PROBABLY TALKING ABOUT A FIX OVER DECADES RATHER THAN MONTHS OR YEARS.
ANOTHER PROBLEM STATEMENT ISSUE WHICH I'M NOT GOING TO TALK ABOUT TODAY BUT WE WILL PROBABLY GET TO MORE EITHER LATER OR IN VANCOUVER IS WE NEED TO BE VERY CAREFUL ABOUT ASSUMPTIONS ABOUT THE DNS THAT INVOLVE ALIASING. ALIASING OFTEN DOESN'T WORK AS EXPECTED. WE DON'T HAVE A GOOD MECHANISMS FOR SAYING TREAT THIS AS ANOTHER STRING. IT'S HARD TO COORDINATE ALIASING EXCEPT UNDER VERY RESTRICTED CONDITIONS. NOTIONS OF TRANSLATED COPIES OF ZONES REQUIRE NONDELEGATED HIERARCHIES AND ALMOST NO ONE WANTS AN UNDELEGATED HIERARCHY.
SOME OF THE NAMING TRICKS WE USE WILL BE QUITE OBVIOUS IF DNSSEC IS ALSO USED. THAT'S NOT QUITE EQUIVALENT TO SAYING THAT THEY WON'T WORK BUT IT COMES PRETTY CLOSE.
SO AGAIN, WE NEED TO UNDERSTAND THESE PROBLEMS IN THE CONTEXT OF USE AND IN THE CONTEXT OF THE OPERATION OF DNS AND IN THE CONTEXT OF WHAT USERS EXPECT, NOT BY TRYING TO FOCUS NARROWLY ON DOMAIN NAMES.
THANK YOU.
>>HUALIN QIAN: GOOD AFTERNOON. I JUST WANT TO GIVE YOU SOME OBSERVATION FROM MY POINT OF VIEW, BECAUSE IN CJK COMMUNITY, ESPECIALLY IN CHINA, WE HAVE DONE THE IDN WORK FOR ALMOST FIVE YEARS. SINCE THE YEAR 2000 WE HAVE LAUNCHED OR TEST BED FOR IDN DOMAIN NAMES.
ACCORDING TO OUR EXPERIENCE THAT THE IDN IS (INAUDIBLE) TOP LEVEL, IDN TO IDN. IT IS VERY CONFUSING TO PEOPLE. BECAUSE, FOR EXAMPLE, IN CJK SITUATION, THE HAN CHARACTERS ARE POPULARLY USED BY DIFFERENT CHARACTERS, BUT IT LOOKS LIKE THE SAME SCRIPTS. BUT WHEN CONNECTED TO DIFFERENT LANGUAGES, ITS VARIANT TABLE SHOULD BE CITED, ARE VERY DIFFERENT.
SO WHEN WE'RE USING, THE MOST DIFFICULT THING IS SET UP THE VARIANT TABLE CORRECTLY. AND THESE ARE -- THE WORK HAVE BEEN DONE BY EXPERTS NOT PERSONS MADE BY ME. THE COMPUTER SCIENCE PERSON, NOT THE LANGUAGE EXPERTS.
AND SO THAT TOOK A VERY LONG TIME. RECENTLY, MAYBE TWO MONTHS AGO, WE HAVE SENT TO THE IANA THE VARIANT TABLE FOR CHINESE CHARACTERS, BECAUSE WE HAD TO THINK ABOUT THE -- TO PROTECT THE END USERS FOR THEIR NAME WITH VARIANTS.
AND ALSO MAKE THE SYSTEM TECHNICALLY FEASIBLE WITHOUT THE VARIANT TABLE TO NORMALIZE THE STRING INPUT BY THE USER YOU CANNOT CORRECTLY RESOLVE THOSE DOMAIN NAMES.
SO SOME PEOPLE MAY SAY THAT THIS DOMAIN LABEL IS ONLY AN IDENTIFIER, BUT I THINK TALKING ABOUT THE IDN, IT'S NOT ONLY THE IDENTIFIER BUT ALSO HAVE SOME RELATIONSHIP WITH THE LANGUAGES, BECAUSE IN DIFFERENT LANGUAGES, THEY WILL USE DIFFERENT LANGUAGE VARIANTS TABLES. THEY HAVE DIFFERENT MEANING.
SO THERE'S A TOTAL DIFFERENCE. SO THAT'S WHY WE THINK THE LANGUAGE SHOULD BE CONSIDERED. AND WHEN WE CONSIDER THE LANGUAGES, THE LANGUAGES TAG MUST BE NEEDED. EITHER EXPLICITLY OR IMPLICITLY.
FOR EXAMPLE, (SAYING WORD) MEANS CHINA IN THE TOP-LEVEL DOMAIN. THAT NAME ITSELF ALREADY DEFINED THE TAG, LANGUAGES TAG. IT MEANS CHINESE. (SPEAKING WORD) IN HAN CHARACTERS MEANS THAT IN JAPANESE CIRCUMSTANCES THEY WRITE IT. THE SITUATION IS ALREADY IMPLIED BY THE NAME ITSELF.
BUT WHEN WE USE THE GTLD NAMES IN IDN, THAT'S THE SAME CHARACTER IN DIFFERENT COUNTRIES, BUT YOU CANNOT DIFFERENTIATE AND YOU CANNOT CITE DIFFERENT VARIANT TABLES.
SO IN ONE OR ANOTHER WAY, YOU MAY BE -- IN THE OPERATING SYSTEM ENVIRONMENT YOU CAN TELL YOU ARE USING JAPANESE VARIANT. THAT IMPLIES A TAG SHOULD BE PUT WITH THE JAPANESE LANGUAGE. SO THAT'S NECESSARY.
SO I THINK FOR -- ON ONE, WE HAVE TO MAKE WORK FORWARD -- GOING FORWARD. ON THE OTHER SIDE WE HAVE TO PROTECT OUR SYSTEM TO MAKE STABLE AND RELIABLE SECURITY. AND THESE TWO THINGS ARE CONTRADICTION WHEN WE ARE USING IDN.
SO MY OBSERVATION IS THAT WE CAN STEP BY STEP DO SIMPLE THING FIRST, DO RELIABLE THING FIRST.
SO THAT'S WHY.
NOW, IN CHINA, WE ALREADY PROVIDE A SERVICE, IDN.IDN FOR YEARS, AND THE RESOLUTION TRAFFIC IS GROWING UP VERY FAST.
THIS DOESN'T MEAN THAT WE HAVE SET ANOTHER ROOT FOR THE TOP-LEVEL IDN. WE JUST SIMPLY, IN OUR PLUG-IN, WHEN PEOPLE INPUTTED THOSE STREAM WITH IDN.IDN, WE INTERNALLY CONVERTED THE TOP LEVEL IDN INTO DOT CN.
SO WHEN A USER REGISTER A NAME IDN.CN OR IDN.IDN, WE GIVE THE OTHER, THE COUNTERPART NAME TO THEM. BOTH THEY CAN HAVE IDN.CN AND IDN.IDN.
INTERNALLY, WE JUST ALL CONVERT IDN.IDN TO IDN.CN. SO THAT'S KEEPING THE SAME ROOT AS IN DOT CN. SO THAT'S NO TROUBLE AT ALL.
BUT WE HOPE TO GO FURTHER, TO HAVE THE ZONE FILE, ROOT ZONE FILE REGISTER THOSE CCTLD IDN NAMES. THEN WE NEED -- WE DON'T HAVE TO NEED THE PLUG-IN TO CONVERT THE NAMES, AND IT'S MORE CONVENIENT FOR USERS.
THIS IS ONE OF THE THINGS.
THE OTHER -- ANOTHER THING IS THE -- WHERE TO CITE TO DO THE NORMALIZATION TO USE THE VARIANT TABLE.
BECAUSE WITHOUT DOING THAT, YOU CANNOT -- A SINGLE LABEL, MAYBE YOU HAVE THOUSANDS OF DIFFERENT COMBINATIONS WITH VARIANTS.
SO WE HAD TO DO NORMALIZATION.
BUT AT THE MOMENT, WE DON'T THINK IT -- BECAUSE IT'S VERY BIG TABLE.
BUT EVERY CLIENT PLUG-IN, THEY WANT TO KEEP IT AS SIMPLE AS POSSIBLE.
SO WE JUST ARE DOING THE NORMALIZATION IN THE SERVER.
WHEN PEOPLE TYPE IN THE IDN.IDN NAMES OR IDN.CN NAMES, THEN AUTOMATICALLY, THE PLUG-IN WILL SEND IT TO A SERVER TO DO A NORMALIZATION, AND THEN GO INTO THE DNS SYSTEM.
THANK YOU.
>>CARY KARP: I ACTUALLY HAVE NO PREPARED STATEMENT.
I AM IN A SITUATION HERE WHERE THE CONCERN IS THAT WE ARE LOOKING AT AN INTERNATIONALIZATION OF THE DOMAIN NAME SPACE WITHOUT THERE BEING ANY PARTICULATION IN THE INITIATIVE.
AND FROM THE PERSPECTIVE OF A CULTURALLY ORIENTED COMMUNITY WEB-BASED GTLD, VERY, VERY INTERESTED IN SEEING HOW, AS THIS DEPLOYMENT PROGRESSES, IT MAY END UP BEING OF USE TO US, AGAIN, WITHOUT THERE BEING ANY PARTITIONING ON THE BASIS OF SOME ARBITRARY TECHNICAL OR POLITICAL OR ANY OTHER PRINCIPLE.
SO....
>>JOHN KLENSIN: IN THAT CASE, I'LL OPEN THE MICROPHONE.
IF ANYONE HAS ANY COMMENTS WHICH TAKE PRECEDENCE OVER THEIR DESIRE TO GET TO LUNCH, OR QUESTIONS, WE'D BE HAPPY TO HEAR THEM.
>> WALTER WU: THIS IS WALTER WU FROM CNIC.
AND, ACTUALLY, HERE I WOULD LOVE TO REPRESENT CDNC TO GIVE POINTS ABOUT THE IDN INITIALS.
ACTUALLY, CDNC IS AN ORGANIZATION THAT'S BETWEEN CNIC AND TWNIC, HKNIC AND MONIC.
AND THEY HAVE A LOT OF EXPERIENCE ON RESEARCH AND DISCUSSION ABOVE THE IDN ISSUES FOR SEVERAL YEARS.
HERE I WOULD JUST WANT TO MENTION ABOUT THE SEVERAL BASIC PRINCIPLES HERE THAT FIT DNC (INAUDIBLE) OF IDN INITIALS IN THE COMMUNITY.
AND THE FIRST ONE IS, WE SHOULD CONSIDER TO GIVE A PRIORITY TO INTERNATIONALIZED CCTLD.
AND THE SECOND ONE IS, FOR CONVENIENCE'S SAKE, ONLY ONE FORM OF THE LANGUAGE CHARACTER VARIANT OF INTERNATIONALIZED CCTLDS IS ACCEPTED.
FOR THIRD ONE IS SUPPORTED BY THEIR OWN GOVERNMENTS, CCTLD REGISTRIES MAKE SELF CHOICE OF IDN CHARACTER SETS FOR CCTLDS.
AND THE FOURTH ONE IS, CLEARLY STATE WHEN NECESSARY PRIORITY AND THE COST FOR IMPLEMENTING INTERNATIONALIZED CCTLD IN EACH PROPOSAL THAT IS SUBMITTED TO ICANN.
AND THE FIFTH POINT IS, REGISTER AN OPERATOR OF INTERNATIONALIZED CCTLD IN THE ROOT DNS SERVER IN THE FORM OF IDNA PUNYCODE.
AND WE ALSO WOULD LOVE TO GIVE A SUGGESTION AND SUGGEST ICANN TO CONSIDER FORMING A COMMITTEE ON INTERNATIONALIZED CCTLDS.
AND WE HOPE THIS COMMITTEE WILL HAVE SEVERAL MISSIONS.
THE FIRST MISSION IS TO CONTROL THE SPONSOR OF CODIFICATION.
AND WE THINK THE PROPOSAL SHALL BE SUBMITTED BY CORRESPONDING CCTLD REGISTRY OR GOVERNMENT-ACCREDITED INSTITUTION WHICH MEETS THE FOLLOWING CRITERIA: THE COUNTRY AND REGION HAS CERTAIN MARKET NEEDS.
AND SECOND ONE IS POSSESSING ABUNDANT TECHNICAL FORCE IN STAFF, EQUIPMENT, AND OPERATION.
THE THIRD ONE IS HAVING BROAD SUPPORT FROM THE LOCAL GOVERNMENT AND THE INTERNET COMMUNITY.
THE FOURTH ONE IS POSSESSING ABUNDANT FINANCIAL RESOURCES.
AND WE SUGGEST THE SECOND MISSION OF THE COMMITTEE IS TO ASK FOR THE GOVERNMENT APPROVAL.
WE BELIEVE, ACTUALLY, THE PROPOSAL SHALL BE FULLY SUPPORTED BY THE CORRESPONDING GOVERNMENTS.
AND THE COMMITTEE SHALL REQUIRE THE SPONSOR TO PROVIDE APPROVAL OF DOCUMENTS ISSUED BY THE GOVERNMENT FOR CHOOSING THE -- A SPECIFIC TLD NAME.
AND THE THIRD ONE IS ABOUT THE PROCESS WE SUGGEST, ACTUALLY, THE REGISTRY WHO SUBMITS THE PROPOSAL TO THE EVALUATION COMMITTEE, AND THE EVALUATION COMMITTEE CAN GIVE A FULL-SCALE EVALUATION ON THE PROPOSAL AND THEN SUBMIT TO ICANN TO -- IF ICANN AGREES TO SUPPORT THIS PROPOSAL, AND THEN REGISTRY CAN CONSIDER TO OFFICIAL LAUNCH.
BUT IF IT WAS REFUSED AND THEN THE LAUNCHING OF THE APPLICATION REGISTRY SHOULD CONSIDER TO MODIFY AND THEN SUBMIT TO EVALUATION COMMITTEE AGAIN.
AND ABOUT THE ORGANIZATIONAL SUGGESTION ON THIS EVALUATION COMMITTEE, WE ALSO SUGGEST THE COMMITTEE SHALL BE COMPOSED OF REPRESENTATIVES FROM BOTH GNSO AND THE CCNSO, EXPERIENCED DNS ENGINEERS, EVEN DIRECTORS OF ICANN BOARD, IF POSSIBLE.
AND ALSO, THE WHOLE PROCEDURE SHALL BE TRANSPARENT AND OPEN FOR PUBLIC INPUTS.
ACTUALLY, THIS PRESENTATION HAS BEEN GIVEN IN THE CCNSO MEETING YESTERDAY.
AND HERE WE JUST WANT TO MENTION ALL OF THE FEEDBACK AND INPUT IS WELCOME.
THANK YOU.
>>JOHN KLENSIN: DON'T LEAVE THE MICROPHONE.
I'D LIKE TO MAKE ONE REQUEST AND ASK YOU TWO QUESTIONS.
THE REQUEST IS, IF I MAY GUESS WHAT ICANN STAFF WOULD SAY IF THEY WERE SITTING HERE, COULD YOU SEND THIS PROPOSAL IN TO STAFF SO THIS CAN BE POSTED ON THE WEB PAGE FOR DISCUSSION, IN ADDITION TO READING IT HERE.
AND THE TWO QUESTIONS ARE WHETHER YOUR COMMITTEE HAS CONSIDERED HOW A COUNTRY WHICH HAS MULTIPLE OFFICIAL LANGUAGES SHOULD BE REPRESENTED IN A SITUATION UNDER THE CONDITIONS OF YOUR PROPOSAL.
AS PROFESSOR QIAN POINTED OUT, YOU'RE IN A VERY LUCKY SITUATION WITH THE CHINESE LANGUAGE IN THE CDNC NICS, BECAUSE WITH EACH OF THOSE AREAS, THE FACT THAT A STRING IS IN THAT COUNTRY MEANS THAT IF IT'S IN SOME PARTICULAR LANGUAGE, THAT LANGUAGE IS ALMOST CERTAINLY CHINESE.
BUT IN A COUNTRY WHICH HAS MULTIPLE OFFICIAL LANGUAGES, THAT MAY NOT -- THAT WOULD NOT BE TRUE.
IN A COUNTRY WHICH HAS MULTIPLE OFFICIAL LANGUAGES WHICH THEY ARE BY LAW OR CONSTITUTION PROHIBITED AGAINST DISCRIMINATING AGAINST ONE IN FAVOR OF ANOTHER, IT IS CERTAINLY NOT TRUE.
AND I'M WONDERING IF YOUR COMMITTEE HAS LOOKED AT THAT QUESTION IN THE CONTEXT OF HOW THEY APPLY FOR IDN TLDS.
AND THE OTHER QUESTION IS WHETHER YOU EXPECT WITH THOSE IDN TLDS ANY RELATIONSHIP BETWEEN THE NAMES WHICH ARE REGISTERED IN THE 3166-BASED TLD IN CORRESPONDENCE OR NOT TO THE STRINGS WHICH ARE REGISTERED IN THE IDN TLD.
>> WALTER WU: WHAT YOU REQUEST, WE WOULD LOVE TO DO THAT.
AT THIS TIME, ACTUALLY, WE HAVE RECEIVED SOME FEEDBACK FROM THE OTHER CONCERNED ENTITIES, AND WE WILL TRY TO HAVE SOME INTERNAL DISCUSSION ABOUT THAT FEEDBACK.
AND WE WILL MAYBE GIVE SOME MODIFICATION, AND WE WOULD LOVE TO SEND TO THE CONCERNED PARTY.
YES.
>>JOHN KLENSIN: DO IT SOON.
IT'S PROBABLY BETTER TO DO IT BADLY BUT GET IT POSTED FOR DISCUSSION THAN TO SPEND SEVERAL MONTHS DISCUSSING IT AND THEN GET IT POST THE.
>> WALTER: YEAH, DEFINITELY.
>>JOHN KLENSIN: THINGS ARE FINALLY BEGINNING TO MOVE QUICKLY.
THAT'S GOOD NEWS.
>> WALTER WU: AND, ACTUALLY, ABOUT YOUR QUESTION, THE FIRST QUESTION YOU MENTIONED, I THINK, IS ABOUT THE -- ABOUT SOME SPECIFIC COUNTRIES, IF THEIR OFFICIAL LANGUAGES MAY BE MORE THAN ONE, HOW TO DEAL WITH THIS SITUATION.
ACTUALLY, WE BELIEVE THAT THE SITUATION IS CURRENTLY THERE WILL BE DIFFERENT DEMAND OF EACH COUNTRY.
YES, YOU'RE RIGHT.
BUT VARIOUS, A LOT OF COUNTRIES, THEY HAVE MORE THAN ONE OFFICIAL LANGUAGE.
HERE, ACTUALLY, LUXEMBOURG HAS MORE THAN ONE OFFICIAL LANGUAGE HERE.
SO -- BUT I THINK WE SHOULD LEAVE THIS TOPIC, THIS IS THE DECISION THAT WILL BE MADE BY THE REGISTRY AND THE CORRESPONDING GOVERNMENTS.
AND, ACTUALLY, I THINK THIS KIND OF DECISION IS, THEY WILL CONSIDER WHETHER TO PROVIDE, YOU KNOW, ONLY ONE LANGUAGE TYPE OF IDN SERVICE OR MORE THAN ONE.
AND THEY CAN SUBMIT THIS KIND OF PROPOSAL TO THE COMMITTEE.
I THINK THAT MEANS EVERY COUNTRY AND EVERY REGISTRY WILL HAVE THEIR OWN DECISION ABOUT THIS POINT, NOT DECIDED OR NOT SUBMITTED BY THE OTHER ENTITY.
SO THAT IS THE FIRST QUESTION.
AND THE SECOND ONE, I'M SORRY, CAN YOU BRIEFLY -- I THINK I DIDN'T RECOGNIZE VERY WELL.
CAN YOU JUST BRIEFLY REPEAT, SAY IT AGAIN, THE SECOND QUESTION.
>>JOHN KLENSIN: THERE ARE TWO WAYS OF THINKING ABOUT THE PURPOSE OF A IDN TLD IN RELATIONSHIP TO THE TRADITIONAL COUNTRY CODE TLDS.
ONE OF THOSE WAYS OF THINKING ABOUT THEM IS THEY ARE TWO COMPLETELY SEPARATE DOMAINS WITH SEPARATE REPRESENTATIONS.
THE OTHER WAY OF THINKING ABOUT THEM IS THAT THE IDN TLD IS SOMEHOW A SHADOW OR TRANSLATED FORM OR EQUIVALENT FORM TO THE TRADITIONAL COUNTRY-CODE TLD, WITH MOST OR ALL OF THE SAME REPRESENTATIONS, ONLY WITH THOSE REPRESENTATIONS INTERNATIONALIZED, IF NECESSARY, AT THE SECOND LEVEL.
AND I'M WONDERING WHAT YOU ARE ANTICIPATING OR DO YOU THINK THAT'S ANOTHER SEPARATE COUNTRY DECISION?
>> WALTER WU: YEAH, ACTUALLY, I THINK IN ALL OUR EXPERIENCE, WE THINK THAT, ACTUALLY, YOU'RE RIGHT, THAT VARIOUS -- IF THE IDN.IDN IS OFFICIALLY LAUNCHED, THERE WILL BE, YOU KNOW, TWO KINDS OF TLDS, THE IDN DOT ORIGINAL ASCII FORM, LIKE IDN.CN, BUT ALSO MAYBE IDN.ZHONGGUO.
BUT OUR SUGGESTION IS, THERE HAS SOME REGISTRY BEGUN THE TEST BED OF IDN BASED ON THE ASCII TLD.
FOR EXAMPLE, CNIC LAUNCHED THE SERVICE OF IDN.CN.
AND, ACTUALLY, IF IN THE FUTURE THE IDN.IDN WILL BE OFFICIALLY LAUNCHED BY CNIC, WE WILL COMBINE THIS TOGETHER.
THAT MEANS, YOU KNOW, ALL THE IDNS -- THE IDN.CN AND IDN.IDN WILL BE COMBINED TOGETHER.
AND EVENTUALLY, ACTUALLY, WHEN IDN.IDN WILL BE OFFICIALLY LAUNCHED, IT WILL BE THE -- ONLY ONE SINGLE SERVICE TO THE USER, THAT WILL BE THE IDN.IDN.
>>CARY KARP: MAY I JUST ADD A REQUEST TO ALL OF THIS.
IF YOU ARE GOING TO BE ADDRESSING THE ISSUE OF HAVING ALL OF THE CC'S THAT HAVE A SHARED INTEREST IN A GIVEN LANGUAGE DISCUSS HOW THAT LANGUAGE MAY BE APPROPRIATELY REPRESENTED ON ALL LEVELS OF THE DNS, WOULD YOU BE WILLING TO SHARE THE RESULT OF THAT WITH THE PEOPLE WHO ARE DISCUSSING IN THE FIRST PANEL SESSION SIMPLY KNOWING HOW A GIVEN SCRIPT IS USED IN GIVEN LANGUAGE SITUATIONS SO THAT THE WISDOM YOU WILL BE GENERATING CANNOT SIMPLY BE APPLIED IN THE TOP-LEVEL IDN SITUATION, BUT CAN BE APPLICABLE, AGAIN, ON ALL LEVELS IN WHAT WILL BE AN INCREASINGLY INTERNATIONALIZED NAME SPACE?
>> WALTER WU: YEAH, --
>>CARY KARP: IT'S JUST A QUESTION.
I'M NOT ASKING YOU -- IT'S SOMETHING I HOPE THAT YOU WILL KEEP IN MIND.
WHAT YOU ARE ADDRESSING IS A FAR MORE COMPOUND COMPONENT OF INTERNATIONALIZING THE INTERNET THAN THE SINGLE APPLICATION OF IT WHICH IS YOUR IMMEDIATE CONCERN.
AND WE HOPE THAT AS YOU CONDUCT THE DISCUSSIONS THAT WILL YOU BE CONDUCTING THAT YOU REALIZE YOU WILL BE DOING SOMETHING MORE IMPORTANT THAN JUST DISCUSSING THE CHARACTERS REQUIRED FOR TOP-LEVEL IDN.
>> WALTER WU: YEAH, I THINK YOUR POINT IS A VERY GOOD POINT.
AND WE WILL CONSIDER, YEAH.
OKAY.
>>HUALIN QIAN: I THINK FOR THE IDN, TOP-LEVEL IDN, IT'S -- THERE'S NO INTERNATIONAL STANDARD FOR EACH COUNTRY'S NAME IN DIFFERENT LANGUAGES.
SO THAT SHOULD BE DECIDED BY THE LOCAL GOVERNMENTS AND THE LOCAL REGISTRY.
THEY CAN, FOR EXAMPLE, SELECT THE WORD REPRESENTING THIS COUNTRY IN THEIR OWN LANGUAGE.
THEN ANOTHER THING, I THINK THE AMOUNT IS NOT VERY (INAUDIBLE) FOR EXAMPLE, IN INDIA, THEY HAVE 18 OFFICIAL LANGUAGES, BUT ONLY 18.
FOR MANY OTHER COUNTRIES, THEY DON'T HAVE SO MUCH. AND FOR MANY COUNTRIES, THEY DON'T EVEN USE IDN TOP-LEVEL DOMAIN.
ENGLISH IS ALREADY THE OFFICIAL LANGUAGE.
SO I THINK THIS IS NOT A TROUBLE, WE DON'T HAVE TOO MUCH -- TOO MANY OF THE NAMES THAT ARE HARD FOR THE ROOT TO DEAL WITH.
>> WALTER WU: COMMENT FOR THAT.
WE REALLY WANT TO POINT OUT THIS PROPOSAL AT THIS TIME, BECAUSE WE THINK THE DEMAND OF IDN IS REALLY VERY -- HAS BEEN VERY GREAT.
AND IF ICANN OR SOME RELATED ENTITY DO NOT MAKE SOME OFFICIAL PROPOSAL OR AT LEAST WE MAKE A FORMAL PROCEDURE AS TO HOW TO IMPLEMENT FOR THAT, IT WILL BE, YOU KNOW -- ANY OTHER KIND OF ALTERNATIVE SOLUTION INCLUDING SOME ALTERNATIVE ROOT OR SOME KEY WORD SOLUTION WILL CONFOUND.
AND, ACTUALLY, IT WILL GET A LOT OF CONFUSING TO INTERNET USERS.
AND SO IT'S THE RIGHT TIME WE MAKE THE DECISION.
OKAY.
THANK YOU.

>>JOHN KLENSIN: OTHER QUESTIONS?
TINA?
>>TINA DAM: I WANTED TO MAKE A COMMENT.
I KNOW THERE WERE MORE PEOPLE IN LINE FOR QUESTIONS.
BUT I WANTED TO MAKE AN ADDITIONAL COMMENT TO THIS TOPIC, IF THAT'S OKAY.
FIRST OF ALL, THE REASON WHY WE HAVE THIS SESSION IS EXACTLY TO GET THE PROCESS STARTED ON GETTING IDNS ON THE TOP LEVEL.
AND THAT IS SOMETHING THAT'S GOING TO BE MOVED FORWARD TO VANCOUVER, WHERE WE'LL HAVE A MUCH LARGER SESSION.
I CERTAINLY HOPE, AND IT'S SOMETHING THAT ICANN STAFF AND THE PANEL MEMBERS AND SEVERAL OF THE SUPPORTING ORGANIZATIONS AND OTHER WORKING GROUPS AND COMMITTEES HAVE BEEN TALKING ABOUT, MAYBE NOT AS FORMAL AS WE DO RIGHT HERE RIGHT NOW, IS TO MAKE SURE THAT ALL OF THOSE GROUPS GET TOGETHER AND MOVE FORWARD IN A COMBINED EFFORT TO REACH THOSE GOALS.
I ALSO WANTED TO MENTION TWO OTHER THINGS.
WHEN WE TALK ABOUT MULTIPLE LANGUAGES AND HOW THAT IS MAKING THINGS DIFFERENT FROM -- DIFFICULT FOR SOME PARTS OF THE WORLD, THAT'S EXACTLY WHY THE PREVIOUS SESSION, AS FAR AS I UNDERSTAND, WAS TALKING ABOUT MOVING FROM LANGUAGE-BASED TO SCRIPT-BASED.
SO THAT'S JUST A COMMENT ON THAT.
AND THE LAST THING IS THAT THE PRESIDENT'S COMMITTEE ON IDNS THAT ICANN HAS BEEN UNDERWAY TO FORM FOR QUITE SOME TIME HAVE -- AND THAT'S A DIRECT COMMENT TO WALTER'S QUESTIONS -- HAVE LISTED SEATS FOR BOTH THE CCNSO AND THE GNSO.
AND WHAT I HEARD THE GTLD REGISTRIES SAY IS ALSO THAT THEY APPRECIATE AND WELCOME AND SUPPORT THAT KIND OF INTERACTION.
SO I JUST WANTED TO PUT ON THE RECORD THAT ICANN SUPPORTS THAT AS WELL.
>>JOHN KLENSIN: DOES THAT MEAN YOU'RE ABOUT TO ANNOUNCE THE MEMBERSHIP OF THAT COMMITTEE?
>>TINA DAM: THE LIST IS IN FRONT OF THE CEO OF ICANN.
AND I BELIEVE HE'S GOING TO APPROVE THAT, AND WE'LL SEND OUT THE INVITATIONS.
>>AMADEU ABRIL I ABRIL: GOOD MORNING.
MY NAME IS AMADEU ABRIL.
AND AS YOU SEE, I AM WEARING MY DOT ORG HAT, SO I AM TALKING AS CHAIR OF THE PIR IDN COMMITTEE.
I AM NOT HERE TO EXPLAIN OUR STRATEGY REGARDING INTRODUCING AS MANY LANGUAGES AS POSSIBLE IN THE IDN PROCESS, BUT JUST TO CONVEY A COUPLE OF ADVICES WE GOT FROM OUR ADVISORY COUNCIL LAST WEEK IN BRUSSELS REGARDING IDNS.
THERE ARE TWO VERY CONCRETE THINGS.
THE FIRST ONE, OUR ADVISORY COUNCIL STRONGLY ENCOURAGED IANA WHEN IANA RECEIVES A LANGUAGE TABLE FOR ANY LANGUAGE FROM A REGISTRY, BE THAT GTLD, CCTLD, OR WHATEVER KIND OF TLD WE MAY HAVE IN THE FUTURE, THAT IANA DOES AT LEAST A SUMMARY EXAMINATION OF THE CONTEXT WITH THE RELEVANT LANGUAGE COMMUNITY THAT THE REGISTRY HAS HAD IN DETERMINING THE LANGUAGES.
THAT IS, JOHN OR ME OR ESPECIALLY MICHAEL EVERSON, WHO IS THERE, MIGHT HAVE SOME KNOWLEDGE ABOUT WHAT ARE THE CHARACTERS USED IN ROMANIAN, PORTUGUESE, OR SWEDISH.
BUT THE QUESTION IS NOT WHETHER SOMEONE WITHIN THE REGISTRY KNOWS IT, BUT WHETHER THIS HAS FOLLOWED SOME PROCESS.
AND WE ALSO PROPOSED TO ESTABLISH A SORT OF LIST OF HIERARCHICAL AUTHORITIES FOR, YOU KNOW, AUTHORITATIVE SOURCE REGARDING LANGUAGES.
LANGUAGE USES THAT REGARDING, YOU KNOW, LEGISLATION, OFFICIAL ACADEMIES, NONOFFICIAL ACADEMIES, OTHER AUTHORITATIVE SOURCES, BE THAT DICTIONARIES OR, YOU KNOW, UNIVERSITIES HAVE A SPECIAL RELATIONSHIP WITH THE LANGUAGE OR WHATEVER.
THAT THERE IS SOMEHOW, OR A CHECK OR A LIST, SOMEHOW A REQUEST TO THE TLD OF WHETHER THIS PROCESS IS BEING FOLLOWED.
THE SECOND CONCRETE ADVICE WAS THAT REGISTRIES AND REGISTRARS SHOULD TAKE INTO CONSIDERATION THAT NOT ONLY THE DOMAIN NAME IN ASCII AND UTF 8, WHATEVER, LANGUAGE CHARACTER SET HAS TO BE ALLOWED TO BE REGISTERED, BUT THEN THAT THE -- THIS PLAYER WHO IS IN FORMATION SHOULD ALSO INCLUDE THE ABILITY TO DO THE SAME FOR ALL THE CONTACT DATA FROM THE REGISTRANT THAT TYPICALLY WILL BE IN THAT SPECIAL CHARACTER SET AND NOT PROBABLY IN ASCII ONLY.
SO THIS SHOULD ALSO BE TAKEN INTO ACCOUNT WHEN WE DESIGN FOR EACH REGISTRY, FOR EACH TOP-LEVEL DOMAIN THE STRATEGY.
AND NOW, WITHOUT THE DOT ORG HAT, MY PERSONAL COMMENT, SOMETHING THAT TINA MENTIONED, SOMETHING THAT'S BEEN HOUNDING ME FOR SOME TIME, WHETHER IT REALLY MAKES SENSE TO DO LANGUAGE TABLES IN A LANGUAGE BASE OR SOMETIMES IT'S BETTER TO GO SCRIPT-BASED, THAT IS, LATIN, LATIN ONE PLUS EXTENDED, ET CETERA.
FOR SOME CERTAIN LANGUAGES, LET'S TAKE ROMAN LANGUAGES IN LATIN SCRIPT.
FOR THOSE DOZEN --
>>JOHN KLENSIN: DID YOU HEAR THE DISCUSSION OF THE EARLIER SESSION?
>>AMADEU ABRIL I ABRIL: YES.
>>JOHN KLENSIN: OKAY.
I JUST -- WE'RE VERY TIME-LIMITED.
SO PIECES OF YOUR SPEECH WHICH HAVE ALREADY BEEN TALKED ABOUT I'D RATHER CUT SHORT.
>>AMADEU ABRIL I ABRIL: OKAY.
I JUST WANTED TO SAY THAT SOMEHOW WE HAVE THIS TYPE OF CONCERN THAT IT WOULD BE BETTER ON THIS ONE.
THE LAST ONE WAS SOMETHING YOU SAID, JOHN.
I GET A LITTLE BIT NERVOUS WHEN WE START TALKING HERE ABOUT OFFICIAL LANGUAGES OR NOT.
I DON'T THINK THAT ICANN OR IDNS HAVE MUCH TO DO WITH OFFICIAL OR NONOFFICIAL LANGUAGES.
YOU KNOW, CHINESE IS NOT OFFICIAL IN THE UNITED STATES, BUT, YOU KNOW, THE PROBLEM IS EXACTLY THE SAME WHEN IT'S -- A CHINESE PERSON IS TRYING TO USE THE LANGUAGE.
I WOULD PREFER THAT WE KEEP TALKING ABOUT LANGUAGES AND SCRIPTS AND WE FORGET SOME LEGAL STATUS OF THEM, EXCEPT WHEN IT COMES WITH CHECKING WITH THE RELEVANT AUTHORITIES WHETHER THE SOLUTION IS CORRECT OR NOT.
>>JOHN KLENSIN: LET ME EXPLAIN THE "OFFICIAL LANGUAGE" COMMENT.
I, TOO, PREFER TO AVOID IT.
BUT IT BECOMES RELEVANT AT THE POINT THAT YOU BEGIN TO SAY CCTLDS SHOULD BE ENTITLED TO IDNS IN ADDITION TO THE LANGUAGES -- TO DOMAINS WHICH THEY'RE RUNNING.
AND THEN YOU GET INTO AN ISSUE OF OFFICIAL LANGUAGES, BECAUSE A COUNTRY HAS TO DECIDE WHAT IDNS IT WANTS, AND ICANN HAS TO DECIDE WHAT THE MAXIMUM NUMBER OF IDNS IS IT'S GOING TO GIVE A COUNTRY, WHETHER THAT MAXIMUM IS SET AS ONE OR THAT MAXIMUM IS SET AS THE NUMBER OF OFFICIAL LANGUAGES OR THAT MAXIMUM IS SET AS THE NUMBER OF LANGUAGES SPOKEN WITHIN THE COUNTRY, OR THAT NUMBER IS SET AS THE NUMBER OF LANGUAGES WHICH ANYONE IN THE COUNTRY IS INTERESTED IN PRODUCE VERY, VERY DIFFERENT IMPACTS ON THE DNS.
SO THAT WAS THE REASON FOR THE QUESTION.
>>MICHAEL EVERSON: HELLO.
IT'S KIND OF AN INTERESTING NEW THING THAT'S JUST OCCURRED TO ME.
THE TOP-LEVEL DOMAIN ABBREVIATIONS DERIVED FROM ISO 3166, THEY ARE IN LATIN, THAT STANDARD DOES MAKE PROVISION FOR THE EXTENSION OF A LIST OF TWO-LETTER ABBREVIATIONS OR TWO-CHARACTER CHINESE ABBREVIATIONS OR WHATEVER -- I MEAN, (SPEAKING CHINESE), FOR IRELAND, IT MIGHT BE "AI-ER" OR IT MIGHT BE "AI-ER-LAN."
BUT WHAT I THINK, SINCE WE ARE GOING INTO THIS AREA WHERE IT IS ACTUALLY SCRIPT-BASED, WE SHOULD PROBABLY BE WORKING -- ICANN SHOULD BE WORKING WITH ISO TC-46, WHICH IS RESPONSIBLE FOR ISO 3166, IN ORDER TO ENSURE THAT, FOR INSTANCE, FOR A GIVEN SCRIPT, THERE WILL ONLY BE ONE ABBREVIATION FOR A PARTICULAR COUNTRY.
WE DON'T WANT TO HAVE -- LET'S SAY I'M MAKING THIS UP OFF THE TOP OF MY HEAD -- LET'S SAY THAT THERE ARE THREE DIFFERENT, IN PERSIAN, ARABIC, AND PASHTO, THERE ARE THREE DIFFERENT WAYS OF SAYING ARMENIA, LIKE HAYASTAN, OR ARMEYNIA, OR SOMETHING.
WE DO NOT WANT TO HAVE IN THE ARABIC SCRIPT, I BELIEVE, MORE THAN ONE WAY OF ABBREVIATING THAT.
AND THIS IS A NEW ACTIVITY WHICH SHOULD BE COORDINATED BETWEEN ICANN, LINGUISTIC EXPERTS, AND THE TC-46 WORKING GROUP, WHICH IS RESPONSIBLE FOR ISO 3166.
>>JOHN KLENSIN: AT THE RISK OF SAYING SOMETHING IN ISO SPEAK, AS YOU PROBABLY KNOW, THERE IS NO SUCH WORKING GROUP.
THEY CONSOLIDATED BACK INTO TC-46 AND THEN HANDED THE WHOLE BUSINESS OFF TO THE GENERAL SECRETARIAT AND THE MAINTENANCE AUTHORITY WHO HAS MADE IT CLEAR THEY DON'T WANT TO TOUCH THAT PROBLEM.
NOW, THAT COULD BE CHANGED, BUT THAT'S WHERE IT STANDS TODAY.
OTHER QUESTIONS OR COMMENTS?
>>MICHAEL EVERSON: IT WOULD BE A NEW WORK ITEM PROPOSAL.
>>JOHN KLENSIN: PARDON ME?
>>MICHAEL EVERSON: IT WOULD BE A NEW WORK ITEM PROPOSAL FOR -- I DON'T THINK THAT WOULD BE A PROBLEM.
THERE'S CLEARLY A USER NEED.
>>CHRISTOPHER WILKINSON: JOHN, JUST TO PUT DOWN A SMALL MARKER THAT -- IN THE AREA OF IDN, I DON'T THINK THERE'S A PRESUPPOSITION THAT THE STANDARD FOR COUNTRY NAMES WOULD BE 3166.
THERE ARE OTHER LISTS OF COUNTRIES WHICH, IN FACT, TODAY HAVE A HIGHER LEVEL OF AUTHENTICATION AND APPROVAL.
>>JOHN KLENSIN: OTHER QUESTIONS OR COMMENTS?
OKAY.
WE'RE GOING TO GET OUT -- UNLESS THE PANELISTS WOULD LIKE TO SAY SOMETHING ELSE, WE'RE GOING TO GET OUT OF HERE ON SCHEDULE.
THANK YOU VERY MUCH.
SEE YOU IN VANCOUVER.
(APPLAUSE.)