GAC Plenary:  New gTLD AG v3
Sunday, 25 October 2009
ICANN Meeting
Seoul, Korea

>>VICE CHAIR DE LA CHAPELLE:   Good afternoon.  If you can please be
seated, we are going to start very quickly, because we have little --
not so much little time, but lots of issues.

Good afternoon.  My name is Bertrand de la Chapelle.  Janis
Karklins, the chair of the GAC, asked me to replace him at least for
the beginning of this session that is devoted to the draft applicant
guidebook Version 3 that has been released briefly before the meeting.

We'll have several presentations this afternoon.  The first one is
by Kurt Pritz that will give an overview of the changes that are
included in the draft applicant guidebook, Version 3, followed by a
series of -- or time for questions.

The second presentation will be done by Kristina Rosette, for the
work done by the IRT regarding trademark protection in particular.

And then there will be a presentation by WIPO on a similar issue
regarding the process that they considering.

Kurt, whenever you are ready, we can start.

There will be probably a lot of questions that will be brought up by
the presentation of Kurt.

He promised that he would make the -- his introduction in about 25
minutes.

If we could devote something like 30 minutes afterwards to the
questions, and if we cannot accommodate all the questions, we'll try
to organize a way to channel those who have not been vocally
expressed to the staff in an appropriate manner.

So without further ado, Kurt, if you are ready to launch, the floor
is yours.

>>KURT PRITZ:   Thank you very much, Bertrand.

And thank you again, everybody, for accommodating me for a second
time today.

So this -- the purpose of this presentation is to provide an update.
ICANN recently published a third version of the applicant guidebook. 
And associated with that third version are an analysis of public
comments, so all the public comments that were made to the second
applicant guidebook, and also to the IRT report, were summarized and
put into categories and considered so that an explanation is given as
to how those comments were accommodated in the third version of the
guidebook.

The third version has in excess of 50 substantive changes.  And
nearly all those changes reflect public comment.  In other words,
they were essentially all made based on comments that have been made
to the guidebook.

Is clean and red-lined version of the guidebook is published.  So
you can see where the changes were made.

And the changes reflected are just between the second and third
version.  I think if you looked at all the changes that have been
made in the guidebook since its outset, it would all pretty much be
red.

And I think that is -- I think that reflects that this discussion
that's happening among the ICANN community, in which I include ICANN
staff, is that it's a meaningful discussion that comments are made
and balanced and taken into account, and to the extent that can be,
are included in the guidebook.

So this process continues now.

And what I'm going to present, if you could go on to the next slide,
what I'm going to present are just a series of discussion topics that
were chosen based on GAC discussion, you know, with ICANN staff and
in mailings, trying to highlight those issues that we think are
important to the GAC.  And then in the questions afterward, we can
discuss these issues in particular or anything you want.

So we'll just start going.  And I'll talk, and then we can stop and
ask questions.

So if you could go to the next slide.

Malicious conduct.  And the next slide, too.

So there's been considerable activity regarding the prevention of
malicious conduct in an environment where there's many gTLDs instead
of just a few.  And it's based on a variety of inputs.  So there's
been a lot of comments made to the ICANN Web site.  There's been
reports from SSAC and others, public consultations, the Anti-Phishing
Working Group; the registry security group, FIRST and CERT; the
banking and finance organizations that all contributed to the
knowledge bank.  And as a result of that, on the next slide, there
were, essentially, four key issues identified with preventing
malicious conduct or the possibility of malicious conduct as the
numbers of gTLDs multiply.

And I think that this presentation will be posted to the GAC, too. 
So that, combined with the transcripts, means I don't have to read
every slide.

But these are the four issues.  And then they were translated into
several specific improvements into the guidebook.  So on the next
slide, there are eight specific enhancements to the guidebook
intended to mitigate the potential for malicious conduct going
forward.  Enhanced background checks.  Those background checks would
include involvement in cybersquatting or over -- involvement in the
UDRP process to the extent that it indicates that the party is an
abuser of domain names.

So that's intended to ensure that bad actors do not run registries --
 Those red lines, there's four sections there.  And each of those
four tie back to the four issues on the previous slide, for those of
you who take this home and look at it.

The next issue was, how do we ensure the accuracy and integrity and
validity of the registry information to the end user.  And here, the
guidebook now requires DNSSEC deployment right at the start.  It's no
longer optional.  And includes a prohibition on wild-carding.  It
also requires -- kind of a two for one here -- includes registries to
remove glue records for deleted names.

And then another issue was, how do we provide enhanced control
framework for TLDs with intrinsic potential for abuse.  So we've --
there's been devised a model where registries can elect to designate
themselves as high-security TLDs.  And they would be verified based
on pre-published criteria that the registries could meet.

And then, finally, as far as combating abuse, the guidebook requires
the thick -- the maintenance of a thick WHOIS database.  The
publication of an antiabuse contact.  And finally an expedited
service where registries can request relief from their agreement
terms in cases where they need to act quickly to combat some sort of
cybercrime or something else.  So these are all changes from the
second guidebook.

If I could have the next slide, it's to discuss root zone scaling.

And that is, can the root zone accommodate a number of new TLDs and
continue to operate in an expected fashion.

And so I think everyone here knows that a couple of reports have
been published to inform the reports or reporting of the SSAC and
RSSAC, both the root zone scaling report and L root scaling report
that was commissioned by ICANN.  And like I said, they're going to
inform RSSAC and the SSAC report to the board on this issue.

If I could have the next slide.

And these are the main findings of the root zone scaling report,
anyway.  That the rate of change is as important as the amount of
change.  And that an early warning system should be implemented in
order to give early warning to any signs of disruption in the domain
name system.  And that, finally, a degree of growth is manageable if
there's a sufficient notice to allow for capacity planning among the
root server operators.  

So the next steps are -- go on to the next slide -- these studies
are going to be evaluated by the RSSAC and SSAC.  We're going to
interpret their results.  Certainly in going forward with new top-
level domains in any environment, that security and stability of the
DNS is paramount, and that, you know, the final approach will be
based on these studies.

So, essentially, we're waiting for the SSAC/RSSAC assimilation of
the reports and their advice to the board.

As far as economic studies -- go ahead -- everybody here has seen
this slide.  So ICANN's released four studies in the past.  And
there's been some other studies done, too.  The OECD did a report in
2009.  There was the Gandi survey.  OECD in 2004, the Gandi study,
and then Michael Kende wrote a report, too.

So what are we going to do?  

I think we've discussed the merits, the benefits, and the potential
gaps in those reports.

So the next steps are on the next slide and that's that ICANN is
going to continue this study area, retain an economist to review and
summarize work to date in context of the open questions.  And then
once doing that, augment these studies to consider the net benefits,
that is, how benefits do or do not outweigh the costs of the new gTLD
program as a whole, and then explore if that can be done on
individual applications.

With -- whoops.  Go back one.  We're not done yet.  There's more
action-packed stuff here.

So, with the goal of verifying the policy conclusion; right?  The
policy conclusion was that new top-level domains would bring benefits
to registrants through competition and choice.  And so the report --
we will ask that the report verify or not verify that conclusion.  

And in the meantime, while we do that remember, the original study
said that ICANN should, you know, realize the benefits of new TLDs,
but also mitigate the costs.  So we will continue the work of
mitigating costs in the area of trademark protections, malicious
conduct, and user confusion.  So the implementation work can continue
in parallel with this additional economic study that will be
undertaken.

>>VICE CHAIR DE LA CHAPELLE:   Kurt?

>>KURT PRITZ:   Go ahead.

>>VICE CHAIR DE LA CHAPELLE:   May I interrupt you just one second.

Can you come back to the previous.

Do you have a time line for the new study on public benefit on new
TLDs?

>>KURT PRITZ:   I'll tell you that we're in exploratory discussions
with a different set of economists, that we furnished them with --
I'll tell you all the detail -- we furnished them with a statement of
work, and they've come back with a revised statement of work.  So now
we're in the process of negotiating an agreement.

I think it --  I think the statement of work includes a two- or
three-phase study, depending on what happens at the end of phase two,
and that some initial reporting or initial work will be done this
calendar year, but then the part that everyone is looking for, I
think, will be done in the first quarter of 2010.

>>VICE CHAIR DE LA CHAPELLE:   Okay.  Thank you.

>>KURT PRITZ:   Yep.

So that's kind of brand-new information.

Okay.

I think we're up to trademark protection.  Yep.  Next slide.

And next slide.

So there's one slide here, but there's a lot to this.

In the new --  So there's -- in the -- the guidebook includes four --
includes two -- in addition to UDRP and what was already in the
guidebook, there's two new mechanisms that are actually in the
applicant guidebook to provide for increased protection for trademark
holders.  One is a post-delegation dispute resolution procedure where
rights-holders can claim -- make a claim not only against the
registrant of the name who might be abusing the name, but also if
certain criteria are satisfied, make a claim against the registry. 
So that trademark post-delegation dispute procedure is detailed in
the applicant guidebook.

Just as an aside, there's also a similar procedure for community-
based applications.  So I think it was a question from this group
earlier that asked about this, and that if somebody claims to be a
community-based TLD, there will be restrictions placed in their
agreement in accordance with that claim, and that the community-based
TLDs will be held to those restrictions throughout the term of the
registry agreement.  So there's another dispute resolution procedure
for that.  But I digress.

Oh, yes.

>>VICE CHAIR DE LA CHAPELLE:   No, you're not digressing.  Actually,
may I ask you one question that I know will come later?

Does the same type of procedure for post-delegation remedy apply,
potentially, on geographic names?  Or is it not included at the
moment?  Because you said it works for community, it works for
trademark.  Is it intended for geographic names?

>>KURT PRITZ:   That's right.  I'm going to talk more about
geographic names in a second.  But it would only apply to geographic
names insofar as it's a community TLD.  So if it labels itself as a
community TLD, then the use of that name would be restricted.

But there's more to say about that.

What happened?  Okay.

So then there were other trademark protection mechanisms or rights
protection mechanisms recommended.  And those are being referred to
the GNSO for review to see if those proposed mechanisms are
consistent with the policy advice received by the GNSO, which was
essentially that -- whoops, I'm sorry -- which was that new TLD
strings should not infringe the rights of others.

And the rights protection mechanisms are intended to be the
personification of that policy recommendation.  But given that the
policy recommendation was broad and unspecific, this is a check to
ensure that these additional rights protection mechanisms are
consistent with that policy advice.

So those two mechanisms are an I.P. or trademark clearinghouse,
which is a database for registered and unregistered trademarks that
can be -- the clearinghouse can then be used to facilitate an I.P.
claims service or a sunrise -- a sunrise service that each registry
would be obligated to implement either on I.P. claims or a sunrise
period.

And then the other mechanism is the Uniform Rapid Suspension system,
whereby names that are clear, blatant violations of trademark rights
can be removed in a rapid, inexpensive fashion.

So if you've read the board's letter to the GNSO, the board asked
the GNSO to undertake this review in an accelerated fashion and
report back in the next 60 days on whether in fact these two
recommendations are consistent with the proposed rights.

Next slide.

So additional issues.

Categories, Bertrand.

So in an effort to open this discussion, what are the goals of the
new gTLD program?  To engender greater participation in the name
space and facilitating the work of communities and organizations by
helping them use the Internet and the DNS to fulfill their missions.

So certainly we've heard that proposals for many types of TLD
categories.  And we think -- interpreting those requests, we think
that TLD categories seek to facilitate those -- those two goals,
creating areas for those seeking to address community concerns.

So what should be discussed is, you know, how should that -- how
should they be defined.

Speaking for myself, I'm sort of reticent to draw the road map for
TLD categories when, you know, the ten smartest people I know sitting
in a room would probably do a poor job of forecasting what categories
might be defined.

And, second, so how do the policy recommendations guide this
rulemaking?  And this -- so this is sort of an important concept,
which is on the next slide.  And that is, there's policy
recommendations that say that, you know, every TLD should sign a
registry agreement that should have these terms in it.  I think
there's four or five of those.  That one of those is to comply with
consensus policy pay fees and use ICANN-accredited registrars.

So the difficulty is that creating categories is -- you know, for
the goals that -- for those goals is -- of benefiting the DNS and for
benefiting the communities is something we strive for.  Whether or
how to give some sort of exemption to the policy recommendations
based on the categorization is somewhat problematic.  And that has to
be done in a careful way so as to not create, you know, from an
separations standpoint, a compliance nightmare for ICANN, but to
ensure that bright-line rules can be written so that it's clear that
we're not violating the policy recommendations and that the
categorization's sort of self-regulating so that ICANN -- the big
ICANN -- does not wind up spending a lot of registrant money on
compliance issues that really don't add to stability and security of
the DNS.

So that's the sort balancing that needs to be done as this
discussion goes forward, I think.

Can I have the next slide?

Geographic names has been discussed quite often and deeply here.

The new guidebook.

(receiving glass of water).

That is just excellent.  Thank you so much.

I start out with an annoying sort of voice and then it goes downhill.

So the new guidebook requires governments to reflect in their
approval of a geographic name TLD that they understand what the
approval is.  So a specific wording in the approval is required.

And then at the second -- I'll just -- At the second level,
importantly, second-level protections are fairly new in the guidebook
since Version 2.  And country or territory names that are on specific
lists can be released -- you know, are all reserved, and then
released through a dot info-like process.

So that's that.

So post delegation.

It's been suggested that we explore a redelegation mechanism for
geographic names that win the approval of the relevant government,
and then act in contravention to the agreement, real or implied, that
they had with the government as far as operating their TLD.

So to start at the bottom, creating a redelegation mechanism as part
of the guidebook is somewhat difficult in that there is a
redelegation process currently that requires the consent of the
government, the local DNS community, and the present operator.  So
that's memorialized in ICP 1 and RFC -- Kim will tell me later.

And to write a redelegation sort of procedure, even in new gTLD land
that is different than that, I think without a policy discussion, is -
- or policy recommendations from one of ICANN's committees or
supporting organizations would be difficult for us.

But in the meantime, we think there's tools at government disposal
for ensuring that the TLD operates in line with its promises to the
government when the government gave its approval.

The government or its proxy could participate in the management of
the TLD in some way, sit on the board.

The government, as a prerequisite to giving the approval, could
require a community-based designation, and then in the new gTLD
process, there is, as I said, a compliance mechanism for ensuring
that community-based TLDs act in accordance with their restrictions. 
So there's a dispute mechanism available in which the government
would not even have to participate.  And then finally, the government
could set itself as the inheritor of the registry in the event of
termination.

So we think there are mechanisms that are available to governments
to ensure that an approval of a TLD -- you know, the good aspects of
that will outlive the moment of approval.

Sorry for my voice.

Wait, where did you go?  Okay.  Go to the next slide.

Thank you.

There's been -- This isn't too much of a slide, but there's been a
lot of discussion about opportunities for new entrants, and
especially small entrants.  So the guidebook is intended to have
criteria, you know, whether or not you get the new TLD, based on some
scaling.  So -- people are giving me candy and water.

So a small -- you know, a TLD that intends to be small does not have
to demonstrate a large investment in security or hardware, but,
rather, the guidebook says -- tells the evaluator to say does the
business plan and the technical plan match.  So if he has got a small
business plan addressing a small community and a small technical
plan, that makes a lot more sense.

So the guidebook is intended to facilitate opportunities for smaller
entrants.  And in fact, is written in such a way to kind of give
advice to the applicant as to how to pass.

And so it's meant as a guidebook.  It's meant as advice to those
applying and to guide them towards -- in the right way.

And then ICANN recently published a communications plan that is
intended to take away the opportunity for surprise.  So the goal of
the communications plan is to avoid a party saying, after the launch
of the new gTLD process, "Gee, I didn't know this was going on."

And this applies to those who would apply for a new TLD or object to
an application for a TLD.

It's not intended to be an advertising mechanism where ICANN is
encouraging applications, but just to educate and make aware, in
every region, that this event is occurring.

So to everybody in this room, it would be helpful to look at that
communications plan and identify regions, geographic regions, or
economic regions, or business segments that are not addressed in the
communications plan and bring that to our attention.

The plan needs to be honed, so there's more detail that will come
out.  But I think all our goal is to have this process be launched
with the cognizance of people interested in every region of the world.

Next.

Registry/registrar separation is not settled.  There was a very
specific proposal in the guidebook, given public comment.  We have
taken that out.  Now there is a range of potential solutions, meaning
to -- that's meant to indicate that the question is still open.  So
there's a very unrestricted model.  There's a restricted model. 
There's a couple in between.

None of those four, which are very terse in description, are thought
to be the final answer.  They are just meant to indicate that the
question is open and there's a range for discussions.

We recently had a Webinar on -- to put positions on the record, and
there will be a session tomorrow -- is tomorrow Monday?  Tomorrow, to
discuss this further.

It's intended that structural separation will always be there. 
There's a requirement for registrar accreditation, and the separate
entity of a registrar that we think brings real benefit to
registrants by having an entity be customer facing towards them.  So
that will remain.

The issue is about not structural separation but ownership or
organization separation.

And the process for closing this out depends on the discussion going
forward.

Next.

The objection procedure for morality and public order.

So what are the options for governments that might object to a TLD? 
Well, the -- one of the issues raised is that an objection requires
payment of a fee.  So while the fee is refundable, so at the end of
the day, a successful objection does not requirement the payment of a
fee to a dispute resolution provider, it still requires a deposit
which can be as difficult for a government to make as a payment.

Other options would include encouraging another to object, such as a
proxy, or to -- there's the -- there's an independent objector that's
more fully described in this version of the guidebook.  And for the
government to inform the independent objector of the objectionability
of the TLD.

And finally, I think this is a review, but the morality and public
order standards and the procedure, too, not just the standards, are
based on a survey of national laws and recognizing some commonalities
across all regions.  And then the policy and procedure is based on
international law practices.

So the parties that had significant input into this process are the
permanent court of arbitration and International Court of Justice,
and, you know, the International Court of Justice doesn't have an
opinion.  But we talked to the justices and the practitioners in
front of the International Court of Justice, and also international
dispute resolution providers.  So it's based on international law
practices.

Next.

Oh, this is -- we're getting near the end.

So how is ICANN going to evaluate -- looking forward to the end. 
How is ICANN going to evaluate applications?

So we are creating a function within ICANN to receive, evaluate and
report on applications.  So it's really important, one, that this
organizational function be transparent and effective.  And, two, the
process for developing this whole function be transparent, too.

So there's four efforts going on.  Processes are being developed,
you know, day-to-day processes for processing of applications. 
There's an application system that will be an online system that's
being developed.

The organization for doing this -- I'm going to show that you
organization in a minute -- and the facilities for it need to be
provided.  But facilities aren't -- it's not necessarily a capital
"F," it's still pretty small but adequate.

And finally, there's a number of evaluation panels, six or seven of
them.  It's important that those panels, panelists and panels, be
retained in a transparent way, and that conflict of interest are
carefully managed in the process.

So we have created a separate Web page for this effort, and that's
at the bottom of the page there.

And so somewhat anticlimactically on the last slide, if we can see
it, is a picture of what that organization looks like.

So could I have the next slide, please.

Yeah, we can't see it.  But there is essentially four major
functions of this process for evaluating applications.  There's a
customer service manager that will be the interface with the
applicants.  There will be an entity that manages the evaluation
panels themselves.  There will be an entity that manages the dispute
resolution panels.  And then there will be a coordination function.

So what's that mean for ICANN?  Well, there's -- so there's five
staff members in that box right there, the director of the office and
those four functions.

Most of the rest would be outsourced based on the extent of demand
and because applications will come, be addressed, and go, than to
shrink the organization back up.

So I think under coordination of applications, there will probably
be a staff member for -- you know, a temporary staff member for every
ten applications or so.  But then the other areas will essentially be
all outsourced.

So the management of all those panels, and the panels themselves,
will be contractors who grow and -- you know, numbers who grow and
shrink, depending on their effectivity and also the demand.

And then, finally, this office will be supported by other ICANN
operations, so there will be some dotted-line responsibility with
finance, the IANA function, information technology and legal.

So three -- I think three of those -- the present office has four
people in it.  Michael Salizar is the director of the office, and he
is building up the organization and the processes in anticipation of
accomplishing this work and retaining the panels.

So that -- as I said, anticlimactically -- brings to a close the
presentation and the information here.  So I would be really pleased
to answer questions.

>>VICE CHAIR DE LA CHAPELLE:   Thank you very much, Kurt.  And I
think your voice is just a testimony of the number of presentations
you have been making on this subject.

In general terms, thanks for the information you provided,
especially on a few points that are new, including on the economic
study, as you mentioned.  I noted as well the issue of post-
delegation geographic names has been incorporated as an issue, and it
will be interesting to see whether the solution that is being
proposed is considered appropriate by the GAC members who were
concerned with this.

Likewise, I noted with great interest, as you may understand, the
introduction of an opening regarding examining categories of how they
could potentially be useful.

Likewise, the root-scaling study will have to go through the RSSAC
report.  So it will not be transmitted directly, as such, and used as
such.

>>KURT PRITZ:   Right.  The RSSAC and SSAC take that report in to
inform them as to their report to the board.

>>VICE CHAIR DE LA CHAPELLE:   Okay.  So thanks also for keeping the
time.

So I'll open the floor for GAC members who may have questions.

I think the general outline is that we're not, if I understand
directly your presentation, we are not in a situation where this is
the final version and, after Seoul, this is closed and goes into
implementation.  I think the general message is the key issues that
are of concern for the different communities, be it on trademark, be
it for geographic names or so, are still being addressed.

So if you can open discussion, this is not the final version and
this is still a space for discussion.

Yes.  Go ahead.

Yeah, Norway.

>>NORWAY:   Thank you.

It's about the categories.  We realize that it is a big task
establishing different categories, defining the limit, setting the
terms and so on.  But on the other hand, Norway has come to realize
that it might be even more challenging to evaluate all the different
kinds of applications that would be sent in to you under the huge
category "open TLDs."  Because as we see it now, you only have --
after delegation, you have two categories, not three.  You have the
opens and you have the one with the community based.  You have the
category saying it's geographical names ahead of delegation, but not
after.

So we are of the opinion that establishing more categories might
also be opening the possibility for making terms and procedures more
predictable for the applicant and other interested parties.

Regarding when it comes to the redelegation procedure, I don't
understand it.  But you said it is solved in any way, because we
don't read it that way.

ICANN's response to the question we had about redelegation is that
there is nothing to prevent the government, a public authority,
conditioning the granting of their approval.  If the operator strays
from these obligations, the government can lodge an objection and the
decision-maker can order a registry to comply with the restriction in
the agreement or face sanctions.  And this is the community-based
gTLDs.

So I don't see if a geographical name is not a community-based gTLD,
then you don't have anything.

We think that -- Could it be possible to draw a parallel to the DAG
Version 3 model subsection 1.2.1 where you say ICANN can deny any
otherwise qualified applicant for a new gTLD if the applicant has
been judged by a court for fraud or anything else by the national
court -- could you use some of the same -- a ruling, a final ruling
from the national court on the agreement, the terms of the agreement,
could that be something you could use in the agreement with the --
your agreement with the registry, to say that you would act on such
decision?

>>KURT PRITZ:   I don't know.

>>NORWAY:   Thank you.

>>KURT PRITZ:   So -- Can I answer?

>>VICE CHAIR DE LA CHAPELLE:   Yes, go ahead.

>>KURT PRITZ:   Yeah, so I don't know.

So that section of the guidebook is intended to address applications
and not a post-delegation environment.

>>NORWAY:   I know.  But it might be possible to draw -- to look at
it.

>>KURT PRITZ:   Right.  And the parallel is that in the registrar
accreditation agreement, you know, if a registrar operator is
convicted of -- if a national court finds against a registrar, that
gives ICANN recourse to pursue -- to pursue compliance actions
against that registrar.  So perhaps -- I'm not sure what the
compliance mechanism is in the registry agreement, but perhaps that's
possible.

>>NORWAY:   Just a quick comment.  Because I think that would really
ease our mind, our government's mind, because they are still asking
us what if we give away this and this, what would happen afterwards? 
And if we could say that we can have an agreement and we can have a
court decision in Norway on the terms and ICANN will respect that
decision, that would be a good answer.

>>KURT PRITZ:   Okay.  So we'll look into that.

And the community-based TLD was meant to be one option.  Certainly
the government could have its own agreement with the TLD operator, if
it desired, as a mechanism for ensuring compliance with the agreement
afterwards, or the government could condition the approval of the TLD
on the government playing a role in the policy-making body for that
TLD.

So it was just minute to provide some options for governments to
pursue their goals.

>>VICE CHAIR DE LA CHAPELLE:   If I may interject here on a
methodological question.  

I think there will be a ceratin number of threads of issues that
will not be solved here.  I think the point you are making is to
indicate that the thread of post-delegation issues for geographic
names has to be continued.

Could we discuss just briefly how the different threads that we are
identifying -- for instance, it can be categories, it can be post
delegation and so on -- how can we organize the interaction between
the GAC and the staff for comments on the different threads in the
future?

How do you see, for instance, the discussion on categories or on
post delegation -- because post delegation is going to be sort of a
common thread in other kinds of requirements, for community or so.

How do you envisage to organize this?  Because the issues are
becoming more and more circumscribed.

>>KURT PRITZ:   Are you talking about a type of mechanism so there
can be --

>>VICE CHAIR DE LA CHAPELLE:   Yes, for the interaction.

>>KURT PRITZ:   That's a good idea.  Let me work with Max to work on
something like that.

>>VICE CHAIR DE LA CHAPELLE:   Because I think that within the GAC
we can identify the few threads, and the other communities can do the
same, that they really want to focus upon, so that the comments are
structured by threads instead of being just comments on the DAG 3, if
I can suggest.

Sorry, Julia and Italy, Stefano afterwards, then Mark.

>>DENMARK:  Yes, thank you. While recognizing the amount of work put
into this, the Danish view is that all comments submitted by the
staff still stands and we are now in a situation where security and
stability of the DNS has been questioned with the introduction of
DNSSEC, IPv4, v6, IDNs, and new gTLDs.  And we are assured of would
not happen earlier.

And on this ground, Denmark is very worried, and these issues should
be taken care of before new gTLDs are introduced.

Thank you.

>>VICE CHAIR DE LA CHAPELLE:   Italy, Stefano.

>>ITALY:   So a similar question, but let me turn in a slightly
different way.

We appreciate the enormous amount of work that you have been doing. 
So first I would like to receive assurance that we are in time for
the schedule, I mean because the hypothesis was at the end of the
first quarter next year, the first goal will be open.

Although we realize that, as you say, that there are a number of
problems that are still under study, like registry/registrar split,
like categories, like market analysis, like the question on security
and stability.

So the figures that were supposed to attract the community to apply,
the last I heard months ago was an expectation of something like 800
or so calls.  And then some, like maybe a couple of hundred
geographical names are going to these divisioning categories.

So looking at the security and stability study of SSAC and RSSAC and
the root server system, so apparently the -- there is an advice not
to start too heavy, let's say.  To have a limited number per year.

And also, I ask you here a specific question.  How ICANN is
organized to handle all the contracts with the new gTLDs?  Perhaps
there would be a realistic constraint that in any case will not allow
to enter with so many in every year, because if you have, for
example, 200, you have to sign a contract every two working days in a
year.  And, frankly, I don't think this is feasible.

So what about the implementation plan you can seek -- you can tell
us in the sense how -- when the process starts, you see 200 this, and
just in case that there are much more demands and possible good
applications that you should put in the root server system.

So this is the question, how you handle an overarching number of
applications, of good applications.

>>VICE CHAIR DE LA CHAPELLE:   Maybe if you allow me, we will take
the other questions, and you can answer afterwards in order to save
time.

Brazil, Vitor, I think you wanted to speak, and then Germany.

>>BRAZIL:   Thank you, Bertrand.

I would like to turn to the geographic names issues.  So I would
divide my question into two parts, the first one referring to first-
level TLDs.  One of the suggestions that were made by the GAC, if I
am not mistaken, in Mexico City is that there should be a sunrise
period in which countries would have the right to object to specific
new gTLDs without any cost.

I would like to understand if this advice was accepted or not.

And when it comes to second-level TLDs, I heard the suggestions that
were made here, but I have to confess that they have not made much
sense to me up to this point.

And this basically for one single reason.  It's a problem related to
jurisdiction.  I think we cannot -- we could not think of calling the
Internet an International Court of Justice to decide on matters
related to redelegation simply because ICANN -- neither ICANN nor
registries nor registrars are -- have international legal
personalities.

So the International Court of Justice would, if such an issue was
presented before it, would just, say, have no competence to make a
decision on that.

And the second is it has been suggested here that whenever a
behavior of a particular registrant does not meet national laws, then
there should be -- then the governments should or would have the
right to call them to their tribunals.  But the thing is we are
dealing with global issues.

Say a Brazilian thinks -- If the Brazilian government considers that
a certain second-level TLD is offensive but this TLD is not run by a
company established in Brazil by a registrar or a company or
whatever, or if anyone involved in this is not based in Brazil, this
enterprise or company that would be sued would not be responsible --
responsive to Brazilian law.

So I think none of the suggestions that have been presented is
really practical.

Thank you.

>>VICE CHAIR DE LA CHAPELLE:   Hubert, Germany.

>>GERMANY:  Yes, thank you.

First of all, I have a more organizational question.

Right now the predictability of the process is very important for
private companies if they want to invest in new gTLDs.  And that's,
especially for the newcomer, very important.

And so far we are a bit worried about the time that already has
occurred in this process, and we ask whether there is a new timetable
which, say, investors can rely on for the process.

The next one, I understand Brazil's position that always mentions
that we don't think an exclusive list for defining geographical names
is sufficient, and there should be some additional, yes, a
possibility to include additional names that are not on this list.

And we expected that in the new guidebook there was some solution to
this problem, but we didn't find any news on this.

One issue I welcome very much is the new communications policy, as
far as I understood, because we, in our country, really realize that
part of the community, especially the business community, is not
aware of the problems or changes that go ahead with the introduction
of new gTLDs.

The Internet-related industry, the registry industries, they are
both very involved in this discussion and even discuss it with ICANN.
But other parts of the industries are not aware of this issue.

We have, as ministries, therefore, arranged some kind of forum in
December to give a brief overview of this gTLD process in our
language, because I think it's also a language problem for many people.

And for the time being, thank you.

>>VICE CHAIR DE LA CHAPELLE:   Thank you.

I will take a following comment for this round by Netherlands,
Thomas.  And maybe, for the sake of the organization of the whole
session, unless there is a pressing question to Kurt, I would like to
close after this round of question -- okay, so Sweden and the
European Commission -- we'll have -- in order to keep at least one
hour among ourselves to further discuss the issue rather than asking
questions.

So finish this round, and then -- Thomas, Netherlands.

>>NETHERLANDS:   My remark is -- it's not so much a question; it's
more a remark and something to think about.

I'm very surprised that in the applicant guidebook there is no --
not even any reference to, let's say, withdrawal ex post of
delegation, let's say, taking out of the root.

That's something which I think in the last month has become very
aware.  I think also the IAB made a statement about the
unpredictability of the system if you introduce hundreds and
thousands and more.  There should be some mechanism to revoke
delegations and get them out of the root.

So, I mean, I respect fully the application in which everything is,
as much as possible ex (inaudible), let's say, objections procedures,
et cetera, not to get any threatening things into the root.  But I am
missing very much the same thing which Norway stated as a point of
view from -- maybe from national point of view, but also from the
point of view of, let's say, you really are granting a resource into
the root which could be potentially damaging.

So I'm missing, really, the mechanisms to revoke and to get
something out of the root afterwards.

Thank you.

>>VICE CHAIR DE LA CHAPELLE:   So if I noted quickly, administrative
scalability, the question of the sunrise and capacity to object for
first level, the new timetable, and revoking mechanisms.

Kurt.

>>KURT PRITZ:   So certainly stability and security of the domain
name system and the Internet is paramount in this.  And so given the
findings of the root scaling study, we expect the SSAC and RSSAC to
react to that and to provide guidance.

So pending that, I don't know what the solution to the issue is if
Stefano would get 800 applications and SSAC says 500 a year.  I know
that, you know, given the timing of the launch of the new gTLD
process, the time to process applications and then delegate them in
the root, that we are looking at, you know, a three times
multiplication of the root maybe in the next 30 months or so.  So I
think the growth is, even unbounded, is slower than anticipated.

Be that as it may, the issue that you raise is a very serious one,
especially if constraints are put on growth.  That would tend to
increase demand, and that would serve to make the problem worse,
because then there will be more requests for gTLDs.

So I think there's a logical process to this, is, first get the
report, and combine that with estimates of demand, and then decide,
you know, more that just staff, but decide with ICANN's technical
advisory committees how that growth can be accommodated and how the
process could be launched.

So security and stability is paramount.  I think that can't be said
too much.

With regard to sunrise periods, the proposal for an I.P. or
trademark clearinghouse that's been referred to the GNSO for
consideration includes as part of that an option for registries to
either have -- provide an I.P. claims service during the launch of
the process, or a sunrise procedure so that -- those were
recommendations of the Implementation Recommendation Team, the IRT. 
And that if we expect that those recommendations make it into the
guidebook, then there would be provision for a sunrise period, or,
alternatively, an I.P. claims period, as part of that.

As --  So as far as the International Court of Justice, I didn't
fully understand the question.  You know, I took that as pertaining
to the -- the objection process.

So what the --  And if that's the case, what the objection process
seeks to provide is a path.  In other words, in launching this
process, the -- us sitting at this table and all the other groups,
ICANN should anticipate problems with different applications, and
that whether or not the paths that ICANN provides are utilized or
not.  I think it's important that it's demonstrated that these sorts
of problematic applications could occur and are anticipated, and so
the best possible path to address those situations in an economical,
meaningful way is provided.

I think courts of competent jurisdiction can always be used as a
method for addressing similar sorts of complaints.  So parties can
always go to the courts in their jurisdictions in order to seek
redress.  And the ICANN process in no way seeks to obviate that.

So I probably did a poor job of answering that question.  And if you
want to reask it, that would be good.

With -- regarding the timetable, there's not one published right
now.  ICANN's published a lot of -- a number of them, as you know. 
And what they've demonstrated is it's very difficult to gauge or put
a timetable on public discussion.  And so public discussion needs to
continue to occur on several of these issues.  And you know several
of the remaining issues were highlighted in the discussion today. 
There's a path, we think, for bringing them to close.  But don't want
to forecast a precise timetable until the exact path is identified.

And the process for geographical names.

So the question was about having a process for adding new
geographical names to the list.  And as --  So as far as I know, the
process right now, since ICANN relies on published lists such as the
ISO list or a U.N. list, UNGEGN list, the process for ICANN would be
to have those authoritative bodies add names or take names off,
because it's better for ICANN to rely on an authoritative body than
be its own.

So I don't know -- and I will take the comment back whether there
can be a process for adding to a -- sort of a reserved names list,
whether it be at the top level or second level.

And then -- did you want to follow up?

And then, finally, with regard to the last question about
withdrawal, there are certain aspects of the guidebook that
anticipate the shutting down of a registry.  A registry is required
to escrow data in a very precise way.  And there's provisions for
transitioning data in the event that the registry closes.  There's
also the requirement of a financial instrument.  In the event that a
registry fails, there is a financial instrument in place to maintain
minimal registry operations for a period of three years so that
registrants are protected at least insofar as having notice that the
registry's going to shut down, and hopefully providing the time for a
successor registry operator to put in place.

So, to an extent, the guidebook anticipates that registries are
going to fail and put mechanisms in place for the protection of
registrants.  But I agree, it's probably not complete.

>>VICE CHAIR DE LA CHAPELLE:   Thank you.

Two last questions.  Maria from Sweden, and Bill for --  Did Brazil
want to come back again on that topic?  No?

>>BRAZIL:   Yes.  Yes, Bertrand.

Just to make it clear, the international courts of justice has no
competence to make decision except in issues involving international
publish institutions.  And this is not the case of ICANN.  And this
is not a case of any registry or registrar.

So I'm just saying that the International Court of Justice has no
possibility to make a decision or to even take into consideration a
problem regarding redelegation or top-level domains in any sense.

So I would like to reiterate the suggestion that was presented by
our co-chair that we seek to find a solution that resembles the one
that's been sought for intellectual property violations in post
delegation.

I think this is the only thing that I see that's viable in this field.

Thank you.

>>VICE CHAIR DE LA CHAPELLE:   Thank you, Vitor.

Sorry.  

So Sweden, Maria, and then European Commission, Bill Dee.

>>SWEDEN:   Thank you, Bertrand.

To start with, I would like to say that I fully agree with my
colleagues in Norway and Denmark, and also the question was raised by
Italy and Netherlands.  Because of the fact that we have the IPv6
development and also the fast-track IDN process, and also the signing
of the root, the DNSSEC, I think that those things must be taken into
account, of course, when we talk about the new gTLD issues.

And if I can say a few words about the root scalability report,
because that is also actually pointing out those things, which I
interpret like it doesn't say you have to stop anything, but you have
to do it in a certain speed that actually is related to the knowledge
we have, which might increase while time goes by.

So that comes to, actually, the question I have, whether the -- what
is actually the mechanisms for being able to tune the speed, maybe
pull it back a little bit or maybe go -- move faster, depending on
the situation, on the stability and the security?

>>KURT PRITZ:   Yeah.

>>VICE CHAIR DE LA CHAPELLE:   We'll take Bill and then -- do you
mind?

>>KURT PRITZ:   No.

>>VICE CHAIR DE LA CHAPELLE:   European Commission, Bill Dee.

>>EUROPEAN COMMISSION:   Thank you very much.  

And thank you for that presentation, Kurt.

Two main points.  One on objection procedures for governments.  I
note with some disappointment, actually, that there's still no
mention of the existing procedures under the bylaws for the GAC to
give advice to the ICANN board, which is an issue I raised in Sydney
with the board directly.

I still think, not least from a legal perspective, I think we need
an answer on why those bylaws aren't applicable in this place and why
there's a need for governments to get in line.

My personal view, actually, is the idea that we pay a deposit, which
we might or might not get back.  It's really not workable, actually. 
It's not salable back home as an objection procedure.  Effectively --
and I said this in Sydney -- this would mean that the GAC end up
being charged for giving the ICANN board advice.  I find that
slightly bizarre.

Second point, you referred to proposals that all gTLDs would have
thick WHOIS registries.  I'd be grateful, actually, if you could
expand on that a little bit.  I'm aware of the historical difference
between thin and thick WHOIS registries.  But what's the working
definition you're using?  Has some -- you might be surprised to hear
it has serious implications, potentially, actually, for jurisdictions
that have data protection and privacy legislation that apply to
registrants' data.  And it seems to me to be potentially a
significant development that we're, if I'm correct, going backwards,
actually.  I think that the original gTLD registries all had thick
WHOIS databases and then moved to thin ones.

So I'd be grateful, actually, if you could tell me a little bit
about what your understanding is of everyone having a thick WHOIS
database in terms of how the data would be processed and published.

Thank you.

>>KURT PRITZ:   Can I answer this question first before I forget it?

I understand your comments about the objection process and the GAC
not paying to object.  So that will have to be taken into account. 
And a mechanism for how the GAC can provide advice on new gTLD
applications.

I'm not a WHOIS expert.  My understanding is -- there are some
things I know and some things I'm not sure of -- prior to this, it
was optional for gTLD registries to maintain thin or thick WHOIS
databases.  So, historically, VeriSign has been a thin database, dot
com, dot net, and one other -- dot name? -- jobs, Dot Jobs, are thin.
Even though it was electable, all the other registries have
maintained thick WHOIS databases.  

The maintenance of a thick WHOIS database is not a publication
requirement, but it's a maintenance requirement.  It's defined in the
registry agreement.  All the data elements to be maintained are
listed in their registry agreement.  But there is a difference
between maintenance and publication which may or may not -- I don't
know -- address the privacy issue.

Also, you will recall the example of Telnic, who requested some
relief of publication of WHOIS data based on national laws.

That, again, they did not reduce the data that was maintained.  They
reduced the publication of that data.

So in this case, the guidebook requests that registries maintain the
thick sort of WHOIS database so that competent authorities can
request that information where they, you know, present evidence that
they can look at that information.

So I don't know if --  A mechanism for slowing down has not been
developed yet.  And as you can imagine, could be very tricky.  A
clear-cut way would be to charge $1 million for every application or
to auction TLDs.  But that's clearly against the policy advice that
the GNSO gave when they said, "Let's create new gTLD registries."

And you can imagine the problems with a sort of first come, first
served or prioritization.  So a lot of thought has to be given to how
that might be done.  But there's not a mechanism for that yet.

>>VICE CHAIR DE LA CHAPELLE:   Do you want to -- a comment?  I sense
you want to comment.  But maybe we keep it for the further discussion
among ourselves.

I'd like to -- Maimouna.  But that's the last question on this
element, 'cause we'd like to move --

>>SENEGAL:   Thank you, Bertrand.  I had one question about the fee.


I know that it is not the last version of the guide, but I think --
my question is, is it possible to have less than one -- $185,000 for
a fee?

If we talk about the community and developing countries, I think
it's a little bit expensive.  And also, if you talk -- if you take
the application, say 15 questions, you have to -- if you have to
answer to all of the --  Is it possible to have some support to do
that?  I --  At the beginning, I think it will not be easy for those
who are not very involved only to proceed.

Thank you.

>>VICE CHAIR DE LA CHAPELLE:   So I will close the questions here.

Kurt, if you can answer.  And I would like to join for the first
question that Maimouna asked.

Could it be connected with the discussion that might be opening up
on the notion of category articulation between fees and categories? 
Thank you.

>>KURT PRITZ:   Yeah, so this is a question that's been asked
several times.  And certainly ICANN's goal is to facilitate regional
participation, facilitate the participation of communities.  It's the
fond goal of the new gTLD program is to grow the domain name system
through this sort of increased participation.

The way we've answered that question in the past is to create a
mechanism for scholarship or forgiveness of fee in the first round is
somewhat problematic, because we don't understand the demand, the
types of customer, or a mechanism that is reliable for granting
forgiveness.  You know, what is the test for that?

Help in devising such a test that cannot be abused in some way would
be very helpful.  And certainly ICANN would seek to help those who
are trying to answer the questions.  And I'm -- you know, I'm just
thinking, you know, we -- we, not ICANN, but ICANN supports ISOC and
other entities in doing regional education in different areas.  And
I'm wondering if that educational process or those efforts can be
augmented in some way to address -- to help potential new gTLD
applicants complete their applications.

>>VICE CHAIR DE LA CHAPELLE:   Maimouna.

>>SENEGAL:   Sorry.  But on this guide, it's written that ICANN will
not begin its evaluation of an application unless it has received the
gTLD evaluation fee by time.

>>KURT PRITZ:   Yeah, that's right.  That's what that says right
now, that everybody pays $185,000 fee.  And up until now, it's been
difficult to develop a means, especially in this first round, for
providing some sort of scholarship for that fee that is reliable.

Maybe the solution for that is to jointly identify or work to
identify some sort of DNS foundation that can be set up to help
applicants in different regions who can't pay the fee but demonstrate
a viable interest in becoming part of it.

>>VICE CHAIR DE LA CHAPELLE:   Okay.  Thank you very much, Kurt, for
the answers to the questions.

It's not closing the discussion, by far, but we need to move into
the next stage.

Thank you for having come to speak to the GAC.

I'd like to ask Kristina Rosette to make the presentation on behalf
of the IRT and maybe to focus on the question of the elements that
the IRT report that have been integrated in the draft applicant
guidebook, as much as you can.

Thank you.

And I give back the chairmanship to the chair of the GAC, Janis
Karklins, who came back.

>>KRISTINA ROSETTE:   Thank you for the opportunity to come and talk
with you and present to the GAC the view of the IRT regarding the
trademark protections in new gTLDs as they stand today.  I know that
you have a very busy agenda, so I will keep this very brief.  And a
number of my colleagues from the IRT are here to answer any specific
questions that you have on specific components.

So where exactly are we now?

Well, ICANN has made a commitment to adequately address the
trademark protection issue.  The first most definitive articulation
of it that I believe we saw occurred in March in Chairman Dengate
Thrush's announcement of the creation of the board resolution
authorizing the creation of the IRT.

More recently, ICANN COO Doug Brent provided in testimony to the
U.S. Congress that we have not and will not allow new gTLD expansion
that does not adequately protect the trademark issues.

And, again, of course, most recently, we've seen in the affirmation
of commitments, paragraph 9.3, that ICANN will ensure that as it
continues to expand the top-level domain name space, the issue of
rights protection will be adequately addressed prior to implementation.

Unfortunately, however, as of today, the new gTLD implementation
plan does not adequately address the issue of trademark protection. 
And that would be true even if every mechanism that was currently in
the draft applicant guidebook were adopted.

What I'd like to do is just highlight some of the key critical
differences from the view of the Implementation Recommendation Team.

At the outset, it's important to note that the IRT recommendations
provided compromised trademark protections.  They were not an ideal
protection from the perspective of many trademark owners, but,
instead, were a compromise, a compromise that the IRT put forth in
the interest of ensuring that they were widely acceptable and
implementable and that the new gTLD process could move forward.

The IRT, which, as you know, was composed of a globally diverse
group of trademark owners, outside counsel, academic and contracted
party representatives, and, to my knowledge, is the first time that
ICANN has utilized such as expert panel in this way.

So what happened?

The IRT proposed the uniform rapid suspension system as a cost-
effective rapid and fair mechanism to combat rampant cybersquatting. 
However, ICANN staff proposes the URS as a best practices only.  If
the URS is not mandatory, it cannot solve the harm that it's intended
to address.  It must be something that is implemented in every new
gTLD registry.

Unfortunately, by making it a best practice, what ICANN is
unintentionally doing is, effectively, penalizing the good actor
registry applicants that will voluntarily adopt, because they will be
placed at a competitive disadvantage against the bad actor registry
applicants who won't.

In order to really --  And I can't emphasize enough how important
this particular mechanism is in the trademark community.

To the extent that there's any analogy to an issue that the GAC is
working on, I would say that trademark owners feel the same way about
the URS being a mandatory mechanism as the GAC feels about geographic
names.

So where are we is now.

The GNSO has been asked to consider the URS.  They've got to provide
a report by December 14th.  And we'll see where that goes.

The IRT-proposed post delegation deputy resolution mechanism
required a specific RALO for ICANN, to investigate contractual abuse
by registry operators and to thwart potential trademark owner abuse. 
It is true that there is a trademark-related post delegation dispute
resolution mechanism in DAG 3.  But it bears very little resemblance,
if any, to our recommendation.  And, in fact, in a crucial aspect, it
removes ICANN from the entire process.  In other words, ICANN is
saying, "Trust us.  We'll enforce our contracts."

And although the I.P. community will be the first to admit that
ICANN has made significant progress as of late in its contractual
compliance, frankly, based on the history and given the potential
scope, this isn't something we're happy and looking forward to
relying on them for.

The IRT also proposed an I.P. clearinghouse as a centralized
database of existing rights to decrease administrative costs of
rights protection mechanisms for all participants.  Trademark owners,
registrars, registries, everyone would benefit.  The current version
of the draft applicant guidebook does not contain the I.P.
clearinghouse proposal.  Instead, the GNSO is considering a trademark
clearinghouse.

The IRT proposed a globally protected marks list in response to
numerous requests in public comment by trademark owners for some type
of list.

The ICANN staff has rejected the globally protected marks
recommendation, despite not having completed promised quantitative
research that would allow us to assess the applicability and scope of
the rights protection mechanism.

So where do we go from here?  That's an excellent question.  I think
the outcome is very uncertain.

The IRT recommendations remain poorly understood.  The IRT,
unfortunately, was unable to brief the board in Sydney.  And, in
fact, the letter that the board sent to the council contains numerous
errors that underscored the extent to which the recommendations still
remain misunderstood.

The IRT remains interested and available to the GAC and its members
to achieve an outcome that adequately addresses the trademark
protection issue in new gTLD implementation.  So if there is any way
that we can be of assistance, don't hesitate to ask us.

>>CHAIR KARKLINS:   So thank you, Kristina, for this presentation.

I think we would listen now to Eun-Joo from WIPO with the WIPO's
views on protection of intellectual property rights on the new gTLD
round, and then we will take questions for both presenters.

>>EUN-JOO MIN:   Thank you for this opportunity to share the
perspectives of the World Intellectual Property Organization in
respect to the trademark-related issues for the ICANN's new gTLD
program.

A very brief introduction for those in the room not familiar with
WIPO.  It's an intergovernmental organization with 184 member states.
So basically covering all the countries, all the countries -- almost
all the countries in the world, dedicated to the promotion of
balanced and accessible intellectual property systems.

WIPO administers 23 treaties, building the international I.P.
normative framework.  And we are an observer to ICANN GAC.

WIPO has been addressing I.P. questions arising within the DNS since
over a decade.  We have undertaken two international processes to
develop recommendations addressing the interface between I.P. and
domain name systems.  And the first intent domain name process led to
what is regarded as a very successful -- ICANN's adoption of the
Uniform Domain Name Dispute Resolution Policy.

And we have also taken further policy developments in this area.

I keep confusing this.

The slides will be distributed, so I'll try to make the presentation
brief and skip some of the points noted on the slides.  The
arbitration and mediation center is a division within WIPO.  And in
the sector of domain names, we are involved in both policy
development and case administration.  And in the part of case
administration, notably we are the principal UDRP provider, having
administered over 16,000 UDRP-based cases involving over 28,000 names.

And we also provide dispute resolution services to 60 ccTLDs.

The following slides I have included just to paint a picture of the
current Domain Name System and why trademark issues are such a big
concern, both to the trademark community but also to the general
public.

You can see from this slide that we have had a constant and even
rising caseload.  In 2009, that's up to September, and we expect to
see around about the same number of cases as last year, which was a
record year.

This pie chart shows the areas of complainant activity.  And
basically, there is no sector that is spared from cybersquatting.

We cover different -- the UDRP cases are conducted in different
languages, in some 15 languages so far.  And we have had cases
basically in almost all gTLDs.

And of the 60 ccTLDs that we administer cases, to we have had cases
from some 35 ccTLDs.

In our opinion, it's a very present problem.  And if we look at one
estimate of domain name abuse that has been provided by MarkMonitor,
I think one of the interesting statistics on this slide -- I'm not
sure if it's legible from your end -- but the second line states
corporate domain name registrations and estimates that 90% of
corporate domain name registrations are for defensive purposes.

So if a brand owner owns some 10,000 domain names, and many brand
owners have such domain name portfolios, 9,000 of them are unwanted
names that they are registering for defensive purposes only.

This is another statistic provided also by MarkMonitor.  They have
estimated the number of instances of cybersquatting covering top 30
brands in the second quarter of 2008 as covering 428,000 names. 
That's about 14,000 cybersquatting names per brand owner.

And so if that brand owner were to file the very -- what is regard
as a very successful trademark protection mechanism, the UDRP, using
WIPO, it will cost -- it's $1,500 U.S. dollars per case.  And so if
they put one domain name per UDRP case, then that would add up to
about $70 million U.S. dollars that they would have to spend to
enforce that mark again cybersquatters.

I did a search -- We provide an extensive search facility on our Web
site, and did I a random search of domain names that were subject to
WIPO UDRP cases containing the brand Tamiflu.  And I see online
Tamiflu, order  Tamiflu, Tamiflu U.S.A., Tamiflu 75 milligrams,
Tamiflu shop, buy Tamiflu online, Tamiflu vaccine, Tamiflu symptoms,
buy Tamiflu online without prescription dot info, so on and so forth.
We have had some 87 cases including the word Tamiflu.  And of course
we would also have had cases including typersquatting of Tamiflu.  So
what do these cases involve?  Cybersquatters, domain name registrants
using these domain name Web sites to sell counterfeit products.  Fake
drugs, that's what they are used for.  And, therefore, this problem,
the trademark issue is very often portrayed as a concern for the big
brand owners; that it is a struggle between the David and the
Goliath, but it is not really so.  We are not really only talking
about the drug company, but we are also talking about everyone who
goes up to these Web sites and buys these fake drugs and gets even
sicker.  I think it is an issue that the governments -- it is an
issue that is very close to the public interest and that the
governments should really take a strong lead in.

So when WIPO conducted its first Internet domain name process back
in 1998, already we had learned that the potential of abusive domain
name registrations that lead to consumer confusion and undermining of
public trust in the DNS is real, and that any introduction of new
gTLDs be in a slow and controlled manner.

Earlier this year we issued a press release, again noting that this
is a watershed moment in the development of the DNS, and that it is a
genuine concern for trademark holders.

We have been engaging very actively with ICANN to provide different
proposals, and we have tried to follow some guiding principles in
making these proposals.

One, to safeguard observance of existing I.P. principles and laws in
the DNS.

To propose mechanism that a workable, efficient, and scalable in
system design.  But in particular, to strike a balance between
legitimate I.P. interests, practical interests of compliant
registries and registrars to minimize their operational burdens, and
also to protect the regional expectations of good-faith domain name
registrants.  And here I would like to emphasize again, because it is
so often misportrayed, that we are talking about legitimate I.P.
interests; we are not only talking only about big brand owners.  The
primary policy basis of trademark law is not protecting the big brand
owners, but it is orderly functioning of market and consumer
protection.  I think we have to really remember that.  That's what we
are trying to achieve when you talk about trademark protection.

We are very pleased to have seen a trademark-based pre-delegation
dispute resolution procedure included in the three DAGs.  WIPO has
been very active in that process, developing the substantive criteria
and procedural rules for this procedure, and the substantive criteria
define their origins in the WIPO joint recommendation concerning
provisions on the protection of marks and other industrial property
signs -- rights and signs on the Internet.  So on an existing WIPO
instrument.

We have also proposed a trademark-based post-delegation dispute
resolution procedure.  Our experience is that trademark abuse is much
more likely to happen and much more likely to be significant once a
TLD is delegated than when it's in the application phase.  And it is
intended as a form of standardized assistance to ICANN's own
compliance oversight responsibilities.  As Kristina mentioned, ICANN
has made great improvement in the recent years in its compliance
matters.  However, it is now dealing with some 20 registries and some
1,000 registrars, and that will probably increase greatly in number. 
And it would be much better if they set up a procedure, a workable --
a working procedure, efficient procedure, to ensure compliance
oversight.  To provide an administrative alternatively to court
litigation and to encourage responsible conduct by relevant actors,
thus enhancing the security and stability of the DNS.

And the type of behavior that we are trying to cover through this
trademark-based post-delegation dispute resolution procedure is when
a TLD applicant makes certain representations in the application
phase and following delegations uses that TLD for a purpose
unreasonably inconsistent with those representations.  For example,
apple growers obtains dot apple and they start selling computers with
that.  That would be one scenario.  Another scenario would be when
the registry operator turns a blind eye to systemic cybersquatting in
its TLD instead of adopting appropriate mechanisms to counter such
abuse.

And this scenario has again been misinterpreted and misportrayed. 
The critics of this procedure state that WIPO is proposing and that
the IRT is proposing, and even ICANN is proposing that registries
police each and every single domain name in their space.  That is not
at all what this is intended to achieve.  Not at all.

What this is intended to achieve is when there is rampant, rampant
cybersquatting in a space that the registry operator is in control,
and that the registry operator is obtaining financial benefit, and
the registry operator knows or should have known of the existence of
such rampant cybersquatting, then, in those instances, then the
registry operator should take reasonable precautionary measures. 
That's what this procedure is seeking.

And regrettably, some ICANN contracting parties, particularly
registrars, have been alleged to engage in contributory
cybersquatting.  There have been court cases, especially in the U.S. 
And also for engaging in UDRP evasion services.  And this is in a
system where we have around 20 gTLDs.  When that goes up to 800
gTLDs, we're not really sure how the ICANN Compliance Team expects to
cover that.

I will not go into the technical differences between the WIPO IRT
and the ICANN proposals.  I think just one thing to note, for the
Uniform Rapid Suspension mechanism that has been submitted to GNSO
for consideration, yes, we also have concern that it has been --
currently appears to be foreseen as a best practice and not as a
mandatory policy.  But even if it were to be adopted as mandatory, we
have some critical design-related concerns.  The URS foresees that
the remedy, the suspension be the balance of the registration period.
Most cybersquatting domain names are registered for a year and not
more.  By the time a URS complaint is filed, goes through the URS
procedure, there will only be a few months left in the registration
period.  So the suspension would be valid only for that three or four
months, and then it will go back to the pool of available domain
names.  They will be registered again.  And the brand -- the
trademark owners will have to file URS complaints again.  It's a
revolving door that we should probably not -- we should probably try
to avoid in the initial stage.

So what are our general concerns?  We have a general concern that is
-- I think that echoes those that were voiced by Kristina.  We're not
sure -- we're not certain that the current ICANN proposals suffice to
address the concerns expressed today by governments.  Some of you
have expressed very clearly these concerns, trademark owners, INTA,
consumer groups, health authorities.

We also get communications from health authorities regarding these
problems that are going on.

We believe that these mechanisms are expected to be built upon the
framework of existing I.P. laws, unfiltered, unfiltered by special
interests.  That the RPM should not saddle brand owners with a range
of unwanted defensive registrations.  Kurt mentioned the proposal to
have a clearinghouse and possibly a sunrise process, which would be a
very good addition, but that, again, if it only contributes to a
brand owner who already has 9,000 defensive registrations to having
90,000 defensive registrations, I'm not so sure that is the most
efficient way to go about it.  

Right holder protection mechanisms should not primarily serve to
facilitate implementation of ICANN's new gTLD program.  And the
adaptations that have so far been introduced risk undercutting the
intended effectiveness of the proposals that have been made.

And lastly, and this is my last slide, before -- I also wanted to
use this occasion to recall that protection of IGO names and acronyms
were included in the GAC principles on new gTLDs.  And we have not,
so far, seen any reflection of this principle in the Draft Applicant
Guidebook and therefore look forward to seeing something that
reflects this principle.

And this may sound self-interested coming from an IGO, but I would
like you to remember that I am not -- that WIPO is not only talking
about the protection of WIPO.  In fact, we are talking about the
protection of Red Cross, human rights organizations, World Health
Organization.  Every time there's a natural disaster, there is a name
of that disaster, like tsunami, Red Cross tsunami, UNHCR, et cetera,
et cetera.  And I think this is something that maybe this group could
also be interested in.

Thank you.

>>CHAIR KARKLINS:   So thank you, Eun-Joo, for this presentation.

And now the floor is open for questions.

United Kingdom, Mark, please.

>>UNITED KINGDOM:   Thank you, Chair.

We in the U.K. certainly placed a lot of value and importance on the
IRT.  It was a good initiative to bring expertise to this very tricky
area.  And though they had to work pretty fast within a tight
schedule, and that meant they had to limit some of the scope of their
work, which perhaps was unfortunate, but at least it, as I say,
brought expertise to this issue.

And the U.K. government, having received representations about the
concerns of brand owners and the trademark community from business
associations and so on, sought to provide input into this whole area.

There was a very good meeting in London with the IRT team that
provided a good opportunity for the team, and also Eun-Joo, I think,
was there at that workshop.

So that was a very good opportunity to get into this issue in some
detail and take account of the concerns of trademark owners and the
big brands who are faced with enormous potential costs in terms of
defensive registrations and so on.

So it's particularly disappointing to hear from Kristina that there
appear to be significant deficiencies in how the teams
recommendations are being implemented or the proposals to implement
them.

I have received representations about this.  In particular, about
the dropping of the GPML and the Globally Protected Marks List. 
Although that was acknowledged to be problematic in terms of criteria
and where you draw the line and so on, still, for the big brand
owners, the big global brands, that was seen as a valuable proposal.

So for the U.K. government, I certainly need some answers as to why
that particular proposal has been dropped.  And also the
transformation of -- from mandatory to best practice for the URS.

So I hope there will be an opportunity to hear more about how this
whole area is going to be addressed in a systemic way, which is going
to serve the business community as effectively as possible.

We -- It's another example, I think, of a major issue where there
seems to be a long way to go, and that is a bit depressing
considering that we have generally supported the principle of
expanding the domain name space.  But this and the other issues we
touched on earlier with Kurt seems to suggest that we do, indeed,
have a long way to go.

Thank you, Chair.

>>CHAIR KARKLINS:   Thank you.

I have France, Germany, and United States on my list.

France, Bertrand.

>>FRANCE:   Thanks, Janis.

I think the -- generally speaking, the introduction of -- the
prospective of introduction of new gTLDs has put on the agenda at a
higher level the whole question of trademarks just as geographic
names and the rest.

So a lot of energy has been devoted to addressing this issue of
trademarks for the upcoming new gTLDs.

The thing that is of interest is that, at the current stage, we have
a domain name space that is largely represented in the dot com space,
and a large number of ccTLDs.  The impact that the new TLDs will have
on those issues will be only progressively growing as they are
introduced.  But it is quite some years ahead.

My question, therefore, is can we draw any kind of conclusions or
lesson from the discussion you have had and the discussion we have on
the new gTLDs that could be applicable to the existing TLD space to
enhance the fight against cybersquatting?  Because let's be Frank. 
The cybersquatting issues that we are talking about, at the moment
they are not in the new gTLDs.  They are in existing TLDs.

Are we fighting them in existing TLDs sufficiently or not?  Because
all the examples that were given are not coming from the perspective
that new things are going to happen.  They are now happening.

So if they are so important, can we draw lessons from this
discussion to help move things forward in the current TLDs that will
help us see whether they are efficient or not?

If you see what I mean.

>>CHAIR KARKLINS:   Thank you.  Kristina, do you want to answer?

>>KRISTINA ROSETTE:   Sure.

That is a difficult question to answer because the IRT had a very
specific scope, which was in connection with new gTLDs.  And by
virtue of the time, frankly, that we had to do the work in, and the
representative from the United Kingdom is absolutely right.  If you
look at the report, there's a laundry list of things that we just
couldn't even start to work on.

Having said that, I think that some of the mechanisms, frankly, are
not applicable in the existing gTLD space in the sense that, for
example, the I.P. clearinghouse -- and again, this goes to kind of
the misunderstanding of some of the mechanisms, because this is
something that shows up in the board letter.  You know, the I.P.
clearinghouse, with regard to the administration of things like the
I.P. clearing process and the sunrise, only come into play for pre-
launch.

Now, the existing gTLD registries have already launched.  So those
would not have any utility.  Similarly with regard to the post-
delegation dispute resolution, existing registries already have
contracts.  So it would be -- you know, I think there would be some
possibility.  I think there would also be different processes and
procedures.  But I think many trademark owners would readily admit
absolutely that there's a problem now.  But when we see this 500 car
train coming barreling down the tracks at us even faster and faster
that's going to multiply the scope of the problem by 500, let's try
and make sure that we can go out of the gate with these new 500 with
mechanisms that will at least help us.  And then we can kind of focus
on what's going on now.

>>CHAIR KARKLINS:   Thank you.

Germany, Hubert.

>>GERMANY:   Yes.  Thank you.  Thank you to these presentations.  I
found them very informative and very helpful.

Trademark protection is, from the point of our government, one of
the crucial issues that have to be resolved before introducing new
gTLDs.  And, therefore, we are a bit concerned to hear now that I
think there is quite a lot of work to be done before we can really
accept the situation ICANN is going to propose.

From my point of view, it was very helpful to have you following
this discussion and informing us.

Thank you.

>>CHAIR KARKLINS:   Thank you, Hubert.

United States.  Suzanne.

>>UNITED STATES OF AMERICA:   Thank you, Janis.  And like my
colleagues who have spoken before, very much welcome the
presentations.  And thank you for taking the time to come and brief
us and keep us apprised.

I think Mark was being incredibly diplomatic when he used the word
"disappointed."  We, too, are disappointed and remain concerned that
insufficient progress has been made on such a key issue.

And note, I think it's regrettable that Kurt has left the room.  But
we'd like to throw a question out there.  At a minimum, there seems
to be a deadline discrepancy.  There are two different dates at work.
And so we're a bit baffled.  There is one deadline for comments on
DAG Version 3, which I believe is November 22nd.  And there is a
subsequent deadline for a response from the GNSO of December 14.

So that leaves a huge question in our minds as to what is the value
of submitting comments on DAG V3, which will include, obviously, the
need to address the inadequacy of the I.P. protection, when there is
a later deadline from the -- for a response from the GNSO Council?

So I'm not putting the question to our guest speakers.  But the
question really does need to go on the record, I think, to ICANN.

And I did want to thank WIPO for their presentation and for
reminding everybody -- because this is exactly where I think the GAC
is coming from -- the I.P. protection issues do relate to consumer
protection.  And it is in fact because of concerns about the level of
abusive behavior and malicious conduct in existing gTLDs that the GAC
is also as concerned as you are about the amplification of these
problems in new gTLDs.

Thank you.

>>CHAIR KARKLINS:   Thank you, Suzanne.

Any other comments or questions to our presenters?

So I see none.

So thank you.  Thank you, Kristina.  Thank you, Eun, for these
presentations.

And now I'm looking whether we have law enforcement representatives
in the room who would -- we have one.  Two.

So do you have presentations?  Or....

>>CHAIR KARKLINS:   So Robert Flaim is from Department of Justice of
United States.

And then we have representative from --

>> Swiss crime (inaudible).

>>CHAIR KARKLINS:   From the Swiss cybercrime (inaudible) unit.

So without further delay, Robert, you have the floor.

>>ROBERT FLAIM:   Hi.  My name is Robert Flaim.  I do work for the
Department of Justice in the United States.

And during this week, we're going to have a few sessions.  The first
session will be malicious conduct of the DNS and consumer protection.
We're also going to talk to the GAC and the ICANN board.  We are also
going to participate at the registrar accreditation agreement
discussion on Wednesday.

And we would like to echo some of the things that have already been
said here concerning new gTLDs.

Our representative, Suzanne Sene, mentioned it, and also Bertrand de
la Chapelle mentioned it, and the fact that we really don't separate
the issue of new gTLDs with the current and existing gTLDs, meaning
that there's a lot of abuse that is occurring on the current DNS
which we would like to see solved before we move forward with the new
gTLDs.

Like Suzanne said concerning current malicious conduct, we just
believe that will be amplified with the introduction of new gTLDs.

Obviously, we have no problem in there being new gTLDs.  We just
want to make sure that whatever gTLDs are out there and whatever is
on the DNS remains safe and secure and is not a haven for
criminality.  So we take the issue as one and the same, old gTLDs,
new gTLDs.  We want to make sure that it's one and the same, that we
have due diligence insofar as registries and registrars; that all the
data concerning registrants is validated; and that there's
transparency and accountability in all the data and all the
registrars and registries.

So I will let my Swiss colleague speak as well.

>> Adrian Koster:  Thank you for giving us opportunity to speak out
here.

I would like to stress again that it's not only about intellectual
property, but it's also about consumer protection.  Because the
consumers out there, they are not so high educated that they find out
about everything that's going on behind the scenes.  They just go
online, and they would like to be safe and protected.  So we have to
make sure that the whole DNS system is safe and secure for the
consumers as well.

Thank you.

>>CHAIR KARKLINS:   So now we have an opportunity to ask questions
to representatives of law enforcement agencies on their statements,
or make comments.

Canada, Heather.

>>CANADA:   Thank you, Chair.

Thank you for coming to the meeting today to raise your concerns. 
I'm wondering whether you had in mind any particular aspects of the
policies under discussion, notwithstanding your lack of distinction
between existing gTLDs and new gTLDs.  You know, how you might want
to advance the discussions so that your concerns are better addressed.

>>ROBERT FLAIM:   Well, we have actually -- excuse me -- we have
actually come up with a proposal that we are going to introduce here
at ICANN.  It has the support of numerous international law
enforcement agencies other than myself and Adrian.  We also have
support from the New Zealand Police, the Australian Federal Police,
the Canadian Federal Police, Japan and Korea are actually reviewing
it.  South Africa is reviewing it.  So there's many other law
enforcement agencies which are going to review this proposal.

And the proposal goes exactly to what I had mentioned, which is,
number one, due diligence insofar as registrars and registries, to
make sure that they are legitimate businesses that are not hiding or
advancing any criminal activity.  Number two, we want to make sure
that we have a public and accurate WHOIS that is in compliance with
national laws.  And, number three, we want to make sure there's
transparency and accountability in the DNS, meaning that anyone who
partakes in it -- resellers, registrars, registries, registrants --
are all who they say they are, and they're engaged in legitimate
businesses that do not further criminal activity.

So we will be discussing that at our DNS abuse session tomorrow,
Monday, at -- I believe the time is 5:30.  We will be speaking to the
GAC again with other law enforcement representatives here from the
United Kingdom, Korea's supposed to be here, Germany is scheduled to
be here, and -- am I leaving anyone out? -- and there may be one or
two others.  I apologize if I leave anyone out.

But we will be discussing those more at length at those sessions.

>>CHAIR KARKLINS:   Thank you, Robert.

Any other questions?

Singapore.

>>SINGAPORE:   Thank you, Chairman.

It is very gratifying to see that we have parties that are coming
forward like the U.S. Department of Justice to look at other aspects
of launching new gTLDs.  And we would very much like to welcome that.

And I think launching a new gTLD is not just an issue of
intellectual property protection, but there are a whole spectrum of
other issues:  The consumer protections, the cybercrimes.  So we very
much welcome, you know, the other parties coming forward and
addressing all these issues.  And we would certainly like to
participate any way we can.

Thank you very much, Chairman.

>>CHAIR KARKLINS:   Thank you.

France.

>>FRANCE:   Thanks.

Just one question.  Is there, in your experience, a relationship
between the degree to which a given registry has a range of
cybersquatting, phishing, or crime-related activity and the
registration policy that this registry has?  In particular, you have
two dimensions.  You have the Gs and the CCs:  The CCs are more
diverse in their registration policy, so maybe it's more interesting
to take that one.  Do you make a correlation, for instance, between
the degree in which the registration policy is monitored,
implemented, and enforced, and the degree of malicious conduct?  If
yes -- that's the first question.  And if yes, how can this be taken
into account when we discuss the new gTLD program if the idea of
categories is being introduced?

>> Adrian KOSTER:  It's always a problem of accountability and
applicable law.

And with a ccTLD, we -- every country can legislate on that.  So
they can decide on whether they want their national law applicable or
not.  But with generic top-level domains, there's no country law that
applies per se.  So we always have to look out who makes decision on
what is malicious conduct and who decides whether there should be a
retraction of a domain.

So if there is a good policy in place with the registrar and they
decide themselves, they always face maybe a tort law, in the United
States especially.  So there's always the question who decides or who
tells what's allowed and what is not allowed.  So it's difficult
sometimes to really address the right jurisdiction.

So if the registrars make policies and enforce them themselves, they
can counter abuse very efficient.  But if you first have to run after
which jurisdiction is it or who is -- who can make the decision on
what is allowed and whatnot....

>>FRANCE:   So if I understand correctly, one of the elements that
you highlight is that the more the registry has actually policies
that are clear and that they take the effort of enforcing them and
somehow have a sort of first level of responsibility in making sure
that the registrations they accept are good, let's put it this way,
the less the problem of fighting whose jurisdiction it is comes into
play when you want to fight it, that's the first.

Second question is a follow-up.

Is there a connection and a relationship with the number of
registrations, i.e., the larger the number of registrations in a
registry, the more likelihood that people who do cybersquatting,
phishing, and so on are going to target this one, a little bit like
there are more viruses on Windows than on Apple, not only because
Apple is or not better, but because the market is smaller.  So the
smaller the TLD, the less a target for phishing and cybersquatting. 
Is that a correct assumption?

>> Adrian KOSTER:  I'm not sure if I can comment on that.

>>ROBERT FLAIM:   Sometimes it's not.  Because sometimes you have
very small, be it a country code registry/registrar, or a small gTLD
registrar/registry.  It just depends on that particular registrar and
registry and what policy, as you were alluding to earlier, what they
have.  Because you can have a registry/registrar respective duties,
of course.  And if they have a good policy, it doesn't matter how
many they're processing.  A good policy is a good policy.

And I think, to answer your earlier question as well, we have seen
that there are registrars that do adopt good policies.  And that's
what we're trying to emulate.  And the question is, trying to get
other registrars to emulate the good ones so that there's a uniform
policy and everything is being done at the very first level.  Because
it's always better to be proactive than reactive.

>>CHAIR KARKLINS:   Thank you.

European Commission.

>>EUROPEAN COMMISSION:   Yes.  Thank you.  I'd like to add my thanks
to everyone else, actually, for your coming along here.  It's good to
hear from some law enforcement people.  Always good for us to have
contact with the real world now and again.

Yeah, I have a simple question.  Well, I hope it's simple.  But
where do you see most of the problems?  Do you see most of them with
gTLD names or ccTLD names?

>> Adrian KOSTER:  I think it's a -- it depends.  Because there are
several country code top-level domains which are merchandise.  And if
it's merchandise, then it's usually not the country's laws that
apply.  So you don't know what -- for example, on the TV domain, what
law is applicable on ABC.TV?  Because it's an American company and
it's on the TV domain, so it's a -- it goes -- the list goes very far.

So from the Swiss perspective, we are trying to adopt Swiss law to
our domain.  But on the other hand, anyway, let's leave registration
open to everyone in the world so everyone could obtain a dot CH
domain.  But if they abuse it, we need to be able to retract it.  And
this needs to be addressed by Swiss law.

>>ROBERT FLAIM:   From the American perspective, we see it more on
the gTLDs than the ccTLDs at this point.  I mean, there are some
ccTLDs which are quite notorious for bad behavior and engaging in
criminal activity or supporting it.  But for the most part, from what
we're seeing, it's gTLDs.

>>CHAIR KARKLINS:   New Zealand.  Frank.

>>NEW ZEALAND:   Thanks.  

Just following up from what you said, my experience is that if there
are contracts in place and those contracts are enforced by the -- as
you said, the registrar/registry relationship, then there shouldn't
be a problem.  The problem is one of enforcement of the contract,
which comes under civil law, generally.

So there's a joint responsibility upon the people who run the
registry to make sure that happens, and for the generics, of course,
that's ICANN.  So the suggestion may be that they're not enforcing
contracts properly.  But that's -- leave that aside for a moment.

Why don't you publish a name and shame list of the worst top-level
domains for criminal activity?

>>ROBERT FLAIM:   We're actually lucky enough there's a lot of
people who do that for us.  So we don't -- we don't do that.  But
that is out there, you know, through a lot of different groups.  

So --  And to go back to your contract question, I think some of the
problem is that those contracts with ICANN, and the RAA in
particular, are not quite as strong as we would like them to be and
as ironclad as we would like them to be.  So the problem that we've
actually heard from the contractual compliance people here at ICANN
is that they need, I guess, better language, better enforcement
language or language in which they can enforce more.

So we're actually trying to help them with that as well.  Or at
least concur with them so that they are in a better position to do
the enforcing that they need to do.

>>CHAIR KARKLINS:   Thank you for answer.

Any other questions?

While people are thinking, I would like to use this opportunity of
your presence here and ask the question related to affirmation of
commitments.

I am glad to hear that you are talking about cooperation between
different law enforcement agencies around the globe on developing
kind of joint approach or best practice initiatives in cooperating on
these sensitive issues.

We know that there are different -- differences in privacy laws in
different countries.  That certainly makes your cooperation much more
difficult.  And we witnessed that during our discussions on WHOIS
questions and developing GAC principles on WHOIS how diverse these
data protection laws can be and how they sometimes prevent law
enforcement agencies from doing the job the way they would wish to do
it.

But point 9.3 speaks about a review of existing policies related to
WHOIS, and the representatives of this review panel should be
selected also from the global law enforcement community.

And the question is, where should we look for global law enforcement
community?  What would be your -- your advice to us?

>> Adrian KOSTER:  Well, there's always the possibility to reach all
law enforcement agencies throughout the world through Interpol.  So
if you contact Interpol, they can distribute for you everything
throughout the whole law enforcement community.

>>CHAIR KARKLINS:   So are you suggesting that Interpol would be the
right organization to address -- to seek representation in this -- on
these issues?  Or they would guide us who would be the best place to
work on it?

>> Adrian KOSTER:  Well, certainly it's -- it's to mention that
Interpol is, first of all, a channel that you can reach all law
enforcement agencies.

>>ROBERT FLAIM:   I would say that since we're talking to the
governments of the world, I think it would be good for the GAC
representatives to reach out to their respective law enforcement and
know who that person would be in this particular area.

To be very honest with you, Internet governance and policy are very
hard for law enforcement, because it's not the -- not deemed to be
the sexiest thing.  Law enforcement likes handcuffs and putting
people in jail and things of that nature.  But we're trying to change
the mindset where, if we can prevent crime, then we're really doing
our job, as opposed to catching criminals after the crime has occurred.

So I think this is a very good place.  Adrian definitely mentioned
Interpol.  I think that's a very good start.  But I think with the
GAC is to reach out to their respective law enforcement agencies and
let them know that this is something that's very critical and that
their know-how and advice is very important.

>>CHAIR KARKLINS:   So thank you, Robert.  Thank you, Adrian for
being with us and answering questions.  And we're looking forward to
participate in the sessions where you will be doing presentations on
the document which I see, and you are hiding the text from me.

So thank you very much, indeed.

So we have now about 20 minutes to discuss issues.  And the question
is, do we want to do that or do we take a break before our meeting
with the GNSO, which will address two sets of issues.  One, new
gTLDs.  And I suspect that some GNSO representatives are already in
the room and they heard this discussion.  And another issue which
will be on our agenda is affirmation of commitment.  So the exchange
of views on this, how we see.  And certainly we will have difficulty
to speak in one voice.  At least I will not even try, since we
haven't discussed the issue.  But the individual interventions from
GAC members are welcomed.

So I -- unless there is a request for the floor right now --
Bertrand, please.

>>FRANCE:   Janis, are you asking whether we should devote 20
minutes to further discussing within the GAC the -- the DAG 3 and new
gTLD issue?  If that is the case, we don't have any other opportunity
to do it apart from this session.  And for the moment, we have only
had presentations and questions.  Maybe it's good to keep at least
one opportunity to exchange among GAC members before the session.

I take the opportunity of a personal remark.

I have left my watch on the -- on the table when I was chairing the
meeting before, and I apparently forgot to pick it up.  So I don't
know if anybody has secured it somewhere.  I'm here.

>>CHAIR KARKLINS:   Okay.

Then let us start this discussion.

So as you remember, the GAC sent the comments on DAG version 2.  And
in reply, we received a letter from the chairman of the board
reacting on our comments.

The Version 3 was published on the 4th of October.  And we --
intersessionally, we had a couple of conference calls where we
started sort of getting acquainted with the proposals in DAG Version
3 and analyzing how far GAC advice has been taken into account, or
GAC comments were taken into account.

So Mark volunteered to try -- volunteered to put our common thoughts
on paper.  And intersessionally, a few substantive submissions were
made on the GAC list.  And, namely, that was Norway, if -- if I
remember correctly, there was Germany who submitted some comments. 
But we are not at the moment at the stage when we have some drafted
text, consolidated drafted text.

So, therefore, I think we should follow the same pattern as we did
with the comments on version 2, that as a result of today's
presentations and a question and answer session, as well as our
discussion today and on Wednesday, when time will permit, we would
formulate our initial and most pressing sort of thoughts related to
DAG Version 3, and the more substantive comments we would develop
after the meeting online, not exceeding three, maybe four weeks.  And
we will submit them for consideration to the board and to the rest of
the community.

So this would be my suggestion on the methodology.  And I don't
know, Mark, whether you want to start this discussion, being the pen-
holder.

>>UNITED KINGDOM:   Well, thank you, Chair.

And I -- first of all, on your proposed methodology, I think that's
a very good one.  We haven't had much time as the GAC, individually,
to go through the DAG Version 3.  It's a long document.  And we have
had a useful digest of the main changes.  But we've only just
received that.  But that is a useful -- provides a useful stair for
colleagues to go into the relevant bits of the DAG, which relate to
the concerns that we've articulated in our correspondence with Peter
Dengate Thrush, your correspondence with Peter, and the reply we've
had from him and so on.

I think, gradually, we are moving forward.  But as we've heard
today, there are issues still very much alive.  And during the course
of this meeting in Seoul, we will, I'm sure, hear further on some of
those crucial issues.  So it is quite a dynamic process.  We all need
time.

Just going through my checklist of the issues we have raised, I
think we've managed, through very effective time management by
yourself and by Bertrand, to raise the live issues.  I think -- I
can't see one which we haven't raised.  So we've had some useful
engagement in the session this afternoon.

So what I would suggest, if there's a colleague here who feels
there's something we have overlooked, or maybe there's a new issue
that has occurred to somebody here, we flag that up now, or that GAC
colleague flags that up.

And --  But as I said at the beginning, your methodology, I think,
is a good one.  I would just urge colleagues, really, to have a
serious go at it.  And we're all under pressure on time in our
administration.  We've all got business portfolios.  And the AOC has
created a lot of work for us.  And there's the IGF coming up next
month -- well, two weeks' time, isn't it?  So a lot of us have been
very busy with that.

So it -- obviously, this is a crucial area of ICANN activity, and we
need to fulfill our role in representing our stakeholders, consumers,
as well as business and others who have a stake in the gTLD round.

Thank you, Chair.

>>CHAIR KARKLINS:   Thank you, Mark.  I have Norway and Netherlands
on my list, and then France.

>>NORWAY:   There's been a lot of talking about protection, the
strong protection of trademarks.  And we just want to put the pen on
the follow-up on protection on national interests regarding
meaningful representation of a country name or territory name, and
that the current formulation proposed for defining country names do
not cover the same terms that was in the DAG before.

So allowing a TLD that is a meaningful representation of a country
to be a gTLD will mean that you have to involve ICANN in the
international policy of a country.  And we don't -- Well, we would
like to have some more discussion on that.

>>CHAIR KARKLINS:   Thank you.

Netherlands, Thomas.

>>NETHERLANDS:   Thank you, Janis.  I think we had no time to
discuss it with Kurt, but I am still puzzled about the reaction of
ICANN in their letter about the question of country names which are
not on the ISO list.

Of course we had the example of dot Holland, dot Ceylon.  We have a
lot of possibilities there.  If there is an applicant for dot
Holland, now ICANN's answer and the staff answer is, okay, you can
object.  There can be a kind of dispute resolution procedure, and
then you pay up to 122K dollar, but you get it back.  That's more or
less the same answer we had before.

But my question goes a little bit further.  If, for example, there's
a dot Holland application for a name which is not on the list, I
would say this should be then referred to as real geographical
application.  And I think this root is not in the application
procedure.  So I have a couple questions about this and I am very
curious about how, for example, the ccNSO is thinking about this
because I think they will probably have the same kind of worries
which we have.

Thank you.

>>CHAIR KARKLINS:   Thank you, Thomas.

Bertrand.

>>FRANCE:   I'd actually like to place this in a slightly broader
context.  After the presentation of Kurt Pritz, my sense is that the
root scaling study is introducing a new dimension in the process of
the introduction of new gTLDs.  What the exact impact will be remains
to be seen.  There are elements of interpretation, and the RSSAC
report is going, actually, to be drawing the conclusions and
providing the final framework.

But irrespective of what the RSSAC report will produce, it is
obvious that the notion of a completely unlimited, right-now process
is not working anymore.  Which raises a few questions.  And I was
very interested to see that in Kurt's presentation, for the first
time, and in the document, the exploration or the little opening
regarding the notion of categories is coming in.  And this is a
positive evolution.

And the sense I get is that ICANN is somehow re-opening a little bit
the discussion on a certain number of issues, which is positive.

Regarding categories, I would like to make an analogy and share it
with people to see whether it can help us discuss those issues in the
future.

There was an analogy that was made during one board meeting by one
board member who basically said the new gTLD program is just like
building a very large plane, and until it is absolutely complete, we
don't launch it.  But when we have completed it, we launch it.

I would suggest that the analogy is probably not as efficient or as
useful as another one, which is an analogy with an airport.

What we collectively are doing is managing an airport that has the
purpose of allowing a certain number of planes to take off.  Some are
very small, two propeller planes.  They will be very small TLDs. 
Some will be very large; potentially, jumbo jets.

I would like to share with you that in most airports, you do not run
the same runways and same procedures for two propeller planes and
jumbo jets, because they have different characteristics, because they
have different security constraints.  They can have different impact
if they crash or if they behave badly.

So when we think about categories, maybe let's not get into trying
to define categories too precisely, but let's explore a little bit
further this notion of different runways and the analogy maybe should
not be pushed too far.  But as far as I am concerned, it is something
that helps me understand the kind of challenges that we have.

Two very quick final points.  I think the discussion we had earlier
is showing that there is a major element that has to be addressed
more thoroughly, which is what happens after delegation.

We have had this post-delegation element coming in and coming in,
and the discussion on law enforcement is highlighting also that the
behavior of the registry itself is a huge element in security and
stability.  Whether the policy is appropriate, whether the registry
enforces the policy is very important.

You can have exactly the same TLD.  Depending on which registration
policy it has, it can be dangerous or very safe.

Whether, with the same policy, it enforces it or not, it can be
dangerous or very safe.

So I think we should pay more attention -- and when I say "we," it's
not only the GAC.  It's the ICANN community as a whole -- to how
ICANN, the registries and the other actors are making sure that the
policy development for inside the registries on the second-level
registration and the enforcement of those policies are currently
taken into account, including, as Thomas was mentioning before, the
possibility of canceling the delegation if there is a clear behavior
that is not correct.

And the final comment is, I was a bit intrigued on one of the slides
to see the expression of customer service, which obviously is the
relationship with the applicants.  I would be more at ease if the
expression was not "customer" but something that deals with
management of relationship with applicants, because it looks a little
bit like a more business activity than regulation.

Thank you.

>>CHAIR KARKLINS:   Thank you, Bertrand.

I think we received some feedback also not only from the board but
also from the community on our comments on version two.  And
particularly on the -- as a categorization of the gTLDs.  And so I
hope that in our session, joint session with GNSO, we will hear more
about that.

I have Australia and Switzerland on my list, and then Italy.

>>AUSTRALIA:   Thanks, Janis.  My comments will be quite short.

There are, as we all know and as we have heard today, a very, very
large number of issues here, some of which we have talked to in some
detail, but there's definitely a lot more depth that we need to go
into.

I really think we need to think about some sort of plan, some sort
of structure to actually look at each of these individual questions
and have the time to examine them carefully.  Otherwise, I think
we're going to be a little bit chaotic in whatever response we come
up from it.

So I would suggest that we start talking about some sort of
structure, whether it's a structure based on the Peter Dengate Thrush
letter, which actually divided things into different issues, or some
other sort of structure.

I actually think we need to tackle this in a very careful,
structured sort of way.  Otherwise, it's going to be very chaotic,
not only for the people who have been in the GAC since the last
millennium, but certainly for those who are more recent as well.

Thanks.

>>CHAIR KARKLINS:   Thank you.

Switzerland.

>>SWITZERLAND:   Thank you.  I think we had a very interesting
session today with lots of new information, at least for those who
were not able to go through the Draft Applicant Guidebook and other
documents in full detail.

I think it's to be welcomed that ICANN has tried to incorporate many
of our concerns.  One of the problems I start to realize is the more
they try to incorporate the concerns and the open questions, the more
open questions arise of these attempts to answer these open
questions, which might force us to find other ways.

In that regard, I would also like to go in the direction that
Bertrand has explained.  And it goes back to what I -- what we
proposed during the last meeting.  The notion of the airport with the
different runways, and maybe introducing a category for geographical
names in a wider sense that might have to be defined maybe even in
different subcategories might be another solution.

Maybe that would help to get these new gTLDs rolling in the sense
that you might not be able to let all the different sizes of planes
take off immediately at the same time, but it might be one way to let
the smaller planes start first, where the damage would be limited in
case something goes wrong.  Or you might want to try a test phase on
a certain level, also taking into account the fact that there are
some doubts about the stability and security if you do not restrict
the number of new gTLDs, at least for a first test phase.

So there are a lot of questions that require new approaches, because
up to now, it has not, at least to me, not become more clear in the
last few meetings and the questions and answers that go back, the
concrete solutions that bring us closer to a starting point have been
found.

Maybe some thinking aloud in terms of working methods, it might be
wise for the GAC to create subworking groups that would deal with
different of these issues that we have been discussing now.

And one last point.  You have been mentioning a draft paper that
Mark and Bertrand have been starting.  Maybe it's because I have been
bombed with e-mails for many different things that I am not aware of
availability of such a paper.  In case such a paper would already be
available, I would be interested in getting it.

Thank you.

>>CHAIR KARKLINS:   Thank you, Thomas.  No, the paper is not
available.  It hasn't been circulated.  There was intention to
circulate paper, but the real life proved that it was not easy to
achieve, to do.

But the notion of categories comes up recurrently in our discussion,
and not the first meeting.

In drawing parallels with the IDN ccTLD process, we suggested that
there should be and we created a fast track for those are ready and
have a pressing need.

In fact, I'm curious to ask a question to GNSO whether the kind of
fast track for gTLDs have ever been discussed, whether that is
feasible or not for apparently those who are ready and waiting and
certainly are not controversial, to be processed and potentially
introduced if they are applicable.  I mean, if they are receivable.

I think that would be an interesting question to ask, at least from
my perspective.

Stefano, Italy, and then Sweden.  And then Singapore and Senegal.

>>ITALY:   Okay.  From the comments I hear, I realize that we are
going well over the content of the DAG documents, because we are
posing here a number of questions that are going to the real
implementation, as I ask to DAG in the previous meeting.

So I don't know if it is the proper approach to now make the plan
how to comment DAG 3, Version 3, and that has also a very short
deadline.  Otherwise, we should restate, maybe after these
discussion, after these encounters with the constituencies, with the
board and so on, and try to make a list of -- not too long a list of
overarching issues, and to start this kind of exercise, to try to
reach a common point in a short time possible, respecting the time
line.

And a comment on the -- of the split -- of the separation of type of
new registries, the categories.  So it is right to ask the GNSO if
they have a plan for selecting new strings that could be ready or
maybe even special categories.  So this is very important.  And it
was also impressing the preoccupation of IPR people, because, also,
they are objecting a large insertion of new gTLDs in a short time
because they have to gain experience and security that they will be
able to provide assurance to the users, in particular.

And if we look at the categories, if we have one category that is a
social utility, new registries, maybe these have not so great impact
on IPR, maybe.  But certainly, the business and the geographic have,
just to make an example.

And I realize that ICANN, up to now, has not gone too far on this,
and even in other aspects that were mentioned by Doug, like
registry/registrar separation, market analysis and things like that.

Thank you.

>>CHAIR KARKLINS:   You thank you, Stefano.

Singapore.

>>SINGAPORE:   Thank you, chairman.  I just want to speak on this
geographical names.  And we remain concerned that issues do not seem
to be adequately addressed by the ICANN, despite the concern
expressed by GAC.  And I recall that in our letters to ICANN board,
and we expressed our concern of protection of geographical names at
the second and the top-level names.  And we also point out that one
of the GAC's principles is particularly on this point.

So I think we will support we continue the dialogue with ICANN, and
we hope that there are some new initiatives that can bring to the
satisfactory conclusion the issues.  And basing on ISO 3166, this we
do not think is very comprehensive or adequate, and I think ICANN
should come out with more new measures, new initiatives to address
the issue.  And I think your latest remark on the fast track of new
gTLDs is an interesting concept, and I think we just -- we are
hopeful that this geographical names issue can sort of be speedily
and adequately addressed by ICANN.

Thank you, Chairman.

>>CHAIR KARKLINS:   Thank you, thank you.

Senegal.

>>SENEGAL:   Thank you, Chair.  I wonder if you can also talk about
the fee, because we support in our letter to ICANN supporting the new
top-level domain, we say that we support these new top-level domain
because it is a demand at the large scale, and it concern government,
community and so forth.  And I think this fee could be a handicap for
developing industry -- I talk about industry, not government -- who
don't -- who could not pay for this fee.

I think that we have to address this issue.  It's important for us,
because, for example, in our region, Africa, we have a lot of traffic
outside because of the domain name.  And if it continue, we cannot
talk about increasing access of Internet if we have to pay all the
bandwidths.  It's very important for us.  And I think we have to take
care of this.

Thank you.

>>CHAIR KARKLINS:   Thank you, Maimouna.  United States.

>>UNITED STATES OF AMERICA:   Thank you, Janis.  I just wanted to
concur with some of the comments earlier by our colleagues about next
steps, I think it was Brenton and Stefano.  And I wonder if it makes
sense for us to comment on DAG v3, actually, in view of the short
deadline and the fact that DAG v3 does not include all of the
protections that that community thinks it needs.  So there is already
a gap.

Currently, we have no way of knowing whether and how the results of
the root scalability study will be factored in, either.

So it just raises a question as to whether that's a practical
approach or whether, instead, we sort of try to set out a list of
issues that we think remain outstanding.  Because commenting on a
text that already -- we have already been informed doesn't adequately
cover certain issues just seems to be an exercise in futility.

Thank you.

>>CHAIR KARKLINS:   Thank you.  And Sweden is the last speaker on my
list.

>>SWEDEN:   Thank you so much, Chair.  And I very much appreciate
this discussion, I have to say, and the concerns and questions that
we apparently share among many of us.  So thank you, everybody, for a
lot of interesting things, aspects.

And I just want to say also that because of the fact that Kurt
answered my question about the speed, if you actually -- well, maybe
I even should exaggerate a little bit, like emergency stop. If you
start something you can't stop, or you can't adjust the speed, I
think we really need to be super cautious before starting that engine
or whatever.

And I also liked the Bertrand -- the similarities with different
airplanes and things.  Maybe  there could be other solutions.  And
once again, enough about this root scalability report, but I think
it's very, very important that it's taken into account.  Thank you.

>>CHAIR KARKLINS:   Thank you.  I think it was an interesting
exchange.  

There is Kenya.  Welcome.  You were not present in the morning, so
welcome to the meeting.  Please, you have the floor.

>>KENYA:   Thank you very much.

Kenya welcomes and supports the introduction of the new gTLDs and
would like to support Senegal's call to, while not ignoring all the
other challenges to be looked into it, would like to support
Senegal's specific request to look specifically at the reduction of
costs for these new gTLDs because of the specific reasons she has
raised regarding developing countries, specifically African countries
getting into this new era.

Thank you.

>>CHAIR KARKLINS:   Thank you.

I think this discussion revealed the most challenging points that we
still remain concerned about, and all the presentations on the issues
starting with protection of intellectual property rights, scalability
of the root, the categorization of new gTLDs, the potential harm in
consumer protection.

So all these issues remain of concern and need further discussion.

At the same time, we certainly would like to see new gTLD process
moving ahead and new gTLDs being put -- being applied for.

And I hope that this week's discussions on different topics,
starting with the next session with the GNSO, will clarify some of
the questions.  And maybe we will come up with some -- potentially
with some solutions to suggest to the board.

I asked Mark to keep track of this discussion and the proposed
language for the communiqué, and I would like to see whether there
are other volunteers who would like to assist Mark in drafting
communiqué.

And I see Stefano and I need another one.  And Suzanne.

So Mark, Stefano, and Suzanne.  So, please, prepare a proposal for
the communiqué for Wednesday morning, most probably.

Yeah, of course, you can.  And Maimouna will join you.

So thank you very much, indeed.  Now we are taking a short break for
about five minutes.  And if I may ask, if there are several
representatives from the same country at the table, if you could be
so find for the next session to leave only one GAC representative at
the table, that we can have, also, representatives of GNSO around the
table that we can have a dialogue.

Thank you very much, indeed.  Five-minute break.

You.

They told me the Secretariat has distributed papers related to
tomorrow's meeting.  So you should have a hard copy of terms of
reference distributed.

[ Break ]