ICANN Meetings in Marrakech, Morocco
GNSO Public Forum
27 June 2006
Note: The following is the output of the real-time captioning taken during the GNSO Public Forum held on 29 June 2006 in Marrakech, Morocco. Although the captioning output is largely accurate, in some cases it is incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the session, but should not be treated as an authoritative record.
GNSO Council Public Forum
27 June 2006
ICANN - Marrakech.
>>BRUCE TONKIN: My name is Bruce Tonkin and I'm chair of the GNSO Council.
The purpose of this forum is to basically cover these topics, which is review
policy development process, to remind people of what stage or what the
different stages are, talk about the work that's ongoing on policies for the
new gTLDs, then there'll be an update on WHOIS, then there'll be an update on
the policy work we're doing on existing gTLDs, and then, finally, an update
on policy work with respect to IDN strings.
Policy development process is probably as confusing to us as it is for other
people because we use all sorts of different names for the different reports.
But at a high level, we start with an issues report.
If the council thinks those issues are sufficiently important from a global
perspective, then we initiate policy development process.
At that time, we call for public comment on that policy development process.
Then, collecting together, I guess, all the input is usually a fairly long
period of discussion within a task force or within a committee.
And then at the end of that process, an initial report is produced, which can
also be called a preliminary task force report.
And that should contain all the recommendations that are under consideration.
That goes out for further public comment.
Then it goes out to council for a vote, and eventually to the board.
That's the process in theory.
The reality is that in some cases, we have broken the work on large policy
issues into phases, and sometimes we issue an interim report to get public
comment on a particularly contentious issue.
And we recently did that with WHOIS with respect to two alternative
formulations for the purpose of WHOIS.
So that was something that was taken out as just one item in a large policy
We got some public comment, and the council made a decision.
But nothing is going to the board until the entire report is produced, which
will probably be later this year.
So the first topic of today is talking about the work that's under way on new
We have a draft of the initial report, and that's available now.
And that draft is probably something that I think the first draft probably
came out in February of this year, around that time frame.
Another draft was produced about a month ago.
And another draft will probably be produced after this meeting.
But, ultimately, that will result in an initial report which will contain the
initial recommendations of the particular committee.
That will then go out for public comment again.
And then we'll have a final report, most likely in October.
And the aim there is to try and have material for the board to vote on in the
Sao Paulo meeting in December.
This is probably the fastest time line, believe it or not, of any of our
policy development processes.
So just picking up where we are on the different terms of reference under
this topic, one of the first terms of reference was the question, should new
gTLDs be introduced?
And it's probably hard to encapsulate this view on a PowerPoint slide.
But, essentially, I think many members of the committee thought as long as
their recommendations were improved in terms of selection criteria and
allocation criteria, they'd say yes.
We still haven't actually gotten to that point where we're able to determine
whether all those recommendations that people have will be supported.
But that's essentially the answer to that question, which, in other words, is
really a "yes, but."
Under selection criteria, we had a fairly in-depth discussion of this in a
face-to-face meeting, and I'm going to probably forget when this was -- but I
think it was in Wellington.
And here we identified some of the criteria related to different parts of the
process itself for applying for a new gTLD.
That included things that there should be application fees and clear
timelines and that the process for applying was clearly understood and all
I don't think there was any great contention around any of those topics.
Then there was a discussion around technical criteria.
And the intent there is that there would be some minimal base criteria.
And this would be a base set of standards such as some of the IETF RFCs and
probably the ICANN IDN guidelines.
And also a requirement to comply with ICANN policies.
Now, these specific standards haven't really been written down yet.
And what we would do as we produce a final report is, we would include,
effectively, a draft RFP in that final report.
And that draft RFP would actually specify the different base standards that
would be required to be complied with.
So then what would actually end up in our sort of final report is not just a
report on the policy recommendation, but, effectively, a reference
implementation of that policy so people can see how do these policies
actually translate into an actual RFP.
Further on collection criteria, the next area is, I guess, more contentious
and it probably breaks down into different elements.
There's how do you select the string.
How would you select purpose or charter for a particular TLD, how would you
select the organizational sponsor for a particular TLD, and what constraints
would you place on registrants.
I guess there's different views on this.
And it probably ranges from, ICANN shouldn't be involved in any of those
things and it's purely as long as the strings don't collide with each other
and that they are visually different, and that would be okay, that's one end
of the scale.
The other end of the scale is we should have something more similar to the
sponsored round, where we have people sort of having to show that they have
the support of a particular community, and we saw dot travel as an example of
one of those, where they needed to demonstrate that they had some kind of
support from the travel industry or travel community as part of their
And that there would be -- it would have a particular charter, and there are
mechanisms for ensuring that they comply with the charter.
Probably at the moment, I think in between those two extremes is this -- I'll
just call it draft, because certainly it hasn't been formally voted on.
But that applicants must offer a clearly differentiated domain name space
with respect to defining purpose.
And that's really very much open for public comment at this stage as to
whether that's a reasonable approach and whether the different extremes are
So here's some of the other proposals that were discussed in the committee
that are probably similar to some of the constraints of the sponsored round.
And these are some of the suggestions that applications for a new gTLD must
represent a well-defined community and registrants are limited to members of
Or a new gTLD must establish a charter that addresses a defined purpose with
eligibility criteria, and registrants must meet the eligibility criteria.
Or that there's an accurate verification of registrant eligibility, or
applicants must explain how the new TLD maximizes the benefits for the global
So we're very much, I guess, interested in comments on this general area, I
think, which is around where should ICANN lean?
Should it lean towards allowing any string?
Should there be restrictions on that?
And we've asked the GAC for their comments on that.
Should we restrict the purpose or charter?
Should we be requiring an organization to show they have some support before
they would receive a TLD?
And these are kind of trial court in addition to others that are perhaps more
objective related to technical, where that would be criteria that you could
sort of say yes or no to.
Other criteria, and the difficulty here has been to get the right wording for
And I think people should read the actual final report or initial report to
get a sense of where we're at on this.
This is kind of a summary.
That the applicant in terms of being selected would need to have the
financial and operational resources to meet the other criteria, being whether
it's technical or whether it's something to do with a sponsoring
This gets called various things.
Some people call this a business plan.
Some people call it an operational plan.
I don't think we've yet really got agreement on the right term.
And this will probably be another one of those things where we really need to
probably specify it in a bit more detail as part of a reference RFP so people
can understand it.
Because I think whatever single term we use will get interpreted in many
I guess the assumption here is that certainly early on, when ICANN has a new
policy for new gTLDs, there may be many applications at once, and so it's
assuming that they will probably be managed in some form of batch process,
which might have a start date and an end date for receiving applications.
There are probably different views of whether that start date should be the
day the board approves the policy or whether the start date should be
And then there are probably other views as to how long that period should be.
I've just thrown up something as an example that have been given examples if
you like, which is three months.
But it could be shorter, could be longer.
And then once the kind of batches have been looked at, then in future, the
process would be in order of receipt.
And this pretty much matches the way, at the second level, names are
registered when a new top-level domain is created.
So .EU has had a very similar process that was one that was recently launched
where there was some sort of process for dealing with the initial rush of
And then on an ongoing basis, it's much, because then you just registry names
as they're submitted.
So we have some sort of batch.
The question is how do you start allocating the ICANN resources to that
So the assumption here is that at the end of the batch, you've collected
applications for three months.
That, in general, you'd process those applications on a first come, first
served order, as long as those applications meet the criteria.
And this, presumably, provides a benefit to those that are more organized
Then the issue comes in, when two parties want the same string, that could
also, by the way, be same purpose, if we are requiring a purpose criteria,
and if you have that contention, how do you deal with that?
You could decide to simply come up with some process that doesn't really
involve trying to select which string -- which proposer is better than
You could either auction it or run some sort of lottery.
It has been pointed out that there could be legal implications for ICANN
doing that as a nonprofit organization.
And we'd need to understand that further.
And as others mentioned earlier this morning, presumably, you'd need some
special license to operate either one of these two options.
The other general approach is to do it via a general comparison, that would
be saying which of these applications is the best.
And if you are going to choose an application which is the best, you probably
then need to have some follow-up process to make sure that if someone's
asserted that best is because I've got the -- technically the best solution,
or best is cause I've got the most money or whatever you want to define
"best" as, then you check that they follow through on that requirement so
they can't just make statements up-front and, hence, win that string or that
purpose, but not follow through on it.
Contractual conditions is the final terms of reference.
And this is acquired from, I think, one of the recent -- I don't know how old
it is, but an OECD report.
And that is that you need to strike the right balance between certainty for
market players and preserving flexibility of the regulatory process to
accommodate the rapidly changing market, technological, and policy
So that the two extremes here, one extreme could be a new TLD operator is
given a new TLD forever and there's never, ever a check after that.
And that could also be that they're only required to comply with the policies
in place at the time that they got this TLD.
The other extreme is to say, okay, you need to renew your agreement every six
months, and any new policies that we create, you need to adhere to.
So that would be a case of saying, well -- it causes a problem for certain --
business certainty for those putting up a TLD.
So that's kind of some extremes there.
So the sorts of conclusions that have come so far is that there's a
preference for a prepublished agreement, in other words, the agreement would
be part of the RFP itself.
And that would be a framework agreement.
And if there's any material alterations to that framework agreement, there
would be some sort of public comment period before approval by the board.
Essentially, what we're trying to do is almost get away from the board having
to approve every new TLD, just as we don't require the board to approve every
new registrar, because registrars are basically signing a standard agreement.
And as long as that agreement hasn't changed significantly, the staff are
authorized to enter into that agreement.
The other thing -- some of the other topics were registry fees, and
currently, that varies by registry.
The view here is that probably not appropriate to fix a registry fee to be
the same for everybody, but at least needs to be some consistent approach,
and that approach would need to take into account regional, presumably where
the TLD is operated, economic, and business models.
There was a fairly strong view that a registry operator should comply with
new ICANN policies that relate -- this is during the lifetime of an
So let's say the agreement was, say, five or seven years, then during that
period, some things can change.
So there's an ability for ICANN to adjust its policies during the lifetime of
But there's a constraint in what policies can be changed that they need to
And these are basically extracted from the recent agreements, like dot jobs.
And I'll just summarize those very quickly.
But they're issues that facilitate interoperability, security and stability,
functional and performance specifications for the registry services, security
and stability of the registry database, registry policies relating to
registry operations, and resolution of disputes relating to the registration
On the topic of renewal, there was a view that there should be renewal
expectancy. And that term, "Renewal expectancy," again came from an
I can't recall exactly which one.
And the concept there was that operators should expect renewal of their
agreements, provided there had not been a material breach of the contract or
repeatedly failed to perform to the standard required in the contract.
And that there should be mechanisms to terminate the contract if the operator
has been found in repeated breach.
And so this was sort of getting over the concept that, okay, an operator
might breach something, and ICANN points that out, and says, "Please fix it,"
and they fix it.
But if a registry operator was doing that continuously, then they would need
-- ICANN should have the opportunity to terminate the agreement.
With respect to use of registry data, it wasn't clear on what the different
types of registry data might be and what the policy implications may be.
There was agreement with respect to personal data.
And that was that personal data, there should be limited use only for the
purpose for which it was collected of any personal data.
And the registry would need to define the extent to which personal data would
be made available to third parties.
So that's basically where we're at.
So I've got four topic areas, three main ones, really.
How do you select an operator, or select a new gTLD.
What do you do for allocation, given that more than one party might want
either the same string or might -- if we have a purpose requirement, the same
And what sort of -- what should be the main conditions in the contract.
So at that point, I'll really just open it up for public comment on the new
>>RON ANDRUFF: Sorry, Bruce, you caught me asleep there, Ron Andruff,
I wanted to say that it's -- as you pointed out, this is probably one of the
most rapid things we've seen, and in the five years that I've been involved
with ICANN and bringing forth the dot travel initiative, this is amazing to
see such speed.
So it's really critical that we get this right.
And so I wanted to bring a couple of points forward to the council's
attention so that as these drafts get prepared, we can make sure that we
understand certain elements.
One point I would bring out is that generic top-level domains, dot com, dot
biz, dot info, dot net, I think that's brought forward the opportunity for
I think it's quite clear that someone can choose to take a dot com or a dot
biz or a dot info at their choice.
And the idea was competition.
And that's why new registries were initially initiated.
The idea of a sponsored domain is very important, because the community
itself can come forward and look for a way to resolve a community need.
In the case of dot travel, it came forward because the global travel and
tourism industry wanted to combine itself and elevate itself out and declare
itself as being a dot travel as opposed to dot whatevers.
So this community is also very well known to itself.
And what I mean by that is that it can do things that a generic top-level
domain today can't do.
And the first of those is the idea of a pristine WHOIS database.
Every dot travel registrant is reviewed by a human being.
In fact, it's reviewed twice, once by an authentication provider, which is
often the Travel Trade Association that reviews that individual who's looking
to register a domain name. And then it's also reviewed by our staff at
So what this means is that we know the exact street address, name, and so
forth, all of the data that we're looking for in the WHOIS, and it's been
checked twice by human beings, so we can take comfort and confidence in that.
The second aspect is, we can bring forward new services in the sponsored
space which serve the community as a whole as opposed to what we just
experienced in this last discussion, which was very valuable.
But the idea that domainers are smart guys that can game a system and put
money in their pockets while the community as a whole really doesn't benefit.
In the sponsored space, all registry services, new services, such as the
directory dot travel just coming forward, benefits the community, and as
such, benefits the Internet users on a global scale at the same time.
And, finally, with regard to sponsored top-level domains, it provides what we
can view as the logical expansion of the name space.
So dot XXX, all the porn is over here; dot travel, all the travel is over
here; dot finance, all the finance would be here.
Those kinds of things are things that Internet users today, tomorrow, ten
years from now, 50 years from now will really understand.
They're intuitive, logical places where one would find information.
And I think one of the other points I would make as a result of this last
session was the fact that users are using the browser bar for search much
more now than one would have thought before.
When search engines first came up, people were going to search engines to
Now, people are doing what some refer to as the grandma research.
They type in "Paris.travel" or "Hiltonhotels.travel," they try that first
before looking at a search engine.
Therefore, the intuitive name space becomes all the more important.
If we can start to work in that direction, I think we'll be serving both the
Internet and Internet users as a whole, and again with a very long-term view.
Adding more generics I don't think at this point builds more competition in
the name space, and neither does it really serve the industry, because of the
WHOIS issues and all of those things that come with it.
>>BRUCE TONKIN: Ron, to summarize, you're more or less supporting these
approaches on the screen now?
>>RON ANDRUFF: Absolutely.
That would be the only way to go forward, as far as we're concerned.
>>DIRK KRISCHENOWSKI: Hello.
Dishing Krischenowski from dot Berlin.
We want to summarize some thoughts we have published a few minutes ago on the
ICANNWIKI, on our Website, dot Berlin, on the ICANNWIKI.
I would say we would fully support the tremendous work the GNSO has done so
far, and we endorse the GNSO efforts to complete the PDP process as soon as
And so we see our following statement as a proposal illustrating how some of
the current concerns with items you presented in the last minute can be
shaped into a process.
But let me start first with the community TLDs.
Today, there are only a few organizations visible that expressed their wish
to get a known TLD.
These organizations mostly represent natural communities with a cultural,
linguistic, or regional background.
These TLDs are dot SCO for the scotch culture and language, dot CYM for the
Welsh culture and language, dot BH, the Brittany culture and language
community, dot GAL for the Galician community, dot LAC for the Latin America
community, dot NYC for the New York City community, and, finally, dot Berlin
for the Berlin community.
Especially community TLDs have the problem that their activities cannot be
treated confidentially, as the community needs to participate and the sponsor
has to fund sufficient support and resources in the public.
Since these communities will play an important role among the future TLDs,
applications, the GNSO should address the context of community TLDs with
One important point is the protection of community strings or TLDs against
misuse in the application process.
An adequate mechanism to protect certain strings could be a preevalluation of
applications and a preevaluation process.
Preevaluation is used very successfully in other industries, for instance, in
the pharmaceutical industry.
The main purpose of a preevaluation process could be to increase the quality
of the final application, to give TLD applicants consulting in their
application, to prevent misuse and speculative TLD applications, and to give
ICANN -- and that's very important -- more certainty in the planning
processing and the planning of resources for the next TLD round.
The preevaluation is not a preapproval.
It's just voluntary for the applicants.
I want to finalize my statement with two points.
One point is regarding the first come, first served.
We think that TLD applications should be accepted on a first-come,
first-served model, let me say, but in a differentiated way.
And we think that TLD deadlines should be three times a year, ideally at a
special day on the ICANN meeting, a special TLD applicants' day for example.
More details on our paper can be found on the ICANNWIKI Web site.
Thank you very much.
>>BRUCE TONKIN: Thank you.
>> DAVID FARES: Thanks, Bruce.
I'm speaking on behalf of the business constituency.
I don't think it's going to surprise anyone that my comments are going to
track Ron Andruff's, since he is a member of the business constituency.
But I will amplify them.
First, the business constituency supports a managed approach to the
introduction of new gTLDs.
And it's our assessment that it's too early to judge the success of the proof
of concept gTLDs in providing additional user choice.
In any case, we believe that an applicant for a new gTLD must assume the
burden of proof to demonstrate that there is actually a market demand for a
With this in mind, though, the BC supports the introduction of sTLDs over new
And any sTLD application should create differentiation, limit the need for
duplicate or defensive registrations, and demonstrate a clear and broad
support from the affected community.
Having said all of that, though, the BC believes that given the changing
demographics of the Internet, that ICANN should prioritize the introduction
of new IDNs over new gTLDs.
>>BRUCE TONKIN: Thank you, David.
Anyone else want to comment on any of these recommendations?
It's been pointed out to me that the presentation that I was working through
wasn't on the screen before.
But I will be mailing that to the council shortly, and it'll also be
published on the ICANN Web site just to sort of get a sense at a high level
of what the recommendations were.
Any other comments?
I guess the main areas of contention or where we are really seeking the input
on is one of this concept of whether we should have something that sounds
like sponsored or not. And the other area is perhaps -- and they kind of
relate, I guess.
So if you're going down the sponsored direction, presumably you're also
favoring a comparative evaluation for allocation method, whereas if you are
preferring something that's more open in terms of allowing people to suggest
new strings and new applications, then, presumably, you're probably leaning
more towards some more objective process for dealing with contention.
But just interested on getting feedback on those two particular areas that
have had, perhaps, the most division within the committee.
>>CARY KARP: Just ought to be pointed out that the issue of IDN appearing in
TLD applications in no way gets past the question of restriction that might
be applied to those domains.
So there's an unrestricted TLD and a sponsored TLD to be encountered in the
IDN realm as well.
So it's not one kind as opposed to two flavors of the other kind, the
>>BRUCE TONKIN: Wendy, go ahead.
Please also bear in mind that the scribes for this session are at the back
and when they are doing the realtime scribing, they put the person's name and
then what they have said.
So if you can just say your name before you speak, that would be helpful.
>>WENDY SELTZER: Thanks, Bruce.
I appreciate that the focus here is going to be on nuts and bolts of
allocation method and contract terms.
And I think that's appropriate.
I was hearing a little bit of leaning toward requirements that new gTLDs be
differentiated from existing TLDs or from one another.
And speaking on behalf of the at-large advisory committee, we have made
statements in the past that -- and continue to make those statements -- that
we don't even see a need for ICANN to prejudge the market in that way, that,
really, the fact of someone's coming up and saying, "We think this would be
an interesting business for us to pursue as a registry," gives us enough
indication that that's something that might be valuable to them, and to the
users whom they think would purchase names from them.
And from the user perspective, having more domain names available where we
can register undifferentiated things like our Web logs and our photo sites
and our video blogging sites and our E-mail exchanges and our bulletin boards
that don't fit into particular chartered categories of things that don't
require us to demonstrate membership in a particular group, but that are open
and accessible, and then memorable to somebody whom we want to say, "Here's
my Web log," is a valuable service that registries for totally generic new
TLDs could provide.
>>BRUCE TONKIN: Thank you.
Anyone else on the council panel?
No, go ahead, ray.
>>RAY FASSETT: Hello, my name is Ray Fassett.
I'm with the dot jobs top-level domain registry. I'm also a member of the
But I really would like to make some comments on a personal level, having
personally gone through an application process of ICANN's.
And my own personal opinion, again, is the number one goal to shoot for in a
new top-level domain round is objectivity of selection criteria.
I think --
>>BRUCE TONKIN: Ray, can I just comment.
Just speak a little closer to the mike, please.
>>RAY FASSETT: Sure, sure.
I said my opinion is that the number-one goal of any new application round is
for ICANN to get itself in a position of objective selection criteria, that
that should be the overwhelming goal to go after.
I can cite that, for example, in this most recent sponsored round, that the
sponsor evaluation team actually cited, at the end of the whole process, that
the criteria put them in a position of having to make futuristic and
subjective conclusions regarding the applications.
And even in the case of dot jobs, where I felt, you know, ourselves that we
did a heck of a job defining our community, having a very qualified
sponsoring organization technically, it was unable to pass the sponsor
And I think part of that is simply for the reason that the criteria was not
objective and extremely exposed to subjective evaluation.
So in any future round and where applicants put forth a great deal of effort
and a great deal of expense, I think the burden is actually on ICANN to make
sure that the criteria is as objective as it possibly can be, including the
fact that I'm not so sure that sponsors are absolutely required as part of
the -- of any next round.
I think dot EDU is an example of a restricted TLD, a guarded name space, if
you will, operated responsibly on the Internet, and is not technically a
So those are my own personal observations to offer.
>>BRUCE TONKIN: Thank you.
>>BECKY BURR: Sorry. I'm much shorter than Ray.
And I didn't talk to Ray before this.
So I didn't put him up to saying what he just said.
But I do want to second, third, and fourth that.
Having represented two applicants in the last round, both of whom had the
uncomfortable problem of, along with eight of the ten applicants, not passing
the sponsorship community or the sponsorship roles the first time.
The difficulty and the expense that was imposed on those applicants as a
result was overwhelming.
It was really a very large piece of work.
And in the case of XXX, is still ongoing.
I just want to raise one other question.
And it relates to Ray's issue about the restricted TLD versus sponsored TLD
versus generic TLD.
In the case of dot XXX, which did make a case that the board accepted, that
it had a sponsored community supporting it, that it passed the criteria for
being -- the evaluation criteria that was set out and still finds itself
without a contract or even the ability to realistically address the deficits
in a contract that ICANN itself drafted and told most applicants was
nonnegotiable, is another source of incredible pain.
But in the aftermath of the board's vote, Paul Twomey has said that the only
way that XXX comes up again -- which I think can't be right, but, in any case
-- is as a gTLD.
So the question I have is, you have a -- you have an applicant who meets the
criteria, who was determined by the board to meet the criteria, who was
willing to agree to a contract drafted, a contract of adhesion drafted by
ICANN itself and still can't get up there except as a gTLD, which we're not
going to have any more of, puts us in a bit of a fix.
And my -- what I would like to say, and this is, I think, true of all of the
applicants, ICANN needs to be more accountable with respect to the way it
manages the new gTLD or new TLD process.
>>BRUCE TONKIN: Thank you, Becky.
>>STEVE METALITZ: Thank you very much.
I'm -- just like to make a brief comment.
And it's just a personal view.
And it really seconds one of the things that David Fares said on behalf of
the business constituency.
And that is on the importance of prioritizing IDNs.
I think it would be very odd if ICANN were to launch another round of new
gTLDs and it was not -- it was opened only to people who had gTLD proposals
in ASCII script.
And even taking on board the views of the gentleman from dot Berlin, it would
basically be saying, well, if you're Welsh or if you're Galician, or if
you're Scottish, our door is open to consider your proposal.
But if you're from a region that doesn't use that script, then you'll have to
wait a little while.
So I hope that -- I think that ICANN's professed commitment to moving forward
on IDNs is very important.
And I think it could be undercut if new gTLDs were launched before we were
ready to also accept proposals for gTLDs in non-ASCII script.
>>BRUCE TONKIN: Thanks, Steve.
I might just comment on IDNs, since that's come up.
It is on the agenda.
And I'll just very quickly go to it.
I think the council would generally agree with you, and one of the things
we're trying to look at is how to almost fast-track the work on IDN so that
it can be almost done and concluded at the time we conclude the new gTLD
work, specifically related to policy as opposed to any of the technical
issues that might arise.
Where we're at with IDNs is we have a draft issues report.
And we formed a GNSO working group to consider that report and liaise with
And at this meeting, we also met with the GAC.
And part of the difficulty at the moment is trying to work out how you split
the work of the GNSO that relates to domain names that are not ccTLD-related,
and what work the ccTLD group would do.
There was further discussion on the ccTLD side, and I -- in the IDN workshop
So I won't try and explain that.
It's not my place to do so.
But certainly from a GNSO point of view, I think the view is that we would --
it's not -- other slides -- the view is that we would work on IDNs using the
same framework as we are working on gTLDs.
In other words, the headings would be the same.
What is the selection criteria for IDNs?
What is the allocation criteria?
What -- you know, should it be sponsored or unsponsored?
All of those things.
But, basically, all the same.
So the only difference is the string.
The string does raise some other complexities.
For example, if you are going to go down the line of saying that you want
TLDs to be separated by purpose, can you really determine whether an IDN
string in another script, in another language is not just another -- is a
translation of travel, you know, like staff may not have that expertise, as
So does add some more complexities.
And I think what we would most likely end up doing is forming a focused
working group -- I'm sorry, focused task force within the GNSO that sort of
looked at all the stuff we're doing on new gTLDs and said, what are the
special cases that IDNs introduce that we need to consider at the same time
and try and, I guess, get those recommendations synchronized.
So I think it's a good objective.
Whether the IDN recommendations come out exactly before or after the new gTLD
ones, I don't think we can comment on at this stage.
But I think as an objective, I think that's reasonable.
>>RON ANDRUFF: Ron Andruff, again, Mr. Chairman.
You brought up a very interesting point.
And I know that IDNs is later in the conversation.
But specifically to what you just said, in this sponsored space, that's the
concept that I was trying to draw out, that a community -- the community is a
There's little doubt that the travel and tourism industry is one industry.
So to have a -- an IDN that would be in travel would be kind of ludicrous.
Here it would be dot safar for Arabic; it would be dot viaggi for Italian,
and so forth.
That's all part of the travel and tourism industry.
Translating the domain name itself into several different languages is a
natural and a logical way to go.
So again, that's one of the things that happens in the sponsored space that
you don't have to worry about in a generic space where you start trying to
separate out, okay, let's take this name, this construct, and I'll use the
construct of dot net, and take that construct and try to figure out a
translation for that in Arabic and in multiple other languages in the world,
and then decide who might run that construct.
So, again, the sponsored space creates a space where you can translate into
every language and this community is well served in that regard.
If I were to use the dot travel example today and the director (inaudible)
it's available in 11 languages.
Someone can load their information in Mandarin, and someone in Portugal can
read it in Portuguese.
So the whole point here is the IDN and the sponsor community really comes
together in a very nice fit.
So I just wanted to comment on that.
>>BRUCE TONKIN: Thank you, Ron.
Any other comments on the topic of purpose or allocation for TLDs?
>> Can I make a comment?
>>BRUCE TONKIN: Yes, go ahead, Sophie.
>>SOPHIA BEKELE: I think I've lost my voice here.
If someone can hear me.
But this is sort of in response to Ron.
I think there are two points that you made which are very, very important in
your first comment about competition, the dot com, the dot biz, and all the
ones that have already been issued that had created competition, and you
appreciated the economic value of that.
And in light of what you just made mention of harmonizing the sponsored TLD
community with the IDNs, I think there is a lot to be said about that
And I can understand the philosophy for what you're saying.
But I think we, as working group, we are also struggling with that.
And just to communicate to you, it deals with the issue of competition again
when we're trying to issue TLDs on the IDNs.
So just the same appreciation you had in the beginning, it would be a concern
on this side if we feel that we have to issue TLDs for IDNs and try to merge
it from the philosophy of, you know, using the sponsored environment as part
So I just thought that it's sort of -- sort of conflicts with each other.
But it's something that we're working on to satisfy, I suppose, the user
community and everybody else.
I don't know if that makes sense to you.
>>RON ANDRUFF: No.
Certainly, it brings up, again, one of the -- the complexities of the task,
So it's good that you brought that out.
But I would counter that by saying that if someone were to have multiple
lingual capabilities and they happened to be sitting in Tokyo and they typed
in in Kanji characters Majestic Travel, or majestic.travel for a travel
agency that happened to be in Hong Kong, the next time they came back to the
net, they'd just -- because they hadn't thought, did I do that in Japanese,
did I do that in English, they type in in an English character set,
"majestic.travel," and that goes to a tour -- travel agent in London.
So you see the conflict -- the opportunity for conflict and collision is
enormous when you start to look at the sponsored space and say, well, we can
sort of divide that up and provide for other operators to run it in other
It really is one community, and that community speaks multiple languages and
is addressed in that one space.
So that's what I was trying to draw out.
But I certainly appreciate the complexities that you brought out.
>>SOPHIA BEKELE: I agree.
I agree with what you say.
>>RON ANDRUFF: Thank you.
>>BRUCE TONKIN: Okay.
Well, the next topic on the public forum, then, is the policy development
we're doing on contracts for existing gTLDs.
I just want to make clear that this policy would really only apply if an
existing operator sought to change the agreement, or if a new entity becomes
the operator for that gTLD.
So that's fairly limited in scope.
And at that point, I'll hand over to Maureen Cubberley to give an update on
>>MAUREEN CUBBERLEY: Thank you, Bruce.
Yes, that is, indeed, the premise upon which the work is based.
As far as the process goes, initially, I'll give you some information on
There is a draft preliminary task force report that's available now.
We met on Saturday here in Marrakech and decided that we would go ahead and
seek further expert advice in specific areas that deal with specific parts of
the terms of reference.
Some of these include competition law, price regulation, definitions of
market power in other sectors.
We have studied some things in other areas.
We're trying to determine where the applicability lies for this particular
piece of work.
The preliminary task force report will be completed in October of this year.
And at that point, it will be available for comment as well.
I'd encourage additional input.
We have had input from the constituencies.
We have had very limited public input.
So we would encourage you to take advantage of that opportunity to provide
some input to us.
Just quickly, on the areas of work, we are examining areas where the
consistency would be useful.
Bruce has outlined the work that's happening with the new TLDs, and clearly,
we're looking for convergence of approaches.
The terms of reference that we have, and there are six altogether, the first
deals with registry agreement renewal.
You can easily find those on the site.
If you have questions, I will tell you the specifics of them.
But the first deals with renewal.
The second is with the relationship between registry agreements and consensus
The third is policy for price controls and registry services.
And the fourth deals with ICANN fees.
And, finally, uses of registry data.
So you can see that the slide that's in front of you pulls out of those six
terms of reference the areas that we've identified so far where consistency
would be useful from our perspective.
We will be meeting in the -- between now and the meeting in Sao Paolo.
Our timeline is to complete the preliminary draft report, and we do hope to
move the process forward significantly by the Sao Paolo meeting and again see
where the opportunities for convergence are between this work and the work of
the new TLDs.
We tend to work in a group on these policy development processes and then
break off with the specifics that are relevant to each area.
So that's all for the update, Bruce.
Thank you all.
>>BRUCE TONKIN: Thank you, Maureen.
Are there any comments or questions about the work on existing gTLD
The next topic covers WHOIS.
I'll just get some introduction stuff there.
There's been quite a bit of discussion already about WHOIS this week.
And so I'll just very quickly, just setting the framework that the objectives
are to improve the effectiveness of the WHOIS service while taking into
account the need to ensure privacy protection for the personal data of
Some of the things we're looking at is access control and processes for
At the GNSO Council level, as reported in Wellington, one of the items that
we pulled out at the council level earlier on was that the task force had
been stuck for several months on agreeing on a definition of WHOIS, and the
task force provided two options to the GNSO Council.
The GNSO Council selected one option so the work with proceed, after
receiving public input.
And we went out for separate public input on that particular point.
So without trying to debate whether that was a good or bad purpose, we have a
And at that point, I think I will hand across to Avri to give an update on
>>AVRI DORIA: A good little bit of exercise.
So this is a WHOIS task force report prepared by Jordyn Buchanan, who is the
I am actually the newest member of that task force.
And it was interesting listening to the new gTLD work be talked about as the
This, I think, is an exemplar of the longest.
I actually believe that I'm a third-generation member of this task force.
So going through the terms of reference, basically, determine the purpose of
the contacts displayed in WHOIS, determine what data should be displayed for
public access in WHOIS.
Determine how to access the data not available for public access.
And improve the process for notifying registrants of inaccurate WHOIS data
and improve process for investigating and correcting the inaccurate data.
So those are basically the goals that are being worked on.
And those really frame the work that needs to be done.
And that's the work that has sort of been enabled by having a purpose to work
The timeline has been shaped -- It's actually a time for optimism in terms of
getting this work done.
There is actually a feeling that it is moving along.
And in October, the preliminary report, including policy recommendations,
will be released.
November, there is -- there's an intent for a longer-than-normal comment
There's just so much interest in commenting.
There's the issue of dealing with the governments in GAC having some chance
So there's an intention of a longer comment period.
So, basically, in November would be the end of the comment period.
In December, the final report would be presented to the GNSO Council to act
on, with the intention that in early of next year, the GNSO would be able to
pass the recommendations on to the board, having gone through its whole
So there's a time line with, actually, an expectation of conclusion.
So there's ongoing discussion in each of the remaining terms of reference.
And, basically, there's an attempt to sort of create a comprehensive approach
as opposed to putting out a solution on each of the individual items or doing
a short report on each item, because there's very much a linkage between the
The -- the task force is basically using a proposal called the OPOC, or
operational point of contact as its template to guide the discussions.
Now, this is just a template.
It hasn't been adopted by the WHOIS task force as a working document, as
It is a template.
All elements of the proposal are subject to ongoing discussion, and there is
a lot of it, and potential changes.
As I said, the proposal has no formal status other than a starting point for
So, basically, a little bit on the OPOC proposal.
And I put a tiny URL up there.
Jordyn and I did, because it was really much too long.
But it can be found going through normal searches.
So the OPOC model replaces the administrative and technical contacts, or
explores replacing, with a new operational contact.
It removes all contact information about the registrant except for their
name, includes full contact information for operational contact.
The noncontact information DNS server status remains unchanged, adds
consistency and validation in the process of responding to inaccurate data,
limits the amount of data displayed by thick registries.
Most information is displayed only by the registrars.
So current issues that we're going through at the moment, what contact
information about the registrant should be displayed?
Should the administrative and technical contacts be replaced with a new
operational contact that handles both?
Or that is a pointer to both?
Should specific time limits govern the process of responding to complaints
for inaccurate WHOIS data?
At the moment when we're talking about it, we say, "In a timely manner."
Does timely manner need to be defined more precisely?
And if so, what do those definitions need to be?
And how should domains with inaccurate contact information be treated?
Should there be a difference between the first instance of an error and
And how should that be described?
So watch for updates. There will be updates. There will be reports coming
And everyone's feedback, both here and continuing out of here, is welcome,
encouraged, and obviously needed.
And I guess we go to questions.
>>BRUCE TONKIN: Thank you. One thing I want to emphasize, although the OPOC
proposal is proposing to change some of the public display of data, it is not
within the WHOIS's task force's scope to change the data that is collected or
stored. So that's all still there.
Are there any comments on the work of the WHOIS task force?
I think particularly, I'd like to get feedback on the work they are currently
Go ahead, Steven Metalitz.
>>STEVEN METALITZ: Thank you very much. Steven Metalitz with the
intellectual property constituency.
I just wanted to say a few things about what we've heard here in Marrakech
about the working definition of the purpose of WHOIS that was adopted since
the Wellington meeting by the council.
And I heard three things. I heard first expressions of concern from many
governments in the GAC/GNSO working session about the impact of this
formulation or the potential impact on enforcing consumer protection laws and
enforcing privacy laws.
Second thing I heard was the expressions of concern from many industry
sectors that have a role in protecting consumers about the potential impact
of this formulation.
And a number of letters have already been -- I think they have already all
been circulated to the council. They have all been presented to the board.
I would just like to mention one in particular, and to emphasize that these
letters are not all about intellectual property issues by any means. But if
you look at the letter from the financial services sector coordinating
council for critical infrastructure protection and homeland security, it
states, we believe the adoption of formulation one would make it more
difficult and time consuming for financial institutions to identify and stop
domain-based scams and the identity theft and account fraud that result.
And they go on to say, agreement from the operators to take down Web sites
quickly when there is a clear violation of trademarks or indications of fraud
is only a partial solution. Financial institutions still need the WHOIS
information to address the other forms of abuse noted above. This is signed
by 30 organizations ranging from the American Bankers Association to the
securities industry association and everything alphabetically in between.
It's a very broad cross-section of the financial services industry.
So that's the second thing I think we heard from industry.
And the third is we have heard a number of new interpretations or
re-interpretations of the working definition that was adopted. Different
perspectives on what was really meant by the council at the time that it
adopted this formulation.
Well, there's obviously been a lot of friction and there's been a certain
amount of stress put on the policy development process but like Avri, I am
trying to be optimistic about the process, too, and hoping that overall,
what's happened here will have a positive effect as we move forward to find
I don't want to speak on behalf of others, and I don't really want to try to
tell the council to do, that's not my place. I just want to say a few things
on behalf of the intellectual property community about things we are going to
do. First, we are going to continue to participate, actively and
constructively in the WHOIS task force as we have throughout, and I think
Avri did a very good job of summarizing where we are and what are the
particular issues that we're dealing with there.
The second thing we're going to do is to step up our level of engagement and
dialogue with interested registrars and registries about what can be done,
what changes can be made to improve the current WHOIS system.
I think we share some common goals. I hope we do. And we're going to try to
And third, we're going to continue to try to correct misconceptions about the
current system, many of which have been repeated here at the Marrakech
I'd like to close by citing an authority that I suspect has never been cited
in an ICANN meeting. Maybe I'm wrong. But it's Justice Louis Brandeis, who
is one of the giants of American jurisprudence. Justice Brandeis is famous
primarily for two things. He wrote the seminal article on the right to
privacy. Indeed, he may have coined that term, right to privacy. He
referred to it as the right most valued among civilized men, and I would say
women. He was writing this in 1890 so he didn't speak quite the way we do
today. But he was also famous for many other law reform activities, and one
of his best-known phrases is sunlight is the best disinfectant. Sunlight is
the best disinfectant.
There are many people around the world in the public sector and in the
private sector who are actively engaged in combating infections, and I use
that term very broadly and metaphorically, on the Internet. Improper,
illegal behavior on the Internet.
The publicly accessible WHOIS provides sunlight in this battle. And if we're
going to block that sunlight in some way through changes to the system, what
are we going to do that will be equally or more effective in bringing light
to these corners of the Internet?
>>BRUCE TONKIN: Thank you, Steve. I think that was all very well put. Go
ahead, is that Sarah Deutsch?
>>SARAH DEUTSCHE: Yes, I am Sarah Deutsch, I am here from the business
constituency, just to echo Steve's statements, our constituency has had a
long-time concern about having an open and accessible WHOIS database, not
openly for businesses but for our customers. They need access to the
information to know they've reached the correct Web site, that a particular
site is not involved in phishing or pharming or other types of fraud.
So this has been very important to us from the beginning.
We, too, were very concerned about the narrow formulation of formulation one
about the purpose of WHOIS. We considered the vote very important because
the purpose, after all, is the over arching framework from which all future
work on the task force goes from there.
And unfortunately, I think that the vote had sent a chilling effect
throughout the business community, which is why we received this growing
package of letters from every possible business, including new entities that
have never before seen the ICANN process, and frankly, I don't know that you
all want them involved in the ICANN process because it's not the friendliest
introduction to ICANN.
But in any event, I would say that we're very pleased in the last day or so
to see some openness in discussing what this definition means, and we, too,
remain available to work constructively and to move this process, going
forward, to reach the right result.
>>BRUCE TONKIN: Thank you, Sarah.
My eyesight is not the best at this range. Is that Maggie?
>>MAGGIE MANSOURKIA: Yes, it is.
Thanks, Bruce. Maggie Mansourkia with the ISPCP and I am also a member of
the task force.
I'm a little embarrassed to say I am thoroughly confused about what the
formulation one means, because we have had many interpretations from the
chair of the GNSO which have been very helpful, but at the same time, I hear
other members of the GNSO advocate vigorously on our task force calls about
the very, very narrow limitation this definition. And yet sit silently while
the chair responds that it's a very, very broad definition and that all the
law enforcement and IP infringement and other uses would be continued.
So I would really urge the GNSO to clarify for all of us, for the task force,
for the community, for each other perhaps, for the GAC, compactly what it is
that you think you voted on because my understanding within the task force is
very different from what it sounds like the GNSO's understanding is. And
that's just not a good way to go forward.
I would like to share Steve and Avri's optimism, but I think based op the
current confusion, there's really no good way to do that.
And I would also like to remind folks of one other important thing, which is
that we have often heard everybody say that law enforcement uses will
continue. I would urge you to keep in mind that many, many, many enforcement
of laws takes place by the private individual and private organizations, not
by law enforcement, and that's a really important thing to keep in mind. We
cannot have a safe and stable Internet if we rely on officials of government
and law enforcement to take care of everything for us. We know that's not
realistic. We know that's not going to happen. Private individuals also
need the ability and information to enforce their own rights in law.
>>BRUCE TONKIN: Thank you, Maggie. Any other speakers on that topic?
>>STEVE DELBIANCO: Hi, Steve Delbianco, executive director of the Net Choice
coalition, which is a lot of e-Commerce firms, including e-Bay, Yahoo!, AOL,
And my concern would be having listened to the confusion about whether
formulation one or formulation two, and more importantly the confusion that
once having selected a formulation, how much does that constrain what else
the task force can do in determining policies with respect to what's
collected, how it's revealed, whether there would be tiered access, and
whether there might even be a differentiator between personal Web sites and
business-oriented Web sites.
So I'd like to ask the council whether the selection of formulation one in
any way inhibits and constrains the task force in trying to develop policies
that are responsive to the needs of both privacy and intellectual property.
Because I listened this morning when we tried to determine what to do about
grace periods. Quite a few speakers got up and reached back into their
memory banks to try to tell you here is why we approved that. Here is where
it came from. We even heard differing accounts of where grace periods came
from on domain name registrations.
So I feel six months or even six years from today I don't want to have to
look back and be constrained on what we are doing on WHOIS because of the
rationale that went into the selection of formulation one. It is not a
casual decision, it's not an arbitrary decision. I think it's important for
the council to clarify why you picked one over two, what that means, or
simply put that ball back into play, and let the task force go on to
determine the rest of its parameters.
>>BRUCE TONKIN: Thank you. Any other comments on formulation one and
formulation two? Go ahead, Wendy.
>>WENDY SELTZER: Thanks. Again, from the At-Large Advisory Committee and
representative on the WHOIS task force, ALAC represents the Internet user who
is both a Spam hater and intellectual property holder and private publisher
of material online, and yet in those multiple roles has concern about the
publication of full personal data in the WHOIS database.
Just to respond briefly to a couple of the comments from the floor, because I
think that the WHOIS task force process has been working, albeit slowly,
making progress and moving toward better understanding and ultimate
resolution of these issues. And so I would hate to see the transparency that
we have been able to introduce into this process by actually publishing
interim reports if that were made into a way to derail the progress that the
committee has been made. The fact that we're telling people what we've done
so far shouldn't mean that we are no longer able to continue with the
progress we have been making.
Just a quick question, back to the intellectual property and business
interests we heard from. When I was with electronic frontier foundation, we
were working side by side with Verizon fighting against the disclosure of
personal information in the case of the recording industry versus Verizon
when the recording industry was on unverified claims that copyright
infringement was going on, was asking for the names of thousands of Verizon
customers. And we were very pleased to stand side by side arguing for the
right and ability to participate in online communications without revealing a
I would hope that there are some who can help us to extend that principle to
online communications that are identified with a domain name as well.
>>BRUCE TONKIN: Okay. Philip?
>>SARAH DEUTSCHE: Can I respond to that?
>>BRUCE TONKIN: Yes, go ahead, Sarah.
>>SARAH DEUTSCHE: Yes, I think this debate does not change that equation at
all in Verizon's view. We are very interested in protecting customers'
privacy. I think the difference, though, is with ISP customers, we have all
the accurate data about people's communications, and we turn that over in the
case of a subpoena. The WHOIS database on the other hand has always been a
public and open database and it's been used for -- mostly for commercial use.
But Jonathan Liebowitz yesterday had suggested yesterday that if maybe some
sort of solution where if people wanted to protect their personal identity
and they were using this in some sort of noncommercial fashion, there could
be some sort of other solution for that. I think there are solutions today,
including as proxy registrations, which people are actively using, sitting on
somebody else's website, not owning your domain name. There are many
solutions but this is something that we need to think about but we can't
mix-up an individual's needs for noncommercial speech purposes with the
entire universe of domain names, many of which are used for commercial
purposes and very harmful purposes. And just because that person is an
individual doesn't mean that they should have a right to privacy,
necessarily. There's no right of privacy to commit a crime, and we shouldn't
be in a situation where registrars are flooded with subpoenas from every
single trademark and copyright !
and business owner and customer out there on the Internet just because they
seek a piece of information about their use.
So I'm sure we will be discussing this more.
>>YOUNESS EL ANDALOUSSI: Hi, I am Youness El Andaloussi. I am from Morocco,
not from any organization. But I wanted to speak as a regular Internet user.
And I've had, over the last ten years I used the Internet, a dozen times to
use the WHOIS to contact Web site or for legitimate purpose.
One was from a defective e-mail server that (inaudible) to me, or another
time it was from a marriage announcement that was put on a Web site and there
was no contact information on the Web site, and I had to gather it from the
I think it is very important to keep the WHOIS information public, not just
for law enforcement purposes but also to everyone.
Now, I understand there is concerns for privacy, but I think it should be
important to have a way to contact the people, the owners of the domain
names, in case there is a problem. Because it's very difficult already now,
because there is bogus -- a lot -- many times, bogus information. If you
take away all information, then it's a bigger problem. Thank you.
>>AVRI DORIA: Bruce, can I respond?
The proposal was not to take away all information. The -- whether it's the
administrative and technical or whether it's the operational point of control
-- of contact, there is, indeed, a set of information for contacting someone
about all technical and operational issues and for contacting them about
anything that has to do with the configuration record.
So there is someone to contact in all cases. There's no -- there's no
intention, I think, on anybody's part in WHOIS to remove the ability of
somebody to contact somebody that can help you. It's just removing personal
information from the public that is possibly in discussion. But certainly
not to remove a point of contact where you can get things taken care of.
>>YOUNESS EL ANDALOUSSI: My point is it is already difficult enough many
times to contact people from -- right now. So if you make it more complex,
then it's going to be even more difficult for the individuals. And as was
pointed out before, if you place the burden on law enforcement agency, then
it's a big problem.
>>AVRI DORIA: Yeah, the question wasn't to place the burden on law
enforcement agencies. And that's one of the other things that's being worked
on, is to make sure that the information for that OPOC or for the point of
contact is, indeed, easy to reach someone. I think it's actually a
simplification of here is one address you go to to contact about
configuration information and problems.
>> Bruce, may have I comment as well?
>>BRUCE TONKIN: Hang on, Tom. Philip, then Tom, then Tony.
>>PHILIP SHEPPARD: This is still on the same string so we are not starting
something new. I think it's very interesting how the discussion which
started off talking about formulation one and two on WHOIS has rapidly
shifted to a discussion about open access of WHOIS. And I think that
characterizes very much the nature and complexity and confusion that's been
in this debate so far.
I have just two questions I wanted to ask. And Wendy, the first is to you
because you started your comments saying about ALAC's view on this was
concern about open access.
Do you believe that the vote that the GNSO took on this assumed open access
as a given, and on that basis the choice was between formulation one,
>>WENDY SELTZER: Well, the -- so in the context of the task force's terms of
reference and scope of work, we were given the request to work on conditions
of access to data. And in order to resolve amongst ourselves on what
conditions data should be available for public access, we were asking the
council to help us by deciding which was its view of the purpose of WHOIS.
So I'm sorry the process is so convoluted, but we felt that getting the
council's sense of the purpose of the WHOIS database would then help us as
task force tasked with describing public access to that database, to describe
public access consistent with the purpose of WHOIS.
Does that answer the question?
>>PHILIP SHEPPARD: Well, Wendy, no, it doesn't. But that doesn't surprise me
in the least.
Let me try our chairman. Bruce, you supervised the vote that we took. Do
you believe that when we took that vote on council between formulation one,
formulation two, there was an assumption of open access as a given before we
took that decision?
>>BRUCE TONKIN: I don't understand the question, I must admit, Philip. What
do you mean by "assumption of open access"?
>>PHILIP SHEPPARD: The way access currently is, without any change to it.
>>BRUCE TONKIN: So are you saying did I assume that making that decision
meant that WHOIS wouldn't change? Is that the question you are asking me?
>>PHILIP SHEPPARD: No. I am saying that did we assume as council, was there
an ongoing assumption as council that the current means of access to WHOIS,
the open access that we have, was somehow, in people's minds, as -- as the
way life is, and therefore, the choice of purpose of WHOIS was built on that
>>BRUCE TONKIN: I must admit, I still don't quite understand the question, so
I will just say no, I don't think that's a correct statement from my
>>PHILIP SHEPPARD: Okay.
>>BRUCE TONKIN: Because you are sort of jumping to the next phase which the
task force hadn't completed yet and is still working on, which is what
changes to WHOIS will be made.
So I certainly expect that there would be a change.
I might add, though, there isn't an existing definition of purpose, so I
didn't have any view on what that change necessarily would be. But I would
expect there would be a change, because the purpose of the WHOIS task force
was to improve WHOIS, so I would assume that the change would improve WHOIS.
That's my -- the best way I can answer that question.
And when you talk about open access, the second, it might be the third terms
of reference talks about it's in two phases. The first phase is should there
be any change to the information that's made public, basically. And
secondly, is if there is change to the information made public, let's say you
remove the piece from public access, what other access mechanisms would be in
place. And certainly I assume that those other access mechanisms would
provide access to legitimate users. And then we might debate what legitimate
users are, but I say the key is in that third terms of reference. That
there's a change to what's made public, and exactly what that change is, I
was expecting the task force to work on. And then if it did change what was
made public, that there was other access mechanisms that they would put in to
So my view is that -- not my view. It's a statement of fact, that the data
being collected isn't changing. So the data is still there. And then it's a
question of should that access be for everyone in the general public or
should the general public have a different level of data compared to others
such as law enforcement, intellectual property ISPs, and others.
>>PHILIP SHEPPARD: I agree completely in terms of that's the way it should
have been. I just wonder if that's the way it was in the minds of all those
who voted at the time.
>>BRUCE TONKIN: I can't comment. I just answered what was in my mind, at
>>THOMAS KELLER: Thanks, Bruce. I just have a short comment on certain
things that have been said, and previous replies to the task force or to the
It's not really about -- only about the definition. We have to find about
what would be the adequate privacy protection, but that there are privacy
laws outside, out in the world already that we have to take into
consideration as well.
So we are not really free to decide anything we want to, actually, in that
WHOIS environment, but there are certain laws and certain countries which we
have to take into consideration, and we have to abide to.
>>TONY HOLMES: Thanks, Bruce. I just want to comment here that it's been
quite an experience over the last few days because as council, we took a vote
on two options and they were going to be used as the basis for moving
Over the past few days, I've heard people comment on those options. They've
expanded on what they believe was contained in those options. And I must
admit that some of the clarifications that have been given have come as quite
a surprise to me. I never understood it that way.
And I voted on this in council.
And I think if you stand back in the cold light of day and you look at the
two options, then you'd think, well, how did we get here? Because there's
confusion around both of the options that are there.
And you have to understand the history of this and the long history of the
task force and how we ended up in the position where those two options were
tabled as the basis of moving forward.
What's come out over the last few days is that there's a lack of consolidated
view within council as to what those options contained. Even sitting here
today, hearing clarity come from the -- from people around the council on
this issue. And I've heard people give comments who are actively involved in
the task force and say they are confused.
What we have done, we've launched something and voted on something that
there's been a tremendous reaction to. And I think as council, we need to be
very careful how we move forward. Now we're looking at OPOC and it's just a
proposal. It's nothing more than a proposal at the moment. It's something
we need to consider. And how council moves forward needs very carefully
constructed, I think, over the next month or so. Because it's quite clear
that what we thought was the basis of moving forward is not really acceptable
because there's confusion over that and we need to find some clarity over
this situation and make sure the next steps we take in moving this forward
are really based on a sound and consolidated understanding, not mixed views.
And that's currently exactly where we are.
>>LUCY NICHOLS: Bruce, may I comment?
>>BRUCE TONKIN: I'll just take a queue again. So that's Mawaki, Lucy.
>>MAWAKI CHANGO: Yes, to respond to Philip's question earlier, personally,
when I voted for the formulation number one, the only thing I assumed was
that the use of the WHOIS data will be -- will be to fulfill the exact
purpose mentioned in that formulation, not more than that.
That was the only thing I assumed.
So if to fulfill those purposes the access to data need to be public, then
wait. If it doesn't need to be public, then we shouldn't put it public. So
that was the only thing I assume. And that was the way I understood it.
Now, I have to say, personally I understand perfectly that law enforcement
may require that the Internet Governance mechanism provides the means to them
to investigate crimes that are committed based on the Internet. But the
thing is, just as they can't go to any -- to any such private property
without warrant, I think it's normal to expect them to have some due process
in order to have access to the personal data in the WHOIS database.
So it's not the mission -- we need to understand that whatever we do, we do
it under the scope of ICANN mission and (inaudible). And it's not the
mission of ICANN, per se to assist law enforcement, per se. It's not its
mission either to oppose law enforcement. But what ICANN needs to do is to
take care of its own mission. Then if law enforcement needs to search ICANN
databases, they need, I think, to go through due process.
>>BRUCE TONKIN: Lucy.
>>LUCY NICHOLS: The comments I'm about to make are my personal comments as a
council member and not -- they do not represent the intellectual property
constituency. They may or may not.
We voted -- the council voted on the formulations, and we're -- we now have
the final vote.
Whether or not I'm happy with the outcome is not the issue but we do have a
And now that formulation, in my opinion, should go to the task force, and
they should be allowed to do with it what they wish.
I think it's really tardy for a lot of these organizations to be now issuing
statements on the formulations and whether the formulation was right or
wrong. Those comments should have been made prior to the vote. I think it's
a waste of time. It's no wonder that this WHOIS matter is still being
In order for us to make progress, we have to respect the decisions and the
votes of the council.
So I would encourage everybody to accept the formulation. If there's any
confusion around what the formulation means, then perhaps that should be
taken up within the task force. But I'm very dismayed by a debate now taking
place on whether this is the right formulation or not.
We voted on it.
We have the results of the vote, so let's go forward.
>>BRUCE TONKIN: Okay. Any other comments from the council members?
Okay, I just want to make a few Comments of my own. And I guess in different
ways, people have been talking about reinventing history after a vote or
having different interpretations.
Firstly, for my perspective, as I understood the council process, we
discussed this in the council meeting in Wellington, and I did go through
what I thought the meanings were of those two formulations. And I said they
were fairly similar except for two areas. And then I clarified at least my
understanding of what the first definition was with respect to issues. And I
did explain why with respect to what a DNS record is.
Now, I can accept that individuals on the council chose to vote based on what
they thought was in the best interest of ICANN, so it's difficult for me to
say why they voted for something. But there is something on the record which
is what I at least explained that I thought those definitions meant, and I
didn't get a huge amount of push-back from anybody that was supporting
formulation one with respect to how I described it.
Since then, people did vote. There was a meeting following the meeting in
Wellington when we simply just put it to a vote, because I felt that we'd had
about an hour of discussion in Wellington and people voted how they saw fit.
Then since then, I've heard wildly different interpretations of what that
meant, and I can't understand where they got those from because there was no
statement that led to some of the interpretations that at least seemed to be
floating around in the community.
Now, I completely agree, though, that given that there has been so many
interpretations in the community, that means that we haven't been clear as a
council in communicating what at least the council thinks that means.
The sorts of things you are raising, Tony, about there were different views,
there's always been different views and that's the challenge. I mean, I
think if everyone had the same view on something and then the council votes
on it, it's pretty easy to, as chair, reflect on that what means. It gets
harder when you have something that's almost split in half and then it's
voted in favor.
The options for the council at this point are to either change the definition
itself or improve the definition itself, or add additional material to
clarify. And I notice, for example, that the ICANN bylaws do that. They
have a mission, which is pretty general, and then they have a set of core
values that are intended to be used to help interpret what that core mission
is. But otherwise, and I'll display in a moment that core mission again,
because that can be interpreted in wildly different ways as well. And the
ICANN bylaws chose to elaborate on that with some values.
So we have some options, I suppose. One is we change the definition as a
council to make it clearer to the community. That still may be difficult to
get that right, and I think it's a question if that's the right thing to do
at this time. Alternatively, the council could decide to elaborate on that
definition with some principles underlying it, and that might be easier.
I'll let you comment in a second, Tony.
So that's one approach.
Another approach, and I think I see this happening particularly with the new
gTLD work is similarly, if you take any single statement that is in --
certainly in the presentation I did earlier, you can interpret any of those
statements in lots of different ways as well. For example, if I said you
need to have a sound business and operational plan, I'll bet there's about
100 different interpretations of what that means in this community.
And one way of doing that is to have a reference implementation that's agreed
by the committee as reflecting what they believe as a committee that that
particular statement means.
So, therefore, I'd like to encourage, at least at the task force level, and
the council can discuss this further later this week in a formal council
meeting, but at least the task force can focus on improving the proposal that
has been put on the table.
And I am very pleased to hear that Steve Metalitz from the intellectual
property community has stated that they will work -- continue to work in the
I really encourage them to focus on the actual data elements themselves and
what should or should not be public.
And if it shouldn't be public, how else should it be accessed?
Because that's really the core.
And also focus on the objective of WHOIS work.
And that is to improve the effectiveness of WHOIS and to protect the personal
data of natural persons.
So if you just take that as an objective and then focus that objective on
your work on OPOC, I think that's the best thing you can possibly do at this
point in time.
The council, however, is at a different level.
We've actually decided to get involved and we've made a decision on the
We need to -- you know, I'll leave it to a council decision later this week
as to whether the council thinks we should add points of clarity on the
definition or whether we should change the definition.
But I just wanted to show two things that at least I thought was part of the
approach, is that we are ultimately work in the constraints of the ICANN
So the ICANN mission says it coordinates the allocation and assignment of
unique identifiers for the Internet.
And it coordinates policy development reasonably and appropriately related to
these technical functions.
That's ICANN's mission.
Just put that up clearly again.
And then formulation 1 says, provide information sufficient to contact a
responsible party for a particular domain name who can resolve or reliably
pass on data to a party who can resolve issues related to the configuration
of records associated with the domain name within a DNS name server.
There's not a lot of difference between those two things.
Now, I've heard people very narrowly interpret this formulation, or at least
narrow in my personal point of view, to say, "Oh, that means that you can
only deal with issues 'cause you put the wrong I.P. address or something in
the name server record, and that won't allow us to deal with any legal or
operational issues with ISPs or law enforcement issues."
But then you can equally say the same thing about the ICANN mission.
So either the ICANN mission should be changed to be much broader or we need
to at least develop text in terms of our definitions that are consistent with
that definition of ICANN's mission.
Because the GNSO is part of ICANN.
And while I can accept a lot of the comments that are made that it's nice to
write a very expansive definition and that the purpose of WHOIS is to deal
with spam, to deal with law enforcement, to deal with child pornography, but
then if we're going to do that, we should stick it in here as well, stick it
right up front in the ICANN mission, and then we can incorporate that into
So that's just food for thought.
Because it may be that we have the ICANN mission wrong.
And just food for thought.
But certainly I think the council needs to consider whether it should change
the definition or whether it needs to clarify what that definition means.
And I think it's going to be difficult because there's obviously a lot of
different views around the table.
But certainly I've heard people make comments about what individual members
And I think it's better that people say what they think or what their group
thinks rather than say, well, the registrars think this, when you should let
the registrars speak for themselves, or, likewise, making a comment about
what the intellectual property group think.
We should let them speak for ourselves.
And we have one here right now.
Steve, if -- and I think I'll let Tony go first and then I'll go across to
>>TONY HOLMES: Okay.
My comment's fairly brief, Bruce.
I do remember your interpretation of both of the options, and they were very
I think if that had been a view that had been generally shared, then probably
with a small tweak on either of them, we would have ended up with one anyway.
The reality was that this was the culmination of a few years of difficult
work in the task force.
People were pretty much talked out.
They understood not only their own views, but everyone else's views.
And they'd made their mind up about these things.
And there was clearly different interpretations put around those.
That isn't the important thing now.
The important thing is about moving forward.
And both of the options that you've tabled, I think, deserve some
consideration, careful consideration.
And out of that I think we'll get our way forward.
And that's the main thing is moving this on and learning from this experience
So we should look ahead and probably not keep going back over hour we got
>>STEVE METALITZ: Yes.
I'll just reiterate what I said at the beginning of my remarks.
I was expressing the view of the intellectual property constituency.
I wasn't trying to tell the council what to do on this.
And I just -- and I'm not trying to characterize the views of any other
I've talked to people in those constituencies.
I've heard some things here.
And we're going to follow up on those discussions.
I would just say, though, in terms of your last comments on the ICANN
mission, it's a big world out there.
There are many people who support the ICANN mission that are also very
concerned about what they read in black and white as far as formulation 1.
There are also a lot of people who have never paid very much attention to the
ICANN mission before.
They haven't had to.
It hasn't really come up to them.
But they are doing some very important things on the Internet.
They are, for example, moving billions of dollars around on the Internet in
all kinds of financial institutions worldwide.
They are conducting travel and lodging businesses, which we heard earlier
today is a global business, and while -- other than through dot travel, many
of them have not been very involved in the ICANN mission.
So what they are hearing when they re- -- or what they are seeing when they
read formulation one with their own eyes and using the best of their
knowledge, they're seeing something that alarms them about the direction that
ICANN is going.
So let's see keep all of this in the perspective that it's a big world out
there, including a lot of people who aren't intimately familiar with the
ICANN mission, and they are concerned about the direction ICANN is going.
>>BRUCE TONKIN: So, Steve, can I -- I might as well just put you on the spot
on this particular point.
What is your recommendation?
Is it to change the purpose or to clarify the purpose?
>>STEVE METALITZ: I'm not making a recommendation to the council on this.
As I said, we're going to continue to work in the task force.
We're going to continue to work in the -- in discussions with the other --
with the other -- members of the other constituencies.
I think you should -- the council should think about what it may need to do.
And I think the GAC may be -- an impression was left with the GAC as well.
And if you wanted them as an active partner in this policy-making process,
perhaps they need some clarification.
So I think you should take that into account, but I'm not making a
recommendation on that.
>>BRUCE TONKIN: Okay. Thank you.
>>JONATHAN ZUCK: Good afternoon.
My name is Jonathan Zuck. And I'm the president of the association for
We're an I.T. trade association representing about 3,000 small and
medium-sized businesses around the world.
And my -- what I'd like to do is revisit the point about conflating access
and collections, because I think that's actually a point that was raised
earlier, and I think a question was asked that we didn't necessarily get a
very clear answer to.
And I'm somewhat concerned about that conflation, because the idea that the
ICANN mission needs to be updated to include a bullet point that says, "Abide
by the laws of the governing nations in which the Internet is used," seems
like it might be gratuitous; right?
>>BRUCE TONKIN: It's actually there.
It's one of the core values.
>>JONATHAN ZUCK: So the extent to which that's already the case, that should
be taken into consideration as well as the sort of technical aspects that
you've outlined here.
So the collection of data might be necessary for the enforcement of consumer
protection laws in one case, and the limitation on access to that data may be
necessary for privacy laws in other cases.
And so that's why I think it's really important and probably worth doing in
writing to make that distinct between collection and access such that those
issues are discussed separately.
Because our job as putting together things from a technical standpoint is to
make sure that all the plumbing is there so the information necessary to
comply with all laws around the world is available.
And then have a separate discussion about who and how that data is accessed.
>>BRUCE TONKIN: Part of the work of the task force is -- it's not in scope
currently to change what's collected.
But it is in scope to help define the purpose of some of the elements that
So today we collect a registrant admin, technical, and billing. And there's
a lot of different -- again, when we start talking about different
interpretations, there's a lot of interpretations on what they mean.
The task force isn't changing what's collected.
But it is looking at help define the purposes of the bits that are collected
and also looking at access.
So I think we are on the record of having split those up.
In terms of your comment about national law and international treaties, the
council does actually have a policy on that which was approved by the ICANN
board, which basically says that -- which, again, is probably stating the
obvious -- but, again, that registrars and registries must comply with
national law and, you know, that there's essentially both at a policy level
and also it's actually one of the core values of ICANN.
>>JONATHAN ZUCK: So defining the purpose of that data could still be
something that puts you in the way of that precept of complying with
international laws and (inaudible).
>>BRUCE TONKIN: Yes.
>>JONATHAN ZUCK: The danger still exists.
>>BRUCE TONKIN: And I agree with you there, yeah.
Partly what we've been hearing, I guess, from the GAC is some advice from
them with that regard.
So I agree, we do need to take that into account.
And it's a requirement, certainly, at the board level, to receive input from
the GAC before it votes on a policy.
And that's the sort of input they'd receive.
And the GNSO Council is trying to receive that input before it gets to the
board so we don't have the situation where you have six months of work and
then something at the final moment you find out that it's in violation of
So one of the things the GNSO Council has actively done -- and this goes back
several years -- has asked the GAC for that advice.
We still, actually, haven't received it.
And we have been told that we will receive that advice in December, I
believe, of this year, on the national law and public policy principles
associated with WHOIS.
So we have asked the question.
And we will certainly be taking it into account.
>>BRUCE TONKIN: Go ahead, Stuart.
>>STUART LAWLEY: Stuart Lawley from ICM registry.
I would just like to say what probably some of you are thinking but nobody
else will say, is that from my position of having been just on the receiving
end of the might of the United States government's interference in the ICANN
process when the ICANN process delivers a decision that they don't like, I
think that's exactly what you're seeing now on the GNSO's formulation of
Clearly, the result has come out at the end of a two- or three-year process,
and the USG doesn't like it, and they will mobilize whatever forces they can
to stop this happening and get it overturned.
And you probably, like me, will waste another six or 12 months, only to be
>>BRUCE TONKIN: Thank you for that comment, Stuart.
Any other comments on WHOIS?
So certainly it will be on the agenda for the council in its discussions
later this week.
Oh, sorry, David.
>>DAVID FARES: I would simply like to make a statement in response to the
I think it's very important to recall that there were a host of governments
who participated in the GAC council meeting earlier this week on WHOIS who
presented their views and their concerns related to formulation 1.
So I don't think it's the position of any single government.
>>BRUCE TONKIN: Sorry, David.
I'm not clear.
Are you -- what I had said is we haven't received information from the GAC as
Let me clarify that.
We have received a number of inputs from government agencies and individual
governments, over six years, I might add.
So to characterize it as though it's the first time we've received input is
In fact, there's been a number of presentations from government agencies and
governments, but we have not yet gotten information from --
>> SHAUNDRA WATSON: My name is Shaundra Watson, and I just wanted to clarify
that I am speaking on my individual behalf.
I'm not speaking on behalf of any law enforcement agency.
But I did want to say, in response to the last -- not the immediate comment,
but the one preceding that one regarding the interference of the United
States government, I just -- I did want to point out that the panel yesterday
really represented representatives from all over the world.
There was a representative from the United States, there was a representative
from Europe, the Netherlands. And there was also a representative from Asia,
So I think it's a little disingenuous to say that this is a response of the
I think the panel intentionally was made, or comprised from people all over
the world to emphasize the point that it really is a global issue and impacts
several different types of agencies.
And I also just wanted to comment on, I think someone had commented about
being dismayed at the timeliness of the comments and how they came after the
But I'm not really sure that it was clear as to when the vote would be made,
first of all.
And, secondly, I believe it was mentioned yesterday that oftentimes the
comment period is a little too short for individual governments to submit
comments, which is also problematic.
And third, in this process, I have found that a lot of people aren't as tuned
in to this process as everyone here is.
So there are several law enforcement agencies and people with other interests
all over the world who don't even know that we're here and don't even know
that we've espoused these positions.
And so I guess I just would encourage you to step back.
And this was a process where I think it was successful in that we were able
to include these views that aren't necessarily here every day before you, and
not to focus on the timeliness, okay, well, it came after the vote, and now
it's an interference.
And I think it's really a way that we can say, hey, well, how can we include
these viewpoints earlier on so that this type of thing doesn't happen again
>>BRUCE TONKIN: Thank you.
Yes, certainly -- I think the thing that's come out of discussions with the
GAC over the last few years is they -- they have said that it's difficult for
them to respond on any specific issue in a timely manner.
And that's why they've been trying to put an effort into developing some
principles around both the topic of WHOIS, and I also believe they're
planning to provide some principles around new gTLDs.
So that I think the feedback that they have provided to us is that they're
intending to focus the GAC resources, if you like, on providing some broad
principles that everyone in the GAC can agree to.
Individual governments can also respond, but also, for an entire government
to respond, it can take a while, because there's so many agencies within
large government structures.
And then you can actually go down to an individual agency responding and then
down to an expert within an agency that might respond on an individual basis.
So there are a number of options for governments to respond, and they have
done so over the time frame.
But at the same time, the GNSO and ICANN has been encouraged to work quickly.
And one of the reasons for setting up ICANN was that it can work faster than
governments traditionally have done.
And that's why it is a private sector organization.
So all these things are a matter of balance.
But certainly when we receive the input, we gratefully accept it.
Any other comments?
Well, perhaps we'll get on to another topic, which is IDNs.
There was a question earlier about IDNs, and I gave a short answer to it
But I'll just give a slightly longer answer to it this time around.
This material was presented in the IDN workshop on Sunday, but not all of you
may have been there.
Certainly at a council level, we've heard some gTLD operators want to
introduce a IDN string that relates to their existing gTLD.
Dot travel could be one such example.
We have ccTLDs operators that want to introduce an IDN-based string,
presumably related to the scripts or languages they use within their country,
and we also have independent parties that are not currently operator of
either gTLDs or ccTLDs that would like to introduce an IDN-based string.
They are all new top-level domains.
And as I had mentioned, we formed a working group within the GNSO to consider
this issue further.
And that working group has recommended studying the IDN gTLDs using the same
framework as used for the new gTLD policy process.
And using IDNs will highlight some of the complexities around introducing new
So the sorts of things that we heard earlier on new gTLDs, clearly, there's
the topic of selection of the appropriate IDN string and the selection of the
appropriate organization to operate that particular string.
There are issues about what requirements an organization to need to make to
be allocated an IDN string.
If strings are differentiated with respect to purpose -- and certainly I
think whatever the policy is at the new gTLD level would apply to new IDNs,
so whether it's a more open process or a more tightly constrained process,
both of the processes would be the same.
It is worth noting that as we move into multiple scripts, it becomes a bit
harder to tell whether some strings are clearly differentiated.
There's the question of how many strings are appropriate for a particular
So let's say it's a new operator, might say I've got this new top-level
Let's say it's dot sport.
I want to have sport in a hundred different scripts and languages.
Is that appropriate?
With respect to ccTLDs, they're currently restricted to choosing strings from
the ISO3166 table.
And the question is what IDN strings should be allowed for ccTLDs.
There's been some different proposals around that coming from the CC
gTLD operators may be prevented from using an IDN string that could be
confused with a country name in a particular script and language.
Then there's the issue of if there are two parties that want the same string,
how do you choose between them?
Even the same string may have different meanings for people in different
parts of the world.
If we consider the WHOIS formulation 1 as a long string, that string has
So certainly when we move into different scripts and languages, they could
And two strings that are technically different, and by that I mean they have
a different ASCII representation in the root zone, may look visually similar
when you actually display those on your Web browser or on your E-mail
application, whether translating the ASCII equivalent into something that
looks like a particular script or language.
From a contractual point of view, I think the main thing that would be
different or additional, if you like, for IDNs would be that an IDN operator
would -- we presume would be operating under the same framework agreement as
a new gTLD with respect to an IDN gTLD.
But they would need to comply with the IETF standards, as well as the ICANN
Some of the other issues that have come up and probably don't necessarily fit
neatly into those categories is, if an operator chooses to have two strings
for the same purpose, should the second-level strings in those TLDs resolve
to the same Internet location?
That might simply be a choice at the operator level, or if people thought
strongly enough about it, maybe that's something that needs a policy.
Should the script used in a string at the second level match the script used
in a string at the top level?
So, in other words, you might have Latin characters Latin characters, then
you might say, well, if we have Chinese characters, it should be Chinese
characters dot Chinese characters.
Again, these are all decisions that could be done at the individual operator
And that's probably the right place for it.
But that's for the policy groups to decide.
Another thing is, should there be competition between two operators of string
for the same purpose?
So certainly we've heard from some in the community, such as Ron from dot
travel saying, no, they shouldn't be.
They should be able to use -- as relating to the travel industry, there
should just be one operator.
It's also worth noting that UDRP can become more complex for trademark
dispute resolution when we start adding additional scripts.
And something else that Vint Cerf mentioned to me last night is that WHOIS
becomes more complicated as well.
The raw WHOIS protocol on port-43 is probably not particularly well suited
for dealing with IDN strings, particularly if you're wanting to visualize
them in some form that's convenient to display.
So certainly the scope for confusing human users can increase, which may lead
to additional opportunities for phishing and spam.
So you certainly have to be very careful in how we introduce these strings.
But that's basically the current status on the GNSO IDN work.
So, again, I put it up to any comments or questions from the audience.
Looks like when we start talking about something that's a bit more technical,
we have less input.
But I think at that point, we can more or less go to an open mike to use the
So if there's any other issue that people in the audience think should be
brought to the attention of the GNSO, now's the time to do it.
Well, that's good, because we've got enough issues to work with at the
At that point, then, I'll thank everybody for attending this session, and I
believe the next session starts in I think half an hour.
Is that right?
3:30. So the next session will start at 3:30 p.m.