Part 2

ALAC Meeting
(Held June 22, 2008)

Legend
Name: First name of member of group
Name?: If we think we might know who's speaking but aren't 100% sure.
Male: Unsure of who's speaking, but know it's a male.
Female: Unsure of who's speaking, but know it's a female.
Name: 52 - 0.31.06 We will identify a foreign language in square brackets with a time code for the beginning and end, so that you can quickly identify those passages. At the top of the transcript, we will advise you of the name(s) of the foreign languages you'll need to search for. For example, some transcripts may contain German, so you'll have to search for "German[ instead of |French" to find all the passages. The speaker identification is the same convention as all the others. i.e. If we know the speaker's name we'll include it, otherwise we'll put a question mark after it, or if we can't even guess, then Male, Female or SP.Translator. If translator actually translates French or German for a participant we will note it as "Translator". It will normally be evident from the dialogue who the translator is translating for. If not, we will include a note in square brackets with respect to who they are translating for.
[cuts out. Audio cuts out for less than 2 seconds. We may have missed something. We do not put this if it's just natural silence where no-one is speaking. We only put this if we can hear that there was briefly a technical problem with the phone line and/or the recording which resulted in temporary loss of sound.
cuts out 2 secs When the audio cuts out for 2 seconds or more we'll show approximately how many seconds were missed. We will most likely have missed something if it cuts out for this long.
-ation If we believe that all we missed was the beginning of a word because the audio cuts out, we will just put the end of the word that we can hear. If we are very sure from context what the whole word is, we'll just put the whole word, to make it easier for the reader. However, if the word could be a variety of different things, then we will show only the part of the word we could hear or thought we heard.
very faint - barely audible We will put this at the beginning of paragraphs where we are having a very difficult time hearing.

********************************************************************************************************

Cheryl: ...your name. Also, speak slowly and clearly. This afternoon, we’ve got a number of very important briefing sessions, and the first one is both by Tina Dam, who is sitting upon my right hand, and Bart Boswinkel, who will be here a little later. The first briefing is on IDNs and ccTLDs, and I’ll now hand over to Tina, but I will ask you, Tina, do you want to take questions during your presentation, or do you want to cluster them at the end.
Tina: So, thank you for having me, and I think I’ll take questions during the presentation. But of course you’re also welcome to ask afterwards, if... like the whole picture sort of gathers and then you have questions coming up. But if there’s anything that I say during the presentation, or there’s anything on the slides that you don’t understand, just raise a hand and ask a questions, because I think it’s much nicer if it’s a dialogue than just me giving a presentation, and we get nothing out of it at the end anyways.
So, the first session, yeah, on IDNs. And I’m going to start, and then Bart is going to join us and go into details of the fast track, which is a process for IDN ccTLDs.
Izumi: 24
Cheryl: Okay, Izumi’s without the microphone pointing our attention to the fact that we have hard copy documentation, which we will ensure is distributed. Thank you, Izumi.
Izumi: 38
Tina: Okay. Yeah?
Alan: I have one other comment. There is an orphan cell phone here. If it’s anyone’s, they should claim it.
Male: 51
Tina: All right, well, we’ll see if it rings.
Alan: In an hour, I’ll go home with it.
Tina: All right. So, on IDNs. I tried to make the slides so that they fit a little bit better for the ALAC and for the user perspectives, but if there’s something that I’m not bringing up that you think is more important for the ALAC, please mention that as well.
So this is what the agenda for my presentation looks like. We’re going to go through quick definitions and how the IDNs work, just to set the stage and make sure that we’re all sort of on the same page. And then we’re going to look at implementation experiences and all the work that is underway. And this is where Bart will then take over at some point and go into the details of one of the things along the way, which is the fast track process.
Okay, so I think everybody by now knows that IDNs are domain names that have characters other than what is in the standard ASCII. So characters outside ABC through Z, 0 to 9, and a dash. And IDNs have a localized solution, so it’s not anticipated that everybody in the world will understand and be able to read or retype all domain names in IDNs. This is for local use, so in... and it’s for user choices. So users in the different regions can decide what kind of characters they want to use in their domain names for the 25 addresses and how they want to use that. And it of course depends on what customers or clients or friends or whatever it is that they’re trying to do with these domain names, so it’s user choice.
So IDNs have existed at the second level since 2003 under the protocol standard, and actually earlier than that in different test beds. And this is in the web protocol standard. Now e-mail protocol standards are on the way, and they actually were just released by the IATF for testing purposes. And what that... when the IATF does something like that, it means that they want application developers to implement them, try it out, see how it works, come back with the experience gained, and then it will go into a standards track, which means it will become a technical standard.
So, you know, there’s no timelines for that, but hopefully that’s something that’s going to happen rather quickly, because I think users... well, you tell me, but I think users will probably appreciate having IDNs to use for e-mails in addition to web.
But we also need IDNs at the top level, which we do not have, and I put up an example up there that should make it apparent why it is that we need that.
Functionality, and some of you may have seen this slide before, because I’ve used it several times. So I’m not going to go into much detail with it unless there are questions about it. I’m just going to say that IDNs work at the user level, at the client. So it works on your own laptop. So it depends on what kind of software you have implemented, and what kind of browser you have, for example, on your laptop. If you have a browser that can handle IDNs, then it will work. If you don’t, then it won’t. It has nothing to do with the DNS per se, because the browser will translate the IDN address, so, you know, the Korean characters, into an ASCII sequence of characters that then gets sent off to the DNS, and the address will get resolve. So everything works in the way that it always did. It works with ASCII characters.
So why am I mentioning this? Well, it’s actually pretty important, because the fact that it’s used... that it works on your client and on the different applications you have on your client is important for users. And it’s important for users to understand that things will work differently depending upon what kind of application you have implemented.
So what that means is that users will get different kinds of experiences, and they’ll get different experiences across the world. And even within one region, they will get different experiences. And having users understand that experience and accept the fact that it is different is really important for us to do in order to make IDNs work in the global market space.
Browsers have implemented IDNs differently, and they have done that mostly for security reasons. And that has to do with, early on, things were registered that shouldn’t have been allowed to be registered, because they looked... things looked confusingly similar. So you know, the PayPal example? Is that pretty known to people here? Yeah, some are nodding.
So PayPal was the mixing of Latin and Cyrillic. So you have two addresses that look the same but go to two different places. Fraud options, things like that. The browser developers responded and said we need to be careful about how we allow our users to access these addresses. So they put up different warnings, and sometimes that means that users don’t want to use it. They get nervous. So we just need to ensure that our users understand that things are different, and that’s okay.
Vanda: 24 Do you release some kind of paper explaining that, or something? Because this is important for circulating a round of posts in the ALAC, or something. Because people need to understand beforehand, that.
Tina: Right, so I probably have to ask Nick for help with this, but I know that we had an ALAC paper. I don’t know if it was posted yet. Because I had lost my computer up to the meeting, and I actually haven’t seen if it was posted, but...
Nick: There is an introduction for end-users and others new to ICANN or new to policy areas of various kinds that were released in time for this meeting. They are being translated into the five UN languages. The English version of the IDN document is on the table behind us. If that isn’t filling the bill, then obviously let us know, but...
Tina: Yeah, so I don’t think it’s sufficient, right? I mean, we have to do more. One of the things that you’ll see later on in my slides, if I actually make it through all of them, is that we’re sort of moving from a development phase and into an outreach and educational phase. And I think that’s normal when you have to do with new technology. It’s sort of like the way things go. So I think we have to do more of that. But we have a first set of that. And I thought it would be useful for ALAC to maybe be working on. I mean, you do have access to users, and it would be a great thing for us to work with you on.
All right. This... I’m just going to skip this. You can go back to it, if you want. This is just to show some of the different ways that IDNs were implemented, and what the different results are you are getting with different browsers. And you can’t really see it from here anyways, but you get the slides, and then you can take a look at it, or you can go and try it yourself. Or you can come and ask me later if you want to try this out and see how it’s different, and I’ll show you how it’s different.
So the next topic I had for you was, or is, implementation experience. And this is really... has much more to do with companies existing, or future registry operators who would like to operate an IDN TLD, or even IDNs at the second level. However, it does have a user component to it, so I decided to include it in this presentation. Because one of the first things a registry operator is deciding, when they want to implement IDNs, is which characters should be offered. And they do this differently. Some do it by, well, we’re going to take everything that’s in the formal languages that we use, or our users are using. Others make surveys to the users available and ask, “What kind of characters would you like to have in domain names under this TLD, or... and others, again, have to do with legal matters.
So things are done differently in different regions. And under different TLDs, I mean, ccs and gs are different, so it will be implemented in a different way. But the first thing is, which characters should be offered. And then it goes into launch procedures and registration policies. Is it first come, first served? Should there be IP rights, or existing registration rights? So if you already have ASCII registrations, should that give you a pre-right to make the IDN registration, and so forth? And you can do this different ways, and it is done different ways. However, what’s common for all of them is that they follow the IDN guidelines, and they follow the IDN protocol. And that puts some limits on the characters as well, and the way that you can mix characters. I don’t know if it’s something that we should go into more details on. I didn’t put more in that what is just on this slide, but you can ask questions if this is of interest. But again, registrar and user education is something that becomes increasingly more important.
Okay, so this is sort of like half way through of what I wanted, and this list... this list that is on this slide is an overview of what still needs to be done. We have the IDN wiki, we have the IDN TLD processes at ICANN and at IANA. I’ve included one of the main policy questions that I get from users, that you may be interested in. There is a protocol revision, and there is the guidelines. And I do have slides for every one of them, but I thought I would take a quick check and see if there’s one of these topics that is of higher interest, or a couple of them that are of higher interest to you, and we can take that first. And if not, then I’ll just go through it.
Vanda: The questions, I believe you could explore a little bit the questions from the users, what is the main question from the users?
Tina: The main question, I see, because we....
Cheryl: Hong, yes.
Hong: Yes, I have a question on IDN technology. Tina, thanks for your briefing. Just now you talked about browsers for IDN implementation. That’s very important. But there is another technological issue that’s equally important, and I guess it has not been focused so much, is now WHOIS. It is so important, and look at the IDN on the second level, and we can see the WHOIS is not readable at all. What we can see is dot com, dot net, at top level. Now if it is a pure IDN, there’s nothing left there. I’m sorry, I really want to know – are you working with IATF on some new standard on this? Thank you very much.
Tina: Okay, so WHOIS is pretty important. I am not working, and ICANN staff is not working with the IATF on a new standard on it. We are looking into what are the different options are available, and should there be a new, but we’re not working on anything new at this point in time. On the web-based, some registries and some registrars will show you the Unicode characters, the local character strings, and the x and dash-dash string. Others will only show you the x and dash-dash string, which in most cases doesn’t make any sense.
So it is an important topic, but not only for the domain name portion, but also for the contact details. Because what if we suddenly have all of the contact details in one language that... you know. We have that today, right, because it’s in Latin characters. So we need to find a way for how to make that usable for the future. And I don’t think that’s an easy question to answer. I know that the GNSO is looking into who is as well, and with IDN relation to it. I haven’t seen any technical standard that is good at solving it. You know, several have been attempted, but it’s an open thing as well. There are several other open topics on IDNs than what I’ve listed as well, but that’s one of them.
Okay, so domain policy question is probably important. So I’m just going to skip quickly through the wiki, and I think, is everybody familiar with the wiki? We took the word “test,” translated it, put it in as top level domains, as IDN TLDs, and opened the wiki at IDN.ICANN.org. And if you have not tried it out, you should go try it out, because this actually will give you a good experience of what is a fully localized domain name. So IDN at all levels. And it works as an open wiki. You can go in and set up your account and set up your pages, and try it out, and it works really well. We keep adding languages. The latest was Amharic, Hebrew, Thai and Urdu. And not all of them are at the top level, but we’re looking into providing them at the top level as well.
We’re trying all kinds of stuff in here. DNSSEC, you know, anything that you can imagine, we’re trying D-Name out, we’re trying all kinds of different things out. We’re trying to figure out if we can do something with the new e-mail as well. So this is a complete test area.
So I’m just going to skip up to... so this is the one main question that I get from users, and it is to some extent policy-related. If I registered and then IDN dot some TLD then will I also be the registrant of IDN dot and then the IDN version of that TLD? So if I have... if I have my name, you know, TINA dot DE, for example. Well, that’s not an IDN, so... Tina with an "a" with two dots over, dot DE, then will I also be the registrant of TINA dot and then DE, and whatever IDN equivalent could be. It’s a bad example, but I’m just trying to not make up a sensitive example. Laughter
Cheryl: Oh, you could use it.
Tina: Yeah. So, and it’s really difficult to answer, you know, on a global scale, because there are policy considerations, right? I mean, first of all, the question sets a precedent that you actually can do, can make an IDN version of the TLD available. So I picked DE, and said that’s for Germany, right? But what about some of the gTLD ones. How do you actually translate some of those? Do they have a... you know, some of them kind of have a meaning you can translate. Others don’t. But if you do, if it is possible to make an IDN version of the TLD available, then in order to be the registrant under that TLD, you know, first of all it requires that it’s the same registry operator. And that’s not clear either, if that’s going to be the case or not.
Under the GNSO policy, it says that there is no precedence for existing registries to become the registry operator of the IDN TLD. So it’s not a guaranteed right. It doesn’t happen very easily. So that means that maybe you won’t be that registrant.
I also wanted to do GNSO holder’s objection rights, which might help. So if somebody applies for something that looks like or means the same as what dotcom means, well then Verisign presumably would make an objection like that, because the wouldn’t want someone else to apply for that, and operate that.
So there’s a little bit up and down in terms of what the answers to your question is, and I think the only thing we can do, at least on the gTLD side, is to wait and see what do different companies apply to, who wants to be a registrar operator for what. Because there is no yes and no on this answer. And that can be a little bit confusing to users because it’s hard to go out and explain to them, “Well, we have these policies, and they are being developed this and this way.” But it’s important for users to understand that that’s how it works.
On the ccTLD side, both the ccTLD operators and the GAC are considering if there is any need for what refer to as an aliasing solution, which means that automatically, if the dot TLD is aliased to a dot IDN TLD, then you automatically become the same registrant. That’s the technical solution behind that. And if that need is requested by the ccTLD operators and the GAC, well, then we need a technical solution for how to enable that. That technical solution does not exist today. d.name has been proposed, but just so you know, initial testing for d.name being used for aliasing did not give very positive results, so... And it was just initial. So we’ve asked technical communities to go ahead and try to get a little bit deeper into the details and see if there’s something that can be done.
So it’s one... it is the main question that I get from users. I don’t know if you recognize it or not, but unfortunately, even though it is the main question, it also is a question that does not have a straightforward answer, so it’s a difficult topic.
All right. We have guidelines... I want to mention the IDNA revision as well, real quick, and then I think I’m going to pass the word to Bart. So the protocol is currently under revision, and one of the reasons for it is that the initial version of the protocol was in 2003. And it was fixed onto Unicode 3.2, and as you know, Unicode keeps expanding. So any new characters that were added into Unicode after 3.2 automatically are invalid for IDNs. So the protocol had to be revised. During that revision, we found some other issues with it, so several things are being fixed in this revision, so that’s good, because it means that it’s a more secure and a more stable technical standard that can be used.
If you’re interested, RFC 4690 is not too technical, and it actually just goes into different issues and reasons and suggestions for solutions to the different problems that exist with IDNs, and I think that it would be a very, very good read, even if you’re not very technical. It is an RFC, but yeah, mainly, the revision means that the protocol will be Unicode version independent. So that means that no matter what kind of version of Unicode you have, as long as you apply the IDNA protocol, it works. And the way that it does that is that it’s using characters by... well, so a character in Unicode has different rules associated with it, and the protocol is based on that, as opposed to just being based on a table of characters. So, when it’s posed on the rules around the character, it means that any character that you put into Unicode will get a result out of the protocol. And the result it will get is either protocol valid, which means that you can use it, or disallowed, or unassigned.
Now what that means is also that that currently is a draft protocol revision completed, that will show you what characters are protocol-valid, and which ones are disallowed. So that might be useful for you to take a look at and see if there are any characters in the disallowed section that you think should actually be protocol valid, and then bring that up to the IATF working group. This is something that is scheduled to be completed by the end of this calendar year, or before even, and the thing is that once a character... once the protocol is finished, once a character has been disallowed, it will never become protocol valid. I mean, this is the change. So if there’s anything in there that you want to make some objections to, you have a chance now, and it’s not going to be changed at a later stage. Because we can’t keep changing it back and forth. This will now be fixed.
In terms of IANA and ICANN, IANA have processes for managing IDN TLDs. And we expanded on those to handle IDNs, and we have this review that is almost done, and I put down before 30th of June this year, and we’re going to publish it publicly. And I think that’s going to partly be done in presentations or in status reports online, but we did review it and we have been... we have identified several different fields of additional information that IANA needs to hold in order to manage IDN TLDs.
And those are going to go into the processes, which are on the next slide, and this is going to lead up to Bart taking over, because there’s different allocation processes being built right now. There is the fast track, that Bart is going to go into details with. That has to do with IDN ccTLDs. There’s the long-term PDP for country code IDN TLDs, that Bart also is the policy manager for, and then there’s the new gTLDs, that Carla will talk to you guys about, I think later today, right?
So these are the three processes. And why do we have these? Well, we have them because there are no existing processes that can handle IDNs, so we need something new. And yeah, I think I’m just going to hand it to you, Bart, and go through the fast track.
Bart: Yeah, that’s fine.
Tina: Okay.
Bart: Yeah, it’s easy, otherwise I blabber away without... Careful with the water.
Cheryl: Yeah, I’m just moving it out of the way. It’s not that I don’t trust you, Bart.
Bart: Well, I don’t trust myself either, at this stage. I hope it works, but now I need to do ugly things which I don’t understand.
Tina: And while Bart is setting up, I’ll just say, the other topics that I had on my slides, the statuses of those are still in the slides that you’re going to get, and I sent it to Nick, so I think they’re going to be published, or some way. Yeah, so if you have any questions on it, you can always send me an e-mail or catch me at some point. So...
Cheryl: Thanks.
Bart: Voila! Magic.
Cheryl: I was going to fill in, but there we are.
Bart: Magic. Magic happens. Um... Let me start... Good afternoon, by the way. My name is Bart Boswinkel. I am, I think my official job title is ccNSO policy advisor, but I do other things besides ccs as well. So thank you for having me here.
Let me start off with explaining a bit about the relation between the fast track and the overall policy. And it will be very brief, but then you understand why there was a fast track needed, and how it fits into the overall policy. And that will... is just a few minutes. And then I will go into more details in where we currently are in regard to the fast track. And I have to warn you in advance, during these days in Paris, say the draft, final recommendation is constantly changing, so we don’t know which, what document will we end up, if we end up with a document at all. So it’s just a... what I tell you is just a snapshot of where we are currently.
Okay. The fast track process, why was it initiated? It is, I think the core of it is, it’s the second bullet point, is there was a need for an IDN ccTLD overall policy, and when we started thinking about it, it became very clear, and that depends very much on the final outcome, that it could take between 2 and 7 years, and it depends very much on the choices made by all the stakeholders during that process. And the reason for that is, if you look how ccTLDs are currently defined, the core of it is the ISO 3166-1 standard. It is to identify territories and to identify the two letter codes which is the ccTLD. And if you talk about IDN ccTLD, there is no such thing. The only thing we know is there are territories, yeah, and the concept of territories is why they introduced the concept of countries. And there are territories out there who express a firm, or express a present need for IDN ccTLDs. And that became clear in the last year.
So when we started with the IDN policy, overall policy, it became clear we cannot wait this long. So something new needs to be designed and adopted by all the relevant stakeholders within ICANN. And that was the fast track. Now, maybe you recall, some of you, that the meeting in Los Angeles was very focussed on process, but that was the reason, is because there is no such thing as all stakeholders working together, say, in developing something like a methodology, or whatever you want to call it, don’t call it a policy, please. We all needed to agree on that process, and that became the fast track process, as defined in the charter, which was adopted in Los Angeles by the ICANN Board.
So that is a bit of the background of the fast track process. Now what is the focus of the fast track process. As I said, and what you can see there is no such thing as the ISO 3166 standard for IDN ccTLDs. So in the fast track process, we needed to design or develop a mechanism to select IDN ccTLD strings. And again, in the back of everybody’s mind who was involved in that process, it’s very clear it can’t be too broad, because then we start pre-empting on the outcome of the overall process, so we need to find... and it is quite the balancing act.
The second thing that needs to be addressed is, how do you designate an IDN ccTLD? Are there specific requirements in regard to the IDN ccTLD manager, yes or no? And fortunately, I think at the end of this process, or at least the people involved in the IDNC working group have a rough consensus, there is no additional need or some extra needs to designate an IDN ccTLD manager. But there are some differing views even within the IDNC working group, and there are some stakeholders in the ICANN community who think there needs to be something additional in place, but I’ll touch on that later on.
Okay, a bit more about the relationship between the IDN fast track and the IDN ccPDP. Again, the IDN ccPDP is to develop an overall policy, and the fast track focus only is 33. Now, the real relation between the two is that as the charter had defined the scope of the IDNC working group pretty narrowly, and some issues could have been and are raised which are outside of scope. The policy development process acts as a kind of... yeah, how do you call it? Bottom line, or, yeah, bottom line. So, issues will be addressed that could not be addressed in the fast track. So that’s one.
The second thing is, and as... probably that’s very clear from Tina’s presentation as well, fast track or IDN or IDN TLDs is something new. Everything is under development, both at the policy end and the technical end, and we don’t know what will happen. So, say, from a policy perspective, it means that whatever the outcome of the fast track is, it could feed, and it will feed into the PDP itself, so we can learn from the experience we went through in designing this methodology. So that’s a bit about the relation about the IDN ccPDP.
A bit of the timeline of the IDN ccPDP is... what will probably happen during this meeting is that the ccNSO council will decide that the issue report has to be published, or has to be available, at the next ICANN meeting in Cairo. It is to allow, as we’ve made so much progress with the fast track, is to allow, to really have the experience of what we went through with the IDNC working group to feed into that PDP, and there is... and I know other constituencies are still working on the answers to the questions raised in the issue paper by the GAC and the ccNSO. And that will feed into the PDP as well, so we have a very broad understanding of what the issues are.
Okay. The methodology itself, what are we talking about with regard to the fast track. I think this is a very high general overview of what is in there. The starting point is, and it will be reiterated and will be in the final report, is the Board set some overarching requirements, you could say parameters, conditions, whatsoever. The IDNC working group has to work within the limits of these overarching requirements. And these are, and I think that they speak for themselves, but they are very important, so I will reiterate them.
The stability and security of the DNS in particular needs to remain, needs to be ensured. Again, this things with the experimental nature, and feeding into the PDP as well. The IDNA protocols and the IDN guidelines need to be adhered to in the selection process, and again, that is one of the issues the IDNC disagrees, what are we going to do on an on-going basis.
Another very relevant element is the input of the technical community. If you look at the final report, you’ll see that the technical community, both from outside ICANN and from the SSAC is working within and working with other members in the IDNC working group. In particular, it’s Steve Crocker, the Chair of the SSAC, and Patrick Feldstrom, and there are others who are acting, say, in more or less a double capacity, or act as an observer. So we’ve ensured that the technical community is involved in the development of the fast track process.
And last, but not least, and I think especially for this group which is relevant, is that, and this touches upon what Tina has said as well, the current practices for the allocations of ccTLD remain overarching. It means that the IDN ccTLD manager has to be selected in territory. It has to go, and it has to have the support endorsement of the local Internet community, which includes end-user groups in that territory. So, and that is one of the main points I think, main differences between gTLDs and ccTLDs, and again, this is an overarching requirement, so this will happen with IDN ccTLDs, if there will be one, as well. So there is a major role, or the end-users can play a major role in the selection of the IDN ccTLD manager, that’s one, and if you look further down the process, in the selection of the string representing the name of the territory as well. Because in designing that mechanism for the selection of the string, we have more or less copied that process into the mechanism.
Okay, so that is the overarching requirements. The guiding principles, I’ll just go through them, but they were very important in order to set, say, context to the methodology itself. There is a debate, should they... are they part of the methodology, are they part of the real recommendations. They set the context. One of the, I think, major guiding principles, which is undisputed, is that the fast track should be an ongoing process until the ccPDP recommendations have been adopted by the Board. Now this is jargon for saying if the process will take five years, say IDN ccTLD managers could be requested under the fast track for this period of five years. And you don’t have to rush into the fast track to get an IDN ccTLD, or say, you should be felt left out because you’re not ready, you’re just starting your process. It is to ensure that territories with different speed can enter the fast track whenever they are ready. And that is the major, I think, one of the major elements in the guiding principles.
Another one is, and that is contentious, there are alternate views on it. Another one is that say the IDN ccTLD string has to be non-contentious within territory. And the IDN... some view that that should be broader, that the non-contention should be across the Board, so outside the territory as well. And one of the discussions we are having within the IDNC and especially with the GAC is this will impinge on the sovereignty of nations to determine what the name of the territory is, so it’s a fun discussion, in a sense.
Cheryl: Oh yes.
Bart: I think these are just the highlights of a few of the major principles. I think one of the other elements which is very important for the methodology, and it re-emphasizes the current practices for the allocation is the working group thinks that the fast track needs to be a three stage process. Let me explain a bit. It think that the major work, and it is to express that the major work to get an IDN ccTLD has to be done in territory. That’s what we call the preparation phase in territory. This is where a string is selected, this is where an official language needs to be identified, this is where you identify and select the IDN ccTLD manager. This is where, and this touches again upon what Tina was saying, this is where a language table or a script table needs to be developed, and be made ready to submit to IANA, because that’s one of the requirements. So that is all in territory, and again, there is again a role for all end-users in territory to participate in that process. And you can drive that process. And the good thing about designing it this way, and that’s one... it allows say that, and it re-emphasizes, that every territory can do this in their own time and in their own manner and at their own pace. We know some territories have gone a long way in their preparation, so they will be ready quickly. We know as well that some territories are just starting that process. And again, you have to go through some steps in order to make it work.
Okay, the second phase is what we call the due diligence. And this is particular for the selection of the string. And there are two issues there. If you will see the report, is the due diligence itself has now turned into being a technical review of the string. Again, this is to ensure that as soon as an IDN ccTLD string is selected, that the security and stability of the DNS will... is ensured. Also it’s to ensure compliance with IDNA protocols and IDN guidelines. This is the technical review, and that is, I think, accepted and adopted by all in the IDNC working group, even the technical, although most people in the IDNC working group don’t understand the technical requirements, they accept that it’s necessary.
If you look at another element of, and that is under heavy debate, especially again from governments, is there is this requirement that the string needs to be meaningful. And meaningfulness is defined in a particular way, but the essence of it is it needs to represent the name of the territory. Now, if you set a requirement like this, you want to ensure, and again, with... in the context of the PDP, that the string really meets that requirement. And you want to avoid that something will be used or entered or delegated which doesn’t meet that requirement. Now, the best way to ensure this, probably, is to have a very heavy committee checking this, but you can see already that’s one end of the scale. And the other end of the scale, which was proposed by some, is if we say this is the name of the territory, that’s it. Because we are sovereign, we are the relevant public authority, and we came up with a name... if it’s the name of the King, that is the name of the territory, because we say so.
So it is, again, a balancing act between, on the one hand, this free for all, and on the other hand, a very, very heavy procedure just to meet these requirements. And again, say this morning we had a discussion with the GAC on this, and yesterday within the IDNC working group as well, and if you have seen the latest draft, what we now suggest as an IDNC working group is that UNESCO will be involved to authenticate that the string really means what it says, that’s one, and that you have documentation and evidence that the string from – so that’s a self-assessment – that the string meets the criteria. So it’s a way around of interfering with the sovereignty of nations again, because that was, I think, the major push-back.
So that’s done in due diligence. One other element which is part... or two elements which are part of the due diligence are, say, the due diligence starts with the submission of the selected string, which is obvious in different formats. And the other one is... we are aware... what needs to happen as well is that the language table, which is developed in the first stage, so in the preparation in territory, that the language table is submitted to IANA in accordance with the IANA procedures for submission.
Now one of the things we already know, based on this methodology, that part of the implementation is changing the current guidelines to enable this, because according to the current guidelines, and please correct me if I’m wrong Tina, according to the current guidelines for submission of language tables, it can only be done by a ccTLD manager itself.
00
Yeah but it’s... by an operator, but if you don’t have an operator, or just an intended operator, he can’t submit and so you’re causing a catch-22 situation. So this probably needs to change, and so it’s part of the implementation plan, and so although the fast track recommendations are there, we know we have to go through an implementation planning phase, as ICANN staff. And I think everybody on the IDNC working group is aware of this as well.
Okay, the involvement of actors, I’ve touched on that bullet point. Then I think what is... what we try to do with the IDNC working group, and if you read the report in that sense, it makes it easier to understand, is to turn it into a checklist type of process. So it allows the people in the territory, the stakeholders in the territory, to check, where are we in the process? What do we need to do in order to make this work? And who do we need to involve in order to make this work? And it also allows, and that’s a good thing as well, and this goes back to the IANA delegation’s principles, or even the underlying principles of ccTLD delegation. If there is an issue in territory, it needs to be resolved before entering into the fast track itself. So if there is a... the end users in a territory do not agree with the intended ccTLD manager, and they can make a strong case for this in territory, and this becomes part of the documentation, or other since it is clear it is a contended IDN ccTLD manager, and therefore not ready to be delegated. And the same is true with regard to the selected string. Okay, I think I covered everything in that one.
Okay, this is, maybe this one is interesting as well. Tina touched upon it the IDN protocol review is still underway. When we started this process on the fast track, everybody was aware that whatever we construe for the fast track is dependent upon a stable IDNA protocol. Now, worst case is, recommendations are out there, but the IDNA protocol is not ready. And by that time it is very clear we are in an awkward situation. I don’t know what will happen, but it will have impact on the timing of implementing the fast track, because you need to have a stable IDNA protocol, you need to have stable IDN guidelines before you can make the fast track itself work.
Another element which was brought up during one of the last IDNC working group calls, and which is part of the recommendations but doesn’t fit in to the methodology itself is that ICANN should organize, or it’s recommended to organize an information process. Maybe some of you know that when we started the whole IDN policy discussions and fast track, the ccNSO sent out a request for information to ccTLD managers, and I think we had about 40 or 50 responses, by heart. But what needs to be done is in order to prepare the implementation plan, the suggestion was to go out and seek information. Who is really interested? What do you want? So you have an idea of scale, what you’re talking about. And again, what should be stressed here, it is not mandatory to participate in the information process, but it would be very, very helpful in order for ICANN and the broader community to prepare on what will be coming. Because it makes a difference if there is a rough guess of 30 to 40 IDN ccTLDs, or if there will be 200.
It’s a scaling matter. It’s resources, and so... but this is not part of the fast track itself, it’s part of the implementation plan. What is also important to stress, I think, is that the IDNC working group recommendations are at a very, very high level. It is... I think I touched upon some of these issues, for instance the change of the IDN guidelines. Details and other processes need to be adjusted, or designed and implemented, and again, the IDNC working group is very aware that this is part of the implementation plan, and it will take time. You can’t do that... and it has to be done thoroughly and so that everyone’s comfortable... it has to be done, say, after this meeting.
And what I just said, adjustment of current practices and processes. Again, adjusting them is part of the implementation itself, but as the IDNC working group is not aware of all the issues involved, it is recommended that ICANN staff goes through all their processes and practices, to assess what needs... if there is a need for change, and what needs to be changed. Then come back with the implementation plan. I think these are more or less the highlights. I can go through all the agreed principles, but I think I touched upon them. Yeah, I will send this presentation to Nick so you have some more background information, but just a pre-warning – this is in flux, so this... say, the first couple of slides will not have changed, but this will have changed, probably, in a day or two. Okay, thank you.
Cheryl: Thank you. Are there any questions? Yes, okay, I see Luc, I see Hong, in that order. And Izumi.
Luc: First is about free speech. The IDN TLDs are for regions, for geographical regions. How will be political hot areas handled? I think about Palestine or Kashmir, yeah. It can be very difficult. But is there any process in planning, or is there something developed – just say yes or no.
And second question, what happens on registry changes? There you have two to five years for new registry to come in, so we have to deal with registry changes about every eight years. What’s with the content of the zone, will it take it over? Or you can just say it’s the topic of the next session.
Bart: My answer with regard to your first question. I think that it is very important to understand, with regard to IDN ccTLDs what we’re talking about. It is very narrowly defined, and very precisely defined. IDN ccTLDs have to do with territories listed in the ISO 3166. So it is arduous, and the reason for saying it this way is that all these regional issues are not IDN ccTLDs. That’s one.
Secondly, your second question. Sorry. Again, I think that is very important, and it’s an overarching requirement, and it’s an overarching principle. Current IANA practices do apply, so in the event of a redelegation, you talk about current IANA practices for redelegation. The IDNC... it is outside the mandate of the IDNC working group to touch this because that is... if you want to have the ccTLDs upset, you touch upon redelegation. So it’s billed within... again, that’s one of the elements of the IANA practices.
Cheryl: Hong, please.
Hong: Okay, I have two questions, and one comment. In our committee, I always want to talk something, but there are so many other people, it is very competitive. And I’m patient, so I’m not that competitive. I’ll leave my question here.
The first question is related to our very recent discussion on the scope of this fast track is it’s limited to non-Latin scripts, or non-ASCII scripts. This is a very critical issue. I guess we can speak German, such as the word können. There is an o with two "i"s on the top. It cannot be registered as such in the present DNS. It is non-ASCII, but it’s Latin. Can such a word be registered?
Cheryl: 00
Hong: Okay, that’s the first question. Can you manage the second one? Well the second one is about a Power Point. I see there’s a very interesting sentence at the end of one slide – I can’t find it now. It means, for the contentious issue now, that our status after very long discussion, the present proposition of our committee is that no action until all parties agree. I want to know – is it within the territory or across the territory? We have been talking about this issue all the time.
I still have a comment or suggestion here. This is about evaluation of the streams and managers. I want to know how the idea of a language committee, linguistic experts committee. Is it still there? If it’s still there... it’s not there anymore.
Cheryl: Hong, I told you today, we got rid of it. LIP is gone.
Hong: Oh, the LIP is gone. Okay, so how about the string verification function...
Bart: It’s... going to your, first the last one. That’s what I said, is if you look at the current draft, there is a role for UNESCO to verify that the string means what it means. That’s one. The second bit of that process is that a kind of self-certification of the selected delegate or somebody else, that the string meets the criteria in linguistic terms. And as far as I know from this morning’s discussion, and this goes, say what I said between the continuum of say, everything goes, and you have a very powerful, veto-like linguistic committee, this is somewhere the middle grounds, where the minds meet, as of this morning.
Okay, so I think going back to your non-contentiousness issue within, in the broader sense. I think based on yesterday’s discussion, and there is still a possibility for the IDNC working group members to submit comments of course, but according to yesterday’s discussion, my impression is that the majority, the vast majority of the IDNC working group is very comfortable with the current definition of that principle is that it has to be within territory, and not outside. There are, however, a few alternative views based for different grounds or different reasons, who think it should be broader. And that is recorded and it’s part of the report, but as an alternative view. And those who support that view are named in that bit as well. And I forgot your third one. Oh yeah. The non-Latin one. Again, since yesterday, the definition of what is non-Latin has changed again.
Cheryl: Sorry, Hong, I did miss telling you that. I apologize.
Bart: It is... let me be clear. I think the difficulty is to find the precise definition of what is non-Latin. We all know, think what we understand by it. It is... is doesn’t include, for example, the German 39. It doesn’t include the Dutch A. It doesn’t include all the characters, the diacritic something... diacritics. So... and if you think about it, if you need to have a language table of just these diacritic characters, you’re going to have weird registrations. So, in that sense, I don’t understand the discussion, really. But it’s very clear it is for non-Latin in the sense of the Western Europeans are out. Yeah.
Cheryl: Izumi, sorry.
Izumi: Thank you very much for the very interesting presentation. Several questions, and perhaps a comment. First, a naive question. Are the evaluation processes for strings and managers separate? Am I correct? Totally separate, so that you don’t give any priority to it?
Bart: Priority is probably the wrong word. It is... there is an order in it. It is... it doesn’t make sense to enter into a delegation request, in say in the sense of current practices, if it is not clear that the string for which it is applied works. So what will happen, as... that’s why, say, it is a three stage process. And I think that the most important ones are the second and the third ones, is first due diligence, so that is checking, I’m using the incorrect word, but checking the string. Does it work, does it meet the criteria, technical and meaningfulness criteria? And then you go into the delegation. So that is the designation process.
Izumi: Thank you. Because in my country’s case, as a territory, that’s pretty interesting. That no other country uses the same name, Japan or Nippon, the same character as well. So I just spoke with the government guys and the ccTLD manager a few days about coming here, about this. But the... our government is considering, for your any reference, to go for very formal process of calling for which parties may become a TLD manager. They suggest it already. It’s almost the first time in Japan to give a very sort of legal, formal procedure to any given ICANN policies, as far as I can remember, for the last 10 years.
18 Japanese character country code and its manager. And they use similar to the telecom law of ICD, sort of policy framework, although there is no specific law in Japan which includes explicitly about ccTLD. They will use some general, very general law of the framework that set up the Ministry. That’s what I heard, first of all.
The question is that we have an existing, incumbent ccTLD manager. It’s a for-profit company called dot JPRS... well, JPRS. And they are legitimate, they were originally redelegated from the original non-profit, and they became sort of a commercial company, but that’s a monopoly. Whether this company will get a new additional one for the new Japan on dot Nippon or not is an interesting question, as you can imagine. So they are wondering how to select that. That’s part of the reason why they may have to go to the formal process, even though the same company will get to the same ccTLD management. But it has to be accountable to the wide public interest. So, just for your information.
But also I’d like to sort of raise the question, I have observed some countries in Asia-Pacific, or Southeast Asia, such as Cambodia, when they decide a Unicode character code set, the International Standard Organization is in charge, together with the Unicode forum, but they didn’t really listen to the native people almost at all, including the governments. And they had strong inaudible later but, you know, once decided, you cannot really change. They made some sort of, I don’t remember, some exception, some very strange characters. It could happen. Natives were not involved at all. That was five years ago.
So I’m sure you will not face a similar situation, or the same, but hope that you do some kind of outreach to make sure all the affected parties... Because, looking at the current IANA policy and guidelines, you take rather reactive, you know, seemingly, unless everybody agrees in the local front. But how do you really outreach those who are excluded from the active members.
And finally, last minute 42, you have only 60 or 70 members to the ccNSO, out of the 200 some ccTLD managers, right? So the majority of these ccTLD managers actually are not really participating in ICANN’s process. So that poses an interesting challenge for all of us to go more in a public outreach, I guess.
Bart: With regard, say, in answer to your comments, this is one of the major reasons, and if you look at the last draft, you will see it, this is one of the major reasons why the request for information process was included. And one of the recommendations in that, of this part, is there should be a real attempt to reach out to all relevant territories, to all territories. And that is, that’s one.
As to the ccNSO process, yeah, it’s... let me be a bit incorrect. laughter
Nick: Politically or factually?
Bart: Both.
Cheryl: Oh.
Bart: First of all, say, it’s over 70. Of the 250 you mention, or 265, some of them do not exist anymore. They are out there, but nobody knows where they are. It’s just a few. Some ccTLD managers are represented by one organization, which is a member of the ccNSO, so that’s on a factual basis, and some ccNSO or ccTLD managers have from day one not acknowledged the existence of ICANN. Now, the problem is, if you don’t acknowledge the existence, but yeah, somebody needs to set the rules, but you don’t acknowledge who sets the rules, then you have a problem. It is very easy to stand around on the outside and say, “Okay, I’m not involved, therefore I don’t recognize it.” But then you have a stalemate, and nothing will happen. And I think this is unexplainable to everybody who is involved, especially in the Asian Pacific region, where there is a real need for IDN ccTLDs. And it’s... you could even change the argument around, is they show irresponsible behaviour.
Izumi: That’s really... for the redelegation I remember you... ICANN requires the endorsement of the government. Is... but that is not in the official delegation policy of IANA, right? Which way are you going to do?
Bart: Let... as far as I know IANA practices, and if you look at it, it is... endorsement is sourced in territory. And if you look what is currently written on the IANA website, there is a role for government, and it’s based on, say... Either way, you can say it’s as part of the local Internet community, because they play a very important role in the local Internet community. Or it is an independent role, and that depends on the territory and how they react as well. And so...
Izumi: Of course, my question is a bit a tricky one. Are you going to... you are not going to request a mandatory endorsement from a government, or...?
Vanda: Yes, just to clarify. There is a three signatures for agreement for three parts – the ccTLD, the govern, and ICANN.
Izumi: For the redelegation or the delegation?
Vanda: The redelegation or the delegation. The first deleg- the problem is, the first delegation was just 39 away by John Postel. The second, yeah, the previous one. So each time they make this formal three signature agreement.
Cheryl: Okay, an intervention. I’m afraid our speaker has to go – we’ve delayed him from his next... So if you desire to have Bart answer these questions, we will get access to him again in the ALAC ccNSO meeting. And just to let you know, I’ve just agreed with him to move the geographic regions briefing into that space and time.
I take it that that meets with everybody’s approval? Thank you very much, and thank you, Bart. I very much appreciate it. Can we just thank you in the normal... 22.
Okay, while we get the technology running, we will note that we are a few minutes behind time, but it is for one of the longest presentations. And I’ve had a little chat to our presenter, and he assures me that we can manage in the time given on the agenda, so it won’t be a problem. And we will be taking a coffee break at one time.
Are we set? Nearly, nearly. Almost there. We’re expecting Carla as well.
Olaf: very faint – not on mic I’m afraid not. She sends her regrets.
Cheryl: Okay, thank you for that. Let’s... right, everything’s working.
Olaf: Ah yeah.
Cheryl: This is good. It gets to a full screen, and we’ll be fine.
It is our pleasure to now have a... inaudible voices – not on mic It is our pleasure now to introduce our next briefing, and we thank Olaf for joining us, very much. We appreciate your time, indeed the great amount of time you’re willing to spend with us this afternoon. And we’re looking at new gTLDs. Karla Valente has sent her apologies, but I’m sure that one presenter will do a more than admirable job. And the floor is yours.
Olaf: Thank you very much, and dot hello, everyone. laughter Or dot bonjour. Or dot hola. So, this may for quite a few of you give some déjà vu experiences, because we met on... well, virtually a couple of weeks ago. And this is essentially the same presentation, given again, for good measure. But, enhanced with the latest developments because not only on the ccTLD side, but also on the gTLD side, this is a subject that is moving. So, there are new... some precisions added to that presentation, in addition to the very catch background, which has now been added to make it fully up to date. And the catchy subtitle remains the same – what kind of Internet do you want? And that’s... well, would prompt, I suppose, a number of comments from most of you, either during the presentation or afterwards. But let’s get on with it.
So, just to jog our memory a bit, well, you’ve heard about the ccTLDs and the up-and-coming IDN ccTLDs, and well, the definition of gTLDs is basically everything that’s not a ccTLD, which is the vast minority of top-level domains as of to date. 21, currently, compared to 249, and that’s also a number that’s changing, of course, because countries come and go. Territories come and go, much like mergers and acquisitions in the business world. laughter
And of course there is the little pizza slice with an 11 on it – that’s the test IDN ccTLDs, which I hope you’ve tried out. Otherwise, please do so. But the little purple pizza piece, that’s what we’re going to talk about now. And, of course, the gTLDs already exist, but there are going to be quite a few more of them, if we get it right.
It’s not the first attempt, of course not. Already pre-ICANN, of course, gTLD existed, dot com, dot net, dot org, to mention a few. There are special ones, as well. We had two, one could say, experimental rounds in order to get a feel for how is it, and how do you do this business of assigning generic top-level domains? And with the so-called 2000 round, or proof-of-concept round, it’s frequently called, and the 2003 round, which is frequently called the 2004 round, and that pre-dates me, so I don’t know why. But we got a few additional ones, then, and well, based on the experiences both positive and negative as to process and how you select, well that formed the basis for what was then to come. But, oh yeah, of course these are bold statements. It’s certainly so that it will change quite a lot, whether it will be the biggest change, one of the really big changes in the Internet history – that is what we believe.
And, well, we’ve been all of the years quite solicited about when will the next round take place, and how will it be constructed, and how will you sort of structure the various categories of top-level domains, and so on. And it ended up being a policy development process, which you all were very well aware of, that was conducted by the GNSO, but involved input from all over the community, including from your side.
A few words on the potential benefits, and this is a bit copying from the outcome of the GNSO work. I mean, they reached the decision fairly early on, because it wasn’t very, at the outset, clear on should we really advance with new gTLDs or not. But it soon emerged a consensus on that, that yes, it’s necessary to foster creativity, innovation, choice of the consumer, and competition, which is one of the key words that we should strive for in the domain name space.
And also, it was clearly stated that the big, perhaps biggest, needs for new gTLDs was, like in the cc arena for international domain names on the top level. Of course, IDNs are available, but only on the second level. And that forces everyone to use the Latin characters, for the top-level domain, so far, for the end-user. Or nice word I heard, listening in to the Arabic script working sitting downstairs. They didn’t talk about the end-user, but they talked about the navigator. I thought that was a nice word. I just want to share that, from a quarter of an hour ago.
So, the policy. Well, at the end of two years, and quite a lot of paper and ink and bits and saliva, the GNSO produced a policy, which was adopted by super-majority, which is, well, quite an achievement on such a, at times, controversial issue. Meaning two-thirds majority, and that was reached last year on the 6th of September, to be precise. The core of it are a set of principles, recommendations, and implementation guidelines to guide what would then become the implementation work.
A few highlights, perhaps, that deserve mentioning, particularly in comparison to the previous rounds, that... well, it is foreseen to be an open approach, where you have necessary minimum thresholds, and rather than having a comparative evaluation for a few slots, this is open. There are not pre-determined limits on it, and there would be ample opportunities to file objections on valid grounds. And it’s not a pre-structuring of the domain name space, so the applicants are to propose the strings themselves.
As said, no pre-defined limit. We have basically no idea how many applications we will get, and there is not a limit on the number of slots for them, and we have the reassurance from the technical community that the domain name system can cope with quite a few more top-level domains, from a technical point of view.
The clear principle from the GNSO was that IDN gTLDs should be included, conditional of course on readiness for introduction of IDNs in the root. And that, also, when it came to more mundane matters like money, that the fees should be on a cost-recovery basis for these rounds that would come. And there was a vision that eventually we would reach some kind of steady state, a delegation procedure. But initially, it would have to happen in rounds.
As you know, ICANN today accredits registrars, and that’s a continuous process, you can apply whenever you want, and that’s what the GNSO had in mind for the long term, but I believe that I will be at least retired and probably dead by the time we manage to do that for the gTLDs.
Cheryl: Olaf.
Olaf: Yes?
Cheryl: Annette, I gather you have a question for this slide? You’re happy to take questions throughout this?
Olaf: Yeah, yeah, absolutely.
Cheryl: Go ahead.
Annette: IDN gTLDs included on readiness. Could you just explain it a little bit further? I just didn’t get it.
Olaf: Well, oh yeah, that was perhaps a bad choice of words, because we had the readiness debate in the ccTLD world, but here it’s nothing about that. It’s on... technically, it’s doable, and we have sort of a clear set of rules. Preferably in the IDN, new IDNA protocol ready for introduction, so there is no question marks in that respect.
So it’s not a matter of readiness in the territory or by the gTLDs or anything of the sort, but it’s sort of the technical preconditions that should be there. Is that okay? You look like you’re happy. Okay. All right.
So, also the policy development, just to rehash a little of that, it was focussed on four themes, and just to... and that’s basically the terms of reference looked like that for the new gTLD committee that GNSO had. First question, whether to introduce new gTLDs. Of course, setting the overall objective, and that was pretty soon consensus on a resounding yes, although that wasn’t clear from the very outset. But after two months of discussions, that was clear.
Allocation matters. How do you proceed - well, should it be allocated by an auction, or by a comparative evaluation, or first come, first served, or whatever.
Selection criteria. Likewise, those set requirements for the process, as well as requirements for the applications. What are the thresholds that the applicants and the applications would have to meet in order to be deemed satisfactory for adoption.
And, of course, contractual conditions, and this was a key piece 37. A streamlined process calls for some kind of baseline agreement, where exceptions and deviations should really be on an exception basis. So that was the objective from the policy development side.
So, the new gTLD program has actually started long before the policy development was finished, because we realized that this called for work in parallel. So that it has been on-going, organized as a project to develop the various criteria and the processes and the tools that we need in order to actually effectuate and realize what the policy outlines then have sketched out.
And that’s conducted by ICANN staff, with the assistance of quite a few outside consultants for various bits and pieces, in order to draw on the available expertise out there. So, let’s start where we... and we, of course, same caveats as have been mentioned earlier. This is based on what we have right now and where we stand right now. It’s not finalized, and final decision have not been taken. But just to give you the flavour of what it looks like, let’s start with a little about the applicant.
Questions could be raised, for example, on who could apply. It wasn’t clearly set out by the policy. A question was raised, should a private individual be able to apply? And the answer to that is quite simply no, for the simple reason that what we do for the registrar accreditation, and that’s sort of seen as a minimum requirement, well, we have to be established, and basically if you run a new gTLD as an individual, and you happen to be run over by a car, and we have to deal with your widow and kids and so on in a contractual relationship, it becomes pretty sloppy, doesn’t it?
So, an ability to have some kind of contract... because this is a requirement which is slightly different from what the situation is for the ccTLDs. Here we have a firm contract, actually a three way contractual relationship between ICANN and the registry, between the registry and the registrar, and between the registrar and ICANN as well. As a basis for what is then the registration contract for the registrant.
So, but apart from that, any public or private established entity – it has to be established beforehand – from anywhere in the world. And there will be application steps and rules, which are encapsulated then in an online interface. So applications will be received online, and I’d say online only. What we’re looking for is a streamlined process.
They must demonstrate, satisfy minimum thresholds when it comes to organizational, operational, technical and financial capabilities in order to satisfy what we call stable registry criterion. I mean, we wouldn’t willingly go into a situation where we know that a registry could collapse within a year or so. So that’s the kind of check and ground checks that are made initially.
And they certainly must pay an application fee, and this is a multi-step process, because you have... in order to... you have an initial application fee, which is most certainly going to be at a reasonable level. But then, depending on what path the application happens to follow... If there are objections raised, well, it should be cost-based or cost recovery, so those will be additional steps that will have to be paid separately. But we will perhaps come back to that.
To have it... Yes, please?
Evan: Are you going to be talking... Sorry, my name’s Evan Leibovitch. I’m the Chair of the North American region of At-Large. Are you going to be going into detail about the nature of what the objections would be? Because there are some issues that are very, very deeply of concern to this group...
Olaf: I know.
Evan: ...regarding the nature of the objections.
Olaf: I know. We’ll get there, but... okay, maybe I’ll try to race it through to that. But you will certainly, it will certainly be covered, yes indeed.
Cheryl: Carlos?
Carlos: 56
Translator: I understood what you were saying with the idea that it would be a legal entity and not necessarily a physical person in terms of the incorporation of a new TLD. But this idea of that it has to fulfill certain criteria, that it not go bankrupt, that, you know, if that’s your logic, I’m not sure that I agree. Because there are so many different companies that go belly up in our region, I don’t know that you could consider that a criteria.
Olaf: Probably. And also, as I tell you, this is the current state of affairs. And we based our considerations, when it comes to physical persons, on the current requirements for registrars. So you cannot become a registrar unless you’re an established entity of some sort, meaning not a physical person. On the other hand, well, there may be numerous differences, and we’ve got plenty of jurisdictions around the world where the forms of incorporation may be very, very different from what we are perhaps accustomed to. So we have to be prepared for this as well. So, as I said, this is not the final word. But it’s sort of the base assumption that we were starting with. Does that serve as a response?
Cheryl: Carlos, 46.
Carlos: Yeah, thank you.
Cheryl: Okay, could you press your button off, so I don’t think you’re... press your button.
Carlos: It looks nice and red, though. I mean, no problem.
Cheryl: Excellent.
Olaf: It was a strong requirement in the policy that information should be available beforehand, for the applicants, exactly on how the procedures... how the application process should handle their application. What the requirements would be, and what the final contract would look like. So this requirement for up-front transparency in that respect, that nothing should be added as an emergency kind of provision, but everything should be there beforehand. It’s something that is putting quite high demands on the implementation process, but it also requires us to have a fully comprehensive request for proposal, with all these pieces in, which will then be put out for public comment. And based on the public comments, will produce a final request for a proposal, and that’s a conditio sine qua non for launching the next round. So we have to have that ready – that’s an essential piece.

We have an application period that opens and closes on defined times. And in the meantime, there will be, one could call it radio silence. We don’t want to... there will be, that... there will not be a transparency, “Who has applied for what string,” until the very end. And then it becomes public, for fear of any copycat situations.
Cheryl: Okay.
Olaf: It will be web-based, and web-based only. That could be a challenge when it comes to multiple documents, and we’re trying to deal with that since we will of course request formal documents of various sorts, and how to have them certified, for example. That’s certainly one of the minor issues that we’re trying to address right now. And there will be customer support available, hopefully through, for example, our regional liaison structure, with outreach, and in various other ways, to help applicants beforehand.
So, if we have... the simple case is really like this. It should be a streamlined process, and it can be a streamlined process. In the case, we have an application that really goes straight through from application received, evaluation done, and then, if it’s satisfying the criteria, all right. The base contract is signed, and we go into the delegation process in a practical sense.
But there are other tracks that can be necessary. So, resolution of string contention, in cases there are more than one applying for an identical string, or for confusingly similar strings. And also, once the applications have been announced, after the closing of the application window, that we do receive, or they do receive, objections of various sorts, on the grounds which we discuss a bit later.
This is a little more elaborate, perhaps we don’t need to go through everything of it, but the opening up of the application, and the close of the application process is the next step.
Review of the applications for completeness, because there can be stupid mistakes. Somebody sent the wrong page, or copied them, or only copying the back side, or whatever. But that kind of curing of applications will be possible. Modifications in a substantial... in a substantive sense, won’t be possible.
And then we have the posting of all applications, which enables those who wish to object, for example, to do so. And I go to initial evaluation, which is then the quick process. We do a first check of whether the string is confusingly similar to any of the reserved names, or to existing top-level domains. Or to other applicant strings.
And there is also a DNS stability check that it satisfies the criteria on what a string should look like from that perspective, and perhaps also with some kind of public call, if there is something we haven’t really realized that we should have checked. If there are problems in that regard, well, it grinds to a halt. But if there is no string confusion or DNS stability concerns, well, we go further, with the business criteria evaluation, and technical criteria evaluation, so that we have a stable registry.
And if they pass, well here it says, “Contract execution,” that’s not... no, no, contract preparation actually, because the Board would have to have the final say, that’s the structure we have, and they would have to improve. They will not be obliged to approve, of course, because this is the structure, but this is the step which is necessary to have.
And there will be an additional check before delegation to verify that everything that has been stated by the applicant when it comes to technical criteria, really is in practice fulfilled, before the delegation has taken place. And then it is added to root zone.
To check a little more in detail on what we do in the initial evaluation, this is the review to see, to safeguard that they do satisfy, at least on paper, the minimum criteria which have been stipulated, when it comes to operational experience, technical and financial capability and such.
And there’s also the check that the string doesn’t lead to technical instability, or we will have some sort of call for whether, or that’s foreseen, whether somebody can find some kind of, well, of course, justified reason why this particular string can create a problem in the DNS. Just in order to say that, well, maybe we don’t know everything beforehand.
It certainly must not be a reserved name or existing top-level domain string, and it must not be confusingly similar to such, either. And it must not cause string confusion, which is, well basically, a consumer protection measure. We don’t want to see things like citbank.c0m, really.
If they don’t pass the initial evaluation, there is a choice for the applicant, in case, for example, “No, no, you got it completely wrong.” And, well, let us explain. “This is our financial stability, and we can really prove that we satisfy all you criteria. “And there is a possibility for them. It will cost them a little extra, because they weren’t sufficiently well prepared, and we should recover our costs. But there is that possibility to go through an extended evaluation – and that’s the choice of the applicant.
And they, again, may pass or fail. The same criteria will prevail, and will be used in the extended evaluation as in the initial evaluation. So it’s more for sorting out misunderstandings of various sorts.
Here we go with the objections. The Objection with a capital O, eh? There will be, following the announcement of the application, a pre-defined period for launching objections. It will be handled outside ICANN. We are a technical coordination body, and these objections are not really within the remit of ICANN’s expertise. So, they will be handled by external dispute resolution providers that then hopefully be the best we can find for those particular areas. And discussions are on-going with such dispute resolution providers.
And a potential grounds for objection, just to get back to what is probably one of the most debated areas. That’s string confusion, perhaps not the most contested, but string confusion can... is checked for visual and alphabetic similarity early on in the process. But there is possibility for launching string confusion on whatever grounds, going to meaning and to such and so on. And to the extent that it would cost, then, for interviews of dispute resolution provider... would actually cause confusion of the end-user or navigator, whatever you’d like to call it, well there may be, it may be upheld or not. But, I repeat, this will be handled then by a dispute resolution body outside of ICANN.
Same goes for infringement of rights of others. If somebody launches, well, take dot Kodak or dot Volvo, and there is objection from, respectively, a major photographic equipment manufacturer or car manufacturer, well, there will probably... well, there will most certainly will be the possibility to launch that kind of objections, which will most likely be handed by another dispute resolution provider. It will be one per each rounds.
So it’s not a forum shopping situation, really, she will be handed one by ne such a body.
Morality and public order. Probably the most debated of all. And community objection, that a community, an established community with a standing finds that the string that has been applied for is directly or indirectly directed to them, and they’re not in favour of that kind of connotation, and it’s not... the string is closely, very strongly and closely connected to the community, but the application and the applicant is not. And these grounds for objection were then straight from the recommendations from the GNSO final report, and the policy developed by the GNSO. So that’s what we’re trying to craft an implementation of.
And perhaps we should stop here to listen to the comments that may have accumulated over the years.
Cheryl: Perfect timing, thank you. In fact, people have been Skyping at me, and I’ve been putting up a speaker’s list. At the moment, I have, first of all, I have the mic, which I’m going to keep for a little bit longer. Then I have Robert, then I have Izumi, then I have Evan, then I have Hong, then I have Annette, then I have Vanda. Latin America, you’re letting me down. Okay, then I have Carleton. And Jacqueline. I’m sure... I cannot believe Latin America. Come on boys, do not disappoint me.
Vanda: I’m from Latin America.
Cheryl: I know, Vanda, but I’m waiting for something over there.
Nick: You might want to make sure that they all realize, if somebody raises their point, just say, “Me too.”
Cheryl: Yes, okay, all right then. Nick just suggested that I do something terribly sensible, which is probably what I will do. You’ve made a suggestion, I’ll take it totally as mine, and that is, if somebody prior to you in that speaking order has raised your point and it has been adequately answered, perhaps you could just say, “Oh, that we me, too.” And I’ll just wipe you off my list here.
Just to begin with, though, because I’m going to take the privilege of doing that. Two things, I’d like to know about. Is who can actually make these objections on these grounds? And whose rights are being infringed, just to begin with?
Olaf: Um... who can lodge an objection? Well, anyone can send an objection, of course, but I mean there will be a check on whether they have standing. And the standing requirements would be quite different. When you come to string confusion, well, any consumer, basically, could say that, “Hey, I’m terribly confused about this.” So what kind of standing requirement could you come up with there?
It’s quite a different thing when you come to infringement of rights. There must be some recognized rights that you base it on, otherwise you wouldn’t have standing. Drawing on my... well, that was a long, long time ago, but I was actually, without having a law degree at all, but I was something called a patent attorney. There! Remember there was something at that time called the Kodak doctrine... famous brands, for example, and I think that that is really the typical case where you would have sufficient standing. Sort of when you would anyway be declined a trademark in another sector, just because Kodak was such a famous brand name.
So, well... just to... so these standing requirements, they do vary. And did I answer your second question? Maybe I didn’t yet?
Vanda: Yeah, just clarify. WIPO has a list of famous names.
Olaf: Oh, that’s very helpful. Thank you. That was a long time ago – early 70s in my case, huh?
So...
Cheryl: The second question was whose rights...
Olaf: Whose rights? Well, if you have rights which are internationally recognized, and I think there is a phrasing there from the GNSO, which they worked on for a long, long time. I can look up exactly how it is – we haven’t changed a word on that, but rather proceeded to... that’s part of the discussion with the dispute resolution service provider on how to interpret such.
But you could, for example... well, those WIPO names, I suppose, would come in pretty handy. Thanks for that 22.
Cheryl: Okay, Robert.
Robert: Good afternoon, and thank you. Hearing your presentation, it’s probably not... it’s probably a big challenge to try and put something like this together. And getting comments from the community I think is good. I do have issues with a couple of points, and a question. I’ll probably leave the first question, which is the easiest one, first, and then I’ll make a kind of a comment on my second one.
The first one is, given the issues that came up with dot travel, and also with triple x, it would be good to know what the lessons learned from that, and particularly the appeals process, and how that’s being built in to the new TLD. That’s the non-controversial question.
The controversial question, I think, is... I’m maybe speaking in my capacity as representing the ALS 17 that’s very much involved in working with human rights organizations around the world around issues of freedom of expression. I’m gravely concerned with the morality and public order section. And so I would like to know two things. One, is a preamble being thought in terms of what are the rights that ICANN adheres to and wishes to promote. Specifically, if it accepts the Universal Declaration of Human Rights as a principle, particularly article 19, on freedom of expression, and maybe article 29, where there are some limits to that. But that is a framework that is internationally accepted, that it would be good if ICANN used.
Just mentioning morality and public order without a reference is dangerous, not only for journalists in countries, but also in parts of the world where morality and public order is used to stifle freedom of expression. This is very worrisome, and so any overarching international covenant that’s being used, I would highly recommend that ICANN use and emphasize. Thank you.
Olaf: Um... I think I’ll take them in reverse order, actually, because there is a reference to the Declaration of Human Rights in the GNSO, so that’s the basis we have. We haven’t sort of brought up any separate, additional such topics in this context at all, but there is also among the principles, the principle of free speech, which is clearly among the declared principles and the GNSO principles. So I think we’re in fairly... well, as stable a ground as we can be with such a delicate matter.
And then your first question was the lessons learned from what has happened earlier. I think that is, if anything, contentious issues that are dealt with at length, and actually absorbing enormous time, or decision time, for the Board. Over, well, more than a year. It’s... that is a lesson learned, and I think that is a very firm basis for having an objection process and trying to resolve such reactions by third parties, which are the closest to experts in those matters you can find.
Cheryl: Just before we move to Izumi, Nick has very kindly reminded me that the summary document is available in hard copy in the three languages that we’re working in today, so please take advantage of that. And I suspect, now we have it, we can have it in soft copy, and it can be promulgated in your local regions as well. And very much, what we’re discussing comes under page 15, for those of you who wish to do a 42 on that now.
Izumi.
Izumi: Thank you, Cheryl. My name is Izumi Aizu. Two questions cum comment. First is, on the website, there is a timeline or anticipated schedule, which you haven’t really shown on the presentation. Is it still valid, or very vague, or what?
Olaf: Actually, it comes later, as time lines tend to do, in a sense.
Izumi: Yeah, but...
Olaf: Now, so it’s a few slides further on, so we’ll cover that.
Izumi: Oh, I see. But the reason is, 20 made a speech at the OACSO meeting on the future of the Internet economy, ministerial debt. Almost a speech that, with a new introduction, of the new gTLDs coming soon or likely next year. Sounded like a big sell, and so what kind of assumption the people around us do have, versus what kind of working assumption you guys and we have is an interesting topic that we have to bear in mind. And when we have to finish this interesting debate of free speech and stuff like that, which are also important. But we need to have some sense of time shared, otherwise it may go on and on, right.
Olaf: It’s key to us all, and well, as said, I’ve got the time plan a few slides later, and so I’d rather take it then, if you don’t mind.
Izumi: Second is my stupid ideal question. Could users propose new gTLD names, without taking or claiming the responsibility of operating them? It’s sort of a... do you have a mechanism to do so?
Olaf: Well, you can easily make... organize such a mechanism. If you’ve got a user organization, for example, and you have, well, which is some kind of community, and it’s organized...
Izumi: I don’t mean any single name, but as a framework, for example. If ALAC, for example, started a campaign. Okay, we will solicit new names.
Cheryl: 51
Izumi: No, no, no, generic. Do we like to have dot music, dot jazz, dot football and stuff like that as ALAC, or any other, or ICANN? There’s a completely different view, on the one hand.
Olaf: Right.
Izumi: But because the current framework given is for the end-users, it’s filled pretty far, on the one hand. But you don’t really have the sense of connected to the selection of proposal. For example, if you have some incentive, like a lottery, you will be one of the 2000 who made a petition to dot music, and the winner will get some money or whatever, that’s a... I understand this is just a completely stupid idea, but why not to consider. It could be running on top of the existing program. It does not give any conflict against. So long as it’s separated from the selection of the operators. But it could demonstrate how much sizeable interest may be there, which could be confrontational. Any third party, anybody aside from ICANN can do whatever they like, but it might be good if we have... it shows our intention that we really would like to confirm what users want, and it doesn’t require to be an ALS to be involved as such.
Olaf: Oh, well. Quick remark on that. First of all, that was the original approach taken by John 20, and to have semantic structuring. So, all right, we have these. And probably, according to his forecast, we won’t need anything else.
So the next question is, how do we know that we don’t need anything else? So that was sort of prompted, perhaps the opposite stance than the GNSO took. No pre-determined semantic structuring, but rather... well, this is based on the market, and the clever applicant would probably have investigated what the user wants. Or the user will vote with his wallet and stay away from it. So, well, we don’t have any particular process to facilitate that, but I suppose that the applicants would be quite keen to get in touch with the users, and in the choice of approach, both when it comes to the string and, perhaps more importantly, when it comes to, well, their registration policies.
Vanda: Evan, your time.
Evan: I guess I’m contrasting the discussion we’re having now with some of the things you were saying right at the beginning of this talk, where it was sort of saying, “Well, we’re erring on being as open as possible, and things are wide open, and as long as you meet these criteria, anybody can do this.” And then the number of buts that come afterwards is just almost so long as to make it prohibitive, both in the combination of the morality and public order, and nobody’s mentioning community objection. But when I bring those two together, I see a concept where conceivably you could say, well, you can make a TLD as long as you don’t offend anybody else out there. And that scares me.
If somebody wants to create, let’s take a hypothetical TLD, dot sucks. And they have the financial wherewithal to do it, and they have stability of an organization. But people know out there that if somebody does something dot sucks, chances are there’s going to be some, shall we say, disparaging comments that might come under that domain. So that is going to offend somebody. And... the point being, and I want to go back on what Robert was saying, right? When you have article 19, the U.N. Declaration of Rights, saying that people have the right to expression, and then here you are saying the TLD process has as a fundamental component of it the potential that somebody is able to object to this because they are offended. I find that offensive.
Olaf: Can I just quickly... well, offensive or not, I mean, this is... was highly debated, of course, within the GNSO, and what was the limits and the extents of that, and of course there may be sort of frivolous objections. Probably won’t be very many such, since there will be a cost involved, because the dispute resolution providers would have to... it’s... there is a fee for launching an objection, first of all.
Second, well, there will be some kind of requirements, of course, so that this... well, the intent of the GNSO policy will be fulfilled. There... maybe... I mean, just cases or non-just cases, and I think that is really also why we have an external dispute resolution provider foreseen for such cases.
Vanda: Yeah, Nick wants to clarify something.
Nick: In connection with this, when they say that strings must not be contrary to “generally accepted legal norms,” related to public order and morality... if you only get that far, it sounds a little frightening. But the examples that are used in the GNSO policy are the Universal Declaration of Human Rights, the International Covenant on Civil and Political Rights, the Convention on the Elimination of all Forms of Discrimination Against Women, etc.
Vanda: I just... United States... United Nation human rights.
Nick: So what they’re talking about is that the... it’s not unlimited, it’s not anyone’s idea of something that’s offensive. It’s actually a bit of the reverse, that the intent is to apply protection for free speech in connection with morality and public order. And only to prevent TLDs which are obviously against all of those instruments protecting international civil liberties.
Now, it also goes on to say that there’s further research needed on the legal application of treaties in this field in a TLD, which is obviously a novel application that was not originally envisioned when many of these treaties were signed. I mean, going back to the 1940s. So I would just encourage... I understand the concern, and I’m not trying to say the concern is misplaced. I would just say look closely if you have not already at how these provisions were framed in the GNSO policy.
Evan?: 06
Vanda: Yeah.
Evan?: Two very brief follow-ups on that. First of all, the idea of having a fee for the dispute resolution, in some ways that scares me even more, because it means that a well-monied organization could take on a smaller organization trying to create a TLD, and literally spend them away from the application by taking them to the dispute resolution process, and out-spending the applicant.
So that... so, I’m sorry. That objection actually creates potentially the reverse of what you intend. And regarding what you were saying, Nick, you know, there are certain things, I guess, that are in the eye of the beholder. I can see certain things, again, I’ll use something like dot Zionist, which to some is offensive, and to some isn’t, and where do you draw the line? The moment you say that there is a line that can be drawn, you’re basically taking ICANN, a technical organization, and whether you like it or not, making it political.
Olaf: Well, quite, quite quickly. Well, there will be, of course, here you’ve got a standing check, and of course also the first check on frivolous objections, because they have to have... the grounds have to be there. So, well, it is to be partly a new territory, but you have to recognize that this is something... part of... well, you don’t want to have something, “I hate 47,” or something of the sort. Just because you do. So protection of minorities is a valid, I believe, a valid concept, and internationally recognized. That’s just to give an example.
There are plenty of things which are internationally very contested on whether it’s a... and I suppose, well there, you couldn’t... you wouldn’t be able to solve it here, you wouldn’t get... If it’s not internationally recognized as, for example, offensive to a minority or something of the sort, then what would a dispute resolution provider do? I think he wouldn’t uphold an objection, but rather regard it as frivolous and for another purpose. So, well, we haven’t seen the end of this, and we haven’t seen how it works in practice, and we should also remember that the GNSO very much requested that, all right, they have a second look on how this works, as soon as we’ve got half way into the first round.
Cheryl: Thank you. I now have Hong, then, just so everyone knows their place in space, Annette, Vanda, Carlton, Jacqueline, Carlos, 17. Is there anyone I’ve missed? Okay, go ahead. Thank you, Hong.
Hong: Okay. My first question is about the correlation between objection procedure and evaluation process. I want to know, are there two processes in parallel, or is there a sequence between them – say, after evaluation, people can raise objections? This is the first question.
The second one – when we look at IDN ccTLD fast track, we can see applications for IDN strings, and applications for IDN ccTLD managers are two evaluation processes. They are separated. So I want to know, in this new gTLD IDN process, the applications should be one, they are combined together. I applied for this string, I’m going to run it, is the second question.
The third one, I want to echo what Annette had said about readiness in the IDN ccTLD fast track documents. This word has been used, and the word has become, the word itself became controversial. Somebody said that readiness means pressing demand, so people have been arguing who is more pressing, and who is less pressing. So I guess that is not a very good term to use here.
The last one is a comment, not really a question. There used to be an GNSO IDN working group, those of us had worked there. And in that working group report, we sort of highlighted some special need from developing countries – applications from developing countries. There will be some, there could be some special consideration for applicants from the developing world. But now, IDN gTLD will be a component in the whole new gTLD process. And I hope those special considerations, the proposed special considerations, could be maintained in the whole process.
All right. But of course, I don’t suggest this should be isolated, a separate process for IDN gTLD.
Olaf: Okay, let’s... well, the sequence. I think it’s been pictured somewhere, I think there are slides which show that they could run in parallel. Of course, they couldn’t. I mean, you have to sort out the reaction processes before you can sort out final string contention, by sheer logic. So, well, there have been... I can’t see how that could ever occur.
When it comes to the IDN ccTLDs and the separate procedures, well, it bases itself a little bit on the separate procedures for strings and operator... when it comes to ccTLDs in general. And copying and taking over as much as it can from the current IANA process. Here, we’re not separating the string as such, and the applicant... the applicant is supposed to propose a string. So, indeed, it’s very different, but they’re two very different animals when it comes to the contractual relationships, as well. Well, yeah. Good observation, I agree.
Readiness – yeah, probably a lousy word to use, and forget we ever had it, but it was a few slides ago, and it didn’t relate to the same thing, and probably controversial.
Applications for developing countries. Well, it’s not only connected to IDNs, of course. There are developing countries which are not using IDNs, so this is a more overall statement, which actually came from among the implementation guidelines from the GNSO, that ICANN may give some kind of support in that regard. And, for example, it has been discussed whether to, for example, to have a special treatment for applicants coming from the LDCs. The identified LDC list from the U.N. Which is possible, at the same time, as soon as we start drawing lines like that, you have to also prevent that, well, it’s used for gaming purposes. That just because it becomes cheaper to apply from there, well, everybody’s going there, and they have a deal with a law firm that happens to be there and apply from there, and so on. So, well, whatever measure you do, and whatever kind of separation and categorization you do, you have to deal with the unintended consequences as well. But yeah, it’s been thought of, already.
Cheryl: Annette.
Annette: Several short points. Some have been mentioned before, yes, also my organization, the Network New Media, and the organization I’m working for, that is a trade union with two and a half million members, we’re actually really afraid of this introduction of a new rule of morality and public order. That’s one thing that has been mentioned already. We are also worried about the speed of the introduction, because we really would like to have a broader domain space and names we could use and get and register. And here is the next problem – what I would like to focus on. How do you deal with restricted top-level domains? I think that these are actually causing a lot of trouble and problems, and I didn’t see – I don’t know if I overlooked or overheard it, but I think that is a problem to address. So someone mentioned the trouble with dot travel and so on.
For me, it was a lot of trouble to come home. Dot jobs got introduced, and I’m working for a trade union, and we could not register. You know, this is not fun. This is really something you take away the possibility for people, for the users, to just use those top-level domains. And actually, I think by not... or really limiting the TLDs, the amount of restricted TLDs to a minimum would be extremely helpful in all these other discussions, on trade, on mar-... all these lawyer stuff here. Intellectual property stuff, you know, I mean, who is calling what, and then there is all this misunderstanding in the names. Maybe what Evan said, dot sucks, means “I love you” in German. You’d never know.
So all these meanings of sounds and words and letters, and, you know, I think all this is just, it doesn’t work that way. And that is why we should keep out of this morality discussion, we should keep out of all these things, we should just allow... you know, put in those names, and let everyone register. Because then, no one can have a hold on a certain thing, you know, so everyone can use the word in the sense they would like to use it. And that takes a lot of pressure away, I think, also from the morality issue, because then you can, you know, use it in a different way. That’s something I’m missing here – how is this discussion going on?
Yeah, and I would like to have an answer for the speed of all that. We want them. Now.
Olaf: Well, that brings us to the time plan. And perhaps, in the view of time, we should proceed with that. But I respond first of all to what you’re saying about restricted and non-restricted, and that’s a categorization we’re not using. I mean, it was not continued for the new gTLD policy that the GNSO put forward. But it did make a distinction between open and community-based. And community-based would have a certain advantage in a string contention situation. That was... and that’s something we’re trying to craft in... but that’s the kind of distinction that’s made right now. So there’s just two categories. And restricted or not, well that’s not the term which is used. Community-based – I mean, if it’s dedicated to a particular community, they will most certainly have a registration policy which is restricted to that particular community. Maybe that’s the very most appropriate thing to do. But then it has to fulfill the... well, in order to get some kind of... reviewed as a community-based application, it has to fulfill standing criteria which are rather similar to the standing criteria which I used if they launch a community objection.
So, I don’t know if that answers all your questions and all your desires, because I can see quite a few different ways that the gTLDs... No, it didn’t? Well then, let’s proceed.
Did you have more in the queue?
Cheryl: Yes, we have more in the queue. And we have a 4:00 coffee break, which I think the queue will take us to nicely. So if you can keep managing to keep on keeping on, I’d like to let everyone on the queue...
Olaf: I won’t stop keeping on.
Cheryl: Okay, good. Okay. That means we have Vanda now. Thank you.
Vanda: Most of my questions are done, but I’d like to know if... well, the sponsorship criterion is still there. Not anymore?
Olaf: No. No. It’s not, it’s community-based.
Vanda: Not anymore? The community-based changed a lot what we are deciding about who is the community that could allow to continue or not. So this raised a lot of issues there. So in my opinion, we’re going to have a lot of dispute resolution. So one thing that I would like to ask is if we’re going to have the same limit, like 1000 dollars we have for dispute resolution, or are we going to have this open? Because this will decide who you’re going... who are going to win the dispute. So that’s... I remember the beginning of this thing with WIPO, and the limit is very important for the developing country. Because if you don’t do that, it’s a money decision, not else. Not the first on the list, nothing, is the money decision. So it’s something that I haven’t heard about that. I don’t know if they make some comments on that.
Olaf: No, but remember, the cost recovery basis is still there. Also, in addition, in most cases, I mean, when you have an objection process, the general idea would be that the loser pays, at the end of the day. And then, but you split... the objection fees are paid by both parties, and then the loser pays for the whole. Well, as a general idea, currently. So, does that answer your question, at least to some extent?
Vanda: What I said, that’s dispute resolution has $1000 limit, one month limit to respond and to coast for the people. So I’m asking if this issue was raised in this discussion with the GNSO? It’s the same one, or it’s going to be open.
Olaf: Not a preset sort of price or fee, or price tag, or anything of the sort. But a cost recovery is there. And that may entail quite a different objection fees and objection cost that need to be covered. Depending, and this is part and parcel of the discussions with the dispute resolution providers, which is on-going... Well, I’ll get to that. We don’t have any final answers on it, but it certainly won’t be the same for all.
Vanda: Yeah, it is something that we and ALAC need to pay attention on.
Cheryl: Carlton.
Carlton: Well, I tell you, I could give you a lot of intellectual reasons why I am fundamentally opposed to this idea of morality and public order being a reason for objection, but I’m going to make it visceral and raw. There was a time when men who were considered men of reason and enlightenment did not think that someone like me is a part and parcel of their god. And the reason why is they thought they had an inherent right to decide what was right and what was wrong. I am fundamentally opposed to anybody, anyone, trying to tell me that they have a lock on morality. And that, for that reason I say... and this is visceral. I don’t need to give you a lot of intellectual arguments. This is a visceral argument.
And for ICANN to continue to say that it is only as a technical remit, and we use a constituency or some part of ICANN to put this forward is both wrong, it is detestable, and it is retrograde. That’s what I have to say.
Olaf: If I may respond? Objections of that nature will be launched with ICANN anyway. We’ve realized that. We’d rather have it narrowly defined and handled by a body which is, well, suited for such disputes than handling it by ICANN itself. Or rather, that was the outcome from the GNSO side. And well, that is the basis for what we are trying to implement.
Cheryl: Jacqueline.
Jacqueline: Okay, a lot of people have said a lot of things about the same thing that I’m going to say, but not exactly really the things I wanted to say. When you look at page 16 in the summary, it goes into great detail about what you mean by morality and public order. Which is not what people think is what you mean by morality and public order. So it might make sense to remove the words “morality and public order” from the heading, so that you won’t get this immediate, visceral reaction from everyone, because this isn’t quite that. Because I had the same reaction, as well. But one thing that does happen with both that and the community objection thing is that you can always take people to mediation, you can order them to mediation, but you cannot order them to mediate or to agree.
So what happens after that? Because you can go to a dispute resolution provider, and both people can sit down for a month and cross their arms at the table and not say a word, and they don’t get to any agreement. And then, according to this, we go back to the Board or somewhere else, which means you throw it right back onto the same group that you didn’t want to... go ahead. I know you want to stop me.
Olaf: Just quickly. We’re talking about arbitration, we’re not talking a mediation. So there will be an outcome. Either it’s upheld or it’s rebutted by the dispute resolution provider. So there will be an outcome.
Jacqueline: Okay, but is that necessarily going to be an outcome that people are going to be happy with?
Olaf: In most cases, when there is a dispute, there is one happy and one unhappy with the outcome.
Cheryl: It’s not necessarily going to be a win-win, indeed. Carlos.
Olaf: 05 Should we proceed perhaps with the inaudible?
Cheryl: Yeah, I understand.
Olaf: Like, could I just rush through the inaudible?
Cheryl: Okay, let me ask. I have questions from Carlos, Luc and Wolf, which are substantive to the morality and public order. I recognize that there are parts in the rest of the presentation, such as dispute resolution, that we are now going to have to rush through, if not just take on the fact that we will have the presentation. And some of us have attended before, but it’s a rare opportunity to get the discussion and I would like to allow at least the speakers so far to continue. Go again... sorry, put my teeth in order. Go ahead, Carlos.
Carlos: Thank you. 53
Translator: On this subject, you inaudible speakers, obscured by muted mic for interpreter
Olaf: ... in relation with this second topic. laughter
Carlos: 21
Translator: When you made your presentation, you, at a certain time, said that we were ready. Now, on the basis of the discussion that was held, I can see that we’re not all that ready as you say, we’re not all that prepared. The reason I’m saying this is that I don’t know what the limit is to the quantity of gTLDs that can be presented. When you say that we’re ready, we’re prepared, you’re saying we’re technologically prepared to receive all the possible new gTLDs. And also, along with what Robert Guerra was saying before, on the Universal Declaration of Human Rights, articles 19 and 29, let me mention article 27 as well, which has to do with intellectual property, creation, and dissemination.
So who is entitled to define what is moral and what isn’t? What will be disseminated and what won’t? What we have to put a limit to, and what we don’t? What is the morality of those who decide on what’s moral and what isn’t? And, in particular, I think that, apart from all that, I think that there are different cultures. And within these cultures, there may be some things which can easily be recognized as, or acknowledged as immoral, but that might be immoral for one culture and not for another one. And Internet, after all, is something global. So who’s going to put limits to some, whereas not to others?
Now, if I’m trying to outline some sort of possible solution to this, well, why can’t we define the legal nature of the domain name? That is a phrase which might have negative connotations. Is it like a brand, or is it simply a distinctive sign that has no connotation other than distinguishing someone on the web from someone else? Because if that’s the case, then we’ve solved the whole problem. That is, we define the legal nature of the domain name. And if you say the domain name is nothing but a distinctive sign with no other connotation, then I think we would be discussing something abstract. But this is just something off the top of my head, off of 27 of South America. Thank you.
Olaf: May I respond to the last question first, because that’s a comment that... I mean, perhaps it was one of the most stupid things in the whole process to call domain names domain names. Perhaps. But, I mean, we have to live with it now. There are labels. And previously, so, when it comes to the interpretation of the various parts of the United Nations Declaration of Human Rights and so on, I am not at all expert on that. But I take your word for it, that it can be used for... And that is, well, one of, and one of the few, bases for the determination that’s been advocated then by the GNSO in their recommendations.
And also, when it comes to our capacity of handling applications, okay. Well, this is something where we would have to be flexible, where studies are made right now, but we don’t know how many applications we’ll get. So we will have to work out alternative plans and how to, capacity-wise, address them. Both process-wise and in manpower, because in many places it doesn’t help if you throw a thousand people on something. You have to follow a certain sequence of events, as in this case. So, well, such plans are most certainly being worked at. And also studies, when it comes to what might the number be. My bet has been 119 applications, and based on absolutely no science at all. It’s just a hunch. But I’m willing to bet on that.
Cheryl: Okay, very precise. Luc, and may I ask Luc and Wolf to be fairly quick, because I have a plead for one more thing, and if we do that we’re going to have no time for any break. Let alone to finish the presentation. So, please go ahead.
Luc: Hello. Um, one of the questions here was what happens if a new TLD is objectable, in any case. But there was a question, what happens if it’s a part of an objectable name? So it can be used to construct an objectable name. What happens? It’s this discussion about sucks.
The other problem we have is, we have a language revolution. We can define a name as clear and without any problems, and two years later, we have a language change, for instance in the teenagers. So we can be faced with a situation that the word became objectable in a very short time. And a problem trademark holders have is that if they register a new word for a trademark, they are faced with problems in different languages. The same word can be objectable in one language and clear and without any problems in another one.
There is a real world solution for this whole area of problems. All these trademarks, all these words are divided by country, by nationality, and by trademark classes. If ICANN goes to merge them in a single root, we open a can of problems. And we know that we have these problems, because on the IDN TLDs there is... there is a restriction included – that was my question about an hour ago, not everything can be registered. But with this gTLD proposal a questionable political area name can be registered the other way, so we have the same problem, only on the other way. And I do not understand this, from a technical point, I do not need any new level domain anymore.
Olaf: Oh, you’re certainly right when it comes to language isn’t all that. If the Italian Cream Manufacturer’s Association wanted to launch dot panna, meaning cream in Italian, of course you get serious objections from the Finns, where it’s a pretty nasty word. But, well, there we go, and that’s where... I mean, to select, or rather, to avoid all those kind of problems is probably completely out of the question. It’s the flagrant, really serious when it comes to something which is by all recognized as, this shouldn’t be like this. Contentious will most, many domain names be, and so on, but... hence the need for some way of dealing with the contentions, or the objections. So, but, well, I won’t solve all those matters right now.
Cheryl: Do I see Wolf? Oh yes, I do. Sorry. Wolf, if you could come forward to a microphone.
Wolf: Well, I just wanted to strongly support the last motion from Carlton, regarding terms like “morality.” If we look back into European history, things like morality was always a pretext for censorship. And we had a famous philosopher, his name was Emmanuel Kant, and he stands for the period of Enlightenment. And if we are now re-introducing terms like morality into contemporary discussions, we are falling back behind Kant. And I think we should... this for me, at least, is unacceptable.
Annette: A period of darkness.
Olaf: I think you should really read the whole paper and see what’s in it rather than hinge it on one particular caption, which is used for shortness and for reference purpose. Just because... so, maybe we should call it “Recommendation Six,” instead.
Cheryl: Yes.
Jacqueline: That was my recommendation.
Cheryl: And...
Olaf: And then it becomes pretty neutral. And then “recommendation” becomes a bad word, in a sense. So, anyway. Could we..
Cheryl: Okay. I know Evan would like to make a couple of comments, but I would like to allow you to zoom through what you feel is still relevant, because much of what is there has probably been covered.
Olaf: Dispute resolution has already been covered. String contention, we probably should mention, and there will be... that’s in those cases when you have a community applicant that has opted for comparative evaluation, there will be a kind of comparative evaluation. In those cases when he can actually prove his standing as a community applicant, and then it’s with a contention set. Meaning those who have strings that are identical or confusingly similar to his or hers, as it may be. Otherwise, when it comes to those not opting to have such, among the community applicants, or for the open applications, well, string contention will be resolved by an auction. And ICANN has recently set up an agreement with a provider of an auction solution.
So, okay, now let’s go very quickly. We’re still waiting for the ICANN Board to approve the policy, of course, so this is in that respect still tentative. And we have to finalize the implementation of that policy once it’s been adopted, and then, of course, very important is to launch an information campaign so that everybody is informed that this is going to happen, and have reasonable time to prepare themselves.
Now, this is the new timeline. It has slipped. We expect the policy to be approved very, very shortly. We expect the DRP providers to be retained very, very shortly, but it hasn’t happened as yet. And by the beginning of fourth quarter, so this fall, we’ll have the draft Request for Proposal issued, for public comments. And based on the comments received and discussions that will take place, we’ll issue a final Request for Proposal. In parallel, the global communication campaign will be ongoing, and we’re aiming for application launch around second quarter of next year. That’s where we stand right now. And there is more information on the ICANN website if you... do read it, eh! Come on. Don’t stop on what you find disgusting headlines. Read the text.
That’s it. Thank you very much. Muchas gracias.
Cheryl: And thank you. I think what’s important is that we recognize, especially in that timeline... and I hope everybody’s ears pricked up - the next opportunity for public comment. This is where the meeting, what we’ve been talking about at this meeting needs to come to a focus. But Evan has what I think is a perfectly reasonable summation statement, which is simply a statement, and it is going to be brief, isn’t it?
Evan: You’ve already seen it, so. Usually dispute resolution mechanisms, they do mediation, they do arbitration. They tend to try to think things of issues of fairness and where they find common ground between two different points of view. I think it’s outside the bounds of conventional mediation to ask them to judge things such as morality, such as offensiveness, which are usually emotional, not based on logic, and usually not based on areas where you can find much common ground. As such, I think it’s a real problem with ICANN getting into politics, getting into morality, and the fact that they choose to outsource it to external bodies does not relieve them of the responsibility for getting into morality and judgements like this, that they really shouldn’t be doing.
Cheryl: So I doubt there will be many surprises in what may be in the comments that come in in the next round. Thank you very much, Olaf. That was a marathon effort, we really do appreciate it. And once again, thank you all for coming up with what I think are very intelligent and well-thought out questions.
Very, very short break – and I’m talking five minutes.
20 – 3.09.25
Cheryl: Ladies and gentlemen, if you’d now take your seats, be aware that we are doing a simultaneous interpretation, and if you are not familiar with the languages of French and Spanish, you will need to have a headset on. The channel is number one for French, number two for English, number three for Spanish. 50 If not, you’ll have to wear a dongle just like me. Okay. What’s the name of this session, seeing as my screen’s not up? Okay, there we go.
Right. I could just let people read what’s in front of them, but because I have a microphone, I’m going to speak into it. It’s the same as if you give me a spotlight, I’ll stand in it. It just happens. We’d like to welcome Patrick Jones to join us this afternoon. He’s going to be covering on the update of the gTLD Registry Failover plan. Something I know the At-Large community is very interested in.
We convinced him by offering him every sort of connection, including the 39, to actually put his presentation up. He was just going to read to the slides, but I think that the pretty pictures that will go with the excellent words will be greatly appreciated.
I might ask you first – do you wish to take questions during your presentation, or at the end.
Patrick: Thank you very much for having me come and speak to you. I think questions would be welcome throughout the presentation. My slides are fairly short, so I was hoping to have more of a free form discussion rather than relying on slides.
Cheryl: Very happy for that. If those of you who are on Skype wish to ping me with your “I would like to speak,” those who are in the Adobe Connect room can literally put up their hand, and as long as I can see the room, I’ll see it. And of course, you can always gesticulate wildly at me and I’ll put you on a list.
Female: 33
Cheryl: Any moment. It’s all happening. The technology will catch up with us, just not right now. So thank you very much, Patrick.
Patrick: Okay? So, Patrick Jones, I’m ICANN’s Registry Liaison Manager. I am also... I wear... I have dual roles. In addition to being the registry liaison manager, I’m also the Staff Coordinator to the ICANN Nominating Committee. And I’m talking to you about the gTLD Registry Failover plan. So I’ve been the project manager for this particular effort since mid-2006. This has been a long-running effort, that’s set forth in ICANN’s strategic plan and operational plan as part of its efforts to ensure protections for registrants.
So the main questions that I want to address in my presentation, and also be able to take questions from you, is to answer what the Registry Failover plan is, what it’s meant to achieve, and who does it impact, and what the current status of the plan is. There’s been... let me say, so the main goals of the plan are first to deal with the protection of registrants and registrant data, and also to ensure confidence in the DNS. Okay.
So a significant amount of work has occurred since the draft Registry Failover plan was posted in mid-October of last year. A revised... so, following a public comment period, the plan was revised, and discussed at the ICANN meeting in Los Angeles. And from the feedback staff received, we did the first internal failover exercise in January of 2008. This is the first time that ICANN staff has done a comprehensive exercise based on cyber-exercise principles, of scenarios that might happen in the launch of new top-level domains, and any situation that might arise with the existing registries. So, from the lessons that we learned from the exercise, we updated the plan and circulated it to a small group of experienced ccTLD managers and gTLD registry operators that we’ve been working with, and took some feedback.

And since then, it’s been an enormous amount of consultation, listening and feedback with that group. And the latest draft is dated June 16th, although I’m sure that I’m going to be incorporating feedback from the next couple of days into a new version, that we’ll be able to release publicly. I guess that’s the latest bit on the plan.
What I want to do is talk a bit more about the exercise, because that’s something that you might be interested in, and then open it up to questions. So we have a new... I guess it’s a collection of information that’s been posted over the last two years about the failover plan. And it’s accessible through the homepage, and also on the registry’s page, and it’s about the registry failover plan. I should have provided a link, but I’ll make sure that Nick... Actually, I think I provided a link to the ALAC list through a response to someone before this meeting.
So there is information available, and it’s some background information on the work that’s been done, plus a collection of all of the articles, presentations and, you know, the discussion that’s happened with the plan. And on that page, there is a copy of the after-action report from the failover exercise that we did. This was a large scale exercise. We looked at five big scenarios. We had staff located in eight different locations, all over the world, from the majority of staff located in Marina del Ray, California, to... I believe we had Cairo, Mexico, Australia, and then multiple other locations within the United States and Brussels. All using Adobe Connect, following the exercise in real time, and looking at the different scenarios and discussing them, using cyber-exercise standards from the National Institute of Standards and Technology – the U.S. NIST standards.
So these were the five areas that we looked at. How the escalation of small outages could result in the large scale failure of a registry. We looked a specific case where... what happens if DNS set keys are compromised in a new gTLD? We looked at a situation involving a fictional IDN registry that is located in an area that is hit by a category 5 cyclone, and then the resulting effects of the government being unable to maintain its infrastructure, and then wanting to nationalize communication in the region. All of the scenarios were building on different things that could happen, and just trying to gauge staff response and emergency response based on the scenarios.
The fourth one we looked at is a complex attack that might happen on a back-end registry operator. And that is of interest to ICANN as we’re getting ready to launch the new gTLD process, and there are currently a small number of back-end providers that provide service to either multiple ccTLDs or in some cases gTLDs, and what would happen if there was an attack on the back-end operator that may not have a contract in place with ICANN.
And the fifth was sort of an example of what happens when there’s a collection of bad acts by a registry, and how ICANN might deal with those acts. So all that work was described in the After-Action report. And from the report we’ve done a lot of revisions to the plan and sort of improved the plan that’s in place now.
Having connection problems. So that’s the end of my slides, and it’s fairly short, but it was intended to have it open for questions that you might have on the failover plan and its efforts. I had some additional slides that I may have been able to incorporate from previous work, but I thought I’d hear some of your concerns instead. And wearing my Registry Liaison hat, for this project, I also manage the process for new registry services. And some of you might have seen that we recently posted a number of registry services requests that deal with... Let’s see, Affilius had one that we posted for an abuse of use policy, dot moby and dot coop are interested in releasing single letter names within their registries, and dot org has a DNSSEC implementation that’s up for consideration by the Board at this meeting. So those are also relevant topics that you feel free to ask me questions about.
Cheryl: Thank you. And obviously, I’d like to see some hands up for questions. Okay, Robert – thank you.
Robert: Thank you for updating on this. My name’s Robert Guerra. I’m on the ALAC representing, together with other colleagues from North America. I saw a number there, number five, was kind of bad faith. Would that cover kind of the registry fly issue in the past? And, so that’s one kind of question on the last slide. But on the larger thing is, in terms of a failover, how far are you away from being able to implement one of these failovers, and how quick would the response be.
Patrick: So the fifth slide, the scenario was... we wouldn’t call it a registry fly situation, but it was built on the lessons learned from that issue in the registrar space. What our intentions are is to compete the current draft of the failover plan, in agreement with, or at least, with the consent of the registries and SSAC, and then take that draft and conduct a large-scale exercise against the draft in 2009, with a group of interested registries and registrars, and then report to the community about that exercise, and then refine the plan. And to do that exercise before we launch the new gTLD. Like, before the introduction of any new gTLD.
Cheryl: Go ahead.
Robert: Just a quick follow-up. As you mention the new TLDs, am I correct then in assuming that the new TLDs will not launch until this failover plan has been tested and ready to go. Will that stop the implementation of gTLDs until...
Patrick: This testing is not going to hold up that effort, but it’s designed to help inform the community and inform staff and inform registry operators and registrars of things that they need to be aware of. And it is also good crisis management training. Internally, we found the failover exercise to be a really good wake up call for things that we hadn’t thought about, or needed to think about, and just hadn’t had time to do in a structured way. It was a really good two day exercise.
Cheryl: Beau.
Beau: Hi, it’s Beau Brendler, of ALAC-D. You mentioned something about dot org looking at DNSSEC. Are there any other registries or, I mean, is that gaining at all in popularity? Or is dot org the only one looking at it right now, of the existing...
Patrick: So far, dot org is coming forward with their proposal, but, you know, their back end operations are conducted by Afflilius, and Affilius manages back end operations for several other gTLD registries. It’s likely that Affilius is comfortable with the implementation end. Dot org... I don’t want to put words in what you were... or suggest that they might then take the lessons learned from dot org and apply it to all the other gTLDs that they manage. But they’re going to be involved in this process. I suspect that Verisign is very interested as well, so I don’t expect org to be the only gTLD registry to look at implementing DNSSEC.
Vanda: Carlos.
Carlos: 01
Translator: Could we have an explanation about the, in government, when you talk about a government takeover, what are you talking about?
Patrick: So, the scenario that was discussed was an example of a government attempting to nationalize a gTLD registry. It’s something that hasn’t happened in a gTLD registry space, but as new TLDs are launched, it’s conceivable that that might happen. So in this scenario, we wanted to explore how that might be discussed internally. And it’s a sensitive issue, and involves a lot of political discussion that tends to be outside of what ICANN involves a lot of political discussion that tends to be outside of what ICANN deals with in its core mission, but it’s an interesting one that may come up in the launch of new gTLDs.
Carlos: 14
Translator: I was asking because in my country, well, we’re one of the only countries where the government enables ccTLDs, but the Argentinian chancellery handles everything that is supposed to be Argentinian domain names. Do you have any information relating to this?
Patrick: I’m not sure it came up in that kind of example, of I guess what you’re asking about. This was more of a general observation. There’s a natural disaster, the government’s having problems with infrastructure, there’s an attempt to nationalize all the media in the country, and that included the registry operator of a gTLD registry. And that’s how it came up in this fact-based but fictional scenario.
Cheryl: Go ahead, Robert.
Robert: Just a quick follow-up. I gather, for the scenarios I think that Carlos was mentioning, two things. One, was it envisioned, for example, to do a failover in case a ccTLD registry goes down, or there’s an issue to do that. I would be really concerned in regards... and maybe ask that if there’s any planning around gTLDs, and how that might affect the community and how it might respond. Though it’s great that staff has gone through this internally, something that affects a gTLD would affect Internet users around the world as well. And it would be interesting, as you plan forward, you know, ALAC may be interested in participating in that because, as representing Internet users, we may be contacted as well. And so part of that contingency planning would help us prepare for that. And I think registry fly was also a learning experience for us. It not only caused us to help start processes at ICANN, but also how to respond to users. So those lessons learned would be useful for us as well. So, going forward, I would maybe suggest, if it’s not mentioned already, or I would propose it to my colleagues that it might be of interest for ALAC to participate in that and be able to contribute. Because I think that users are affected as well, and as we represent them, it might be useful for us to be involved. Thank you.
Patrick: So, I have, I guess, two points in response. The registry failover plan has some communications elements that involve ICANN and registries being able to communicate out to registrants what the facts are with an event that’s happened. And so ALAC would definitely be involved with any dissemination of information about an event, and communicating that to the users. But another, I guess, it’s related to efforts dealing with ccTLDs, I think it was a couple of weeks ago, or in late April, there was an apTLD session on disaster planning, that ICANN staff was involved in. So that’s an example of current efforts by ICANN to work with ccTLD operators to learn more about how to deal with cyber attacks and disaster planning and recovery, and those efforts are going to continue, and we’ll do that in conjunction with experienced operators, whether they’re in the g space or the c space.
Cheryl: Izumi.
Izumi: Thank you, Cheryl. My name is Izumi. I may have a wrong question, but allow me. What you have prepared, does it relate to the registrant data escrow arrangement mandated earlier this year?
Patrick: So, the registrar data escrow program is a different effort, and that’s being handled through ICANN’s registrar team. I know that is an effort that’s moving along very well, and there will be more information provided about the current status of the registrar data escrow program this week.
Izumi: Uh huh. Currently, the... so, your plan of registry failover preparation, whatever you call, will not contain the escrow data for that registrar matter, or is there a plan to use that? Or how do you separate them? Or am I making a long question?
Patrick: No, so in most cases, registrar information is held by registrars. But many of the gTLD registries are thick registries, so they have the information as well. So they have the same information that a registrar has, except in the case of dot com and dot net. Those are operated as thin registries, where the registrar has the authoritative data, but the registry has the information that displays who the registrar of record is. In order for this plan to work, there needs to be cooperation, coordination and joint efforts of registries and registrars, so that there is... the registrar data escrow program is related to this. And tomorrow, at, I believe it’s at 11:00, there will be another open session, and it’s focussed on protection of registrants with the launch of new gTLDs. So we’ll give an update then.
Izumi: So, just to give background on why I asked the question, is in Japan, several registrars gathered together with myself and the government people, because at a certain date, ICANN mandated all the all the ICANN accredited registrars to find their own escrow, or if they can’t, 43 is the officially designated American company, or global company, who gets all the depository, all the escrow. Which actually, in our national law, is sort of... touches against this law of personal data protection, so we are murmuring what to do with that. And, of course, we... any registrar can choose other entity that meets the criteria that ICANN set. But there aren’t that many yet, and also there are some cost problems associated. So the registrars in Japan are now talking to each other to set up some sort of, you know, common repository of sorts.
So I think some other countries’ registrars may have a similar if not the same challenges. And also, that concerns our national government, also about national interests. Whether it’s valid or not is another question, but that’s the kind of discussion we are having about registrars. So about registry, it may be extended to that, that we need to have some kind of legal implications for that.
Cheryl: Nothing wrong with a no comment. That was the end of my speakers list. I’d call for any questions, including from the wider audience, those of you not at the table. Going once, going twice. I see no hands.
Therefore, it falls to me to thank you very much, Patrick. I think we’ve certainly got a lot out of that. But what I don’t have, and I’ve just gone through the at-large list, is the e-mail you referred to. So if you wouldn’t mind sending the link, we will ensure that it gets attached to the presentation. But I just went through by name. I don’t think you would have got caught in my spam filter, so...
Patrick: I might have had it sent through... I think John Levine posted it to the list.
Cheryl: Okay, I will recheck as well, but perhaps it would be expedient...
Patrick: I will send it to you.
Cheryl: Yeah, measure twice and cut once. Let’s get a copy. Because I’m sure we’d all like to read further on the matter. So thank you very much. If I can ask you all to thank him in the normal way. applause
Okay ladies and gentlemen. We’re perilously close to back almost on our currently and frequently modified agenda in terms of timing. We really only have a session two of what would normally have been what we do with less briefings and less time spent talking about an ALAC review in the morning. And this is to look at the RALO activity reports, the ALAC working groups, and we have any questions or comments from ALS representatives that are around the room.
We had planned small team activities, which would be identifying main challenges and possible solutions in ALAC activities since New Delhi. We discussed earlier this morning that perhaps focussing on small group activity with a view to a response to the ALAC review would be a worthwhile topic. And I would also suggest that perhaps preparation for our inevitable responses to the next public comment period for the new gTLDs might also be an interesting topic to look at. So we might be able to split into two separate areas.
There was also to be a discussion of the process used for the development of statements, and Nick has prepared a delightful flow chart for us, and the normal winding up discussion of results and lessons learned.
What I’d like to do, seeing as small group activities give us the opportunity to get up and move around, and I don’t know about you, but I’m very keen to get up and move around very soon, is I’d like to propose that the RALO and working group activities, which really then become part of our ALAC report, following on from what I started last year, which is to allow RALO input as well as working group input. Could I suggest that we take those on line? But remember, if I don’t get them, they don’t get included. And I’m planning on writing this sooner rather than later. So perhaps finding me at the back of a workshop room and giving me a thumb drive with your statement on it. The long form statements will form part of the full report. The short form statement may make it into a presentation piece, but I’ve not been informed what our presentation piece will be. As you remember last year, all chairs of the various supporting organizations and advisory committees were asked to use a...
end of audio