Participation and network issues for Delhi meeting

by Kieren McCarthy on February 12, 2008

As many of the remote participants have noticed, we have been having network problems at the ICANN meeting in Delhi, which is impacting remote participation.

As soon as the technicians have sorted out the problem, we will come back with an explanation and update this post with details.


[This has come from the technical staff to explain the issues there have been over the past few days]

We have been faced with many challenges here.

The week before the meeting there were a total of five undersea fibre cuts in the Asia Pacific region. Because of this, a lot of the traffic of the region has been aggregated and is all traveling through the remaining fibre paths (all at capacity at the moment). As the fibres are replaced, it will certainly impact the region (in a positive way!) but unlike when your computer crashes at home, deploying the “tech support” to a cut undersea fibre can take up to a week to arrive at the destination.

We’ve run into some hurdles with the local transit as well in the way of trying to get them to sort out our AS number and properly announce (and transit) our traffic.

We have also had a challenge with stable power in the hotel. There have been a number of power outages (either partial or full) and blown circuits. Some of these have directly impacted the ability to feed power to the routers we are using to egress the hotel. Though we have UPS (uninterruptible power supplies), we’ve found a number of them have been mis-wired ore plugged into the hotel UPS, which can cause a “blowback” and cut the power.

Although service has not been stable, the team that is here locally has done an amazing job of adapting to each new challenge and providing a solution in the most timely fashion that they were able to.

We apologize for any inconvenience and assure you we are doing everything within our power to try to keep the services running so that the meetings can be effective. It should also be noted that the local staff have provided incredible support dealing with multiple challenges.


热门榜单 02.13.08 at 12:34 am

remote participants ? Who and where can participate ICANN meeting in Delhi?

Kieren McCarthy 02.13.08 at 1:07 am

Please start at


Leo Vegoda 02.13.08 at 3:03 am

I’ve been listening to the audio over the last few hours and it has been very good quality. The content was good, too.

Anonymous 02.13.08 at 7:37 am

where can I find the audio?

Leo Vegoda 02.13.08 at 8:29 am

Sessions with video or audio streaming have links to the streams available in their description page on the meeting web site. The sessions I’ve looked at so far have links to translation streams, too.

Bret Fausett 02.13.08 at 1:24 pm

Thanks for the update. I wonder if it would be possible to convert the RAM file to mp3 or leach an mp3 file off the live stream for archival purposes?

Paul Levins 02.13.08 at 7:38 pm

Hi Brett

We do archive presentations, transcribed material and also video and – I think – audio.

I think it’s very important that we archive the contributions.

Let me check and come back.


Harry Strane 02.13.08 at 8:13 pm

Kieren, many thanks for the participate ICANN link.

Kieren McCarthy 02.13.08 at 9:58 pm

All the material provided is always archived.

This time around we have worked on recording audio – in the different languages – and ge a system where we can upload that as fast as possible.

I am very keen on getting to mp3 but for reasons too boring to go into at the moment we have Real Audio streams.

The video archive takes much longer – and there’s very little way to speed that up. Assuming that is we do streaming video. I want to end the webcast – and make up for it by supplying a feed of the scribes and early presentations.

Then we can record video with a camera and do a fast turnaround and upload the files that night.


哲学.cn 02.15.08 at 7:32 am

what do you mean?


Giải pháp văn phòng chia sẻ (Virtual Office), công nghệ tiên tiến, thích hợp cho: thành lập công ty, chi nhánh, VPĐD, giúp giảm chi phí tối đa cho doanh nghiệp, tại cao ốc 19 tầng Indochina Park Tower, 4 Nguyễn Đình Chiểu, Quận 1 (có sẵn line điện thoại, Fax, vào hoạt động ngay).Giá thuê văn phòng chỉ từ 50 USD/1 công ty/tháng.

David Wrixon 02.17.08 at 3:25 am

Communication isn’t exactly ICANN’s strong suit.

Final Press Communique didn’t even mention the topic that has preoccupied you all week. No, on this weeks performance you are not ready for independence!

Paul Levins 02.17.08 at 3:23 pm


Not sure what you mean by ‘didn’t even mention the topic that has preoccupied you all week’. There is not one topic that preoccupies these meetings.

If by ‘this week’s performance’ you mean running a meeting in India for 720 people, 290 of whom were from the host country, we plead guilty.

Paul Levins
Excecutive Officer and Vice President
Corporate Affairs

Kieren McCarthy 02.17.08 at 8:06 pm

Included in the final press release:

* New gTLDs
* IPv6
* ALAC/RALO changes
* Accountability and Transparency Frameworks
* IDNs

The only thing I can think of not on the list is domain name tasting, front-running etc.

They formed a part of the Board resolutions. Is your point that a link to the Board’s resolution would be helpful?

If so, that’s a good idea. We’ll add one. But if there is failure to communicate in this case, I am afraid it is coming from your side since your message is very unclear.


Kieren McCarthy 02.17.08 at 8:13 pm

A link to the Board resolutions has been added.


David Wrixon 02.18.08 at 3:49 am

Well it seems to be there, but I am pretty sure I searched on IDN earlier.

The one thing that ICANN constantly fails to communicate is that IDN already exist at the second level. Failure to deliver IDN TLDs in a timely manner means that will be a defacto standard by the time the “Fast Track” ccTLD belatedly arrive some time next year.

ICANN fails to appreciate what every manufacturer in World has to understand to survive. That is that failure to get you product to market on time means that it becomes irrelevant.

Lets examine what is required for the Fast Track. What is actually required is to agree what the best representation of the name of a dozen or so countries is in the most popularly used script in each of those countries, and then compile a list for rubber stamping by the Board. As there are actually very few countries trying to lay claim to exactly the same name as another country, and indeed in most of the cases under consideration, they all use completely different scripts, just how bloody complicated can you make it?

This really could have been sorted out and done and dusted by now. The Board should have been presented with a list and the IDN ccTLDs should have been in the process of being inserted into the Root Zone by now.

Frankly, I wonder why everyone is so concerned that you might be taken over the UN. Even if they were in charge something concrete would have been determined by now!

Paul Levins 02.18.08 at 8:27 am


We do communicate that IDNs exist at the second level. In fact if you look at the videos on the topic on the ICANN website and on Youtube as well as written material you will see that we talk about it a lot.

The process is not ‘easy’ it is extremely complex – a fact that we have been stating now for at least the 18 months I’ve been working for ICANN.


Bill Winans 02.18.08 at 2:48 pm

FOR WHAT EVER REASON…. it appears that ICANN is not interested in letting worldwide internet users and native language speakers know about,,,, etc. etc.

The big question is WHY NOT?? I have sure been trying to figure out the answer. They may not be “pure” IDN.IDN, but for many the wait is going to be 1, 2, 3 who knows how many years. Why not kick things off with, then upgrade the program with the rollout of IDN.IDN.

I know you mention that this is publicized by ICANN, but I sure challenge you to show me a few links, ICANN articles, or anything relating to this. I have searched the internet and can’t find them. Even the first few Tina Dam IDN videos don’t mention a single word about Below is a comment that may be one of the many reasons why…

Comment by bwhhisc
2008-02-17 04:33:22
You can use,,,,,, etc. etc. TODAY. Many of these have been registered since 2000′. Only now with the worldwide release of IE7 and Vista will the browser be able to read the unicode and resolve to punycode., etc. can be registered at Dynadot, Domainsite, Moniker, ENOM, and now I think even GoDaddy.

IDN.idn will be a wonderful thing when it arrives over the course of the next few years (sunrise for a few are “projected” for late 2008 and 2009.

In the meantime, ICANN would do a great service to the world internet communities to inform them about their IDN choices that they can use TODAY!

QUOTE: Comment by Tina Dam
2008-02-18 00:31:32
Hello bwhhisc, ICANN is not a marketing service for domain related products. This is handled by registrars or resellers and to some extend the registries.

However, in the work related to idn tlds icann staff is trying to be careful about mentioning that we offcourse have had idns for quite some time and that this is the extention of the second level introductions.


Paul Levins 02.18.08 at 7:54 pm


Tina’s right.

“ICANN is not a marketing service for domain related products. This is handled by registrars or resellers and to some extend the registries”

Paul Levins

David Wrixon 02.18.08 at 10:14 pm

Paul, regardless of whether you see yourselves as a marketing organisation or not, you need to be creating the perception of actually doing something, to offset the commonly held view that you nothing other than an Expenses Gravy Train. At the moment. you seem to be losing that battle. It just surprises me that you have managed to keep Russia and China on board for so long, but it is still not inconceivable that an alternative root could be established if this one doesn’t deliver.

Kieren McCarthy 02.19.08 at 1:56 am


Would you consider two dedicated public sessions on IDNs at the most recent ICANN meeting – one running for three hours [details | transcript] and the other for two hours [details] – evidence of work?

Or perhaps, the two Governmental Advisory Committee sessions [here and here], both running for four hours? Or the IDNC Working Group meeting [details]?

Perhaps we should advertise the work being done. Like in this press release of 12 February entitled: “ICANN Explores India’s Pioneering Work on IDNs”.

Maybe evidence of real work – such as a draft initial report on the ccIDN fasttrack [available here]. Why doesn’t ICANN create a Working Group solely to work on the questions – such as the one draw up by Charter here?

But what about public input? Surely ICANN should allow the public to comment on this vital topic? Maybe these two comment periods on the ccIDN fasttrack and the introduction of IDN ccTLDs might suffice.

The least we could do is advertise these public comments periods. And not just in English. But also in Arabic, in Chinese, in Japanese, in Russian, in Korean.

Why doesn’t ICANN test IDNs in the root – and then advertise the test results?

Where aren’t there any joint papers between ICANN supporting organisations and advisory committees – like, say, the ccNSO and the GAC?

Why haven’t we posted and advertised the IDNA guidelines? Or the protocol revision proposals?

You’d think the least we could do was post a status report or something along those lines? And how come it’s never occurred to ICANN to post IDN glossaries in English, Arabic, Spanish, French, Portuguese, Russian, or Chinese.

Why don’t we have a dedicated IDN webpage linked to directly from the front page where we outline all the latest developments?

And what’s the point in having a blog if we don’t provideusefulsnippetsofinformation?

You’re are right, David, perhaps it’s about time ICANN actually *did* something.

[I should point out that IDNs are only one part of what ICANN doesn’t do every day.]


David Wrixon 02.19.08 at 11:37 am


It is 6 year on here. Yes, I will admit there has been a great deal of chatter, but what have you actually delivered to date? Well not so much as programme schedule for delivery in reality. It is like a re-run of the “The Life of Brian”!

Kieren McCarthy 02.19.08 at 12:29 pm

I’m not sure that “chatter” adds test TLDs to the root. Unless of course you are making a wider philosophical point about the nature of human beings and advancement.

If the nub of your complaint is that IDN top-level domains don’t exist yet, you’re quite right. Someone would have said. Other things people are pretty keen on that don’t exist yet include flying cars and a vaccine for AIDS.

I don’t know where we are at the moment with flying cars, and I was disheartened to hear last week that scientists say a vaccine for AIDS may never be possible, but we can be pretty certain that IDNs will exist in the next few years. So, aside from a hell of a lot of work in the meantime, I’m not sure what else there is to be said about it.


Paul Levins 02.19.08 at 9:14 pm


I would just add to what Kieren has said here by saying IDNs exist at the second level and they have been tested live in the root at the top level. A lot of hard work has gone into those results.

So there is way more than ‘chatter’ going on. The IETF, the IAB and – yes – ICANN amongst many others, are working hard to deliver on the technical insertion of IDNs in the root hopefully by the end of this year.

In the meantime if you have the solutions that have alluded the finest minds in the technical, legal and civil community then speak up.

To paraphrase the Python classic you reference, there are no Messiahs here and there are no very naughty boys either. Just people doing hard work. And they have results to show they are getting somewhere.

Paul Levins

David Wrixon 02.19.08 at 9:59 pm

Get a list of each country that wants a fast track TLD in their own script and stick it in the Root. Which bit exactly is defying the finest minds of a generation?

Paul Levins 02.19.08 at 11:14 pm

The bit that was called ‘breaking the Internet’ by one of the guys that invented it:


David Wr 02.20.08 at 11:03 am

It complete rubbish that IDN will break the Internet. So something nobody is using doesn’t resolve, so what! What ICANN has just achieved is spend a year proving that the Root can resolve ASCII strings starting with xn! Isn’t that basically what it has been doing over a decade?

And by the way, your blog doesn’t render properly in IE7! I am seeing all the comments super-imposed on each other. Now getting to Microsoft to do something that is helpful for its customers, now there is a challenge!

Steve Clarke 02.20.08 at 12:19 pm

I believe both Dave and Bill are correct in their contention that ICANN has not been telling anyone that IDNS, in .com/net/etc have been in use for several years now.
I can’t remember hearing Tina’s video mention it.
If I didn’t kwow any better, I would have thought the ability to use IDNS was still only a future event.

Mike Jackson 02.20.08 at 12:23 pm

Vince knows that inserting a new gTLD that is named for example .xn--xylzf is not going to “break the internet”. Period.

What inserting IDNs into the root will do is cause people to use them. When people start to use them, it could very well break some of the NSA and CIA’s screen-scraping data collection programs as they will need to add new tables to their database, port their tools to be UTF-8 enabled, and so on. Maybe they’ve been talking to Vince, asking for some more stall-time, you know?

Honestly, is ICANN going to require root insertion test periods for every new gTLD string which doesn’t begin with .xn--, such as .asia? Will the finest minds be required to give their blessing that .asia as a string won’t break the internet?

It’s time for some straight talk. People deserve it after this wait. If the root gets split, the game is over.

mdw 02.20.08 at 12:44 pm

There is a considerable portion of the IDN community, perhaps the majority, that do not like the rhetoric used by our most vocal member. However there seems to be widespread agreement among the IDN community that as various ICANN members talk about IDN issues, they often and deliberately avoid the mention of, net, biz, and the various ccTLDs.

We often hear that native language domains are needed and coming soon. That small omission of the fact that many other scripts can be used is a source of great frustration to those of us who use domains like this, own domain names in other scripts, develop websites on “foreign-language” domains, etc.

Any efforts to propogate the fact that there are native language solutions that work today would be welcomed by IDNers. It would also be beneficial to speakers of many of the world’s major languages and local business communities.

Bill Winans 02.20.08 at 1:48 pm

It seems that,,,, etc. are 1000% better that NO idns at all. The word is clearly not out to native language speakers around the world, based on reports from native speakers in various domain forums and unofficial polls conducted in Japan, China and elsewhere. It is further verified by the slow uptake by large numbers of different internet users in the various countries. etc. are a good stepping stone to lead to IDN.IDN in the next few years. They will no doubt allow people worldwide to see that the IDN program ICANN is building can be used today, and is indeed underway. Granted these are not “full” idn.idn but still available to use and build websites on TODAY.

ICANN would get positive publicity from letting the internet users around the world know they can participate and use the IDNs today without waiting for another year or two, or more.


Kieren McCarthy 02.20.08 at 1:55 pm

Hi Steve,

This simply isn’t true. I have heard Tina and Paul Twomey and Peter Dengate-Thrush *continually* refer to the fact that you currently have IDN at the second level.

They also continually make the point that there is nothing at all to stop people producing content in different languages right at this moment.

I don’t know where this concern has come from and I have to say I haven’t ever heard it expressed before, but if it’s a genuine concern, we will endeavour to make sure that ICANN representatives continue to make the point in future.


Kieren McCarthy 02.20.08 at 2:07 pm

Given a choice between believing that the co-inventor of TCP/IP and one of the leading voices of the Internet is:

a) Trying to protect the security and stability of the Internet using several decades’ worth of hands-on in-depth technical knowledge, or
b) Delaying expansion of the Internet for a short period of time so that the CIA can upgrade its software

I think I know which one I’ll go for.

We thank you for your conspiracy theory but we must apologise if ICANN just gets on with making full IDNs a reality.


Mike Jackson 02.20.08 at 2:18 pm

Tell me, after you read this, do you get the impression that IDNs are available and working today?

I sure don’t. And this is the same type of information that is repeatedly published from ICANN.

The general public doesn’t understand all the technical talk about levels, roots, strings, scripts, etc. – they just want to know when they can start to use web addresses such as lääkä so they can resume typing like they are taught in school, instead of using misspellings such as They read that article and think “Hmm, the inventor of the internet said that it can break the internet if they put in non-english character domains”. The truth is that there are thousands of Chinese domains already developed and working today, modern web browsers can reach them, and the internet hasn’t broken yet.

What is the IDN hype being saved for?

Kieren McCarthy 02.20.08 at 2:37 pm

This thread started out complaining that ICANN wasn’t doing enough with IDNs, then it became that ICANN maybe wasn’t doing the right thing with IDNs, and now it’s become ICANN isn’t giving the right impression to non-technical people about what IDNs could be used for.

It doesn’t, hasn’t, and never will take ICANN pointing at things to get people to use the Internet in whatever way they can. And when the work is completed on full IDNs, there will be plenty more that people can do with the network.


FrankieSaysImpeach 02.20.08 at 3:02 pm

So was there any progress made on the ccTLD fasttrack process?

Did they come up with at least a small list of non-controversial non-latin ccTLD labels that can be put in the root NOW, or are the ccTLD managers in those countries going to continue splitting off from the ICANN root, sick of the delays?

bwhhisc 02.20.08 at 4:24 pm

Frankie, this article summed it up:


“Weeklong discussions in India about the creation of online domain names entirely in languages other than English led to greater understanding but no major decisions, the chief executive of the Internet’s key oversight agency said Friday.

At issue is a proposed “fast-track” mechanism for specific countries to get non-English suffixes — the “.com” part of Internet addresses. For example, many Internet users in China would like to see a Chinese-character equivalent of the “.cn” suffix.

“At the beginning of the week, various groups mapped out their extreme positions,” said Paul Twomey, chief executive of the Internet Corporation for Assigned Names and Numbers, the U.S. group tasked with overseeing domain name issues. “By the end of the week, it became a more reasonable conversation.”

Twomey said a working group may make recommendations by ICANN’s next major meeting in Paris in June. ICANN’s board took no action on new domains Friday as the meetings wrapped up in New Delhi.

Twomey said he does not expect ICANN to start accepting bids for specific names until the end of this year at the earliest.


Paul Levins 02.20.08 at 4:48 pm

This is wrong. We say it in press releases and in speeches and in the most recent video by Tina Dam on the website she says:

“So we’ve had IDNs at the second level for a long time but we don’t have them at the top level and that’s the new thing”

In addition to her saying it in English and posting it on Youtube and dotsub, to make sure we got this message out it is transalted into: Farsi, Greek, Hindi, Tamil, Chinese, Chinese (traditional), Korean, Russian, Yiddish, Arabic, Japanese, Spanish, French, Italian, Portugese.

I am going out now to get a skywriter and a blimp with a neon sign.

Paul Levins

Kieren McCarthy 02.20.08 at 9:05 pm

Yes, it’s a fair summary.


Kieren McCarthy 02.20.08 at 9:07 pm

Please see below FrankieSaysImpeach.

Frankie – are you looking to impeach anyone to do with ICANN, or do you just have a general impeaching stand?


David Wrixon 02.20.08 at 10:17 pm


This is what is making people so angry. Representative it may be. Fair it is not. The Board should be guillotining debate and insisting that a list be put forward to them for approval by June. If both sides cannot agree a list, then each side should forward their own list and the Board should choose the most reasonable. A bit of pendulum negotiation is what is required here.

bwhhisc 02.21.08 at 2:06 am

Regarding Tina’s video, do you really think the “general public” would have a clue about what this statement means. Not 1 in a 1000 I would bet.
“So we’ve had IDNs at the second level for a long time but we don’t have them at the second level and that’s the new thing”

This would be a much more positive statement of IDN progress by ICANN and others:

“IDNs are live today under many global top level domains in dozens of languages. In the near future, some of the top level domains themselves will be constructed from non-english characters and will be displayed as idn.idn.”

Kieren McCarthy 02.21.08 at 2:56 am

David, I think you have really misunderstood a fundamental element of the ICANN model. That is: that as far as possible, and in as many ways as possible, it works on consensus.

If there is disagreement, then there a variety of ways for getting people to grow to understand the others’ perspective and then prod them into coming to agreement.

The Board tries as far as possible to avoid exactly what you are suggesting: acting as some kind of Grand Tribunal of the Internet. And for good reason – the Internet relies on people all agreeing to do something in a certain way. The only reason the Internet as we know it exists is because everyone agrees to do things in a certain way.

The concept of 22 people taking the place of the entire model of consensus-building is appealing on a shallow level because it is faster but it would set a terrible precedent and risk undermining the entire model.

And if the model is undermined, then the whole issue would be entirely moot anyway.


Kieren McCarthy 02.21.08 at 3:03 am

This is just getting silly.

Now you are rewriting paragraphs in a video that was made months ago because of a personal perception that ICANN isn’t pushing something clearly enough to a general public.

How many levels of abstraction do we have left?


bwhhisc 02.21.08 at 3:11 am


This is meant to be constructive criticism. I am not ‘re-writing” anything, just suggesting that most of the publicity and press for IDNs leave the readers believing they are a “thing” of the future.

Bill Winans 02.21.08 at 3:34 am

To back up my comments about the press take on ICANN statements, here are a few good articles from October 2007 from the Washington Post and Wall Street Journal that tend to leave the readers believing that native script domains (IDNs) are a thing of the “future”.

Steve Clarke 02.21.08 at 6:08 am

As Bill has clearly stated, the “GENERAL PUBLIC” or (the majority of Internet users) would have no idea that already exist, and have for several years now.
If it is an oversight, than you can surely rectify it in future press releases or videos.
I don’t think that is too abstract…do you?

root 02.21.08 at 7:25 am

And if the root is split, then the whole issue would be entirely moot anyway.

Paul Levins 02.21.08 at 8:28 am

I just got back from our India meeting.

The majority of people who had access to the Internet knew about IDNs at the second level. This includes not only people attending the ICANN meeting (as you’d expect) but a straw poll of cab drivers, guides, shop owners, journalists etc.

They have existed at the second level for years. Our task with others is to get on and make them happen at the top level.

My point about quoting Tina’s video was that the claim was made that you didn’t recall her even mentioning it. It is obviously there.

I’m not sure we can add much more to this debate. I might ask Tina if she has a perspective on this though.


David Wrixon 02.21.08 at 8:34 am

People have been pushing to get this implemented for over decade.

Most of the first five years was met with total indifference from ICANN.

The bottom up model is admirable, but there will come a point when things need to get done otherwise the various groups which you are trying gain consensus from will just wander off in different directions.

You might say “Bottom Up”
Others might say “Institutionalised Anarchy”
I would suggest “Lack of Leadership”.

If you cannot sell a vision, then you are unlikely to gain a consensus.

If you cannot impose a timetable, you may never arrive at a solution.

Kieren McCarthy 02.21.08 at 9:31 am

As luck would have it I am currently at Domainpulse in Vienna.

Up on stage a moment ago was Sabine Dolderer, the CEO of Denic, which is the third largest registry in the world with 11 million .de domains; Urs Eppensberger, head of business at Switch, which runs Switzerland’s .ch and Liechtenstein’s .li; and Richard Wein, the head of which runs Austria’s .at domain.

Also there was Peter van Roste, the head of CENTR – a regional registry organisation that represents 47 different countries registries.

I asked the panel about second-level IDNs and whether they had been successful or not; if people were aware of them or not.

Previously Peter van Roste had given a presentation in which he said that he was surprised to find that the introduction or availability of IDNs on a certain registry was not a factor in people deciding a ccTLD’s success.

He qualified this when I raised it and said that, that’s not to say people weren’t registering IDNs at the second level – they could be choosing an IDN and then not renewing an ASCII domain instead – but that the introduction of IDNs was NOT a factor in a ccTLD’s success.

Also, Urs Eppenberger mentioned in his presentation that Switch had had a promotion in which an IDN domain was given for free – but that nearly all of them were returned. I will ask him about what exactly happened later but it appears at least that Switch actively promoted second-level IDNs to the extent that they were giving them away for free and it didn’t have much of an impact. I will check though.

Sabine Dolderer answered my question. She said that there are 600,000 IDN .de domains – which Denic has estimated is around five percent of all the possible IDNs that could exist in the German language. She was pretty content with this.

However, she stressed that the real value of IDNs comes not with a second-level IDN but when there are full IDNs because then people not familiar with the Western alphabet at all, such as Chinese and Arab speakers, will be able to interact easily and send email easily.

So, what does this all mean?

Well, one ICANN is not the body to come to complain one way or another than IDNs should be like this or like that, or we should say this or we should say that. It is not up to ICANN to promote second-level IDNs.

If VeriSign thinks there is a market in then it is up to VeriSign to market it. To think that anything ICANN – an obscure technical body – does would have the slightest impact on this is just ridiculous.

Secondly, the news from the people actually selling IDNs in markets where people are not native English speakers is that IDNs are doing fine but there is no pent-up demand from people who simply aren’t aware that IDNs don’t exist.

Thirdly and lastly, why do people – including the press – look forward to full IDNs, why do they make a big deal of it? Because it *is* a big deal. If you had to type Arabic characters at the end of every email and the end of every website that you have no idea what they mean, you would not find the Internet that friendly either. Until we have full IDNs a huge percentage of the world’s population is at a disadvantage.

Why do people look forward to the next big thing? When was the last time you saw a news article, or, for that reason had a conversation with someone, about your landline phone?

People are interested in what is new and what is coming – that is human nature.

So in summary, if you think there are millions of people just waiting to get an or, go talk to VeriSign or Afilias or GoDaddy or Tucows – they will welcome you and all their new customers with open arms.

But the word from those doing it is that there is no army of IDN buyers just waiting to be told they can do it.

But, let us make it clear one more time – this is not what ICANN does. And nor does anyone – apart from the three commenters on this blog post – want it to do so either.


Tina Dam 02.21.08 at 10:17 am

Hi Paul, thank for mentioning this to me this morning. Its hard to keep an eye on everything and this post was not in particular orginally focused on IDNs.

In any event, i find it hard to understand why some people complain about our communication efforts about IDNs. We obviously need to mention that we are working on IDNs at the top level because it is the next step.

At the same time IDN second levels are and haver been included in our communication plans, we mention it to reporters that ask about IDNs, it is part of the development both on technical and policy fronts focused at TLD level for the simple reason that IDN second level have taught us a lot that we can take advantage on. this is also part of my status update from Delhi:

As paul and Kieren already have mentioned loads of other items all i can say is that i could continue on and on and on. Instead of doing that let me mention – again – that we will take the input from all of those that are concerned about this and adjust things to make it even more clear.

That said, I wonder if it is possible that those complaining does not want IDN TLDs, and are trying to attrack attention and thereby slow things down. Sorry, if that is the case it is not working. The hundreds of emails that go in my inbox every day on solwing the issues is a much better DDoS attack than the comments on this blog.


David Wrix 02.21.08 at 10:37 am

No Tina, we are not trying to slow things down. Frankly, it would be hard to slow things down much more, and often we just seem to be drifting backwards.

When ccTLDs were first established somebody just got hold of a list and banged it into the Root Server, with little regard for the fact that dot CN didn’t mean a damned thing to the people in China. But now and for non-Latin script only we have to be Eeeeversooo Caaarefuuulll!

What I am concerned about is double standards. In anyone’s fair opinion because the first dozen or so gTLDs coming out of the trap are in Latin Script which is only used by half of the globe, then the next dozen ought rightly to be in non-Latin script, but the truth is that we are going to get inundated with even more virtually worthless Latin extension before there is even consideration of implementation of IDN.IDN.

As for the term “Fast Track” in ICANN usage, it is is a complete Oxymoron. Or is that deliberate?

David Wrix 02.21.08 at 10:45 am

Yes, and because it is such a big deal in Arabic, ICANN should be doing everything in it power to make Top Level IDN available to the Arabs as soon as possible, not only by issuing IDN ccTLD in Arabic, but also by providing Aliasing to existing gTLDs. This is much much much much more important than trying to promote a whole load of junk gTLDs whose only real purpose is to get yet another set of fees out of companies that only buy them to defend their Trademarks.

Tina Dam 02.21.08 at 11:09 am

Hi David, when you say “we” on behalf of whom are you speaking?

I am glad to know that you do not want to slow things down. Your comment also shows why it is that we need processes that work for everybody – namely that the existing one does not. Keep in mind though that having latin scrip on the top level does not mean “not IDNs” – it can easily be represented as another language and there are plenty using the latin script that uses IDN at the second level succesfully.

However, the notion of making things fair is soemthing that the policy development bodies have been very focused on – as such implemtning things to make sure we are avoiding “ascii squatting” has been a core IDN focus for them.

Regarding ‘fast-track’ – this was not a term defined by icann staff if that is what you mean? sometimes it is hard to know what people mean when they talk about icann – for me icann is the entire community, including users, and as such including both you and I.

In any event the fast-track was named as such because it is a way to try to do something faster about implementing a process for IDN ccTLds than what the standard ccNSO policy development process could do. The timelines are in my mind very optimistic – ending in June 2008 (following that is implementation etc…see one of my IDN blogs on “when to register IDNs under IDN TLDs) – but they are supported by staff and we are doing the best we can to make this work. It is important though to keep in mind that these processes have to be developed by the community. I am sure you would not like a group of icann staff to sit around and make these decisions. the community being where processes are being defined is what makes things work well.


Alias 02.21.08 at 11:36 am

What can you say about aliasing? Are existing gTLDs going to be aliased to IDN extensions or not?


David Wrixon 02.21.08 at 11:44 am


We is whoever you where accusing of trying to slow things up.

Yes by all means. Ask the relevant national authorities how they want their names represented. Make a list. Get it ratified and stick it in the Root Zone. That is really what they want and far more than they got with the existing ccTLDs. As for everyone else provided they are only asking for something that means their country in script that is in current use, and provided you don’t have a situation where Russia suddenly wants to be China or China wants to be India, I really don’t see who else has a vested interest that needs to be taken into consideration.

Tina Dam 02.21.08 at 12:07 pm

Hi david, i was merely trying to figure out the reason behind the complaint. The suggestion you mention is not that easily carried out. we have been down that road several times. Some questions around it: who are relevant national authorities, what do you do when more than one want the same string, what does it mean that a script is in current use, what about those that have multple scripts used within a country, what about those where the script in use is used by some other “country” as well and the use creates visual confusion (as we had in the initial phases on 2ndl elevl implementation). There is no list. It is not icann’s business to determine who should be on a “list of countries” or a “list of relevant national authorities” and who should not. Icann also dont want to be in that business. There are plenty of international organisations that have enough difficulty with that as it is.

So, that is why we need a process, where those interested can show that they fullfill technical criteria as well as any other criteria set by the community. Then they can have the string in the script they want. But the process is important because contrary to what is used today on the ccTLd front, no list exists, and so far nobody have been wanting to make such list. The few suggestions that have been on the table have esitmated 5-7years as the minimum to create such list – and these are organisations that have a great deal of experience in this field. Obviously that is not timely enough – and so the community is instead trying to create the fast track.

Maybe you want to participate in some of this work instead of being unsatisfied with the road it is taking?


David Wrixon 02.21.08 at 12:59 pm

So apart from perhaps North and South Korea, just exactly which other countries have an identity crisis. And of those how many would be likely to get fast tracked anyway? These kind of arguments just don’t wash. I could more or less work out what the final list is now.

Tina Dam 02.21.08 at 1:02 pm

I guess you missed a lot of my points. and my points were just examples. If you feel like you have the solution the why dont you join the work that is going on.

David Wrixon 02.21.08 at 2:22 pm

I think we are starting to get disillusioned with the process.

We spent the whole of last year proving that the ICANN Root can handle ASCII strings starting with XN.

What we are looking for is a commitment to produce results within a specific time frame. Even some clear indication of what the ultimate objectives are would be most useful.

Nobody following the process can figure out where the process is suppose to be heading. That is because there is no clarity in the process. I have read nearly all the documents that have been produced front to back several times. Frankly, they tell us very little. I am sorry, but I thought the idea of developing policy was to produce clarity!

Tina Dam 02.21.08 at 3:26 pm

-and so i suggest to you again to participate in the work. commenting negatively on a blog post that originally had nothing to do with IDNs and hence got my attention late will not help.

the protocol is alos under revision, maybe the technical side is more firtting for your expertise than the policy and processes. i think the roadmap for the ccTLD fast track is quite clear. perhaps you can be more specific as to which parts are confusing to you.

Tina Dam 02.21.08 at 3:33 pm

Hello Alias, this is difficult to answer and might end up being different for the various TLDs. Let me try and see if this explanation helps.

In the policy work related to IDN gTLds it was determined by the community that just because a company is the registry operator for a gTLD today is not the same as saying that it will have pre-rights to all or any translated or transliterated versions of that TLD. However, as part of the process for introduction of new gTLds there is also an objection based process. As such if someone applies for a certain string others can object to it on ground of for example “confusability” with the existing string. So one can imagine a situation where a .tld is tranlated into another language, represented by a corresponding script and applied for in the process. If it is not the same operator then the current operator of .tld could possible make an objection.

Now that said, offcourse current gTLD operators can also decide right away to apply for a translated string and to operate that as an aliasing to the exisiting one.

So in other words untill we see applications i cant tell what it will be.


sean 02.21.08 at 6:18 pm


Can you clarify what you mean buy a current gTLD operating a new string as “an aliasing to the existing one”?

Do you mean:

1: The given registry will somehow “alias” the new IDN gTLD to an existing latin TLD so that typing one domain name + gTLD automatically directs to the same location as the other gTLD? My understanding is this solution (whether DNAME or equivalent) would require further ICANN testing, approval, etc.

2: The registry will effectively grant registrants in one gTLD the same domain in the IDN gTLD. In this scenario it would be the registrant who decides whether the domains “alias” or not as they would decide if they should point to the same location. This scenario would not need the approval of anyone.

sean 02.21.08 at 6:25 pm

I rereading my comment i realize it might be confusing.

To avoid confusion, and purely as an exampl:.

let’s say that Verisign applies for and is granted a hanzi equivalent of .com. For arguments sake we’ll call this “.公司”

Are you saying Verisign can then somehow choose to alias these domains so that

points to the exact same location as

without any intervening technology? (i.e. DNAME)

or, are you saying Verisign can choose to grant the stakeholder in “” the domain “”. This would effectively alias the domain (assuming the registrant pointed them to the same location) without any new technology, ICANN approval, etc.

David Wrixon 02.21.08 at 10:43 pm

Well I have to disagree, and the discord between the ccNSO and gNSO underline my case. If they actually had a clear understanding of what each party stood to gain or loose from this then they would not be at each others throats. There really is no clear reason for Countries to fight over names except perhaps the Koreas, but that didn’t happen to any great extent with .KR. I am sure the prospective registry owner could, if they think hard, come up with one or two interesting extensions that did not actually mean somebody else’s country. It seems to me that the only real grounds for conflict are those with existing rights over names and those that are trying to usurp them. The purpose of the policy to the extent it is need is to prevent that from happening. Publicising more widely, and don’t mean in California in English, what is happening and how it affects everybody, especially in Asia is essential, if you consultation process is going to have any credibility, otherwise you might as well just take a straw poll of those Taxi Drivers that ferry from the airports to the conferences. To the extent it is possible this should be being widely publicized in the prospective markets of Asia, where people are still bitter from being sidelined in the first Land Rush. This at the very least means public announcements in Japanese, Chinese, Russian, Korean, Hind and Arabic in local publications where these languages are spoken. And pleased try to convey a clear unambiguous message for once!

Kieren McCarthy 02.21.08 at 11:41 pm

This is bordering on a rant, David, and there’s no useful way for me to respond to it.

All I can say is that the IDN process is happening, it has been taken up by all sides of ICANN, it is working through the multitude of issues, it will produce IDNs as soon as it can.

If you think you can solve the problems faster, you have already had two explicit invitations from Tina to join one of the working groups. If you *can* solve them faster, everyone will benefit.


Kieren McCarthy 02.21.08 at 11:48 pm

Since we are gradually going through each aspect of the enormous issue of IDNs: you making accusations; ICANN responding with links and explanations, can I suggest we try to approach this is a more positive manner?

Is what you are asking for a breakdown of the IDN issue into different elements with an explanation of how we got where we are, the assumptions going forward, the future issues, and a timeline?

Or would something like this, which would be more of a report, not be interactive enough?

Would you prefer it if we have Tina take questions in a live Q&A session online some time soon?

Or is there some other method by which we can provide useful information on the work ICANN is doing without it having to be this adversarial, aggressive approach?


Kieren McCarthy 02.21.08 at 11:53 pm

I’m afraid you’re not right on this David. The ccNSO and GNSO are not “at each others throats”. In fact, the IDN session in Delhi was very interesting for that reason – the different bodies understood the others’ perspectives and in fact it was a hugely positive step forward.

I think Tina has made it extremely clear that while it may look easy, when you actually try to implement something that you outline above, the people involved themselves aren’t happy with the approach.

Towards the end of this latest comment, you start making assumptions and outlining accusations that are proveably untrue. I won’t stir it up by going into them because it is clear that you feel strongly about this and we should therefore be trying to work together rather than throw bricks at one another.

Can I please suggest you take up Tina’s offer to get involved in the work going on – we could do with all the people we can – and I have also outlined some ideas for getting more and clearer information out to you and others in a response to another of your posts above.

Do take us up on the offer.


Tina Dam 02.22.08 at 9:12 am

I am saying that it could be either one. It is important to talk about alaising and what functionality is really meant behind it. This is what the policy development bodies are doing today. Then once the policy is decided and the term defined then we can talk about what technical solution can provide that function. DNAME has been reviewed a few times as a potential solution to aliaising, but as you point out with your examples, it depends on how one defines aliasing.

David Wrixon 02.22.08 at 11:05 am

So Tina, are we saying that Aliasing will be clearly defined in the policy documents, because until now it seems not to be mentioned at all?

Are ICANN going to take make it clear in this “policy document” whether or not Aliasing can be done at the Back-End of the Registry or not before they agree the policy, or are we just going to leave it in the lap of the God’s and wait to see what innovative schemes the registries for these new extension dream up to exploit the punters and implement off their own backs?

David Wrixon 02.23.08 at 3:03 am


I would love to take you up on your offer, but unfortunately personal circumstances do not permit.

I understand the need for fairness. Indeed, implementing IDN ccTLDs at this current point in time would probably act against my own vested interest, but frankly having analysed the situation in depth, I cannot think of a single constructive argument for delaying the introduction of IDN ccTLDs, even though there is a risk that would take market share from my own IDN dot com investments.

Fairness, however, should not be sought to the Nth degree at the cost of progress. Progress itself is a massive fairness issue. Many economies in the World are being denied an Internet where direct navigation is not an every day part of the Internet experience, and that is costing the World economy billions of dollars, at a time when we cannot afford for things to be under-performing.

To be fair IDN.IDN is not the only progress issue, the wanton neglect on Microsoft’s part to introduce a browser that supports the IDNA protocol has been a very serious burden for everyone. That problem, however, is now thankfully be laid to rest, and that will mean the widespread adoption of IDNs whether with IDN extension or with existing ASCII extensions. However, to provide an equal playing field for all, it is of paramount importance to get IDN.IDN for right to left languages which cannot realistically be expect to function with bi-directional strings.

Progress is about fairness. You cannot achieve fairness without progress, and whilst I have severe time constraints to deal with, I may yet come to Paris and lamblast you all if you are not showing signs of delivery. And beware because I am more than capable of making the delivery in French should the need arise.

Anonymous 02.23.08 at 3:36 am

how dare you question the almighty icann?

bwhhisc 02.23.08 at 4:35 am

On behalf of the vast majority of people interested in the IDN topic “thanks” to Tina, Kieren, and Paul for answering questions and taking time to provide us with a better understanding of the IDN process.

Tina Dam 02.27.08 at 4:54 am

Hi David, I guess i will have to wait and see what the policy comes out with. staff does not decide on policies. We have staff supporting the policy development bodies, however their role is limited to that or support.

In terms of defining how aliasing can be done – – this is not a policy question. If the policy side clearly defines that they want aliasing and what they define by that, then we can ask the technical community to provide some guidance on how such alaising can be tecnically provided. DNAME have been one such suggestion that to date have not been verified.


Tina Dam 02.27.08 at 7:56 am

Thanks bwhhisc, much appreciated. It is off course what we are here for, but it is always nice to hear a positive feedback :-)

Personally i enjoy having the discussion and getting the issues and concerns on the table – that is the only way staff at ICANN can at least attempt to address them. I also realize that it is not easy to understand what staff is doing and what the various statusses are. Believe me, the communication aspect around soemthing like IDNs is something we try really hard to do well. Hopefully blog posts on some o fht emain questions and issues help. If there are suggestions for other activities then by all means let us know.


Paul Levins 02.27.08 at 1:40 pm

What she said ;-)


Comments on this entry are closed.