Archive for August, 2007

.test IDN TLD (For Real!)

Tuesday, August 28th, 2007

Tina Dam and I caught up during ICANN San Juan on the current state of IDN TLD work within ICANN and that was when I first found out that she was putting together a live IDN TLD test bed which includes translations of the string .test into eleven written languages (Arabic, Chinese-simplified, Chinese-traditional, Greek, Hindi, Japanese, Korean, Persian, Russian, Tamil and Yiddish) and ten scripts (Arabic, Cyrillic, Devanagari, Greek, Han, Hangul, Hebrew, Hiragana, Katakana, Tamil).

I was very excited to hear that and wanted to blog about it but there was much to catch up at work so I let it slide.

Two days ago, an update from ICANN on this project:

ICANN today finalized the IDN .test Evaluation Plan and continued taking steps toward insertion of IDN strings in the root zone. Recent changes to the plan are based on comments received on the IDN public forum and also from consultations with ICANN Technical Advisory Committees. This last version was approved by the ICANN Board at their 14 August 2007 meeting. The resolution directs ICANN Staff to implement the IDN .test Evaluation Plan, and report back to the ICANN Board following the conclusion of the evaluation.

Specifically, the Board approved the delegation of eleven evaluative top-level domains representing the term ‘test’ translated into: Arabic, Persian, Chinese (simplified and traditional), Russian, Hindi, Greek, Korean, Yiddish, Japanese and Tamil. Following this ICANN Board approval, the delegation request will now go through standard IANA procedures for insertion of top-level domains into the root zone. The technical evaluations of IDN TLDs and their usability in various applications will proceed following their delegation.

This is a major milestone in the IDN Program Plan and signals a significant step forward towards Internationalization of the DNS. It is currently anticipated that delegation of these TLDs and the evaluations, as described in the plan, will commence in September 2007.

I would like to applaud ICANN (specifically Tina) on bringing the project this far. This is a major undertaking, and an important one too, for the following reasons:

  1. It works across the Internet — not just an isolated test that involved a couple of PCs running in a lab environment; they are going to insert these labels into the root zone! This is the only way to involve the most diverse mix of participants in the test bed. The fact that it requires no special setup (like tweaking of hint files) may lower the barrier for entities such as ISP’s, who are traditionally seldom interested in the development of IDN, to participate.
  2. As a result, it allows the most diverse array of test cases possible. Just imagine how an innocent domain name gets passed around applications, resolvers, recursive and authoritative DNS servers through a myriad of functions, API calls and networking protocols. It’s supposed to work but who knows for sure?
  3. It tests major scripts used by regions with an immediate need for IDN-TLDs (Latin would have induced yawns)

I’m sure there are skeptics and ICANN-haters out there who will dismiss this for time-wasting activity, etc. The truth is, no one has actually tested IDN TLD’s on Internet-wide scale before. And no, country-wide deployments will not suffice because the diversity of software environments and cultures simply isn’t there.

And if you’re a proponent of the Just-do-it school of thought, this should be seen as a move in the right direction. If no major problem was found during the live test, it will shut the mouths of those who are doubtful.

We should take advantage of this test to file bugs for your favourite software vendors and get them to support IDNs! When was the last time IANA inserted a test TLD to the root? Tell them that if IANA agreed to put these test strings in the freakin’ root, this is not a default ignorable technology alright!

So, help spread the word and test the Internationalized Internet!


Why 1/5 of Americans Can’t Locate The U.S.

Sunday, August 26th, 2007

Found this gem on Digg, made my day!

.BIZ Korean IDNs Just Launched

Monday, August 20th, 2007

The second piece of news following the localization of EchIDNA into Korean is that we have just launched Korean IDN registration in .BIZ three hours ago. I won’t divulge the numbers here but let’s just say that it far exceeded my expectations.

You may be wondering if there is any connection between our Korean launch and the localization of EchIDNA. Aside from yours truly being involved in both projects, it is purely coincidental. We have been preparing for the Korean launch for a few months, and in doing so have consulted with NIDA (National Internet Development Agency in Korea) and have adopted their language table (with 2,350 KS X 1001 Hangeul characters). The table is available here: and will be posted to the IANA Repository of TLD IDN Practices soon.

If you have a .BIZ Korean IDN and would like some free links, submit it to!

EchIDNA Speaks Korean

Monday, August 20th, 2007

안녕하세요! There are two pieces of great news I’d like to report, both of which relate to the Korean Internet users community.

First of all, EchIDNA is now localized to Korean thanks to Jaeyoun Kim (www.김재연.kr) who took the initiative that triggered a collaboration between us where he translated the EchIDNA user interface into Korean and I added localization support to the code. While I have long been thinking about localizing EchIDNA, I never did get around to doing it so I’m really glad that it’s finally out.

Here’s a screenshot of it in action:

EchIDNA Korean UI screenshot

This move was largely motivated by the disheartening news that Verisign discontinued its IDN Web Navigation system on July 25th 2007. A bit of explanation of the so-called IDN Web Navigation System is in order. Basically, every IDN registered in the com and net TLD gets one or more A records inserted into the zone that points to a Verisign web server(s). These are not punycode domain names but binary representation of the IDN in local code pages where it is likely used, plus maybe some IE-specific behavior. The latter refers to an IE5/6 quirk that will encode the individual bytes of the domain name in the system code page to UTF-8 by treating them as ISO8859-1 characters. By doing so, users who are using non-IDN aware browsers will be redirected to Verisign’s web service which will then offer them the ability to download its i-Nav plug-in, and at the same time select the correct IDN domain to navigate to. Since the A records are binary representations, there is a possibility of collision which explains why the users have to select. It’s not hard to imagine that the general Internet users may mistake i-Nav to be the Web Navigation system that they were using, and since it works why bother downloading anything? So my guess is that many users simply ignored the i-Nav download part and click through to the site they wanted to go to. When the service was discontinued these users were lost. It didn’t take long for the domain owners to notice and complaints started flooding into the registrars and hosting providers. Some of these complaints also made it to NIDA which had an agreement with Verisign to distribute the i-Nav plug-in (and since users confuse i-Nav with the IDN web navigation system, it just looked like “i-Nav stopped working”.)

While I agree that this is really a hack and that the service cannot run forever, it is simply too soon as IE7 is not even one year old in Korea. Even though there is technically nothing wrong with discontinuing a hack, I can’t help but question the ethical validity of the decision, given the obscene amount of cash made from these TLDs. I wonder if the transition could be made smoother by first disabling the navigation feature and explaining to the users that they really need to be using an IDN-aware browser or download a plug-in.

Many ccTLD registries (CN, DK, JP, KR, TW to name a few) that offer IDN registration also run a similar web navigation system for their TLD, and AFAIK are still running it.

So, back to EchIDNA. The localized version was released on August 15th, 2007 and is available on

GearSaver v0.4

Sunday, August 12th, 2007

I’m pleased to announce GearSaver v0.4. This version has been tested under Firefox and IE7 on Windows with Gears installed. Most importantly, I’ve finally ironed out all the bugs in the various usage scenarios x both browsers. The usage scenarios I’m talking about are:

  1. By including the script on the page
  2. By using it as a bookmarklet

Along with it, I’ve put up a proper page for it: GearSaver Home.

Here’s a screenshot:

GearSaver Screenshot

If you’ve tried GearSaver before and it didn’t work for you, I would encourage you to give it another try and submit a comment here.

IP Migration

Saturday, August 11th, 2007

Our excellent hosting provider, Simpli.BIZ, has had in the past a few glitches due to the fault of AboveNet, their upstream provider. Yesterday, I finally bit the bullet and migrated all our “hosts”1 to the new IP range owned by Simpli. It wasn’t as smooth a migration as I’d hoped, but the Simpli team was very helpful and attentive. Knowing how much I suck at system administration, Erica had already arranged for KVM attached to my console so even though I managed to lock myself out, I could still do everything through the console.

The old IPs are still available, but I’ve already updated some DNS records to point at the new IPs. The old IPs still work via PF redirection rules.

Simpli must be the most personal commercial hosting provider out there, it’s nice to have your web hosting support person (in this case it happens to always be Erica the founder) to be on IM and maintains a blog. What more can you ask for?

1 I said “hosts” because while I only rented a measly Celeron from them for USD99/month, I have 3 FreeBSD jails running.

GearSaver Reloaded

Monday, August 6th, 2007

The GearSaver Bookmarklet was not that useful since there was no easy way to use the saved data. Since this is an itch that I’d really like to be able to scratch, I reached further and improved on it to the point that I can now write this post in a GearSaver-enabled textarea. Added features are:

  • uses jQuery – I have no intention of working around browser DOM quirks more than I already have to
  • displays previously saved versions
  • ability to preview a version and apply to current textarea content
  • ability to delete versions

Eventually, I hope to be able to provide the script as:

  • ad-hoc inclusion into any page – this already works in the current version
  • as a bookmarklet – so that you can enable any page
  • as a greasemonkey script – so that you can automatically have it turned on on the sites that you want it

So, without further ado, you can try it out…

1. Here’s a textarea

2. Make sure you have Google Gears installed.

3. Click here to enable GearSaver.

4. Gears will ask you whether to allow to write data

Please report any bugs / suggestions.

Update 20070812: Updated to use new and improved version 0.4