nav search
Data Center Software Security Transformation DevOps Business Personal Tech Science Emergent Tech Bootnotes BOFH

Dot, squiggle, plop

Unencoding the web

By Guy Kewney, 23 Oct 2007

Column My first response to the idea of a dot-sex web domain was: "Well, that's a rip-off". And when I heard about dot-tv and dot-euro I wasn't the only one to say: "Here we go again..." And I bet most of us were equally cynical about dot-asia.

I mean, most people in Asia don't actually use the Roman alphabet and so (I sort of assumed) they would see this as just another way of lining the pockets of domain registrars. So I checked with Kieren McCarthy.

Kieren, as you may remember, was a reporter on the Register for some time and then, in a fit of insanity, put himself up for election to the council of ICANN. When that didn't work out, he accepted an actual paying job with them. Turncoat, I said. Not a bit of it, he said. "Well, can you justify this latest scam for taking money out of people with websites? I asked."

The problem, apparently, isn't with the desire of domain registrars wanting to make money. It's the fact that the internet is, inherently, unfair to the people of the East. Come to that, it's also unfair to those who use Cyrillic or other non-Roman alphabets. So ICANN can justify dot-asia, and it's going further with a project (unofficially) named dot-squiggle.

Squiggles are the characters which most English-speaking people simply can't decipher, and which make a walk down the main streets of Tokyo and Bangkok such a mystifying experience.

In the same way that I can't read Japanese street signs, the average Korean struggles to understand the URLs of the World Wide Web. ICANN is tackling this.

ICANN has organised a conference, to be held in Taipei next month, to try to find out where the world ought to go with dot-squiggle. It is an upgrade to the encoding of domain names and DNS to allow Chinese characters, Korean characters, Singaporean characters, and all non-Roman alphabets to be uniquely decoded so that ordinary citizens of the various Asian countries can search the internet without needing to know how to write in English.

It doesn't take much imagination to see how useful that could be. But does it explain dot-asia?

"Public demand," said the official spokesman I was referred to by ICANN. "It was the same with dot-euro," she said. "We were approached by domain registrars who suggested there was a need for it, and they demonstrated their technical ability to handle the job. Some did say it was uncalled for, and would just force the big domain holdes and trade-mark owners to register names like to cough up. But it turns out that no; there really are lots and lots - millions - of people who want to be "me@justme.euro" or run www.justme.euro as websites; and that in fact, Sony didn't feel even slightly bothered by the idea."

And, apparently, there is an equally enthusiastic requirement for dot-asia domains, and ICANN expects ordinary people to go for it on an even greater scale. The main pressure to provide it, it says, actually came from Asia - specifically, from Korea.

The move towards non-Roman domains and characters, however, is just the second step. From ICANN's point of view, the real goal is a domain name system which allows almost anybody to create and administer a top-level domain which is entirely original. If you didn't get a dot-com name, suddenly, who cares? Create dot-anything.

For sailors, perhaps, there would be a value in a dot-boat domain. For bee-keepers, a dot-honey domain; for gardeners, a dot-seed domain. If someone wants to have a domain and has the skills to administer it, the plan is to make it possible. Why not?

But of course, this opens several cans, many of which have worms.

Developing a system of character encoding which enables all these domains to exist in any number of languages, was a challenge. But now it's been done (and is in beta) the next step is going to be, surely, to decide whether we can open up the rest of programming technology? Was that a worm I saw?

Today, HTML is not only written in Roman characters, but uses English (and American-spelled English, at that). That means if your education doesn't include English, you can't be a website manager (worm).

What about Pascal? C#? C++? Javascript? Why should it not be possible for people who use different alphabets to write software in them? Ooh, look, another worm...

It'll be interesting to see how this one plays out. ®

The Register - Independent news and views for the tech community. Part of Situation Publishing