The web is about to change in a very profound way – maybe it already has. The IANA has responsibility for the root name servers, the most important thing on the Internet
http://www.iana.org/domains/root/db/
Down near the end, it mentions a series of countries experimenting with Unicode top level domains. What are they and how does it affect things?
Let’s say you’re in China and you have a web site aimed at mandarin Chinese visitors. Today your web site name has to have .cn on the end, using two Latin characters that indicate the web site is in China. This is a legacy of the Internet and the later web being designed by Americans and Western Europeans.
In the beginning was ASCII – the 7 bits per character official US Government standard. All American. None of those silly Frenchy swirly characters or German guttural markings. We are America and that’s all that mattered. To be flexible, the government did allow UPPER and lower case (Don’t tell the FCC) With 128 choices – 52 for the alphabet, 10 for digits, a couple dozen punctuation marks and 32 characters to tell mechanical teletype machines to return the print head, advance the paper, and stall while the print head travels back, the world is complete.
http://www.ascii-code.com/
Then along came the IBM PC that would change the world. While ASCII takes up 7 bits, messages are sent in groups of 8 bits and a bit is a horrible thing to waste – so IBM took upon itself to create 8 bit ASCII – and built into character mode graphics adapters important characters that allowed you to draw boxes without having to switch into that complicated thing called graphics mode.
http://www.ascii-codes.com/
It also included some of the characters needed by European Latin based languages. Since the world of 1980 is run by NATO, that’s all that is needed. IBM’s version of ASCII quickly vanishes with the end of MS-DOS. Microsoft realizes there is a world out there beyond America and people in Russia, India, China, Saudi Arabia and Israel are not going to create Word documents using ASCII – and some languages are written from right to left or top to bottom! OMG – my head hurts!
By 1990, it was obvious that forcing the world to spell out words using Latin characters would not work – Russia, China Japan and others would just develop their own computer industry, incompatible with everyone else.
So Unicode was designed
http://en.wikipedia.org/wiki/Unicode
It allowed every possible character or symbol in every language to be represented on the same computer. The world is one.
Then around 1994, this thing called the Web arrived – from your account at the Well, Mindspring or AOL, you could connect to computers all over the world using the HTTP protocol and HTML to design web pages. The world was changed forever.
But the most central parts of the Internet remain US centric. The core concept of the Internet (not just the web) is interoperability – everything on every computer should be accessible by everything else. Names like URIs (http://blahblah…) should be limited to the lowest common denominator – capabilities that every computer has – so names are limited to only a subset of ASCII – letters and numbers and a few special characters. If you want a Unicode character in your web page name, you have to encode it %[encoded character] and you can’t do that at all in your domain name.
So what does Wan Chang do? He runs a shrimp farming business in China and wants a web site to tell people in his city Bout his shrimp. Today’s standards force him to have a name that is Latinized – http://www.wan-chang-shrimp.co.cn/ which may make sense to a university student who has learned English, but not the bulk of people using iPhones. Chinese language search engines mask this to some point, but the ultimate I usability would be that the entire Internet is based on Unicode rather than Latin ASCII.
In 2003, the IETF published an attempt at encoding Unicode characters in domain names (wan-chang-shrimp.co) in a way that was compatible with DNS
http://www.ietf.org/rfc/rfc2396.txt
Many of the hacking exploits over the past 10 or so years have involved using Unicode sequences to slip past edits designed to lock out improper ASCII attacks.
so the IANA is exploring allowing Top Level Domains to be in the characters of the language of a country – instead of being .cn, the name might end with .中國
Welcome to the global community.