On Wed, Jan 20, 2010 at 03:01:41PM -0500, Alex Fink wrote:
> There's something basic I'm missing about the design of ISO 639 and ISO
> 15924 and whatever else is out there.  Why in the world do they use tiny
> fixed-length letter codes?

Because they're easier to use than numbers, I suppose.  What would you prefer, arbitrary-length strings?

* we'd have a lexicon of standard names for languages rather than a code as such (it might say, for example, that Welsh should always be 
called "Cymraeg" instead of "Welsh") which would look like an attempt to dictate usage in everyday language;

* people who have to type ISO 639 codes (which is most of us, since they're used in things like Wikipedia DNS labels) are presumably quite 
glad that they don't run into the tens of characters;

* there are still very many applications out there for which a guaranteed limit on code length is a good idea: not just legacy systems, but 
also wire protocols which need fixed-length records.

You might as well ask why the US post office maintains a two-letter code list of states.  Why should we bother writing "PA" on an envelope 
when we can write "Pennsylvania"?

> (And scripts get four-letter codes but
> languages three: what, does one expect there to be 26 times more scripts
> than languages that people would want to localise into!?)

ISO 15924 has nothing to do with ISO 639, except perhaps that the choice was presumably made in order to make ISO 15924 codes visually 
distinct from ISO 639 codes.