kenw at sybase.com
Thu Nov 23 01:05:58 CET 2006
> I think my main points are still valid:
> - There may well be a need to tag orthography distinctions that are
> particular to specific languages, but IMO the fact that they are
> language-specific automatically disqualifies them from any possible
> coding in ISO 15924 (with or without some re-statement of its scope
> and purpose).
> I think you'd probably agree with the first point.
Well, even this first point is problematical, because it immediately
runs afoul of Hans and Hant, which is an orthographic distinction
in Han, encoded as a "script variant" in 15924, despite the fact
that it applies only to one language. (You could quibble that "Chinese"
is actually an entire family of languages, which is genetically
true, but nearly everybody in IT, as well as the Chinese government,
behaves as if written Chinese were a single language, and "Hans" and
"Hant" are just two ways to write it.)
This point is probably on thin ice for the Syriac script variants,
as well. And what about systems which *aren't* encoded at this point --
perhaps Traditional Malayalam versus Reformed Malayalam. That is
basically an orthographic distinction within a single script for
a single language, rather
than a script variant on the order of Latf, but it has about as
much justification for script codes as Hans versus Hant, and if the
political stinkus about it got hot enough, people would show up
at the door of the ISO 15924 JAC demanding that script variant
codes be registered, so they could appropriately tag text represented
in one form versus the other.
> - A system for written notation of language that is not specific to
> one particular language is, IMO, a potential candidate for coding
> in ISO 15924 (with or without some re-statement of its scope and purpose).
> The second point is vague enough that it's probably not worth debating,
> though I think you probably wouldn't object to that statement given its
Yep. Effectively that is what is happening now, anyway. As I said in
the first response on this topic, people are coming in with
"distinctions in writing systems" that they want to tag, and they
want script codes to do so.
> - And, more specifically (this is a bit of an amplification of what I had
> said in the earlier message), I don't think it is an unreasonable stretch
> of the scope and purpose of ISO 15924 to codify as a script variant
> "Latin-based phonetic transcriptions".
> The third is my particular variant of a proposal to code something in
> ISO 15924 -- I am not inclined in favour of coding specifically IPA
> (or Harrington's system) in ISO 15924. And clearly this is subject to
It is not an unreasonable stretch. Reasonable people are clearly making
that stretch. But it lets the orthography camel stick his nose further
in the tent, and I think it is more or less inevitable that, to
paraphrase Mark, that camel is going to come all the way into the
tent, drink all the date wine, and then end up lying in the gutter.
More information about the Ietf-languages