- From: Jukka Korpela <jkorpela@cc.hut.fi>
- Date: Fri, 26 Sep 1997 07:48:53 +0300 (EET DST)
- To: www-html@w3.org
On Thu, 25 Sep 1997, MegaZone wrote: > Once upon a time Jordan Reiter shaped the electrons to say... > >semantic-specific derivation; instead, it would be used only to indicate: > >"Here is something different, that has a language associated with it but is > >being placed directly within the text itself." > > I don't see why <SPAN> with a LANG attribute DIFFERENT from the document > default LANG is not enough to implicitly imply this. (Nit-picking: how would you handle English text within some German text in a document with English as default language.) Perhaps we could handle things like the plural of <span lang="la">populus</span> is <span lang="la">populi</span> that way, but how would it work for the plural of <span lang="en">ox</span> is <span lang="en">oxen</span> ? There are things involved here: the logical distinction being an object language and a metalanguage (in the linguistic sense), and the distinction between various languages. If one cannot do the former distinction using HTML elements (or perhaps attributes), one has to resort to some form of physical markup at least when the object language is the same as the metalanguage. I regard explicit quotes in the text as comparable to physical markup. The lang attribute should _only_ define the language used in a part of a document. It should not imply things like quotation or reference to words or phrases as linguistic objects. Yucca, http://www.hut.fi/home/jkorpela/
Received on Friday, 26 September 1997 00:49:11 UTC