On Jan 10, 11:13am, Nick Arnett wrote: At 07:07 PM 1/10/97 +0100, Chris Lilley wrote: > > >Yes, it is worth stressing that language is a much more important > >tag or attribute when using speech synthesis. It is easy to tell > >que j'ecrire en Français pour un peu, and then switch to back to > >English. A speech synth would get very confused without explicit markup. > > That's not really a good assumption. Morphological analyzers have very > little trouble identifying languages. It's a lot easier to toss the text > against something that'll automatically recognize it than to expect writers > and editors to insert such tags. While I have no problem with people using morphological analyzers to *insert* such tags, that is just one option. Far simpler is to have the author of the document insert an attribute. The thought of Alta Vista performing morphological analysis on each document per search, or the browser performing it before starting to display, makes me shudder. <body lang=fr> is hardly a great burden, compared to writing the rest of the document. > A standard set of tags is a good idea, but one shouldn't assume that a lot > of this can't be automated and reviewed by humans. Automation and review is a good model, particularly for large publishers with large stocks of legacy material to repurpose. The fruits of this process should, however, be marked up using the existing lang tag and attribute, so it does not have to be repeated. -- Chris Lilley, W3C [ http://www.w3.org/ ] Graphics and Fonts Guy The World Wide Web Consortium http://www.w3.org/people/chris/ INRIA, Projet W3C chris@w3.org 2004 Rt des Lucioles / BP 93 +33 (0)4 93 65 79 87 06902 Sophia Antipolis Cedex, France
Received on Monday, 13 January 1997 08:56:47 UTC