RE: can we get an i18n look at this Compound Document issue?

Hello Al, WAI-folks,

The i18n WG discussed this today during our telecon.

We originally raised a similar point to the CDF folks, but in the end we
accepted that we cannot expect them to control this when addressing
combinations of markup as-is.

Thus we expect that a document included by reference with no language (or
other similar) information of its own, will be treated as if we cannot say
anything about the language (or other characteristics) of the included
content, ie. it would be handled in the same way as the document would be
handled if read on its own rather than included.

Note that the ITS WG had to address a similar issue (raised by the DITA)
folks, with regard to translatability declarations in documents included by
reference, and came to the same conclusion.

We think the best way to address this issue is to strongly encourage people
to include language (or other) information in the documents to be
referenced, and the ITS WG will be adding a Best Practise to this effect to
one of its documents.

We expect that content included in a document not by reference, eg. a
fragment of SVG inside an HTML document, *will* inherit any language (or
other similar) information declared higher up in the HTML tree.  What we
describe above only applies to documents included by reference.

Hope that helps,
RI


============
Richard Ishida
Internationalization Lead
W3C (World Wide Web Consortium)

http://www.w3.org/People/Ishida/
http://www.w3.org/International/
http://people.w3.org/rishida/blog/
http://www.flickr.com/photos/ishida/
 

> -----Original Message-----
> From: Al Gilman [mailto:Alfred.S.Gilman@IEEE.org] 
> Sent: 03 June 2006 21:03
> To: ishida@w3.org
> Cc: wai-liaison@w3.org
> Subject: can we get an i18n look at this Compound Document issue?
> 
> 
> Reference: Find "i18n" in
> http://lists.w3.org/Archives/Member/w3c-wai-pf/2006AprJun/0131
> 
> The issue is roughly that accessibility wants the natural 
> language to be documented (identified in a format-defined way 
> amenable to machine
> recognition) in all contexts where there is natural language text.
> the i18n group are the experts as to how to do this.
> 
> CDF group took the position that this should be entirely 
> managed at the constituent document level (particularly here 
> for compound documents *by reference*) An alternative would 
> be to define heredity within what the user will perceive as a 
> single page.
> 
> Please refer as stated above for the discussion to date.
> 
> Is there a clear precedent in what i18n has done that resolves this?
> I see in the draft XML internationalization BP the tools and 
> author imperatives to do it document by document. Have you 
> given any thought to heredity in the CDF context? Etc.
> 
> Whaddya [as in, what does i18n] think?
> 
> Al
> 
> PS: Sorry I haven't gotten this to you sooner.  I missed you 
> at the 19 May Hypertext CG call.
> http://www.w3.org/2006/05/19-hcg-minutes.html#action02
>

Received on Tuesday, 13 June 2006 15:47:25 UTC