Re: Keeping the structure, Moving the meaning Re: samp, kbd, var

Le 25 août 06 à 15:32, David Woolley a écrit :
>> The original SGML specification is in front of my eyes right now and
>
> I suspect that you don't actually mean the SGML specification, but  
> rather
> the DTD for one of the W3C languages.  Not having paid for the real  
> SGML
> specification, I don't know if it says much about intended uses, but
> my perception is that the typical intended use was for things like
> Electronic Data Interchange, where the detailed structure of a  
> business
> document, like an invoice or bill of lading, would be marked up,
> or for documents to be placed into an indexing system, where it  
> would be
> important to identify things like addresses.

SGML defines a mechanism to create markup languages, exactly like XML.
HTML is about Text with links.
An HTML specification tries to define the nature of a text aka  
"Structure"

> What I've seen as typical examples of roles, many of which actually
> represent primary structure, like navigation bars, are things that
> would need to marked by elements in the above sorts of SGML
> applications.  It's actually things like strong/bold that wouldn't -
> in fact, in EDI, I doubt they would have any presence at all.

The discussion is about role/property attributes, not only role.

It's another issue which has been raised
The spec is not very clear about the proper role of each attribute.
http://lists.w3.org/Archives/Public/www-html-editor/2006JulSep/0107

It seems reading the specification that "property" is the right  
attribute for the proposal.


-- 
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager, QA Activity Lead
   QA Weblog - http://www.w3.org/QA/
      *** Be Strict To Be Cool ***

Received on Friday, 25 August 2006 07:25:02 UTC