W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 1998

Re: introducing new semantics to HTML

From: Al Gilman <asgilman@access.digex.net>
Date: Tue, 17 Nov 1998 09:19:04 -0500 (EST)
Message-Id: <199811171419.JAA00480@access2.digex.net>
To: nir@nirdagan.com (Nir Dagan)
Cc: w3c-wai-gl@w3.org
to follow up on what Nir Dagan said:

> I am not a supporter for using WAI defined 
> classes for semantics, or for inventing special 
> anchor names.

What is the actual problem caused by these practices?

> In my view, if WAI wants to add semantics to HTML, it should
> propose new elements (or attributes) to extend HTML4.0. These
> elements should degrade gracefully to HTML4.0 browsers.  E.g.,
> a block level element NAVBAR with content %block; is fine with
> me.

The W3C is having trouble getting people to implement even all
of what is in HTML4.0.  If there is a reason to prefer one 
means of extension over another, we may find it by asking the
people who would have to implement it what is easier for them.

> This is better than giving semantics to the class
> attribute in contradiction to the HTML spec. after people 
> have been using it for more than two years, and accidentally 
> may have used one of the WAI classes.

Where is the contradiction?  CLASS was put in HTML4.0
specifically to be an extension mechanism, to support semantics
not covered in the DTD.  This is just the tip of the iceberg.  In
XML all semantics is after the fact, except for one very abstract
inclusion tree and cross-reference via X-pointer.

> With named anchors there is the same problems but also
> an additional one: 
> When I write a website I choose the URLs and 
> fragment identifiers to have some meaning in the 
> human language of the document. It makes is easier 
> to remember URLs or understand where links go to 
> by looking at the status bar. That's one reason why XML 
> allows ID to have non-ascii characters.

Yes, anchor names that are in the language of the document could
be better.  I think that we are debating how the examples are
constructed anyway, not rigid presecriptions in guidelines.

> If you define a recommended anchor name, in 
> what language should it be written?

1. Define what it should mean

2. Translate this idiomatically into the [five?] core langauges
used at the U.N.  Offer all these as examples of good anchor
names, not required.  The requirement is to make logical starting
points within the document easy destinations by the use of
A[nchor] NAMEs.  Supported by strategically defined examples like
"start-reading," "&agrave;-propos," etc.  

The growing edge would appear to be in XML appications, more than
in adding features to HTML.  We can benefit from reviewing what
works in successful XML applications like the NISO Digital Talking
Book application before drawing conclusions as to what extension
techniques offer the greatest chance of success.

Received on Tuesday, 17 November 1998 09:18:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:28 UTC