W3C home > Mailing lists > Public > www-html@w3.org > August 2006

Re: XHTML2 and JS ads

From: David Woolley <david@djwhome.demon.co.uk>
Date: Thu, 10 Aug 2006 21:39:26 +0100 (BST)
Message-Id: <200608102039.k7AKdRs12767@djwhome.demon.co.uk>
To: www-html@w3.org

> 
> It would have a lot more than just an <adverts> element -- commercial Web 
> sites would need elements for things like context menus, asides, footers, 
> meters, etc, and would need to define APIs for how these things interact 
> with the DOM, as well as having APIs for graphics, sound, drag and drop, 
> networking, and so forth.

Yes.  It is true that if, as I was doing when I said that one would need
an advert element, assumes that the basic principle of XHTML2, that it
represents structure.  However, I don't believe that commercial web
designers care about representing structure, and, in general, representing
semantics can be counterproductive, because it shifts control from the
producer to the consumer.

There is also an imperative in commercial web site design to be 
different, preferably from the competitors, although in reality  more
to be different from last months new web site.  This does defy the
usability principle of having a consistent user interface, which would
be the potential benefit of having structural elements for them.  When
they were first introduced, they would be new, and would be used, but
that wouldn't last long.
 
> As far as I can tell, actually, XHTML2 is aimed at simple documentation 
> pages, and complex tax forms.

I get the distinct impression that you think that there is no real
use case for XHTML2.  I would suggest that the number of in house
content management systems that seem to be around suggests that there
is a demand for something closer to the original aims of HTML.  In 
such an environment, the obvious, and WYSIWYG, alternative would have
been MSWord.

On the other hand, if it is not used much in anger, it can still be
worthwhile in terms of keeping alive the original, pre-commercial,
goals of HTML, some of which may still influence the commercial track.

>From a commercial use point of view, I think all versions of HTML are
an abberation.  The network hyperlinking and, in the beginning, the
fact that you could author using tools that you already had, proved 
sufficiently attractive to outweigh the fact that it is a markup
language and not even a page descriptions graphics one.

However, this has had beneficial consequences in making search engines
easy to develop (whilst they now cope with PDF as well, I'm not sure
that would have happened so fast if the web had been based on PDF) and
it has set a standard in terms of creating accessible documents, which
has even resulted in PDF now having an HTML-like structural overlay,
even if its use is not that common.  

My feeling is that if it hadn't have been possible to show that
the underlying text and structure could be extracted, awareness of the
possibility of providing alternative means of accessing a document
wouldn't have been as great.

> The biggest reason _not_ to have an <advert> element is that people 
> wouldn't use it, thus making it useless. This is the same argument against 
> having <img longdesc>, <object standby> (or in XHTML2, <standby>), etc.

Yes.  Basically they benefit the consumer, and, in particular those
consumers who are likely to have low buying power, or be discriminating
in what they purchase, rather than following fashion.  They don't 
benefit the seller, with the fast connection and the desire for 
purchases.

longdesc has the particular disadvantage that images are typically used
either, in a catalogue, as an alternative to spending the time to provide
an informative description of the product, and in pure advertising copy,
to send messages that wouldn't be allowed in text, because they would
be making unsubstantiated claims about the product.

The big danger, if XHTML2 is produced, is that it will become fashionable
and lots of people will try to use it for commercial sites, for that reason,
rather than because it is suitable.
Received on Thursday, 10 August 2006 20:40:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:07 GMT