W3C home > Mailing lists > Public > whatwg@whatwg.org > January 2005

[whatwg] Web Forms 2.0 - what does it extend , definition of same,

From: James Graham <jg307@cam.ac.uk>
Date: Mon, 10 Jan 2005 12:20:45 +0000
Message-ID: <41E2731D.8000008@cam.ac.uk>
Jim Ley wrote:

>On Mon, 10 Jan 2005 10:33:19 +0000, James Graham <jg307 at cam.ac.uk> wrote:
>  
>
>>However the existance of non-semantic uses of HTML only proves
>>that these are possible in the language, not that well written examples
>>are not common.
>>    
>>
>
>When looking at what Web-Application Developers need to create better
>web-applications, surely looking at case studies of example
>web-applications is highly relevant to the discussion? 
>
[...]

>It's my conclusion from seeing the available web-applications today,
>that the web document semantics of HTML are almost completely useless
>and un-used, what's needed is application level semantics.
>  
>

Case studies are relevant. Taking a case study from a vendor known to 
use poor markup and asserting that it is evidence that HTML has the 
wrong kind of semantics for Web Applications is a nonsense however 
(irrespective of whether the conclusion you happen to reach is correct). 
The interesting question the case study provokes is this: why do Google 
(and presumably others) ignore the semantic richness of HTML? Why do 
they, in 2005, use <font> tags when they're a totally unnecessary and 
useless construct? If we had "application level semantics" could we be 
at all sure that these would be used in the correct way given that many 
HTML documents (that are really documents and not applications) ignore 
the document semantics of HTML?

My answer is simply that authors see no obvious benefit from the use of 
semantic markup. Common browsers (not browsers tuned to an accessibility 
requirement) provide no mechanism to get at any of the interesting 
information in a document (and CMS systems abstract away the ugliness), 
so what's the point? Would that be different if we had "application 
level semantics" (I'm not sure I understand exactly what that means)? If 
so it's certianly a worthy goal. But how can we get there from here? 
I've already said that I think XML breaks too many of the things that 
web developers rely on (ability to hack together quick solutions to easy 
problems) to be very useful, so is there a way we can do better?
Received on Monday, 10 January 2005 04:20:45 UTC

This archive was generated by hypermail 2.3.1 : Monday, 13 April 2015 23:08:20 UTC