W3C home > Mailing lists > Public > www-html@w3.org > May 2007

RE: "Pave The Cowpaths" Design Principle

From: John Foliot - WATS.ca <foliot@wats.ca>
Date: Tue, 15 May 2007 09:14:24 -0700
To: "'Frank Hellenkamp'" <jonas@depagecms.net>, "'Laura Carlson'" <laura.lee.carlson@gmail.com>
Cc: <www-html@w3.org>, <public-html@w3.org>, "'Roger Johansson'" <roger@456bereastreet.com>, "'Lachlan Hunt'" <lachlan.hunt@lachy.id.au>
Message-ID: <007001c7970c$1f18a8e0$6501a8c0@bosshog>

Frank Hellenkamp wrote:
> 
> Somewhere in the discussion, the seatbelt-example came up:
> 
> In germany for example [1] there is a law for institutions of the
> state (and some federal states too), that needs their websites to be
> completely accessible. That includes most universities, many museums
> etc.   
> 
> For these scenarios it is not enough to "pave the cowpaths" of the
> usual blog/private website/corporate website. And you do that, if you
> try to get most things from the web, how it is today.  

And in some territories, this may soon "spill over" into the private sector
as well.  Many Accessibility advocates are watching and waiting the outcome
of the Target.com challenge with great anticipation - it's outcome *may*
have far-reaching ramifications for web developers and large corporate
sites.  

> 
> The inaccessable web might be enough for most websites, but there are
> many more cases, in which web-developers are forced to generate
> semantic code.  

Thank you Frank for echoing one of the themes I have been trying to put
forth.  Not every website nor web author can just go ahead and do what they
want - often they are mandated to do something, even if it isn't the best
way, or the fastest, or the coolest.  They don't care what color you paint
your bike shed, but in some territories, your bike shed *MUST* be up to the
building code, or it's torn down.

HTML5 could help here, but only if *all* aspects of the accessibility debate
are listened to with calm and receptive ears.  Comments like "Semantics for
semantics sake" often leave us feeling that this is not the case. (And I'm
sorry Lachlan, those were your words - they are out there and like the
future code we are talking about now, once it's out, you can't recall them
back in for fine tuning... Something to also consider in this discussion) 

> 
> Obviously there is a lot to learn, but both HTML5 and XHTML2 could
> improve that, 
> 
> - if they are well defined,
> - got implemented somewhere in the future and
> - well documented, so that book-authors and then web-developers could
> base their work on it. 
> 
> If HTML5 is the way to go: Fine with me.

And Frank, I suspect that done right, most accessibility advocates would
also agree.  We want improvements to the language as well - we have a
constituency that is also lacking some tools with the current spec, and be
it XHTML 2 or HTML5, improvements are improvements.  

> 
> * the role-attribute e.g. is *much* cleaner and less error-prone than
> redefining the class-attribute. 
> 
> [1] http://www.einfach-fuer-alle.de/artikel/bitv-reloaded/

Thank you for your statement of support for @role.  

I sincerely hope that the @class proposal for semantic definition goes away.
Some of the newer element attributes such as @role and @profile (right Dan
C.?) may not yet have well established cowpaths in the wild, but how about
we try and herd the cows in the right direction once and a while?
Once-upon-a-time the most traveled cowpath was using tables for all layouts,
and we've almost completely finished re-growing grass over that path...

JF
Received on Tuesday, 15 May 2007 16:16:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:10 GMT