W3C home > Mailing lists > Public > w3c-wai-ua@w3.org > July to September 2001

Re: HTTP USER-AGENT header question

From: Al Gilman <asgilman@iamdigex.net>
Date: Wed, 18 Jul 2001 18:29:13 -0400
Message-Id: <Version.32.20010718072012.042a3f00@pop.iamdigex.net>
To: Louis Atkinson <louis@highermind.org>
Cc: <w3c-wai-ua@w3.org>
>On Wed, 18 Jul 2001, Louis Atkinson wrote:
>
>  My other main problem is that many people seem to be using screen reader
>  plug-ins for IE and Netscape which do not allow for easy access to the
>  relevant user agent string, but I am still convinced that there is a need
>  for there to be some intelligence on the server side for these things.
>

AG::  

What Charles sent you to, in terms of CC/PP, is where to look for something
usable to possibly emerge.  But this does not give you something, nor is there
anything else, that you can use today.  The fact that the user is using AT,
never mind precisely what AT, is something that they have good reason not to
advertise in random web hits.

Further, some intelligence would be fine, but an automatic redirect without
recourse is not.

Build the choices into the site in static HTML forms.  Look at the "universal
usability checklist" from University of Maryland for things to cover in detail
pages accessible from a [required] "how to use this site" guide section of the
site and an [optional] "access features" section.  If you do the latter, look
at the self-representation template from the GSA for equipment vendors to
publish how they address Section 508 requirements.  Adopt a similar story line
to say how your site approaches the access requirements of the WCAG.

Do not put any look and feel variations into your site that operates by
sleight
of hand that are beyond the reach of some 'manual' control or other you give
the user.  You will just make more enemies than you win friends.

The feasible success rate of automatic 'intelligence' stuff in estimating what
the user will find advantageous is currently at a performance level where
it is
essential to make whatever the automation comes up with enter into the process
as nominations for, not constraints on, the choice/profiling of the view
through which the user operates.

See also

Web Content Accessibility Guidelines Working Group (WCAG WG) Home Page
 <http://www.w3.org/WAI/GL>http://www.w3.org/WAI/GL/
-- in particular find "server-side" on that page.  This group is where
there is
the most committed, continuing interest in the kind of technique that you are
interested in.  But it's not that active a thread.

giving the user the last word
 <http://lists.w3.org/Archives/Public/w3c-wai-ua/2001JulSep/0099.html>http:
//lists.w3.org/Archives/Public/w3c-wai-ua/2001JulSep/0099.html

Pre-testing for user-profiles to enhance accessibility of  ...
 <http://www.csun.edu/cod/conf2000/proceedings/0078Velleman.html>http://www
.csun.edu/cod/conf2000/proceedings/0078Velleman.html

Al

At 01:29 AM 2001-07-18 , Charles McCathieNevile wrote:
>Hi Louis,
>
>I think you should look at the CC/PP work being done at W3C. This came from
>the world of mobile user agents, and is designed to allow for content to be
>served to a wide variety of systems, according to their needs.
>
><http://www.w3.org/Mobile/CCPP>http://www.w3.org/Mobile/CCPP
>
>cheers
>
>Charles McCN
>
>On Wed, 18 Jul 2001, Louis Atkinson wrote:
>
>  Hi,
>
>  apologies if this is not apropos, but searching the list archives gave me a
>  lot of hits, and the ones I browsed did not answer my question.
>
>  Question:
>  Has there been any discussion on creating a standard for how accessible
user
>  agents will identify themselves to the server?
>
>  I have been trying to compile a list of the USER-AGENT HTTP headers used by
>  accessible browsers. It struck me that if they all contained a string in
>  common (similar to the way that many browsers use "Mozilla") it would make
>  it easier to send them to a version of the site that has no graphics or
>  specialized scripting. Something like having the string "WAI", rather than
>  "Mozilla", and then the various product tokens, is what I am thinking.
>
>  I am holding text content in a database and wrapping the interface
around it
>  to suit the capabilities of the user agents which access it, so can and do
>  do this.
>
>  My other main problem is that many people seem to be using screen reader
>  plug-ins for IE and Netscape which do not allow for easy access to the
>  relevant user agent string, but I am still convinced that there is a need
>  for there to be some intelligence on the server side for these things.
>
>  The page I have made to gather and list USER-AGENT strings is located here:
> 
<http://www.highermind.org/design/user_agent/>http://www.highermind.org/desi
gn/user_agent/
>  so if you know anyone who wants to add to it, it has a script that will get
>  the string and add it to the database.
>
>  Thank you very much.
>
>  Best regards,
>  Louis Atkinson
>  Higher Mind Productions
>  <http://www.highermind.com/>http://www.highermind.com/
>
>
>-- 
>Charles McCathieNevile   
<http://www.w3.org/People/Charles>http://www.w3.org/People/Charles  phone: +61
409 134 136
>W3C Web Accessibility Initiative    
<http://www.w3.org/WAI>http://www.w3.org/WAI    fax: +1 617 258 5999
>Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia
>(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex,
France)
>  
Received on Wednesday, 18 July 2001 18:17:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 06:50:58 GMT