W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2006

RE: "untill user agents" retired?

From: Alastair Campbell <ac@nomensa.com>
Date: Wed, 1 Feb 2006 10:39:52 -0000
Message-ID: <2A876A583754DD4E8E03CFE899FA1606397477@saturn.intranet.nomensa.com>
To: "wai-ig list" <w3c-wai-ig@w3.org>
Cc: "Patrick H. Lauke" <redux@splintered.co.uk>

Patrick wrote:
> it can't be "all", if the user agents don't follow the basic W3C 
> standards, have major flaws and bugs, or simply require non-standard 
> hacks to work.
(snip)
> How's this as an inclusive limit: user agents that adhere to the W3C 
> specifications.

I wholeheartedly agree, and would like to add that user agent
technologies are unlikely to improve without prodding, and the most
effective kind comes from their own customers/users.

If there are examples where people *have* to use a particular technology
(that does not meet the UAAG) and there is no choice, fair enough, it's
probably a case for an "until user agents" clause or part of a
'baseline' level. (Screen readers having to use Win/IE at the moment for
example).

However, for new UAs to come along and not meet the basics or not be as
good as current UAs is no excuse. The user has to meet the content
author halfway along the guidelines continuum. Complaining because their
new, whizzy but inaccessible user agent doesn't work with a perfectly
good site is not progress.

Following the WCAG guidelines should protect an organisation from
invalid complaints.

The UAs have done some great things to cope with the inaccessible sites
out there as their primary focus is that their customers can use sites
in general, not just accessible ones.

Roger makes some very persuasive points, the main rebuttal of my line of
thinking is that:
"there appears to be no requirement for a nominated baseline technology
to be supported by a significant proportion of assistive technologies
that are in current use.
This could result in sizable shift in the onus for accessibility away
from the site developer and proprietor and onto the users of assistive
technologies: That is, it will be up to the disabled person to obtain
(purchase) the appropriate technology to access a site, rather than the
responsibility of the site proprietor to ensure their content is
accessible to users of a wide range of current assistive technologies."

I largely agree, the important aspect is the baseline level. 

User agent technologies are a moving target, and I don't think it's
reasonable for every site developer to understand all the technologies
that might be used. A site developer should be able to follow the WCAG
guidelines & techniques without having to learn to use umpteen different
UAs. (I see more harm than good come from developers that try screen
readers briefly.)

If the version 2 guidelines did use a baseline or "until user agents"
clauses, I'd hope that there was also a mechanism to inform people of
changes! E.g: 
"Latest news from the WAI: user agents no longer need text within inputs
to understand them" ;-)

David wrote:
"I know there are new and wild technologies coming and I don't know what
impact they are going to have, but we are loosing valuable ground on
*accessibility* if we do not leave a door open for users to access
content and ui functions"

I don't really understand this, new technologies have a responsibility
to be at least as good as the old. For a user to *choose* a new
technology that doesn't work as well as the old it is really their own
problem and not one of accessibility. 

There is always going to be a conflict between people using old
technologies (for familiarity or expense reasons) and the duty on site
owners to support a wide range of technologies. I think the only
reasonable approach for the WCAG V2 is to assume that UAs should meet
the UAAG guidelines, with specific (current) exceptions for certain
groups of people have no choice in what UA they use - which is then an
accessibility issue.

So, who's got a list? ;-)

Kind regards,

-Alastair

-- 

Alastair Campbell   |   Director of Research & Development

Please refer to the following disclaimer for this message:
http://www.nomensa.com/email-disclaimer.html
Received on Wednesday, 1 February 2006 10:40:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:24 GMT