- From: David Clark <dmclark@cast.org>
- Date: Fri, 3 Sep 1999 09:22:26 -0400
- To: "Charles McCathieNevile" <charles@w3.org>, "Kynn Bartlett" <kynn@idyllmtn.com>
- Cc: <w3c-wai-ig@w3.org>
Kynn & Charles,
The issue is the confusion between "incompatible" and "unrecognizable".
Attributes and values can be unrecognizable in a particular browser, but
have no negative effect on the rendering or the page in that browser.
Another way to think of this is as an "and vs or" problem. What these
validators are doing is checking to see if a tag or value is valid in all
(or a majority of) browsers, rather than if it is valid in any browser.
David M. Clark
CAST, Inc., 39 Cross St., Peabody, MA 01960
Tel 978-531-8555 x236 - Fax 978-531-0192
Email dmclark@cast.org
http://www.cast.org/bobby/
-----Original Message-----
From: w3c-wai-ig-request@w3.org [mailto:w3c-wai-ig-request@w3.org]On Behalf
Of Charles McCathieNevile
Sent: Friday, September 03, 1999 12:16 AM
To: Kynn Bartlett
Cc: w3c-wai-ig@w3.org
Subject: Re: Fwd: NetMechanic Results on www.hwg.org
Kynn,
The issue of greatest concern is that their advice seems to be based on
misunderstanding of how browsers work. To say that longdesc is incompatible
with browsers is vague enough as to be misleading. To advise against using
it
on this basis is, as you point out, extremly detrimental to accessibility
and
in my humble opinion to the development of the web itself.
I'll take a little time to compose a polite letter and make the points as
clearly as I cn with appropriate suggestions and references, so I exect I'll
send it some time next week.
Charles McCN
On Thu, 2 Sep 1999, Kynn Bartlett wrote:
I ran a test on http://www.hwg.org/ at http://www.netmechanic.com/
today.
They told me that I'd "misspelled" a lot of words, like XML and
XHTML, plus they kindly informed me that <!DOCTYPE> is not a valid
HTML tag.
The thing that scares the heck out of the accessibility instructor
within me is rating of our "browser compatability". The things it
identifies as "incompatible" all fall into one of the two
categories:
(a) Style sheet support, such as SPAN elements or CLASS/ID
attributes, which specifically degrade gracefully when CSS
is not present, and
(b) Attributes/elements added for accessibility's sake, such
as IMG LONGDESC, HTML LANG, ABBR, or LABEL.
Now, I don't mind being identified as "incompatible" with old
versions of browsers, but the PROBLEM here is their "advice" to
web designers:
"Try to avoid using a tag or attribute if it is incompatible with
more than 10% of your audience."
THIS IS SO UTTERLY WRONG.
Here's what they say for LONGDESC:
Tag: IMG
Attribute: LONGDESC
Lines: 78
Visitors Affected: 99.00%
Microsoft:
3: N
4: N
5: N
Netscape:
2: N
3: N
4: N
In other words, since 99% of browsers out there don't use the
information, you shouldn't include LONGDESC.
In short:
Their advice is actively ANTI-ACCESSIBLE.
Would someone care to join me in writing to these people about the
inherent problems in advising against the use of proper HTML code
as they are doing here? It might mean more if we all got together.
This could also be an issue for the evaluation/repairs tools to
look at, but I don't know what exactly you'd want to do.
PS: The url below expires in 2 days.
--Kynn
>Date: Thu, 2 Sep 1999 22:59:18 -0400
>To: kynn@kynn.com
>Subject: NetMechanic Results
>From: webmaster@netmechanic.com
>
>[ad snipped]
>
>NetMechanic has completed the tests you requested for:
>
>http://www.hwg.org/
>Job Configuration: One Page, Local Links, Remote Links, Images, HTML
Standard: HTML Version 4.0 Standard
>
>You can find your results at:
>
>http://beta.netmechanic2.com/summary.cgi?f=244225225-06389&s=NetMechanic&fv
=2
>
>Reports will be stored at this URL for the next 2 days.
>
>
>
--
Kynn Bartlett <kynn@idyllmtn.com> http://www.kynn.com/
Chief Technologist, Idyll Mountain Internet http://www.idyllmtn.com/
Catch the Web Accessibility Meme! http://aware.hwg.org/
--Charles McCathieNevile mailto:charles@w3.org
phone: +1 617 258 0992 http://www.w3.org/People/Charles
W3C Web Accessibility Initiative http://www.w3.org/WAI
MIT/LCS - 545 Technology sq., Cambridge MA, 02139, USA
Received on Friday, 3 September 1999 09:16:50 UTC