W3C home > Mailing lists > Public > www-validator@w3.org > July 2001

Re: W3W and Bobby Validation

From: Kynn Bartlett <kynn@idyllmtn.com>
Date: Mon, 9 Jul 2001 12:18:11 -0700
Message-Id: <a05100323b76fb1c36756@[10.0.1.19]>
To: "Allan Smith" <allans@thad.demon.co.uk>, www-validator@w3.org
At 6:13 PM +0100 2001/7/09, Allan Smith wrote:
>Bobby 3.2 comes states that

You might want to direct this to the Bobby admins, or perhaps the
W3C's WAI-IG list.

>It suggest the following remedy
>Add a "title" attribute to each FRAME element to describe the 
>purpose and content of the FRAME. For example,
>   <FRAMESET ROWS="100,*">
>     <FRAME SRC="navigate.html" TITLE="Navigation Bar">
>     <FRAME SRC="content.html" TITLE="The Navigation Bar changes the 
>contents of this frame.">
>   </FRAMESET>
>I am not aware that name is a valid attribute. Surely title is correct?

"@name" is a valid attribute for computer-readable identification of
a frame.  The @title attribute is a human-readable description of the
frame, intended for users who can't access the frames.

>Bobby also lists the following as a minor problem:-
>It suggests the following as the format for resolving this problem:-
><HTML lang="en">
>However if I put this attribute in then Bobby gives the following message:-
>5.Unknown attribute LANG in element HEAD. for browser(s): 
>Explorer4.01 (1 instance)
>      Line 6: <HEAD lang="en">
>Why would a suggested solution create another error????

Because Bobby is measuring different things -- accessibility vs. browser
compatibility/support.  IE 4 doesn't "support" the @lang attribute on
the head element.

How should you interpret this, then?  Easy:  Ignore the browser warning,
and go with the accessibility recommendation.

>Bobby also lists the following "Browser Compatability Errors", many 
>of which I feel do not make sense!

That's because browser support doesn't make sense.  Bobby is just
reporting this, but you should typically take it with a grain a salt.
Lack of browser support should rarely be a reason to avoid Doing The
Right Thing, because in most cases -- such as the @lang attribute on
head -- the browser will just ignore the unknown attribute anyway.

>    3.Required attribute LANGUAGE is missing from tag SCRIPT for 
>browser(s): Lynx2.7 (2 instances)
>Surely in item 3. Language is deprecated???

Yes, but "deprecated" doesn't mean the same thing as "supported by
browsers."  (However, I have to admit that a complaint that Lynx won't
understand javascript seems misplaced.)

>However Bobby also states
>An accessibility rating will be given only if the Bobby Approved 
>accessibility rating option at the top of the dialog is checked. 
>When this is checked, the HTML 4.0, Netscape  Navigator 4.0, 
>Microsoft Internet Explorer 4.0, and Lynx 2.7 will also be checked, 
>as these are considered a base set of browsers against which 
>accessibility should be tested.
>How is it, thus, possible to obtain a "clean" Bobby report and still 
>obtain a clean "W3W" report???

I dunno, ask the Bobby people, I guess.  That seems a strange restriction
especially as Navigator4, IE4, and Lynx2.7 are all not compliant with
HTML 4.01, CSS 1 (or 2), or User Agent Accessibility Guidelines.

Practical advice time:  Look at the warnings about various browsers, but
ignore them if you feel like it.  Pay more attention to the accessibility
results and ESPECIALLY the accessibility _questions_ that need to be
human-judged.  And if you don't get "Bobby Approval", who cares?  What's
most important is to Do The Right Thing as far accessibility goes, and
Bobby Approval is just an -approximation- of that.  (It's perfectly
possible to be accessible and not be Bobby Approved, or to be Bobby
Approved and not be accessible.)

--Kynn

-- 
Kynn Bartlett <kynn@idyllmtn.com>
http://www.kynn.com/
Received on Monday, 9 July 2001 15:46:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:13:59 GMT