W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > April to June 1998

Re: Any browser? table

From: Alan J. Flavell <flavell@a5.ph.gla.ac.uk>
Date: Sat, 25 Apr 1998 15:20:46 +0100 (BST)
To: Jason White <jasonw@ariel.ucs.unimelb.edu.au>
cc: WAI Markup Guidelines <w3c-wai-gl@w3.org>
Message-ID: <Pine.OSF.3.96.980425143127.560B-100000@a5.ph.gla.ac.uk>
On Sat, 25 Apr 1998, Jason White wrote:

> The guidelines should definitely not recommend the use of NBSP or any
> other non-standard technique for achieving particular visual effects in
> user agents that have inadequate support for HTML table markup.

I have a great deal of sympathy with what you're saying here, which is
one reason I've been careful to point out that my own paper was
originally composed for a different purpose, and why I've tried to
steer the discussion away from the tricks that can't pass syntax

However, I'm not sure in what sense you mean that stuffing with
no-break spaces is "non-standard".  It is syntactically valid, and I'm
not aware of anything in the HTML spec with which it is
_inconsistent_, albeit there is nothing in the spec that _guarantees_
that it will work.   

That having been said, one would have every right to argue that the
technique/trick is inappropriate to be recommended in the present
context.  That it is in the nature of a trick is not contested! 

My perception is that Lynx is used by many users who are in the kind
of situation for which the WAI guidelines are intended.  In an ideal
world, Lynx would support TABLEs properly when it was needed; the fact
is that it doesn't.

> It is the
> responsibility of the user agent to provide proper support for HTML
> constructs, not that of the author to compensate for the idiosyncrasies of
> implementations that fail to meet the standards set forth in W3C
> recommendations.

Again, I've much sympathy for the principles, but in a practical sense
one may sometimes have to compromise.  After all, the HTML4.0 spec
says noble things about how speaking browsers are to make tables
accessible, but in practice today such support would be the exception
rather than the rule, and it would be inappropriate for authors to
behave as if such sophisticated browsers as described in the spec were

While I absolutely would not expect to see table tricks in the main
body of the guidelines with a "RECOMMENDED" status, I submit that it
can be useful to offer practical advice to authors, as is presently
done in a series of appendices, and that this kind of advice could be
appropriate there.  The exact nature of the advice offered is of
course fully open to debate, but I suggest you're going too far when
you ask to rule it out of court entirely. 

best regards.
Received on Saturday, 25 April 1998 10:20:56 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:27 UTC