W3C home > Mailing lists > Public > w3c-wai-au@w3.org > April to June 1999

Re: null alt revisited

From: Charles McCathieNevile <charles@w3.org>
Date: Thu, 8 Apr 1999 13:12:55 -0400 (EDT)
To: William Loughborough <love26@gorge.net>
cc: "w3c-wai-au@w3.org" <w3c-wai-au@w3.org>
Message-ID: <Pine.LNX.4.10.9904081309040.15377-100000@tux.w3.org>
It is an error to leave out an alt attribute on various elements.

It is, from an accesibility point of view, an error not to put in a value
which is relevant. So unless a relevant value is assumed to have been
provided, the tool should not that there is an outstanding error. The
easiest way to do this is by making a validity error. I don't see that
this is a problem from an accessiblity point of view. In terms of actual
validation, the specification of HTML says some stuff about what should be
in there. The fact that it is not machine-checkable whether it is
appropriate doesn't make it correct to have "some unknown image" as the
value for each image on a website. It is just that machine validation
can't tell, whereas it can tell if there is nothing there.


On Thu, 8 Apr 1999, William Loughborough wrote:

  A friend writes "I think alt=" " would be effective in forcing the
  webauthor to do SOMETHING but I also think we should be hounding browser
  vendors to break any page on which NOTHING was done with it".
  This was in response to my proposal that the default in an authoring 
  tool be alt=" " *provided that this will force modification* to either
  enter non-white-space text or, in the case of "trivial" graphics, a null

--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 Thursday, 8 April 1999 13:13:01 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:39:42 UTC