W3C home > Mailing lists > Public > public-html@w3.org > August 2008

Re: Flickr and alt

From: Sam Kuper <sam.kuper@uclmail.net>
Date: Tue, 19 Aug 2008 15:53:11 +0100
Message-ID: <4126b3450808190753h6a5b9f4bh379a3319039801d0@mail.gmail.com>
To: public-html@w3.org
2008/8/19 Smylers <Smylers@stripey.com>

> Sam Kuper writes:
> > 2008/8/19 Boris Zbarsky <bzbarsky@mit.edu>
> > > I do want some recognition of the fact that it's quite easy to make
> > > [accessibility conformance] too much pain for a large fraction of
> > > authors, and that this is something to avoid.
> > I think the level of pain would depend not just upon what is mandated,
> > but also upon what could be used to make the mandated things less
> > painful. E.g., if Flickr not only asked users for alt text, but also
> > explained what alt text should be like and why it is important,
> > Flickr's users might mind entering it less than they would if it was
> > simply insisted upon with little or no explanation. I think users
> > should be able to temporarily override requirements, too. In the case
> > of Flickr, this might mean that users could upload images without
> > providing alt text, but that Flickr would put a little exclamation
> > mark (or suchlike) next to each such image, to remind the user to
> > enter the missing alt text.
> OK.  If Flickr did all the above, do you think their output should be
> deemed to be compliant HTML?

I think it's important to distinguish between compliance in an authoring
tool and compliance in output.

In the above case, Flickr (and any other AT) would be a compliant authoring
tool if it:

   - checked the output for validity,
   - notified any authors creating invalid output of the fact, and explained
   the kind of invalidity and why (in layman's terms, but with links to
   technical documents) it is considered problematic,
   - provided reminders of the invalidity on each occasion an invalid
   document was re-edited.

As for the output, it would still not be valid HTML unless, well, it was
valid HTML.

Received on Tuesday, 19 August 2008 14:53:52 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:37 UTC