W3C home > Mailing lists > Public > wai-xtech@w3.org > August 2008

Re: some reflections on @alt usage (and summary of research so far)

From: Ian Hickson <ian@hixie.ch>
Date: Sat, 23 Aug 2008 10:23:39 +0000 (UTC)
To: Steven Faulkner <faulkner.steve@gmail.com>
Cc: public-html@w3.org, W3C WAI-XTECH <wai-xtech@w3.org>
Message-ID: <Pine.LNX.4.62.0808231005091.14795@hixie.dreamhostps.com>

On Sat, 23 Aug 2008, Steven Faulkner wrote:
> In your righteuos efforts to retrofit some 'research' to your 
> unsubstantiated claims you have chosen to conflate the image gallery use 
> case with the 'simply is no text that could do justice to the image' use 
> case.
> Or is it your contention that the images in the photo sites you cite are 
> of the category 'simply is no text that could do justice to the image' 
> if so, your abilites or your motivation to provide text alternatives is 
> very limited indeed.

Speaking with my Google hat on for just this paragraph, I can assure you 
that with Picasa Web Albums, if we offered our users the opportunity to 
specify alternative text, most wouldn't use it, if we required them to 
provide it, most would provide bogus text, and if we forced them to 
provide useful alternative text, they would all find one of our 
competitors' sites and give up on Picasa altogether. (Google hat off.)

In practice, photo sharing sites will never have alternative text 
available for the vast majorty of their images. Pretending otherwise is 
neither realistic nor productive.

> The example from the spec that was the subject of your statement was
> not one identified as from a photo site, it is the rorshach example.

There are five examples of images that fall into the "Images whose 
contents are not known" category in the spec:

 1. A photo upload site (actually a real image on Flickr)

 2. A photo from a blind photographer's blog

 3. A captcha with a link to an audio captcha instead

 4. A page like Henri's image report tool

 5. Another photo upload site.

> I suggest the basis for something that could actually be described as
> relevant research in this case would something like
> a) select from a random sample of images out of the context in which
> they were published as web content, those which you consider fit the
> category you defined "Sometimes there simply is no text that can do
> justice to an image."
> b) go and look at the images in context and see if descriptive
> identification is provided elsewhere on the page that is unambiguously
> associated with the image (not just an implied visual association).

It's not clear to me what point this would provide research for.

The point that I had been arguing which started this thread was that for 
the three cases where using alt={} would be valid per the current spec, 
namely automated image collection (e.g. Webcams), tools dealing with 
unknown imags (e.g. photo upload sites, image-report-like tools), and 
authors who haven't seen the image (e.g. blind photographers asking their 
friends for descriptions), the metadata about the image would always be 
included near the image and would therefore always be redundant if 
included in the alt="" attribute.

I have shown that this is true for photo upload sites (at least, I have 
shown it to my satisfaction, and nobody has shown a counter-example). I 
could go on and show it for webcams, blind users' blogs, and so forth, but 
it would probably be easier for everyone for you just to show a few 
counter-examples, if there really are any.

Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Saturday, 23 August 2008 10:24:00 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:25:22 UTC