[uf-new] figure microformat

[Benjamin Hawkes-Lewis]

I wrote:
> For a captioned photo I'd recommend either using a concise label 
> (alt="Einstein") or a brief description of what you can see, depending 
> on your editorial focus and what information is provided by text elsewhere.
> 
> ALT="Einstein photographed at 68, eyebrow arched as he looks out to the 
> camera, face creased with wrinkles, with an impressive mustache and a 
> scraggy mane of white hair."
> 
> would be one attempt to provide an actual text equivalent for (say):
> 
> http://en.wikipedia.org/wiki/Image:Albert_Einstein_Head.jpg

You may have noticed that my example features a different photograph 
than that referred to in the draft's example. This was because there was 
more than one photograph of Einstein by Ehrenfest, and there wasn't 
enough equivalent text to identify which one was indicated. It occurred 
to me afterwards that this is actually a rather nice example of how the 
example markup /didn't/ provide enough text equivalence, at least not 
for this reader. :)

> Whatever you think of the necessity of alternative text for content 
> images, such supplementary uses of ALT and LONGDESC need (it seems to 
> me) to have some sort of place in the proposed draft.

To rephrase this in a (sightly) more actionable way: if we're talking 
about extracting content images and information about them from web 
documents, then we either need to try to dictate how authors should 
provide text equivalents for images or we need to cater for the 
different ways they already do (and don't) provide text equivalents. The 
first is hard because it's hard to achieve consensus (see the differing 
opinions in the original reading list I provided) and the second is hard 
because practice is variable.

--
Benjamin Hawkes-Lewis