- From: Al Gilman <asgilman@access.digex.net>
- Date: Wed, 22 Oct 1997 08:16:02 -0400 (EDT)
- To: w3c-wai-ig@w3.org
Interesting. With your two edits, you have succeeded in bracketing where I would draw a line. to follow up on what Harvey Bingham said: [quoted twice is me> > > >Is that what you want, or would it be better for the comments to say > > > >"To avoid problems with non-visual UAs, provide a brief text > >alternative for the image via ALT, and avoid server-side image > --- [quoted once is Harvey> > I'd be more explicit: substitute for "for" > "describing the information conveyed by" In other words, "To avoid problems with non-visual UAs, provide a brief text alternative describing the information conveyed by the image via ALT, and avoid server-side image..." That one is OK, if not entirely an improvement. > > alt CDATA #REQUIRED -- text alternative to image -- > > > > Agree: but how about -- text alternative describing image -- I would not be happy with that one. The classic example of the icon in the software manual illustrates the difference. "Exclamation point in triangle (in red)" describes the image. "Warning! " captures the information conveyed. For images whose function is primarily symbolic (they allude to a concept) a verbal expression of the same concept is to be preferred to a verbal description of the visual expression of the concept. IMG is too broad in its range of application for one to know from just the IMG element type if it should have a description or an equivalent in its ALT string. Over the long haul maybe we could look at defining elements which are more specific concerning the role of the image. For now, we still need Gregg's CASE statement in the guidelines. As I see it. -- Al Gilman
Received on Wednesday, 22 October 1997 08:16:31 UTC