- From: David Dorward <david@dorward.me.uk>
- Date: Mon, 20 Jun 2005 13:58:59 +0100
- To: w3c-wai-gl@w3.org
On Mon, Jun 20, 2005 at 08:44:30AM -0400, Bailey, Bruce wrote: > (2) We can't require the presence of ALT attributes on images to be P1 > because we know that is not *really* necessary. After all, one can > find many examples where the root file name is plenty meaningful. That assumes the user agent displays the filename to the user, not all do. Links (for example) treats an image with no alt attribute as if it had alt="", this is a great boon when faced with a page that uses a hundred spacer.gif files, but not so when important content is hidden in an image. Can we really depend on specific error recovery features of particular user agents? Especially when they are harmful in some circumstances? Or do we expect end users to try to access a page in as many user agents as possible until they find one which can compensate in a way that is useful for that page? -- David Dorward http://dorward.me.uk
Received on Monday, 20 June 2005 12:59:04 UTC