- From: Henri Sivonen <hsivonen@iki.fi>
- Date: Sat, 3 May 2008 13:08:45 +0300
- To: Daniel Glazman <daniel.glazman@disruptive-innovations.com>
- Cc: "public-html@w3.org" <public-html@w3.org>
On May 3, 2008, at 12:23 , Daniel Glazman wrote: > A conformance checker will never know what precisely is author's > intent. Indeed. But a conformance checker (piece of software) can't check if a page is accessible, either. However, software can help a *person* assess accessibility. I think the Image Report feature of Validator.nu does more to help a person who wants to make images accessible than HTML 4 validators checking merely for alt presence do. (The feature is also pretty discoverable for people who don't yet know that they will want to use it.) > And I don't expect non-visual user agents to implement such image > analysis heuristics just because spec authors think they'll do it. In general, investments in software don't just happen because spec authors said so. This an economic problem: Does the benefit of implementable image analysis (and indeed *some* analysis is implementable given what kind of signal processing is known today) justify the opportunity cost? That is, do AT vendors get a bigger win by doing something else instead? Fortunately, image analysis code would likely be relatively separate from the other code in a screen reader, so even if AT vendors didn't have economic business incentives to ever pull this off, there's opportunity for outside economic disruption by contributing code to NVDA or Orca without having to tie up core developer time too much. (I don't think the code size is a big deal these days.) -- Henri Sivonen hsivonen@iki.fi http://hsivonen.iki.fi/
Received on Saturday, 3 May 2008 10:09:30 UTC