- From: Gez Lemon <gez.lemon@gmail.com>
- Date: Tue, 19 Aug 2008 01:38:13 +0100
- To: "Patrick H. Lauke" <redux@splintered.co.uk>
- Cc: wai-xtech@w3.org, public-html@w3.org
Hi Patrick, > The problem, it seems, is that the pain (failed validation?) won't > automatically result in the correct behaviour (providing relevant @alt), but > only in the minimum effort required to make the pain go away (putting > *anything* into @alt, even adding a null @alt, just to get the thumbs-up > from the validator). That's a good point, but it would be tragic if conformance requirements were lowered to satisfy validators. Maybe the conversation should be based around conformance, rather than validation. Validators are easily fooled, and when they're the sole method of quality assurance, aiming to please validators is likely to result in a poorer experience - particularly from an accessibility viewpoint, as the nature of ensuring content is suitable for humans means that very little can be automated. If you do bulk uploads on Flickr intended for your friends and loved ones, it's reasonable that you might decide to add text alternatives later, or maybe never get around to adding them. But do you think the resulting content should be considered compliant? Without text alternatives, the content will not be perceivable for some people. It seems reasonable to me that appropriate alternative text is required for conformance. Lowering conformance requirements seems a dangerous route, as not only does that lock out people that cannot readily change an aspect about themselves to be able to perceive the content, but it's also a route that is open to abuse. Why is it so important that inaccessible content should be considered compliant? Why not allow these edge cases to be considered non-compliant, and have authoring tools encourage authors to author accessible content? If an author chooses not to provide text alternatives because they're writing for themselves, close friends and relatives, that's fine; but lowering conformance requirements in order to declare a structure that is incomplete as compliant doesn't seem reasonable. Cheers, Gez -- _____________________________ Supplement your vitamins http://juicystudio.com
Received on Tuesday, 19 August 2008 00:38:50 UTC