Re: Proposal for <canvas src> to allow images with structured fallback by Tab Atkins Jr.

I've lost the original context, but the argument that linkrot
will occur just because sighted authors will not actively see
some portion of HTML is specious.  Content management systems
don't need eyes in order to check links.  Maintenance robots
don't need eyes in order to check links.  And, perhaps most
importantly, non-sighted users of the Web don't need eyes in
order to test the accessibility of important websites.

An organization truly concerned about Web accessibility,
including the ones that build sites for information that
is government-mandated to be accessible to the greatest
extent possible, will have both automated and human
procedures in place to make sure that is so (or will soon
find themselves on the losing end of a class action lawsuit).

Our job is to provide the means to make HTML accessible to
the general public by those who have a will to make it so.
Removing features that are used to improve accessibility,
without providing a suitable alternative that covers the
same use cases, is wrong.  Providing new features that may
be used to convey important information without also providing
a mechanism to make that information accessible is wrong.

Arguing that some people might use the feature ineffectively
is no more sensible than arguing that we can't standardize
the script element because many authors can't program
correctly in javascript.

Cheers,

Roy T. Fielding                     <http://roy.gbiv.com/>
Principal Scientist, Adobe Systems  <http://adobe.com/enterprise>

Received on Wednesday, 9 March 2011 02:09:43 UTC