> On 27 Jul 2017, at 07:37, AUDRAIN LUC <LAUDRAIN@hachette-livre.fr <mailto:LAUDRAIN@hachette-livre.fr>> wrote:
>
> I like this idea Luc, but want to point out that we also need to keep in mind that there may also be metadata stating that a specific version of this publication may have been audited by a third party and “certified” to be accessible and now that this publication has been modified those sections changed may not comply with this certification. This does raise the point that if a publication that was certified to be accessible and hence modified that would nullify the certification and that metadata should be removed or modified in some way to reflect this change. This is probably something that the accessibility Task Force will need to take up.
Note that this isn't specific to publications, as the very same issue can happen with the general Web; but I agree it is particularly relevant to the publication space where we expect a11y statements to be discoverable.
On the Web, I believe the way to deal with it is with trust and with processes. Ideally, the owner of an accessible web site can be trusted to maintain an accessible process and/or to update her accessible information (e.g. VPAT) when the site changes.
In practice, I have no idea if this is working, or if this is a known deficiency in the Web. Any insight or data?
Romain.