New comments received on the CT guidelines

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

New comments received on the CT guidelines

Francois Daoust
Hi,

We have received 3 more comments on the Guidelines for Web Content
Transformation Proxies from the Protocols and Formats Working Group
(PFWG), see mail on public-bpwg-comments:
http://lists.w3.org/Archives/Public/public-bpwg-comments/2010JanMar/0000.html

I have created LC-2358, LC-2359 and LC2360 in our tracking system as a
consequence.


LC-2358
-----
The comment suggests changing the title of the spec to make it clear
that we are talking about mobile:
http://www.w3.org/2006/02/lc-comments-tracker/37584/WD-ct-guidelines-20091006/2358

Wait a minute, I vaguely remember we had extensive discussions about the
title in the past ;)



LC-2359
-----
The comment wonders about the impact of "Cache-Control: no-transform" in
a proxy-based accessibility transcoding solution:
http://www.w3.org/2006/02/lc-comments-tracker/37584/WD-ct-guidelines-20091006/2359

Two notes on the "Cache-Control: no-transform":
- it is defined in the HTTP RFC. We advertise it in the guidelines
because it was neither well-known among content providers nor respected
by content transcoding proxies, but we do not alter its meaning. In
other words, an HTTP proxy should already respect this directive.
- I think we realize that it is a "heavy" switch, but we cannot invent a
switch that would better suit our purpose, e.g to narrow the scope of
the no-transform to "no desktop-to-mobile transcoding".

One way to address the comment could be to add a note to call out that
"Cache-Control: no-transform" is not limited to desktop-to-mobile
transcoding: it also prevents desktop-to-accessible,
mobile-to-accessible (and actually any type of transcoding), and should
therefore be used with great care.


LC-2360
-----
The comment says that it is sometimes preferable to introduce new
elements and attributes that do not belong to the DTD, such as ARIA
attributes in the case of accessibility:
http://www.w3.org/2006/02/lc-comments-tracker/37584/WD-ct-guidelines-20091006/2360

I think that is precisely the reason why we put grammar validation at a
SHOULD level in the first place.

Thanks,
Francois.