Re: 48-Hour Call for Consensus (CfC): Publish "ARIA D-Pub Roles" FPWD

Rich Schwerdtfeger

James Craig <jcraig@apple.com> wrote on 07/01/2015 08:15:15 PM:

> From: James Craig <jcraig@apple.com>
> To: Richard Schwerdtfeger/Austin/IBM@IBMUS
> Cc: janina@rednote.net, W3C WAI Protocols & Formats <public-
> pfwg@w3.org>, W3C PF - DPUB Joint Task Force <public-dpub-aria@w3.org>
> Date: 07/01/2015 08:16 PM
> Subject: Re: 48-Hour Call for Consensus (CfC): Publish "ARIA D-Pub Roles"
FPWD
>
>
> > On Jun 30, 2015, at 1:17 PM, Richard Schwerdtfeger
> <schwer@us.ibm.com> wrote:
> >
> > I can respond to the following now:
> >
> > dpub-toc: directory maps to a list in all the platforms.
>
> If you think that's a problem, address it via vendor ERs or in the
> Core AAM, but documents should still use "directory" role instead of
> a redundant "toc" role.
>
I am not convinced. We map it to lists to lists and directories to lists.
What if the table of contents is not build like a list? What if it is built
like a Tree? This is too restrictive. They are thinking of toc as a region
versus a structural object.

> > One would expect that lists can appear throughout a book. If I
> wanted to go to THE table of contents it would not work.
>
> Sounds like a problem with your book reader. Which one are you
> using? Navigation to the table of contents if standard in every
> reader I've ever used. It's standardized in other ways that do not
> require role overlap.
>
> > There is no distinction. I could see us subclassing directory however.
> > mapping: http://rawgit.com/w3c/aria/master/core-aam/core-
> aam.html#role-map-directory
> >
> > dpub-part: is a name, apparently well known in the publication
> space. I will let Tzviya respond to that.
>
> "dpub-chaptergroup" might be a better name.
>
The the task for on this will want to respond formally but the problem I
see arising is that we are trying to redefine terms commonly known by the
publishing industry. My concern is that we put too tight a vice-grip on
things that we cause people to not use the technology. I think these are
good suggestions but we need to recognize that they have a terminology they
are accustomed to using. If we define new words for that it will lead to
confusion by book authors. They are going to use these roles for more than
AT interoperability. They are going to be exposed to book publishers in
authoring tools.

Let's see what they come back with. They did put this in their vocab space
so at least we are not going to see conflicts.



> > bibliography is not abbreviated.
>
> :-/
>
> > However, we can look at the abbreviations. I can see where
> changing qna to questionsandanswers may get pushback as q and a are
> pretty well understood.
>
> It's still unclear why an extra role is needed. This is just a <dl>.
> or a "group" that surrounds a <dl>. The clarification if how this
> group is unique would be made apparent by the group's label. e.g.
> aria-labelledby="qnaheader"
>
> If there is an extra role, it should be to include the 1:1 mapping
> to HTML5: "descriptionlist", perhaps.
>
> > Rich
> >
> > Rich Schwerdtfeger
> >
> > From:   James Craig <jcraig@apple.com>
> > Date:   06/25/2015 02:25 PM
> >
> >> I support the FPWD with these comments/issues.
> >>
> >> 1. Please link all the ISSUES notes to their respective issues.
> >>
> >> 2. dpub-toc is unnecessary. Use the general ARIA 1.0 role
> "directory" whose description is "Authors should use this role for a
> static table of contents, whether linked or unlinked. This includes
> tables of contents built with lists, including nested lists."
> >>
> >> 3. Role name abbreviation is still very inconsistent.
> "bibliography", "biblioref", "biblioterm", "qna"... Please note that
> ARIA does not abbreviate role names (except for one oversight, "img")
> >>
> >> 4. "dpub-part" is still too generic a name, and the previously
> filed issue/bug is not mentioned.
> >>
> >> 5. "dpub-noteref" and "dpub-glossref" are unnecessary. These are
> simple links. There is no need for another role.
> >>
> >> Most of this and other feedback was previously provided but does
> not appear to be addressed or mentioned.
> >>
> >> James
> >>
> >>
> >>> On Jun 24, 2015, at 8:44 AM, janina@rednote.net wrote:
> >>>
> >>> Colleagues:
> >>>
> >>> This is a Call for Consensus (CfC) to the Protocols and Formats
Working
> >>> Group (PFWG) to approve a First Public Working Draft (FPWD)
publication
> >>> of "WAI-ARIA Digital Publishing Module, Version 1.0" as requested by
the
> >>> ARIA-Dpub Subteam at:
> >>>
> >>> http://lists.w3.org/Archives/Public/public-pfwg/2015May/0027.html
> >
> >
>