Add doc-pageheader, doc-pagefooter

[aleventhal]

Closes #10

Also deals with some of the use cases in w3c/aria#1044

---

Preview | Diff

[aleventhal]

Also seeking reviews from @mcking65 @carmacleod @cookiecrook

[carmacleod]

@aleventhal Looks good so far! Just a few little things to fix up. See comments.

@jnurthen dpub-aria roles that inherit from section or landmark (maybe others) are missing "(deprecated on this role in ARIA 1.2)" on aria-disabled, aria-errormessage, aria-haspopup, and aria-invalid. Do you know what needs to be done to fix this?

[Edit: Opened #31 for flagging the 4 deprecated aria-* global attributes]

[aleventhal]

Thanks @carmacleod and @mattgarrish, I've incorporated your suggestions. Please take a look.

[TzviyaSiegman]

Matt already added most of the comments I was going to make. I think it would also be worth explaining that this is for circumstances when contentinfo (which is how I treat running headers now) is insufficient)

[carmacleod]

Almost +1, but I'll approve now. Thanks for all the fixes!

It just needs @mattgarrish or @TzviyaSiegman to comment on what element/role to use for the title/author content of a header/footer. Right now, it's just div (no role) which is fine with me, but I don't know if there may be something better.

[aleventhal]

Matt already added most of the comments I was going to make. I think it would also be worth explaining that this is for circumstances when contentinfo (which is how I treat running headers now) is insufficient)

Thanks Tzviya. Two problems with contentinfo:

1. It will add lots of extra items to the landmark navigation cycle in a screen reader. If there are 500 pages, this is a lot of landmarks!
2. It doesn't indicate that the content can be optionally skipped over while reading

I would avoid using contentinfo for these. Because of #1, it would be better to use group or section, until doc-pageheader/doc-pagefooter are available.

Where did you want us to add this info to the PR.

[mattgarrish]

on what element/role to use for the title/author content of a header/footer

I don't think you'd want to identify these in a running header/footer. I would think it would have the effect of repeating this information on every (virtual) page of the document. As the information may be truncated, it's not always equatable with the formal author/title designation.

We were also advised against using ARIA for general metadata when we did the 1.0 release. Once we open the door to author and title tagging, there's a never-ending queue of others that we may get requested (publishers, imprints, illustrators, tranlators, etc., etc., etc.). There are other technologies to support this information, too (package metadata in EPUB, jsonld/RDFa/microdata in HTML).

[aleventhal]

As such, I think using a plain <div> is fine for the author or title.

[aleventhal]

I'm thinking we should get rid of doc-pagenum.

• It's already possible to specify the page number as the name of a doc-pagebreak
• In the case of there only being a page number in the repeated content, the author can still wrap it in a doc-pageheader or doc-pagefooer

[aleventhal]

I'm thinking we should get rid of doc-pagenum.

• It's already possible to specify the page number as the name of a doc-pagebreak so removing doc-pagenum avoids a confusing redundancy
• In the case of there only being a page number in the repeated content, the author can still wrap it in a doc-pageheader or doc-pagefooer

@mattgarrish WDYT?

[jongund]

@aleventhal
Looks good to me.

[sinabahram]

This seems quite helpful. Thank you @aleventhal for your work on this.

[aleventhal]

Thanks for these email +1s. Is the GitHub UI letting you mark it as +1 in the interface for the PR? These reviews aren't showing up there. You may need to add yourself as a reviewer.

…

On Fri, Oct 2, 2020 at 2:49 PM Sina Bahram ***@***.***> wrote: This seems quite helpful. Thank you @aleventhal <https://github.com/aleventhal> for your work on this. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#28 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKQAZRAFDVVM3MORTXUVCTSIYOELANCNFSM4RYZPTPA> .

[sinabahram]

@aleventhal not sure I can add myself as a reviewer. Looking to see if I have permission to do so.

[aleventhal]

@mattgarrish @michael-n-cooper we got +1's in the email list from @sinabahram and @jongund, but they couldn't be added as reviewers, presumably because they aren't in the project.