W3C

– DRAFT –
PDF-AAM Meeting

16 July 2025

Attendees

Present
Daniel, jcraig, Paul, Zakk, Duff
Regrets
-
Chair
Daniel
Scribe
Daniel

Meeting minutes

First import of PDF-AAM mappings https://github.com/w3c/aria/pull/2466/

Zakk: Thanks for your commits Daniel
… I mentioned a couple of lines we could look at

<Zakk> lines: 264, 278, 361, and the table starting on 396

Line 400 -- label role

Zakk: We wanted to try this one out as there is not a clear correspondence
… We may need to create a new pDF row
… How could we provide guidance?

JamesC: In this case you would point to the HTML-AAM, you could just say "use the HTML mappings"

Daniel: See my comment

Daniel: That's how this should look like

JamesC: You don't need to map a pdf- role. If you did, we should probably decide if we need to put it somewhere in the dpub-aria
… HTML has one and I am not sure if it predates this

Duff: I was reading a GitHub discussion about this. But we should cover other use cases of the `lbl` element that don't have a good paralel on the HTML element

Paul: We have both of them taken into account

JamesC: In HTML you have `a` element with and without and `href`. That could be the approach

<jcraig> https://w3c.github.io/html-aam/#el-label

Zakk: I just need to change this

Zakk: What do we want to put on line 412 for the non-HTML elements?

JamesC: in this case it would be `<code>label</code> element`
… For the others, use HTML-AAM mappings

JamesC: We sshould probably leave the note

JamesC: If no link, we should probably remove the class

Duff: We have use WAI ARIA mappings. Should we use PDF mappings?

JamesC: "In the context of forms" is a bit tricky. We should be more explicit. I think in HTML it calls out the relationship
… Is there something that the PDF author should take into account

Zakk: Label as a child of a form element?

Duff: You have to give the correct semantics for it

JamesC: Child of, group with, or maybe a tagging relationship. Maybe in this case the parenthetical can be replaced with "label which is associated with a form element"

Paul: that makes sense and plays well with the other ccontexts where lbl is used outside forms

Zakk: This is role label forms, and the other can be role label list

Paul: Label can be used for headings as well, not just lists

Zakk: Yes, I was just thinking from an organizational perspective, we can get as specific or as general as we like

JamesC: Anything that requires a different mapping needs to be called out separately: form elements, lists, headings, and maybe a catch all

Duff: There are three: forms, lists, and content. Content could be headings, footnotes, etc

Zakk: There is also the generic which maps to generic

JamesC: I think you want to add the parenthetical outside the `code` block: "label other than in the context of lists of forms"

Paul: We haven't broken out labels that match

Zakks: List numbering or the other various ones we are going to have to provide guidance for some of the unique labeling.

Paul: If they are going to be generic do we need to separate them out?

Zakk: No. We only need an additional mapping if the platform mapping is different

Paul: So I would say forms and out of forms for now

Duff: Don't we want to clarify that there is additional consideration for labels that may not map to generic?

JamesC: We could include it in a note

Zakk: Ideally is to have these eleven entries ready for the Editor's Draft

Duff: It's in scope and may create misunderstandings if not addressed appropriately

Daniel: I think priority for us is to have the import script working. We could clarify the situation on the status of this document so that it's clear upfront for everybody

Wrap up

Zakk: Let me know what we need to do to get this merged

Daniel: I'll re-review these last additions, then ask Valerie for review and if she's happy we can merge

Minutes manually created (not a transcript), formatted by scribe.perl version 244 (Thu Feb 27 01:23:09 2025 UTC).

Diagnostics

Failed: s/..s/...

Succeeded: s/412/412 for the non-HTML elements/

Succeeded: s/We need to pull out the notes probably/We sshould probably leave the note/

Failed: s/<Zakk> present//

Maybe present: Duff, JamesC, Zakks

All speakers: Daniel, Duff, JamesC, Paul, Zakk, Zakks

Active on IRC: Daniel, jcraig, Paul, Zakk