Meeting minutes
identifying conditional states&properties in characteristics tables
<ZoeBijl> show agenda
pkra: I filed this from Sarah's observation and questions from validator work. It's probably just relevant for ARIA but I just wanted to note it here in case it effects other specs
cyns: maybe WPTs can help. They have plenty of conditionals.
pkra: good point. Hopefully a proper markup should help clarify what's tested (and possibly learn from WPT what we might consider in the spec).
WIP: Add CSS-AAM editor’s draft
pkra: yay, we have a spec! And editors!
ZoeBijl: this is the first step. Coming out of tpac, hoping to get back into it. First to document. Hopefully to make this part of the CSS process, that CSSWG contacts to make sure updates are made for CSS changes.
… Jamie has also started to make edits, too.
… silly first question: editor config. Should it align with aria settings?
jamesn: we run prettier anyway, so it doesn't matter too much.
Daniel: but we only really do printwidth.
ZoeBijl: right. I thought it would pick up on editor config.
… just want to align.
Daniel: should be fine either way.
ZoeBijl: great, I'll check with Jamie.
Daniel: did you already add something substantially new?
… and we should also check the conformance section to align with the other (recently aligned) specs.
ZoeBijl: any pointers?
Daniel: just copy one of the others.
ZoeBijl: will do.
… regarding edits, so far I purged the old repository of PRs and issues. I made 2 PRs to clean up main and gh-pages branches.
… then jamie made some recent suggestions. I'll review those tomorrow.
pkra: I should look into that roleinfo check failing.
… I think it just needs a prettier addition. I'll file a bug for myself.
Kicking off SVG-AAM work
cyns: we haven't done a PR yet but was hoping to look into what people are expecting or hoping for.
… this seems to not have been touched since 2021.
Daniel: I merged a PR for some of the broken references so at least it should run through respec now.
cyns: is there some guidance doc?
Daniel: there is a little bit in the main repo readme file.
cyns: thanks. I'll start with that.
markrogers: one of the problems with the svg2 spec seems to be that browser haven't really implemented it.
<ZoeBijl> https://
markrogers: There was a disconnect between spec editors and implementors. Trying to work out what's implemented seems like the first important step.
… for the svg element there's been some disagreement on which role to expose.
cyns: I recall that it's tricky to map since platforms don't have a lot of graphics stuff.
… I don't know how much graphics-aria is implemented and how it helps (or not).
markrogers; I think some of it is and that's in part what leads to differences on e.g. svg role.
cyns: I don't know how much OSs can or will do something.
… once we know what's implemented, we can clean up the document and possibly move it forward.
Daniel: re graphics-aria if guidance is in different documents, we might need to go back to those.
cyns: does anyone know if it is implemented?
jamesn: I don't think it does anything useful.
<cyns> https://
pkra: I think section 5 with the inclusion/exclusion is a bit of a mess. If that could be clarified from implementations, that would help a lot, esp. for testing tools. There are also a few open issues from jcraig which might worth attention. Also, browser+AT behavior is sometimes "surprising".
markrogers: SVG is just a very large spec. Working out what if shapes could be exposed might be worthwhile.
cyns: we should look at good use cases like charts.
pkra: every other year someone brings up a data-viz mode but that seems separate. There's also regular interest in flowcharts but those always run into circles / focus traps. The last time there was some discussion around aria-flowto and the old SVG connectors(?) proposal.
cyns: ok. let's start by clean up and then take it from there.