Meeting minutes
<Daniel> agenda: 376853
<spectranaut_> w3c/
w3c/aria#2744
spectranaut_: I wrote a script to extract data from the tables.
… the relevant part is generating the mappings from the json file.
… my motivation was that it might be easier to edit the json file.
… it's still awkward due to having a second step.
jamesn: why is it easier to edit json?
Daniel: it's easier to have fewer text ni context.
… but I suppose yaml is even better.
jamesn: wouldn't be escaping make things difficult?
spectranaut_: good point. Maybe yaml would be better.
jamesn: I'm skeptical if that benefits editing
spectranaut_: right. another motivation was that Scott wanted a source of truth from core-aam to feed into html-aam.
jamesn: right, that makes sense. That could be a file in the build step to create a json fle.
… the spec is never going to be just json, no?
… if we split it up, this sounds like a step backward.
spectranaut_: feels like the HTML has more room to make mistakes.
… already a lot of inconsistencies in the tables that would need fixing up.
jamesn: fair. It's not my spec after all.
spectranaut_: right. thanks for the feedback. I'm hoping for more
jamesn: something like roleInfo seems great.
spectranaut_: then maybe cleaning up the table is the key thing
pkra: we also have inconsistencies in roleInfo. E.g. conditional statements in the characteristics tables.
spectranaut_: good to know. I'm still not sure if it's worth the effort.
… needs more thinking, and input.
… besides Scott's request, there's tests. I'm not sure yet how feasible that is.
revisit: tooling not strict https://github.com/w3c/aria/issues/1877
pkra: I was going through old issues. This seems like worth another look
… the particular case was solved back then.
Daniel: the larger issue remains. If we turn on the linter, we'll get a lot of noise.
… but those would show up on TR
… we could also decide which kinds of PRs we run this on so as to limit the noise.
… I'll try a couple.
revisit: replacing IDREFs with modernized language https://github.com/w3c/aria/issues/2556
pkra: I unassigned myself from this since the scale is very large
… essentially all specs need to be checked.
spectranaut_: definitely not a first issue but a deep dive.
jamesn: not sure what to do this.
spectranaut_: sounds like medium issue.
jamesn: not sure. It sounds like someone would need it.
… to know the spec well enough
spectranaut_: maybe it's primarily about finding the right language.
revisit: better dev tool support https://github.com/w3c/aria/issues/1962
pkra: I tried to re-use this old issue to see if we can talk about dev tools.
jamesn: let's open a new issue. This one has too much history.
pkra: will do.
machine readable source of truth for core-aam mapping tables https://github.com/w3c/core-aam/issues/258 agendabot]
revisit: role=none prohibited attributes https://github.com/w3c/aria/issues/1983
pkra: this is an old one that specifically asked for editors' meeting agenda but was never added.
jamesn: I think the original idea was to avoid that prohibiting means that the attributes are ignored rather than the role being removed.
… we should talk about it in a WG meeting.
stale PRs
pkra: a few stale PRs caught my eye
… github.com/w3c/aria/pull/2181
… I was surprised this hasn't been merged yet.
… I don't know what the status.
spectranaut_: should we ask Scott if someone else should take it on?
jamesn: does this need actual implementations?
pkra: w3c/
… this is a stale draft that didn't go anywhere. Can we close?
spectranaut_: yeah, there was no response.
pkra: ok. What about w3c/
jamesn: let's close it.
pkra: next w3c/
spectranaut_: I'll think about it.
pkra: great. Would be nice to find someone to take this over.