Meeting minutes
Accessibility (A11y) of Mermaid diagrams
Ivan Herman: it's not the mermaid, its the specification that we write that uses mermaid
… what I'm worried about is problems because we don't have alt descriptions of those diagrams, which makes our documents inaccessible
… So, how might we work on that? I don't have a solution. but it would be good to have a common approach
… Yesterday, Mermaid has introduced title and desc that gets promoted into the image
<Kevin Dean> https://
Ivan Herman: I raised an issue that we could use this info to put it in proper context
Phil Archer: do we know if Mermaid input documents are a list of instructions. Can that set of instruction be "accessible"
Ivan Herman: I spoke with an expert yesterday and she said that would not be enough
… it's not very readable by itself
… but maybe putting a description tag into the diagram, that propagates upward
<Michael Shea> JoeAndrieu: ... all Mermaid diagrams are their own bespoke language. No semantics that are common.
Phil Archer: clearly this is a real issue. it must be faced by others
… where do we raise this?
Ivan Herman: maybe we first contact the group that does the reviews and bring up the issues ahead of time.
… we aren't the only working group creating mermaid
… This comes up more and more with AI producing mermaid code too
<Kayode Ezike> Did we not use Mermaid in earlier versions of our specs?
VCDM 2.1 PRs https://github.com/w3c/vc-data-model/pulls
Phil Archer: There are currently nine open PRs on VCDM
… with Manu not here, I wonder how much progress we can make. But I'd like to take a look at some of these.
… Does anyone know who Astro Gilder is?
[crickets]
… This might be a bot rather than a person
… Editors will take a look
… next is one from JinDX
… Anyone know this person? Were they at TPAC?
PR 1663 w3c/vc-data-model#1663
PR 1664 w3c/vc-data-model#1664
Ivan Herman: He submitted over 100 PRs in the last few weeks in many WGs.
… Turns out to be a real person, likely helped by AI, somewhere in China. And one of our W3C colleagues have talked to him
<Kayode Ezike> Yeah, we’ve received a few PRs from JinDX as well.
Ivan Herman: He is experimenting with using AI to translate documents
<Kayode Ezike> in the VCALM spec
Ivan Herman: He's translating all the W3C specs into Japanese, Korean, and Chinese.
… During this translation, his machine finds a bunch of small editorial problems.
… I have the dubious privilege of going through and making sure there are no IPR issues
… They have generally been harmless, useful, and should be merged.
… I have asked him to batch edits together to help
Phil Archer: sounds like a good thing to me if there's a way to check if our text is easily translated
Wesley Smith: +1 to robots helping us. During our task force meeting, is a normative change automatically not appropriate from an IPR perspective?
Ivan Herman: depends on the change
… so far, it was mostly spelling mistakes and syntax mistakes in JSON, even if those changes are in normative sections, the question is whether or not the change itself creates IPR problems
Wesley Smith: so a human makes a subjective call
Ivan Herman: yes. I push the button at the end of the day
PR 1665 w3c/vc-data-model#1665
Ivan Herman: we put in a new version of the vocabulary for 1.1, that's already been merged. But that also means the github workflows have to be changes to generate things correctly.
… I spotted a problem in the last round of that. Some who understands the gitflow should take a look, please.
Phil Archer: there are review requests
Ivan Herman: that was just because of CODEOWNERS
… but I'd like Manu and Dave Lehn to take a look.
Phil Archer: It's been there a few weeks, but let's get a look at it
… a couple others that need the editor to chime in
Phil Archer: redirections for 2.1?
Ivan Herman: same story as before. we need to set up redirects for context file and vocabulary.
… I set up a proposal to do that. Manu gave feedback and I updated, but I need a review since I'm not confident enough in what I do.
Phil Archer: looking at the labels on outstanding issues (#1619)
Issue 1619 w3c/vc-data-model#1619
Phil Archer: This looks like a stale issue that should be closed.
… It says "PR Exists"
… Dmitri created a PR. So let's check in with Dmitri
… Can anyone take a quick look at PRs and see if there's anything of value today with our attendees?
… Only one more topic for today.
Meeting schedule
Phil Archer: The meeting schedule for the rest of the year.
… This time next week, America is on holiday.
… Do we want a meeting this time next week?
… I'm guessing people on the West of the Atlantic won't be keen.
<Michael Shea> they'll all be sitting in airports on Wed
<James Richards> I will be working
Phil Archer: seems like general support for cancelling
… So, let's cancel.
<James Richards> I miss the UK
Phil Archer: Last working day for me is 18th December
… so unless anyone objects, I'd like to cancel the Dec 23rd and Dec 30th meetings
<Michael Shea> +1 to cancel
Phil Archer: ok. Cancelled those two around year end
… nothing else queued up.
… Anything else to discuss?
… Next time we do meet, Wed Dec 2nd
… Hopefully Brent can chair, I will not be available
… Head of scribe queue is either Michael or Jennie, in two weeks time. Hopefully one of you will be here at the start to help out.
… Thanks, all. Have a good Thursday next week!