Meeting minutes
Agenda Review & Announcements
<matatk> ICCHP coming up later this month: https://
janina: But no remote participation, papers not yet published either
TPAC 2026
<matatk> Apply for funding! https://
New charters review
<Fazio6> +1 TPAC is an awesome experience
<Fazio6> A Nerd Utopia
ATAG
<matatk> w3c/
Group is supportive overall; would like to know the expected date for ATAG-next, but not a blocker.
New on TR
WebMCP - should it be adopted as W3C work?
Email for context: https://
<Fazio> via prompt to the user?
Good RQTF discussion today logged here: https://
<Fazio> no different than latover dilemma
<Fazio> layover
<Fazio> +1 to excitement
<Crispy> matatk: Interested in adapting the interface to the user. This is a huge architectural overhall. Concerned about potential misuse and issues for users with AT. Maybe only for devs? What is the risk to the A11y tree?
Dr_Keith's view: https://
<Fazio> Dr_Keith email was very well worded
<Crispy> PaulG: AI adopters possibly to get special access? Thinks we should take this on.
<Fazio> why does it need to be chartered?
<Crispy> matatk: Question is that this spec as is, do we adopt it?
<Crispy> PaulG: What are the options for us?
<Crispy> matatk: Adopt and it requires review, etc.
<Fazio> do all specs need to be chartered specifically?
<Crispy> Not adopt as it is, propose a different way.
<Fazio> If it is chartered it can still end up being deemed inappropriate in the end
<Crispy> They want to atleast investigate
<Fazio> W3C needs excitement
<Fazio> This is exciting
<Crispy> PaulG: Concerned without eyes on it may get abused
<Crispy> Browser vendors may use this or partner with AT companies. It should be ubiquitous, yet private, and secure
<Fazio> All reasons to charter and police it
<Crispy> matatk: Mentioned that there are position statements on this. Google are in favour, Mozilla said nothing to date, Safari object strongly.
<Fazio> charter for exploratory review
<Crispy> Inclined to have discussion at least, but where and who. Not enough info for core group to handle this?
<Fazio> TPAC meeting on Wev MCP?
<Fazio> Web
<Crispy> Needs W3C involvement, but in what format?
<Fazio> Do we have a say in what WG takes it up?
<Crispy> matatk They already have the bandwidth and are already looking into it
<Fazio> charter us deep into the cross review
<Crispy> matatk: tremendous pressure to do this. Currently great risk to move this to recommendation.
<Crispy> janina: If they won't listen to the broader public maybe they aren't the right group to take it forward.
<Crispy> Fazio: Not too concerned with objections, but we should support the exploration
<Crispy> Chartering isn't endorsing, but must be explored
<Crispy> matatk: possible to charter exploring an issue we're not sure how to solve yet
<Crispy> David gave an example of this being done elsewhere
<Crispy> matatk: This isn't an example of exploratory work. They have created the spec, no more research is due to be done. What they propose has too many issues from an A11y and security perspective. We need to clarify what we need to see happen.
<Crispy> David agrees
<Fazio> I support Dr_Keith's email
<Crispy> Dr_Keith: agrees we should support it but we need to involve our comments as not fit for purpose as it stands
Crispy: I've gone through the emails; I think Leonie had some good input in the threads; I agree it is interesting but certainly from an accessibility point of view needs work.
<Crispy> PaulG: WEbMCP can be helpful, gave a workflow example, highlights loads of other areas it can help. Get it to help the larger audience. Doesn't want to shelve it.
Issue tracking
Open UI ::interest-button
relevant bit of explainer: https://
<Crispy> matatk: Interest invokers - Explained the scenario for a pop-up on hover with stats on a person, press on link, go to full page user profile. Similarly, link on Github,gives info on pop-up, saves moving away from current place. Wiki does the same with references. This pattern to not follow a link to a page is quite common, trying to standardise
<Crispy> this. A11y management will all just work, but there are challenges. 1. Desktop with mouse works, on mobile hover might not be available. How to implement it? 2. Immersive platforms like Vision OS and Android XR - Look at link for a while and pop up may appear. Privacy issue on where the users are looking.
<Crispy> One technique - provide button - look at link and button pop asks if interested and user selects the button.
<Crispy> Ways around this. Button maybe provided by the user agent
<Crispy> How should this be styled by default? Opt in or Opt out by a media query?
<Crispy> PaulG: Concerns for User agent controlling that button, etc, and the default. Not enough modality use considered, we need to look at more.
Please comment with use cases/Qs/concerns on w3c/
<Crispy> matatk: Please comment by next week and add use cases if you have any.
<Crispy> We can illustrate to the group what they may not have considered.