– DRAFT –
Accessible Platform Architectures Working Group Teleconference

Synchronization Accessibility User Requirements.

JW: Asks about publishing and review periods

We have the SAUR ready for a descision - about whether we publish a working draft or make it ready for wide review.

If you are not keen on publishing let us know

+1 Scott and Steve

+1 from Janina - also to wide public review

SH: Acks Steves great work

JOC: +1 from me

JS: Is the 6 weeks necessary

JW: We did propose 6 weeks review time but can be adjusted

JS: <Discusses messaging etc>

Earlier in May for publishing would be good.

SN: I'm fine with that

JS: Lets do the 13th?

JW: What would we like?

SH: Mentions Easter Holidays globally..

RESOLUTION: RQTF will publish SAUR https://www.w3.org/TR/saur/ for a second public review with comments due midnight Boston time Fri 13th May

JS: Will flag this to Roy, that we have permission to update public working drafts

Lets update the drafts and get messaging ready to roll - Shawn will see the update working draft

JS: Hands off Github for now

JW: Any final comments?

Accessibility of Remote Meetings.

JW: I'm the one with the report here..

I've opened some issues after a review - had some discussion with Scott. Have created some PRs.

Scott has reviewed, thanks.

JW: <Gives overview of PRs>

Mostly editorial but some may be more substantive

SH: Great work Jason

Happy to have that input on testing video

JOC: Here are the PRs https://github.com/w3c/remote-meetings/pulls

JW: Any other comments?

<GH banter>

agenda

JW: Had Judy been asked to review?

SH: Yes

Review of draft ethical machine learning consultation document.

JW: I've some comments on this..

The link to the latest draft in agenda

There are meetings next week - with two sessions run by Machine Learning group

APA will be attending a session - I'll do another. Am meeting with MattTA to review his draft

If anyone has input here, we can feedback to them

JS: Anyone can attend Monday April 4th, 6am Eastern, and April 5th at 16:00 UTC

JW: Any other comments? Or we can take our thoughts to those meetings.

Natural Language Interface Accessibility User Requirements.

JW: Once we get some of the other publications under control we can get this to the top of the agenda.

If there have been updates let us know..

JOC: Has no subterranean updates

JW: Its good to have the walkthru of the COGA issues done.

JW: We can re-engage with MITRE also.

JW: Lets discuss next week.

JW: Final words?

Inaccessibility of CAPTCHA.

JW: There was a meeting with Cloudflare last week?

LW: We had a good meeting with Cloudflare - they are commited to helping with our agenda.

CAP is the technology that we see as a replacement for CAPTCHA..

LW: They mentioned privacy pass - they are interested in unblocking traffic and reducing friction

As well as Zero knowledge extentions - support privacy and get user agents to support

Tara Whelan is involved in privacy and security

LW: I've been reading up on standards etc

Janina is to reach out to WebAuthen - to also cover attestation

We dont want to go too far with only one companies stack.

JS: Thats important

JS: There are terms that are not in our document that we should track

There is privacy and security jargon that we should mirror

Important for implementors to see

JW: We could cite relevant issues.

JS: We may have a new layer on historical or legacy tech that we dont recommend are used.

JS: I will send a new reference to the list

LW: I can add to the minutes

LW: <talks about CSUN experience with Sign Language>

Miscellaneous topics.

JW: I want to say that the maturity model discussion is continuing, the co-chairs of APA aims to bring this into that group.

Language identifiers

<Lionel_Wolberger> Lionel adding to the minutes some more details on the discussion with CloudFlare:

JS: In pronunciation, it is a good research question - would like validation

<Lionel_Wolberger> While currently CAP relies on Webauthn attestation, CloudFlare is proposing a ZK-based extension to Webauthn that will not rely on Webauthn attestation

<Lionel_Wolberger> (ZK = zero knowledge)

<Lionel_Wolberger> - The IETF Privacy Attestation Tokens (https://www.w3.org/TR/turingtest/) demonstrates that a task has been completed, or that you have met a given challenge. A quote from the spec, "token authentication can be used in similar ways to CAPTCHA validation today, but without the need for user interaction"

JS: It seems BC47 spec relates to an IETF doc RFC 5646

<Lionel_Wolberger> Tara who is on W3C privacy WG, will read, comment, edit the CAPTCHA W3C paper

<Lionel_Wolberger> Lionel to read PrivacyPass (https://privacypass.github.io/) IETF Privacy Attestation Tokens (https://www.w3.org/TR/turingtest/)

Regarding ancient languages.. some are to be supported by extensions

<Lionel_Wolberger> Thibault (from CloudFlare) to advise on what other companies/corporations could implement CAP. RQTF and W3C is always very pleased to see multiple companies involved.

But if we want to specify a language there are different issues..

Variations in regional pronunciation etc

JS: We need some time based way to specify where we have rules to distinuish English over different time periods.

Same with ancient languages - vernacular etc

Is this accurate? That our language identifiers dont support language variation in historical contexts.

JS: Would like us to look at adding time/context considerations into these languages.

JW: Thats clear.

JOC: Good catch, time is another axis.

JW: I'll have a look - lets discuss more.

JS: Something useful for TPAC.

JW: I've got some insight into a related group and can share.

<SteveNoble> one resource that mentions some ancient language codes: https://sites.psu.edu/symbolcodes/web/langtag/#langcode