W3C

- DRAFT -

Accessible Platform Architectures Working Group Teleconference

10 Apr 2019

Attendees

Present
janina, Joshue108_, Judy
Regrets
Chair
SV_MEETING_CHAIR
Scribe
Joshue108

Contents


Further discussion of issue #16 and issue #26.

<jasonjgw> Janina suggests these are serious issues regarding emerging protocols that are privacy preserving. After completing a CAPTCHA, the user receives tokens that can be 'spent" in subsequent interactions. Janina raises the question whether a biometric identifier could serve as an alternative starting point for token generation.

<jasonjgw> Janina further notes the withdrawal of audio CAPTCHAs, as they are becoming easily attacked by speech recognition tools.

<Joshue108_> Scribenick: Joshue108

<Joshue108_> JW: I agree those are the issues.

<Joshue108_> JW: I also think we should have something to say about these protocals.

<Joshue108_> JW: Will review again.

<Joshue108_> JW: need to be clear about solid alternative vs lighter solutions.

<Joshue108_> JS: I'm going to have a look at some low hanging fruit.

<Joshue108_> JS: Recapthca fallback etc..

<Joshue108_> JS: Some users will just go back to v2, if they dont like v3.

<Joshue108_> JS: We do need to say more about logic puzzles and video.

<Joshue108_> JW: Can we postpone that discussion?

<Joshue108_> JW: Lets move on, ok?

<Joshue108_> JS: Yup

CAPTCHA - any other updates or issues to discuss.

<jasonjgw> Discussion of #16 and #26 postponed to the next meeting.

<Joshue108_> JW: The logic puzzles idea.. we are not clear enough on this.

<Joshue108_> JW: We need to highlight the probs for people with cognitive issues.

<Joshue108_> JW: We don't want to imply that all are disadvantaged by this..

<Joshue108_> JW: But we want to highlight there is a prob.

<Joshue108_> JS: There should be different logic puzzle types..

<Joshue108_> JS: So a user may not be challenged inappropriately.

<Joshue108_> JW: I will look at text and work something out.

<Joshue108_> JW: Then take it to the mailing list..

<Joshue108_> JW: I just got your list..any stand outs?

<Joshue108_> JS: Video?

<Joshue108_> JS: Not sure what more we need to say about it.

<Joshue108_> JW: Half of Issue 4.

<Joshue108_> JW: Video game CAPTHCA..

<Joshue108_> JW: I think Scott may have found it..

<Joshue108_> JW: Need to review..

<Joshue108_> JS: Not sure if we added much too it..

<Joshue108_> JW: Some of the others look easily addressed, matter of detail.

<Joshue108_> JS: Lets look at 6 and 7 today.

<Joshue108_> JW: Issue number 6?

<janina> https://raw.githack.com/w3c/apa/editorial2-captcha/captcha/index.html

<Joshue108_> JS: Yup

<Joshue108_> JW: reviewing now.

<Joshue108_> JW: Interesting one..

<Joshue108_> JS: Is it a kind of logic puzzle?

<Joshue108_> JW: Could be..

<Joshue108_> JS: A non-sighted person wont solve it very well.

<Joshue108_> JW: Is it a jigsaw puzzle, or testing a drag and drop action?

<Joshue108_> JW: Looks like a paste two items on top of another?

<Joshue108_> JW: Ok

<Joshue108_> JW: The puzzle section could be expanded with issue 24?

<Joshue108_> JS: Yes, I suggest thinking of it like that.

<Joshue108_> JS: Like the folding on the dotted line things, so you can conceptualise the final thing.

<Joshue108_> JS: Visual excercise - but measures cognitive ability etc..

<Joshue108_> JW: Would be nice to have reference, rather than open issues but makes sense..

<Joshue108_> JW: We could say in absense of real examples, we shouldn't say much..

<Joshue108_> JS: I'm happy to keep looking for examples.

<Joshue108_> JW: Interesting, I support Janinas approach.

<Joshue108_> JS: Can I asign you number 7? The PKI?

<Joshue108_> JW: There is a lot here in this issue..

<Joshue108_> JW: How much of this do we want to address in the doc?

<Joshue108_> JS: Yeah, thats true for lots of these.

<Joshue108_> JW: I'm wondering if it should be a note?

<Joshue108_> JS: Well it sounds like we should talk about it more. If you dont see a quick path thru.

<Joshue108_> JW: The question is to what extent do we want to address privacy concerns here?

<Joshue108_> JW: Need to set boundary.

<Joshue108_> JW: If we acknowledge them appropriately then we have achieved our mandate.

<Joshue108_> JW: The auth acks the issues and avoid strong rec of CAPTCHA techniques that brings privacy issues..

<Joshue108_> JS: That could be a user choice.

<Joshue108_> JW: Yes.

<Joshue108_> JS: We do point out this is a treasure throve for Google.

<Joshue108_> JW: I need to go back and look at draft, and maybe bring it back for descision on way to solve.

<Joshue108_> JS: We may need to do a similar thing for PKI.

<Joshue108_> JS: Or do less for reCAPTCHA..

<Joshue108_> JW: Thats fine..

<Joshue108_> JW: These merit a more generic response.

Update - accessibility of emerging web technologies.

<Joshue108_> JB: I'd like to talk later about processing of video stream issue.

<Joshue108_> JW: Yes, when do you want to raise it?

<Joshue108_> JB: Keep going and I'll add something soon.

<jasonjgw> Josh plans to work on emerging technologies and gaps in accessibility standardization (i.e., how the gaps relate to the emerging technologies).

<jasonjgw> Artificial intelligence, machine learning, etc., are in scope - looking at issues not addressed in current specifications.

<jasonjgw> There are opportunities to ensure that the experiences of these technologies are inclusive of people with disabilities.

<jasonjgw> Josh also plans to contribute to curriculum development, authoring tools, etc.

<jasonjgw> Judy notes the "industry verticals", and the challenge of creating intersections between the work of APA and these areas in which activity is taking place. To build this intersection, we need to understand user requirements and the experiences of people with disabilities in using the technologies included in the verticals.

<jasonjgw> Judy: Example - novel applications for accessibility of the web of things. Furhter example: XR and accessibility - enhancing the dialogue with this community and associated W3C groups. A further opportunity for informing the scope - the upcoming workshop on web games.

<jasonjgw> There was a clear need for focused effort in this area, hence the work described by Josh.

<jasonjgw> Janina notes the APA effort to participate in the Web games workshop, development of a position paper.

<jasonjgw> Judy notes the process with respect to the web games workshop - enabling the issue to be identified early and included in the scope of the event at the planning stage.

<Joshue108_> JW: It seems that some of the issues from verticals would be appropropriate for this TF.

<Joshue108_> JW: There is stuff IN CG around convo and speech based UI, machine learning etc.

<Joshue108_> JW: So any a11y concerns would be fitting for both APA and RQTF.

<Joshue108_> JW: All sounds good.

<Joshue108_> JW: There are interesting areas such as improvements to validation and appropriateness of text alternatives..

<Joshue108_> JW: Going beyond pattern based stuff..

<Joshue108_> JB: We can put that on the agenda..

<Joshue108_> JB: Bringing new discussions in will be great. There may be IP concerns however..

<Joshue108_> JW: Yes, those conversations will help solidify priorities and help implementation advise.

<Joshue108_> JW: Some of those issues could be brought here.

<Joshue108_> JS: A la personalisation TF..

<Joshue108_> JS That needs to be explored, we are looking at redefining the concept of alt.

<Joshue108_> JS: Alt is an alterntive to graphical images right, the new COGA idea is that we may have different categories of alternatives..

<Joshue108_> JS: say a blob with numbers, may have an easy language or other alternative..

<Joshue108_> JB: I can see the need for that.. @alt text is complex with nuances..

<Joshue108_> JB: Does this make something more complex?

<Joshue108_> JS: We are redefining the structure..

<Joshue108_> JOC: Sounds great. Very interesting.

<Joshue108_> JW: One way is to generate it automatically where you can. There is a question of how you define responsibility..

<Joshue108_> JW: But you have author intervention..

<Joshue108_> JW: Image rec, speech rec and other fields raise questions about how to include that in a11y solutions..

<Joshue108_> JW: How do you maintain a high level of quality and include it in the author process.

<Joshue108_> JOC: Very interesting, the idea of providing semantically suitable alternate forms for different user requirements such as COGA, or other personalisation needs.

<Joshue108_> JS: Could be alternatives for images and blocks of text.

<Joshue108_> JOC: Symbolsets etc could be served as alternatives etc.

<Joshue108_> JB: There are converstations that may need to be embedded into HTML, or other content.

<Joshue108_> JB: Sandboxing of rapid changes to contrast ratios etc that doesn't leak_in video.

<Joshue108_> JB: I'd like to see if we can define the need etc.

<Joshue108_> JB: reaction?

<Joshue108_> JB: Would that be ok?

<Joshue108_> JS: Yes, more APA.

<Joshue108_> JS: We need more experience with implementation and user agents.

<Joshue108_> JS: I think the spec is ok, even for encrypted media.

<Joshue108_> JB: I've got some updates on that.

<Joshue108_> JS: We need to stop being theoretical and get real user agent experience, we dont have enough.

<Joshue108_> If there is a desire to find out if research is published.. we can do a review on the research carried out.

<Zakim> Joshue108_, you wanted to say there is a need for a11y use case and user needs research methodolgies - we hope to look at this as a research question.

<jasonjgw> Josh notes the desirability of developing better methods for creating use cases and investigating user needs with respect to accessibility.

<jasonjgw> This coheres well with the need for more user agent experience and implementation experience.

<Joshue108_> JW: We can look at the agenda next week.

<Joshue108_> JOC: I can go thru my role again new week for the TF.

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.154 (CVS log)
$Date: 2019/04/10 14:07:28 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/lans/plans/
Succeeded: s/brough/brought/
Succeeded: s/JS: I can see the need for that.. @alt text is complex with nuances../JB: I can see the need for that.. @alt text is complex with nuances../
Default Present: janina, Joshue108_, Judy
Present: janina Joshue108_ Judy
Found ScribeNick: Joshue108
WARNING: No scribe lines found matching ScribeNick pattern: <Joshue108> ...
Inferring Scribes: Joshue108

WARNING: 0 scribe lines found (out of 153 total lines.)
Are you sure you specified a correct ScribeNick?


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 10 Apr 2019
People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)


[End of scribe.perl diagnostic output]