AGWG Teleconference

03 Oct 2023


Chuck, ShawnT, Rachael, alastairc, Francis_Storr, bruce_bailey, Bri, sarahhorton, kirkwood, Makoto, giacomo-petri, kevin, mbgower, Azlan, frankie, Laura_Carlson, DanielHE, dan_bjorge, maryjom, Ben_Tillyer, scotto, Poornima, ljoakley, tburtin, GreggVan, jeanne, ljoakley1
Jennie, DJ


<scribe> scribe: bruce_bailey

new members and topics

Rachael welcomes any new members, ask for future topics?


WCAG 2.2 will publish Thursday!

<alastairc> https://github.com/w3c/wcag/pull/3439

alastairc: Will be note added for internationalization

<alastairc> https://github.com/w3c/wcag/pull/3437

alastairc: links to github if anyone curious, please say so today. SLH helped with editiorial.

Rachael: Any questions?

Makoto: I am curious about ISO update ?

<mbgower> present_

alastairc: There is an interest with updating ISO reference.

kevin: 16 oct is deadline for ISO submission, and we will work out what goes into ISO version..
... I believe that the ISO version will have only normative text.

<Rachael> Survey for the last 2 subgroups https://www.w3.org/2002/09/wbs/35422/Sept_subgroups/

Rachael: Looking for volunteers for last two subgroups please.
... as with this week and last week we are just working on exploration topics...
... we will alternative between these groups and conformance over the next several meetings.

Francis_Storr: Can you explain "control semantics"?

<Rachael> https://docs.google.com/document/d/1EsNS1z_WBt3Ey30m-At8V87jYeM65KoWSR7L3SRR3T4/edit#heading=h.dqb2ye84xt4r

Rachael: Based on earlier work on grouping items.

<Azlan> I don't have access to the survey

<Rachael> Aid navigation: https://docs.google.com/document/d/1EsNS1z_WBt3Ey30m-At8V87jYeM65KoWSR7L3SRR3T4/edit#heading=h.ipa8mijomyem

Rachael: Definitions for aid to navigation also part of that draft

<Azlan> Thank you

Rachael: It includes breadcrumbs and outlines and other things

new members and topics

Phrasing outcomes https://docs.google.com/document/d/1JlOTK9m9DJjh6RqRsLupKfE5AuRK7nlu7IB6vpRDXOM/edit

Rachael: We have done enough work that we want to have exercise for phrasing outcomes.

<Rachael> https://docs.google.com/document/d/1sugAtqie_x1XqHDZo1Im7ftDNllWeRV_ty4PULeoTV0/edit#heading=h.q6kvhdps0qv4

Rachael: we have subgroup which worked with ACT and we have that for background,

but remember that we are in exploratory phase, things will change.

Rachael: We want to have a uniform way of phrasing outcomes so that might facilitate card sorting or other later activities.
... doc also includes examples from other groups

<Rachael> As a person who cannot see an image containing text, I need to know what the text says.

Rachael: but do we want more of a story opening? I will put examples into IRC.

[rachael reads user stories]

<Rachael> Users of Assistive/Adaptive Technology can control content through their devices

<Rachael> Provides the ability to transform content based on user personalization

Rachael: these are user-center examples
... another is more application oriented.

Chuck: I do think writing as user story is valueable for guidelines, but I thing I prefer technical approach for outcomes

<kirkwood> user centric approach has a tendency to be more plain language. if that’s a factor

jeanne: I prefer user needs in outcomes because it has real value...
... it keeps user included even when technical and stripped down to check list.

dan_bjorge: I mostly agree with chuck in that I have strong preference for normative requirements being as clean and precise as we can make them.
... so I prefer outcomes be clinical. I recogonize user stories are so important, so am all for keeping User Stories nearby.

<Zakim> alastairc, you wanted to comment on the 'as a user' format, and why we've gone with outcomes

alastairc: To caveat slightly, I do have concern that examples will not be only aspect which is normative, but we can wait on this.
... I also am not a fan of the "as a user, I" approach but User Need is quite important.
... for example "user can control content from their devices" is an example that is concrete but include user need...
... I prefer content does X so that user can do Y format

GreggVan: User Need is important so developer meets the the user need while meeting the normative requirement....

<Ben_Tillyer> +1 to Gregg

<kevin> +1

GreggVan: but user need belongs in Understanding type document since they will be long, subtle to people new to accessibility, and many SC address multiple disabilities...

<kirkwood> +1 to Gregg

<jon_avila> I agree with Gregg that multiple disabilities will be covered by outcomes and it will be problematic to have it in there as people will get left out.

<Zakim> Rachael, you wanted to say there may be a lot of user needs

GreggVan: Normative parts need to be unambiguous but User Needs may not be testable.

<jeanne> -1 to the mocking tone that I heard from the imitation of a person whose disability would not be included. I found it offensive.

<Zakim> alastairc, you wanted to comment on overlap of requirement & user-need

<laura> +1 to Gregg

Rachael: I really like exploring User Needs as an important approach, but the FAST frameworks has so many categories. I worry for length.

alastairc: In the way we have done before, each requirement (SC) has multiple audiences. I suspect though if its swapped around, that the document won't get much longer.
... my preference would be for the tests to be the normative parts.

Rachael: Please follow the W3C eithcs work requirments

<jon_avila> I agree with Gregg that the mapping to disabilities should be able to be updated to add additional mappings.

GreggVan: Many of the requirements are going to be cross disabilities. Taking cognitive as example, many structural SC important for those...

<Zakim> mbgower, you wanted to say it is possible to include user need, but we need to get agreement on it being okay to be more generic most of the time, where it is broader, and to

GreggVan: AI will need that structure to reformat content fo address language and learning.

Mike Gower: From working on the "in brief" statements (for Understanding) I can say they are helpful but can be very long if try to address all the Whys..

scribe: With the WCAG2 approach, so many SC apply to so many functional limitations...

<Zakim> kevin, you wanted to ask if the outcomes can be 'testable'

scribe: concure with Gregg that it would be valueable to have mapping showing the relations. But I think it will be hard to do in text alone.

kevin: The full richness and breadth of whats is address might be done through tags...

<Zakim> jeanne, you wanted to propose that we do an experiment with trying several wordings and see what works?

kevin: I think Outcomes -- how does it benefit users? But as one has test that becomes more specific and perhaps the normative parts.

jeanne: I am strong believer in trying things out, see how they work. Hypothetical discussions are not likely to be productive. I would suggest writing outcomes multiple ways.

<GreggVan> +1 to examples

Rachael: We might come back to that, not today though. I want to get back to AC suggestion.

sarahhorton: There are mutliple ways to approach and thing about Outcomes but we almost need different term...
... describing what might happen if we apply methods to a digital things, that leads to outcomes.

<Zakim> alastairc, you wanted to try examples

Rachael asks Sarah to re-draft Alistars list with that approach.

alastairc: There are different approach with outcome -- what is the outcome is not neccessrily the user, more that the outcome is possible...
... then describe in a more technology specific way how that happens.

<Zakim> GreggVan, you wanted to ask a testing question. If a user need is in the normative text of a requirement - would that mean that a test would need to include whether the user

Rachael: I am re-writting a few which we might consider as examples. Sarah as well. Please be encouraged to offer others.

GreggVan: Experience with WCAG2 was that as soon as User was part of the SC, the requirement became untestable...
... We would tend to think of the most competent users, e.g. TV Ramen, then it was possible...

<Zakim> alastairc, you wanted to say that this is a different context, the ability is there, but could be accomplished in different ways (user, user-agent, auth text etc)

GreggVan: came appart when we tried to think of more typical users. Having User in SC never worked even though we tried.

alastairc: I think we may be able to include Users might if the content or whatever provides the feature. Under the hood, could be multiple ways to achive.

Rachael: We have 3 ways to approach.

<alastairc> https://docs.google.com/document/d/1JlOTK9m9DJjh6RqRsLupKfE5AuRK7nlu7IB6vpRDXOM/edit

sarahhorton: The difference is the verb approach gets at what is being effect. Seems like it may not work for all outcomes.
... Images with text have problematic text descriptions -- versus text description is the equivalent the visual presentation.

<Zakim> GreggVan, you wanted to say

sarahhorton: so having User in the Outcome seems like it could be effective. The "thing" is defined by the story which meets the user needs.

GreggVan: Might we consider a two stage approach?...
... First stage is not testable, describe what the User is trying to achieve...
... next stage is the tests and multiple testers must be able to come to same conclusion. Those would be testable Outcomes.
... that would allow us to advance something which would get adopted.

<mbgower> +1 that user centric seems more plain language

<Zakim> alastairc, you wanted to comment on the levels, do we need another level? Plus should we move onto the main exercise and try these out?

kirkwood: Thanks for this discussion, there are two things, but I have to point out that the User centric approach also address Plain Language.

alastairc: This conversation is making me wonder if we need another level. Guidelines and Outcomes at present are still quite broad...
... not suggesting that for now. I would like to try approach with our work for today.

<Zakim> Rachael, you wanted to move us to resolution

Rachael: We will be getting to a card sort, so the levels can be part of that.

Please try different way to approach and we will return to this conversation.

Try the ones most promissing

<Chuck> bruce_bailey: I was wondering, the activity last week, some of the work groups were better than others. I love the template, but I also do well with an example of using the template.

<Chuck> bruce_bailey: Mentioned that no prior WCAG 3 research, that was a placeholder, because there was research mentioned.

<jeanne> It was a placeholder "no prior research" from Silver

<Chuck> Rachael: Meant that we couldn't find something relevant to the current conversation, but people were adding content. If no prior research from Silver. It's good to identify if there's no research so we can advance to the research task force.

<Chuck> Rachael: We want to identify the gaps.

<Chuck> bruce_bailey: For prevent harm, there's lots of comments about moving some user needs to other subgroups and other guidance.

<Chuck> Rachael: As groups identify needs, we encouraged groups to write those down so that we can review later and not lose the thoughts.

<Chuck> Rachael: It is better to note something and where it should go, we can remove duplicates later. Better than to lose.

<Chuck> Bruce: For text symantics, there's a lot about reflow. Today should we skip the reflow if we are in that group?

<Chuck> Rachael: Leave the note in the outcomes list so we don't lose it. If you are certain it's covered elsewhere, you don't need to cover it deeply.

Racheal: Four groups. If you have not participated previousl, please just pick the topic as you like.
... please just self-asigne

Scribing concluds for now, notes go into scratch pad document.

Rachael: We do not have number at the moment.
... if no preferences , please just hang back and chairs will assign

<alastairc> you have to scroll down...

ljoakley1: Hang back and you can see rooms fill up.

<jeanne> Prevent harm scratchpad https://docs.google.com/document/d/1S9K4i0CAiTHCWXD0_LnpS90E2l9VC_Xs18UmOXhGzaw/

Racheal: All back.
... in our group, we did not through template.

How far did groups get?

Should we repeat next week?

alastairc: I would like people to keep working through off line.

jeanne: Prevent harm decided we wanted to meet outside of the main meeting.

Mike Gower: Last week we had intent to meet between Tuesday meetings. That did not happen.

scribe: We have had two weeks of discussion, it might be productive to continue.

ljoakley1: One week is not enough time to get through a guideline.

<mbgower> +1

<dan_bjorge> +1

<frankie> +1

<kirkwood> +1

<Makoto> +1

Rachael: Any objections to returning to two hour meetings?

<jeanne> 0 - continuity of people helps a lot

<ShawnT> +1 to 2 hour meetings

Bruce: We did not get far, but phrasing as Users Needs seems to work well.

GreggVan: It worked to say Users provided a control... versus broad users CAN ...
... seems to work if more narrow User need versus general abiltiy.

alastairc: We do not have ability to summarize groups, but agree with spending more time next week.

Rachael: Please be encouraged to meet or add notes to documents.

GreggVan: We had example of doing outcome but then user need...
... example was "turning off audio" then several user needs followed easily from that.
... Do we need more discussion about that? We will have four or five User Needs per outcome with how we are going.

jeanne: This came up in discussion for prevent harm...
... at this stage seems more important to have exhaustive User Need rather that risk skipping because Outcomes are too technical.

GreggVan: So don't hang up that the end document might look different.

mbgower: Similar to jeanne , where you decide user need is out of scope, be specific about that , do not just delete or ignore

<kirkwood> +1 to reference other groups if we can

mbgower: but do not need to address

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2023/10/03 16:33:45 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision VERSION of 2020-12-31
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/we will to cap what goes into/we will to work out what goes into/
Succeeded: s/we will to work out what goes into/we will work out what goes into/
Succeeded: s/ISO version/I believe that the ISO version/
Succeeded: s/suggest adding outcomes multiple ways/suggest writing outcomes multiple ways/
Succeeded: s/q_//
Succeeded: s/presennt+//
Default Present: Chuck, ShawnT, Rachael, alastairc, Francis_Storr, bruce_bailey, Bri, sarahhorton, kirkwood, Makoto, giacomo-petri, kevin, mbgower, Azlan, frankie, Laura_Carlson, DanielHE, dan_bjorge, maryjom, Ben_Tillyer, scotto, Poornima, ljoakley, tburtin, GreggVan, jeanne
Present: Chuck, ShawnT, Rachael, alastairc, Francis_Storr, bruce_bailey, Bri, sarahhorton, kirkwood, Makoto, giacomo-petri, kevin, mbgower, Azlan, frankie, Laura_Carlson, DanielHE, dan_bjorge, maryjom, Ben_Tillyer, scotto, Poornima, ljoakley, tburtin, GreggVan, jeanne, ljoakley1
Regrets: Jennie, DJ
Found Scribe: bruce_bailey
Inferring ScribeNick: bruce_bailey

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

WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

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]