<Chuck> meeting: AGWG-2024-04-02

<Chuck> Continue text alternatives work https://docs.google.com/document/d/1wkQ8ZKiRnNnax9ENaDXfQhl6kh-iYlXx7gMEhKMTQBU/edit#heading=h.cu8l0ohk9n2q

<Jennie> scribe+

<Jennie> Chuck: We will get started in 2 minutes.

<Jennie> Chuck: Welcome.

<Jennie> ...Is anyone on the call who is new or has a new role?

<Chuck> Welcome Aditya!

<Jennie> Aditya: I am a UX designer, working with Knowbility. I am also a grad student.

<Jennie> ...I am working on a caption project.

<Jennie> ...I am working on an accessibility assistant tool.

<Jennie> Chuck: Next: any new topics?

New IRC client

<Jennie> Chuck: We discovered this last week.

<Jennie> ...This is an acknowledgement that the switch introduced some accessibility issues.

<Jennie> ...W3C is aware and is working on that.

<Jennie> ...We will get an update from Kevin.

<jaunita_goerge> Can someone write a dark theme!!??

<jaunita_goerge> We can input custom CSS

<Jennie> ...If you have joined and you are having trouble, we can use the Zoom chat to help.

<Jennie> Kevin: There is the possibility that we can look at other IRC clients that are available.

<Jennie> ...I need to take time to look at them and see if they could be any better.

<alastairc> In the settings there is a dark theme, called 'morning', but it doesn't make the contrast much better.

<Jennie> ...I told the team I would gather more information about the accessibility issues.

<Jennie> ...Daniel Montalvo identified some issues previously, and documented those in a wiki.

<Jennie> ...We have shared them with those who make "TheLounge"

<Jennie> ...I have a Google document - if having an accessibility issue or identifying any

<Jennie> ...please record them in the document.

<kevin> Capture TheLounge Accessibility Issues

<Jennie> ...There are some issues we can address ourselves, but others may need patching from the service

<Jennie> ...The Google doc also links to Daniel's work to see what has already been identified.

<Jennie> ...This is a work in progress.

Subgroup updates

<Jennie> Chuck: 2 subgroups are in progress.

<Jennie> ...1: the publication approach subgroup

<Jon_avila> The wiki linked page of issues requires a W3C account - my account doesn't seem to have permission for that.

<Jennie> Juanita: We didn't meet this week.

<Jennie> ...We resume meeting on the 9th. We have an agenda set.

<Jennie> ...If anyone would like to join: Tuesdays, 9am.

<Jennie> Chuck: 2: focus appearance subgroup

<Jennie> Rachael: We have met 2 weeks now. We meet Mondays at 10

<Jennie> ...We are progressing on writing testable outcomes.

<Jennie> ...We are adding pass and failure conditions for visual focus.

<Jennie> ...This is our working document.

<Jennie> ...We will also be kicking off a voice-input subgroup in the future.

<Jennie> Chuck: For our new members - time of day.

<Rachael> focus appearance document: https://docs.google.com/document/d/1IgNr6WAd9ovk7t4sMrrZkLJzsFBDROOpw3i1oCxlCVQ/edit#heading=h.juf86zc2cv61

<Jennie> ...When we speak in "time" - we are using United States - Eastern time zone.

<Chuck> Jennie: Did someone message me privately? Someone is asking for the link to join subgroups.

<Jennie> Chuck: If subgroup leaders could put their links in that would be great

Continue text alternatives work https://docs.google.com/document/d/1wkQ8ZKiRnNnax9ENaDXfQhl6kh-iYlXx7gMEhKMTQBU/edit#heading=h.cu8l0ohk9n2q

<Jennie> Chuck: This is a bit of review, then we will start working sessions. We will divide into 4 groups at that time.

<Jennie> ...This continues our text alternatives group.

<jaunita_goerge> Publications sub-group working doc: https://docs.google.com/document/d/15P9HBcYKmHqkedRDT_LIHwJAG4EEYUjHyUSBmIAZ7yQ/edit#heading=h.gm9vi8afej1t

<Jennie> (note: work, not group)

<Jennie> AlastairC: We have been looking at various parts of non-visual alternatives.

<Jennie> ...Images have text alternatives is our draft outcome.

<Jennie> ...This was to try to flesh out something that may not be a typical content image.

<Jennie> ...We have a couple of procedures from the subgroup - we will focus on the revised procedure.

<Jennie> ...The procedure is the default starting point.

<Jennie> ...Next: we will try to flesh out the examples.

<Jennie> ...There are links to the different groups.

<Jennie> ...(displaying group 1's document)

<Jennie> ...Some examples are crossed out.

<Jennie> ...This is so we maintain our numbering.

<Jennie> ...We left the procedures and decision tree, slightly refined.

<Jennie> ...This helps scoping what goes into or not into this outcome.

<Jennie> ...We have not figured out how this relates to the tests - we will be working on that.

<Jennie> ...Reminder: if not part of an operable control, it is within scope

<Jennie> ...(reads from the decision tree)

<Jennie> ...Today: working through examples. Example 1 is filled in for each group.

<Jennie> ...(Reviews information under example 1)

<Jennie> ...This includes what would pass, what would fail

<Jennie> ...There are examples of alternative text that would pass, others that would fail.

<Jennie> ...We are checking if we can say what would pass, what would fail.

<Jennie> ...Underneath: examples that are blank.

<Jennie> ...Today: carry on filling these in.

<Jennie> ...If they work well with the revised procedure, great.

<Jennie> ...Are any examples which don't work for this procedure?

<Jennie> Chuck: Are we implementing the revised procedure, then the decision tree?

<Jennie> Alastairc: question 1 - does this revised procedure work?

<Jennie> ...Does this work for these examples?

<Jennie> ...If not, why doesn't it? What would work?

<Jennie> ...Look at where each example falls within the decision tree.

<Jennie> ...Where it doesn't it also good information.

<alastairc> https://docs.google.com/document/d/1wkQ8ZKiRnNnax9ENaDXfQhl6kh-iYlXx7gMEhKMTQBU/edit#heading=h.acg4ly89e4m6

<Jennie> ...Now: we will distribute everyone into 4 groups.

<Jennie> ...There are 25 examples.

<Jennie> ...There are links to each of the documents.

<Jennie> ...Pick 1 group at random.

<Jennie> ...If any group has more than 14 or less than 9 we will shuffle people.

<Jennie> Chuck: We are going into exercises now (so scribing will stop).

<scribe> scribe: mbgower

Chuck: Group 1 did not finish, but we had a good time going through the examples. We think we made process.
... We think we found some bugs in the decision tree.
... I'm going to ask for each team to do a brief update, and give a brief example of a few things they found.
... [Reads out text in Example 2] We added in information.
... The image of three people does not match the caption.

<Graham> That caption is a fail, it doesn't explain the image, it is commentary as part of the article.

Chuck: But the article text did infer there was celebration.
... We did not conclude example 14 maps.
... We did find maps we thought would be worthy of review.
... Someone from group 2?

Alastair: We got most of the way through things at a slightly more draft level.
... We got through example 3, but found some that fall in between
... If you have an image name with a description. In the NYC example, there may be an in between.
... We also had AI generated images. In general, we didn't think they were particularly different. It may or may not be relevant whether it was AI generated.

Chuck: In regards to the last NY one with the scale. Did you come up with a way to tweak the decision tree?

Alastair: It depends on how you describe purely decorative. That image could fall into meaningful or decorative.
... For ASCII art, there's not too much you can do to make it accessible. We added something in the decision tree to try to tackle this.
... Our next one was example 15 where the image is the whole content
... Example 19 is Math served up as an image.
... There are potential ways to doing this, but whether it would be understandable was a question. You may need something beyond alt text.

Example 23 Dynamic images that change could use a few more examples

Chuck: Group 3.

Rachael: We got less far than you.
... Two images of different sizes that have the same ALT.
... In one, there is a thumbnail with the same image as a larger image. Where one is operable and the other is not, they should not have the same ALT.
... Where both are either operable or non-operable, it's a different consideration.
... If you have one high resolution image, and then a bunch of thumbnail of different resolutions, there is a need for a lot more information to match the high resolution image.
... For the thumbnails, infomration on the different sizes seems relevant

Chuck: Group 4, Kevin

Kevin: We feel Rachael cheated by using the Library of Congress.
... We looked for examples on the web. Images critical to the page use/task completion was relatively straightforward.
... For the decision tree, it seemed to be the same whether the image was operable or not operable.
... Example 9 Font icons were relatively simple and similar, in that they could be anything: informative, decorative, etc
... The revised procedure was more useful, since you would be looking at it with the context in mind.
... Is the ALT for screen reader users, or does it serve another purpose? For example, understanding the meaning.
... Example 13, we got nowhere
... Example 17 warranted a bit of discussion: images provided with the intent of exploration
... I had the idea around a map, where you're exploring a symbolic space, which led to a different discussion.
... [SHows I spy examples]
... Example 25 Infographics, including the many varieties of beer.
... We have conclusions around that.
... The revised procedure calls for including words in the text alternative. The words in an infographic became problematic. The beer example has a lot of text, but you don't need it for the context.

Glenda: I just wanted to say that I added in some comments, including alternate exploration methods.
... We may want to consider this for Bronze level.
... For example, raised tactile graphics.

Juanita: There is new research on haptic and multimodal that may render similar results.

<Zakim> Chuck, you wanted to compare approaches

Chuck: It seems as though groups took different approaches. Group 1 found an example, dug up the alt text and applied it to the decision tree.
... We spent time where we found the decision tree wasn't helping.
... It looks like some of the other groups were more hypothetical.
... I don't think there is a right or wrong. Just noting the difference.

Detlev: I wanted to talk about one, interactive images. The hamburger and search icons seemed like they should be easy, but they appeared in different contexts.
... They may have different roles. The same icon used for search can also be used to zoom in.

<Zakim> dan_bjorge, you wanted to discuss 1 more point from group 4

Dan: I was going to add from Group 4's notes, especially around info graphics.

<alastairc> Hmm, good point, need to separate images as-content for a control, vs graphics having controls.

Dan: What counts as being operable? There may be individual data points with tooltips. It wasn't clear if that counted in the decision tree.
... It would be helpful to have more examples.

Chuck: We also had some situations where it would have benefited from more definitions.
... Tomorrow when the chairs get together, we are going to look at results, evaluate, and plan for the next step.

Alastair: Feel free to continue if you didn't get through, but plan to be done before Friday, when I plan to merge into one big document.

Charlie: This was fun

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/Kevein/Kevin/
Default Present: Jennie, Laura_Carlson, alastairc, tburtin, JakeAbma, Rachael, kevin, JeanneEC, Graham, julierawe, mbgower, Francis_Storr, Frankie, giacomo-petri, Poornima, jtoles, gpellegrino, ShawnT, AlinaV, dan_bjorge, avk, ScottO, Azlan, sarahhorton, Makoto, maryjom, JenStrickland, ljoakley, mgarrish, jaunita_goerge, Detlev, Glenda, mike_beganyi, dj
Present: Jennie, Laura_Carlson, alastairc, tburtin, JakeAbma, Rachael, kevin, JeanneEC, Graham, julierawe, mbgower, Francis_Storr, Frankie, giacomo-petri, Poornima, jtoles, gpellegrino, ShawnT, AlinaV, dan_bjorge, avk, ScottO, Azlan, sarahhorton, Makoto, maryjom, JenStrickland, ljoakley, mgarrish, jaunita_goerge, Detlev, Glenda, mike_beganyi, dj
Regrets: Ashley Firth, Roberto Scano
Found Scribe: mbgower
Inferring ScribeNick: mbgower

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.)