<jeanne> feasibility test
<jeanne> Interruptions
<scribe> scribe: Chuck
Janina: Shawn will be the in
person person. Will there be wider agwg participation? Alistair
or Andrew?
... For Monday & Tuesday. I've an invite for later in the
week, related to XR.
<CharlesHall> I will be at TPAC Monday–Wednesday. but traveling locally Thursday–Monday. so I will miss the XR mtg.
Janina: Emergency technologies
and emersive web is part of our remit for Silver. If you are in
Japan thu or fri you are welcome to join.
... I've 2 conflicts on Mon and Tue.
Shawn: We do expect agwg to join
silver.
... Several will make a point to attend, some will filter in
and out depending on other meetings.
... APA side sounds interesting.
... I put link in irc for our page, we have a section for
groups to coordinate with.
... Has act, agwg, education and outreach.
... With agwg it's generate participation both directions. ACT
we'll say "we have groupings, we have user needs...." for
these
... ... let's build up the tests to see if the needs have been
met.
Janina: Shadi will be there?
Shawn: No idea.
<jeanne> Jeanne found the link to the AGWG TPAC agenda https://www.w3.org/WAI/GL/wiki/Meetings/TPAC_2019#Thursday.2C_19_September_2019
Jeanne: I found the agwg agenda.
Thu afternoon will be about silver.
... does that conflict with emersive web?
Janina: Not on thursday with emersive web.
Jeanne: Should we extend invite to Andrew and Alastair for Mon and Tue.
Shawn: We can, but we've already
offered invites to everyone for those 2 days.
... We specifically made the days not overlap.
... But reminder would be good idea.
Rachael: Where will call in info be posted?
Jeanne: will be a section on the wiki page.
<jeanne> https://www.w3.org/WAI/GL/task-forces/silver/wiki/2019_September_16-20_in_Fukuoka,_Japan
Jeanne: I still need to set that
up.
... Remember Japan is a day ahead of us.
Janina: 13 hours from east coast.
Jeanne: Do images of text still
apply to canvas. Short answer is that ...
... Canvas example in 1.4.5 is no longer relevant.
... Fallback content is fully supported. Not the shadow dom
(different meanings).
Shawn: There are caveats to that. I can talk about it with interested parties.
<jeanne> "I believe the canvas example in WCAG 1.4.5 is no longer relevent. As you have aluded, fallback content is fully supported for the canvas element. It’s important to refer to it as fallback content, and not the shadow dom, as they have very different meanings, as far as I know. You can easily provide alternative content for images represented in the canvas, using the fallback content of
<jeanne> the canvas element."
Jeanne: Essential to the user/functional needs?
Shawn: Yes. Context of text is
very important. As a for-instance. A video game rendered in
canvas. Probably have a bunch of labels.
... It's not enough to dump that stuff. Needs to have same
symantec relationship for content to be meaningful.
... It's not just a matter of saying "is there fallback
content".
Jeanne: About images of text.
Shawn: With canvas the entire
thing is an image. Will contain a bunch of text.
... Another example, google sheets. Sheets rendered visually is
in Canvas.
... Very important that fallback content has the structure and
relationship that is expressed visually in rows and
columns...
Janina: Interesting!
Shawn: We had to make that explicitly correct as far as relationships go to make braille displays render properly.
Janina: 3d canvas?
Shawn: canvas and 3d canvas not terribly different.
Charles: Note on the conversation
about canvas, started from text spacing and not images of text,
because understanding doc...
... <reads from>
... Canvas doesn't allow user to render that text
differently.
Shawn: You would have to detect spacing adjustments and apply within canvas rendering.
Jeanne: Good point to go back and look at the continuing the functional needs (next agenda item).
Jeanne: Here's a link...
... I have some other things open on my desk that I want to
mention...
... We have new proposals coming in on perception, and on
content side personalizing content. They are related. After we
finish this topic.
<jeanne> WCAG 2 Silver mIgration map https://docs.google.com/document/d/1aCRXrtmnSSTso-6S_IO9GQ3AKTB4FYt9k92eT_1PWX4/edit
Jeanne: Here's the migration map
(above)...
... Shawn, want to review work on 1.4.5? I think that starts
with 1.4.4.
Shawn: This is the grouping 1.4.4 ...
<Lauriat> Grouping: SC 1.4.4 Resize text, SC 1.4.5 Images of Text, SC 1.4.9 Images of Text (No Exception), SC 1.4.8 Visual Presentation, SC 1.4.12 Text Spacing
Jeanne: We worked on it while you were away.
Shawn: There's a lot of user
needs in here.
... Would someone talk through it?
Jeanne: We worked on it as a
group. No single individual. We did it over 2 meetings.
... We went to understanding docs for each individual sc last
Tuesday.
... We did merging, re-aranging and word smithing. That's where
canvas came up.
... Charles, do you know which one refers to canvas?
Charles: Which user need?
Jeanne: yes.
Shawn: I've some q...
... For this grouping, for the methods we have, it's around
text rendering customization. Virsual customization of rendered
text.
... Seems like there's a lot of things that overlap.
Charles: We thought there would
be another round of editing to merge bullets.
... I don't see specific need as it related to canvas. More
than one of them do.
... Issue was trying to determine if there were additional
needs if text is rendered through canvas.
... Docs specifically say it's considered an image.
Shawn: User needs aren't any
different if the rendering is handled in a different way. We
want tech agnostic user needs.
... Within browser you can render text in a bunch of different
ways. We look to see if that way of rendering meets these user
needs.
... HTML gives you a lot for free.
Charles: The other things we
identified that we don't have the best phrase to describe,
visual tracking.
... Within the realm of limited vision there's this concept of
visual tracking. Things that are too narrow or close
together...
... Need to have customization because tracking is the
issue.
Shawn: With tracking this is something reading text and their ability to read across line, get to next line, and not lose there place?
Charles: Or readability in
general. Assumes usage with vision, but tracking within their
vision is the issue.
... Could be character to character and/or line to line.
Shawn: Makes sense.
Jeanne: Do you want to add that explanation to 3rd from the bottom Charles?
Shawn: We added usage with
limited tracking.
... Maybe remove start.
Jeanne: We won't do editorial aspects today. We just want to capture the ideas.
Shawn: What is the user need
specifically? I don't see it written down.
... Is the need to customize?
Charles: That sounds like a
solution, the needs are written "usage without..." the need
here is "I need to track between characters and lines."
... The solution is to allow customization.
Shawn: Got it!
Jeanne: Other than these adjustments, are people happy with this list of user needs for 1.4.4-1.4.12?
Shawn: I've a granular question.
There's very specific things here.
... Should we break this up or generalize the user needs? Open
question, I don't have a feeling one way or another.
Charles: My gut says the process
of editing and combining will organically produce that
result.
... We need to combine the ones that say the same things.
jeanne: I like the granularity. We may need to break this one up some. But maybe not as done in WCAG.
Charles: I think this is still the correct grouping of sc's.
Jeanne: Hoping someone will take
this up when we finish up with the other ones that people took
on as homework.
... Someone can take this one on and review and edit.
... Anybody on the call that wants this grouping?
<no takers>
Jeanne: Moving on to 1.2.2 and....
<Lauriat> SC 1.4.2 Audio Control, SC 2.2.4 Interruptions https://docs.google.com/document/d/1aCRXrtmnSSTso-6S_IO9GQ3AKTB4FYt9k92eT_1PWX4/edit#heading=h.8ioplsj35s7t
Jeanne: Janina brought up
captions.
... Do these look new to you Shawn?
Shawn: They do, but I've been 6 days away, no computers.
Jeanne: I'll read through
<reading>
... Any others people thought of?
Janina: Not really a captcha issue, it's really an issue of the first point, you don't want overspeaking the text to speech engine.
<pkorn> Am I audible?
<pkorn> Skip me while I work that out
Shawn: There were 2 in the middle
that seemed same user need.
... <reads>
... Seems like same user need.
Charles: I think the second one
if we are going to name a specific disorder type opens up the q
of whether or not we describe the core need behind it.
... I think these are different because one is visual plus
audio, second one is not the combination, it's distraction
period.
Shawn: I can get that difference, but the second one is a superset of the first one.
Charles: I think the reason that
I am concerned about describing a.d.d. specifically is that
there is a general trend to not name specific things.
... It opens the door for another specific one, autism users
may have a compelling...
... I've read a lot, they state that they need to read every
word on the page. There's a specific challenge when that
process is interrupted.
Shawn: That sounds...
... Like a consequence when interrupted rather than a user need
to avoid interruption.
Charles: I think the simplest way to describe is that understanding is impacted if reading is not completed.
Jeanne: Put these examples in the bullets. Coga will know that we were considering these issues (autism and a.d.d.)
Charles: We have 2 other groupings that come to mind.
Shawn: 2.2.4 has interruptions as
its own thing. We may consolidate them. We have animation from
interactions (2.3.3).
... the visual interruptions corresponding to the audio
interruptions this one is addressing.
... We could consolidate around "interruptions".
<Lauriat> SC 2.2.4 Interruptions https://docs.google.com/document/d/1aCRXrtmnSSTso-6S_IO9GQ3AKTB4FYt9k92eT_1PWX4/edit#heading=h.6alx9fshrle
Shawn: All of the methods are
"interruption" based.
... Overlaps with a lot of what we are discussing now.
... Non consentual sound.
Charles: Are you saying we have the appropriate groupings and interruptions in the first case we reviewed, specific to audio?
<pkorn> Regrets; I need to leave a bit early.
Shawn: More of an open question. We could say that the grouping is specific to audio, and we can also do the same for visual.
Janina: Diving in.
... I'm tantalized by considering to combine them.
... Animations and spacings could be controlled through CSS.
Would be an interesting point if css want's to have impacts on
the various environments.
... I don't know the outcome, but that will be a conversation
Tuesday morning.
Shawn: Thanks.
... I'm going to copy in the list.
<Lauriat> Methods: Visual interruptions, Audio interruptions, Focus interruptions, Announcement interruptions
Shawn: I'm of 2 minds. I don't
want something that has too many tests and methods though
conceptually its all the same kind of user need.
... Maybe this will be "here's ways you can test this." may be
a good test case going through ACT
Janina: Don't know how the screen reader will handle this, or webex if the screen reader is talking.
Shawn: Not to high on webex.
Janina: The situation exists for any post connect on telephone. And people struggle.
Shawn: There's some things that could improve it, but would require platform changes.
Janina: Exactly.
Jeanne: Want to continue on this
or take a few minutes on conformance?
... It would be an update rather than new work.
Shawn: Let's do an update.
Jeanne: Chuck's a GREAT scriber!
Jeanne: We started working on
issues list for conformance. We are trying to address some of
the big picture items before
... We dive into how we measure things. In the hope that by
solving big issues we will not have to redo the point system
over and over.
... The one we were working on was what do we want to measure.
Suggestion came up last week that we make a list of
... The things that we can measure and do a feasibility test
where we do a rough scoring of these individual measures
against WCAG and the 4 new items.
... We can see how these measures work.
... I want to make the list of measures that we want to
test.
... We have some proposals we can tap into.
... I also want to add those from this group.
... Some examples are "ease of implementation"... "ease of
testing", "effectiveness", "# of groups affected", "severity (I
don't like this one)"
... Any others?
Chuck: Peter brought up "work-around".
<jeanne> Chuck: Peter Korn suggested the availability of workarounds
Janina: I want to suggest a straw
man: there may be several situations where we want to split a
single sc historically into a quantitative and qualitative
test.
... Alt text is great example.
... review text is not machine testable yet beyond testing for
structures. We should score those separately.
... And come up with different approaches.
Shawn: I look forward to working
through this with ACT.
... Testing if alt text exists is one step in a line that you
go through to complete the test.
... If no alt text, you have a conclusion, if there is alt text
but blank, you need an assessment, and if there but not blank,
need to evaluate the text.
Janina: That whole process applies to captions and described media, if tab order is meaningful, and a bunch more.
Shawn: I see that method of describing testing as very applicable to everything we are doing.
Janina: Silver will be great!
Shawn: 1-2 minutes left.
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/ regrets+Bruce// Succeeded: s/ext/text/ Present: jeanne janina Chuck Lauriat CharlesHall pkorn KimD Rachael WARNING: Replacing previous Regrets list. (Old list: Shawn) Use 'Regrets+ ... ' if you meant to add people without replacing the list, such as: <dbooth> Regrets+ Regrets: Bruce Found Scribe: Chuck Inferring ScribeNick: Chuck Found Date: 30 Aug 2019 People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. 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]