<scribe> Scribe: Cyborg
Shawn - draft of FPWD. Jeanne, you linked to Editor's Draft.
<Lauriat> Editors' Draft (raw git version): https://raw.githack.com/w3c/silver/ED-Draft-js-Oct/guidelines/index.html
Jeanne: work in progress, I did
work on it last night and today. The abstract still needs work.
If someone could help me draft W3C version, we recommend
someone else use the plain language version, we haven't written
yet.
... I put in some guidelines and I took 2 things I've been
involved in, so I knew where they were. I'd like to include
more as people work on them. The name of the work and the
actual guideline itself. These can be changed as each group
does more and as things are further along, you can tell me what
you want this to be.
<Lauriat> Introductions, copied from elsewhere and in need of review/proposed changes: https://raw.githack.com/w3c/silver/ED-Draft-js-Oct/guidelines/index.html#introduction
Jeanne: the status of the
document required by W3C, we can't change that, Table of
Contents is automatically generated. Intro, I'd welcome
recommendations, comes from old Editor's Draft. Includes
Headings and Clear Words with explanations. This brings up
plain language prototype. This is tabbed panel that gives more
info about each of these pieces of guidelines.
... for everyone who is working on various drafts, as you work
through process document, you get down to part where you are
filling in this information based on earlier work you did on
user needs, so know that is what is the goal of what you are
sending. what I get in time will be included.
... any questions?
Shawn: Guidelines section 2 of draft, currently headings and clear words, those have numbers next to them, is that something we can't do anything about?
Jeanne: auto generated and there should be a way to turn them off.
Shawn: with WCAG we have numbers for SC, and since we were getting away from that, is there a way we can just have headings and not numbers.
Jeanne: great point. i will look
into it.
... i will remove the numbers.
... any other comments?
... i originally wanted to put the plain language right into
the document, as an accordion and tab panel, and I could not
find an accessible version of the code that would work well for
this, so it is simpler to just put a link.
... conformance - we worked on this last week in the
conformance meeting, mostly the goals, and I hope I caught all
the edits from the meeting.
PeterKorn: user oriented vs page oriented, site wide vs page based - both are saying don't be page based. what the person is trying to do is the stronger of them, so potential rewrite of bullet point or making 6 and 4 more harmonized with each other.
Jeanne: was going to just delete that one. for now, I will combine them...
PeterKorn: maybe for the 6th bullet, better to say...
Jeanne: they are actually different things
PeterKorn: for 6th bullet, want
to say support measuring or looking at subsets rather than
entirety
... support for organizations to define logical subset
Jeanne: I like that
<PeterKorn> "Support organizations in defining a logical subset of a site for conformance
PeterKorn: in 7th bullet, not a fan of word penalize
Jeanne: ok?
<Fazio> +1 to changing or removing penalized
PeterKorn: write as if
organizations are trying to do the right thing, rather than
avoid penalties
... help organizations focus on things that have large impact
vs things that don't
Jeanne: I don't know that we're doing that though
PeterKorn: I'm wondering how this can be written to frame it to help organizations focus on the most important stuff. don't do things that have no significant impact.
Jeanne: maybe we need to say something regarding 100 percent or nothing
PeterKorn: if we think of what sub bullets are trying to do to support top level bullets, something more flexible, maybe the word is prioritize
<Fazio> holistic and incremental maybe?
PeterKorn: address things with
large impact vs small impact
... the second sub bullet feels like a second sub bullet of
third sub bullet, not sure if it adds value on its own
Jeanne: I think I put it in twice
PeterKorn: prefer measuring than
claiming, claiming it focuses on public statements rather than
stuff that clearly helps people
... the other key issue in the challenges I've talked about are
that many of the criterial require human intervention. need
sub-bullet that speaks to that
Jeanne: I didn't put that in
because even if we go to task completion, that will be
manual?
... not sure it's an issue we can actually solve
PeterKorn: does everything we state in goals have to be achieved to deliver or can we mention things we are trying to do but not succeed at?
Shawn: good point to call out and some of the goals may not be met, some are aspirational and less measurable in terms of success.
Janina: partway on some of these goals and that is valuable
<Lauriat> Cyborg: Based on what I heard, I wanted to raise an element of concern about focusing on what's important and what's not, around people and their needs.
<Lauriat> Cyborg: Consensus, with requirements, we had work to build requirements on consensus, but not on conformance goals. I don't know if that represents consensus for conformance goals.
<Lauriat> Cyborg: That work that we've done around potential impacts through the conformance model on people with disabilities, whether we can include something on that in the editor's draft. What can we do around incentives?
<Lauriat> Cyborg: The impact specifically around the impact to specific groups whose needs may be deprioritized as a result of certain aspects of motivation for what organizations should do.
Jeanne: what you're referring to is related to work on points system, didn't include that as we don't have consensus on that. i would be willing to include something in goals about transparency and removing structural bias, would be happy to include that, we can work on that and talk about it on tuesday.
PeterKorn: for what it's worth,
user-oriented rather than page-oriented is supporting the idea
that a lack of something (e.g. alt-text) might be violation,
and alt-text is v important for important images, but if it is
clear in context, then that is less important if it is
missing?
... does a failure to meet a success criteria prevent meeting
what they are trying to do?
Jeanne: structural bias inherent
in WCAG 2.0 A, AA, AAA, where people with no vision had A SC,
because complete barrier, people with cognitive disabilities
more AAA, because the idea is if they try hard enough can work
around it.
... several proposals around different point systems that may
continue structural bias, put it in goals that as best as we
can, we are eliminating structural bias and treating all
disabilities equally
<Lauriat> Cyborg: Yes and no. Yes, that's why I'm referring to. But there's many ways that that could creep back in. We might flatten A, AA, and AAA, but then allocate points to each in ways that then reinforce that same bias.
<Lauriat> Cyborg: Where's the threshold where burdensome meets prevention. Things are "just tougher", but they can be done, end up with end results of task failures because of that added burden.
<Fazio> +1
David Fazio: this is back when we were talking about penalize and aspirational goals, iso continuous improvement, providing incremental and wholistic approach is closer to what we are trying to achieve
David: accessibility throughout product lifecycle
Jeanne: would like to encourage you to write up some of those ideas as a proposal. i would be very interested in writing up something that is based on that for conformance.
David: if you shoot me an email about what you want me to include, piggybacking off standards and definitions would be great
<Fazio> I would love to be involved too
<Fazio> major concern of mine
<Lauriat> Cyborg: I was just speaking with a startup trying to make a process more accessible to people with cognitive disabilities, with a system that has aspects of it that poses barriers. I'd like to talk through things with you (Jeanne) maybe offline to explore this more.
Jeanne: to see this is going to
help people working on different guidelines, to see where we
are going.
... content updates?
... in that case, let me ask...for each of the groups, then,
who thinks they have text to be included in this format, within
the next 10 days?
... for everyone who suggested re-organizing wiki, that was
very helpful, and i've done those
... section headings, from plain language prototype, Jan and I
were going to bring that up to the process we've been working
on now, to evolve this. we will set a time to work on section
headings.
<Lauriat> Cyborg: Language of the user needs hasn't been verified for accuracy and the form has changed. If you want to get something in, we can write up a draft of something brief that we could include (with maybe something explaining how it will probably change).
Cyborg: intro with intent is possible
<Lauriat> Cyborg: We [alt text] have shared the version of the user needs, but it hasn't been written into that format. I can see if I can get it more into that format by Monday, our next meeting.
<Lauriat> Cyborg: We can go postpone the test stuff and on this first and get something together by the end of Monday night.
Jeanne: clear language, a lot
more done that is not yet in HTML and github, should have more
of that by Tuesday.
... point of regard?
Charles: no volume of work sufficient for deadline of this draft.
Jeanne: that's fine. but i appreciate you starting working on it, hopefully we will have it for next draft.
Shawn: wanted to follow-up on email thread on date to present FPWD to AGWG - Tuesday or Nov 19?
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) Present: jeanne PeterKorn Cyborg CharlesHall Lauriat Fazio janina johnkirkwood Rachael shari Found Scribe: Cyborg Inferring ScribeNick: Cyborg Found Date: 01 Nov 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]