<jeanne> TPAC?
<jeanne> TPAC?
Jeanne: Looking at Plain Language prototype today
<jeanne> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Main_Page#Plain_Language
<jeanne> https://docs.google.com/document/d/1BGr0XSQgjBSVDG_Xn9MoodEq01cJxgg_EE0bniXONXM/edit
Jeanne: Realized the section
headings were a little random, so I organized it
... as I was doing that, I realized it really isn't clear how
this would be used. Cyborg has links to the drawings in the
comments, but people probably might not see the comments.
... so I put that in the proposal section. As we're thinking
about showing this to people, how do we organize it so that it
makes sense?
... I think that would be an important thing to do before TPAC
and seems like the next step to improve how we're presenting it
to people
Lauriat: My gut feeling is to
take the design work in the drawings and turn that into a few
flat HTML pages with content within it
... but steer clear of including technology aspect, since
that's a different set of thigns we want to get feedback on, so
we don't want analysis of this prototype to get conflated with
what we're not trying to test
... want to make it clear there are three different
prototypes
Jeanne: I think we should do Information Architecture first since it ties things together. As people see it, they're going to need to know how it's put together
mikeCrabb: Do you need someone to make the HTML pages?
Jeanne: If you have a good accessible tab panel...
<Cyborg> sorry, just got on call
mikeCrabb: I could find one
hi Cyborg!
<mikeCrabb> https://inclusive-components.design/tabbed-interfaces/
<Charles> Heydon could also be invited to help
mikeCrabb: I'll make it this weekend and put it up on Github
Jeanne: any other thoughts about the prototype? and what we want to do with this?
Cyborg: Not sure how much is needed for the meeting; in our last call I was on, there was a discussion about feedback fitting into three categories. I've added a fourth.
<shari> present_
Cyborg: there were a couple of points to make sure get addressed. I think we were going to go with developer tasks first
Jeanne: Want to add to remember
that we dont' have to make this match WCAG exactly since we're
doing it differently.
... some of the success criteria are broken down by
level...this at Level A, this at Level AA, etc. we don't have
to do that
... we can merge things more than WCAG has done.
Jennison: That's something worth
stressing...that the intent isn't to mirror WCAG.
... We're not trying to be WCAG. We would honor the work that's
been done, but we're not treating things as set in stone.
... the more we say that the better
+1
Shawn: We just need to do it wisely. Focus on the tasks at hand, keep the "not doing things like WCAG" conversations high level.
Jeanne: I think the hard part will be getting the structure right.
Lauriat: I think we should remind people that structure was one of the main drivers for Silver
<Cyborg> I think if we can get the developer tab done on the plain language prototype well, and the get started tab is both technically precise and following all best practices for plain language, we will solve a lot of the concerns.
<Cyborg> And the draft is ready for the recruitment email for people to help with the developer tab -- developers/auditors who are good at writing for a general developer audience, who are good at training new people.
Jeanne: I was assuming the tests would be part of the methods, but as I read the comments, I see we're going to need some type of normatic test statement to go with the guidance.
mikeCrabb: this was brought up a few weeks ago. It would make sense to have a test on the guidance sense "conforming with this guideline will results in this experience"
Jeanne: So instead of the test,
it would be the result. Interesting.
... anything else on Plain Language?
Lauriat: Thanks to Cyborg
<Cyborg> thanks, my pleasure
Jeanne: let's take a look at the Information Architecture
mikeCrabb: spent a few hours on the IA this week. And coming up with database models and started creating an API
<jeanne> https://github.com/mikecrabb/silverTaggingAPI
mikeCrabb: I have a database and
root API, after that I can start building an interface to work
with the API
... for data, I have dummy lorem ipsum text. all of the
guidelines have standard lorem ipsum, and ?? have pirate ipsum
so we can tell the difference.
... if we go straight for the plain language content, people
might get distracted by that
... hoping I can spend some time actually coding and then by
the meeting on Tuesday afternoon I'll have something to show
off how it's all working together
<mikeCrabb> https://github.com/mikecrabb/silverTaggingAPI
Lauriat: One nitpick. It looks
like you're using numeric IDs. I would suggest using short
names for now since using numeric IDs might be confused with
WCAG references
... if we stick with short names, we side step any
confusion
mikeCrabb: currently I'm using some technology related tags, but I could change it to tags which are shapes and colors
Lauriat: Could we use POUR
principles? so we can show people how it works
... having a few real examples of things that could fit more
than one principle
... it would show that we can carry the POUR principles
forward
<Cyborg> Shawn, how does the email I sent you about usability fit into task-based guidance/methods and then possibly into POUR? How does that relate to information architecture?
Lauriat: without it requiring you to organize them by principle
Cyborg: In principle, task based guidance fits into IA by the tags, so it's could maybe address usability concerns between usability vs. accessibility
<Charles> I’m still not clear what the discrepency is between usability and accessibility
mikeCrabb: it should work
<Charles> or who has this discrepency
Lauriat: At a high-level, basically looking at the conflict between "meeting the guidelines" but having something unusable and having something that's usable, but not meeting the guidelines.
<Charles> so that discrepency is between WCAG and accessibility and NOT between usability and accessibility
Cyborg: Something that is an accessibility issue, but doesn't have a guideline.
Lauriat: It might fit better in conformance.
Charles: If we define these things, it makes it easier to have the conversation
Cyborg: I like the task-based approach.
Jeanne: Anything else for IA?
Jeanne: Started working on some
test examples. I think before we get too far into that, we need
to talk about how we're going to present it.
... It's complicated and how do we explain it to people?
Lauriat: I think what we have we
should present it as "early explorations" but not what would be
as fleshed out as the other prototypes
... because it's complicated and there is a lot involved. As
much as I'd like to push it forward to get people on board, I
don't think it's solid enough to make the case yet
... we have some good directions, but need to present it as
such. TPAC might be a good place to get feedback from specific
individuals.
<Cyborg> How do we move that conversation in such a way that conformance model can be innovative and supportive of new approach?
Jeanne: I have an early diagram we might be able to use.
Lauriat: While you were out, we
couldn't get to the point where we could describe the
boundaries.
... it was like going the more granular route instead of the
generic route. The guidelines are pretty general now. Even
though you could apply it to native mobile app, they weren't
made for that.
... when it comes to conformance and allocating points based on
what you are, we need to be able to handle "this is a native
mobile app with a cart and social aspects" and the conformance
model changes based on the platform, implementation, etc. and
we'd have to define what each of those means because the points
apply differently based on the contexts
Charles: And there has to be
criteria that helps me determine which categorization my site
fits into.
... 8 of the categories could accurately describe my site
Lauriat: We have interesting
directions to explore for the conformance model, but it's still
early for us to really present it as a solid prototype
... it would be "here is what we're looking to explore, here is
out plan"
... For the context of presenting at TPAC, I would want to
present on conformance model last. It's the least fleshed out
and it's the logical end of "the story".
... and the conversations around it will be long, so better to
not have them overwhelm the other topics we want feedback
on
<jeanne> https://docs.google.com/document/d/1D_d5qrQtoX4BbNcbx4v0hjnkN0wPE3PLI8S5tY8ZPME/edit#
Jeanne: Here is the document I've
started working on with the test in it
... an Example of an introduction and then started into the
different tests we discussed. Stumbled with how it really fits
in.
<Cyborg> Jeanne can you please repost the conformance links?
<Cyborg> i got booted somehow by mistake
<Cyborg> oh i saw that
<Cyborg> it's ok
<jeanne> https://docs.google.com/document/d/1D_d5qrQtoX4BbNcbx4v0hjnkN0wPE3PLI8S5tY8ZPME/edit#
Jeanne: I was getting tangled up
in what was a "test assertion" and what was a "test"
... wonder if it would make sense to put test assertions in the
guidelines. And we could use them to nail down the precision,
make it normative, and we could have the actual tests as part
of the methods that would be easier to change.
... and the test assertions, I think are more related to what
mikeCrabb was talking about with the results.
<Cyborg> I like the idea of test assertions rather than too granular on guidance.
<Cyborg> and keep the granularity to methods
<Cyborg> Will task-based and product/project-based guidance fit into the conformance presentation?
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: Charles jeanne Lauriat LuisG Jennison mikeCrabb Cyborg shari No ScribeNick specified. Guessing ScribeNick: LuisG Inferring Scribes: LuisG Found Date: 05 Oct 2018 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]