See also: IRC log
<trackbot> Date: 09 February 2015
<scribe> Meeting: ARIA Authoring Practices TF
<scribe> scribe: jamesn
MK: Presidents Day next week
no one gets the day off
MK: we will meet next week
... Week of CSUN... March 1
JN: Prefer not to meet that week
<mattking> https://www.w3.org/WAI/PF/Group/track/actions/1574
ACTION-1574?
<trackbot> ACTION-1574 -- Matthew King to Propose wording for a browser and assistive technology note for example pages -- due 2015-02-09 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1574
MK: The list is whatever the
people who coded and reviewed the example used.
... There might be some we don't list on purpose.
... Just listing for people's reference. We are not making any
other guarantees
... right now we are mostly looking at JAWS 16 and NVDA
q4.2014
JG: Seems like a lot of
noise
... Maybe the examples should stand on their own. Could get
=confusing and distract from what we want is how people build
the widget
... if we have a standard thing - but if we are letting someone
decide what to list
MK: could guarantee plaforms but not versions
JN: perhaps we should say where it doesn't work in modern browser UA combinations
MK: potentially there is some difference.
<Birkir> What about a show/hide "assistive technology testing" section for each example. Offers possibilities to add, can be viwed if interested, but does not get in way of those who do not want to concentrate on taht.
JG: devs want some level of
confidence that if they use a practice they will get some level
of compatibility withj AT
... Is this part of the groups resopnsibility
All: no
<Birkir> Yes, 919 is me, just like 007 is Bond .. he is cooler but less interested in accessibility
<Birkir> I have once implemented a set of widgets, I had to make extensive notes on responsive behavior and a.t. compatibility, paritcularly voiceover as opposed to Windows, of course that was a client ask, but those notes turned out a lot more extensive than I ha anticipated
MK: if someone wants to learn what the people who wrote the example expected the experience to be - they could then reproduce exactly
LW: these just need to be examples of the design patterns
JG: Want to be clear what it is
we are trying to do here
... Just want to be clear that if we are going to publish
compatibility information why we are putting it in there
... have kind of a contract with people using the examples
LW: have to make it
complrehensive to be useful
... if it just says JAWS16 and FF then how much practyiocal use
id that in the real world
MK: think it is useful as it tells you that it wasn't written for jaws 3 or ff99.
LW: what good does it do in isolation
<janina> Isn't the purpose of W3C specs/notes to be agnostic about UAs and ATs?
Bryan: if it doesn't work in an AT / browser then it is the AT / browser fault that it doesn't work.
MK: then should put a date in there
Bryan: taking a forward looking approach then it is up to the AT to convey whether they support the use of aria
JN: shadi's work on database
<scribe> ACTION: James to talk to Shadi about the database to mark accessibility support of the examples with crowdourcing [recorded in http://www.w3.org/2015/02/09-aria-apg-minutes.html#action01]
<trackbot> 'James' is an ambiguous username. Please try a different identifier, such as family name or username (e.g., jcraig, jhawkins2, jnurthen).
<scribe> ACTION: jnurthen to talk to Shadi about the database to mark accessibility support of the examples with crowdourcing [recorded in http://www.w3.org/2015/02/09-aria-apg-minutes.html#action02]
<trackbot> Created ACTION-1577 - Talk to shadi about the database to mark accessibility support of the examples with crowdourcing [on James Nurthen - due 2015-02-16].
LW: make a fork of the aria repo
JG: can you turn a clone into a fork?
LW: don't believe you can
<Birkir> What I dream about, as a SME/user, would be to use these examples as test harness .. and report bugs to vendors of browsers/AT based on these examles, har to argue with them being best practice.
LW: take a copy and paste into
the fork
... send a pull request back to the main aria repo
MK: does the fork end up in his own space
LW: a fork is your own local copy
completely seperate then once comitted then you issue a pull
request to the main repo
... the chanbges then get merged back into th main copy
JG: will try to get in a fork
this week
... should have an internal standard that we should test things
on A, B and C.....
MK: one of the things I am
concerned about in general is if we were to document expected
results then just to say code it like this and then run it with
some AT or even no AT. People will experience something but the
question is whether what they experience is what we
expected.
... If we are relying on there being no bugs in the version
they are using.
JN: how would you possibly document that?
MK: spinbutton for example
... trying to think how we could communicate this. for link and
button maybe straightforward but for others maybe more
complex
JG: could have pseudo code for like if move slider user should be made aware of new value
MK: if press down arrow and screen reader says nothing. Is that a bug? Does the person who experiecnces that for the fiorst time know that there is a bug
JG: should look a the authoring
practices to see if there should be a place where we state what
should happen on certain interactions
... what are we trying to do and what are we not trying to
do
... going to be a lot of work for us. What developers want to
know is how should I be using aria
... if s dev doesn't know how a screen reader works then our
doc isn't going to be much use
... Should do the best to help developers use aria technology
the best we can
... if we are doing something to get around limitations we
should document it.
MK: have talked about our
audience being devs and toolkit devs & also AT vendors
being part of our audience too
... brand new 1.1 features - we need to help AT devs know how
they are going to be used and what is the repsonsibility of
AT
... we were going to put a note that something was a new
feature and not yet properly supported by AT
... I think you are spot on Jon - need to decide what we are
doing and why we are doing it
JG: it is not the job of this
group
... think we need a minimum set of standards for this
group
... some groups talk about current and last vcersions of AT
MK: we need to be developing code examples for (exmaple) aria-current with code samples where we can point the AT vendors to it.
JG: At vendors don't want to be
told how to create a UI
... anything like that would be interpreted as beyond the scope
of the group
Bryan: how would you know if
there was a max value of 11.
... could put in the valuetext
... if you have a time slider the only way to convey AM and PM
is to add it all to aria-valuetext
JG: not sure what the APG can do
to help with that
... close to consensus. Don't want to talk about what AT should
and should not so. Want to have some list of what we test
against. question is if we include that information in the
note
JN: agree with pretty much all of that
<janina> Log a resolution
<janina> Indeed!
RESOLUTION: We will
not dictate AT behaviour.
... We want to test our examples with a core set of browser and
AT combinations to be enumerated later
... Sometimes we may want to suggest outcomes that browsers and
AT may want to ensure are communicated but without suggesting
exactly how they do it.
proposed resolution: we will generally not document workarounds where the browser or AT is not following the standards correctly
RESOLUTION: we are looking for the current best thinking in aria development for what the features are meant to achieve.
discussion over a note to state that people are encouraged to use recent versions as stuff changes....
JS: reason stuff changes is that aria is in current active development
RESOLUTION: we will generally not document workarounds where the browser or AT is not following the standards correctly
This is scribe.perl Revision: 1.140 of Date: 2014-11-06 18:16:30 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Found Scribe: jamesn Inferring ScribeNick: jamesn WARNING: No "Present: ... " found! Possibly Present: All Birkir Bryan Bryan_Garaventa IPcaller JG JN JS James_Nurthen Jemma Jon_Gunderson LW MK Matt_King P3 aaaa aabb bgaraventa1979 https inserted jaeun_ku janina jongund mattking trackbot You can indicate people for the Present list like this: <dbooth> Present: dbooth jonathan mary <dbooth> Present+ amy Regrets: Leonie_Watson Ann_Abbott WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 09 Feb 2015 Guessing minutes URL: http://www.w3.org/2015/02/09-aria-apg-minutes.html People with action items: james jnurthen WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]