/invite zakim #aria-at

Introductions

<Scott_O> I can hear you

Jemma: work for University of IL extension web manager. Co-facilitator in APG Task force.
... Motivated by difficulty working with A/T as a developer.

Joe H: Work for TPG as developer.

scribe: Motivated by desire to understand how A/T supports ARIA after a talk made at Accessing Higher Ground

Jon G: University of IL, disability services, member of APG TF

scribe: Interested in how much of the work on examples for APG actually works and want to feedback to ARIA WG and how it impacts users.

Kathy M: Arizona State U, was colleage of Jon at U of IL, now a11y specialist at ASU.

scribe: I want to just give back how I can.

Scott o: Work for TPG as a11y eng. Previously editor for html 5.3 and now on html AAM.

scribe: Have been doing a lot of this kind of testing and writing about it and want to share back to this group.
... Did some work on Michael's prototype.

Shadi: Work for W3C WAI. Based in Austria.

<shadi> https://www.w3.org/WAI/accessibility-support/

Shadi: Back in 2010 had a exploratory attempt at a similar project.
... Integrates with W3C systems.
... Supports creating and running tests.
... Invested heavily in prsentation of results.
... Was focused strongly on WCAG conformance.
... Related to concept of accessibility support in WCAG.
... All source code is available.
... Currently focused on a11y conformance testing.

<shadi> https://auto-wcag.github.io/auto-wcag/pages/rules.html

Shadi: Being done in auto-wcag CG.

<shadi> https://auto-wcag.github.io/auto-wcag/rules/SC4-1-2-aria-hidden-focus.html

Shadi: These are sme-automated and automated rules. Some related to ARIA.
... Example related to aria-hidden.
... These rules have test cases and some may be reusable.
... Happy to provide staff support.

Tzviya: Work for Wiley, publishing company. Was editor of digital publishing module of ARIA.
... Have heard some reluctance among A/T for uptake/adoption.
... This may be a good place to help resolve that.

Terrill: Work at University of WA, technology specialist, and do web development.
... Web devs are often not sure why there code does not work with A/T.
... Long been in search of a resource to help with this.
... Was part of a 2014 CSUN hackathon that was on a similar track to this group.
... Want to support an effort in any way I can.

Keith Bundy: Work with Site Improve as accessibility consultant and trainer.

scribe: Would like to help with further dev and improvement of ARIA.
... I have seen the good bad and ugly.

Corbb O'Connor: Over see accessibility testing at Site Improve.

scribe: Want to find better ways of explaining aria/js interaction

Elaine Rettger: Arizona State University, manage web site for non-academic side of university, e.g., benefits and fcilities.

scribe: Joined to support Kathy Marks. We work together and have a vested interest in making sure oru sites are accessible.

Background, objectives, and workstreams

https://github.com/w3c/aria-at/wiki/Background

Discussion

<jongund> MK: Any concerns about objectives

<jongund> MF: Do you want me to talk about the tool I put together

<michael_fairchild> https://a11ysupport.io/

<jongund> MF: This is a prototype that I built on y own time

<jongund> MF: The perspective of this tool is for the web developer, not an AT developer

<jongund> MF: It shows how a feature (e.g. aria attribute) performs versus a set of test cases

<jongund> MF: Depending on the command you use there may be different results and the tool helps document these variations

<jongund> MF: I want it to be reproducible, people fill out forms on the site

<jongund> MF: The summary looks at test results

<jongund> MF: Do this project has helped me understand the problem

<jongund> MF: We need to discuss what support means as a community

<jongund> MF: We need a process and methodology so people can reproduce and have confidence in the results

<jongund> MK: If people do have questions about the objects and scope, there alot of details that I did not talk about on purpose, because I want to get feedback from the community on priorities

<jongund> MK: Can we go around and get feedback

<jongund> Corb: What is the role for people who are not developers in each work stream

<jongund> MK: The majority of the work is not for developers, most of the work is project management, communications, quality assurance, do the testing

<jongund> MK: We want confidence in the process and how do we respond to feedback and disputes

<jongund> MK: the 3rd work stream on the how to build the test cases and building them out, is more on AT..

<jongund> MK: You may have the similar test conditions for different widgets so we can use instructions across multiple patterns

<jongund> MK: We have lots of work for non-developers, developers mostly need for infrastructure to hold test and test results

<jongund> Scott: How we will report the results, what does it mean to support and how to report bugs to AT developers

<jongund> MK: One of the primary outputs is bug reports

<jongund> MK: I am think about trying to get browser and AT developers in the bay area and look at different combinations of AT with a browser

<jongund> MK: We can generate bug reports, for example there is a bug reporting system for JAWS

<jongund> MK: We can submit defect bus and we would coordinate with them so they know when and where they are coming from

<jongund> MK: We need to coordinate with browser developers and AT developers to get stuff fixed

<jongund> Scott: we could also file bugs with specifications too

<jongund> MK: That is a result too, I have fixed bugs in the practices based on people creating issues

<jongund> MK: If we find more bugs in ARIA we can get them fixed

<jongund> T: Do you have plans to get AT developers involved in this group?

<jongund> MK: Yes I have had conversations with Glenn Gorden and other about participating in this group

<jongund> MK: Sometime people go AT developers and say they have a bug because it doesn't work on their site, so AT developers have a place to send people

<jongund> ??: What about Narrator testing?

<jongund> MK: That raises a great question on what are we testing, we do not want to limit the testing to just a subset, we want to get information on both desktop and mobile and if someone want to work on Narrator we would want a way to have that contributed

<jongund> MK: The 3 most popular screen readers are JAWS, VoiceOver and NVDA, so we want to make sure those are covered, but we also want VoiceOver on iOS and TalkBack on Android

<jongund> MF: This does bring up a good question about scope

<jongund> MK: I want to be open, but I have a clear priority of getting 10 patterns tested with the top screen readers, this is my personal goal

<jongund> MK: If you look at the WebAIM survey, we can get alot of coverage with those screen readers

<jongund> TT: I want to go back to the CSUN event, MK I think you were there

<jongund> MK: I might have been

<jongund> TT: There was alot of energy there, and JF bought pizza

<jongund> MK: I remember the room and the pizza, I think it was in San Diego

<jongund> TT: I don't think there is much left from that day, and the repo seems to be gone

<jongund> TT: that effort was about crowd sourcing the results, if the community would build the back end and the actual testing was users

<jongund> TT: You can get alot of versions of ATs and Browsers, so you would have more data

<jongund> TT: The problem is are users qualified to test, but if you have enough volume ...

<Scott_O> apologies everyone, I have to drop. will review the remaining minutes once they're posted. thank you again. and for those that may have been unaware, here's the link to JAWS bug tracker MK mentioned: https://github.com/FreedomScientific/VFO-standards-support

<jongund> TT: MF tool there already alot of things that are objectoves, rather than start over from scratch

<michael_fairchild> Shadi: crowed sourcing is good, but last time we didn't build strong community. This is a great opportunity to build a community.

<michael_fairchild> Shadi: Keeping scope small at the start would be a good.

<michael_fairchild> Shadi: we found that contradicting results were often very helpful. This sort of information often revealed the fact that many A/T users didn't know about modes or features in A/T.

<michael_fairchild> Shadi: We explored many different aspects of collecting this data, including many different events and venues. There was a lot of energy at in-person community events, and that is something we might want to do.

<michael_fairchild> MK: learning from failure points and past experiences is very helpful

<michael_fairchild> MK: the third workstream would hopefully address the problem of testers not understanding how to use A/T or what is expected for certain tests to pass.

next steps

<michael_fairchild> MK: One idea: One of the most important things is figuring out who is interested in leadership roles and establishing meeting times. Ideally an hour a week over the next few weeks. After we map out goals, we can involve more people.

<michael_fairchild> Shadi: interested in infrastructure and process

<michael_fairchild> MK: does anyone else want to be part of a core group that will figure out next steps for the process works stream?

<michael_fairchild> Terrill: I'm willing to help out

<michael_fairchild> MF: we could follow up via email and as for volunteers too

<michael_fairchild> MK: Michael and I can put together some specific asks and send to the list to find out who can help and when

<michael_fairchild> MK: Thank you all for your time and interest.

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/[[currently on WebEx and not yet spoken: Corbb O'Connor, Elaine Retteger, Keith Bundy, Keyonda Smith]]//
Default Present: shadi, mck, tzviya, Scott_O, JoeHumbert, michael_fairchild, jongund, jemma, Kathy, Marks, Terrill-thompson
Present: shadi mck tzviya Scott_O JoeHumbert michael_fairchild jongund jemma Kathy Marks Terrill-thompson
No ScribeNick specified.  Guessing ScribeNick: mck
Inferring Scribes: mck

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.)