W3C

- DRAFT -

ARIA and Assistive Technologies Community Group

27 May 2020

Attendees

Present
Matt_King, zcorpan, michael_fairchild, jongund, isaacdurazo, shimizuyohta, Jemma, s3ththompson, Joe_Humbert, rob-fentress, simon, spectranaut
Regrets
Chair
Matt King
Scribe
michael_fairchild

Contents


Test writing update

<scribe> scribe: michael_fairchild

<siri> how to see shared screen

Matt_King: I could not get combobox ready for the pilot
... I changed the title to let people know not to test it
... Once we start a cycle, we can not change the test plan
... good news: I feel great about both checkbox and menubar
... there are now 40 tests for menubar, because I made a version for every test that applies to VO
... I also hacked the harness so that it does not mention anything about mode for VO tests
... I also made a lot of grammar and editorial changes

Pilot overview

<siri> How to see the screen?

<siri> Can you send me link?

https://bluejeans.com/678068494

(Discussion about sharing screen)

(discussion about recording the demo - we were unable to record, but we do have instructions that can be shared)

<zcorpan> https://github.com/w3c/aria-at/issues/162

zcorpan: first step: accept invite to the GitHub team
... second step, go to the site and click sign in with github
... next step, is to go to settings and select the AT that you use, then save.
... initially, nothing is selected
... there will be no responsive after clicking save, but it works. this is a known bug.
... next, navigate to the test queue page
... there will be a table with test cycles. There is only once cycle called 'test pilot'
... click 'open test cycle'
... next, assign yourself to the tests
... since we removed combobox, we might want to revisit assignments
... At the moment, we need to use the admin interface to assign more then two people. Up to two people can self-assign.
... step 8 is to click the link to the test plan. It becomes a link after you are assigned.
... when you click that, the first test for the test plan will open
... now, follow the instructions for the test. All fields are required.
... there is a button 'open test page'.

Matt_King: step 1 is always to ensure that your screen reader is in its default settings
... step 2, is to open the test page. If there is a setup script that runs, step 2 will describe this.
... for example, a specific element might be focused on page load
... If you are using NVDA or JAWS, step 3 will tell you which mode you need to be in.
... if you are using VO, we do not include instructions for mode

Rob-fentress: Should we include instructions in step 1 to remove plugins from the browser/AT?

michael_fairchild: yes, I think that's a good idea.

zcorpan: good idea, but I don't think we can make that change for the pilot tests

Matt_King: we should create an issue for that

(all) TamperMonkey (unsure of spelling) is an example, or a landmark extension which provides keyboard functionality to jump between landmarks, or even password managers which inject functionality into forms

siri: when should we raise an issue when running a test?

Matt_King: if there is a failing test, don't raise an issue. But if there is a problem with the test, or you don't understand the test, or you can't complete a test, raise an issue.

zcorpan: assuming you have followed the instructions, you fill out the results.

Matt_King: There are certain ways to copy screen reader output in some screen readers
... I think it is ctrl+option+shift+c to copy in VO, and in jaws it is insert+space,h in JAWS

michael_fairchild: I think it is preferable to use screen reader commands to copy actual output, otherwise manually type what you heard.

zcorpan: After filling in all fields, there is a button 'submit results'. If you missed something, you will get an error.
... after submitted you will get a success screen
... and you can move on to the next test in the plan
... You can navigate to the previous test, skip a test, and there are 3 buttons to raise an issue, redo a test, and (missed the last button name)
... when you raise an issue, you are prompted to describe your issue/question. This will help you create a github issue.
... redo test will reset the test results
... save and close will take you back to the list of test plans

siri: when I entered all the required fields, but when I say submit issue, I got an error saying that I can not continue.

spectranaut: thank you for reporting that issue, we did get that issue fixed

siri: and I can save and close at any time?

zcorpan: yes

Matt_King: what if you have partial data?

spectranaut: we do not have partial saving at the moment. We will soon, but not for the pilot tests

<Jemma> what is the deadline to finish the testing?

zcorpan: deadline is June 2

<zcorpan> http://aria-at.w3.org/

<jongund> https://github.com/w3c/aria-at/issues/162

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version (CVS log)
$Date: 2020/05/27 20:05:39 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision of Date 
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: Matt_King, zcorpan, michael_fairchild, jongund, isaacdurazo, shimizuyohta, Jemma, s3ththompson, Joe_Humbert, rob-fentress, simon
Present: Matt_King zcorpan michael_fairchild jongund isaacdurazo shimizuyohta Jemma s3ththompson Joe_Humbert rob-fentress simon spectranaut
Found Scribe: michael_fairchild
Inferring ScribeNick: michael_fairchild

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: 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]