<scribe> scribe: michael_fairchild

overview of meetings with AT vendors

<isaacdurazo> https://docs.google.com/document/d/1rI8UlsG9gZr86sJjcf07a7c-a2u3mx8dIejUEKP3xe4/edit#

matt: we met with 2 of 3 screen reader developers that are on our list so far. Apple and Vispero
... we have not met with NVAaccess, Google, or Microsoft
... high level - Apple is on board. James Craig will review test plans, but had some concerns about our schedule (potential confects and events)
... they had some concerns about assertions and instructions

Isaac: they also suggested that we start with simple examples and work toward more complex

matt: feedback could result in us changing APG examples
... for example, the semantics of the group role and fieldset are not exactly the same. ARIA suggests that they are, but screen reader manufacturers don't agree.

group: oh

matt: freedom scientific is very much on board
... Eric even personally ran some of the tests and gave positive feedback
... (discussion of jaws history)
... freedom scientific wants to be involved in the publication of our results, because they want the opportunity to fix issues before the results are widely published.

valerie: was anything mentioned about the format of the report that needs work?

matt: there was some feedback, and I spent some time reviewing the report. There are some things can be improved, but we don't have a solution to yet.
... some other feedback: using the word 'nice' can also convey that a screen reader is doing something 'bad' if they are not doing it.
... we should simplify to 'must' and 'optional' instead of 3 priorities

valerie: I'll make an issue for that right now

matt: great
... any other feedback?

Isaac: freedom scientific thinks the chromium version of edge will become much more popular than firefox

matt: so they want us to test chrome and edge, not firefox

<spectranaut> michael: I think the amount of feedback and involvement AT developers want to have will lead us to work more slowly and iteratively on the test plan creation

<spectranaut> related issue: https://github.com/w3c/aria-at/issues/104

overview of project progress

valerie: we are starting work on the production runner. We are doing research this week, and will start development next week.
... we should give the runner application a cool name

matt: that sounds fun
... let's make an issue and put it on the agenda for next week
... so far Isaac has done some work on use cases. we expect to turn those into wire frames very soon. whats the time table on that?

Isaac: I'm estimated about a week. the wireframes will be done in HTML so that we can all review them.
... I'll probably wait until next week to start those wireframes so that everyone has a chance to provide feedback on the use cases

matt: in the next 3 weeks, we will have use case review, and two rounds of wire frame review
... ongoing at the same time, development of the production system will be started.

matt/valerie (discuss potential github integration into the production system)

matt: does anyone have any questions about status?

yohta: with regards to test running, are we still going to test the other examples?

test plan development, review, and execution

matt: valerie asked about test authoring in an issue

<spectranaut> issue: https://github.com/w3c/aria-at/issues/85

matt: my answer was to not have bocoup spend more time involved on helping us write tests
... but as we were working on the last tests, it became more obvious that a test authoring system would be very helpful. So it might be worth while to consider building such a solution.
... in the meantime, it would be good to figure out a process to use the spreadsheets

Jon: i'm working on a script in node to help with that, and it will have some better validation. Hopefully we will have this very soon.

matt: that's great! my thinking is for the near term we will use those csv files as the single source of the tests, and we will review them in the csv files, and when we think they are ready we can review them again once they are in the system.
... but if we are going to make changes, my suggestion for the near term is to make the csv files

jon: are we meeting next week because of CSUN

matt: I just found out that I'm not going

Jon: I'm not going

Michael: I'm not going

matt: so I think we will plan on having a meeting next week
... but we will find some other ways to broadcast news about this project
... we do want to get NVDA/FF results for checkbox done
... I think we should hold off on combobox and menubar for now
... and our assertions for grouping of checkboxes, I want to discuss

yohta: I did run NVDA/FF last weekend

matt: thank you!

use cases

<isaacdurazo> Issue link to use cases: https://github.com/w3c/aria-at/issues/102

Isaac: I worked on some different use cases. This is focused on the test runner. I conducted this research from having conversations with Valerie, Matt, and Michael
... actors = tests and admins
... (discuss document and test cases)
... this is the high level of the use cases

matt: this last use case is one where we haven't discussed what our process will be and what our roles will be
... we have always assumed that tests will be published right away, but after our discussion with freedom scientific, we may need another step for AT developers to review before final publishing
... so we might need to do more research and develop a process map
... and then I think we need to get feedback on that process from the AT developers

Isaac: I did have the idea of having a draft mode, where the admin can review before they are published

(discussion of data management and storage)

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: jongund Matt_King michael_fairchild spectranaut shimizuyohta isaacdurazo Erika_Miguel
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.)