TPAC

<jeanne> https://www.w3.org/2018/10/TPAC/

Jeanne: Lets talk TPAC (link above) . Email was sent out with lots of info about TPAC, registration is now open and for those that are interesting in attending there are 2 days of face to face meetings.

Days are October 22nd and 23rd in Lyon (France), lots of details on the W3C TPAC pages including hotels, airports, transport. Registration is open, there is a fee for each day. Normally a week long conference:

--Mon nad Tues (working group meeting)

--Wed (group meetings)

<Charles> Charles thinks it is not clear how CG members attend WG Plenary sessions on Wednesday

proposed Silver meeting on Wednesday

Charles: How do community group members fit into plenary sessions? Some sessions are "by invite of the chair only"?

Michael Cooper: Wednesday is called plenary, its for anybody really. A few presentations from management and then breakout sessions nominated by the participants

Charles: I'll be there Mon-Wed

Jeanne: similar

;)

Requirements survey

<jeanne> https://www.w3.org/2002/09/wbs/35422/silver-requirements/

Jeanne: Members of the Accessibility guideline working group (i.e. not community group), we have presented silver requirements to accessibility guidelines working group. It is a survey being presented so anyone that has access to this, please complete it. We would really like to have comments on it from the people that are familiar with it.
... Please do this before 11 EST today - that is when the AG Working Group meeting is

WCAG Structure document

https://github.com/mikecrabb/silver_guidelineAPI/tree/master/documentation

<jeanne> MikeCrabb: Start with the existing structure of WCAG as a basis of creating a new structure for Silver

<jeanne> ... there is the SC, the way you can test them, and ??

<jeanne> ... there are techniques and failures with examples

<jeanne> ... passing technique has a technology associated with it.

<jeanne> ... there can also be a situation where the SC applies

<jeanne> ... I have put it into a database and created an API

API Link: https://github.com/mikecrabb/silver_guidelineAPI

<jeanne> ... so we can get the information for each success criteria

<jeanne> ... it makes it easier to build the prototype

example Passing Technique call: http://35.178.161.137/passingTest/ARIA6

example for all success criteria: http://35.178.161.137/example/allSuccessCriteria

<jeanne> ... it gives a model of the data that we can use to start displaying the information

<jeanne> ... none of it models the guidelines, just how we meet the guidelines

<jeanne> ... there is a demo of it in github

<jeanne> MichaelC: I do want to be careful not to be so focused on WCAG2, that we end up basing all the Silver work on it.

<MichaelC> WCAG 2 JSON: https://rawgit.com/w3c/wai-wcag-quickref/gh-pages/_data/wcag21.json

<jeanne> ... I haven't had time to look at the JSON, but that link is an existing JSON file that drives the WCAG 2 QuickRef

<Charles> Agree with Michael C. This is very useful for comparative testing with proposed / prototype structures.

<jeanne> ... the SQL diagram is missing Advisory techniques

<jeanne> ... we classify all techniques as either Sufficient, Advisory, or Failures.

<jeanne> ... Failures should never have been called Techniques

<jeanne> ... some success criteria have situations, others don't. The situation SC can have different sets of Techniques based on the situation.

<jeanne> MikeCrabb: We shouldn't use this as a template for Silver, we should look for what information that people are after, and then use that information to inform SIlver

<jeanne> ... do we have log files?

<jeanne> MichaelC: There are log files for the W3C site, but I don't know if we would be able to use them.

<jeanne> ... there is some logging for WAI resources, but I don't know what it captures and whether we could have access to it.

<jeanne> ACTION: MichaelC to look into the possibility of log files on use of WCAG2 information

<trackbot> Created ACTION-193 - Look into the possibility of log files on use of wcag2 information [on Michael Cooper - due 2018-06-26].

<jeanne> Charles: Another handy datum would be the originating link

Jeanne: I'll spend some time looking at the API and how we can look at the information differently. I agree that we need to look at a different structure but this is cool!

exceptions: https://www.w3.org/TR/UNDERSTANDING-WCAG20/text-equiv-all.html

Jeanne: it would be great to get a tool that could look at different components and how they could be made accessibly, it would be difficult, but when we are looking at things like tagging it would be useful
... when Charles Hall was looking at word analysis for WCAG2, one of the things that I thought about was to capture examples of where the techniques apply. I think there is a lot more to do with this

MichaelC: I agree it would be useful, my concerns about this and implementability - we dont have full information for all widgets and it would be a lot of work to obtain it. Also concerns around exclusion by ommision. Anything that we dont know about could lack guidance. Tagging is good whre we have that information but at that granular level it would be difficult

Jeanne: Exclusion by omission is important, need to ensure that whatever we do we keep that in mind
... any other questions? Exciting to have a tool that means we can create prototypes that have data in them

Charles: I'll need some time to wrap my head around looking at the structure and looking at best path forward. One of the things that we've talked about is looking at comparative usability testing (take prototype of new structure and compare against old structure). Need to look at how we describe the deltas - what is the difference between this structure and current structure

Jeanne: I can see us looking at 1 prototype and another, but what is the importance of comparing to WCAG2?

Charles: e.g. parent child relationship of some categories. Existing structure has a clear relationship. If we reverse that or have a different structure or one that is not parent/child then we have 2 things to compare in the context of a task, but should we also describe the difference?

Jeanne: at an early stage, yes. But for the *last* one, it will be difficult. We woudl have to spend a lot of time if we wanted to continually compare to original WCAG. We're going to have a lot of differences so may not want to do that?
... Peter, any comments or impressions?

Pete_kennaugh: Having something that better describes the structure in WCAG is good and an interesting tool to have, will help when it comes to looking at how to put guidelines into a different format. By understanding this we will avoid repetition.

Jeanne: Also want to introduce Peter to Mike and to Charles. 3 people that are most interested and have expertise in the data design.
... please feel free to connect outside this meeting also

Charles: I also completed the list of contacts to look at PwD with guidelines

Jeanne: Great! Is it worth having a separate meeting about IA?

mikeCrabb: +1 yes!

pete_kennaugh +1

Jeanne: Timeline is to have a few prototypes in July, settling on one in September
... is this doable timeline wise? If not please let me know

Charles: do you mean that this is done enough to test?

Jeanne: testing in August and then end of september is the one that we want to propose to AGWG in October

Charles: What volume of prototypes makes that reasonable/unreasonable to do? 3? Sounds like a good number to test and talk about, but if we go beyond that it will really impact on the timeline

Jeanne: If we have multiple prototypes beyond 3, we should be looking at picking best features of each one based on requirements and making sure that when we get to user testing we have best 3, or best of the 3
... if this isnt realistic then we can look at it then
... anything else on IA?

mikeC: gitHub pull requests

Jeanne: have an action to deal with these
... do you want your accepted?

mikeCrabb: Yes, please. Just gives a bit more detail to pull requests

<Zakim> KimD, you wanted to ask if 2.1 was included

Kim: Does the prototype look at 2.0 or 2.1?

<jeanne> Kim: Does it include 2.1?

MikeCrabb: just 2.0. Took a deep approach to look at IA first then will work on getting data in after

<jeanne> chair: jeanne

This is scribe.perl Revision: 1.152  of Date: 2017/02/06 11:04:15  
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: KimD Charles jeanne MikeCrabb kirkwood MichaelC imelda Peter
Regrets: Jan Shawn
No ScribeNick specified.  Guessing ScribeNick: mikeCrabb
Inferring Scribes: mikeCrabb
Found Date: 19 Jun 2018
People with action items: michaelc

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.)