Re: R&R Update

I can't remember the status of https://www.w3.org/WAI/EO/wiki/Role_definition_document

What stage is it in? How much has EOWG reviewed it?

I think EOWG did some review, and we also wanted to get even more international input?

Do you want to call it down for now, and maybe refine as you learn things throughout the project? Did we want EOWG to do thorough review and sign-off, or not ready yet?

Is there benefit of getting the wider review -- beyond EOWG -- now? or wait for later?

~Shawn



On 10/5/2018 7:20 AM, Denis Boudreau wrote:
> Good morning folks,
> 
> Adding Bill and Sean to this email so that they can see the progress report as well.
> 
> We are slowly making progress. We were scheduled to meet today, but I got slammed with last minute client training and I won’t be able to meet with them today as planned. This past month has been completely out of control, with lots of work, so my ability to volunteer and contribute to the R&R project has been very limited. But I expect things to slow down a bit as of next week, so I’m hoping to catch up a little. Soon. Story of my life.
> 
> Here’s where we are right now: we started reviewing our checkpoints based on our decision tree models, to determine who should be the primary, secondary or tertiary owner of each checkpoint. This is how we will get the role based profiles to bubble up.
> 
> While doing so, we are reviewing the checkpoints and testing our logics. We decided to break our checkpoints by content types, so we could deliver the results in smaller, more manageable chunks to the working group. We have completed our first content type following this approach, and we’re about ready to share that with the group. We decided to focus on images first, and that amounts to about 24 different checkpoints.
> 
> We have a total of 10-12 such content types, and well over 300 such checkpoints to go through for WCAG 2.0 level AA. There are a few more for AAA, and then there would of course be additional ones for 2.1 - and while those are currently not within our scope, obviously, as were making progress and as Deque defines its methodology to assess 2.1, I expect to bring those into the fold as well.
> 
> The challenge we face right now is that by going from one contentvtype to the other, we are no longer focused on coming up with a list for UX, but instead, we are focused on getting our raw data in place, so the checklists by roles naturally emerge. I actually think it’s a better approach, but that means that we’ll need more time before we come up with our pilot. It’s all going in the right direction though, because we’re defining an approach that is based on a method as opposed to informed gut feelings. By using these decision trees to categorize each checkpoint, we’re following a rigorous process which we feel will give even more credibility to our work. But it does add to the workload. We feel like this is fine, because we are in for the long run.
> 
> Our plan right now is to better define the decision trees so we can share that with the working group. We will use the images content type as our proof on concept for the use of the methodology to define checklists, and we will soon share that with the group through GitHub.
> 
> We figured that it would be best to iterate with 1/10th or 1/12th of our checklist at a time, supported by those decision trees, and slowly have the checklists take shape in front of everyone’s eyes, instead of coming up with a list out of thin air and ask everyone to trust us with our breakdown. We expect that each content type will bring its own share of comments and questions through GitHub, and that it will be more manageable for everyone if we break it down this way instead of waiting until we’ve covered all 200+ checkpoints we’ll be using to inform our process.
> 
> Doing it this way instead informs on our process and we feel the decisions we made for each role are going to be easier to understand for everyone, and consequently, challenge If anyone feels differently about any of it.
> 
> I am hoping that by the end of the month, we will have made enough progress to share this first round of checkpoints. Doing so will also allow us to create partial lists for each role, and as I said, these will eventually take shape over time as we share additional content types with the group.
> 
> I hope this makes sense. This is our “small update” for now. I meant to share this with you all on the call today, but unfortunately, I will be training from 8:39 to 11am today, and then just catching up on my regularly scheduled work after that because none of this training I had to ta joe this week was even on my calendar when the week started on Monday.
> 
> Take care everyone and talk soon. Now that some hard deadlines have passed, including CSUN’s, I’m fairly confident that I will have more time to return to this work soon.
> 
> /Denis
> 
> 
> 
> 
> On Thu, Oct 4, 2018 at 13:25 Bakken, Brent <brent.bakken@pearson.com <mailto:brent.bakken@pearson.com>> wrote:
> 
>     Hi Denis,
> 
>     I know you are meeting on Fridays with your small team. Any updates on progress for Sharron and me? A short summary would be sufficient.
>     We are wondering if your team would have anything that you would want a group of people to look over and provide feedback on at the F2F at TPAC. If so, we are happy to add it to the agenda and we could put in on the afternoon of the 23rd so any of you interested could call in and discuss/listen.
> 
>     Let us know what you think,
>     Brent
> 
> 
>     *__*
> 
>     Brent A. Bakken
>     Director, Accessibility Strategy & Education Services
>     Psychometrics & Testing Services
>     *Pearson*
> 
>     512 202 1087
>     brent.bakken@pearson.com <mailto:brent.bakken@pearson.com>
> 
> 
> 
> 
> -- 
> /Denis
> 
> 
> --
> Denis Boudreau
> 514-730-9168
> 

Received on Friday, 5 October 2018 17:49:42 UTC