mary jo: Now that we have our spec published, we need to make sure that people are looking at it. Wilco, have we sent out any emails to any groups orgs or people besides W3
mary jo: We want to make sure that it is reviewed and make sure we have a good version for publication and that people are trying to use it and that it works for them
mary jo: We want to make sure we reach out and find contacts to reach out to for publicizing the spec
wilco: I think we had a list last time
mary jo: Could have been a wiki?
mary jo: If not in a wiki, it may be in previous meeting minutes
wilco: Another things we could do is go see who gave us feedback last time around
<shadi> https://www.w3.org/WAI/GL/task-forces/conformance-testing/wiki/List_of_related_groups
shadi: Link above, we have a small page on the wiki, but I think we had something more comprehensive at some point
wilco: That is still a useful list.
shadi: There is also the aria AT group
Anne: What about the IAAP community?
shadi: Who there? Do they have an active list?
mary jo: I am a member, and know some folks higher up in that org
wilco: Are you suggesting we get into the news letter?
anne: Yes, and I don't htink it will be so hard.
maryjo: We can also point them to the W3 newsletter as well
wilco: Alright, so lets create a list and devise some tasks.
anne: If we have specific orgs we would like to invite, how do we do that? Any shared communication or just reach out individually?
wilco: Just reach out individually
... I can do aria AT, web aim.
shadi: Can send to john gunderson, he may be interested? I can take that
wilco: I will reach out to Annaca as well
... I don't think we got back from spec ops or test infrastructure group
shadi: I can resend to test infrastructure group.
wilco: Who would like to reach out to IAAP?
mary jo: I can do IAAP
wilco: I need to reach out to tool developers as well
anne: I would suggest we reach out to the groups in the white tools project
... Maybe also getting some consultants for feedback
... Will reach out to my contacts here in denmark
alistair: We could reach out to the testing symposium
trevor: Trusted tester, Kathy Eng is a member of this group and part of that.
wilco: Should also get some of the large tech companies like microsoft and amazon
shadi: Should we try to reach out by early next week.
wilco: *Reads of list of assignments*
anne: There are a number of gov bodies starting to worry about accesssibility, should we contact them
wilco: Should we talk about this in the meeting at brussels
shadi: Yeah we can, they may not be as technical
wilco: Still sharing with them will be useful
mary jo: Wilco do you know the answer to that
wilco: So anne, myself, Jay, and carlos have been looking at updating the ACT TR rules
... There are some actions for me and Jay to make some structural changes. I have updated the templates.
... Anne has some people working on updating the existing published rules.
anne: It has been pushed one day farther so we are hoping to start working tomorrow.
... Has the new template been updated?
wilco: If it hasn't, it will be very soon.
anne: We are expecting it will only take a few days
mary jo: Rules are going to be developed, almost there.
allistair: The question I still have, and sent it in a summarized version of the list. After looking at the ACT rules, we found some of them to be advisory over testing pass/fail
alistair: I don't think we are going to implement all of them, at least not the once not within WCAG conformance. We need to justify every single test
shadi: I don't know if everyone has to agree, not everyone agree on WCAG.
alistair: There are still things that are out of scope, like the aria stuff. Why would we implement that?
... There are things with aria. I can put you in touch with someone that can walk through why these things are advisory and not conformance.
... Why would we spend any time on things that are questionably advisory instead of just rules for conformance.
shadi: I think this is a question for the group as what should be accepted. But what one person thinks is conformance may not be shared by others.
alistair: The example I used isn't arguable. The ACT rules also mean that anyone can implement the rules, and thus may have some incorrect rules
shadi: We will have some gatekeeper to decide which rules to accept. There is some debate on who the gatekeeper. We could add disclaimer that some of the rules are not finalized
... I think this group is very good at getting people to be transparent and put out how they are implementing these rules. The more people that get involved the more people we get in this group the better we get.
alistair: I am just trying to not give people undue burden. We need to worry about scope creep
... There are tools implementing these rules and eventually goes back to W3
... We need to make sure that like the EU monitoring is within the conformance scope.
shadi: I think we are getting a bit in to the weeds, so I think we can take it offline.
wilco: So I think this is actually a sign of this working the way its intended. Due to the transparency we are now seeing the new opinions on the rules that may not have been thought to be controversial.
... It allows us to take the feedback and go rework our own rules.
shadi: I think the issue, is that the fact these rules are coming from a w3C project, and we know these are not normative or widely approved, others may not have that perception.
... We need more clarification on this issue. We should look at how the rules are presented. That they are how this project is presenting them, but there is no approval state.
mary jo: Perhaps some state notification for the rules
alistair: They have already been implemented by the project, which is some sign of approval.
wilco: Part of our ACT work is figuring out what we agree on and not.
alistair: This brings us to the next item about what the group's future would be. Seems like need industry experts to give the "8 our of 10" experts agree with x
shadI: It is likely we would have a similar structure as others were a group of 3 or 4 agree and then send it to the working group.
... would be better to have a strong community. Save the working group from many feedback rounds
... My hope is that there would be multiple groups. The more people involved in these communitites the better we will be
... Going back, do you want to raise an issue on this?
alsitair: Yes, the question still is why we would put effort into the advisory rules, when there are other conformance rules to be doing.
wilco: I do believe in these rules, they are used by millions and have been used for some amount of time.
mary jo: Can I make a suggestion? Before you go into the writing a rule, have a quick proposal and have the group agree on if it is a conformance rule or just more advisory
anne: We have lately implemented such a phase.
... What we have found is that it is hard to get opinions on it in such an early phase.
shadi: Perhaps we could have a time like once a month where we update the working group on which rules are currently being proposed. It may get noisy
alistair: The problem is that alll of that front upload makes it very time costly. Hopefully you will get some accordance between the players. It needs to be done at a higher filter level, but you don't want to stop people bring in tests.
... The problem is, instead of just a few players disagreeing with, if you can see problems in court, you would have wanted the rule to go through all those filter phases.
shadi: It is all being done in public, and the earlier you bring in your disagreement the better. The published rules will still have to go through a filtering process.
wilco: I would rather have ore rules even if they are controversial. But I think it helps with transparency a lot. Not every rule needs to be published. There are developers with varying degrees of tolerance to false positivies. Just getting insight into who is doing what is very helpful.
... I would rather know that, have people make those proposals, and see where they land instead of pushing them away initially.
alistair: As a company we look at testing for conformance unless the customer wants to run the other lovely tests, but knowing they are above WCAG conformance.
wilco: We run these rules because they are indicators of bad code, and possible conformance problems.
alistair: So they shouldn't be in an exercise looking for conformance.
wilco: Probably not, there are assumptions made in those rules.
mary jo: They are more a unit test type rule.
alistair: Some of them are down to bad code. Which isn't really for conformance. If you put it into official audit, and they start wacking fines on people, and those people disagree because it isn't conformane you are going to bet blowback
wilco: No disagreement. I think we disagree on if that rule in practice does find conformance issue. I can see the argument of why it is not 100% inline
shadi: I think this is addressed by clarifying the authoritativeness of these rules in a better way.
wilco: I don't think we have that much control over things. If monitoring bodies aren't fine making assumptions because they may be improperly fining people they shouldn't use the rules with assumptions.
alistair: By implementing the rules you have chosen, you should be taking those responsibilities off of them.
wilco: I don't htink that is the task of the implement or the ACT rules community.
mary jo: The issue is, we are considered the expert, and if the eu is tyring to levy fines using our tools, they aren't experts and they won't know if the tool is using advisory techniques to test. They will rely on us to do the right level of conformance testing. It needs to be clear what the level of rules to test with is
<maryjom> https://github.com/w3c/wcag-act/issues/353
mary jo: The working group will be looking at how rules will be published and incorporated into the materials. How they will make it into the techniques document
mary jo: I don't know if everyone has had a chance to look at the process, did anyone have any further points of discussion on this?
wilco: I wanted to ask Shadi what he would like to do with the comment he put in.
shadi: So basically, I broke down the tasks that Wilco laid out into more specific sub tasks. What does each mean. I hoped to address some of alistairs inputs on that. So its a smaller breakdown, but doesn't suggest which one.
... From breaking it down, it does look like there is some substantial work to be done.
... I am leaning towards needing an ongoing filter group
wilco: I think that is probably right. There is just a lot and I think to alistair's point we will probably need a group to pre-sort. I think newer groups will need some hand-holding and get them set up
alistair: There is also the idea of making some notional idea of the scope of the rules.
... We talked about how we need to maintain scope over a period of time so that scope doesn't change
wilco: Yes, I think we would need something like that. When you say ICT baseline, are you talking about ICT testing baseline?
alistair: Yes, the one used by trusted tester.
mary jo: I think we can continue this next week as well, and refine this for presentation to the ag working group.