- Minutes -
Education and Outreach Working Group Teleconference
12 June 2020

<shawn> scribe instrucitons -- https://www.w3.org/WAI/EO/wiki/Minute_Taking_Tips (may be a bit out date)

EOWG Charter

Shawn: Charter review is closed, 34 people completed the survey, very few comments, thanks everyone for the contributions. Replies to everyone still be sent. Current charter until 30 June 2020. It's good to go, there weren't any substantive changes.

Brent: Only one comment to change to a community group?

Shawn: The question was an interest group or working group. There were a couple comments preferring IG. It will stay for as WG for now.

Kevin: Are any implications shifting from an interest group to a WG

<shawn> https://www.w3.org/WAI/EO/wiki/EOWG_Charter_2020_Additional_Information#Analysis_of_Working_Group.2C_Interest_Group.2C_Community_Group

Shawn: Not significantly. See above what was written a few years ago. There might be a change in the next three years.

Brent: From a chair perspective, Sharron and I have thought about this. One point is how would content be perceived coming from an IG rather than a WG. Will it be seen as serious and official or authoratativeif it comes from an IG.
... We are concerned that if it is not a WG, that the content and resources that are created would not be authoritative or seen as important as things are perceived from WG.

<shawn> W3C Process on Groups is at https://www.w3.org/2019/Process-20190301/#GAGeneral

Shadi: In my view, it is more than just the members, it's actually a structural issue. Another group called "PING", they have a similar issue. There is a risk that if we become an IG, essentially, we may lose resources. Either the definition of IG needs to change or the definition for WG needs to change. It is much more than just a desir. There are structural issues.

Shawn: This is something that is acknowledged and I expect there may be changes to the group definitions in the near future.

Kris Anne: We are a very important group that makes things happen, make people understand better how to implement the guidelines. I struggle with the importance of this group, it is important.

Shawn: The value of the group is definitely valued. Every WG, except us, produce technical reports. The whole idea of a WG is that you have to go through patent policies, etc. and we don't. There is a thread on this. Two charters ago, there was whole bunch of stuff on this. Last charger (2017), we struggled alot. With this charter, we did not have any critique on our work. We got a lot of positive comments. As AC rep, in their mind, if they see WG,
... they feel there is a lot of legal stuff. People love the work of this group.

Shadi: It wasn't said that this should be an IG. [correction: at least 2 AC Rep preferred an Interest Group] It was just that this group doesn't match the definition of WG. We don't do W3C standards so this is where the question on the structure of WG applied to this group comes.

Shawn: Another point: in the past, we had been asked to clarify our name. This time, it was more justified because there are other groups doing education and outreach in other areas. We were asked to add "Accessibility" to our group. So, our name will change to Accessibility Education and Outreach Working Group -- still abbreviated (EOWG).

Brent: In IG, there is a lot of turnover compared with a WG. People in an IG come and go.
... we need a WG with consistent participants.

<Zakim> shadi, you wanted to explain Interest Groups

Shawn: There is a discussion on the type of groups and I am in this discussion. By 2023, there may be something different and something positive for us.

Shadi: I think we shouldn't wait until 2023. I think the members in this group should be involved. It is a public discussion. In answer to Brent's question, IG's were to incubate work, were to bring members together, have a discussion. They backfill things which don't fit in other groups. That is part of the discussion of redefining group structures. I really encourage organizations here to be engaging in the discussion. Shawn may come to you chime in
... best to funnel these thoughts.

<shawn> this issue is under discussion. Shawn will watch this discussion, and will encourage W3C members to engage as helpful

Shawn: There is movement to change the description of groups. I think we probably need something other than IG. I will definitely watch this. I will let you know. You and your AC rep are welcome to watch. And, when I think it is useful to have your voices, I will let you know.

Brent: Obviously, there is a lot of interest in this issue. It comes up frequently. So, thank you for the explanations, Shawn and Shadi. As and when you need our support, just bring us into the conversation.

ATAG Report Tool

Brent: A lot of great work on this.

Report Summary

<hdv> github: https://github.com/w3c/wai-atag-report-tool/issues/132#issuecomment-641955358

Hidde: A couple of changes since last time. Two specific changes: the report summary (above link), in particular, the results page. New section heading: Missing results (exp/collapse). Your feedback on this?

Helen: What's the difference between "not checked" and "missing results"?

<shawn> Shawn points Helen to https://atag-report-tool.hdv.now.sh/"Result Choices" section at the bottom

Hidde: Not checked (can be checked/has to be checked).

Helen: Maybe it's not clear enough

Shawn: the descriptions are on the first page.

<brentb> "not evaluated"

<shawn> "This check intentiaonlly left blank" ;-)

Kris Anne: Maybe someone doesn't understand how to check it.

<Howard> how about "ignored"

Kevin: Do we have a use case for which the existence of that serves?

Shadi: I think these values are coming from the WCAG tool. All the fields that are in scope should be marked as "not checked". Those which are not in scope should not be in missing. Under the missing results, there should only be two, the two that are not checked.

Hidde: The default value is the two dashes.

Shadi: The WCAG EM tool, the default is not checked. The ones that are not in scope could have the double dash. ..... (vick is lost)....

<Zakim> shadi, you wanted to ask about definition for "missing"

<kevin> s/..... (vick is lost)…./The default in ATAG Report Tool should be ‘Not checked’. If it is out of scope then it should default to null/

Hidde: at the moment, when you change conformance target, we don't change the... so this could result in some bugs.

(thanks, keven)

Hidde: what you see under the summary, this is what the person has selected.

Howard: I wanted to suggest the word "ignored"

Laura: Missing means that you have to do something. Skipped is clear, too.

Hidde: Maybe we need to change the drop-down value.

<Zakim> shawn, you wanted to say -- following along this thread "Not evaluated" since "checked" can mean passed and to say "not checked" could mean a checkbox that is deselected

Hidde: If we change "not checked" to "skip", would that be okay?

Shawn: If you say "not checked", then it could be ambiguous with a check-box that has not been selected. Maybe, it should be "not evaluated" to remove that ambiguity.

<Helen> +1 to Shawn's comment

<Zakim> Daniel, you wanted to say wording for "Results for these criteria are still missing:" may need some tweaking

<shawn> 1+ to delete that line

Daniel: I am also a little confused. If you expand the missing results, I think this doesn't add too much to the heading. I'm just wondering if we should remove this completely. I think also it is a bit confusing. It is not clear.

Hidde: Good point. I will remove the line as it doesn't add anything.

Shadi: I'm confused between the "not checked", "missing", and "skipped". I would suggest to have just one term. And it should be the heading of the exp/col. It needs to be clear that this is what has not been done.

Hidde: It's not the same. There are two different things happening. If we want to merge them, then, we have to merge them in the data.

<Zakim> shawn, you wanted to notes out option

Shawn: I get very frustrated when I select from a check-box, I change my mind, and then, I cannot undo this. For me, in the way I use check-boxes, I want to undo and I cannot.

Hidde: The double dash was included to address this. If someone wants to change their choice to nothing then they can change it to the double dash

Daniel: I see the point of not checked. So, my results are clear because I deliberately checked "not checked".

<shawn> [ Not checked -> Not evaluated. Missing results -> No results ]

Hidde: One other big difference is that the missing have links.

<Zakim> kevin, you wanted to say there isn’t any substantive difference between missing and not checked from a users perspective

<shadi> +1 to kevin

<Howard> +1 to kevin

<eoncins> +1 to Kevin users will get confused and results might not be consistent

<krisannekinney> +1 to kevin

Kevin: I think when you were describing the difference between "not checked" and a double dash, from a user's perspective, what is the difference. I don't understand what the double dash is for. I understand the problem of unselecting a check-box (as explained by Shawn). But I don't get the double dash. I don't see the point.

Hidde: It's very subtle.

Kevin: Perhaps, too subtle. It seems to cause confusion.

<shadi> +1 to howard

Howard: I think Kevin articulated it well. Either get rid of the double-dashes and make "not checked" the default or change the wording or functionality. Otherwise, it is really confusing.

Laura: I think I'm with Shawn. "Not evaluated" is clear. "Missing" means you missed something and you need to go back and fill it in.

Hidde: Would you say that there is a use case that you "skip" it.

Laura: If I'm evaluating to AA and I'm addressing AAA, then, I won't evaluate that.

Hidde: I'm starting to be convinced with Kevin's argument.

<Zakim> shadi, you wanted to express concern about "--"

Shadi: The use case Hidde that you brought up. That is "cannot tell" so you pass it on to a colleague. We can discuss changing the name. The use case of Laura that would be "not applicable" and use case of Shawn of reversing an entry, the two dashes are confusing. "Not checked" is a default value.

<Zakim> shawn, you wanted to say I *really* like the out option; however, I will go with group preference and to say (and y'all don't have to listen to me yell at my computer when I get

Shawn: It's a null. I personally like having an option. However, I agree that this complicates it overall. I can make do.
... A lot of times, I will come to a drop-down, and then I cannot go back. I can live with the default. As long as I can put some stuff in the observations, I'm okay.

Hidde: If we get rid of the two dashes, we will have no results. The same confusion may exist here. For either, you can add an observation.

Howard: The use case I can think of: what if someone wants to do AAA, even though they selected AA. I guess there may be a use case for "come back to later". Or, maybe just get rid of the double dashes.

<shadi> +1 to brent

Brent: To me, it seems like we are calling a subset of these three different things. When you go to the pages, if you haven't done anything yet, you have this double dash. So, we have called it "double dash" and "missing". If it is going to be in a list of missing results, then in the drop-down, it should say "missing", in the results, it should say "missing". Because that is what it will be called in the report. The same name should be used in al[CUT]

<Laura> +1 to Brent

+1 to Brent

<Daniel> +1 to Brent

Eoncins: where are the cannot tell?

Hidde: Cannot tell .... (pse fill in later)

<shadi> +1 to estella

Estella: I see problems in consistency in the end. The more values you have, the more confusion created. If you have an open field, then, you can solve it that way. Now it is more clear than before but still I am a bit confused with all the values. Sometimes I may struggle with which value to select.

Hidde: Does the first page help or would you like the explanations in another place?

Estella: Then the double dash is an extra confusion
... there is a confusion also between organizations, how this is perceived and checked.

Hidde: I think we are getting consensus that "not checked" is not helpful. The double dashes shouldn't show up in the summary. We should come up with one name for these things.

<Howard> I see 'cannot tell' as useful.

<shawn> [ Shawn notes "Cannot tell" is an option in WCAG-EM report tool ]

Brent: I think I agree with what Estella is saying. When you get to a criteria, you will have a positive result. Or, you will say that this doesn't apply, so you will select "not applicable". I understand what Estella is saying. So do you choose, "not checked", "cannot tell" . I'm wondering whether "cannot tell" is adding confusion.

<shadi> qq+

Hidde: "cannot tell" : you've looked at it but there are certain issues with the authoring tool

<Zakim> shadi, you wanted to react to brentb

Shadi: There is a difference between "cannot tell" and "not checked". “Cannot tell” indicates that the status of the criteria cannot be determined by the assessor (whether automatic testing tool or real person). “Not checked” indicates that the criteria has not yet been assessed.
... we can use different labels. To the user, we can present whatever labels you think best.

Hidde: Issue is so many people are using it now to do evaluations.

<Zakim> shawn, you wanted to say +1 . "Not checked" [Not evaluated] - maybe include the number in the heading? And possibly under there, include a list of "Not in scope".

(thanks kevin)

Shawn: if the default is not checked, that's fine. I don't feel strongly about the name. New points: Under "not checked" it would be nice to have underneath "out of scope". I don't think it needs exp/col.

<brentb> +1 to Shawn

Shadi: This last bit about putting it under "not checked" but rather listing it separately. And also the "not checked" for, and underneath that the list. I would suggest not nesting it. What about the "cannot tell"s. Do these need to be highlighted?

Shawn: Then, I take back the listing of "out of scope". Good point made. Let's keep this simple.

<shadi> +1 to only "not checked"

Brent: This goes back to Kris-Anne's use case, if I had a list of what I missed, then, it's useful. This is the list where you can go back and check them.

<Zakim> shawn, you wanted to ask about visual design

Shawn: We were talking about some specifics from the wording. I would like to get people's input on the visual design. And, they are very interesting, Hidde.

Laura: I like the visual design. Clean and easy to understand.

<kevin> +1 to Laura, I like it :)

+1 to Laura

<Helen> +1 to Laura

<eoncins> +1 to Laura

<Mark> +1 to visual design

<Howard> +1

<shadi> +1

Shawn: I think it is very pretty. I do wonder about the pass/fail. It seems to get lost to me.

Laura: Maybe you get the font size one step up.

Shawn: works with bold or larger font.

Evaluation box

<hdv> github: https://github.com/w3c/wai-atag-report-tool/issues/60#issuecomment-642563974

Hidde: here there are a couple of different proposals. On the right, an evaluation box. The new thing is that you can see the progress, in percentage but visually represented as a bar. The difference that there is an exp/col in one of the designs. Question is: Do you want to have exp/col. Next question: Do you want the number linked. Next question: Do you prefer a single or double column display?

<hdv> a. Expand/collapse

Shawn: I thought that the decision last week was that the entire box would be hide-able.

<Howard> Would default to visible with option to minimize

Hidde: In our tool, it's a bit harder to do as there is a space reserved for this box.

<shawn> +1 for whole box being show/hide. -1 for a expand-collapse within the box

Hidde: Do you mean like a "hide" button?

Howard: I like way it is now but maybe I would have the details shown by default and then you can hide the details. I'll go with the consensus.

Laura: I agree with Howard. I wouldn't hide the whole box. On mobiles, what would happen, would it drop underneath. If you have 2 columns on a desk top, does it collapse on mobile?

Hidde: on very small screens, it still fits.

Laura: I have no strong preference but I still like the 2 columns a little bit better.

<Howard> also lean towards 2 columns - less real estate

Shadi: I don't feel strongly. What is the use case for showing/hiding the details? I heard several times the desire to hide the whole box. However, if you have the widget on the right hand side, I don't understand hiding just the details. I can see the use case to hide it completely and use the full width.

<Helen> +1 to the full width content when hiding the box

<Zakim> shawn, you wanted to remind that that box is optional anyway. being able to hide the whole box gives cleaner view

Shawn: Some people might want to remove the clutter of the box. Some will want full width, especially so they have more space for the Observations box.

Hidde: If we make the it wider, it might make it less readable.

Shawn: But it's a user option. You are giving users control of their environment. Maybe they need it/like it that way.

Hidde: It is tricky to build this option.

Shadi: It would be an important investment of time for future tools.

<shadi> +1 to Helen!

Helen: Although it might be seen as some as not essential, personally, I think it is important that it is there. But I think it is important to give the user the option to decide if it should stay there.

<shawn> +1 for having it there by default, and allowing the user to hide it

Hidde: Would you have a say on if it should be wider?

Helen: The only thing is if the observations can be dragged across. Often people want to resize their observations to add more text. The observations re-sizing is important for me. If people truly want it wider, they could change the resolution. The show/hide of the whole widget is important.

<shawn> +1 for needing to make the Observarions box bigger

Laura: I don't know how I feel. I understand the readability issue. Then, if you hide the entire tool, well... I don't feel strongly about it. When we are building a web site or a form, we do not collapse. My feeling is we should have some consistency in what we do.

Next question: is what about the links as they stand. Do we need to make this clearer in any way? I had thought of using the chevron.

Shawn: The other point is that those links are redundant with those at the top (and the bottom).

Brent: On Shawn's point, what does this do for a screen reader. If a screen reader is pulling up the links, does this make the information doubled, will it be confusing?

Shadi: This may be a bug in screen readers. Daniel > if the links are provided twice, are they repeated twice?

Daniel: If they are adjacent, I think they are not repeated. If there is text between, they are repeated and they are announced as many times as they appear but I will send a screen shot of a page I am looking at it.
... I don't think repeating them will be an issue.

<brentb> +1 to not indicate links

Hidde: So, it seems we don't need to indicate the links.

Helen: Technically, I think that they are okay. They are shown in a navigational panel. I did like the chevrons (arrows). It's nice to know that you have other alternatives.

<shawn> [ to me, the chevrons do not signal links here, they signal twistie ]

Shadi: I don't feel strongly about the chevrons. Some sites use chevrons as exp/col. At a glance, it looked as col/exp.

<brentb> +1 to keep progress bars

Helen: Maybe the confusion is the thick line, the progress bar, under the link.

<shadi> +1 to keep progress bars

<Laura> +1 to keep the progress bars

+1 to keep progress bar

Helen: As long as the design is consistent, there is no issue.

<brentb> +1 to no chevrons

<Daniel> Would hurt if you'd put a label to the aside adding and id to the heading and referencing with aria-labelledby, anyway

Shadi: I like it the way it is without the chevrons. I like the progress bar, too.

<eoncins> +1 to progress bar

<Daniel> Would hurt/Would not hurt

Shawn: Brent has to leave. So, to conclude on this: we don't need other indication of links.
Columns?

<shawn> +1 to 2 column

<Helen> +1 to 2 cols

<eoncins> +1 to 2 column

+1 to 2 columns

<kevin> +1 to 2 column

<Mark> +1 to 2 cols

<Laura> +1 to 2 cols

Shawn: Okay, we have consensus for 2 columns.

<brentb> +1 to 2 columns

Shawn: What about to have one of the cool progress bars for the overall progress?

<shadi> +1 to overall progress bar

Hidde. Already done. What do yo think?

<kevin> +0.5 to overall progress bar

+1 to overall progress bar

<eoncins> +0.5

<Helen> +1 to the progress bar

<shawn> +0.5 to overall progress bar - kinda nice, yet more clutter

Hidde: When I listen to audio books, I like to see the chapter progress as well as in the whole book.

<Laura> +1 to overall progress bar

<shadi> +1 to kevin

Shadi: Should we address issue 144? Should we have text on the start page and where?

start page

<hdv> github: https://github.com/w3c/wai-atag-report-tool/issues/144

<shawn> +1 to keeping the box consistent. those instructions are not needed

Kevin: Don't feel strongly, but probably under the heading.

Laura: There should be consistency. Same display as the Overview page.

<shawn> +1 to Laura

<eoncins> +1 to LAura

Hidde: There is strong consensus to remove the text.

Shawn: Further, the text is redundant.

Work for this week

Shawn: Work for this week: Survey, review of the curricula. The TF has been working on it. Please schedule some time for that next week.