Meeting minutes
Chuck: House keeping, anyone new to the call?
Chuck polls chair and group for announcements
Items https://github.com/w3c/wcag3/discussions/301 and definitions scratchpad
Chuck: Start with 301 discussion
Rachael shares screen
Where we left off last week, was discussing UI Unit or UI context got us closer to consensus, but we did not conclude conversion.
Rachael: we need something between interactive and static...
Please weigh in on GitHub thread
… We need something which can be used for things like image or words in a sentence or paragraph...
Components test at different granularity.
We also need some backward compatibility with terms and concepts from WCAG 2.
Chuck: Are we hoping to finalize terms?
<Zakim> Graham, you wanted to say, unit examples are very varied in size, is it test / method specific that the word changes
Rachael: We need shared terminology and only change if needed, but still in the middle of process, and can return to this problem if needed.
Graham: Could terms change depending on contest? For example media player has lots of parts to it.
… I agree we are processing on list of terms.
Rachael: Media player is good example of content versus UI component and need for nesting.
[scribe misses Chucks comment]
<Chuck> Chuck: Does this mean that the definition of "unit" depends on the guidance/method/test, and could change when another guidance/method/test is defined?
giacomo-petri: Agree that we are on correct track, term like unit will mean something different for menu item for example.
<Zakim> Graham, you wanted to say, could we add examples within each method / guidance part then perhaps?
<Jennie_Delisi> * Each requirement defines unit for that specific requirement.
Rachael: I think we can terms defined while having different implications for different methods or contexts.
Graham: Going word by word in one context makes sense, but that would not make sense for a media player.
[rachael making notes in Google doc]
Rachael: We can get something close-ish today, sub groups can start working with new vocabulary.
Chuck: Graham mentioned that examples will be very helpful, but are examples part of the working document and intended as part of definition?
Graham: If we are having flexibility, it gets tricky for subgroup working on things like media player.
Chuck: I agree that examples will help inform our decision.
Chuck: No one else in queue. So is everyone on call comfortable with working definition for today?
<sarahhorton> Have we looked at Atomic Design for inspiration? https://
Rachael: If we have UI Unit do we need Unit?
Graham: Doe we have have vocabulary to talk about components and buttons?
Graham: Doe we have have vocabulary to talk about components and buttons?
bruce_bailey: We still need to define "part".
<Zakim> bruce_bailey, you wanted to say we need part
<Graham> could we not say "element" (as that is pretty universal) for things like buttons. The lowest technology native part essentially and then "unit" becomes smallest testable part in each context?
bruce_bailey: The smallest testable piece is a button. In a media player, yes, we need nesting terms.
giacomo-petri: Talking about one of the examples, drop down menus, unit could be entire menu, collection of units within menu column contain, but also one atomic unit in the menu.
… If intention is to test container, we we need to be able to talk about pieces of the container.
GreggVan: The only thing that will be viable is whatever is being tested...
The navigation bar or skip link is going to be part of the page. All of the parts could be unit.
… Hopeless to say sub unit, sub sub unit, etc.
Poll: Does the group support continuing forward with development of the terms "Unit" and "Process" found on this definitions scratchpad: https://
Chuck: Rachel is capturing notes, do we have enough in the doc to let people go forward?
GreggVan: Lets call it something like evaluation unit or something that is not a plain english word...
… we could have Evaluation Unit or evalu-unit but it needs to be unique...
… a button could be both unit and a UI Unit and it will be confusing.
<mbgower> "unit" seems relatively safe to me
<Zakim> joryc, you wanted to say smallest testable piece of content will always be any tag, right? Should we discuss unit as the scope of a test?
Rachael: I am okay with trying to use a particular term, and that can be an option for the poll.
joryc: I struggle a bit with small testable piece of content, maybe the small testable unit for the assertion or test?
<mbgower> +1
Detlev: I think it acceptable to use "unit" without any qualifier because the the context will be self explanatory...
<sarahhorton> I like the proposal of something like "evaluation unit", which removes ambiguity
Detlev: Rachael had example with media player, it is no problem to talk about audio description even when the concern is the media player.
… It is not a problem in really life.
GreggVan: It is okay to use word in common usage, but we should not use glossary terms which we feel compelled to clarify with our definitions.
<Detlev> @gregg offen, the same lexical term has a range of meanings in different contexts - that is completely normal.
GreggVan: We can use standard words, so long as we use standard definition....
<Detlev> ...and there will be several definitions even of 'unit' in a dictionary...
GreggVan: With regard to menu example, we call talk about choices in a drop down, and even characters in the words in a drop down choice.
<Laura_Carlson> Merriam-webster definition for "unit": https://
GreggVan: Example should include both menu item and characters in a menu item.
<Detlev> ..so even Webster has more than a handful of definitions depending on context!
<Zakim> mbgower, you wanted to say maybe "against which an assertion can be made"? A unit may still have attributes we would want to inspect, but we would report against a unit (at most)
mbgower: I think it is okay to use a common term in our glossary, we can make it link (or not) for the TR doc.
mbgower: I thought we were looking for a term which we don't go below?
Poll: 1) Unit, 2) Evaluation unit, 3) User Interface Unit
<Zakim> Chuck, you wanted to ask for the Poll: 1) Unit, 2) Evaluation unit, 3) User Interface Unit
<GreggVan> 2
<Francis_Storr> agree with the Mike's "composite" comment for media player. Related: see ARIA's composite role: https://
<Graham> 1,2,3
2, 1, 3
<mbgower> 1
<sarahhorton> 2
<Azlan> 1
<joryc> 2,3,1
<Jennie_Delisi> 1, 2
<Rachael> 0
<Makoto6> 2
<Detlev> 1,3,2
<CHall> 2,1,3
<MJ> 2
<ShawnT> 2,1,3
<filippo-zorzi> 2,1,3
<Francis_Storr> 3,1,2
<Laura_Carlson> 2, 1 ,3
<jtoles> 2, 1, 3
<mbgower> (but I can live with any of them)
chuck: As chair, I will have trouble with live tally, but looks like 2 is more popular
Rachael: I agree that 2 has slight edge. Any objects to using Evaluation Unit for now?
<CHall> as a literal thinker, I like evaluation unit because it's specific (we're going to test that)
<Zakim> GreggVan, you wanted to say 1) if you want to have this adopted in any regulatory 2) to make editorial comment. Only the definition should be
Detlev: +1 to mike, I think it may be more contract to have composite term.
GreggVan: For an SDO, we can't use "Unit" as is being proposed.
… As everything is composite, even button has aspects (for example active/inactive area) which are important
… Also, important that definition can replace term -- and everything else is a note
<Francis_Storr> https://
Francis_Storr: Disagree that concept of composite is problematic, as it works for context.
<mbgower> I was just thinking the same thing, Giacomo. Isn't the thing being assessed determined by the outcome being assessed?
giacomo-petri: Maybe we can clarify unit within scope of test. For example an image of text would have different meaning for "unit" but that is okay.
<Zakim> Chuck, you wanted to say I think we need to start somewhere to help in the crafting of the guidelines
GreggVan: +1 to giacomo. My point is that when we are talking about all of them at once. Within get the provision, we should not use "the evaluation unit is xxxx (for this requirement)".
Chuck: CHO we need a term while we are doing drafting. Maybe once we have drafting mature, we might find we can leave definition out.
Rachael: +1 to chuck. Do we need Unit if we have Evaluation Unit?
giacomo-petri: the part which is conceived by the user as whole is different from what a provision might be referring to...
… if native HTML, we talking about elements -- and that is meaningful to author -- but not the end user.
GreggVan: +1 to giacomo-petri -- We cannot reference "perceive by the user" because that is an unknown. We might write "intended to be perceived as a unit by the user"
Graham: Could we scope to technology? In HTML the smallest unit would be tag or element. For an audio program it might be different. For document it might be paragraph.
GreggVan: What is the purpose for defining the smallest element of a user interface? Where we use it?
… We need scope for a provision, but otherwise we don't need "smallest element".
Graham: It gives the a ground to for building a a definition which is structured. For web units build up to be a page.
… Media player could be a combination of units in a way which make sense to talk about them.
GreggVan: 1st we are not going to be able to name all of the technologies, because technologies evolve.
… On a provision, in the provision, we must write what you must do and what you do to it. The bits and pieces can't be part of that, because it won't make sense as linked defined term.
Francis_Storr: It is good to be talking about what the user is going to perceive as unit, design or intended as unit is important distinction.
<Zakim> Chuck, you wanted to summarize
Francis_Storr: Going back to media player, consider drag bar or volume control, it makes sense for each of those to be considered as a component for a provision.
Chuck: CHO -- I understand Gregg's concern but I still think we need a term to help with sub group work...
<Zakim> GreggVan, you wanted to say agree and std practice
Chuck: Chairs will flush out examples and come back to group next week.
GreggVan: What you propose is fine and is common with SDO work. Can come back to end and see if terms are not used or not needed because they can be deleted at end...
Rachael: What to poll group on how to move forward.
<GreggVan> +1 to moving on and getting experience using them before further discussion
Rachael: If people are comfortable going forward where we are, we might not need to poll.
<mbgower> +1 to moving on and getting experience using them before further discussion
<Rachael> Option 1) Bring it back for more conversation next week or 2) Continue the discussion on GitHub and bring it back only when needed
chuck: We either continue on live calls or work continue just in git hub.
<GreggVan> 2
<sarahhorton> 2
2
<ShawnT> 2
<Jennie_Delisi> 2
<giacomo-petri> 2
<Makoto6> 2
<filippo-zorzi> 2
<Detlev> 2
<Azlan> 2
<Rachael> 2 (chair hat off)
<LenB> 2
<Graham> 1 (just toi be awkward, actually 2)
<Francis_Storr> 1
Chuck: Seeing lots of votes for 2, but please note this only works if people contribute offline
Francis: I think bring back to meeting has some utility, but okay with 2.
kenneth: One factor is people who have trouble with GitHub or are not on call today.
GreggVan: If you sign up on GitHub, replies can be in email.
GreggVan: Problem is finding issue.
… find the issue and comment.
bruce_bailey: Gregg has been an individual who replies via email, but was asked to respond in the discussion. Using email is aspirational.
Gregg: If you do sign up, you do get an email, and you can use the link and fill out the comment in the discussion.
<Zakim> bruce_bailey, you wanted to say GitHub via email is scrambled.
<Zakim> Rachael, you wanted to say we can move this to our orientation calls
Rachael: I will add to orientation call.
Chuck: First meeting of the month starts early for anyone looking for orientation.
Subgroup work
Rachael: We have another GitHub discussion process.
Rachael: We will continue discussion on the GitHub issue thread.
… Still unresolved for view whole view part of view
<kenneth> Process discussion I think is https://
<Zakim> Makoto, you wanted to ask co-chairs to let me have an opportunity to ask a question from our subgroup to check consensus as the whole group
Makoto6: My subgroup is stuck on this and can't go forward without some additional clarification....
<Makoto6> "Method list" section in the Guideline template https://
Makoto6: The template is not working for us. We are trying focus appearance, and even with Alastair helping, we have hit a dead end.
<Makoto6> Text Appearance
Makoto6: it the text appearance template, which we also tried to work from, still has subgroup stuck.
Question can we change the sections and subsections on our own?
… It would be better for sub groups to follow template , but we need to customize template for our use to keep work. Subgroup asks for help with that. Frankie, Laura?
<Chuck> +1
Rachael: Chairs talking about this directly before this meeting. Does not help you today, use template as you have it. We will talk more about this next week,
<Laura_Carlson> Thank you.