14:46:55 RRSAgent has joined #ag 14:46:59 logging to https://www.w3.org/2023/03/21-ag-irc 14:46:59 rrsagent, make logs world 14:47:16 rrsagent, generate minutes 14:47:17 I have made the request to generate https://www.w3.org/2023/03/21-ag-minutes.html Chuck 14:47:24 chair: Chuck 14:47:24 Zakim, start meeting 14:47:24 RRSAgent, make logs Public 14:47:25 Meeting: AGWG Teleconference 14:47:25 meeting: AGWG-2023-03-21 14:47:46 agenda+ Google Focus appearance research presentation [30 minutes] 14:47:46 agenda+ WCAG 2.2 Focus appearance Survey: https://www.w3.org/2002/09/wbs/35422/focus-appearance/ 14:48:00 agenda+ WCAG 2.2 issues, Questions 1-5 Survey: https://www.w3.org/2002/09/wbs/35422/wcag22-misc5/ 14:48:25 agenda+ WCAG2ICT Survey: https://www.w3.org/2002/09/wbs/55145/WCAG2ICT-First-AGWG-review-4-sc [Last 30 minutes] 14:51:26 Alastair, Michael or Rachael, I can't get logged into the telecon info page, with the zoom link. If you have it, can you email it to me? 14:54:33 Can you send it to me as well? 14:55:04 will send direct. 14:55:17 AWK has joined #ag 14:55:49 OK, we should not send to list 14:55:54 so send them to IRC maybe, and then we can update the minutes and remove the link from the minutes after. 14:58:28 dshoukry has joined #ag 15:00:16 GN015 has joined #ag 15:00:36 https://mit.zoom.us/j/583945521?pwd=TE9HM1BSN1N4Ri9rMHdPU3pVZ0RoQT09 15:01:03 s/https://mit.zoom.us/j/583945521?pwd=TE9HM1BSN1N4Ri9rMHdPU3pVZ0RoQT09/ 15:01:22 maryjom has joined #ag 15:03:03 bruce_bailey has joined #ag 15:03:10 present+ 15:03:12 present+ 15:03:29 zakim, generate minutes 15:03:29 I don't understand 'generate minutes', Rachael 15:03:47 rrsagent, make minutes 15:03:48 I have made the request to generate https://www.w3.org/2023/03/21-ag-minutes.html Rachael 15:05:09 Detlev has joined #ag 15:05:16 present+ 15:05:16 agenda? 15:05:41 I can't get to the zoom call-in info page, it times out... 15:05:56 +1 15:06:35 zakim, pick a scribe 15:06:35 Not knowing who is chairing or who scribed recently, I propose Detlev 15:07:28 Scribe: Detlev 15:07:28 mbgower has joined #ag 15:07:58 Chuck: Want to introduce yourselves? 15:08:08 sarahhorton has joined #ag 15:08:08 ...Announcements? 15:08:10 q+ 15:08:19 present+ 15:08:19 ack Rach 15:08:34 +AWK 15:08:35 zakim, take up item 1 15:08:35 agendum 1 -- Google Focus appearance research presentation -- taken up [from 30 minutes via Chuck] 15:08:36 Rachael: There will be a 2 weeks survey tomorrow or Thursday for WCAG 3 15:08:39 present+ 15:08:46 present+ 15:09:16 Julia: I am researcher at Google 15:09:42 Julia: (sharing screen) 15:10:01 Focus states: research and solutions 15:10:35 URL: https://docs.google.com/presentation/d/1luwiP519TITHicm0g-5xnE_l47Yyw96nUJSNW8Z2wEU/edit?resourcekey=0-AHcVjjJ5TIM-9VJmxpiKlQ#slide=id.g1b8ea21c7ad_0_6 15:10:37 Chuck: can you share URL 15:11:41 MelanieP has joined #ag 15:11:53 Julia: sharing Google research on focus states 15:12:04 present+ 15:12:19 ...first share research, clarify understanding that may contribute to WCAG 15:12:40 The meeting page on w3c is timing out for me and Wilco. Can you please drop the zoom link info here? Thanks! 15:12:45 ...what is a focus indicator? (Julia explains) 15:13:18 Julia: sharing SC text 2.4.11 15:13:29 present+ 15:13:35 Thank you! 15:13:38 understand, align, even exceed guideline 15:14:12 Julia: contrast and failing that, 2 pixel width 15:14:20 ...may be difficult to see 15:14:42 ...want to research optimal focus state indication 15:15:11 ...part 1 behavioral task (fundability) 15:15:25 part 2 user opinios 15:15:39 part 1 was focusing on response time and accuracy 15:16:06 ...click highlighted element as past as possible 15:16:34 ...starting at the centre, one of the buttons around it being highlighted 15:16:47 ...modifying color shape outline weight 15:17:01 ...1080 participants 15:17:14 some with some without disability 15:17:39 ...no particular design was helpful for one group over another 15:18:30 ... 1dp outline inside and outside button did not perform well 15:18:50 q? 15:18:51 ...confirms WCAG stance on non conformance 15:19:10 ...1 dp seemed to not perform well throughout 15:19:41 ... with extra padding and more weight targets performed better 15:19:55 (please refer to presentation for details :) ) 15:21:07 ...dropping weight of outlines increases response times 15:21:53 laura has joined #ag 15:22:02 Luis: Example of solution 3 dp outline with 2 dp padding and good contrast 15:22:13 ShawnT has joined #ag 15:22:17 Wilco has joined #ag 15:22:46 present+ 15:22:46 Luis: (summarising SC 2.4.11 requirements) 15:22:47 present+ Laura_Carlson 15:22:57 dan_bjorge has joined #ag 15:23:01 kirkwood has joined #ag 15:23:07 ...some components can't have padding 15:23:19 chinshaw has joined #ag 15:23:26 like list and menu components - so 3 dp outline rremains 15:23:28 mgarrish has joined #ag 15:23:46 J_Mullen has joined #ag 15:23:50 Present+ 15:23:51 present+ 15:23:56 Luis: showing different schemes and what targets look like there 15:24:01 present+ 15:24:07 present+ 15:24:12 SuzanneTaylor has joined #ag 15:24:14 present+ 15:24:15 ...referencing the way it meets SC 15:25:27 GreggVan has joined #ag 15:25:27 Julia: what does this all mean? 15:25:27 ...WCAG allowed design may lead to focus states that are difficult to find 15:25:38 sorry for being late -- the w3 server isnt giving out the dial in number -- and it only just occurred to me I could get it here. can someone post it please? thx Gregg 15:25:45 ...recommendinng requiring a weight of 2 dp instead of 1 15:26:06 ...but could be tough as a minimum requirement 15:26:14 jeanne has joined #ag 15:26:45 joweismantel has joined #ag 15:26:48 q+ to ask what is a dp 15:26:49 Julia: there are benefits to exceed requirements - (google intends to do that) 15:26:57 ack Ch 15:26:57 Chuck, you wanted to ask what is a dp 15:27:11 jon_avila has joined #ag 15:27:11 Chuck: Julia - what id dp? 15:27:11 present+ 15:27:28 Julia: it is a virtual pixel measurement 15:27:30 device independent pixels 15:27:31 Q+ to ask what we are allowed to do with this presentation, as it is marked "confidential and proprietary" 15:27:37 present+Jon_avila 15:27:39 ack AWK 15:27:39 AWK, you wanted to ask what we are allowed to do with this presentation, as it is marked "confidential and proprietary" 15:27:40 q+ 15:27:41 q+ to ask about non-outline indicators 15:27:58 roughly equals CSS pixel (?) 15:28:16 AWK: is it confidential - you showed it to a public WG... 15:28:27 ack Jon 15:28:33 Julia: There is legal approval to share the doc 15:28:38 q+ to ask about the final recommendation 15:28:50 Jon: Goof to take into account the to find indicators 15:28:59 present+ 15:29:04 ...where participants Google employees? 15:29:33 ...what requirements you find difficult to meet? 15:29:48 Julia: This was not Googlers, but anyone 15:30:40 ...second question: problem when indicator does not fully enclose the target 15:31:13 Luis: in some components the focus ring could not encompass / be visible all around the component 15:31:22 40 dp equals 24 CSS px 15:31:27 q+ 15:31:35 ...but contrast and size criteria where met 15:31:47 ack ala 15:31:47 alastairc, you wanted to ask about non-outline indicators 15:31:51 Jon: (inaudible 15:32:10 Alastair: Thanks, good to see, lines up with session on developing criterion 15:32:13 Can you please share any details on the percent of participants with low vision? Thank you. 15:32:30 ack Ch 15:32:30 Chuck, you wanted to ask about the final recommendation 15:32:33 ...you were looking at outline - did you also look at non-outline? 15:32:52 Caryn has joined #ag 15:32:52 Julia: this focused on outline, most common 15:33:37 present+ 15:33:51 present+ 15:34:01 ack dan 15:34:02 q? 15:34:07 Luis: Main concern wanted to look at good differentiation of focused / non focused - change of background seemed less useful 15:34:50 Dan: one case was against mixed background - these were not uncommon in reality, and most difficult to treat 15:34:58 ...any recommendations for this? 15:35:15 Julia: this had light theme / dark theme focus 15:36:30 Michael Gilbert (Google): Goal of having inner and outer ring was to make it accommodate variable backgrounds and with adjacent contrast it would always be sufficiently contrasty 15:36:39 So, the recommendation from the research was minimum 2dp, which would be equivalent to 1.2px (CSS pixels)? 15:36:41 q+ for slide 27 do we agree that 3 of 4 fail 2.4.11 ? -- I think top two are presently a pass ? 15:36:55 q? 15:36:55 q+ to say can you comment on the level of difficulty your team had interpretting the SC wording 15:37:09 ... wogen on white and black background because you'd see at least one or the other of the outlines 15:37:28 ack bru 15:37:28 bruce_bailey, you wanted to discuss slide 27 do we agree that 3 of 4 fail 2.4.11 ? -- I think top two are presently a pass ? 15:37:37 ...works on arbitrary design without impact on design 15:37:43 https://w3c.github.io/wcag/guidelines/22/#focus-appearance 15:37:49 agreed 15:37:50 q+ 15:38:01 Bruce: don't agree the top two examples fail the current SC language 15:38:15 I agree, I think the top two pass clause 2 so long as the rings meet 3:1 with the button color 15:38:20 q+ to say that's down to the 4px shortest side, not encloses. 15:38:32 ack ala 15:38:32 alastairc, you wanted to say that's down to the 4px shortest side, not encloses. 15:38:32 if subcomponent is too weak it would pass on the main component 15:38:33 Would be good to know what the button looks like unfocused 15:38:44 encloses the user interface component or sub-component that is focused, and 15:38:49 abbey has joined #ag 15:38:59 Alastair: they probably pass the 4 px along the shortest side metric 15:39:24 ...that's wha I asked about the non-online indicators 15:39:34 zakim, close queue 15:39:34 ok, Chuck, the speaker queue is closed 15:39:48 ...people may use different quite visible indicators 15:39:48 ack mbg 15:39:48 mbgower, you wanted to say can you comment on the level of difficulty your team had interpretting the SC wording 15:40:22 Mike Gower: would like to hear about difficulty of interpreting the SC text in the team... 15:40:42 ...agrees that example does pass SC via the 4px case 15:41:07 ...while others (?) might fail (unclear which examples this refers to) 15:42:31 "is at least as large as the area of a 1 CSS pixel thick perimeter of the unfocused component or sub-component, or is at least as large as a 4 CSS pixel thick line along the shortest side of the minimum bounding box of the unfocused component or sub-component, and" 15:42:34 Michael (google): we need some level of interpretation - the req that the size of the focus state inside the component would need to be equal or greater than the perimeter of the component 15:42:37 Maybe they were looking at old version of the criterion 15:42:41 ...is that correct? 15:43:13 Alastair<<<<. only correct for perimeter but SC can be met also by the 4px at shortest side of the perimeter case 15:43:30 Two interesting things there - it doesn't perform well, and it wasn't understood. 15:43:36 Chuck: Thanks for presenting the research 15:43:39 Thank you Google team. 15:43:44 yes, thanks for this work and feedback ! 15:43:54 zakim, take up item 2 15:43:54 agendum 2 -- WCAG 2.2 Focus appearance Survey: https://www.w3.org/2002/09/wbs/35422/focus-appearance/ -- taken up [from Chuck] 15:44:17 present+ 15:44:41 I was unable to save my responses due to the w3c timeouts 15:44:45 Alastair: results of focus appearance survey - currently at risk - issue is complexity 15:45:23 ...Microsoft raised an issue that was resolved - different type of research, testers rating different examples 15:45:49 ...dotted and dashed outlines presented a problem because different browsers present this differently 15:46:33 ...some mistakes happened, like thinking perimeter was required, failed color inversion 15:46:53 ...another problem was determine whether default browser indicator is at play or not 15:47:22 ...various issues, that may be covered in understanding doc - table of safe and non-safe indicators 15:48:29 ...in summary we can go through comments - the survey seems to indicate that testability is the main issue 15:48:50 q+ to ask about Alastair's proposal 15:49:03 ...no proposal to solve that without creating some fall big loophole or constrainig design to using outlines 15:49:17 ack Ch 15:49:17 Chuck, you wanted to ask about Alastair's proposal 15:49:18 ...could be triple AAA 15:49:48 Chuck: So your proposal would be to move it to AAA? 15:50:05 Alastair: there are three objections to continue with it as is 15:50:17 ...no objection to move to AAA 15:50:54 ...Does it impact on moving Focus visible - does it need to be moved back to AA? 15:51:17 Chuck: Going through comments.. Dan 15:52:34 Dan: Fine with with move to tripple A or remove - but should never mandate outline - cases that uses grading non solid background would be outruled 15:52:44 q+ to suggest the alternative option (move to AAA and keep second part only). 15:53:05 Chuck: Reading Kiara's comments 15:53:26 Alastair: Option suggested was not on the table 15:53:41 Chuck: (reads Gundula's comment) 15:54:16 Chuck: (reading Jon's comment) 15:54:41 Chuck: (reads Melanie's comment) 15:55:20 jon_avila - would that suggestion be for if we were continued with it a t AA? 15:55:39 Chuck: (reads Alastair's comment) 15:55:43 q? 15:55:45 ack ala 15:55:45 alastairc, you wanted to suggest the alternative option (move to AAA and keep second part only). 15:55:49 q+ to say if we move this to AAA, can we increase the requirement to better match the user needs? 15:56:19 q+ to say I'm concerned about moving and tweaking 15:56:44 Alastair: Have spent a lot of time on this would like to keep guidance on level AAA focusing on the second part of the S 15:56:51 i did not submit survey but +1 with some move to AAA 15:56:52 ack Rach 15:56:52 Rachael, you wanted to say if we move this to AAA, can we increase the requirement to better match the user needs? 15:56:55 ...make it shorter (if not simpler) 15:56:57 q+ 15:57:28 ack Ch 15:57:28 Chuck, you wanted to say I'm concerned about moving and tweaking 15:57:33 q- 15:57:36 +1 Rachael 15:57:39 Rachael: Would rather keep it in - but if AAA is the way to go we should increase requirement to move thickness to 2 or 3 CSS pixels 15:57:46 +1 Rachael 15:58:03 Chuck: Concern about moving I level up AND taking it - need to be careful 15:58:04 q+ 15:58:11 +1 Rachael 15:58:16 +1 to rachael comment that by moving to AAA we can strengthen and simplify 15:58:17 q+ 15:58:22 ack mb 15:58:30 ...lov to in-areas weight requirement 15:58:51 Mike Gower: Thanks to all who did this even if it is to become AAA 15:58:55 q+ to ask for a scribe change 15:59:09 ...it is still valuable, can be pointed in, used in in-house testing 15:59:38 ...increasing the threshold - needs clarification from MichaelC if this is an option 15:59:42 ack dan 15:59:46 q+ on the options at this stage 15:59:54 Ben_Tillyer has joined #ag 15:59:57 present+ 16:00:29 q+ 16:00:33 ack Ch 16:00:37 Dan: big +1 to folks who worked on this - good to move this forward - interesting that perimeter requirement was misinterpreted even by an expert 16:00:57 +1 to Dan Bjorge observation that google SME has trouble reading SC 16:01:22 q+ to also mention a quick update. 16:01:25 scribe+ dan_bjorge 16:01:27 ack Ch 16:01:27 Chuck, you wanted to ask for a scribe change 16:01:31 ack ala 16:01:31 alastairc, you wanted to comment on the options at this stage and to also mention a quick update. 16:02:10 Alastair: on options to make changes: hard to make changes at AA and higher now. will check with Michael Cooper tomorrow 16:02:31 q+ 16:02:31 ...did previously have a version of focus appearance at AAA, was part of first wide review, so wouldn't be completely unexpected 16:02:34 q+ to say that if we are tweaking, we need to craft it and review it as is 16:02:37 ack mb 16:02:43 ...think a AAA change would be feasible if it's what group wants 16:03:09 Mike: Suggest pulling out entire first clause, create a technique for two-color contrasting indicator that meets updated requirement 16:03:14 Suggest: When the keyboard focus indicator is visible, an area of the focus indicator meets all the following: 16:03:14 - is at least as large as the area of a 2 CSS pixel thick perimeter of the unfocused component or sub-component, and 16:03:17 ...maybe offer google research as supporting doc 16:03:46 ack Ch 16:03:46 Chuck, you wanted to say that if we are tweaking, we need to craft it and review it as is 16:03:55 ...Clarifying that such a two-color indicator would fail part 1 of the existing requirement text. Think that eliminating clause 1 might cut down on interpretation difficulty 16:04:09 q+ 16:04:17 ack Wil 16:04:37 Charles: Unsure how to craft a poll about exactly how to change, may need to craft if offline and come back 16:04:43 q+ 16:04:50 ack ala 16:04:50 Wilco: Thinks we would likely need another round of CR if making substantive changes to the text 16:05:13 q+ to say is it Wilco's feeling that moving to AAA is also going to trigger another round? 16:05:25 Alastair: Thinks in AAA area we have more flexibility, likely to be non-problematic if we're taking substantially the same at-risk requirement and moving from AA to AAA 16:05:32 ...Will double check, but thinks it's likely the best option 16:05:39 q- 16:05:48 q+ to ask if we should consider a very simple sub-set of the 16:05:50 No, I don't think changing to level AAA requires another CR Mike 16:05:57 ack Suz 16:05:57 SuzanneTaylor, you wanted to ask if we should consider a very simple sub-set of the 16:06:25 Poll: Assuming we don't trigger another round of CR, do you approve of moving a sub-set of the SC to AAA? 16:06:43 Suzanne: Asks if we can take a simple subset of the requirement and put that at AA. Maybe "some portion of the focus indicator has 3:1 contrast, either change or adjacent". 16:06:48 q+ 16:06:56 q+ to say we have that in non-text contrast 16:07:04 ack ala 16:07:04 alastairc, you wanted to say we have that in non-text contrast 16:07:27 +1 to poll at 12:06 16:07:30 Alastair: Think we already have that as part of non-text contrast. Lacks change-of-contrast, but is close to that suggestion already 16:07:41 Poll: Assuming we don't trigger another round of CR, do you approve of moving a sub-set of the SC to AAA? 16:07:54 +1 16:07:55 +1 16:08:00 +1 16:08:02 +1 16:08:05 q+ to ask if we already polled for AAA ? 16:08:07 -1 not without knowing which kind of subset 16:08:07 +.9 16:08:10 0 joined call too late to understand the nuances 16:08:15 .5 if we can't change the requirements, .9 if we can increase them 16:08:17 ack bru 16:08:17 bruce_bailey, you wanted to ask if we already polled for AAA ? 16:08:27 -1 Agree with GN 16:08:29 +1 16:08:33 +1 in principle, but depends on sub-set 16:08:42 +1 16:09:01 q+ 16:09:02 +1 in principle, but depends on sub-set (thanks Dan for the copy/paste) 16:09:05 ack ala 16:09:37 Bruce: Should we do separate polls for "move to AAA" vs "what update to make at AAA"? 16:09:43 q+ 16:10:49 q+ to say that this is a "spoiler" of what an AAA might look like, and we are not voting on this NOW, just that we will try to craft one that may look like this. 16:10:54 ack Bru 16:10:57 at the moment we have the area of a 1px thick perimeter, haven't we? Increasing to 2px means a substantive change. 16:11:21 q+ 16:11:29 Alastair: suggests proposal for AAA text that is essentially "the second clause of the current text", but increasing the area requirement to "area of a 2CSS pixel thick perimeter" with no "smallest side of minimum bounding box" clause 16:11:45 s/proposal/preview of proposal/ 16:11:53 This is for those who weren't sure what the sub-set would be 16:12:00 q? 16:12:09 ack Ch 16:12:09 Chuck, you wanted to say that this is a "spoiler" of what an AAA might look like, and we are not voting on this NOW, just that we will try to craft one that may look like this. 16:12:14 ack mb 16:12:55 q+ to ask for clarity on the Bruce and Mike ask 16:13:10 ack Ch 16:13:10 Chuck, you wanted to ask for clarity on the Bruce and Mike ask 16:13:20 Mike: Can see scenario where moving AA to AAA might be fine without CR, but changing wording along with move might trigger new CR. From that standpoint, would be nice to see if group approves "move to AAA" separately from "change text" or not 16:13:24 Unfortunately changing the text of the SC doesn't seem to be allowed as it is a Substantive Change: https://www.w3.org/2021/Process-20211102/#substantive-change 16:13:51 After CR you can go to "Proposed Recommendation, if there are no substantive change other than dropping at risk features" 16:14:07 Bruce: agree with Mike's restatement 16:14:19 q+ 16:14:19 Poll: Agree to explore moving a modified version of the SC to AAA 16:14:25 ack Ben 16:14:30 +1 16:14:32 +1 16:14:34 +1 16:14:42 q+ 16:14:44 ack Ch 16:14:48 q+ 16:14:48 +1 16:14:52 ack ala 16:14:54 +1 16:15:03 +1 16:15:17 +.5 only if it is ensured that the new version is not more restricting than the current version 16:15:19 q+ 16:15:21 +1 16:15:25 ack Rach 16:15:47 q+ 16:15:59 ack mb 16:16:25 Mike: Would like to get a vote in on moving it to AAA. Can do a separate resolution on changing text. 16:16:28 proposed RESOLUTION: Move the SC to AAA 16:16:41 +1 16:16:41 +1 16:16:43 +1 16:16:44 +1 16:16:45 +1 16:16:47 +1 16:16:49 +1 16:16:50 +1 16:16:53 +1 16:16:54 +.5 16:16:55 +1 16:16:57 +1 16:16:58 +1 16:17:06 -1 without knowing petnetial changes, dropping might be better then, so the option to drop should be kept 16:17:12 Just so I'm super clear, voting a -1 would most likely end in removal of SC? 16:17:20 yes 16:17:40 RESOLUTION: Move the SC to AAA 16:17:53 0 16:18:00 q+ 16:18:04 ack ala 16:18:24 Chuck: Not sure we need a second resolution at this moment, may want to craft proposal for language update and come back 16:18:42 Alastair: Wants to discuss "if we are able to modify it, what modifications would we want to make?" 16:18:53 q? 16:19:05 might have resolution to agree to continue considering phrasing -- so long as CR not risk 16:19:15 q+ 16:19:50 q+ 16:20:05 Corey: Loves increasing the minimum area like Alastair's preview from earlier. Wants to discuss dp vs px from google talk earlier 16:20:31 proposed RESOLUTION: Explore removing the first part of the SC and strengthening the second part without trigging another CR 16:20:39