Accessibility Guidelines Working Group Teleconference

06 Jul 2017

See also: IRC log


AK, AWK, jasonjgw, david-macdonald, MichaelC, shadi, Detlev, MikeGower, KimDirks, steverep, Pietro, alastairc, Joshue, allanj, kirkwood, Laura, Makoto, Rachael, chriscm, marcjohlic, Joshue108
Pietro, EA_Draffan, Glenda, Shari_Butler, David_MacDonald, Mpluke, Chris_Loiselle, Crystal_Jones



<KimDirks> *w3c worked for me

<lisa> sorry I am getting the password wrong for the webex

<lisa> thank u t

<lisa> the capture and passwords are horrid

<lisa> when you get the password wrong

<lisa> ie - every time

srcibe: allanj

<scribe> scribe: allanj

Undo: https://www.w3.org/2002/09/wbs/35422/COGA_undo/

3 responses, 2 having issues

<AWK> https://www.w3.org/2002/09/wbs/35422/COGA_undo/results

<Rachael> Where users are required to enter data to complete a transaction, a mechanism is provided to undo an action or correct an error and return to the place where they identified the error through at least one clearly labeled action without data loss, except when the data loss is part of the correction.

<chriscm> Is anyone else getting digital noise???

RM: hope above addresses concerns

JOC: what is difference between current and your version.

rm: reduce the number of interactions. rest is clarification to questions

ls: scope for proposed is very limited.
... make sure 33 and 38 are covered, with the exception. they came from different use cases
... want back always working

joc: is LS ok with RM draft?

ls: no, if we go with proposed we loose too many use cases. perhaps can add an exception

rm: ok with adding exception

ls: explains exception,

joc: related to unwanted loss of data

<alastairc> The structure seems a bit odd, I think this says the same thing without as many clauses: "Undo: a user can go back steps in a process or repair information via a clearly labeled action without unwanted loss of data."

joc: adding exception to rm proposal.

<Joshue108> +1 to Alastair

cm: why no have BACK button listed instead of "clearly labeled action". or add in BACK to SC
... add bullet point - BACK button

<Joshue108> +1 to Chris

cm: better implementation. need to know when data will be lost and not

<Zakim> AWK, you wanted to say that we need to be able to evaluate the SC on a page by page basis. Which page would fail if the back button wasn't working?

ls: ok with the BACK addition.

awk: we need to be able to evaluate the SC on a page by page basis. Which page would fail if the back button wasn't working?

joc: would failure be on output page.

awk: not sure what is involved with making the back work in all cases

ls: user can go back, if they have the wrong browser, its a user issue. support back button... name the browser
... during testing
... awk? ok?

awk: yes, theoretically

cm: implementation wise... a little UA dependent. want consistent functioning. there are callbacks for the BACK button.

<kirkwood> The page would fail that doesn’t allow you to return to the previous filled out page, no?

<kirkwood> +1 to chris

cm: with new applications... BACK is now dependent on context in the application. app is in charge of doing something sensible with the back button... do something special or let the browser do its thing

<chriscm> @kirkwood yes.

jw: when can user do an undo. the proposal does not address this scenario. there are some actions that are non-reversible. no provisions for which actions are reversible and which are not.
... should be restricted to current session (privacy, security issues)

<alastairc> "except where an action has been confirmed by a second action..." or something like that?

jw: all irreversible issues are clearly identified by the author, help user make appropriate actions

<WayneD> * That looks good.

jw: need to create list of criteria for what is and is not reversible, based on context. need definition and qualification

joc: a scope issue. current wording is very broad. may need to restrict.

<WayneD> Maybe the formal term 'commit' should be used.

ls: agree. could jw propose some wording.

<chriscm> Can we define that in terms of "Changes of Context" which is wording used throughout WCAG already...?

jw: this is hard. put it on the list to see what group comes up with. needs some work

<Zakim> Greg, you wanted to say that "preserving data on return", "provide undo", and "auto-correct data" are independent features, so combining them into a single SC is needlessly complex

jw: pointed out issues, don;t have solutions.

<Joshue108> +1 to Makoto comment on Error Prevention

gl: issues with auto-correct, and repair, and incomplete process. too many in 1 sc. perhaps 3 sc

ls: time is short. may have to leave a loophole

<Greg> My comment was that I think "preserving data on return" is independent from "provide undo" and combining them into a single SC is needlessly complex and potentially confusing.

<Joshue108> https://www.w3.org/TR/UNDERSTANDING-WCAG20/minimize-error-reversible-all.html

joc: ls explain difference between this and 3.3.6 error prevention.

<Greg> Also, that my survey comments were submitted late so would require refresh to see.

ls: 3.3.6 is about submitting information. hit something by accident, how do I get back. that is not is 3.3.6

Adapting Text: https://www.w3.org/2002/09/wbs/35422/AdaptingTextJuly6/


<lisa> also problem is if a submission is reversable

<lisa> this offent includeds using small print return people which coga groups can not do

<lisa> so 3.3.6 is not helpful for coga

<lisa> even for finacial transaction

joc: 12 ready to go, 2 with issues

wd: had bullet for font family. done lots of research. little impact on developer.
... no deterministic way to change font family. issues with font-icon.s, etc.

<laura> @alastc has filed Issue 297: [Add technique for identifying CSS

wd: lost ability to change font, no font attached to <i>

<laura> generated content-images https://github.com/w3c/wcag21/issues/297

<laura> I have drafted several techniques for icon fonts one of which is: Providing a Semantically Identified Icon Font with

<laura> role=img

<laura> https://www.w3.org/WAI/GL/wiki/Providing_a_Semantically_Identified_Icon_Font_with_role%3Dimg

<laura> The objective of this technique is to show how to provide a semantically identified icon font that does not disappear if a user overrides font-family via user stylesheet.

wd: we may loose ability to change fonts. need to put in the GL to ensure ability to change fonts
... combine font spacing, letter and word spacing and limit to 400%

<laura> WCAG also has related failures: F87 F3

<alastairc> It came up repeatedly, we still can't see how it is a content issue. See Laura's extensive comments.

joc: was this discussed in LVTF. merged many SC. was font removed for a good reason.

ac: users can change fonts. don't need to add that ability. mitigate the impact of font changes.
... have techniques about font-icons. lots of research on font changes.

<laura> LVTF RESOLUTION: "removing font family from Adapting Text SC text, because font width is very similar to letter spacing. we will address spacing and font family in the understanding document: https://www.w3.org/2017/04/27-lvtf-minutes.html#resolution01

ac: impact of font change was incorporated into letter spacing, word spaicing
... not talking about font family, addressing the impact not a user agent issue.
... added lots of information in the Understanding document.

wd: there are many fonts that are more readable. they also need spacing. conflict with fonts and letter recognition.

ac: there is nothing to limit the users ability to change font
... we are talking about buffering around characters. preventing the layout falling apart.
... give the author information about how to buffer the content to not break the layout.

wd: if you change the font, then you can read. this is not about size.

joc: issue of changing font family, is not crucial to this SC. this SC is about spacing.

wd: I will file an objection on this SC

<WayneD> I file a formal objection to adapting text based on omission of font family

<alastairc> Wayne: Given that users can change font, what are you asking authors to do regarding fonts?

<laura> Wayne had mentioned Web Components may be a blocker for overriding font family. So I left the editors note in the SC to give him time to research and test that aspect.

<Zakim> Greg, you wanted to support Wayne: I believe (and have stated before) that I believe the concerns about font family can be addressed, and that it is important, and that making sure

sr: AC covered all of my issues and comments. may need a bit of info in the Understanding about fonts

<WayneD> Ok I'll romove my objection. Grr

gl: concerns - units on line spacing.
... wording has reintroduced

This draft has brought back the problem that it inadvertently requires pages to provide their own means to override author styling, rather than merely requiring content to still work when the formatting is overridden at the client. Specifically, it says that if the technologies being used allow the user agent to "adapt style properties of text"--*ANY* style properties--then it has to support...

scribe: "all" of those listed. It should instead only apply when the *corresponding* styles are adapted; that is, when the technologies (format and user agent) support adapting line spacing, then the content has to remain usable when line spacing is adapted. I still think my wording was cleaner, but understand the desire to start with more explicit scoping, so perhaps reword as "If the technologies bei

ng used allow the user agent to adapt any of the following style properties of text, then no loss of essential content or functionality occurs by adapting those properties, as follows: * if line spacing (leading) is adapted, it can be adapted to at least 1.5 lines without the loss of essential content or functionality * if letter spacing (tracking) is adapted, it can be adapted to at least...

scribe: 0.12 em without the loss of essential content or functionality * if words spacing is adapted, it can be adapted to at least 0.16 em without the loss of essential content or functionality

gl: first sentence says all styling is over-rideable
... need definition for adaptable
... support WD about font family, I think. agree with AC. user should be able to change all fonts.

joc: need to iterate SC, if potentially good, get it to the public for comment.

wd: withdraw objection

<Greg> I made four points, 3 from the survey, plus that Wayne and Alastair both have valid points but I think it'll be possible to come up with a solution that addresses Wayne's concern.

ls: thought COGA asked for increased paragraph spacing.

AC: LVTF reviewed. this is text level. para changes have layout complications.

<lisa> could we at least alow people to set phragh specing to 1.5?

<Zakim> steverep, you wanted to try to address Greg's concerns

lc: see survey for additional comments about GL and WD concerns.

<alastairc> Lisa: LVTF (mostly coming from me) didn't want to add in a layout-level change, it takes the complexity of the testing up a level.

sr: spacing... any style property. changed wording. "following" can be added.
... don't need def for "adapting" just a synonym for "changing"

joc: are we happy for this to go into the draft? public review?

<lisa> -1


<steverep> +1

<alastairc> +1 for this to go in

<laura> +1

<WayneD> +1

<Detlev> +1

<chriscm> +1

<marcjohlic> +1

<Rachael> +1

<KimDirks> +1

<Kathy> +1

<Joshue108> +1

<Greg> +1 for inclusion in draft, although not as written in a final version

RESOLUTION: Adapting Text accepted into editors draft

<lisa> josh, if there is an objection it should have some discusion time

<lisa> IRE that we can see if it can be addressed

open item 3

kw: addressing issues

<lisa> need to go to aria call now. leaving the call

<Joshue108> We will acknowledge objection

#62 keyboard, #63 touch functions, #71 non-interference with AT, #46 standard API

<WayneD> https://www.w3.org/2002/09/wbs/35422/Issues64-67-68/

kw: should these be clarifications to conformance claim?
... please add comments, let's have a discussion

<Detlev> she took it back, Jossh!

<laura> Thanks.

<Joshue108> trackbot, end meeting

Summary of Action Items

Summary of Resolutions

  1. leave open UNDO
  2. Adapting Text accepted into editors draft
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2017/07/06 16:34:36 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.152  of Date: 2017/02/06 11:04:15  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: AK, AWK, jasonjgw, david-macdonald, MichaelC, shadi, Detlev, MikeGower, KimDirks, steverep, Pietro, alastairc, Joshue, allanj, kirkwood, Laura, Makoto, Rachael, chriscm, marcjohlic
Present: AK AWK jasonjgw david-macdonald MichaelC shadi Detlev MikeGower KimDirks steverep Pietro alastairc Joshue allanj kirkwood Laura Makoto Rachael chriscm marcjohlic Joshue108
Regrets: Pietro EA_Draffan Glenda Shari_Butler David_MacDonald Mpluke Chris_Loiselle Crystal_Jones
Found Scribe: allanj
Inferring ScribeNick: allanj
Found Date: 06 Jul 2017
Guessing minutes URL: http://www.w3.org/2017/07/06-ag-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]