User Agent Accessibility Guidelines Working Group Teleconference -- 29 Aug 2013

<trackbot> Date: 29 August 2013

<AllanJ> scribe: allanj

1.8.12 reflow - level determination

from shawn: should be A not AA

<shawn> based on http://www.tader.info/scrolling.html

<jeanne> Today's Master document <- http://www.w3.org/WAI/UA/2013/ED-UAAG20-20130828/MasterUAAG20130829.html

group feels that AA is fine, more information is welcome

shawn comment delete 1.4.2

<shawn> [ Shawn feels discussion wihtout background is not useful ]

SH1 1.8.12 reflow

sh: discussing email. understanding users with low vision and dyslexia and text customization

kf: is there something else we need to consider

sh: email was quick
... things need to be added
... complexities to consider

kf: start smaller and work to the bigger issues

<shawn> http://www.tader.info/scrolling.html

js: zoom is basic level A
... advanced zoom is AA

sh: references research
... avoiding scrolling is the 3rd highest rating, 1st was text size
... EO has also been researching

kp: scrolling is important to the RSI community also
... would be happy to move to A

jr: text rescaling or whole page rescaling

sh: mostly scaling blocks of text - separate from navigation.

jr: text resize, rescale viewport are different
... this will be very hard for browser to do
... text, images, etc all rescale at the same time. hard to do without horizontal scrolling
... if text only ok at A, but not whole viewport

gl: but if UA want to do reflow on viewport nothing prevents them from doing it.
... "at least text" wraps

jr: "reflowable text content"
... sometimes author says "don't reflow"

gl: could use user-stylesheets to override reflow.

sh: this is only on html, not pdf

gl: will not help folks with pdf

sh: this must cover pdf

kf: do we have implementations?

sh: yes

gl: 1.8.3 add ", overriding author specified wrapping settings"

<shawn> implementation: at least Opera. plus all browsers with user style sheets

<jeanne> GL: Microsoft Word draft mode, Browsers can turn off styling

gl: word draft mode, in most browsers turn off styling

<shawn> 1.8.12 delete "reflowable"

<shawn> text content

<jeanne> +1 to delete reflowable

<shawn> and change to level A

<greg> Add: ", overriding any values specified by the author."

<greg> ("overriding" and "author" are links to the glossary.)

pdf reflow: http://help.adobe.com/en_US/acrobat/X/pro/using/WS58a04a822e3e50102bd615109794195ff-7d19.w.html

gl: some option, wrap to window. Word also has wrap to window

<jeanne> 1.8.12 Reflowing Zoom:

<jeanne> The user can specify that when text content in a graphical viewport rescales, it reflows so that one dimension of the content fits within the height or width of the viewport, overriding any values specified by the author. (Level AA)

jr: this is just reflow, no reflow zoom

<shawn> [ Shawn agrees reflow regardless of zoom]

gl: right.

<greg> The user can specify that text content in a graphical viewport reflows so that one dimension of the content fits within the height or width of the viewport, overriding any values specified by the author. (Level AA)

gl: stem - Reflow Text

kp: +1 to A

gl: need to fix examples

js: examples already talk about it.
... need pdf example

<shawn> ...lines of text go beyond the visible window (or viewport) and people have to scroll back and forth horizontally to read the text. ...For some people with disabilities, it is impossible to adequately read and comprehend text that requires such scrolling. Finding and activating scrollbars, then finding the next words in the sentence requires more effort. This causes a lose of reading flow and

<shawn> degrades comprehension, especially for people who have general difficulties in understanding written text....

sh: use any of this language
... suggests taking out height - The user can specify that text content in a graphical viewport reflows so that the content fits within width of the viewport, overriding any values specified by the author.

<Zakim> shawn, you wanted to suggest try width and check with Richard Ishida or other I18N folk

<jeanne> ACTION: jeanne to check with r12a on need for vertical fit of 1.8.12 [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action01]

<trackbot> Created ACTION-878 - Check with r12a on need for vertical fit of 1.8.12 [on Jeanne F Spellman - due 2013-09-05].

<jeanne> +1 for level A

<jeanne> JR and JA for level A

resolution: Text Reflow: The user can specify that text content in a graphical viewport reflows so that one dimension of the content fits within the height or width of the viewport, overriding any values specified by the author. (Level A)

<jeanne> ACTION: jeanne to renumber 1.8.12 to Level A, and adjust the summary. [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action02]

<trackbot> Created ACTION-879 - Renumber 1.8.12 to level a, and adjust the summary. [on Jeanne F Spellman - due 2013-09-05].

gl: is turning off all styles enough, or get specific

<greg> I'm concerned that browsers will claim they comply because the user can override author nowrap by turning off *all* author formatting, even though that may make the page unusable.

jr: not an issue

sh: good thing to consider, add to implementing--best practice to the wrap without changing other stuff rather than turn off all formatting

js: what about on mobile, no feature to turn off all styles?

<jeanne> 82941

SH2 - 1.4.2

<jeanne> scribe: jeanne

SH2: I do not think 1.4.2 Preserving Size Distinctions is necessary at all. I'll provide some support for that perspective from a survey (introduced at <http://www.tader.info/survey1304.html>).

<AllanJ> http://www.tader.info/survey1304.html

<AllanJ> 1.4.2 Preserving Size Distinctions:

<AllanJ> he user can specify whether or not distinctions in the size of rendered text are preserved when that text is rescaled (e.g. headers continue to be larger than body text). (Level A)

<AllanJ> sh: there are no implementations. may be an extension.

<AllanJ> ... users should be able to change size at the element level, then this is not necessary

<AllanJ> gl: in word, reflow to window, or draft and everything is same size.

<AllanJ> js: this is something Wayne Dick proposed. this would make everything the same size.

<AllanJ> sh: user can set text size then covered.

<AllanJ> js: what have something on customizing text sizing

<AllanJ> kp: wayne was adamant about this.

<AllanJ> gl: reads 1.4.1, but does not get specific about keeping all text the same size

<AllanJ> ... 1.4.2 extends 1.4.1

<AllanJ> js: what about bumping it down a level?

<AllanJ> sh: will check with wayne, then come back to it.

<shawn> ACTION: shawn to work with Wayne on 1.4.2 [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action03]

<trackbot> Error finding 'shawn'. You can review and register nicknames at <http://www.w3.org/WAI/UA/tracker/users>.

<scribe> ACTION: jeanne will discuss with shawn and Wayne on 1.4.2 and come back with a proposal [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action04]

<trackbot> Created ACTION-880 - Will discuss with shawn and wayne on 1.4.2 and come back with a proposal [on Jeanne F Spellman - due 2013-09-05].

<shawn> http://www.tader.info/requirements.html

SH3 1.4.1 redux challenges, etc

3. I'll have several suggestions (and challenges) for 1.4.1 Configure Rendered Text, for example:

3.a. I think 1.4.1 will need to be more specific -- for example, will need to specify range of fonts and colors to be available.

3.b. I think character spacing can go down to Level AA, and possibly line spacing, too.

3.c. There are several other things I suggest for Levels A, AA, and AAA.

<AllanJ> sh: element level customization is most important.

<shawn> Element [type] level Customization http://www.tader.info/element-level.html

<AllanJ> gl: element type customization

<AllanJ> does user styles cover this

<AllanJ> sh: yes

<AllanJ> gl: yes for html, but not for pdf (no user style sheets)

<AllanJ> sh: the point is PDF would have to do this.

<AllanJ> ... pdf has tags, can structure those to control presentation

<Eric> Eric needs to leave. Will return later

<AllanJ> kf: pushback, this seems aspirational

<AllanJ> gl: things that implement stylesheets could easily comply.

<AllanJ> sh: lets all things with no stylesheets less accessible

<AllanJ> kp: if we had SC wording and IER would be helpful

<AllanJ> ... if we have a proposal we can take it up next week,

<AllanJ> kp: have lots of questions.

<AllanJ> kf: this will take TIME, must be last call in Sept

<AllanJ> js: will work with shawn to get it done

<AllanJ> kp: want user perspective, good examples - use cases

<shawn> http://www.tader.info/element-level.html

<AllanJ> sh: importance of printing customized text. I increased font size I want to print them.

<AllanJ> kf: need fully baked proposals

<AllanJ> ja: no printing SC currently (loosing text on left edge of page is issue), was deferred to UAAG next

<AllanJ> sh: in 1.4.1

<AllanJ> ... what if browser only offers 2 sizes

<AllanJ> ja: for browser they already do with current settings, and user style sheets

<AllanJ> gl: this doesn't cover other user agents

<AllanJ> sh: you have to do this

<AllanJ> ... for AA need to choose line spacing at 1, 1.5 and 2

<AllanJ> ... AAA character spacing in .1 ems

<AllanJ> gl: want a minimum set

<AllanJ> sh: yes have a draft ready

<AllanJ> ... need to be more specific, will be a challenge

<shawn> http://www.tader.info/requirements.html#lnspace

<AllanJ> sh: actual increments, points, ems, %% need more research

<AllanJ> ... perhaps this is not necessary

<AllanJ> ... browsers, pdf, etc already do increments

<AllanJ> ... are there other UAs (ebooks, closed systems, etc) that this would need to apply to.

<AllanJ> sh: color contrast (f/b). best for users, easiest for wording

<AllanJ> ... users can select f/b color from the palet of the OS.

<AllanJ> what about things without an OS.

<AllanJ> what about a devise (ebook reader) without OS

<AllanJ> gl: this thing runs on hardware

<AllanJ> js: kindle on IOS allows no color changes

<AllanJ> sh: if UAAG applies other devices then you need to word to include these devices

<AllanJ> js: uaag would apply to these type of apps

<AllanJ> jr: epub is webbased

<AllanJ> gl: kindle app for IOS UA or not UA

<AllanJ> kp: yes it is

<AllanJ> kp: caution, industry may push back

<Eric> Eric has returned

<AllanJ> gl: edge case, need to call out in definition of UA, include ereaders as example

<AllanJ> sh: you have to be able to choose color from OS

<AllanJ> gl: don't say OS we say platform...hardware, os, other layers.

<AllanJ> sh: fonts?

<AllanJ> sh: need 2 serif, 2 sans-serif fonts

<AllanJ> jr: these seem to be edge cases

<AllanJ> sh: go with the simple

<greg> I think Shawn is saying that the SC would set no minimum on UA in the marginal case of where there is no OS, so not platform-supported fonts. The UA would provide it's own fonts, and there would be no minimal set specified by the SC.

<greg> Scribe: greg

Greg: The correct way to word it would probably in the form of "the set of system fonts, or this specified set, whichever is larger".

<jeanne> JS: The real problem is not when there are no system fonts, it is the platforms that do not expose the fonts to the applications

<shawn> shawn encourages you to read:

<shawn> * Text Customization for Readability <http://www.tader.info/index.html>

<shawn> * Understanding Users' Needs to Customize Text Display <http://www.tader.info/understanding.html>

Discussion of priorities for the remaining time today.

<Eric> Eric signing off

<jeanne> http://www.w3.org/WAI/UA/UAAG20/

<jeanne> 5.1.2 Intent - new

Most content specifications include features important to users with disabilities. Users may find it difficult or impossible to use a product that fails to support those features. Conformance claim documents list the technologies that the user agent fully supports, such as WAI-ARIA, so that users can make informed decisions about whether or not they will be able to use, and therefore should...

scribe: install, a new product or version of that product.

<jeanne> ACTION: jeanne to convert links to the W3C format for Bibliographies [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action05]

<trackbot> Created ACTION-881 - Convert links to the w3c format for bibliographies [on Jeanne F Spellman - due 2013-09-05].

<jeanne> Today's Master doc <- http://www.w3.org/WAI/UA/2013/ED-UAAG20-20130828/MasterUAAG20130829.html

<jeanne> ACTION: Kelly to address with UAWG adding "description" to 4.1.6 (KF, see review notes from F2F day 3) [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action06]

<trackbot> Created ACTION-882 - Address with uawg adding "description" to 4.1.6 (kf, see review notes from f2f day 3) [on Kelly Ford - due 2013-09-05].

<jeanne> ACTION: Kelly to bring up SC 3.2.5 Settings Change Confirmation to the group. [recorded in http://www.w3.org/2013/08/29-ua-minutes.html#action07]

<trackbot> Created ACTION-883 - Bring up sc 3.2.5 settings change confirmation to the group. [on Kelly Ford - due 2013-09-05].

<kford> We need to fix this sentence.

<kford> 'While it may improve accessibility for some users on some pages to have the page to***delete*** set focus to specific link***s*** or field***s*** when the page loads, it can also be detrimental for some users, and therefore users need to***delete*** control of this behavior.

<Jan> Unrecognizable content: When success criteria require authoring tools to treat web content according to semantic criteria, the success criteria do not apply when these semantics are missing (e.g., text that describes an image is only considered to be a text alternative when this role is encoded within markup).

<Jan> http://www.w3.org/WAI/AU/ATAG20/#part_a

<Jan> http://www.w3.org/WAI/UA/2013/ED-UAAG20-20130827/MasterUAAG20130828.html#def-recognize

<AllanJ> js: put this information from the definition in the introduction and link to the defintion of recognize

<Jan> UAAG 2.0 success criteria only apply to web content that can be *recognized*.

<Jan> UAAG 2.0 success criteria only apply to web content that can be *recognized* by user agents.

<AllanJ> put this in a new section

User agents are not expected to comply with success criteria when content document does not include information needed to satisfy the criteria (e.g. neither HTML nor video metadata provide any way to notify a user agent that a video contains flashing images, keyboard shortcuts are implemented using Javascript so they are invisible to the user agent, or an author denotes a heading only with...

scribe: larger font rather than heading tags).

<jeanne> http://www.w3.org/WAI/UA/2013/ED-UAAG20-20130827/MasterUAAG20130829.html

<jeanne> http://www.w3.org/WAI/UA/2013/ED-UAAG20-20130828/MasterUAAG20130829.html

<Jan> scribe: Jan

Things still to do

Reconcile table of contents

Consistent stems list (Kim)

Fix resources at end of document (Jeanne)

Read the glossary (Everyone)

Finish making editorial comments (e.g. from EO) (Jeanne)

Reply to all outstanding comments (Jeanne)

Renumbering SCs because they changed priority (Jeanne)

Bibliography (Jeanne)

Any other applicability notes to address (Jan)

Kim can reconcile Table of contents

JS: Aiming for Sept 24

Look up links in Resources

CSS changes to look more like ATAG2 but with WCAG2 type spacing (Jan)

Kelly to finish GL's 1 and 2 ("sanity check")

<AllanJ> removing @@ from doc-- still some floating around

Resolution: the Last Call will be 6 weeks

<AllanJ> priorities:

<AllanJ> table of contents

<AllanJ> stems

<AllanJ> proof glossary

<AllanJ> css changes - jan

Look through open action items (Kelly)

<AllanJ> close action-798 8/29/2013

<AllanJ> close action-798

<trackbot> Closed action-798.

- DRAFT -

User Agent Accessibility Guidelines Working Group Teleconference

29 Aug 2013

Attendees

Contents

1.8.12 reflow - level determination

shawn comment delete 1.4.2

SH1 1.8.12 reflow

SH2 - 1.4.2

SH3 1.4.1 redux challenges, etc

Summary of Action Items

Scribe.perl diagnostic output