Low Vision Accessibility Task Force Teleconference -- 12 Jan 2017

hyphenation

http://w3c.github.io/low-vision-a11y-tf/hyphenation-test.htm

there is a soft hyphen generator - http://yellowgreen.de/soft-hyphenation-generator

and a javascript generator - https://github.com/mnater/Hyphenator

General discussion

<scribe> scribe: allanj

wayne: concerned about SC.
... other folks don't see user CSS as a mechanism for achieving.
... LA times uses 25 character columns.

<laura> http://www.d.umn.edu/~lcarlson/wcagwg/tests/user_styles/important_spacing.html

Laura: for Spacing SC. want examples of author styles breaking user styles.

<laura> Patrick wants to see hard examples of where authors are breaking user styles…

<laura> and where it's not simply the user styles that need to be written more specifically with their selectors - e.g. not just setting style rules purely on body or html.

from Jonathan - https://lists.w3.org/Archives/Public/public-low-vision-a11y-tf/2017Jan/0008.html

"that is the point I was attempting to make-- from what I know stylish adds in the styles as document level styles and they are not seen as "user level styles". Many browsers don't let you add user level styles and thus we are stuck with tools that only allow us to enter document level styles."

<laura> http://www.d.umn.edu/~lcarlson/wcagwg/tests/user_styles/spacing.css

jim: with css specificity, they user must look at the code and fined the classes and id associated with all elements on the page, then build the 'stylish' rule to make a page readable. Very onerous for users

laura: stylish is not a user stylesheet.

<laura> From the CSS Cascading and Inheritance Level 3 specification: https://www.w3.org/TR/2013/CR-css-cascade-3-20131003/#important0

<ScottM> The basic problem is that the majority of users are not able to create their own style sheets and the closest they will get is high contrast mode in Windows

wayne: author's should not be using !important in CSS

<shawn> [ /me looking for examples in the wild... https://wellnessmama.com/2261/water-kefir-soda/ -- my Stylish changes line spacing in most of the text but not in the recipe box ...]

<shawn> ^^ wrong, ignore that

IE worked with user style sheet

shawn: stylish doesnot override author !important
... user stylesheets in safari and IE do override author !important

firefox extension - https://addons.mozilla.org/en-US/firefox/addon/user-style-manager/

above is a manager for stylish

wayne: authors using !important is a FAIL case
... at the element level on a page.

shawn: what is the limitation on developers of not using !important on elements in a page
... the user can use common tools for most elements (stylish). if author uses !important on an element then user must switch to IE or Safari
... to use user stylesheets

scott: most users will not use user stylesheets
... requires lots of technical acumen

shawn: +1
... how do we make it possible for those who have the technical skills to use them to write styles.

wayne: majority of folks with low vision don't use screen magnification

<alastairc> I'll be referring to this example in a moment: https://alastairc.ac/tests/layouts/percentages-rwd.html

<laura> Would be good if Shawn commented on the issue with her examples: https://github.com/w3c/wcag21/issues/78

shawn: Alistair - what is impact if authors not using !important?

alistair: user styles are a poor way of fixing author styles.
... been experimenting with using javascript to apply user styles.
... line length - when the user puts max-width by whatever mechanism, what do authors need to do to not break the user demand
... the question is "what are we asking authors to do"
... back to question.... not much of a limitation. the issues is frameworks, extra plugins, server side styles, etc.
... authors are writing very specific styles for get around the plethora of css in use. which makes it hard for users to override styles

https://alastairc.ac/tests/layouts/percentages-rwd.html

links can work as bookmarklet to work on any site.

because it is javascript it can be more selective

wayne: line length is not only about word wrap

shawn: accessibility support. these work with html. what about pdf?
... if pdf user can't do reflow (if forms) and no line spacing on any pdf

is that sufficeient justification for including it in WCAG. authors must provide this, because it only works in html

<shawn> also do-able in word processing formats

alistair: if something is only doable in html and not anywhere else, is that reason to reject from WCAG?

shawn: mobile is an exception. I have other ways of looking at html other than mobile device. if it is in PDF I can't read it.

alistair: the old hands don't reject "mechanism". hard to get consensus. we are not asking for widgets on pages.
... font family, line length and spacing..... are for users overriding settings. need to rephrase for things the author can do, not what user overrides.
... what things break when user tries to override and they can't?

shawn: then user can't read the content. especially in PDF

wayne: every WCAG 2.0 had techniques for things that users could do.

shawn: what can we do?
... having good well thought out examples, and communicating what "mechanism" means

<shawn> (e.g., "mechanism" doesn't mean you need a widget)

scott: this is a hard thing. huge variations in vision. users need flexibility. must keep after it. get what we can and try for more later.
... high contrast hasn't changed in years.

<David-MacDonald> I'm just joining in

<laura> Welcome, David.

david is facilitator for Line Length and Reflow.

david: what is thought on reflow language.

alistairs reflow j https://alastairc.ac/tests/layouts/percentages-rwd.html

<David-MacDonald> For all visual presentation of text, a mechanism is available to adjust the line length to a maximum of 25 characters (13 if CJK) without increasing the font size in the user agent, and without requiring two-dimensional scrolling, except where: * The user-agent provides no means of re-flowing content. * The spatial layout of the text is essential to its use. * A line has a word which is longer than 25 (13 if CJK) characters.

david: above stands by itself.

shawn: explain UA exception.

david: that is the mobile exception.

shawn: issue: content in html, with limitation in mobile...so use desktop. Fine. if content in PDF with a form...can't access it at all.
... it is not fine to have an exception for PDF. author must provide in an alternate format.

david: pdf xml forms is going away next year.

shawn: tons of legacy content. need a remediation path?

david: what about reflow in pdf.

shawn: doesn't work with forms, signatures, etc. has specific cases

david: what about export to html

shawn: functionally not workable

scott: reflow is based on old document structure. security will block export.

shawn: has examples of files that don't export

<alastairc> question: why the mobile exception?

shawn: goal is user can get to information. if info is only available in PDF - need an alternate format or a different tool

david: if it conforms to wcag, I can get to information. venn diagram - technically feasible from accessibility business case.
... if authors do this ... then problem solved. this is what SCs do.

shawn: today... authors need to provide info in html.

<laura> David’s Venn diagram is on his spreadsheet:

<laura> https://docs.google.com/spreadsheets/d/1XShLFX8fxHYYLn8A6avDwu37w9JfnZCGWvAKBpK9Xo4/edit#gid=684261396

<Wayne> q

<Zakim> shawn, you wanted to ask related point about accessibility supported

Alistair: can make pdf that reflow.
... why exception for mobile? line length and reflow allow users to override. we want to keep authors from preventing user override

<alastairc> https://alastairc.ac/tests/layouts/percentages-rwd.html

david: reflow. if I make window narrower the text will rewrap. but eventually will stop. and cant make it change.

<shawn> shawn clarifies -- can make *some* pdf relow, not all "Text that does not reflow includes forms, comments, digital signature fields, and page artifacts, such as page numbers, headers, and footers. Pages that contain both readable text and form or digital signature fields do not reflow.". Cannot change line spacing http://tader.info/support.html#PDFviewers

the bookmarklets work fine on CNN

alistair: if you can linearize then linelengh is unnecessary.
... only works if page linearizes well
... the mechanism is not a widget. its a bookmarklet or extension. The author just has to write in a way that does not break linearization

some pages works well. other pages things overlap badly.

we have a good user requirement. how to get authors to write appropriately.

alistair: chrome doesn't allow less than 400 pixel window.
... can you zoom in to get certain line lengths is irrelevant.

david: so drop 57 and reword 58

<David-MacDonald> tinyurl.com/jmo9st4

<David-MacDonald> LVTF Issue #58: Content can be viewed as a single column, except where: 1) The content is in data tables which have more than one column 2) Content contains interactive controls that cannot be reflowed 3) Reflow would cause distortion or loss of information

<alastairc> sorry, wife ghas to leave right now!

<David-MacDonald> [OR The spatial layout is essential to the function and understanding of the content.]

wayne: reflow causes distortion? what does that mean?

david: datatables, and <pre> and <code>

<David-MacDonald> [OR The spatial layout is essential to the function and understanding of the content.]

<shawn> Bottom Line -- Some things are an automatic pass in basic HTML. The issue is that users can't do it in other technologies (e.g., PDF). Thus, we need the requirement in WCAG, so that the user need is met -- (which means that while it is not accessibility-supported, then alternative version is needed for info in those technologies (PDF))

wayne: need to write <pre> and <code> better
... +1 fo spatial layout is essential
... main problem with reflow ... positioning ... you can get elements in a column that looks like linearization in 1.3.2 ... the only way to get column narrow enough
... is to linearize all block content (gmail breaks when linearizing)

david: 1.3.2 is all about MEANING, not linearization.

wayne: we want to be able to linearize page in the same order they appear is accessibility API

david: current language is OK?

wayne: yes.
... but it is difficult to linearize page. will take time to tease out what authors need to do to fix linearability
... new technologies are preventing setting position to static, and turning off grid.
... need to tease out what authors are doing that break linearization.

david: is it doable, technically possible, and business case feasible

wayne: some pages dye when linearized

david: I think we can put this out. get feedback. search for solutions in mean time.
... 58 (reflow) out, retire 57 (line length).
... use current language. keep spatial layout for later

wayne: would like to change reflow.... people get confused with word wrap. perhaps linearize...

david: linearize goes back to WCAG1 and has issues.

wayne: need better wording for title.

<shawn> [ /me too -- that's why I *really* appreciated Laura's succinct summary of the issues earlier in the call !!!]

shawn: me to

need summary of points that LVTF folks need from emails and calls

david: summary - Propose 57 is retired. present 58 with bullet 3 removed.

wayne: have alistairs code to check for failures.

<Zakim> shawn, you wanted to ask about spacing and to cooment about bottom line

david: moving the needle on responsive design.

shawn: does that address issue with spacing.

laura: if UA prevents spacing adjustment, content is exempt.

shawn: ok for mobile, not for PDF

<laura> For the visual presentation of blocks of text:

<laura> * character spacing can be adjusted by the user

<laura> * word spacing can be adjusted by the user

<laura> * line spacing can be adjusted by the user

<laura> * paragraph spacing can be adjusted by the user

wayne: really need spacing in PFD.

<laura> with following the exceptions:

<laura> * If the user agent prohibits spacing adjustments the content is exempt.

wcag are pushing back saying is a UA issue.

alistairs js works on mobile.

scribe: but not on PDF

shawn: why exception?

wayne: we are working on an AUTHOR level.

shawn: require authors to provide content in a way that is accessibly supported.
... thats what wcag2 said about flash. right?

david: all technologies must meets all of wcag...unless there is an exception.
... if the technology doesn't apply then its an automatic pass

wayne: mechanism is NOT a widget.

<shawn> [ Shawn is not cool with that exception ]

<shawn> Content is not accessible if users cannot change line spacing

wayne: they just have to be able to do something. If author uses !important then it breaks user control

david: need to be able to swap out stylesheets.

SC vs what is available. or what can be done today.

we want authors to not prevent users from changing their content to meet their needs

<shawn> [ /me debates saying that the only way to get user needs met right now is WCAG ]

wayne: because not many people can use user stylesheets, then UAs removed styleability of form controls and form controls are broken for lv folks

<alastairc> still going?

yes.!

<Wayne> still going

wayne: some places can't write javascript to modify form controls

david: reading mode in Edge. can modify spacing etc

shawn: zoomtext or AT is not solution for MOST users with low vision (ie moderate low vision)

david: have to prove SC is implementable.

shawn: what about alistairs language to not get in the way of CSS.
... can we propose spacing without the exception. it can be met automatically by html unless you use !important at page level element.
... it works on mobile with alistair's bookmarklet

alistair: bookmarklet will work on any website.

wayne: at least getting it in WCAG and recognized is important.

laura: need examples in the wild to prove sc

perhaps spacing as AAA.

wayne: can't control font family in select/option

<shawn> THANKS Alastair!

alistair: if we can prove spacing works in html
... opera on android, pinch zoom reflows text. sets a max width to text block

<laura> Bye. Thanks all.

trackbot, end meeting

<shawn> laura - please spacing *not* at Level AAA - it's a solid AA for many people

- DRAFT -

Low Vision Accessibility Task Force Teleconference

12 Jan 2017

Attendees

Contents

hyphenation

General discussion

Summary of Action Items

Summary of Resolutions

Scribe.perl diagnostic output