RE: Proposals for revision of the Plain Language SC proposals for WCAG 2.1

Hi John,

The example you use about the Japanese language is actually covered by one of the exceptions:

“In languages where present tense and active voice do not exist, or are not clearer in the language of the content, use the tense and the voice that are clearest for the content”.

I appreciate that internationalisation is vital to the adoption of WCAG 2.1 and that clear, concise, machine testable SC are going to be very difficult to create around the use of language.

One thing to keep in mind is that there are three different SC around Plain Language, each with different scopes:


·         SC 30<https://github.com/w3c/wcag21/issues/30> (A) is restricted to  instructions, labels, navigational elements, and error messages, which require a response to continue

·         SC 41<https://github.com/w3c/wcag21/issues/41> (AA) is expanded to include language in headings, error messages and important information<https://rawgit.com/w3c/coga/master/extension/index.html#dfn-important-information> (with the removal of the requirement of a response to continue)

·         SC 42<https://github.com/w3c/wcag21/issues/42> (AAA) is expanded to include all content

The intent of the first two SC is to allow people with cognitive issues to functionally use a web-page, the third is to help them to fully understand the content of the page.

The exceptions are designed to accommodate different linguistic and stylistic requirements (such as games, where working out how to continue or being cryptic are part of the design intent or scientific works where a particular style is required to maintain credibility) and the requirement for exceptions is likely to increase through the set of SC, reaching a peak in SC 42. The danger here is that if the number of exceptions increases too much, the SC becomes pointless because everything is exempt.

I’m very new to the world of SC, so I’m not sure at what point is no longer acceptable for human testing, particularly when the volume of testing increase (if at all). Any guidance here would be greatly appreciated as these SC will be difficult to create robust automated tests for.


Regards,
Jim

Jim Smith
Accessibility Specialist
Accessibility & Digital Inclusion
Workplace & Service Desk Services

T +353 (0) 1 2469714
M +44 (0) 78 91138262
smithjs@atos.net<mailto:smithjs@atos.net>

Atos IT Solutions and Services Ltd.
Regus
Phoenix House
Monahan Road
Cork
T12H1XY
Ireland

http://www.atos.net<http://www.atos.net/>

[cid:image002.png@01D28332.128F9990]

Registered in Dublin, Ireland. Registered No.: 484835. VAT Registration No.: 9749790O
Directors:  A. Gregory, J. Finnerty

Important notice: This e-mail and any attachment thereof contains corporate proprietary information. If you have received it by mistake, please notify us immediately by reply e-mail and delete this e-mail and its attachments from your system. Thank you

[AtosPride-Ally]




From: John Foliot [mailto:john.foliot@deque.com]
Sent: Thursday, February 09, 2017 11:15 PM
To: Steve Lee <steve@opendirective.com>
Cc: lisa.seeman <lisa.seeman@zoho.com>; EA Draffan <ead@ecs.soton.ac.uk>; Milliken, Neil <neil.milliken@atos.net>; Thaddeus . <inclusivethinking@gmail.com>; public-cognitive-a11y-tf <public-cognitive-a11y-tf@w3.org>; Jeanne Spellman <jspellman@spellmanconsulting.com>
Subject: Re: Proposals for revision of the Plain Language SC proposals for WCAG 2.1

Hi Steve,

From my perspective, do not be confused by low levels of discussion on any single new SC - we are all struggling to keep up with the flurry of correspondence at this time.

The latest PR for this new SC is simply the latest PR - it in no way means that the SC is "finalized" - only that it is now going to the larger WCAG WG for more review before it is "baked" into the 2.1 FPWD. (Note that the full Working Group is not copied on this email, only the COGA TF)

I have a number of concerns with how this is emerging right now, including some centered around internationalization (for example, my early research shows that the use of the Passive Voice is not only common, but often "required" in the Japanese language, and insisting on a non-passive voice in that language may actually introduce *MORE* confusion for Japanese with learning disabilities. Surely we don't want that!)

Additionally, I personally believe that statements such as "It is expected that natural language processing algorithms will be able to conform to this automatically with reasonable accuracy." (Future tense) means that we do not have this ability today - but I am not sure, do such tools exist today? (Later, the draft suggests that IBM has "a tool" that can perform this today, but dependency on a single tool for testing is problematic, especially if it is a "for-profit" tool. Additionally, does that tool also support multiple languages? My colleague Birkir Gunnarsson is Icelandic - does the tool support his mother tongue as well?)

NOTE - I am not for an instant suggesting that the spirit of this SC, or the Needs Statement that is driving it, are not valid, only that the proposal as writ right now will have a hard time passing the wide review that FPWD brings, and if we cannot answer the types of questions I am asking now, in our more closed environment, then this SC will likely not make the final cut, sad as that is.

So let's get it rock-solid now, ya?

JF

On Thu, Feb 9, 2017 at 3:14 PM, Steve Lee <steve@opendirective.com<mailto:steve@opendirective.com>> wrote:
Yes, my bad. I forgot where I was in the process of managing these 2.

The reason for my reticence was the very low level of discussion. These were my 1st as a SC manager and I really expect more push and shove. I guess that means they are good.

Sorry again for the confusion due to being new to the process.


Steve Lee
OpenDirective http://opendirective.com


On 9 February 2017 at 21:01, lisa.seeman <lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>> wrote:
The pull request was done before Jeene made her suggestions so it is really too late. The issue is closed.
My 2 cents - The Success criteria was pretty clear, measurable and testable  - more then a lot of what is in WCAG 2.0

All the best

Lisa Seeman

LinkedIn<http://il.linkedin.com/in/lisaseeman/>, Twitter<https://twitter.com/SeemanLisa>



---- On Thu, 09 Feb 2017 20:46:03 +0200 John Foliot<john.foliot@deque.com<mailto:john.foliot@deque.com>> wrote ----
Hi EA,

Thanks. I don't see this as "causing trouble" - I see this as having an open, honest and candid discussion. We need to balance the needs of many disparate groups, including content authors who are not experts (and never will be). I've tried very hard to stay on top of the COGA requirements, and one of the larger take-away's I've learned is that individual personalization is and will be the Holy Grail for COGA issues.

But we simply aren't there yet, not at anything that would scale, and I think we do ourselves a dis-service if we don't accept that truism today.

Re: Innovation - I fully support that 100% - YES. We have a number of user-needs today, however the technology still isn't mature enough to start mandating that site-owners do "X, Y, Z", and frankly I think that if we ever got to the point where WCAG became that prescriptive we'd loose more ground then we've gained.

This is one of the reasons why I suggested that for the release of 2.1, any User Requirement that was still unattainable at scale be none-the-less published as an official W3C Note, as we did with the MAUR (https://www.w3.org/TR/media-accessibility-reqs/) - not everything in that list is achievable today, but the needs still exist, and what the 'expectations' are have been collected and published. To my happy discovery, there are now technologists out there taking these Requirements and then working on Proof Of Concept solutions. This has to be a positive thing!

I sort of think of it like American Football - not every play is going to score a touch-down, but if we are successful in moving the ball closer to the goal line, we're still "winning". WCAG 2.0 had little-to-nothing to address the needs of the core constituency of the COGA and LV Task Forces when it was published in 2008, and we've done a good job collecting the User Requirements (Gap Analysis), but I also think we've got plenty more plays ahead of us before we score touch-downs there. But if, with 2.1, we move the ball forward closer towards the goal-posts, I think we're doing well - the goal now isn't "the touch-down" but rather "How many yards can we advance forward with this play?"

For me, it keeps on coming down to "Don't let Perfect be the enemy of Good".

Cheers!

JF

On Thu, Feb 9, 2017 at 10:47 AM, EA Draffan <ead@ecs.soton.ac.uk<mailto:ead@ecs.soton.ac.uk>> wrote:
Thank you for all the trouble you have taken John,  and I certainly did not expect such an amazing reply this was just me researching it all a bit more.

Apologies for causing trouble.  Lets just see if we can find a better way to test readability to suit all users.  Perhaps we can be a bit more innovative as Lisa suggested, but I appreciate we will have to make it robust and go through validation tests - thoughts of crowdsourcing help across different languages etc.

Best wishes
E.A.

Mrs E.A. Draffan
WAIS, ECS , University of Southampton
Mobile +44 (0)7976 289103
http://access.ecs.soton.ac.uk<http://access.ecs.soton.ac.uk/>
UK AAATE rep http://www.aaate.net/



________________________________
From: John Foliot [john.foliot@deque.com<mailto:john.foliot@deque.com>]
Sent: 09 February 2017 16:18
To: EA Draffan
Cc: Milliken, Neil; lisa.seeman; Thaddeus .; public-cognitive-a11y-tf; Jeanne Spellman
Subject: Re: Proposals for revision of the Plain Language SC proposals for WCAG 2.1

TL;DR:
   WCAG Success Criteria need to be measurable, and while Reading Scores have their issues, they are at least measurable and repeatable, and will be significantly more palpable to the millions of content authors we will be asking to meet this Need.

***

Hi EA,

Thanks for those links. After reading through them (and yes, I read all 3), I am struck by one of the conclusion statements of the third reference (https://www.ideals.illinois.edu/bitstream/handle/2142/15490/why-rf-fail.html?sequence=3)

"The real factors that affect readability are elements such as the background knowledge of the reader relative to the knowledge presumed by the writer, the purpose of the reader relative to the purpose of the writer, and the purpose of the person who is presenting the text to the reader. These factors cannot be captured in a simple formula and ignoring them may do more harm than good."

While we cannot discount this expert opinion
​, it also leaves me wondering how we can ever hope to "standardize" and quantify/measure something that is clearly not scientific​? Dissecting the statement above:

  1.  background knowledge of the reader relative to the knowledge presumed by the writer - unknown and unknowable at scale (i.e. sites that get hundreds of thousands of unique visits a day)

  2.  the purpose of the reader relative to the purpose of the writer - again, unknown and unknowable at scale

  3.  the purpose of the person who is presenting the text to the reader - this is the only factor apparently under the control of the content author, and in scope for the Web Content Accessibility Guidelines, and thus the only thing a WCAG SC can address.

​
My fear here is that there seems to be 2 opposing goals that we are trying to meet: one is a "testable" and measurable *standard* that can be taught and applied​ to millions of websites (the science piece), and yet "writing" and writing for specific audiences is an "art" (my distillation and take-away of those three articles).

I get "art", and art is important, but art cannot be quantifiably measured, it cannot be "taught" (outside of principles - the science of painting with oils versus drawing with charcoals), but actual "art" certainly cannot be standardized or measured (unless you are shopping at Walmart, and purchase "Pastoral Scene #3 - 40" X 60"")

What do I tell a Fortune 500 company they should do, if not try and meet some kind of standardized reading level? When you are authoring content for a million people, you cannot know all of your readers. I was more encouraged by one of the conclusions of the Leeds paper (
http://www.leeds.ac.uk/educol/documents/213296.pdf

​)​


"In conclusion, we want to emphasize that formulas are not invalidated for the great majority of writing. On the other hand, what they cannot measure should make clear that they cannot make writing a science."



So... what can we do?

In controlled environments, you may be able to ensure more attention is applied to the "art" side of the problem statement, but for a company like Tesco, what would you tell Tesco's editorial staff (where there is more than one editorial person) to do? Tesco proudly claim to serve "...millions of customers a week in our stores and online." (https://www.tescoplc.com/about-us/our-businesses/), and so all they can "know" about their audience is generalized data (likely determined by user-logs on their website, coupled with possible surveys and focus-group testing).

Large organizations like this also generally use Style Guides (AP, The Oxford Style Manual, The Chicago Manual of Style, etc. See: https://en.wikipedia.org/wiki/List_of_style_guides) as well as often they will have internal "Voice of the company/Voice of the client" guides as well (when I worked at JPMC they had such an internal document).

However, outside of specialized environments, getting any kind of buy-in from the millions of content creators out there will necessitate some form of measuring methodology, and while reading scores have their issues, they seem to be better than nothing at all, and so I am concerned that COGA experts are pushing back on this. I will posit that Jeanne's re-writes, while not 100% "perfect", brings the authoring solution a lot closer to what is required based upon the research provided.

Add to that the increasingly litigious environment around web accessibility, and ask yourself how will a judge (who is neither an accessibility expert nor a language expert) going to judge whether a site "fails" or not? (For this reason alone we need standardized testing of some fashion or other, and if not readability scores, then what?)

JF

On Tue, Feb 7, 2017 at 8:31 AM, EA Draffan <ead@ecs.soton.ac.uk<mailto:ead@ecs.soton.ac.uk><mailto:ead@ecs.soton.ac.uk<mailto:ead@ecs.soton.ac.uk>>> wrote:
I vote 3

Holiday reading or references!

Readability: The limitations of an approach through formulae (this paper has a definition of readability)
http://www.leeds.ac.uk/educol/documents/213296.pdf


Another very readable discussion about readability and the limitations of scales,  but also measuring sentence length by number of words etc.
http://www.impact-information.com/impactinfo/Limitations.pdf


old one
 https://www.ideals.illinois.edu/bitstream/handle/2142/15490/why-rf-fail.html?sequence=3




Best wishes
E.A.

Mrs E.A. Draffan
WAIS, ECS , University of Southampton
Mobile +44 (0)7976 289103<tel:%2B44%20%280%297976%20289103>
http://access.ecs.soton.ac.uk<http://access.ecs.soton.ac.uk/>
UK AAATE rep http://www.aaate.net/



________________________________
From: Milliken, Neil [neil.milliken@atos.net<mailto:neil.milliken@atos.net><mailto:neil.milliken@atos.net<mailto:neil.milliken@atos.net>>]
Sent: 06 February 2017 23:13
To: lisa.seeman
Cc: Thaddeus .; public-cognitive-a11y-tf; Jeanne Spellman
Subject: Re: Proposals for revision of the Plain Language SC proposals for WCAG 2.1

I vote 3


Kind regards,

Neil Milliken
Head of Accessibility & Digital Inclusion
Atos
M: 07812325386<tel:07812%20325386><tel:07812325386<tel:07812%20325386>>
E: Neil.Milliken@atos.net<mailto:Neil.Milliken@atos.net><mailto:Neil.Milliken@atos.net<mailto:Neil.Milliken@atos.net>><mailto:Neil.Milliken@atos.net<mailto:Neil.Milliken@atos.net><mailto:Neil.Milliken@atos.net<mailto:Neil.Milliken@atos.net>>>
http://atos.net/iux

http://atos.net/accessibilityservices

@neilmilliken



On 6 Feb 2017, at 22:35, lisa.seeman <lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com><mailto:lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>><mailto:lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com><mailto:lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>>>> wrote:

I am changing my vote to 3 as well.
The SC as it  is incredibly easy to write testing tools for. there are a few open source  language processing tools that you can use to count cluses actureltys. Testing against a word list is also something that exists already in restricted language tools and is very easy to program. It cant be that we need to have a worse SC and use archaic reading level tools because WCAG are to set in their ways to accept any new technology.

All the best

Lisa Seeman

LinkedIn<http://il.linkedin.com/in/lisaseeman/>, Twitter<https://twitter.com/SeemanLisa>




---- On Mon, 06 Feb 2017 21:55:36 +0200 Thaddeus .<inclusivethinking@gmail.com<mailto:inclusivethinking@gmail.com><mailto:inclusivethinking@gmail.com<mailto:inclusivethinking@gmail.com>><mailto:inclusivethinking@gmail.com<mailto:inclusivethinking@gmail.com><mailto:inclusivethinking@gmail.com<mailto:inclusivethinking@gmail.com>>>> wrote ----

I vote 3

On Feb 6, 2017 11:08 AM, "lisa.seeman" <lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com><mailto:lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>><mailto:lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com><mailto:lisa.seeman@zoho.com<mailto:lisa.seeman@zoho.com>>>> wrote:
We had issues with reading level , for example the word "mode" is a lower reading level than "hot or cold" . the lower reading level is much harder to understand.
The reason to go with Jeanne's proposal is because wcag _might_ find it more testable. This would only be, in my opinion, because they have not bothered read the whole proposal and testability section  (or they do not want new tools) Also i am not sure it is more testable in different languages and that is essential for WCAG. Wordlists requiremnts however, can work easily in any language and wordlists can be automatically generated by parsing a few sites.

I agree that the "unless..."  clause is only human testable but that it very typical for wcag.


I want to suggest three options

1 -  we retract our current pull requests and put these in instead

2 - we go with the current pull requests. If they fail and the comments are hard to address then we go with Jeanne's

3 -we go with the current pull requests. we can revisit this if needed

My vote is 3, to go with the current wording and see what happens


All the best

Lisa Seeman

LinkedIn<http://il.linkedin.com/in/lisaseeman/>, Twitter<https://twitter.com/SeemanLisa>




---- On Mon, 06 Feb 2017 20:00:24 +0200 Jeanne Spellman<jspellman@spellmanconsulting.com<mailto:jspellman@spellmanconsulting.com><mailto:jspellman@spellmanconsulting.com<mailto:jspellman@spellmanconsulting.com>><mailto:jspellman@spellmanconsulting.com<mailto:jspellman@spellmanconsulting.com><mailto:jspellman@spellmanconsulting.com<mailto:jspellman@spellmanconsulting.com>>>> wrote ----

A group of us at The Paciello Group (TPG) have been meeting every week in January to comment on the WCAG 2.1 proposals.  Because we test WCAG 2.0 all day, every (business) day, we have a lot of experience with both the language of WCAG and the testing of WCAG.  What we decided this week is that we want to focus our efforts toward helping COGA TF draft success criteria that will get into WCAG 2.1 and will accomplish most of what you want -- even if it is phrased differently.

We started with the proposals that we thought would be the least controversial to the WCAG WG to include.  I looked at the Plain Language proposals and did my best to look at the needs identified by COGA TF, and craft language that I thought would be acceptable to the WCAG WG and be included in the first draft version of WCAG 2.1.

The wording is quite different, but in my opinion, addresses the needs identified.  I chose reading level, because it is internationally standardized, and there are automated tests already available.  When I look at Technique  G153: Making the text easier to read  https://www.w3.org/TR/WCAG20-TECHS/G153.html , it covers most of the items that the COGA TF identified.

Issue 30 Proposal:

Understandable Labels:  Navigation elements and form labels do not require reading ability greater than primary education level.  (A)  [link to WCAG’s definition of primary education level from UNESCO standard]


Issue 41:

Understandable Instructions:  Headings, error messages and instructions for completing tasks do not require reading ability greater than lower secondary education level.  (AA)  [link to WCAG’s definition of lower secondary level from UNESCO standard]


Delta 3.1.5 (rewrite of existing WCAG 3.1.5)

Understandable Content: Blocks of text either:  (AAA)

·        have a reading level no more advanced than lower secondary education, or

·        a version is provided that does not require reading ability more advanced than lower secondary education. [links to WCAG’s definitions of lower secondary education and blocks of text]








Atos, Atos Consulting, Worldline and Canopy The Open Cloud Company are trading names used by the Atos group. The following trading entities are registered in England and Wales: Atos IT Services UK Limited (registered number 01245534), Atos Consulting Limited (registered number 04312380), Atos Worldline UK Limited (registered number 08514184) and Canopy The Open Cloud Company Limited (registration number 08011902). The registered office for each is at 4 Triton Square, Regent’s Place, London, NW1 3HG.The VAT No. for each is: GB232327983.

This e-mail and the documents attached are confidential and intended solely for the addressee, and may contain confidential or privileged information. If you receive this e-mail in error, you are not authorised to copy, disclose, use or retain it. Please notify the sender immediately and delete this email from your systems. As emails may be intercepted, amended or lost, they are not secure. Atos therefore can accept no liability for any errors or their content. Although Atos endeavours to maintain a virus-free network, we do not warrant that this transmission is virus-free and can accept no liability for any damages resulting from any virus transmitted. The risks are deemed to be accepted by everyone who communicates with Atos by email.




--
John Foliot
Principal Accessibility Strategist
Deque Systems Inc.
john.foliot@deque.com<mailto:john.foliot@deque.com><mailto:john.foliot@deque.com<mailto:john.foliot@deque.com>>

Advancing the mission of digital accessibility and inclusion


--
John Foliot
Principal Accessibility Strategist
Deque Systems Inc.
john.foliot@deque.com<mailto:john.foliot@deque.com>

Advancing the mission of digital accessibility and inclusion






--
John Foliot
Principal Accessibility Strategist
Deque Systems Inc.
john.foliot@deque.com<mailto:john.foliot@deque.com>

Advancing the mission of digital accessibility and inclusion

Received on Friday, 10 February 2017 00:23:04 UTC