Re: [Protocols] Agenda for April 22th, 2022

Hello,

We created a protocols email list to reduce the back and forth on the main lists. I am moving this conversation to that list.

If you are on the silver or AG list and would like to continue to see the protocols conversation and have not attended a protocols meeting, please email AGWG Chairs group-ag-chairs@w3.org<mailto:group-ag-chairs@w3.org> to be added to this list.

Please respond to this conversation using the list in the to line.

Thank you!

Rachael

From: Michael Gower <michael.gower@ca.ibm.com>
Date: Tuesday, April 26, 2022 at 9:58 AM
To: John Foliot <john@foliot.ca>
Cc: Jaunita George <jaunita_george@navyfederal.org>, "public-silver@w3.org" <public-silver@w3.org>, "w3c-wai-gl@w3.org" <w3c-wai-gl@w3.org>, Benjamin Feigel <Benjamin_Feigel@navyfederal.org>
Subject: RE: [Protocols] Agenda for April 22th, 2022
Resent-From: <w3c-wai-gl@w3.org>
Resent-Date: Tuesday, April 26, 2022 at 9:56 AM

It sounds like maybe there is a point of confusion going on here that will likely get resolved as the two ways of thinking about protocols (as represented by the breakout groups on Friday) work towards a merged understanding.

Simply, I’m thinking of the protocol establishing both the desired outcome and the preferred method for something too subject or varied to be a prescriptive WCAG 3.0 method but for which 3.0 provides some suggestions on which an organization makes its own guidance to follow.

So when I list a protocol as including the outcome of “minimize abbreviations” and the method “use only abbreviations on this common list”, what I pictured was WCAG publishing a bunch of suggested outcomes (a lot like some AAAs now) and some generic suggested methods, which are then made specific for the organizations. So  I think maybe the confusion is me saying “this common list”. I’m imagining that team publishing a style guide that lists allowed abbreviations for that site/organization for their language and controlled vocabulary. I can see some ‘common abbreviations’ lists being published an adopted by many (or refined to a subset). I’m not suggesting everyone would use the same lists.

A style guide is how a lot of things to do with more accessible language could be achieved by an organization. Many organizations, for example, follow Chicago Manual of Style. WCAG doesn’t say ‘you have to use Chicago for __’ but does say following a style guide is a method of wholly or partially meeting a number of desired outcomes.

Mike

From: John Foliot <john@foliot.ca>
Date: Wednesday, April 20, 2022 at 9:21 AM
To: Michael Gower <michael.gower@ca.ibm.com>
Cc: Jaunita George <jaunita_george@navyfederal.org>, public-silver@w3.org <public-silver@w3.org>, w3c-wai-gl@w3.org <w3c-wai-gl@w3.org>, Benjamin Feigel <Benjamin_Feigel@navyfederal.org>
Subject: [EXTERNAL] Re: [Protocols] Agenda for April 22th, 2022
Hi Mike, [Comments inline] ‘by using only abbreviations on this common list’ I suspect this will run headlong into internationalization issues very quickly. And, with multiple and various subject-matter domains, a "common" abbreviation ZjQcmQRYFpfptBannerStart
This Message Is From an External Sender

This message came from outside your organization.



ZjQcmQRYFpfptBannerEnd
Hi Mike,
[Comments inline]

‘by using only abbreviations on this common list’

I suspect this will run headlong into internationalization issues very quickly.
And, with multiple and various subject-matter domains, a "common" abbreviation (or acronym) to one group or community may be a complete mystery to another. Or there may be instances where the abbreviation or acronym has multiple meanings - for example the IAAP is *also* the International Association of Administrative Professionals<https://www.iaap-hq.org>. I'd be concerned if we get too prescriptive here, which this seems to suggest.


'by always using a spelled out version of the term in the first instance on a page or group of pages followed by the abbreviation in parentheses’

While I agree with this recommendation in principle, you are (IMHO) introducing a severe bottleneck with that kind of phrasing and the 'absolutist' tone of the requirement - it is presented in MUST-like language rather than SHOULD-like language, which will result in the same problem at evaluation time.

If the content author does not *ALWAYS* use a spelled out version, what are the consequences? Fewer 'points'? How/why? And how is that then evaluated or measured at scale? Absolutist language - a.k.a. MUST language - will (logically - at least to me) always resolve to either True or False, which, at the highest levels, is the problem that Protocols is seeking to resolve (at least in my mental model) - what to do when the real answer is neither True or False, but far closer to "...it depends"? Protocols, by design, would be far more flexible - the goal being that they help inform decision-making in the most positive way, but at the same time recognizing that "...it depends" is, and always will be, part of the mix.

To my mind, Protocols seek to solve the problems you point out via (what I think of as) formally documented "scenario recommendations" (and they could be applicable at both the micro and macro levels), which are the instructions that would be given - here, to content authors - at creation time, but without the need to have their output rigorously 'policed' with MUST-like requirements, which simply re-introduces the subjectivity concerns that I was hoping Protocols would address.


Where those protocols are published, the methods could help anyone evaluate the relative success of the outcome against the protocol.

To date, as 'examples' we've been using plainlanguage.gov<http://plainlanguage.gov> as our primary straw-man example, which is a site owned and published by the US Government (thus the .gov domain). I had wanted to also include "Making Content...COGA" as an example as well, but for reasons that I personally do not agree with, COGA has requested we not use that document at this time (because they believe there remain requirements in that document that can be 'measured' - but I don't see it myself). Additionally, I personally do not see a Protocol with *some* measurability for some requirements as being problematic: that document may have some additional sub-points that could be measured, but Outcomes like "Help Users Understand What Things are and How to Use Them<https://www.w3.org/TR/coga-usable/#objective-1-help-users-understand-what-things-are-and-how-to-use-them>" will always be a subjective determination.

FWIW, I've also always thought of the alt-text decision tree<https://www.w3.org/WAI/tutorials/images/decision-tree/> as a rudimentary Protocol - again, that guidance does not drive to a "measurable" outcome, but it has been created/authored in such a way that using that "Protocol" will significantly contribute to a "good" (better?) text alternative, based simply on the guidance and instruction provided by that document. That to me is the real value of a Protocol - it is the "steady coaching hand" that helps content creators "get it right" (to the best of their ability - which also presumes that over time their ability will improve with experience, as in real life). And in that model, I am proposing we reward entities for "trying", as opposed to penalizing them when they fail to meet a MUST-like technical requirement.

(I say this while the Conformance Model is still very much under-developed, but in principle, with the current Gold/Silver/Bronze proposal I see 'scoring' as being mostly a subtractive process, with Gold being "100%", and Silver and Bronze being "less than 100%" - you lose 'points' when you fail an ACT-style test, which seems to be the intent at this time, but... all TBD.)

However, I am proposing that with Protocols, instead of losing points, you gain points for making the commitment of adopting the Protocol "formally" - which is the accountability piece. For example, if IBM were to "formally" adopt Plain Language as part of IBM's authoring practices, that would also be introducing "institutional risk" to the company if IBM then failed to follow through on that commitment.

So the "assertions" piece that was part of my larger proposal will be key there, although I am not sure if everyone agrees that this kind of "risk" is or will-be sufficient in a conformance model. It certainly does anticipate a certain level of trust that, when an entity makes the assertion, that they will actually follow through. But, just like VPATs today, a certain amount of trust is required when it comes to "self-certification", which to me is the only way this will truly scale. (Does this leave open the door for "gaming" the score? Sure, but at the end of the day if an entity wants to cheat, they are cheating themselves first and foremost - and the public assertion piece potentially makes that cheating a risky proposition in any regulatory scenario. Having spent a fair bit of time consulting private industry, I know from my experience and perspective that corporate lawyers tend to be risk-averse at the best of times - that's what they get paid to do.)

But questions remain:

  *   What constitutes a 'valid' Protocol, and where should that 'official' and valid protocol 'live'?
  *   Who writes Protocols?
  *   Can we reference 3rd-Party protocols hosted external to the W3C if it/they have been vetted by our group?
  *   Do we envision a mechanism where 3rd-Party entities could introduce/propose additional Protocols (and how would we roll that up into our spec)?
  *   Or should all Protocols be hosted 'normatively' by the W3C?
(I see pros and cons for all of these options, and/but the group has not really tackled those questions yet, as we are still struggling with what constitutes a Protocol and how we will be using them.)

However I also see these as process and policy questions we can tackle once we have agreement on what a Protocol is and does in the context of WCAG 3.

JF

---------- Forwarded message ---------
From: Michael Gower <michael.gower@ca.ibm.com<mailto:michael.gower@ca.ibm.com>>
Date: Wed, Apr 20, 2022 at 9:26 AM
Subject: RE: [Protocols] Agenda for April 22th, 2022
To: John Foliot <john@foliot.ca<mailto:john@foliot.ca>>, Jaunita George <jaunita_george@navyfederal.org<mailto:jaunita_george@navyfederal.org>>
Cc: public-silver@w3.org<mailto:public-silver@w3.org> <public-silver@w3.org<mailto:public-silver@w3.org>>, w3c-wai-gl@w3.org<mailto:w3c-wai-gl@w3.org> <w3c-wai-gl@w3.org<mailto:w3c-wai-gl@w3.org>>, Benjamin Feigel <Benjamin_Feigel@navyfederal.org<mailto:Benjamin_Feigel@navyfederal.org>>

I largely agree with what John is saying, especially from this point onwards: “Protocols will not / should not "evaluate guidance" - Protocols *ARE* guidance – “
But how I’ve thought of the, protocols can potentially encapsulate both the guidance (‘minimize abbreviations’) and one or more adopted methods (‘by using only abbreviations on this common list’ and ‘by always using a spelled out version of the term in the first instance on a page or group of pages followed by the abbreviation in parentheses’). Where those protocols are published, the methods could help anyone evaluate the relative success of the outcome against the protocol.
So perhaps that is what is meant by “evaluate guidance”? Now that you’re holding protocols meetings and a later time on alternating weeks, I hope to attend to better understand where the group is at.

Thanks!
Mike

From: John Foliot <john@foliot.ca<mailto:john@foliot.ca>>
Date: Tuesday, April 19, 2022 at 4:15 PM
To: Jaunita George <jaunita_george@navyfederal.org<mailto:jaunita_george@navyfederal.org>>
Cc: public-silver@w3.org<mailto:public-silver@w3.org> <public-silver@w3.org<mailto:public-silver@w3.org>>, w3c-wai-gl@w3.org<mailto:w3c-wai-gl@w3.org> <w3c-wai-gl@w3.org<mailto:w3c-wai-gl@w3.org>>, Benjamin Feigel <Benjamin_Feigel@navyfederal.org<mailto:Benjamin_Feigel@navyfederal.org>>
Subject: [EXTERNAL] Re: [Protocols] Agenda for April 22th, 2022
Hi Juanita, You state that one view is: "Protocols are above or beyond any type of conformance, aka “extra credit”" I do not remember the group agreeing to that at all - and in fact I was slapped down for bringing up conformance during ZjQcmQRYFpfptBannerStart
This Message Is From an External Sender

This message came from outside your organization.



ZjQcmQRYFpfptBannerEnd
Hi Juanita,

You state that one view is: "Protocols are above or beyond any type of conformance, aka “extra credit”"

I do not remember the group agreeing to that at all - and in fact I was slapped down for bringing up conformance during that call.


JF: I still feel strongly that is not reasonable for us to expect much progress on protocol until we have better clarity with the WCAG3 conformance model...
… just too hard to know how to talk about protocol without metrics of scoring and points.

Rachael: I hear you John, and this is something the chairs have discussed.

JF: It seems like a blocker to me.

Jeanne: Point noted, I would ask that you let us move on. (JF agrees.)

(source: https://www.w3.org/2022/04/15-wcag3-protocols-minutes.html)

(I note that according to the minutes<https://www.w3.org/2022/04/19-ag-minutes.html>, the topic of the need for a workable Conformance Model came up again on today's WCAG call, as well as in the WBS survey for Charter Extension, with more than one person commenting on how critical it is we start to address this gap. But I digress...)

I disagree in the strongest of terms that Protocols are "extra credit" because:
a) we do not yet have any idea what "credit" means in the context of conformance, and
b) making Protocols "extra" in any context will likely serve to relegate Protocols to the WCAG AAA Success Criteria pile - lovely to look at but never really adopted at scale. I believe that would be a horrible end-state.

I personally do agree with a point made by Jeanne over the concern that Protocols not be "too weighted" and that using Protocols not be a means to game the final score in whatever conformance model we ultimately come up with.

***********
Continuing on the broader point, I do not agree with either of the "views" you presented here - and did not on Friday either, at least not as articulated here.

While I strongly disagree with the 'second' view, I also disagree with how you are characterizing the 'first view' - "Protocols can evaluate guidance where the outcomes cannot be measured, such as evaluating whether a process has been followed."

Protocols will not / should not "evaluate guidance" - Protocols *ARE* guidance - guidance that then allows content creators to evaluate their final outcomes when faced with a given scenario. The outcomes arrived at (in context) cannot be 'measured' in any consistent way, but they can be contrasted and compared to the published examples and expected outcomes articulated in a Protocol, with the content creation goal being "Get as close to this expected outcome as we can in our context".

So for example, when Plain Language<https://plainlanguage.gov> states "Choose your words carefully<https://www.plainlanguage.gov/guidelines/words/>" and then provides the following recommendations:

  *   Use simple words and phrases
  *   Avoid hidden verbs
  *   Avoid noun strings
  *   Avoid jargon
  *   Minimize abbreviations
  *   Minimize definitions
  *   Use the same terms consistently
  *   Place words carefully
...we cannot accurately or consistently "measure" the outcome, but we *CAN* evaluate content against those recommendations: Protocols help inform 'opinion'!

Example: One of the recommendations is 'minimize abbreviations' - which again cannot be measured or counted, but most people will be able to get a sense of whether there are "too many" abbreviations or not. But... how many is "too many"? Protocols cannot answer that question, because the most truthful answer is "it depends"... 10 abbreviations in 10 sentences may be too many (or not), but 10 abbreviations in 10,000 sentences is a whole different scenario.

Perhaps another way of thinking about Protocols is when they are 'used' by the content author(s)/creator(s).
ACT-like rules are run *after* there is some code or content to evaluate or test, but Protocols should be used far sooner than that (Shifting Left) - Protocols provide the appropriate guidance to avoid issues once content starts to be created. Protocols guide 'planning' far more than as part of the finished product (e.g. when writing content, "Choose your words carefully" - which is guidance given before the content is written, and not afterwards).

However you, as a 3rd-party evaluator can also use the Protocol to evaluate my content. You may not agree with each individual decision I made (subjectivity at play), but after us both reading the Protocol, we both should have a broadly-shared understanding of the goal of any given Protocol, that we can then use to guide our subjective (opinionated) evaluation of the content.

This is how I view both the goals and application of Protocols in WCAG 3 - which neither of your expressed perspectives seems to capture.

My $0.02

JF

On Tue, Apr 19, 2022 at 4:33 PM Jaunita George <jaunita_george@navyfederal.org<mailto:jaunita_george@navyfederal.org>> wrote:
Hi all,

We had a very productive meeting last week. Here are some of the decisions from our meeting last Friday:


  *   Decision: There are two views about how Protocols can be used:

     *   Protocols can evaluate guidance where the outcomes cannot be measured, such as evaluating whether a process has been followed.
     *   Protocols are above or beyond any type of conformance, aka “extra credit”

  *   Decision: Work will continue on both areas of consensus.
  *   Decision: Meetings will alternate between 8am and 12pm EST.
  *   Decision: Table discussion of splitting discussion during meetings until we have mechanics worked out but can split the time to both theories next week to keep discussion and work moving.

To continue our work, we’ll be using Zoom’s breakout group feature during the first half of our meeting and then reporting out on what we worked on and having a brief discussion to iron out the mechanics of our meetings moving forward.


Next Meeting:

The next meeting will be on Friday, April 22th at 12:00pm EST.

*** Agenda***

Here’s our agenda for this week based on our discussions:

Agenda+ Breakout groups (30 minutes)
Agenda+ Report out (10 minutes)
Agenda+ How to run future meetings

Meeting info:

The Zoom teleconference data is provided at this link: https://www.w3.org/events/meetings/f828eba1-cf1d-44ce-b68e-b22406b25ae1/20220422T120000


We will be on IRC using the W3C server at https://irc.w3.org<https://secure-web.cisco.com/1bbHg1R0NpE7wu_f231-AnlS80Y-y5vJTvdOEWJENhb8A72iXd8LCnf3ggBw4-smyidfgNxC8x8umv7E05ehkUF2bZBz2YRztLLv4RKaBjPX52PKUDxmvfMGEBjoozskkcieYlkp03z0RNZpT4OYcOd4hVzq8R7ZxdOFKZWkBKST8tH_692bct2eWCZGqxEH2CqP3AYWJvaQCFfDc0IutM6Rj-U09KIPjPRnc7FV26Wdj10BaRhnKyirh5JjBQBiUijZJ9qa0yX2qAvF17eId_iMnLYZJUiOJFMccNsS6Lx1I4XZzBbvOxVip14Ng8wGLJfXvx15u76cOB7dAeH7XvgpoWs6R_tsGa3dqAcv4R_28w7gc_bd-NL4b_nyhs9ocwFRphYixX3lTUlkEH0Nokh_o6hHppEenQieUgyV9aZx_s1E0SmDtT5e5jzWNn35PIXxCXehg937DHDABmN_O-g/https%3A%2F%2Furldefense.com%2Fv3%2F__https%3A%2Firc.w3.org%2F__%3B%21%21ACWV5N9M2RV99hQ%21ZvVx1wh89EAXhBiorHpgvdpQRlEtQPxaEsJbJ7_Q3MrxtnQGs5lwbIC34ybOl3ZsYw%24>, in channel #wcag3-protocols

Where to find more information:

These and additional details of our work, including minutes, current, and archived draft documents are available on our subgroup wiki page here: https://github.com/w3c/silver/wiki/Protocols<https://secure-web.cisco.com/1Hf7szbnvXj1QLrtDEE6rG7_4x2W4mjJ7gAH2zvGGTBAbMmVNyz6tNiVnJpgd9TFfgBYw-5X5beCKFq8To4sjXq5xKpPTu6lTPbVtQCROAEQX2H14eAfdGsixIcNkARfu_TdzOlXB2BtBTemQPxcXpo4S12Y-4CgkPapDOcFlsz_e2DPX921mCkxlRfEG3xA4WGbR2wnaXOq-GC0VY3HU4Sck16zbHzKiI8QodbeWTkTMGdniQitpC8iSoGJF48TnI5OZ8tbhVcMjwXUOQt14z-iNhOIUsIpFz-5ow-jynMKi6Mq_7zdAGSVV6G1TTXsuWCPJHlx1EjNvWpzbsC1zUPg5HaKNl-wEDX_9mZYhXaZcZZ-zl4Be7FvruDKdvZKxh4VXzFMOXaVjWLwcXaplsafajbRCfp75gHc6l-hgavxjhZapqtoVDIzI58t9uZTDbunnYgDhQmDavi7BtMoboQ/https%3A%2F%2Fgithub.com%2Fw3c%2Fsilver%2Fwiki%2FProtocols>

If you have any questions at all, please don’t hesitate to reach out. Looking forward to our next meeting!


Jaunita George, JD, PMP, WAS (she/her)
QA-ADA Analyst III, Product Engineering & Delivery Services (ISD)
DHS Certified Trusted Tester (TTV5)
[IAAP WAS circular badge and horizontal name logo for International Association of Accessibility Professionals (IAAP) Web Accessibility Specialist (WAS) credential. To the left is a dark blue circle with three lines of centered white text that read: IAAP Certified WAS. There is a smaller light blue circle that surrounds the dark blue inner circle that designates the WAS credential color scheme. To the right, two lines of dark blue text. Top text reads Web Accessibility Specialist, second line reads International Association of Accessibility Professionals.]<https://www.accessibilityassociation.org/s/wascertification>
Navy Federal Credit Union, 820 Follin Lane, Vienna VA 22180
w 571-391-0356 • c 571-422-2661 • nfcu.webex.com/meet/jaunita_george<http://nfcu.webex.com/meet/jaunita_george>
Facebook<http://www.facebook.com/navyfederal> | Twitter<http://www.twitter.com/navyfederal> | YouTube<http://www.youtube.com/user/navyfederal> | Instagram<http://instagram.com/navyfederal> | LinkedIn<http://www.linkedin.com/company/11568?trk=tyah>

[Navy Federal Credit Union: Our members are the mission.]    [Digital A11ies -- Working Together for All]

[Fortune 100 Best Companies to Work for 2022]
©2022 Fortune Media IP Limited. Used under license.



--
John Foliot |
Senior Industry Specialist, Digital Accessibility |
W3C Accessibility Standards Contributor |
"I made this so long because I did not have time to make it shorter." - Pascal "links go places, buttons do things"


--
John Foliot |
Senior Industry Specialist, Digital Accessibility |
W3C Accessibility Standards Contributor |
"I made this so long because I did not have time to make it shorter." - Pascal "links go places, buttons do things"