Cognitive and Learning Disabilities Accessibility Task Force Teleconference

05 May 2022


Lisa, Rashmi_, Jennie, ShawnT, Roy, JustineP, Le, julierawe, BeccaMonteleone, cweidner, kirkwood, Rachael, Jennie_, Rain, Becca_Monteleone, JohnRochford, Rashmi, DavidSwallow
Kris Anne, David F, Justine


<Rain> trackbot, start meeting

<Rain> meeting: COGA TF Weekly Teleconference

E.A. present

<Rain> scribe: EA

Deborah Bosley has also joined the session as a guest.

<Jennie> preset+

New members intro?

Rain described her jobs and her work with WCAG and then introduced new members

Becca Monteleone - Asst Prof at University of Toledo research interests in clear language and works in Disability Studies, Feminist Science and Technology Studies, Sociology of Disability

ICCHP conference (hybrid and/or face to face)

Deborah also introduced herself as having similar interests and in paricular in Accessibility studies.

ICCHP / AAATE taking place 11-15th July in Lecco Italy

<Rain> Link to ICCHP - https://www.icchp-aaate.org/

David F and Rain co-chairing a session at the conference. Several others will be attending including those working in immersive captions and Lisa was mentioning that there might be worth having a task force presence at the conference - hybrid and face to face?

<kirkwood> hybrid sounds interesting.

This has been mentioned to Michael and Janina - this would need to be announced by next week. This allows others to participate. So would this group like to plan a hybrid session?

<kirkwood> +1

+1 as I am attendin


<Jennie> +1 to hybrid, not travelling yet

<julierawe> +1 to hybrid

<JohnRochford> +1 I may go

Rain asked how would be thematically approach the session - other than educating people that the coga task force exists?

<DavidSwallow> +1 likely hybrid. Haven't attended ICCHP in years!

<Jennie> *Thank you for clarifying!

Jennie asked about hybrid face to face and coga having a presence. Rain wanted whatever is created is available both face to face and hybrid.

Rain then took Jennie's other idea of having two types of events 1. us as a coga group interacting without external people. 2. a session that would include external people

Jennie suggested that having a community group sponsored event that coga taskforce members join in order to get more buy-in

<ShawnT> What about TPAC?

and quality time as Jennie found this very beneficial when she last attended an event with Coga members

TPAC is Sept/Oct is more controlled but ICCHP/AAATE is less specific.

John K big +1 to leveraging the community group - thinking through them to make our work more effective - get more involvement.

Julie suggested that this could be a moment that one could collect more support for internationalisation

ShawnT mentioned that some people may have difficulties getting funding for conferences

Rain mentioned that Lisa was exploring some funding avenues

<Zakim> Rachael, you wanted to say the survey for tpac is out though

Rachael mentioned that it is important to keep TPAC in mind.

Update from AGWG

Rachael updated AGWG - visible controls has been deferred and then there was the discussion about the next charter - WCAG 2 looking ahead to WCAG 3. There is a survey for people to complete

<Rachael> https://www.w3.org/2002/09/wbs/35422/one_or_two_groups/

This survey closes Monday night before the next meeting. This a draft - nothing finalised. Rachael is happy to answer questions.

Julie mentioned that she had not way to gauge whether splitting into two groups would ease the burden. There is so much to do but if you split into two may cause more work rather than less - not sure would be better.

Rachael said there may be a don't know - legitimate to admit not knowing what would help

<kirkwood> +1 to Jennie very good point

Jennie mentioned that if something is missed and you are not really understanding what has been taken forward - do put it in the survey as by any thought or idea that is brought forward this would be helpful

Rachael is stressing that surveys are taken note of as usually survey responses are read aloud at the meetings but this time this survey is so big things will not be read aloud

John K very much agrees with what has been said but also could there be a pros and cons of splitting into two groups.

Rain said that it was helpful to know what would come under each model

Rachael - previous survey had the pros and cons..

<Rachael> WCAG 3, W3C Decision Policy Minimal WCAG 2 maintenance, WCAG2ICT, All taskforce work

<Rachael> WCAC 2 final, AG Decision Policy, Extensive WCAG 2 maintenance, WCAG 2 support materials, WCAG2ICT, ACT work

<Rachael> WCAG 3, W3C Decision Policy, COGA, low vision, and mobile work

There are over 500 issues for WCAG 2 so with two groups this could be achieved whilst the other group dealt with WCAG 3 - use W3C approach

John K thanked and Julie asked if Rachael could share her thoughts.

Benefits of staying as one group - long history and no change allows for real cohesion - less risk

Two groups have value in doing both bits of work - WCAG 2 still the standard so needs to be maintained. One group cannot do it all and there are different work processes one is transformative and the other is tried and tested.

These are different ways of working and it hard to flip between the two. Decision policy is good for a process that works in law as it does in WCAG 2 at the moment but this does not help with a transformative process such as WCAG 3 - needs more iterative and not a final decision making process. T

This allow for better public perception

Julie asked about risk factors with the splitting possible diminishing the process and work ahead.

Rachael admitted this has been asked if there is a lack of support for WCAG 2 but people are being lost because it is not working at the moment

Risk to coga is more about how we can get more into WCAG 2 but we can achieve this in WCAG 3 as it will have to adapt. WCAG 2 will clean up what is there but not add new items.

Julie still concerned about the amount of momentum in each group - one will atrophy possibly and this could be a worry for coga.

Coga will be going towards WCAG 3 so Rachael felt there was no risk

John K suggested that the risk when breaking away may well be more about the legal aspect of WCAG which exists are present. This will be difficult to maintain if we move away from the WCAG 2 foundation.

John R added that if it goes to two groups how will the chairs work

Rachael said that the present chairs herself and Chuck would go to WCAG 3 and Alastair would stay with WCAG 2

Check in on subgroups

Rain mentioned the subgroups images with Jennie and they will have an ongoing meeting with another artist who has joined

Jennie mentioned that they will be regrouping later in the month

Rain said that the coga work with WCAG 3 are seein in the Clear Language subgroup - test cases and the other subgroup is access to help - drafting guidance on WCAG 3 - what is available. The third subgroup is research into mental health and its intersection wtih cognitive accessibility.

Rashmi mentioned that they are looking into articles related to elearning, social media and others that are available in the research domain and still looking for help with this work.

Rain said that there will be three new subgroups namely: explore structural changes to Content Usable, research plan and strategy and the third one is a test plan strategy.

Anything to do with WCAG 3 is how success criteria will be tested.

Rain asked if people could let Lisa and herself if anyone has any interest in the groups.

John asked if there was a voice interaction or an AI subgroup.

There is an AUR - subgroup working AI voice interactions

<JohnRochford> FYI: I am working with Microsoft on A1-powered voice interactions.

John R said this is very important as large companies are working with these technologies and especially when it comes to chunking information and thinking about cognitive overload etc. Media fragments

<JohnRochford> AI not A1

Clear Language - tests plans, scoping - https://www.w3.org/WAI/GL/task-forces/coga/wiki/Subgroups/WCAG_3_Coordination

<julierawe> WCAG 3's Clear Language scoping exercise: https://docs.google.com/presentation/d/1PsPVWTEjDwZUfevlvywI6NeepAxkveU8_SxYuCVAY80/edit#slide=id.g120a12cbea9_0_152

<JohnRochford> John K, not John R said the most recent minuted comment.

<kirkwood> Sorry that was JOhn K: said this is very important as large companies are working with these technologies and especially when it comes to chunking information and thinking about cognitive overload etc. Media fragments

Julie asked Rachael whether there should be more work on the scoping exercise.

Rachael really wanted to know how it could all be tested.

Julie suggested it was proving hard to get down to the smallest level - what do we mean by many of the concepts - often there are issues with the number of tests needed for each of the concepts

Julie suggested next step would be to take one concept and try to find suitable testing mechanisms.

Rain wanted to wrap up the discussion in order to transition to the next topic working with the plain language

Short break - back in 5 mins

Rain introduced Julie to frame the conversation

Subgroup meeting (11am-Noon ET): Clear Language, special guest: Deborah Boseley from Plain Language Group

Meeting is now being recorded by Rain.

Julie introduced Deborah as a plain language expert and is now willing to share her knowledge especially in terms of testing. Chair https://www.theplainlanguagegroup.com/

Deborah mentioned her work with the international organisations and stated how important the accessibility aspect is to all elements of our lives

Deborah is going to talk about 4 areas including testing, grey areas where there are no literal rules and then nationalisation of plain language.

definition of plain language https://plainlanguagenetwork.org/plain-language/what-is-plain-language/

https://www.iplfederation.org/ also has the definition

Testing - first using readability testing which is limited - Flesch Kincaid - counting syllables - fewer syllables easier to understand, words and words per sentence - 15-18 for easy understanding

But issues are that low syllables do not necessarily mean it is easy - ennui v boredom - opens the door - just shows grade levels (USA)

Flesch reading ease 0-100 higher score the easier to understand 90-100 comic books, 60-70 local newspaper - standard for Adults in USA (8th grade) 30 and below e.g. Harvard Law Review.

<Becca_Monteleone> I worked with some data visualization journalists to talk about some of the limitations of readability algortihms that rely on syllable counts/word difficulty recently, similar to the issues Deborah was discussing: https://pudding.cool/2022/02/plain/

Again this just opens the door to conversations but limited. More significant is usability and qualatative - one on one interviews of the intended audience - if you test with 8-10 people you will get around 80% of what you need to find out.

qualitative - read/ listening and ask about understanding then also find out about items that give clues about finding the answers - so have both open and closed questions

<Rain> a?

Not everyone in the audience is going to understand something. transcripts and notes are made from the information gathered in the interviews. Revised content is also tested so that there is a comparison with the original. Testing is iterative with retesting

Grey areas include the qualitative testing - use of active voice is what is used when learning to speak and is easier to understand. Shorter than passive voice.

Not all practices are going to be perfect for what you are communicating. Use passive voice when you do not want to blame someone - it disguises agency.

Also if you do not know who was the agent then passive is allowed. Active voice can be used when you have identified the subject.

Also if you do not know who was the agent then passive is allowed. Active voice can be used when you have identified the subject.

Vocabulary - jargon, acronyms etc need explanations - you do not know what people may understand - avoid idioms and metaphors.

Do not translate well into other languages.

May also confuse the average person

John asked about legal content - Deborah said that the legal folk need to minimise risk for the company - terms and conditions, privacy documents, pensions these can be in plain language

But companies do not get into trouble for lack of compliance so Deborah tries to seek out empathy within a company to ensure that there is an understanding about the right to understand

It is possible to make legal language plain - such as taking out the double negatives etc

Breaking up sentences can also help - reformating in ways that help such as lists.

Deborah also said her company approaches with empathy and - can show how the more legalese is used the higher grade level is needed

Last area content length - legacy language, redunancy can cause more complication - it may be that plain language may mean examples are needed or extended by acronym explanations. But on the whole the aim is to offer shorter content witih less overwriting - too much knowledge rather than keeping it simple.

Internationalisation - PLAIN https://plainlanguagenetwork.org/plain-language/what-is-plain-language/

Many countries across the world have their own plain language organisations with documents working to simplify government documents

USA has the Plain Writing Act and CLARITY - works with plain legal language https://www.clarity-international.org/plain-legal-language/

Plain language does not just mean plain English it applies to all languages in different ways and culture.

If you have plain language at the outset it is much cheaper to then translate the content.

Deborah ended by offering a chance to ask questions.

Julie asked about the user testing - if you were user testing 8-10 bank customers do you consider age?

How do you select a diverse group

Deborah said it depends on the document and do not look at race but English as a second language and have an age range unless there is a specified group.

Some may not allow for user testing even though it is the best way to know whether the information is understood.

Real data is only possible through user testing but hard to achieve with companies rather than the quanitative tests

sorry - quantitative

<Becca_Monteleone> +1 to John R's comment - and of course recognizing that people with ID are part of the "general public"

John R asked about using cognitive impairment (intellectual difficulties) and learning difficulties as being able to provide information about content and therefore help everyone

Deborah offered to run any testing with groups or with content.

<Zakim> kirkwood, you wanted to ask best practice with acronyms: does one just put the definition in line, in a footer, once, each time? second question, do you have any numbers around

John K do you have best practice for handling acronyms - where and when should the definition be provided?

Also easier aspects of internationalising content for plain language

Deborah said first time acronym always explained and then depending on the length of the content you use the full name and acronym again in context to aid memory later in the text depending on the length.

<ShawnT> +1 to not using footnotes and putting things in context!

Acronym used on multiple pages or sites - give the definition each time for each web page - always in context

John K asked are there any numbers of people saying 'Translation from plain language is easier' - are their any studies on the subject

Becca asked about the conversations Deborah as with clients as to who really are their audience.

Deborah said it is ongoing as there is a tendancy to think that all customers may be of one type - but there is a need to ensure that a wider knowledge base is considered

They need to be thinking about accessibility and perhaps only think about education levels or gender - common demographics not the ones that are relevant to understanding - the types of people.

specific populations have different needs and these criteria need to be recognised but not every aspect can be tested - Deborah said she felt lucky if she could just test once on a general group

<Jennie> Fantastic information!

Rain, Lisa and Julie want to continue to discuss with Deborah how collaboration can occur

<Rain> RRSAgent: make minutes

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2022/05/05 16:06:07 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision VERSION of 2020-12-31
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/over 1000/over 500/
Default Present: Lisa, Rashmi_, Jennie, ShawnT, Roy, JustineP, Le, julierawe, BeccaMonteleone, cweidner, kirkwood, Rachael, Jennie_, Rain, Becca_Monteleone, JohnRochford, Rashmi, DavidSwallow
Present: Lisa, Rashmi_, Jennie, ShawnT, Roy, JustineP, Le, julierawe, BeccaMonteleone, cweidner, kirkwood, Rachael, Jennie_, Rain, Becca_Monteleone, JohnRochford, Rashmi, DavidSwallow
Regrets: Kris Anne, David F, Justine
Found Scribe: EA
Inferring ScribeNick: EA
Found Date: 05 May 2022
People with action items: 

WARNING: Possible internal error: join/leave lines remaining: 
        <scribe> Rain mentioned the subgroups images with Jennie and they will have an ongoing meeting with another artist who has joined

WARNING: Possible internal error: join/leave lines remaining: 
        <scribe> Rain mentioned the subgroups images with Jennie and they will have an ongoing meeting with another artist who has joined

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)

[End of scribe.perl diagnostic output]