Silver Conformance Options Subgroup

18 Nov 2021


GreggVan, janina, JF, maryjom, PeterKorn, shadi, ToddLibby, Wilco
Aslan, Jeanne
PeterKorn, shadi

Meeting minutes

Agenda Review & Administrative Items

[Janina walks attendees through the agenda]

JS: Shadi also has a partial write-up on the wiki which we can review in the coming weeks
… no meeting next week, several will be comatozed by Turkey


SAZ: first pass, all examples are there for discussion
… not sure how to get first reviewed

JS: will come back to this on 2nd Dec meeting

GV: coming back to the sampling
… also need to talk about mass sampling
… testing massive number of pages
… maybe a different discussion item for the future

JS: agree, we will probably need to tackle this
… but might not be able to get to it this year

Analyzing conformance Considerations https://www.w3.org/WAI/GL/task-forces/silver/wiki/Conformance_Short_Summary_Strawdog

GV: was not intended as a proposal
… just an attempt to find a common thread
… what is it that we're trying to do?
… understand the different needs and desires
… the second part is to summarize the understanding
… initially doing it for myself
… first parts is the higher-level goals
… in WCAG 2.x tried to have Level AAA
… but that mostly just got cut away
… and people didn't do it
… another goal is to get accepted by other regulatory and standards groups
… third goals is what content can influence and what it can't
… for example, can't be responsible for the accessibility of emails written by users of a webmail site
… fourth goal is that page that is part of a process needs to be considered together
… like a checkout page in a shopping site to complete the transaction
… fifth is to recognize there is no absolute accessibility
… then lastly, to make it as simple as possible
… already complaints about the currently complexity of WCAG
… sometimes even debates within the Working Group about what something means

JF: agreement with much of this


JF: propose goal for bronze level to encourage people to implement things like COGA guidance

GV: sounds like a how, rather than a goal
… maybe part of the first goal?

JF: looking for something more specific

JS: concerned we can get into the weeds
… for example, which level and such
… discussion in process, let's not define it that specifically

GV: why is the first goal not sufficient

JF: want people to actually do these things

PK: goal 4 seems overly page-centric

<Wilco> +1

PK: later on you speak of parts of a process, which is less page-centric

PK: another goal was to better reflect the lived experience of people with disabilities
… to reduce definition of requirements that do not really impact accessibility

JS: technical violation that doesn't really have a real impact, like empty iframe?

PK: yes but trying to stay higher-level than that level of detail

PK: might be other goals from Silver research work
… would be good to look back at these

<Wilco> https://www.w3.org/community/silver/draft-final-report-of-silver/

GV: now numbered goals, and added new goal #5 to address your latest comment

PK: not very settled on the language

<PeterKorn> Suggested alternate text for #5: "That something that fails technically, but which does not have any real world impact on the lives of people with disabilities, can still conform"

GV: would be good to re-review these

WF: how do we define "objective"?

GV: something is objective when it has a high inter-rater reliability

<JF> if we are wordsmithing, then I'd like to see #1 modified to: "Will get web designers to do more than what has been able to be put into WCAG 2.x"

GV: people who know what they are doing agree to a large degree

<JF> i.e. do more than just "look" - DO

WF: agree with high inter-rater reliability as a goal

<Zakim> JF, you wanted to note that US federal government sites TODAY have to apply Plain language

JF: understand the difficulty and the many attempts of trying this
… governments have to do this today

GV: which law is that?

JF: plainlanguage.gov

GV: will check it out

JS: not sure we need this deep dive now

<JF> A HUGE +1 to Wilco

<JF> this is *exactly* my concern

WF: one thing that may raise concerns, is that things that are testable will make into regulations and other things will be on a level not required by law
… not sure if this moves us forward from the current situation

GV: concerned about reducing testability may rule out adoption by policy makers
… do people think we can get passed this?

<JF> @Gregg, I *DO* believe we can get there

JS: working some kind of an understanding to get more requirements

PK: lack of clear testability is a huge problem
… like the idea of medals or such
… think that the level for requirements will depend on the nature of the product
… what is within reason or outside reasonable control
… for example, healthcare might be different from entertainment
… does not change the definition of what technically conforms
… but maybe which levels of conformance are acceptable for each
… not convinced that 3 is the magical number

WF: don't think we can put normative requirements that can't be tested reliably
… the question is why can't we figure out how to make these more testable

GV: added definition for objective and testable

<Zakim> JF, you wanted to disagree cite: ISO 9001

JF: disagree with Wilco, you can't test or measure ISO 9001

GV: you can absolutely test and measure ISO 9001
… and you can get certification and such

JF: not based on sloppy unit tests

JF: different way of looking at conforming

GV: is ISO 9001 required by law?

JF: no but adopted by companies

SA: discussion of process-based standards, that would be a separate standard.
… maturity model could be that separate standard. Feel comparison between a bulding code & ISO 9001 is Apples & oranges

GV: very big difference between voluntary and regulatory standards
… as soon as requirements are going to be regulated, a whole set of different rules come in

<JF> https://blog.libryo.com/iso-9001-2015-legal-requirements

WF: would love several more people debating this
… seems only few of us discussing it

DL: my first time here, hello everybody
… from Toronto area
… education and digital publishing
… previously involved with IDPF and now W3C on digital publishing

MJM: first time I've seen this proposal
… not sure it addresses web content and web apps
… always have bugs, and WCAG has been all or nothing

<PeterKorn> +1 I think there are important things not yet in this doc., and they are things we have been working on previously in this group.

MJM: not conducive to progressing accessibility
… also some things are not easily testable
… becomes daunting on large-scale websites
… would be helpful for developers to show progress over time
… would help get more buy-in

TL: conversation is great, don't have much to add to it today

<JF> @Gregg: The Plain Writing Act of 2010 requires federal agencies to write “clear government communication that the public can understand and use.” https://www.dni.gov/index.php/plain-language-act

Glossary Initial Draft Definitions https://www.w3.org/WAI/GL/task-forces/silver/wiki/Conformance_Glossary_Candidates

<PeterKorn> A potential missing goal is around helping / inventing continual improvement.

JS: let's try to advance the terms "conformance" and "compliance"
… helps us to distinguish these two concepts
… what we expect from a technical spec vs regulation

<janina> https://www.w3.org/2021/09/draft-wai-glossary

JS: hopefully not controversial concept
… but need to work on the definitions
… tried to integrate existing definitions
… will discuss in 2 weeks

GV: we ought to get some definitive decisions on the goals

<ToddLibby> I've got to get to another meeting. Thanks, everyone.

GV: if we can't get agreement then will go in circles
… need clarity if regulation requires testability or not

PK: invite folks to look at the other doc shared at the beginning of the call

PK: https://www.w3.org/WAI/GL/task-forces/silver/wiki/Substantial_Conformance/Example_Scenarios

<JF> We have evidence today of a regulation with Unmeasurable requirements: https://www.govinfo.gov/app/details/PLAW-111publ274

Minutes manually created (not a transcript), formatted by scribe.perl version 159 (Fri Nov 5 17:37:14 2021 UTC).


Succeeded: s/WF/JF

Succeeded: s/requirements come in/rules come in

Maybe present: DL, GV, JS, MJM, PK, SA, SAZ, TL, WF