W3C

– DRAFT –
Protocols subgroup

15 October 2021

Attendees

Present
Fazio, GreggVan, jeanne, jenniferS, JF, Rachael
Regrets
-
Chair
-
Scribe
jeanne, JF, Rachael

Meeting minutes

JF: no agenda this week. I want to continue last week's agenda by trying to define protocols
… it could be processes. There are different types of processes.
… Outcome-based processes gains momentum
… we don't want it to be too broad, but it should lead toward better outcomes
… I want to test that concensus

JA: Whatt we want to achieve is to design content that improves testability
… COGA document we agree should be a protocol
… there are other documents that might be protocols
… we know Inclusive Design, BBC guidelines, they are all documents that are broadly used and improve accessible
… are they too small, are they not protocols? Or are we looking at ISO standards?

JF: We need to decide if external protocols are in scope. What specific things must a protocol include to assess it?

<Zakim> jeanne, you wanted to suggest a narrower scope than a broad definition of protocols

Jeanne: want to suggest a multiple part process
… let's look only at W3C protocols - go through process on how to handle and score
… then next step: what did we learn? Apply to "what ar the considerations of a protocol to make it more broad"

GV: If it has an outcome, it has to be measurable. If the outcome is measurable, it is accessible. If everyone comes out with the same outcome than it is reliably measurable.
… I want to ask for examples for a process of steps.
… I don't think tthe COGA document is a process, it is advice
… Protocols is equal to guidelines, because what was mentioned were guidelines
… I thought a protocol was more tightly defined
… how would we be using them if protocols are a group of recommendations?
… If you have certain things you have to do and get credit for doing more, then that makes sense
… What are our examples? Start talking about it in the abstract and trying it in reality
… then we can see what works and what doesn't and make progress

<Zakim> MichaelC, you wanted to suggest requirements for protocols

MC: I suggest a hybrid of what we are hearing, which is that this group develop a list of requirements

jeanne: -1 for us developong a protocol, rather than approving existing protocols

MC: I think this group could develop a W3C protocol

JF: My mental model would be COGA Content Usable and Plain Language.gov
… there are two types of guidance, Testable, measureable, repeatable and guidance that is less measurable
… that is the origin of Content Usable, because they couldn't write testable, measureable, repeatable true/false statements
… scoring is off the table
… unfortunate, because scoring is integral

GV: Even if we only decide internal vs. external, we have to decide how we are going to use it.
… how do we incorporate really good non-measurable advice?
… if you can reliable score, then you have measurability and testability, then that is guidelines
… if you can score, then it is testable, and that doesn't address this problem
… there is a lot of good advice for making things more accessible that are not measurable
… how are we going to use this?
… these appears to be that we can't get important things into measurable form
… we require the things we can require, and we give extra credit to people who try to do more
… I don't know if we can require extra credit. COGA would object, because they don't want their needs to be optional.

JF: We encourage people to do it, by making it worth their while

<JakeAbma> presdent+

JF: proposal is entitled "Protocols and Assertions"
… ePub has a manifest file with meta-declarations, which include assertions of the accessibility
… the digital entity could have a meta data document -- a declaration of adopting this
… "I promise to use the COGA document while developing content". Then people can see whether they are following the COGA document
… it is subject, so we can't measure it reliably
… "we know it when we see it" but it is difficult to measure
… WCAG3 is moving toward ACT true/false statements, so it is no longer guidelines, it is testable statements
… [scribe missed final statement}

JA: We in the Netherlands have a model where if you don't comply to WCAG, you have to state what you are doing to mature in your process.
… you are judged by what you are doing
… the UN using a similar process, you must provide reasonable accommodations. You have to make a statement on what you are doing.
… be judged by the public

<JF> +1 to Jake - this is the mental model I had too

JA: in social media, in the public view, and maybe in court

<Zakim> GreggVan, you wanted to talk about assertions

GV: speaks to history of assertions in the past

<jeanne> GV: It is fine to make a claim, but then they aren't going to claim that they meet WCAG because of the legal oblications. The VPAT doesn't say you meet WCAG, but that you support it. It has no legal meaning

GV: may have a path forward however

<jeanne> ... I see a path, that the Netherlands don't have the same legal system where you could be in court and have to show that you meet the guidelines

<jeanne> ... if there was a way perhaps that you do the testable and then consider or incorporate the stuff that you can't meet but you have to use

<jeanne> ... like take a COGA and push it out in your organization

DFR: talks about ISO 9000 - "Say what you mean, and do what you do"

it's a framework for policy adoption - voluntarily

So how do we award a point?

what does a point mean?

<Fazio> ISO 9000 https://www.iso.org/iso-9001-quality-management.html

GV: ongoing concerns about measurability and points

Jeanne: Really good discussion. Influencing my ideas of how we can do this.

<Zakim> jeanne, you wanted to propose that we put Protocols at Silver level and we don't give points, we just say "implement 2 or more protocols from this list"

Jeanne: when we originally looked at the silver architecture and, instead of making SC A, AA, or AAA took the totality of what can be measured and allowed pieces of it to meet different levels (bronze, silver, gold). The silver and gold were intended/discussed as going beyond the testable. I was thinking of protocols as points but Gregg's argument is a good oe.
… what if we use protocols as the silver level? You must use at least 2 of the following protocols to claim silver. If you want to claim silver, then you should...

JF: I want to push back a bit. If I can get to bronze without protocols, then why bother?I would like protocols to be part of the score regardless of level.
… I am thinking about the netherlands. We may not be where we want to be today but we commit to doing this. We let people be bronze because they adopt a protocol. I think that is a way to normalize protocols. Avoid putting them in the rare level.

JakeAbma: We talk a bit about measuring. I hope you the point. Measuring for the BBC has never been their priority. Instead they care about accessibility. They create protocols and following them and because of that they become known for their accessibility.

<JF> +1 to Jake - this is *exactly* how I was envisioning this activity going forward

JakeAbma: In the NEtherlands, the director changed the system. Following WCAG didn't work because noone could pass it. People didn't feel motivated to work on accessibility. Now Its not that bad if you don't score 100% on WCAG, just tell us what you do and place it in a public statement. The groups that care about accessibility has never been bigger. Changing the goal increased the participation. That moved the needle. I care about the

measuring but it doesn't move the needle.
… that has been proven many times. Not measurable stuff. Crediting that you do your best.

GreggVan: Jeanne, what you were talking about was really good. You said if they implement a protocol. What does implement mean and what is enough? When do you get to check it off?
… you mentioned Lainey and use of milestones. If you look at them, they are very measurable. This % done by this date. Interesting but still measureable
… goal is to get to required but not measurable. Problem, you can't do it. I do want to think about what Jake just said. I want to think about, can we have both these levels and points. Some things are important and some are not. If people are not going to get 100% of WCAG which they don't, how do we ensure they get the really important ones.

Rachael: Someone needs to take all this and put it in a wiki. Capture the main points along with the pros and cons.

JF: I agree. Have a wiki. Teased out a lot of good points.
… we've teased out a lot of good points.

Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).

Diagnostics

Succeeded: s/"Say what you do, and do what you mean"/"Say what you mean, and do what you do"

Maybe present: DFR, GV, JA, JakeAbma, MC