Protocols Sub Team - Dec. 17, 2021

17 December 2021


Chuck_, Jaunita_George, jeanne, Jennifer, JenniferS, JF, Mary_Jo_Mueller, ShawnT

Meeting minutes

<JF> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Protocols#10_December_2021

<JF> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Protocols#10_December_2021

<JF> https://www.w3.org/2021/12/10-silver-protocols-minutes.html

<JF> https://www.w3.org/TR/coga-usable/

suggested short topic: What's our calendar for the rest of this year and into next year?

<Zakim> jeanne, you wanted to say that there are volunteers who are testing experts to help make them testable

<JF> [discussion about 'repeatable']

<Zakim> MichaelC, you wanted to talk about documenting, following, qa, and auditing of protocol

<JF> MC: question around quality assurance

<JF> MC: also questions around auditing

<JF> who has responsibility for what?

<JF> JG: thinking about orgs - holding partners and suppliers to the same standard

<JF> we need a means to generally measure some type of compliance

<Zakim> MichaelC, you wanted to separate auditing adherence to protocol from auditing success of protocol and to define ¨following¨

<JF> MC: what is being audited is the "steps that the protocol requires"

<Jaunita_George> If we're using protocols to define adherence to a standard, then we need to have a way to ensure someone is meeting the spirit of the standard in a way that's measurable and repeatable

<Jaunita_George> +1 for decision trees

<Zakim> Chuck_, you wanted to say that in order to be crisp... we need an agreed upon defintion

<Zakim> MichaelC, you wanted to realize implementers might use multiple protocols, not ¨a protocol¨? and to introduce ¨evaluatable¨ alongside ¨testable¨

jf: I've referenced 3: alt text tree, plain language, content usable.

+1 to stating the unstated assumptions!

<Jaunita_George> +1 for adding protocols to MM

<Jaunita_George> We need to write a definition of "protocol," how we will use it and intended outcomes of using it

<Jaunita_George> Protocols would allow someone to conform to a guideline

mc: Protocols need measurable steps that can be audited.

<JF> Protocols need measurable steps to use for auditing

mc: "Verifiable" that the steps are being followed.

<Jaunita_George> Conformance with the protocol should be measurable

jf: Making content usable, they don't have steps, they have user stories.

<JF> Verifiable that you are following the steps

mc: protocol is like "%" of web dev resources to implementing guidance, or you are making sure developers have read doc, ensure that the guidance gets meaningfully followed.

<Jaunita_George> +1

<JF> MC: that would be why Making Content cannot be a Protocol (no steps), but the process and policy that ensures that guidance is meaningfully follwoed

jf: I like that we can.... to use a doc and build a protocol around a doc.

mc: That's an example of what we might do.

<Jaunita_George> Protocols can take the form of decision trees, perhaps? That was mentioned before

jf: You've tapped into something. Process and policy that ensures guidance is followed, that's what we want.

jf: Part of it was making the public declaration. You've put a "stick in stand", saying you are doing. But doing what?

jf: In maturity model, we can define the process and policy for making content usable. If you implement that doc, here's the process and policy.

jf: It's demonstrating that you understand the thrust of the doc.

<Jaunita_George> Process for adopting policy -- that's a great idea. Can be a supplement to a document like "content usable"

mc: There can be different process and policies. Training... audit points, testing points.

mc: There could be different protocols for achieving the same guidance.

mc: Measurable steps of protocol are what you are following.

jf: The steps themselves, the outcomes is what you can't accurately measure.

mc: And that's where we get into evaluation.

mc: I had in mind a user testing example, with steps such as "here's some abilities that need to be included". We should be able to define to get meaningful results.

mc: A public statement... following a protocol requires a statement, I don't think a public statement is required.

jf: I'll push back on "public statement".

jf: The public statement piece is how we work it into scoring and conformance. That is what contributes to bronze/silver/gold.

mc: If for conformance.

jf: If we don't get legislatures behind us, we will not go anywhere.

Juanita: A supplement maybe to content usable for example. If there are multiple protocols, we need something in the definition that all protocols are of equal quality.

Juanita: So that there aren't 2 protocols where one is easier but doesn't achieve same quality.

jf: I think that there may be 2 separate protocols that get us to where we want to be. Plain language, there's something in US and equiv in EU.

<Jaunita_George> +1

jf: From my perspective as a Canadian, either one would suffice.

jf: There may be times we have more than one protocol that drives to a single outcome.

Jaunita: Can we come up with draft defintion?

<Zakim> MichaelC, you wanted to s/equal/sufficient/ quality and to ask if we want to gatekeep protocols, or set requirements for them? ref techniques

mc: We can move to a more formal doc. Equal quality, I prefer "sufficient". A very high quality protocol and a good enough quality.

mc: Need a minimum quality. We can't require them to be equal. We might circle back to later. Should W3C be a gate keeper?

mc: I think a lot of big orgs would prefer to develop their own in-house protocols. In such case we need a defined set of requirements.

mc: That's a bigger conversation, but I want to put a finger on it.

<Zakim> Chuck_, you wanted to say lets talk about our calendar

jf: In my original preso, I accounted for multiple protocols, vetted and non-vetted. The idea I had included a scoring piece, where different protocols would accumulate points.

jf: vetted protocols would get more points than non-vetted.

jf: Accumulating multiple protocols builds points.

jf: Jaunita - we've more ideas on table. We left last week with very broad definition.


jf: Are we happy with that as a starting point that we can add to?

Jaunita: <screen sharing>

scribe note: rapid brainstorming occuring

<Zakim> Chuck_, you wanted to say it's adoption is measurable

JenniferS: If a protocol is used, I expect the user to show their homework.

jf: Concerned with including "legal scrutiny" into the requirements.

jennifers: I'm ok with pushback.

<Zakim> Chuck_, you wanted to say some of us need to jump over to silver.

Twas the night before chrismas, and all through the house, not a standard was stiring, not even WCAG

Minutes manually created (not a transcript), formatted by scribe.perl version 185 (Thu Dec 2 18:51:55 2021 UTC).


No scribenick or scribe found. Guessed: Chuck_

Maybe present: Jaunita, Juanita, mc