<Rachael> https://www.w3.org/WAI/GL/wiki/Scribe_List

<jon_avila> scribe:jon_avila

Content Usable Update

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

Rachael: comment period on COGA document closes September 4th.
... Places were wording agreement was reached - COGA is setting up weekly meeting on Thursdays. Reach out to chairs or Lisa if you are interested in participating in COGA discussion.

Silver Survey Responses https://www.w3.org/2002/09/wbs/35422/2020-08-silvereditorsnotes/results

Rachael: will focus on Silver one week and WCAG 2.2 the next - pattern that has evolved. Today is a joint meeting AG and Silver TF.

Representative Sampling

Rachael: Want to make sure equal representation and conversations.

<Rachael> Note from survey: The AG WG and Silver Task Force recognize that representative sampling is an important strategy that large and complex sites use to assess accessibility. While it is not addressed within this document at this time, our intent is to later address it within this document or in a separate document before the guidelines reach the Candidate Recommendation stage. We welcome your suggestions and feedback about the best way to

<Rachael> incorporate representative sampling in WCAG 3.0.

Rachael: From results - 12 people - yes as written.
... 3 comments for changes. Peter are you here?

<Rachael> Peter's suggested addition: Representative sampling is an important tool in a strategy to assess how well an overall website meets a particular level of conformance by looking at a small subset of the site.

Sajkaj: Mindful that we have not considered - which of these specific proposals intend to be normative or informative to the claim and why. glad we are talking about scoring and sampling - but the 3rd question struck me. Are we really saying we are going to timestamp a conformance claim and then if it's challenged a major web presnence and recreate those conditions from that time - that might be difficult to work through. Maybe possible - bu[CUT]

sajka: We need to say something normative - but wondering if actual scoring should be normative.

Rachael: these are editors notes we are putting - agree that we don't all agree. An editors note is not normative.

Sajkaj: Part of how we help people assess or defensible in a conformance claim. Don't know how it fits.

Rachael: Purpose of survey was to continue momentum on deep dive. We could not agree on sampling for conformance. We see that in comments. We see importance of sampling. Comments from Wilco and Laura from removing statement in this document as they don't believe it belongs here in this document.
... You will see it in context when we provide a draft document.

MichaelC: Reluctant to educate in a note - prefer to link. We do choose to educate - Peter's isn't quite the wording. We shouldn't provide a value judgement when we are collecting feedback.

Michael: Purpose of note is that we don't have consensus rather than the remove clarifications

Sajkaj: on July 29th we had an idea of a broad note - that there would be more to come and that would be an opportunity to point to the challenges to document. That would be useful as we'd like to address some of those in WCAG 3. Maybe there is more than these notes we should consider.

rachael: Agree more notes will be needed
... Start with straw poll to move this forward - to move into draft without changes. Recognize that you will have draft in 2 weeks to provide detailed comments on.

<Rachael> straw poll: Move forward with editors note on representative sampling as written. +1 to moving forward, -1 to not, 0 to abstain

<MichaelC> +1 to ednote, -.5 to defining representative sampling, strong -1 to including a value statement in the definition

Rachael: This is not a CFC - we are saying the wording is conformant to be put in the draft of the first public working draft - it will come back to this group again.

<sajkaj> +1 to seeing it in context before final disposition

<Wilco> +0 would like to see it in context of a doc, still concerned

<alastairc> +1

<pwentz_> +0

<Laura_> 0

<sarahhorton> +1

Rachael: Go ahead with poll. 0 to abstain.

<jeanne> +1 to move forward

<Sukriti> +1

<Brooks> 0

<Lauriat> +1

<Detlev> +1 with Michael's suggestions

+0

<Todd> 0

<KimD> +1

<Raf> 0

Rachael: Seeing notes that it needs to be in context. Seems like we are ok to move forward recognizing that it will come back.

RESOLUTION: Include note on representative sampling in draft recognizing it will come back to this group for further review.

Testing Editors Note

<Rachael> One the goals of WCAG 3.0 is to expand scoring tests beyond a binary true/false choice at the page level. We have included tests within the sample guidelines that demonstrate alternatives such as rubrics and scales. We are also exploring including these with an ACT structure where possible. We will include additional example tests in a future draft of this publication. Our intent is to include detailed tests and techniques for each functional

<Rachael> outcome within the Silver framework.

Rachael: 11 to including as written. 4 with changes. Peter stated - I'd like to see larger context in which it sits. Wilco - like to remove where possible.

Wilco: May need to extend for more variety of formats - not limit our tests to current format.

<MichaelC> We are also exploring integrating these options into <a href="https://www.w3.org/WAI/standards-guidelines/act/">Accessibility Conformance Testing</a> format.

<alastairc> +1 we don't know if we'd update ACT or Silver, or even find a dead end

MichaelC: sentence got dropped related to ACT - pasted above. Good version we may need to version ACT or if we will run into road blocks. Where possible allows us to not be committed to a specific direction.

<Lauriat> +1 to Wilco, I think we can remove "where possible", especially since the start of the sentence notes "We are exploring…"

<Rachael> One the goals of WCAG 3.0 is to expand scoring tests beyond a binary true/false choice at the page level. We have included tests within the sample guidelines that demonstrate alternatives such as rubrics and scales. We are also exploring integrating these options into <a href="https://www.w3.org/WAI/standards-guidelines/act/">Accessibility Conformance Testing</a> format where possible. We will include additional example tests in a future draft

<Rachael> of this publication. Our intent is to include detailed tests and techniques for each functional outcome within the Silver framework.

<jeanne> +1 Michael

Wilco: Exploring is already conditional enough. We need to make sure some wording about requirements that it isn't just about true false conformance.

<jeanne> -1 to true false conformance -- we have a lot more nuance in scoring conformance

<Wilco> suggest: One the goals of WCAG 3.0 is to expand scoring of test methods or success criteria beyond a binary true/false choice

AlastairC: we are exploring including these in ACT structure. If you stop there you are either going to ACT or not - it is not a reasonable limit at the moment. We may find it suits some tests but not others. But theorectically some can be broken done into true and false - but not sure we want to limit our guidelines to that.
... Rather leave wording as is

<Rachael> straw poll: +1 including where possible, -1 to removing where possible, 0 no strong opinion

<jeanne> +1 to include "where possible"

<sajkaj> +0

<MichaelC> -.5 (lean against but not gonna die on that hill)

<Wilco> -1

<alastairc> + 0.5

<Detlev> +1 include "where possible"

<JustineP> 0

<Sukriti> 0

+0

<KimD> 0

<Todd> 0

<Laura_> 0

<sarahhorton> -1

<Lauriat> -.5, seems like wordsmithing to clarify something already non-committal…

Rachael: Remove for time being at negative .5 tipped scale toward removal.

MichaelC: We need to track on someone to bring it up again.

<Wilco> One the goals of WCAG 3.0 is to expand scoring of test methods or success criteria beyond a binary true/false choice

<Rachael> Suggested wording with change: One the goals of WCAG 3.0 is to expand scoring tests methods or success criteria beyond a binary true/false choice at the page level. We have included tests within the sample guidelines that demonstrate alternatives such as rubrics and scales. We are also exploring integrating these options into <a href="https://www.w3.org/WAI/standards-guidelines/act/">Accessibility Conformance Testing</a> format. We will include

<Rachael> additional example tests in a future draft Our intent is to include detailed tests and techniques for each functional outcome within the Silver framework.

Rachael: proposed wording by Wilco pasted above. Does anyone have concerns with that change? Updated wording:

<sajkaj> We don't have SC in WCAG 3?

MichaelC: Should be test methods or test of methods.

Rachael: may be just say test methods or guidelines. May not have guidelines either.

<Rachael> One the goals of WCAG 3.0 is to expand scoring tests of methods beyond a binary true/false choice at the page level.

<Wilco> +1

Rachael: any negative 1s for that wording?

Francis: three different definitions of rubric - not sure what which ones you are going for. Not sure what a good replacement word would be.

<Brooks> How about "grading tips"?

Rachael: Does anyone have a suggestion for another word? We should revisit for plain language. Brooks has suggested grading tips.

<alastairc> grading rules? Not a big issue for now

<Laura_> Scoring guide?

<Detlev> grading scale?

<Rachael> Our intent is to include detailed tests and techniques for each functional outcome within the Silver framework.

<JustineP> * need to drop for a bit

Rachael: Shawn had suggested revision of last sentence regarding detailed tests and techniques.

<Rachael> Our intent is to include detailed tests for and methods to support each functional outcome within the WCAG 3.0 framework.

<sajkaj> +1 to the change

Rachael: Any concerns with that suggested change that was pasted in. They are editorial.
... Any other concerns before we move to straw poll?

<Zakim> Lauriat, you wanted to add the link?

<jeanne> +1 to using the Requirements for more info

<sajkaj> +1 to better contextualization

<Rachael> Suggested Wording: One the goals of WCAG 3.0 is to expand scoring tests of methods beyond a binary true/false choice at the page level. We have included tests within the sample guidelines that demonstrate alternatives such as rubrics and scales.

<Rachael> We are also exploring integrating these options into <a href="https://www.w3.org/WAI/standards-guidelines/act/">Accessibility Conformance Testing</a> format. We will include example tests in a future draft. Our intent is to include detailed tests for and methods to support each functional outcome within the WCAG 3.0 framework.

Rachael: Do we need to replace sample guidelines with sample functional outcomes?

Lauriat: Should be test for methods.

<Rachael> Suggested Wording: One the goals of WCAG 3.0 is to expand scoring tests of methods beyond a binary true/false choice at the page level. We have included tests within the sample functional outcomes that demonstrate alternatives such as rubrics and scales.

<jeanne> +1 to Functional Outcomes. I think we need to have some higher level organization for usability, but I want to wait until we settle on the Conformance to do that.

<Rachael> We are also exploring integrating these options into <a href="https://www.w3.org/WAI/standards-guidelines/act/">Accessibility Conformance Testing</a> format. We will include example tests in a future draft. Our intent is to include detailed <a href="https://w3c.github.io/silver/requirements/#multiple-ways-to-measure">tests for methods</a> to support each functional outcome within the WCAG 3.0 framework.

Pwentz: Is wording for scales remains -- not certain if I can map to - scales isn't such a good wording.
... Most issues with scales as it is so open.

<sarahhorton> How about "ranking" or "rating"

Rachael: Do we feel examples are needed or should we remove wording? Does anyone have alternative wording?
... Sarah has suggested ranking or rating.

<stevelee> doesn't scan quite right

Rachael: Do a straw poll on this wording and that we should not that we may need to continue working on wording.

<Rachael> Straw poll: +1 to using last wording with a note that we need to work on "rubrics and scales", -1 to keep editing here, 0 to abstain

<jeanne> +1 to use wording we have. Editors will be making a pass on it anyway/

<Lauriat> +1

+1

<Detlev> +1

<sarahhorton> +1

<Ryladog> +1

<Sukriti> +1

<stevelee> +1

<Francis_Storr> +1

<Raf> +1

<Laura_> +1

<alastairc> +1

<Brooks> 0

Rachael: overwhelming support to add wording and move on.

RESOLUTION: Use edited wording for Testing note and work on "rubrics and scales"

Rachael: Next topic on depreciating scores. A resolution was passed at task force level. Within the model should we date stamp claim or look at some depreciation of scores over time.

Depreciation

Rachael: Claim only with date - no depreciation. Date stamp and depreciation. Date and mandate end of life or something else.
... Peter's comment indicates that some sites will not align with date stamps - but ok with raising it as a question in the working draft.

MichaelC: favor date stamp only as well. I am supportive of editors not to recognize other opinions.

Sajka: Not sure how depreciation would help - other than we should not trust a score that happened a time ago - scores that are old may be less reliable. Doesn't seem to be defended.

Rachael: question over the date - is the date of the claim or when the testing was complete?

Lauriat: Stake in ground as of this time these are the results - date of claim and not of testing date.
... would be for whatever date you are claiming it for - could be for 2 months or next week when you have fixes in places. Leave it up for person who is making claim for the scope of what is covered.

<Zakim> alastairc, you wanted to say it is an MOT approach

Francis: Seems more complicated if you are future dating it -but that clarifies.

AlastairC: vehicle check is valid until you drive away - valid at that time only. From the next day it says nothing about suitableness for later days. We are trying to provide foundation. It is up to regulators for how long they want that to remain valid.

<Lauriat> +1, I think Wilco's comment in the survey put it quite well.

Sukriti: See point of regulators might be responsible for time frame - so what is suitable for which industry. That shouldn't be something we are deciding. The point I made would no longer hold. I'm ok with it.

<Rachael> Straw Poll: Include Date Stamp, Not depreciating, and adding an editor's note to gain more feedback on date stamp

<Detlev> +1

<KimD> +1

Rachael: Including a date stamp, no depreciation, add editor's note.

<Sukriti> +1

<Laura_> +1

<Ryladog> 0

<sajkaj> +1

<sarahhorton> +1

<pwentz_> +1

<AWK> +0

+1

<Lauriat> +1

<jeanne> +1

<Wilco> 0

<alastairc> +1

Rachael: Todd can you speak to the end of life reasoning?

<Raf> 0

Rachael: Since we don't have some speaking and it looks like we have overwhelming support we should add a editor's note. We can address later. Any objections?

RESOLUTION: Including a date stamp, no depreciation, add editor's note

Revised Silver Proposal https://docs.google.com/presentation/d/1yZLF29X3SoDyn1ph8KSTn5nYCC_P4E6fw1zbrUo35uA/edit#slide=id.g91ee1de303_0_16

Rachael: Revised proposal based on deep dive. We moved forward with combing Detlev's excellent 3rd suggesting - we picked what worked best from each proposal and created this updated proposal as a way forward to get to a public working draft. We would continue to work on as well.

<Todd> 0

Jeanne: was planning to do functional outcome portion.

Rachael: Starting with slide 3 - key points - bring down to 2 test levels - 1 is traditional tests - atomic and comparable to WCAG 2 - may be A, AA, AAA may include methods and techniques. Those would get us to a bronze which is minimal.
... Next have holistic tests which are done on a complete path or process. This concept might be done throughout tte process that would give you points to get to silver or gold.
... functional needs and outcomes. First score would be applied at functional outcome level. Those contain methods and techniques. 2 options at this point - adjectival at coutcome level and the other is to aggregate into a percentage.

<alastairc> scribe:alastairc

<jon_avila> Rachael: We will propose adjectival but propose an alternative. We will still look at critical failures being defined.

Rachael: In either case, still looking at critical failures
... (slide 5) Functional outcomes, the level where we do the adjectival ratings, similar level to SCs. Plainer language, more granular.
... (slide 8) People will declare the [path/view], and those are tested for conformance. An example of a scope might be website of ACME with 10 views and 4 paths. Views selected by representative sampling.
... conformance will be at the step or view level.
... the decisions about the 'borders' of the scope depend on the tester.
... some states called on top of a view would be separate, would be defined in the conformance scope.

Ryladog: Functional outcomes are normative, which include tests & techniques. Previous trouble with techniques as normative.

<jeanne> Tests and Techniques are not not normative

Rachael: Intent was that tests & techniques would be informative.
... for holistic testing (slide 9) these would be additive points. Needs more work, but will need to document that more.
... (slide 19) Select and run tests based on tech used. Score functional outcomes (reads from slide)
... (slide 11) scoping at the top, testing by func outcome, then scoring by func outcome, then total scores. Little decision tree to get to a final level.

<david-macdonald> presnt+

Wilco: Had a million thoughts in the last 5 minutes, what sticks out at the moment: The idea of percentages or adjectival testing, would we be able to give a score at an element level?
... in ACT we're working towards having a score at the page level, consistently. That's very difficult. Breaking it down to the element level? Concerned would be next to impossible.
... e.g. measuring colour contrast, 2 of 3 paragraphs pass, there is a fail there. But in % method, what level do you take that at?
... being able to do that in a way that can be tested consistently, not sure we can do that?

<Zakim> jeanne, you wanted to answer wilco

Jeanne: Thank you Wilco, lots to think about. Starting with concern about whether we can do it?
... e.g. for things that could be tested by tags, that seems it would be easier? For things where there is more judgement, more time consuming, we tried to address that by having the critical failures on the path.
... you could test until you got a critical failure, and then stop, you then don't have to do a %
... that was part of the attraction, we can test up to there and then stop.
... but I'd like more details about what we can't test today.

Detlev: From a mere conformance perspective, it might be ok to stop if you find a crit failure. E.g. with threshold on contrast etc. Let's assume that's possible. If you are testing to provide feedback, you'd want to cover everything, not stop.

<jeanne> Thanks, Detlev, I was speaking of testing for conformance purposes, not for identifying accessibility bugs -- even though I realize they are done together in real life.

Detlev: on Wilco's point about assessing quantity + quality. That's true, it's not easy. However, we have to hold that against a situation where you have to decide whether something is a pass/fail, and then you have situations where there is little problem for the user. You have to find some way of providing an overall value judgement.
... e.g where you have a set of footer links that are just under the contrast threshold.
... that's a fail now, but in the new system we could be more qualitative. That means looking at the results of things that can be calculated and work out whether it is a big or small problem. It's difficult, but it's better than going by hard pass/fail threadholds.

<Zakim> alastairc, you wanted to ask about 'guidelines'

AlastairC: Agree with detlev, and think it changes the conversation from what we currently have.

Wilco: I'm not saying we can't do testing at element level consistently, but we've never tried it before.

<Detlev> UWEM??

<Detlev> UWEM tried it

<Rachael> alastairc: : Asked about whether guidelines would be included. They are in the name.

Wilco: never had to do this before, so assuming that we can put it in a doc and it will work is dangerous.

<Rachael> Jeanne: Yes but not normative or in line of conformance

Wilco: I like the image example, where WCAG doesn't really say what the boundaries of non-text content are. If you are looking to test an image from a tools perpective, could look at accname, role etc. Some tools only have access to some aspects.

<Detlev> if quantitative counts just *support* your eventual adjectival judgment, they don't need to be perfect....

<jeanne> Wilco, but don't you have to do that today?

Wilco: there are images type things that are not marked up as images, e.g. backgrounds, canvas, SVG. It gets complicated very quickly. If you lock that details down into a standard, requiring tools to do things the exact same way, would lock down the standard a lot, it would be very hard to change.
... don't think we should lock it down at that level.

<Zakim> Brooks, you wanted to ask whether or not we want to encourage simple pages versus more complex pages in the way we shape this part of the new standard that addresses scoring?

Brooks: Has the group addressed the issue of benefiting pages that are simple vs those that are complex?
... How do the rules incentivise that?

<Detlev> if there's less stuff on the page, less can go wrong...

Rachael: The way it is structured has been tested for validity on simple vs complex pages. However, there is no encouragement for simple or complex.

Brooks: Wondering about a simple page where one issue might be a blocker, but on a complex page it might not be a big thing.

Rachael: Regarding Wilco's (internet problem, missed a bit), you do raise some good points, but that's why the cut-off of the page help.

Wilco: How is that different from a percentage, still need to count?

Rachael: The differnce is that at the point you put in an adjectival rating, you aren't putting in a hard judgement that carries all the way up.

Detlev: The example I used was a simple form, simple step by step process. But, you have a footer with 30 links all slightly below the contrast level. You wouldn't just count elements, you'd take everything you get back from the automated tests and look at the main elements on the page to advance. It requires value judgements, which the counting supports.
... it helps to judge the impediment to the user with an adjectival rating.
... with a sample of 10 pages, if some are 'fair', some good, one bad, you get a mix of that. At that stage you might be in the percentage world. Also, any crit failure would need to stand out, so it isn't masked.

Wilco: Difficult to explain with using percentages? Not opposed to using these adjectival buckets, but we need to know it can be done consistently across organisations.

<Lauriat> +1 to Wilco's point about demonstrating how this can work, I think we have tests coming that'll help there.

Detlev: repeatable is a bit of a red herring. If you only have pass/fail judgements, you often find differences. Repeatability is an issue, but have to live with some differences, have to find ways (possibly not automatable) to deal with it. We have a scheme with 2 testers doing a test, sometimes we do find differerences. We have an arbitration procedure.
... as soon as human judgement comes in (even supported by automation), you will have differences.

<Zakim> Lauriat, you wanted to speak to "repeatable"

Lauriat: Repeatable, for our context, means that the tests are repeatable. I.e. two testers looking at the same thing with the same tests should come up with the same thing, which is difficult now.
... want to improve the decision making support for the judgement calls.

Wilco: Don't think that WCAG today isn't repeatable. Also, it sort of undermines WCAG if the results are not repeatable. I don't know what repeatable tests are if they are not repeatable. Important thing not to lose site of is that organsiations will get results from different places. We don't want each to have different outcomes.

<Zakim> alastairc, you wanted to say theoretically it should be smoother.

<Detlev> +1 to Alastair

<Brooks> inter rater reliability is supported by providing quality grading tips that draw on examples of commonly encountered real-world scenarios that testers are likely familiar with.

<jeanne> +1 Alastair. Thank you Wilco for raising important points that we have to test.

AlastairC: Having category rating (1-5, bad-good etc) should smooth things out, have less differences.

DavidM: Not sure I understand, in the new model: Will we have to count all the items part of this? That would make testing a lot longer. It's much easier to pick out all the errors and focus on those, rather than comparing to the ones which pass.

Rachael: When it can be automated, yes, but not in all cases.

Wilco: If we do the adjectival route, it is not something tools can do unless it is based on hard numbers. If not based on hard numbers, then tools are not going to be able to help us with that.

<Zakim> Lauriat, you wanted to respond on testing vs. conformance claim

Shawn: Tools today already give us data we make judgment calls about.
... For David's comment, the steps we go through is for creating a claim for conformance. For those who just want to test what they need to fix, it doesn't change much. Adds steps only if you do a conformance claim.

<Zakim> alastairc, you wanted to say it doesn't change the level of automated testing

<Rachael> alastair: I don't think this model changes the amount of automated testing we use. It just means that automated testing will be more encouraged/required to speed up certain tests or methods. It might make it more obvious that certain manual testing is required to have a conformance statement. Which wouldn't bother me.

Discuss Functional Outcome Presentation (Slides 5-7)

Jeanne: Looking at how the functional outcomes (FO) differ from success criteria (SC).
... as we were writing them (slide 6), we wanted to say what the result the tester would want to see, and then why they want to do it. Aware that some people would extract just the FO, so wanted to include some reasoning in it.
... we sent this around, and got some feedback suggesting we put it into columns for usability (slide 7). So took the content from slide 6 and presented in the table view on slide 7. Should we continue with this?

Bruce: What was new to me, was that the FO are different from functional criteria, these are much more granular.

<jon_avila> Yes, WCAG 2 is actually a functional standard.

Jeanne: We have far too many things starting with "functional", which we should address in future.

<Rachael> alastair: I'm not sure right now, but I find that when you convert something from one structure to another like this you find gaps which happen to matter. I think the best thing to do is to scan down and work out whether it fits.

Rachael: Any other thoughts?

Sukriti: I think this is relatively easy to understand, I wanted to work on something like this on MAFT.
... e.g. for video, for page heirarchy etc. Makes it easier to test.

prefer the slide 7 table format.

Detlev: I agree with Alastair, there might be things we discover things that don't fit. Big fan of taking 1.3.1 & 1.1.1 and splitting those up as most failures come from those big SCs. Good to split them up.
... in future it would be good to separate outcomes from the guidelines we have now.

JonA: Clarifying, does this map to specific disability types anymore?

Jeanne: It's in a different place, we haven't lost the mapping. The mapping happens at the methods level, so they can be carried through to a total, when you get the total score. Then you can break the score down by disability categories.
... want to be able to carry the data through to the final score.

Wilco: Comment, I like the way this is breaking down, helps with consistentcy. Question: Are we going to put this in normative language? Or will the testing be in informative.

Jeanne: Current plan is that this is the normative level, and the more granular breakdown is at the informative level.
... wanted to keep the normative level more inclusive.

<Wilco> -1 to not defining in normative

Detlev: It would be good to keep an open mind about how these work. If we had about 200 of these outcomes, compared to 50 guidelines, then ... (sorry, missed the thread of this)

DavidM: Testing could be done by various stakeholders, how do we manage "We provide guidance on scoping a step/view but the decisions are made and declared by the tester"

Jeanne: What people do today, is claim conformance by individual pages. They decide on the pages. We aren't trying to fix that, so what we're doing is acknolegding the reality of how people do that today. Typically in US, most are done by VPAT, where people state what they are writing their VPAT for.
... It is up to them to scope those pages. We're trying to move it beyond just using a page, so we can address more varied scenarios/technologies. Have to address it.
... simplest way is to have it be like today, where people declare what they claim conformance on.
... what people want is transparancy, where you can tell what was put forward for conformance.

<jon_avila> So a claim could be on only part of a page .

[Scribe missed section due to internet dropping briefly.]

Rachael: Will continue to explore this, there will be a survey for two weeks from now.

<jon_avila> thank you

<Sukriti> thank you

<Rachael> David: Is the scope for the conformace only done by the creator?

<Rachael> Jeanne: No.

rssagent, make minutes

This is scribe.perl Revision of Date 
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/Resolution/RESOLUTION/
Succeeded: s/we is not a reasonable/it is not a reasonable/
Succeeded: s/words for/definitions of/
Succeeded: s/it in/in it/
Succeeded: s/fconformance/conformance/
Default Present: Rachael, pwentz_, Lauriat, alastairc, JustineP, Raf, sajkaj, mattg, KimD, Francis_Storr, Laura_, jeanne, Brooks, Detlev, jon_avila, sarahhorton, Sukriti, Todd, Caryn-Pagel, stevelee, MichaelC, Wilco, 0.5, Katie_Haritos-Shea, bruce_bailey, david-macdonald, AngelaAccessForAll
Present: Rachael pwentz_ Lauriat alastairc JustineP Raf sajkaj mattg KimD Francis_Storr Laura_ jeanne Brooks Detlev jon_avila sarahhorton Sukriti Todd Caryn-Pagel stevelee MichaelC Wilco 0.5 Katie_Haritos-Shea bruce_bailey david-macdonald AngelaAccessForAll
Regrets: Chuck A Bruce B John K Nicaise D
Found Scribe: jon_avila
Inferring ScribeNick: jon_avila
Found Scribe: alastairc
Inferring ScribeNick: alastairc
Scribes: jon_avila, alastairc
ScribeNicks: jon_avila, alastairc

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

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.)