W3C

- DRAFT -

User Agent Accessibility Guidelines Working Group Teleconference

20 Dec 2012

See also: IRC log

Attendees

Present
Jim_Allan, Jeanne, Greg_Lowney, Kim_Patch, kford, [Microsoft]
Regrets
simon, mark
Chair
jimallan, kellyford
Scribe
kippatch, kimpatch, -kippatch

Contents


<trackbot> Date: 20 December 2012

<JAllan> new editors draft: http://www.w3.org/WAI/UA/2012/ED-UAAG20-20121220/

<KimPatch> Jim: end of levels discussion – if you have any more about levels send to list

<JAllan> close item 1

<KimPatch> Jim: winter break – I propose that we don't meet the next two Thursdays

<KimPatch> general agreement

<JAllan> close item 2

<JAllan> open item 3

<JAllan> information to come in the new year.

<KimPatch> Jeanne: definition of levels – more on the first of the year

<JAllan> close item 3

<KimPatch> Jim: success criteria, conformance

<jeanne> http://lists.w3.org/Archives/Public/w3c-wai-ua/2012OctDec/0052.html

<KimPatch> Jim: ARIA role navigation - trying to tease out whether someone did aria right is difficult

<KimPatch> Jim: these are pretty minor otherwise

<KimPatch> Jeanne: 2.1.3 I recommend we drop that – there are a lot of WC3 documents on compound documents

<JAllan> ACTION: jim to do 2.1.1 urls for resources [recorded in http://www.w3.org/2012/12/20-ua-minutes.html#action01]

<trackbot> Created ACTION-787 - Do 2.1.1 urls for resources [on Jim Allan - due 2012-12-27].

<KimPatch> Jim: 2.1.3 is no keyboard trap, so it has to do with passing off the keyboard focus inside a compound document, which was a big thing a couple or three years ago. Maybe our examples are good enough. This is really more of the application of the keyboard trap in flash or the keyboard trap and some other embedded application which is essentially what SVG is and all the rest of that. So I would...

<KimPatch> ...be okay with removing it.

<KimPatch> no objections to removing related resource for 213

<KimPatch> Jim: 4.1.2 2.5.3 and issue 87 which are all related to aria navigation

<KimPatch> Jim: it's a AAA and this would be being able to configure landmark navigation and roles. We could add a resource, but want to avoid getting into a rats nest.

<KimPatch> Jim: 2.5.1 is location and hierarchy, we have already done that. The issue is navigate by element and semantic role.

<KimPatch> Jeanne: recommend we close it

<KimPatch> general agreement

<KimPatch> Greg: so we have a definition of structural and operable elements?

<KimPatch> Greg: there does not seem to be a definition – but it's only used in summaries

<KimPatch> Jim: we have important structural elements

<KimPatch> Greg: structural elements is in one note and one SC, and is not defined

<KimPatch> Jim: important elements is defined

<KimPatch> Greg: structural elements as mentioned in the glossary under outline view

<KimPatch> Greg: the term structure is used in the note with important structural elements, but the word structural there. It was changed in a note from 191 from important structural elements two important elements. The other case 252 where only in the title of the SC, and that's okay – it looks like we don't need to define it there, already in SC

<KimPatch> Greg: 191 changing to important in the note, 252 we can just leave it

<KimPatch> Jim: 253 can be marked as done

<KimPatch> Greg: however I do know that the term structural navigation is used – closest thing in the glossary is structural navigation command – that's probably okay

<KimPatch> Jim: I think we are good

<KimPatch> Greg: so we can close action 500

<JAllan> close action-500

<trackbot> ACTION-500 Kim And Greg to draft definition of Structural and Operable elements closed

<JAllan> close action-545

<trackbot> ACTION-545 SC and EIR for 2.5.1 Provide structural navigation [NEW] closed

<JAllan> close item 4

<JAllan> open item 5

partial conformance

<Jan> http://lists.w3.org/Archives/Public/w3c-wai-ua/2012OctDec/0054.html

<KimPatch> Jan: action to propose a type of partial conformance for mobile apps

<KimPatch> Jan: example, airline app on your phone and it uses the whole browser window to display only the current state of your flight, the name of the flight where it's from and if it's running late and by how long – something like that. Even though it is a full browser window potentially it knows where it's receiving data from – but basically it's not getting just any old image from the web.the...

<KimPatch> ...proposal was at a partial UN 2.0 conformance level

<Jan> Partial UAAG 2.0 Conformance - Constrained Content (Level A, AA, or AAA)

<Jan> ---

<Jan> This conformance option may be selected when the user agent is deployed such that it can only be used to display a tightly constrained set of content (e.g. as part of a mobile app that only displays text messages). The conformance claim must list those success criteria which are judged not applicable due to the nature of the constrained content.

<jeanne> ACTION: jeanne to add "Evaluating Web Sites for Accessibility" to the bibliography. [recorded in http://www.w3.org/2012/12/20-ua-minutes.html#action02]

<trackbot> Created ACTION-788 - Add "Evaluating Web Sites for Accessibility" to the bibliography. [on Jeanne F Spellman - due 2012-12-27].

<KimPatch> Jan: in theory this is a browser that could display time-based but it never will –because of the content we don't have to make this

<KimPatch> Greg: where would this be in the conformance section

<KimPatch> Jan: would say is claiming partial conformance to UAAG, and claiming partial conformance based on constrained content

<KimPatch> Jan: right now there's no section but in ATAG you can see how they are done

<Jan> http://www.w3.org/WAI/AU/2012/ED-ATAG20-20120924/#conf-levels

<KimPatch> Jan: the intro text is out of date – two types of levels, regular and partial

<KimPatch> Jan: regular, 3 A levels, then two types of partial.

<Jan> http://www.w3.org/TR/WCAG/#conformance-partial

<KimPatch> Jan: we got this from WCAG – here's how they do it

<KimPatch> Greg: ATAG has levels, WCAG has requirements

<KimPatch> Jan: H3– at the same level. The point is it's just a modifier, it is in full conformance, we know it's not, and here's the reason. It could be because some of the content is third-party content or the language

<KimPatch> Jan: could say we are not a general content provider – it's constrained and constrained in these ways

<KimPatch> Jim: works for the American airline app example – this will do a fine job

<KimPatch> Jim: list of the things they are complying with a list of the things they aren't complying with:

<KimPatch> Greg: they could use the same form, but simply fill in NA with explanations

<KimPatch> Greg: if they are making a claim at all, which is optional. But if the claim is there do they have to list every SC?

<KimPatch> Greg: #7 says to state whether conforms. So how is this different – requires to state, when you read through section 7 you discover that half of them are NAs

<Greg> That is, section 7 in a conformance claim requires listing for every SC pass, fail, or N/A and why.

<KimPatch> Jan: the browser component would not be a very good implementation of UAAG if it were by itself – for example doesn't have highlighting options for active keyboard focus, recognize enabled input elements were recently visited links – but that's okay because it knows that the only text that's going to come through and be displays is flight numbers

<Greg> Thus if it knows from constrained input content that there will be no images, then every SC involving images can already be marked N/A with explanation of "because of constrained input set does not include images".

<JAllan> scribe: kippatch

<JAllan> scribe: kimpatch

Jan: if the web browser is going to display content that includes whatever those classes should be highlighted – you can just offload that entire scenario to a special type of conformance – this is not a great example because it doesn't do this stuff, but it's such constrained input that can get away with it

Greg: but can they already get away with it under the current method – in the list of SC's they would say NA because the constrained input does not include

Jan: that becomes a testing question – I don't think it was clear before
... it's kind of an edge case that needed to be stated

Greg: I'm not disagreeing necessarily, just trying to work out the implications
... I can understand that the high level wanting them to identify that it's a different kind of case, but in practice I wouldn't want to let them out of section 7 where they list every SC and say yes no or N/A – so that I don't think should change

Jan: I agree – just want to make sure there's a meaningful distinction between a tool that – example a tool that does not support video – it can say not applicable to anything that has to do with video. and then this other browser complement, HTML 5 browser complement which is being used in this particular situation where it's text. I agree with you it doesn't change it all going...
... through the...
... whole document and saying yes no N/A

Greg: if were still going to do all the steps in the conformance claim including listing everything yes pass fail or NA, but a at a high level you want to call it out, how manyNAs does it take to make it have to be called out
... do many user agents have a least one NA?

Jan: this is a very different type of browser than Internet Explorer or a Firefox where you can browse anywhere on the web

Jeanne: one of the other use cases that we have been talking about for partial conformance was the American Airlines app – where success criteria – conformance is coming from the platform level but it might not work on a different platform. So the American Airlines app on iOS might meet the success criteria because of features of iOS and it might not need it on the android platform or vice...
... versa. I believe that was one of our use cases

Jan: that's handled in the next type of partial – there are three types

Jim: must listen to all of Jan's levels and then come back and see how we can resolve Greg's issues

Jan: reading second one

<JAllan> Partial UAAG 2.0 Conformance - User Agent Component (Level A, AA, or AAA)

<JAllan> ---

<JAllan> This conformance option may be selected when a user agent would require additional user agent functionality in order to conform as a complete user agent. This option may be used for components with very limited functionality (e.g. a plug-in) up to nearly complete systems (e.g. a user agent that only lacks mouseless browsing).

<JAllan> The level of conformance (A, AA, or AAA) is determined as above except that, for any "no" answers, the user agent must not prevent the success criteria from being met by another user agent component as part of a complete user agent system.

<JAllan> Note: User agents would not be able to meet partial conformance if they prevent additional user agent components from meeting the failed success criteria (e.g., for security reasons).

<JAllan> Partial UAAG 2.0 Conformance - Platform Limitations (Level A, AA, or AAA)

<JAllan> ---

<JAllan> This conformance option may be selected when a user agent is unable to meet one or more success criteria because of intrinsic limitations of the platform (e.g., lacking a platform accessibility service). The conformance should explain what platform features are missing.

Jan: this is so a browser can say I don't do mouseless browsing, but some other plug-in could

Jim: so you could say I need the extension in order to do full compliance
... so you can say I meet this one success criteria and nothing else

Jan: yes, important because if the company says we don't want to step on the toes of the checking company… this lets everybody do it they do and do no harm to others
... you can let a small company do only their part. You can let another company say we do everything but we don't do mouseless browsing
... large companies don't want to have to rely on small companies for conformance

Greg: in terms of partial compliance – partial implementation user agent component – so in that case Firefox would claim partial compliance as a user agent component

Jan: yes it is a loophole, but if they are claiming user agent partial conformance and all the sudden there in some RFP and Microsoft comes in and says actually were doing a full conformance claim – either because they built all the stuff natively or maybe they did with the last-minute agree to rely on a secondary vendor, then it looks better, they have the full story to tell

Greg: assuming mouseless browsing is AA, they can't claim AA level?

Jan: they are free to claim a bundle – it's just that in the atag world we ran into strong – we don't want to

Greg: Where do they claim add-ons as a component?

Jan: it's something that is going to have to be built into the conformance claim

<Greg> In the section titled "Required Components of an UAAG 2.0 Conformance Claim" we need to add an additional item where the claimant identifies first, or third party components that are required to meet some of the SC requirements.

<Greg> For example, Firefox would list the Mouseless Browsing extension.

<Jan> Analagous ATAG2 wording: "Note: If the authoring tool is a collection of software components (e.g., a markup editor, an image editor, and a validation tool), then information must be provided separately for each component, although the conformance claim will treat them as a whole."

Greg: ATAG wording doesn't really make it clear that some of them might be third-party components

Jan: we've left that unstated – what if the mouseless browsing people wrote something about themselves in the context of Firefox. It's allowed, but Firefox might not like it so we dialed it down because of that concern.

Jim: maybe leave them in and see if anybody complains

Jeanne: different situation, vendor dominated versus accessibility

<JAllan> greg: may need use cases for each of these

<JAllan> will take this up at the next meeting.

<JAllan> scribe: -kippatch

Summary of Action Items

[NEW] ACTION: jeanne to add "Evaluating Web Sites for Accessibility" to the bibliography. [recorded in http://www.w3.org/2012/12/20-ua-minutes.html#action02]
[NEW] ACTION: jim to do 2.1.1 urls for resources [recorded in http://www.w3.org/2012/12/20-ua-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.137 (CVS log)
$Date: 2012/12/20 19:51:45 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/muscles/mouseless/
Found Scribe: kippatch
Found Scribe: kimpatch
Inferring ScribeNick: KimPatch
Found Scribe: -kippatch
Scribes: kippatch, kimpatch, -kippatch

WARNING: Dash separator lines found.  If you intended them to mark
the start of a new topic, you need the -dashTopics option.
For example:
        <Philippe> ---
        <Philippe> Review of Action Items

Default Present: Jim_Allan, Jeanne, Greg_Lowney, Kim_Patch, kford, [Microsoft]
Present: Jim_Allan Jeanne Greg_Lowney Kim_Patch kford [Microsoft]
Regrets: simon mark
Found Date: 20 Dec 2012
Guessing minutes URL: http://www.w3.org/2012/12/20-ua-minutes.html
People with action items: jeanne jim

[End of scribe.perl diagnostic output]