JR: Jan Richards (Acting Chair)
JT: Jutta Treviranus (by Conference Call)
JB: Judy Brewer
CMN: Charles McCathieNevile
LN: Liddy Neville
PJ: Phill Jenkins
GB: Giorgio Brajnik
AA: Andrew Arch
LH: Lofton Henderson
MR: Madeleine Rothberg
CB: Cathleen Barstow
HK:Hazel Kennedy
NL: Natasha Lipkina
KHS: Katie Haritos-Shea
IJ: Ian Jacobs
HB: Helle Bjarna
Heather Swayne
(Liddy Nevile is minuting)
Attending: JR, JT, PJ, MR, CB, LN, CMN, HK, LH, AA, GB, NL, JB, KHS
JR: Sums up what ATAG has been doing.
MR, CB: Who IMS are and what they do in collab with NCAM - What's BlackBoard doing? - expenses limited so it is not doing so well but planning a more accessible version plus a special extra version for people with special needs - but they are also trying to do things about interfaces, we think. WebCT and e-College - nothing heard from them direct but they have tried but we are not clear how far they have got.
JR: WebCT evaluation is on the ATAG web site.
Macromedia are obviously doing active improvement to their environment and they are aiming to make their Flash etc work with MSAA (Microsoft Screen Active Accessibility) - and those with intellectual disabilities are doing better.
PJ: to demo Lotus Learning Space - Same time web conference - a product from Lotus. 3 windows - chat, who's on line and then a window as a whiteboard, or running an app, eg PowerPoint, or as a desktop. Navigation is impossible, no speech output because client machine does not get access to action, only the large bitmap, (not like Netmeeting). Lotus has as authoring tools - all the Macromedia products (See Phill's email on the relationship). So the authoring is as ever for Macromedia tools and the user gets to work with Learning Space - now in Version 5. IBM so far recommend ATAG but don't yet enforce it - maybe by enforcing it, they would both influence Lotus product and the Macromedia product.
The fundamental problem is that some of these things use real interactions and then you could possibly call off live streams of captions etc but if the client only has access to an image of the action, there will need to be real radical re-design of the system. There are social and other issues in this context and timing is one major problem. Sometimes it is up to the instructor to think at the time about how to use the facilities ie thinking at delivery time as well as at development time.
IBM have lots of hints about at the time - on their Internal web sites. EOWG has had lots of requests for work arounds - IBM might be able to help with hints when authoring tools don't work. There is an accessible presentation about how to give an accessible presentation!! Delivery tools are like 'live authoring' tools.
Action Item: JR make edits to Techniques draft on the basis of these discussions.
Should we add to this? 1.1 says make it possible for an author to do it - but what if the tool is one that controls the publishing/ then the technique is to look at the automatic publishing bit.
So far the checkpoint talks about the format that the tool supports - but should it say you MUST provide an alternative accessible format? This only applies in as much as the author is using the tool - other checkpoints catch them but this is limited to what they do. EG Flash passes here but fails later.
So we should say - an eval. technique is that if the tool does not support accessible languages, 1.1 is not applicable but the tool will fail because you have not provided alternative formats.
It could be that what you do is provide for metadata to be added to show the limitation of the product - ie, a technique for coping is to do this and that will lead on to an accessible product.
Example for meta-data is to reference Dublin Core
JR - are you proposing another technique under 1.1?
CMN - yes, as recommended, not required - hard to argue that it is a necessary technique (see CMN's email, see CMN's second proposal)
JR - as a technique for automatically generated? Could add in elearning section - at delivery time you could provide in a way that is not available at time of authoring - a warning when the author is authoring that they will need to do something extra at delivery time. And tell the author at authoring time that the format being used is not going to be accessible.
Course management tools often adapt content for inclusion in the system. The author needs to know what is happening to the content and be able to control it. The author should be told what is happening to the content.
The one about tables - ensure that the order is not changed - should say that the author should be shown the new order so they check it is what they want…
The footnotes one should include annotations, call-outs, etc.
JR: · Add "inclusion of documents within other systems" i.e.WebCT
PJ - as a non-text elements we have frames and scripts - does this make sense? How is ' 'non-text' to be interpreted?
1.3 is wide in that its scope goes from a new image, eg to one that is in a library already - but it is narrow in that the prompting etc for the author comes from 3.1, not here.
What about when content is generated and it is not good - eg that simplest language should be used. Keywords should be made available, using words from a graded dictionary, etc…. this is pretty hard and the checkpoint looks at markup ONLY - eg there are content generating tools being developed and it is important that these tools are included here - should this go in the section on this WCAG 14.1
Eg an automatic translation tool might need to be able to provide several levels of vocab.? Is this a matter for ATAG? or WCAG? We need a technique here that suggests that the content may be more accessible if the translation is done in simpler language
Big issue is does 1.3 cover content at all? If so, does WCAG 14.1 have real techniques and use cases? - yes, eg in a translation tool as it can make the content more or less accessible.
This looks like a new technique for suggestion ie language simplification. Should it be recommended or required? See T77. Recommended seems best.
JR: In Wombat, 'Template' has been expanded to include previously authored pieces of content.
The long list of WCAG checkpoints should be moved from here - this was resolved last year.
Add in pointers to the accessibility features bit of the W3C recommendations that we point to.
Is IMS developing an 'eML' that one day may be pointed to?
T84 is just more of T82 - can they be combined? Provide some info about how to do this using W3C recs.
Update list of W3C recommended specs. And annotate the list. Esp. show the ones with accessibility notes. Break the list into groups.
T85 is about 'markup languages and formats '- not just 'languages' - this might need to be attended to throughout the document as it can be muddling.
IMS and the idea of User profiles
IMS has a two-pronged approach - to specify characteristics of the user and to specify metadata that will be matched
CMN: CCPC could be used for this. Composite Capabilities Preference Profile is a technology for achieving this and used by other communities. It is extensible RDF for encoding this and there is architecture for using it.
JT: The preferences piece is not complete enough so is there going to be more done with preferences? The detail of the structure is basic - is it preliminary???
Metadata spec should perhaps be included in WCAG and then on from there. Then ATAG would pick it up.
If metadata exists that could be used for transformations, how should content be developed to facilitate this? It is another reason for following
W3C complete set of guidelines.
Note that Dublin Core is also working towards accessibility metadata.
The IMS LIP specification (learner information profile) foreshadows transformation things - eg Guideline 7 things. IMS is due to work through the LIP spec and add in the behaviours that can be associated with each node.
IJ: WCAG 1.1 talks about things that are non-text and includes frames, scripts,
…
Does this mean noscripts as well or what?
If we better define a non-text element we can answer the questions more easily.
New defintioin is anything that can be rendered as speech, Braille or an image.
Eg ascii art is not text. Some scripts have no rendering effects and so are
OK but others need to be handled differently. The UAG glossary contains a definition
that might be helpful. Encrypted content - it could be rendered in the ways
specified and so is not necessarily non-text. Accessible content can be rendered
in all of those modes.
Frames are caught in the ATAG. It is not that you can't have frames. I t is
the information that is embedded in the particular frame structure that cannot
be accessed ie visual association is doing the job of content - and it cannot
be rendered in the three formats - so maybe we need a clearer definition in
WCAG and therefore also in ATAG.
We should adopt the definition that has been put in UAG and link to it from
our guidelines.
Frames are a problem and it is not clear yet - will the WCAG group want to open
this up again?
Looking again at ATAG Techniques
nothing there
equivalent alternate content - see now the appendix for user prompting
When you are doing live presentations ie authoring while you are publishing,
there are things you can do to make this more accessible if the tool does not
produce something accessible - so there should be a prompt to remind author
to complement content.
What is equivalent might be critical in some circs - eg providing something
that is equivalent that is intended for a test … this might be best practice
stuff and very 'education' oriented.
Required is being removed so now there is recommended and suggested - because
these are not normative documents - but are these two expressions too similar?
Do WCAG and UAG doe the same?
Wombat works on minimums … etc
Note also the stylistic change - if there are not sub-texts, the WCAG checkpoint
is altered but where there is info added, it might be quoted in original form
- this is helpful as it makes for a single reading place. Add note to 3.1 that
there should be prompts for the long list and then leave the list of n/a things.
Perhaps number them off the ATAG checkpoints and make them eg 3.1.2T should
there be a co-ordinated numbering system.
Present : JR, PJ, MR, CB, LN, CMN, HK, HB, AA, GB, JB, KHS
(Phill is minuting)
IMS group leaves
add text to NOT include "equivalent alternatives" to Wombat, remove
T120
T0121 prompt the author to include a color independent identifier such as
an ordered list, Asterisk, wingdings,
T0122 prompt for a contrasting background or foreground color when the
background or foreground is changed.
T-137 need to be expanded (Phill work item)
Errata request to WCAG 1.0
Change incorrect use of "Header" as in WCAG 1.0 3.5 - should be
"Heading level element"
UAAG 2.1 technique - add one for Author to "override" user's dictionary
with Acronym/ABBR tag
All agree to read 9.4 to end as homework for Tuesday.
(Andrew Arch is minuting)
WCAG 9.4 / T0157 - does this below with 'skip nav' section instead? discussion was around implementation of tab order in user agents - often broken, sometimes locks non-tab-indexed links out of any tabbing access.
aside - What is status of discussion with WCAG? Are issues being tracked properly?
Change T0157 wording to reflect use of tab-oder for logical tabbing in form/page
Add - provide a 'skip programatic objects' link/ancjor to stop people getting trapped by these
Action - check in with WCAG w/g re status of tab-index and how addressing in 2.0
WCAG 12.2 /T0164 - ammend to specify that frameset is a collection of images
WCAG 12.2 - add technique to prompt for an HTML file instead of an image as tyhe source for a frame
suggest/required/recommended - be consistent
WCAG 13.6 Add
skip navigation
put group of related links in a frame
Action item - reality check (ie annotations/warnings) will increase credibility of this doc
Action item - consider code examples required for each technique, eg working web site with all techniques implemented as show case.
WCAG 12.3 / T0166 - applies only to unorganised (incl non-alpha) lists
WCAG COORD ISSUE: WCAG should look at 'how many' in an unorgasnised list before it needs to be broken into sub lists - ie what is 'natural and appropriate'
WCAG 14.1 - T0169 - refs required. What about other langs?
more kinds of pre-packacged content to be added?
ok
ok
T0196 - clean up and fix reference link
add best practices from some of the ER tools out there
Action item - Phill to id some of these and forward to the list
move t0205 to ATAG 3.2
ok
t0211-218 - break in to 2 gps - technologies and ??
t02221 - split into 2 techniques (CMN to clarify intent of part of this)
===== morning coffee =======
HB joins group
T0222
Phill - eclipse.org - open tools platform
T0222 - Phill suggests word changing to allow for toolkit building options
Action item - check out referencing open source tools in examples (but not proprietry ones) - refer to Judy for policy discussion/checking; Phill has details as above
T0224 - downgrade to [suggest]? for all?
Lots of discussion about differentiation for commercial/marketing reasons, usability improvements, etc
=> seemless integration of functionality into the products work-flow model of the tool is a minimum, look and feel consistency is encouraged, but not to the detriment of product differentiation and improved accessibility/usability
JR to sub-gp the existing Ts - suggestions:
t222 - worksflow
t223 - l&f
t224 - l&f
t225 - workflow
t226 - how invoked/added (probably workflow) - alos change to 'default settings'
vs 'default installation' (laso this is about content accessibility)
t227 - workflow
combine t229 into t228
t230 --> prompting techniques under 3.1
add tech - help serach should find information in the accessibility doc if it can't find it in the core helop doc
possibly need to differentiate between core tool and third-party plugins
t236-239 needs to be cleaned up and reference EO docs and WCAG techs
WCAG to include rationale for techniques in 2.0
Consistent style required in this doc and its component parts (eg placement of icons)
t241 - but not to the detriment of market differentiation
t238/239 reference http://www.w3.org/WAI/bcase/benefits.html
=======LUNCH=============
(Liddy Nevile is minuting)
Context sensitive help
244 - put in a link to references that show examples
243 - point to a table of attributes…as well as specs.
Tutorials
246 - include tutorials for any accessibility feature eg checking and improving…
247, 248 - fix formatting - otherwise happy
Looking at Jan's proposal for an appendix for relatives… Phill suggests
the idea is good but maybe order by ATAG checkpoints, not WCAG ones…
Action item: KHS to provide nicknames for all of WCAG checkpoints…
based on some already developed.
7.2 - add something to show this is zoom -
add to 282 that …?
Lots of discussion … trying to know what to do with whiteboards as an example
of problematic areas - Realtime? Synchronous? Collaborative?
What is the scope of tools for which ATAG is relevant??? Maybe there was an
oversight and ATAG 2.0 should perhaps have looked at these tools… include
MS Paint?
Should there be an appendix with 'best practices' for authors working in real
time??
Suppose a collab. tool has a prompting system that looks at what's happening
and suggests that you eg describe an image when you drop it into the whiteboard.
When it appears on the whiteboard it is covered by WCAG.
Phill agrees that if we know enough to say something, we should put it in. This
is real-time prompting… There is a need for collab with WCAG WG here -
they need to look at requirements …
Madeleine pointed out that this is a great intersection problem - real time
has UA, AT and XAG/WCAG all going at once!!!
All: Action item: read through ATAG and see where there is something
that needs to be added/changed or whatever in order to cover real time authoring.
16:00 (Andrew Arch Minuting)
T0142 - If you know it is a layout table, do not allow author to include structural table markup for visual presentation
T0139 - ask if table is layout OR data table (nb layout includes mixed/hybrid purposes)
WCAG2 - needs to discourage hybdrid tables
At front - need "how to use related techniques docs"
wcag 6.1 T0144 - make sure that if the S/S is taken away the essential content and nav is still available and meaningful and the structure is still evident - technique is torovide a draft view with styling turnedd off.
send clarification back to WCAG 2.0
WCAG 6.3 - split in to 3 separate gps of techniques
scripts - prompt for server side alternatives for essential scripts
- provide a draft view which disables scripts (suggest)
-
applets - prompt for alt content for essentail applets
- (for 6.4) make the applet directly accessible by following the JAVA accessibility
guidelines
-
CMN - EG ABOUT Client SIDE AND SERVER SIDE validation scripts
wcag 6.4 - T0147 - change 'applet' to 'script' (waht about auto incvlusionf keybd/mouse equivalent actions?)
CMN - action item - write something on flickering/blinking (7.1/7.2/7.3)
-> Wendy - need some vision research about this (eg. epliepsy foundation)
t0152 - meta-refresh - see AERT for technique - detect and tell author to provide a static refresh link
- 'until user agensts' - suggest using cookies (or other profile techniques)
to manage time sensetive issues
- ask WCAG for prefered techniques
wcag 9.1 - ok
wcag 9.2 - see 6.3/4
wcag 9.3 - ditto
t0157 - monash uni have done something here (CMN)
============Close ATAG notes at 1740hrs================
(Madeleine Rothberg Minuting)
wai au 2/26 4:30 pm
LN: IMS planning to use WCAG to create metadata on learning materials, match up with a learner profile to deliver accessible content to learners.
JB: concerned that we are splitting up the checkpoints to form groups optimized for users with X or Y disability. encourages companies to ignore specific groups.
LN: but would help users find stuff on Google if they need it.
JR: why not fix problems while doing the evaluation?
LN: some metadata provided by people other than the author. some automatically generated, some in libraries that catalogue materials that belong to others.
JB: does this support segregation? encourages disabled users to only use specific pieces of content. Stress preferentially ranked content rather than filtering out inacessible content to avoid ghettoizing.
LN: Please present this at WWW2002 to state these principles.
LN: may wish to have ERL report in composite point scores rather than three levels. Go through checkpoints and give weights and then report composite score.
PJ: consider the AIR rankings for the Acessibility Internet Rally. They weight accessibility and also attractiveness. We struggled with it alot. but in the end it may not help the person with a disability, if they got a good score but messed up one thing that is crucial to a blind person.
Helle: Best on the Net project from teh Danish government, rated 6000 web sites. Grouped into three areas: quality (including attractiveness, level of interaction with constituents (including simple language, organized for citizens to understand), usability. Rated to maximum of 100 points. Groups that get over 90 points are evaluated by end users (w and w/o disabilities) and webmasters can respond to evaluation. Evaluation of this whole project led to addition of some accessibility issues.
MR: Back to IMS goals, we can't use composite ratings because to take advantage of the learner profile we need info on which checkpoints it failed.
JB: Shouldn't you be encouraging all materials to be accessible to everyone?
MR: Yes, but there is legacy content out there that isn't accessible and users may want to know which is more accessible to them. The profile will also be used to transform accessible, flexible content into the correct view for the user.
JB: Concerned about slippery slope toward avoiding fixing all the content. You do need to deal with reality -- a deaf student in a classroom needs an interpreter and an online classroom has similar requirements. But be sure to stress the longterm goal of full access.
LN: IMS aims to provide high-quality educational experiences. Whitepaper gives guidance on that.
JB: in higher ed, i've seen people rework curriculum to meet the needs of a particular student, and think they'd made the whole thing accessible. Web encourages repeated reuse of educational materials. See Alison Littlejohn on accessible, reusable learning materials. DIfferent people can add different accessibility features to the same piece over time until it is all fixed.
LN: what is the purpose of the profile? 1st, have you paid, are you in the right grade, speaking hte right language, and then, is this material accessible to you?
JB: Watch out that you don't take a prescriptive approach, using info to exclude people rather than to help them. Use a philosophy of ablism, or a library model to accept who comes in the door and work with them.
JB: What is specific coordination issue for IMS/AU?
LN: IMS folks, when we talk about courseware we are always talking about authoring tools.
MR: We are always talking about content and user agents, too.
MR: As a specific coordination item, once we finish the metadata spec, we would like ATAG to encourage autoring tools to put metadata on content.
JB: If needed, we could do some additional coordination through the WAI coordination group. What are the coordination goals? For example, IMS could drive ATAG implementation. We could check in periodically.
Action AU: address question if additional techniques or checkpoints on metadata are needed. They may already be covered through WCAG.
Action Judy to pose to the ERT group: look at metadata in ERL
Action Judy to pose the metadata question to WCAG for 2.0
Action IMS to review WCAG conversation on metadata and give feedback.
Action Katie: provide info to IMS group on WCAG's conversation on metadata.
Action Phil to provide comments to IMS group on the whitepaper.
LN: joint meetings?
JB: AU might want to meet in Scotland in June with WCAG.
LN: www2002 in May, has lots of sessions including one Liddy is chairing. A full day workshop, morning on who's doing what. afternoon on metadata.
JB: Future co-meetings should be kept in mind.
JB: EO coordination issues. (She gives complete review of current EO docs, how they scope creep) lately lots of conversation on encouraging implementation of ATAG. Probably write a business case for adoption of ATAG in 3rd quarter. Can ATAG help?
ATAG: sure.
JB: we have a non-technical presentation of intro to access. another thing we could do is a simple, non-technical set of slides on authorign tools accessibility. also adding AU concerns to any other document they edit, for example the doc on organizational policy should have a section on selecting authoring tools.
PJ: And also on user agents.
JB: yes. The one pager "Getting started" could say something about choosing a user agent. "How PWDs use the web" needs updating. Authoring is one scenario. Could be useful to au tool developers. FAQs need updating. Look through EO deliverables and find additional ways we could reference ATAG, or other pieces we need to add. We could have more joint calls like the recent one.
PJ:
JB: EO goes where ever help is needed. WCAG was first. UA is in candidate rec so we need implementations. ATAG is very important since if a few of the major tools would implement it there would be a huge improvement. Also need major conversion tools, i.e. MS Word to HTML. Hope to get soem double A implementations.
GB: Why double A rather than single A?
JB: Single A would be nice, double A better. Single A seems to be on the horizon, maybe a couple of them in the next 6 months. Awareness is drastically better than a year ago. Set the bar high for motivation. Appropriate to promote single A for orgs requiring or recommending purchase of Authoring Tools since there are no double a, don't want to discourage them.
PJ: could we push on gls 1-6, since some groups are only working on gl 7 (the interface itself) which doesn't help with need for accessible content.
GB: How about certification process? Any org stamping tools with a lable?
JB: One in Europe, one in US, talking to about product reviews that would be more explicit than AU group's reviews. Would pay attention to 7 as well as 1-6. Need another party to come in and do those reviews. but they may need test suites. Could ATAG partner with QA working group?
--Meeting adjourned-----