W3C User Agent Teleconference for 1 December 2005 -- 1 Dec 2005

ok so it wasn't logging? I'll cut and past

<JR> JA: Only new column in roadmap is: Components Related

<JR> UAAG Checkpoints

<JR> JA: Hasn't send roadmap comments to PF explicitly.

<JR> JA: But received a few comments (not many)

<JR> AG: OK but you should ping AG with this.

<JR> AG: Thinks someone from UAAG should come and brief PF group.

<JR> JA: We have had a couple emails on this.

<JR> AG: Too narrow band for this.

<JR> JA: AG and JA to coord for future work

<JR> JA: Next step: review the Structural Roles table to ensure definitional alignment

<JR> between UAAG and the Roadmap

<JR> JA: Last week we were unsure of a few details.

<JR> AG: What is in draft for that section is pretty raw.

* dpoehlman has joined #ua

<JR> AG: Neither part of doc is definitive.

<JR> AG: Send a mail to xtech about structure.

<JR> action AG: send email to JA on structure

<JR> action AG : send email to JA on structure

* jallan invite rrsagent #ua

<JR> AG ACTION: send email to JA on structure

<JR> oh well

<JR> AG: to send Cathy info on joining xtech

<JR> AL: Should UA subscribe to xtech?

<JR> AG: No

<JR> AG: PF will be monitoring xtech so Cathy can send her comments to roadmap.

<JR> AG: Starts on top of roadmap -> Events -> 1.2

<JR> AG: Looking at uaag checkpoint to consider how it affects roadmap work.

<scribe> ACTION: AG to send email to JA and CL about xtech joining [recorded in http://www.w3.org/2005/12/01-ua-minutes.html#action01]

AG, AL: Discussion of events..

AG: Roadmap blank on ua reqs...

CL: Is AL's implementation covered by uaag?
... Trying to figure out if anything needs to be added to uaag?

AL: Maybe could be handled by some new uaag techs?

JA: wanted to bring this uaag gap analysis back to pf

DP?: Also are there contradictions between them?

AL: No change in DOM when CSS changes.

JA: Is that not value change?

AL: What I had to do in layout engine - when there is a re-style looks for things that were invisible and are now visible or viceversa - if this happens AL's system fires an MSAA event.
... But no DOM event happens. Not even a mutation event.

CL: Really?

AL: Style change that hasn't changed a node.
... Because style can change for reasons outside the DOM
... Against states that could disagree with reality -e.g. no focus state

AG: Should check back with PF about ua requirement that these things be generated by events.

DP: No parallel in Linux etc?

AL: Yes think there are events for show/hide.

in linux

AL: So previous discussion was on Windows
... Does ua say anything about 3 types of nav reg, container (treeview) with internal nav, non interactive content (e.g. progress meter)

JA, CL: ua does say that all info must be available to nav

AG: Has issue with this model - what happened in uaag dev people had lots of nav schemes - but in end only minimum came out.
... Roadmap will support richer nav methods.
... So for TAB key is like continuous play in DAISY
... vs. structured DAISY nav

AL: Already like this - in HTML we tab between lists and arrow within
... SR user will be informed of navigation options by the role (e.g. "listbox")
... In uaag does all interactive stuff have to be in tab order?

CL: Think uaag covers most of this nav conceptually.

JA: SR's have provided richer nav than regular ua's

AG: But should SR detect roles? Move to next same, move to next different, etc.

AL: This should actually a requirement for any new renderable markup language at w3c.

AG: DAISY has useful precedent -"escapable structures"
... eg. Help I want out of structure.
... Not just go to, but also excape from.
... scope is necessary for escaping (but not for going to)
... Is there ua-wg support for escape method?

DP: On uawg and pf
... would like to volunteer

CL: Is ecape clause in roadmap?

AG: No
... it's an example of kind of thing UAWG could decide should be added to roadmap.

JA: COuld be technique in navigate structure

AL: Possible issue: need labelled by and described by - DOM already has label-to etc
... ANything DOM lacks seems to be an issue
... Same prob in opposite direction HTML (which has label-for)

<jallan> Al: easy to go to element with id=x, but harder to go to "what points to id=x"

AG: COuld be new requirement on ua's to enrich dom with backwards relationships

AL: Do think hidden-visiblew issue is one for ua-wg
... Special case is Alerts.
... e.g. someone filling out form.

issue DP: cue 2 issues to xtech hidden/visible, and second is escape from container

<scribe> ACTION: DP to cue 2 issues to xtech hidden/visible, and second is escape from container [recorded in http://www.w3.org/2005/12/01-ua-minutes.html#action02]

JA: Need to continue this discussion
... Get sense that its too early to start building test cases.

<scribe> ACTION: AL to share 3-tier nav slides to ua-wg (on next uaag call - Dec 15) [recorded in http://www.w3.org/2005/12/01-ua-minutes.html#action03]

<scribe> ACTION: JA to take definitions concern to CG [recorded in http://www.w3.org/2005/12/01-ua-minutes.html#action04]

<jallan> emma - extensive multimodal alignment?

Jim - will you wrap up?

<jallan> scribe: Jan Richards

- DRAFT -

W3C User Agent Teleconference for 1 December 2005

1 Dec 2005

Attendees

Contents

Summary of Action Items

Scribe.perl diagnostic output