Web Components & ARIA

[rniwa]

Session description

Session to discuss ARIA and web components:
https://github.com/WICG/webcomponents/blob/gh-pages/proposals/reference-target-explainer.md

Session goal

Concrete proposal for ARIA across shadow trees

Additional session chairs (Optional)

No response

Who can attend

Anyone may attend (Default)

IRC channel (Optional)

#webcomponents

Other sessions where we should avoid scheduling conflicts (Optional)

#26, #27, #28, #29

Instructions for meeting planners (Optional)

No response

Agenda for the meeting.

No response

Links to calendar

• 2024-09-25, 11:15 - 12:15

Meeting materials

• Minutes

[tpac-breakout-bot]

Thank you for proposing a session!

You may update the session description as needed and at any time before the meeting, but please keep in mind that tooling relies on issue formatting: follow the instructions and leave all headings and other formatting intact in particular. Bots and W3C meeting organizers may also update the description, to fix formatting issues or add links and other relevant information. Please do not revert these changes. Feel free to use comments to raise questions.

Do not expect formal approval; W3C meeting organizers endeavor to schedule all proposed sessions that are in scope for a breakout. Actual scheduling should take place shortly before the meeting.

[sanketj]

Minutes from TPAC breakout session discussion:

Sanket Joshi
dandclark: presenting https://1drv.ms/p/s!AtmAxiCADIM8nq09hubFJwqEsUqBDg?e=gmm2J1
11:18 AM
→ aaronlev, rniwa and xiaochengh joined
11:18 AM
Di set the topic: Breakout: webcomponents & ARIA - Catalina 3 - 11:15-12:15
11:18 AM
rniwa set the topic: "Custom elements and ARIA"
11:18 AM
→ siye, Michael_Warren0 and spectranaut_ joined
11:19 AM Sanket Joshi
dandclark: Shadow roots provide encapsulation, but sometimes ID-based refs needed to cross shadow DOM, ex. ARIA
11:19 AM
dandclark: Without totally exposing IDs and breaking encapsulation
11:19 AM
— spectranaut_ can someone link the presentation again?
11:19 AM Sanket Joshi
...: Reflected ARIA properties work for references out of shadow, but not into
11:20 AM
...: Focused on other direction today - unsolved problem
11:20 AM Jeffrey Yasskin
Am I reading right that the first direction only works with Javascript enabled?
11:20 AM
→ sarah joined (~sarah@9d40a032.publics.cloak)
11:20 AM
— jyasskin spectranaut_ https://1drv.ms/p/s!AtmAxiCADIM8nq09hubFJwqEsUqBDg?e=gmm2J1
11:20 AM Sanket Joshi
...: No way to associate label outside shadow with input inside shadow
11:21 AM
...: Proposal is to add a property on the shadow DOM called referenceTarget to forward external ID references into shadow DOM
11:21 AM
q+
11:21 AM
— Zakim sees justinf, rniwa on the speaker queue
11:21 AM Sanket Joshi
...: shadowrootreferencetarget for declarative shadow DOM
11:22 AM
...: Want this to work generally for all ID-based attrs, ie. aria, form, popovertarget, for, etc.
11:22 AM
...: aria-owns is an exception, reparents parts of accessibility tree, hard to get right
11:23 AM
— alice aria-owns may be feasible later on, but there are a lot of intricacies and we don't want to block the rest of the attributes on being sure we've got them all right
11:23 AM Sanket Joshi
...: so aria-owns excluded, no use case found for this either
11:23 AM
...: Would like to hear if there is a reason to support aria-owns
11:23 AM <Michael_Warren0>
ive never even heard of aria-owns :)
11:23 AM Sanket Joshi
...: Properties are reflected in the DOM, but cannot leak shadow contents
11:23 AM Lea Verou
present+
11:23 AM
q?
11:23 AM
— Zakim sees justinf, rniwa on the speaker queue
11:23 AM Lea Verou
q+
11:23 AM
— Zakim sees justinf, rniwa, lea on the speaker queue
11:23 AM Sanket Joshi
...: Quering properties in DOM will just get the host element's info
11:24 AM
...: Works with nested shadow tree, can pass through to deeper shadows
11:24 AM
...: Prototyped in Blink already
11:24 AM
→ ethanjv joined (~ethanjv@9d40a032.publics.cloak)
11:24 AM
— jyasskin Thanks alice. Having 1 exception seemed scary for developer understanding, but maybe there's a semantic reason to exclude it beyond just implementation difficulty.
11:25 AM Sanket Joshi
...: Forward looking, think about expanding to multiple ref forwarding via a ReferenceTargetMap
11:25 AM
...: Different attributes needed to target different elements
11:25 AM
...: Can name each property individually
11:25 AM
...: Will have both declarative and imperative versions
11:26 AM
...: ReferenceTargetMap and ReferenceTarget can be used together
11:26 AM
...: There is an open issue about reflection of properties with invariants
11:26 AM
...: form attr reflects form relationship, ex. a button associating with a form in shadow DOM
11:27 AM
...: need to reflect attr without leaking shadow contents
11:27 AM
→ jcraig joined (~sid363695@9d40a032.publics.cloak)
11:27 AM Sanket Joshi
...: One solution is to have property to reflect the host element (loosen invariant)
11:28 AM <bkardell_> Brian Kardell
null seems bad
11:28 AM
q?
11:28 AM
— Zakim sees justinf, rniwa, lea on the speaker queue
11:28 AM Sanket Joshi
...: Another issue is association with invalid id
11:28 AM
q+
11:28 AM
— Zakim sees justinf, rniwa, lea, masonf on the speaker queue
11:28 AM Sanket Joshi
...:
11:28 AM
...: No associated created, how to reflect that?
11:28 AM
...: One option is to return null, since no relationship
11:28 AM
...: But that is kind of leaking info about breakage in the shadow DOM
11:28 AM
q+
11:28 AM
— Zakim sees justinf, rniwa, lea, masonf, sorvell on the speaker queue
11:28 AM Sanket Joshi
...: Other option is to always reflect the host.
11:29 AM
q+
11:29 AM
— Zakim sees justinf, rniwa, lea, masonf, sorvell, aaronlev on the speaker queue
11:29 AM
Put this in formAssociated custom elements and provide a way to set the form the element is providing
11:29 AM Sanket Joshi
...: Looking for feedback on problem, solution, specific issues, etc.
11:29 AM
q?
11:29 AM
— Zakim sees justinf, rniwa, lea, masonf, sorvell, aaronlev on the speaker queue
11:29 AM
q?
11:29 AM
— Zakim sees justinf, rniwa, lea, masonf, sorvell, aaronlev on the speaker queue
11:29 AM Lea Verou
q-
11:29 AM
— Zakim sees justinf, rniwa, masonf, sorvell, aaronlev on the speaker queue
11:29 AM Lea Verou
q+
11:29 AM
— Zakim sees justinf, rniwa, masonf, sorvell, aaronlev, lea on the speaker queue
11:29 AM Jeffrey Yasskin
q+ jyasskin to ask about CSS alignment
11:29 AM
— Zakim sees justinf, rniwa, masonf, sorvell, aaronlev, lea, jyasskin on the speaker queue
11:29 AM
ack justinf
11:29 AM
— Zakim sees rniwa, masonf, sorvell, aaronlev, lea, jyasskin on the speaker queue
11:29 AM
ack rniwa
11:29 AM
— Zakim sees masonf, sorvell, aaronlev, lea, jyasskin on the speaker queue
11:30 AM Sanket Joshi
rniwa: One interesting thing with this feature is that there is a parallel to delegatesfocus
11:30 AM
...: That also forwards reference inside shadow tree
11:30 AM
...: There was a proposal for each attribute to have a target variant.
11:30 AM
...: Any reason for not doing that?
11:31 AM
q?
11:31 AM
— Zakim sees masonf, sorvell, aaronlev, lea, jyasskin on the speaker queue
11:31 AM Sanket Joshi
dandclark: Ben might know, but seems like dev ergonomics and future looking are good reasons
11:31 AM
...: Maybe others have more context
11:31 AM
→ CurtBellew joined (~CurtBellew@9d40a032.publics.cloak)
11:31 AM Sanket Joshi
alice: That's right. Want there to be a default forwarding option.
11:32 AM
...: Ex: the label/input example is a good one
11:32 AM
→ kschmi joined (~kschmi@9d40a032.publics.cloak)
11:33 AM Sanket Joshi
keith: having microsyntax of id-to-map is novel and might be better to avoid
11:33 AM Greg Whitworth
rniwa are you referring to cross-root aria? https://github.com/leobalter/cross-root-aria-delegation/blob/main/explainer.md
11:33 AM Sanket Joshi
s/keith/rniwa
11:33 AM
westbrook: microsyntax can help prevent explosion of attributes
11:33 AM
...: lots of new attributes being brought in
11:34 AM Jeffrey Yasskin
There's the data- -> dataset precedent. Could do shadowrootreferencetarget-aria-controls=foo
11:34 AM Sanket Joshi
masonf: exception would be listing them all
11:34 AM <bkardell_> Brian Kardell
+1 to what masonf said
11:34 AM Sanket Joshi
...: like that you only list one where the target is different from fallback
11:34 AM Lea Verou
q-
11:34 AM
— Zakim sees masonf, sorvell, aaronlev, jyasskin on the speaker queue
11:34 AM Sanket Joshi
rniwa: agree with mason
11:34 AM
→ TylerWilcock joined (~uid669735@9d40a032.publics.cloak)
11:35 AM
q?
11:35 AM
— Zakim sees masonf, sorvell, aaronlev, jyasskin on the speaker queue
11:35 AM
q?
11:35 AM
— Zakim sees masonf, sorvell, aaronlev, jyasskin on the speaker queue
11:35 AM Sanket Joshi
dandclark: imperative side of reftargetmap allows you set separate properties for each of them, but listing only those differing from fallback for declarative one seems like a good thing
11:35 AM
rniwa: maybe update delegatesfocus too in the same way
11:35 AM
masonf: like this overall, but tricky to make sure we don't expose the shadow root
11:36 AM
...: more of an implementation concern
11:36 AM
q?
11:36 AM
— Zakim sees masonf, sorvell, aaronlev, jyasskin on the speaker queue
11:36 AM Sanket Joshi
...: for form property, would be compatible to switch to a different property, but not if used in wrong ways
11:36 AM
ack masonf
11:36 AM
— Zakim sees sorvell, aaronlev, jyasskin on the speaker queue
11:36 AM
ack sorvell
11:36 AM
— Zakim sees aaronlev, jyasskin on the speaker queue
11:37 AM Sanket Joshi
sorvell: element could provide real form inside
11:37 AM
...: if there was something the element could opt into, that would solve it
11:37 AM
...: want a real form since this is a form API
11:38 AM
...: correct me if i'm wrong, but we shouldn't block the core feature on this
11:38 AM
ack aaronlev
11:38 AM
— Zakim sees jyasskin on the speaker queue
11:38 AM Sanket Joshi
masonf: +1
11:38 AM
aaronlev: noticed a pattern of creating a custom element and putting an aria label on it
11:38 AM
...: when widget is used, attr is copied
11:39 AM
...: problem is that there are two separate object in AX tree, can cause double speaking
11:39 AM James Nurthen
q?
11:39 AM
— Zakim sees jyasskin on the speaker queue
11:39 AM Sanket Joshi
...: worried that this is problematic since very hard to fix once component has starting doing this
11:39 AM Lea Verou
q?
11:39 AM
— Zakim sees jyasskin on the speaker queue
11:39 AM
q+
11:39 AM
— Zakim sees jyasskin, sorvell on the speaker queue
11:39 AM
w3c/aria#2303
11:39 AM Lea Verou
q+
11:39 AM
— Zakim sees jyasskin, sorvell, lea on the speaker queue
11:39 AM
⇐ sarah quit (~sarah@9d40a032.publics.cloak)
11:39 AM Sanket Joshi
...: filed above issue to give component authors to put role: none
11:40 AM
— jyasskin That problem strikes me as independent of this id-retargeting topic.
11:40 AM
q?
11:40 AM
— Zakim sees jyasskin, sorvell, lea on the speaker queue
11:40 AM Sanket Joshi
...: will allow removing element to be removed from AX tree
11:40 AM
ack jyasskin
11:40 AM Bridge daemon
jyasskin, you wanted to ask about CSS alignment
11:40 AM
— Zakim sees sorvell, lea on the speaker queue
11:40 AM <Michael_Warren0>
definitely a real issue, but imo thats separate from idref
11:40 AM Sanket Joshi
+1 to what jyasskin
11:40 AM
said
11:40 AM <Michael_Warren0>
aria-labels arent idrefs etc
11:40 AM Sanket Joshi
jyasskin: ref target seems inconsistent with how CSS exposes parts
11:41 AM
...: why?
11:41 AM
keith: was proposed that way previously, don't remember why we moved to this
11:41 AM
alice: that was more verbose
11:41 AM
Can we add my issue to a future agenda since I basically just don't want it to be dropped
11:41 AM
→ sarah joined (~sarah@9d40a032.publics.cloak)
11:41 AM Sanket Joshi
masonf: this is more generic, other approach needs to know more about component
11:41 AM
I guess i can just file an issue in web components
11:42 AM Westbrook Johnson
q+
11:42 AM
— Zakim sees sorvell, lea, westbrook on the speaker queue
11:42 AM Sanket Joshi
rniwa: in the case where you want to forward all refs to one element, much more verbose
11:42 AM
q?
11:42 AM
— Zakim sees sorvell, lea, westbrook on the speaker queue
11:42 AM
ack sorvell
11:42 AM
— Zakim sees lea, westbrook on the speaker queue
11:42 AM <Michael_Warren0>
+1 I would argue that label -> input is a lot of the implementation that would be used
11:42 AM Sanket Joshi
sorvell: Seen pattern of moving instead of copying
11:42 AM
...: Point is we need to get this feature landed
11:42 AM
ack lea
11:42 AM
— Zakim sees westbrook on the speaker queue
11:43 AM Sanket Joshi
lea: Much better than prev solutions
11:43 AM
...: in response to Aaron, what you said highlights a big pain point, need to find a solution
11:43 AM
...: not just about ARIA, for native elements, need to copy properties, etc.
11:44 AM
...: wanted to ask if reftarget can be read outside the shadow DOM somehow
11:44 AM Westbrook Johnson
Level 2 converastion: WICG/webcomponents#1068
11:44 AM
q+
11:44 AM
— Zakim sees westbrook, keithamus on the speaker queue
11:44 AM Sanket Joshi
...: would provide a natural solution to the 2nd open issue, will get host normally
11:44 AM
...: would also be nice for authors to define their own id ref down the line
11:44 AM
...: much much later
11:44 AM
...: shouldn't make that impossible though
11:45 AM
...: map needs to be able to ref the element itself, may need to specify the host as the reftarget
11:45 AM <Michael_Warren0>
if the host is the reference target, wouldnt that just be adding an ID onto the host which is already doable?
11:45 AM Sanket Joshi
dandclark: if reftargetmap doesn't specify reftarget explicitly, will fallback to the host
11:46 AM
ack westbrook
11:46 AM
— Zakim sees keithamus on the speaker queue
11:46 AM Sanket Joshi
...: for open shadow root, reftarget reflects actual thing in shadow tree
11:46 AM
...: obviously doesn't work for closed shadows
11:46 AM
westbrook: web components CG has been reporting since 2021 that this is one of the top issues that needs to be addressed
11:47 AM
...: want to make sure that feature expansion doesn't block progress
11:47 AM Greg Whitworth
+1 to westbrook
11:47 AM Sanket Joshi
...: still good to think about level 2 and get agreement from implementors about direction
11:48 AM
masonf: agree we are going in the right direction
11:48 AM
smaug: me too
11:48 AM
masonf: is reftarget is v1 and reftargetmap v2?
11:48 AM <Michael_Warren0>
+1 phased approach
11:48 AM Sanket Joshi
dandclark: yes
11:48 AM
q?
11:48 AM
— Zakim sees keithamus on the speaker queue
11:48 AM Lea Verou
q+
11:48 AM
— Zakim sees keithamus, lea on the speaker queue
11:48 AM Sanket Joshi
rniwa: need to check on support
11:48 AM
westbrook: when?
11:49 AM
rniwa: will try to get an answer ASAP
11:49 AM
ack keith
11:49 AM
— Zakim sees lea on the speaker queue
11:49 AM Sanket Joshi
keith: for form, submit bubble isn't composed
11:49 AM
...: is form already exposed via other means?
11:49 AM
...: if not, returning host element is fine
11:50 AM
I don't think submit is composed, but not sure.
11:50 AM Sanket Joshi
...: for open shadows, if submit happens outside form, can get the form?
11:50 AM
dandclark: don't want it to be different between open vs closed, but interesting thought
11:50 AM
sorvell: is submit really composed?
11:51 AM
ack lea
11:51 AM
— Zakim sees no one on the speaker queue
11:51 AM Sanket Joshi
[keith is looking]
11:51 AM
q+
11:51 AM
— Zakim sees sarah on the speaker queue
11:51 AM Sanket Joshi
lea: if reftarget ships in level 1 and reftargetmap in level 2, what are semantics of level 1?
11:51 AM
...: worried it won't cover any use cases
11:51 AM
masonf: still useful when forwarding different things to different elements
11:52 AM
rniwa: reftargetmap solves this problem
11:52 AM
lea: but ships in l2
11:52 AM
dandclark: reftargetmap broader solution, but reftarget still addresses simpler cases
11:52 AM
I filed WICG/webcomponents#1073 for my issue about duplicated properties
11:52 AM Sanket Joshi
...: will eventually need l2
11:52 AM
q?
11:52 AM
— Zakim sees sarah on the speaker queue
11:53 AM Sanket Joshi
masonf: would be useful for minimal leaf nodes
11:53 AM
westbrook: we think 80-90% of use cases to be covered
11:53 AM
...: perhaps authors are not making more complicated things because not possible
11:53 AM
ack sarah
11:53 AM
— Zakim sees no one on the speaker queue
11:53 AM Sanket Joshi
...: maybe after reftarget/reftargetmap, we'll see more of those
11:54 AM <Michael_Warren0>
https://lit.dev/playground/ submit event is not composed
11:54 AM Sanket Joshi
sarah: have a component where we can mark one element as target, works well
11:54 AM
...:is it possible to have a default role for custom element with reftarget/reftargetmap defined?
11:54 AM <Michael_Warren0>
per a playground I just did haha
11:54 AM
Yeah it's not composed. Click events are so if you had the button inside the form you could still get the composedpath
11:55 AM Sanket Joshi
dandclark: feasible, need to understand about need
11:55 AM
...: see this as a transparent pass through shadow DOM boundary
11:55 AM <Michael_Warren0>
bah, sorry for the bad link haha
11:55 AM
q?
11:55 AM
— Zakim sees no one on the speaker queue
11:56 AM Sanket Joshi
sarah: idea is that aria-label, etc. apply to outer element but you want it applied to the inner element
11:56 AM
...: really want to forward everything to the shadow node
11:57 AM
...: would be nicer to implicitly forward rather than explicitly forwarding everything
11:57 AM
...: may encourage right patterns for authors
11:57 AM
..: proposal is to have a different default AAM mapping
11:57 AM
alice: implicit role none?
11:57 AM
rniwa: can be overriden?
11:57 AM
sarah: yes and yes
11:58 AM
dandclark: would only be if you supplied reftarget but not reftargetmap?
11:58 AM
sarah: yes
11:58 AM
⇐ sarah quit (~sarah@9d40a032.publics.cloak)
11:58 AM
q?
11:58 AM
— Zakim sees no one on the speaker queue
11:58 AM Sanket Joshi
...: current default role is generic
11:58 AM
dandclark: will follow up with sarah
11:58 AM
q+
11:58 AM
— Zakim sees rniwa on the speaker queue
11:58 AM Sanket Joshi
...: would be possible
11:58 AM
ack rniwa
11:58 AM
— Zakim sees no one on the speaker queue
11:58 AM
— alice note that referenceTarget won't automatically forward attributes on the host like role
11:58 AM Sanket Joshi
rniwa: Webkit supportive of proposal
11:59 AM <bkardell_> Brian Kardell
🥳
11:59 AM Sanket Joshi
lea: Support from Mozilla?
11:59 AM Westbrook Johnson
q+
11:59 AM
— Zakim sees westbrook on the speaker queue
11:59 AM <Michael_Warren0>
woo! ship it!
11:59 AM Jesse Jurman
q+
11:59 AM
— Zakim sees westbrook, JRJurman on the speaker queue
11:59 AM Sanket Joshi
everyone: Olli said yes earlier.
11:59 AM Lea Verou
🚀
12:00 PM
q+
12:00 PM
— Zakim sees westbrook, JRJurman, sorvell on the speaker queue
12:00 PM
→ sarah joined (~sarah@9d40a032.publics.cloak)
12:00 PM Sanket Joshi
rniwa: should we move the list of attributes onto the shadow root instead of the map?
12:00 PM
...: would resemble delegatesfocus
12:00 PM
lea: maybe rename delegatesfocus?
12:01 PM
alice: I think delegatesfocus is doing something different. Picking focus based on direction.
12:01 PM Lea Verou
delegatesReferences?
12:01 PM
ack westbrook
12:01 PM
— Zakim sees JRJurman, sorvell on the speaker queue
12:01 PM Lea Verou
delegatesIds? delegatesTo?
12:02 PM
rniwa: I think it might be better to have individual attributes instead of a map with a micro syntax to be consistent with delegatesFocus.
12:02 PM Sanket Joshi
westbrook: do we have a spec writer on board? that would be helpful
12:02 PM
dandclark: i'll do it, unless others want to be onboard
12:02 PM
q?
12:02 PM
— Zakim sees JRJurman, sorvell on the speaker queue
12:02 PM Sanket Joshi
keith: i'll do it with Dan
12:02 PM
ack jrjurman
12:02 PM
— Zakim sees sorvell on the speaker queue
12:02 PM Sanket Joshi
alice: i'll help Dan and Keith work on the spec
12:03 PM
jesse: when custom element wraps an input, can we just inherit the role onto the host?
12:03 PM
keith: shadow roots are transparent
12:03 PM
...: want the opposite, role should be none or generic
12:03 PM
ack sorvell
12:03 PM
— Zakim sees no one on the speaker queue
12:04 PM Sanket Joshi
masonf: yeah, would cause two nested elements with same role
12:04 PM
sorvell: can we avoid long, verbose attribute names?
12:04 PM Lea Verou
q?
12:04 PM
— Zakim sees no one on the speaker queue
12:04 PM Sanket Joshi
...: can we revisit something like data-?
12:04 PM
q+
12:04 PM
— Zakim sees keithamus on the speaker queue
12:04 PM <Michael_Warren0>
q+
12:04 PM
— Zakim sees keithamus, Michael_Warren on the speaker queue
12:04 PM Sanket Joshi
...: no concrete proposal though
12:04 PM
ack keith
12:04 PM
— Zakim sees Michael_Warren on the speaker queue
12:04 PM Sanket Joshi
keith: don't mind on the attribute names, but microsyntax is worse than attributes
12:05 PM Lea Verou
qq+ to say +1 for shorter attributes, but we need to avoid hyphens so we can have the option of custom attributes open
12:05 PM
— Zakim sees lea, Michael_Warren on the speaker queue
12:05 PM Sanket Joshi
...: easy to bundle attrs using templating libraries
12:06 PM
sorvell: isn't that what happens with export parts?
12:06 PM
rniwa: semantics very different
12:06 PM
ack lea
12:06 PM Bridge daemon
lea, you wanted to react to keithamus to say +1 for shorter attributes, but we need to avoid hyphens so we can have the option of custom attributes open
12:06 PM Sanket Joshi
...: may be an argument for not doing microsyntax
12:06 PM
— Zakim sees Michael_Warren on the speaker queue
12:06 PM
q?
12:06 PM
— Zakim sees Michael_Warren on the speaker queue
12:06 PM Sanket Joshi
lea: custom attributes should still be explored
12:07 PM
ack Michael
12:07 PM
— Zakim sees no one on the speaker queue
12:07 PM Sanket Joshi
Michael_Warren: shorter attrs would be nice, especially people that hand write shadow DOM
12:07 PM
...: not everyone uses a build tool
12:07 PM
q?
12:07 PM
— Zakim sees no one on the speaker queue
12:08 PM Lea Verou
+1 to "not everyone uses a build tool"
12:08 PM Sanket Joshi
dandclark: Next step is to write the spec, continue to work on prototype, potentially consider OT in Blink
12:08 PM
masonf: Can we come back to the form question? Since we didn't get an answer.
12:08 PM
dandclark: Also the question about broken ref.
12:09 PM
masonf: Should just return the host?
12:09 PM
lea: yes
12:09 PM
lea: AX tree would also fallback to host, right/
12:09 PM
q?
12:09 PM Sanket Joshi
?
12:09 PM
— Zakim sees no one on the speaker queue
12:09 PM Sanket Joshi
dandclark: yes
12:09 PM <Michael_Warren0>
q+
12:09 PM
— Zakim sees Michael_Warren on the speaker queue
12:10 PM Sanket Joshi
keith: would be good to try this on a test bench
12:10 PM
...: hard to come up with the expected answer
12:10 PM
ack Michael
12:10 PM
— Zakim sees no one on the speaker queue
12:10 PM Sanket Joshi
...: weak argument though
12:10 PM Alice
q+ to go back to the

question if we have tiem
12:10 PM
— Zakim sees alice on the speaker queue
12:11 PM Sanket Joshi
Michael_Warren: Someone said id ref different than other ARIA references. How different are the implementations?
12:11 PM
dandclark: reftarget as proposed is not limited to ARIA
12:11 PM
...: supports all id based lookups
12:12 PM
masonf: string based ones could technically be adapted too
12:12 PM
rniwa: are we adjusting svg use too?
12:12 PM
lea: uses href, not id
12:13 PM
rniwa: resolved to change semantics, does use id for things with that SVG tree
12:13 PM
dandclark: not adjusted those
12:13 PM
rniwa: may need to review where id lookup is done
12:13 PM
q?
12:13 PM
— Zakim sees alice on the speaker queue
12:13 PM
q?
12:13 PM
— Zakim sees alice on the speaker queue
12:14 PM
ack alice
12:14 PM Bridge daemon
alice, you wanted to go back to the question if we have tiem
12:14 PM
— Zakim sees no one on the speaker queue
12:14 PM Sanket Joshi
alice: RE form question: Not sure about issues from leaking, but tells info about host, which means it can 'act' as an element you are targeting
12:14 PM
⇐ ethanjv quit (~ethanjv@9d40a032.publics.cloak)
12:14 PM Sanket Joshi
keith: what about .form today on an element?
12:14 PM
...: that's undefined, right?
12:15 PM
dandclark: yeah, might be compat since it returns null today and now invalid ref would cause it to return the host
12:16 PM
keith: Restating... .form returns null if element not form participant, but if reftarget always fallbacks to host, then .form will always fallback to the host element and look like a form participant
12:17 PM
dandclark: Maybe a good principal is that reftarget won't change property reflections or other internal associations
12:17 PM
q+
12:17 PM
— Zakim sees xiaochengh on the speaker queue
12:17 PM Sanket Joshi
...: For aria properties, would just change those internal associations, but not the property reflections
12:18 PM
keith: No way to use form property on custom element, right?
12:18 PM
⇐ kschmi and justinf quit
12:18 PM
ack xiaochengh
12:18 PM
— Zakim sees no one on the speaker queue
12:19 PM Sanket Joshi
xiaochengh: Maybe shouldn't push reftargetmap to level 2?
12:19 PM
lea: That's a lot more complex. Forwarding everything to one element is simpler and addresses a lot of use cases.
12:19 PM
xiaochengh: Risky. Ex: How do we address form?
12:20 PM
⇐ jarhar, masonf, jocelyntran3, alisonmaher, Di and xiaochengh quit
12:20 PM Sanket Joshi
masonf: Maybe we try to address all properties first and strive for reftarget only in level 1.