IRC log of components on 2020-10-26

Timestamps are in UTC.

17:01:12 [RRSAgent]
RRSAgent has joined #components
17:01:12 [RRSAgent]
logging to https://www.w3.org/2020/10/26-components-irc
17:01:14 [Zakim]
RRSAgent, make logs Public
17:01:15 [Zakim]
please title this meeting ("meeting: ..."), plh
17:01:36 [plh]
Title: Web Components
17:02:17 [caridy]
caridy has joined #components
17:03:01 [boazsender]
boazsender has joined #components
17:03:05 [bkardell]
bkardell has joined #components
17:03:09 [bkardell]
present+
17:03:14 [boazsender]
present+
17:03:18 [masonfreed]
present+
17:03:19 [plh]
present+
17:04:09 [smaug]
present+
17:04:14 [Lars]
Lars has joined #components
17:04:15 [leobalter]
leobalter has joined #components
17:04:15 [npm]
npm has joined #components
17:04:17 [hdv]
present+ Hidde
17:04:18 [JohnRiv]
JohnRiv has joined #components
17:04:28 [Rob]
Rob has joined #components
17:04:29 [leobalter]
present+ Leo Balter
17:04:45 [Rob]
present+ Rob Eisenberg
17:04:55 [dandclark]
present+ Dan Clark
17:05:00 [jensimmons]
jensimmons has joined #components
17:05:13 [Jemma]
Jemma has joined #components
17:05:45 [plh]
https://github.com/w3c/webcomponents/issues/877
17:05:50 [plh]
scribeNick: plh
17:06:01 [kenchris]
kenchris has joined #components
17:06:05 [plh]
Agenda: https://github.com/w3c/webcomponents/issues/877
17:06:13 [zcorpan]
zcorpan has joined #components
17:06:22 [EliHowey]
EliHowey has joined #components
17:06:39 [zcorpan]
present+
17:06:46 [zcorpan]
Simon Pieters, Bocoup
17:06:57 [graynorton]
graynorton has joined #components
17:07:22 [plh]
ryosuke: any topic to discusss at this meeting?
17:07:29 [jensimmons]
present+
17:07:32 [jensimmons]
Jen Simmons, Apple
17:07:50 [plh]
brian: can we do them in reverse order?
17:07:57 [plh]
ryosuke: sure
17:08:12 [plh]
Mason: also, we can talk is #45
17:08:47 [bkardell]
https://docs.google.com/document/d/1tWvWBuPtQiJFhOEEfMiLtPZuxag1CV_tbTJHE187B_g/edit#heading=h.9ut5sja3keva
17:08:49 [leobalter]
https://docs.google.com/document/d/1tWvWBuPtQiJFhOEEfMiLtPZuxag1CV_tbTJHE187B_g/edit#
17:09:07 [leobalter]
You've got my by the hair, bkardell
17:09:20 [kim_wooglae]
kim_wooglae has joined #components
17:09:52 [justinfagnani]
justinfagnani has joined #components
17:09:53 [Kunal]
Kunal has joined #components
17:09:54 [kenchris]
present+
17:09:55 [pea]
pea has joined #components
17:10:01 [plh]
Topic: SVG unknown element
17:10:49 [plh]
Brian: SVG and MathML are both in HTML spec under embedded content
17:11:02 [plh]
[Brian Kardell, Igalia]
17:11:41 [plh]
Brian: they lack basic IDL definition until recently
17:11:55 [plh]
... SVG has the concept of SVGUnknownElement
17:11:58 [yuzhe-han]
yuzhe-han has joined #components
17:12:05 [plh]
... Iagalia would like to do the same for MathML
17:12:32 [plh]
... both of those are interested in taking advantages of custom elements and ShadowDOM
17:12:53 [plh]
Justin: this is realted to do custom svg element in the future
17:13:05 [plh]
... we've been working on a polyfill
17:13:17 [plh]
... SVGUnknownElement is spec'ed out in SVG 2.0
17:13:28 [plh]
... make it behave like a g element
17:13:28 [Jemma]
s/realted/related
17:13:38 [plh]
... would be prefer to custom elements
17:13:46 [plh]
s/Iagalia/Igalia/
17:13:57 [bkardell]
q+
17:13:59 [plh]
... but nobody working on SVG anymore, and no implementation for SVG 2
17:14:08 [plh]
... so I wanted to check in
17:14:15 [plh]
... and if anybody knows what's going on
17:14:50 [Jemma]
present+
17:15:15 [plh]
Tess: I agree with Justin. SVG is not high priority for implementers
17:15:19 [Westbrook]
Westbrook has joined #components
17:15:20 [justinfagnani]
present+
17:15:34 [plh]
Tess: SVG has 2 kinds
17:15:43 [plh]
... SVG document as an XML doc
17:15:46 [plh]
... and inline in HTML
17:15:53 [plh]
... in the XML, it's obvious
17:16:00 [plh]
... in the HTML, less obvious
17:16:08 [plh]
... wouldn't you get HTMLUnknownElement?
17:16:17 [plh]
... are you proposing to change the HTML parser?
17:16:35 [plh]
Simon: when you have an SVG start tag
17:16:46 [plh]
... so any unknown elements get the SVG namespace
17:17:12 [plh]
Tess: so the only would be in the SVG space.
17:17:23 [plh]
Simon: it would appear in the DOM inteface
17:17:38 [plh]
Alex: wouldn't be a breaking change?
17:17:47 [plh]
Simon: not aware of any
17:18:07 [plh]
Brian: I agree it's not making changes to the parser
17:18:20 [plh]
... worked for our prototype in MathML
17:18:29 [plh]
... would be the same for SVG
17:18:41 [plh]
... we are interested in SVG and are actively implementing
17:19:01 [plh]
... so it's about going through custom elements and shadowDOM
17:19:54 [plh]
Anne: Wouldn’t there be a change to the parser given that it introduces script execution at those points?
17:20:13 [plh]
Simon: script execution doesn't really match what browser implementers
17:20:25 [plh]
... we'd need to update the spec anyway to match implementation
17:21:09 [plh]
... for the DOM interface, it's not a material change to the HTML parser
17:21:21 [plh]
... script eexcution would need to change
17:21:50 [plh]
Olli: list of possibilities or problems? like IDref?
17:22:11 [plh]
Justin: there are several SVG issues in the webcomponents repo
17:22:12 [muodov]
muodov has joined #components
17:22:25 [plh]
... should we readjust them and/or open new ones?
17:22:40 [plh]
Olli: having the big picture would be useful
17:22:52 [plh]
Brian: is there any interest to do that?
17:23:24 [plh]
Justin: it seems a big downside to the world at the moment. blocking use cases
17:23:42 [slightlyoff]
+1 to making this possible
17:23:58 [Westbrook]
+1 to making this possible
17:24:01 [bkardell]
+1 to moving on - I thought it would actually be quicker than that :)
17:24:07 [plh]
Ryosuke: let's move next topic. maybe best to open issues and track them
17:24:34 [plh]
Justin: there are some issues open already
17:24:40 [plh]
Ryosuke: have an umbrella issue
17:24:58 [plh]
Topic: Scoped custom element registries
17:25:16 [plh]
(Justin presents slides)
17:26:17 [plh]
Justin: idea to create a registry, associate with shadowroot
17:26:33 [Joshue108]
Joshue108 has joined #components
17:26:35 [plh]
... shadowdom is scoped to a registry or not
17:26:41 [Joshue108]
present+
17:27:06 [slightlyoff]
present+
17:27:08 [plh]
Justin: seems good consensus on the MVP
17:27:19 [plh]
... some open question
17:27:24 [plh]
... #895
17:27:41 [plh]
... Upgrade Ordering
17:27:42 [Rob]
q+
17:27:49 [plh]
q- bkardell
17:28:12 [plh]
Justin: would be nice not have to walk the entire document but just a subtree
17:28:23 [plh]
q- Rob
17:28:46 [plh]
Rob: status on how this would interact with static shadowDOM?
17:29:04 [plh]
s/static/declarative/
17:29:09 [plh]
Justin: still open question
17:29:57 [plh]
... [quoting Ryosuke]
17:30:30 [plh]
Ryosuke: what about constructors?
17:30:43 [plh]
... not making constructors would be ok?
17:31:09 [plh]
... there are some ways when we couldn't make it to work
17:31:15 [kevinpschaaf]
kevinpschaaf has joined #components
17:31:30 [plh]
Caridy: that case would be tricky, ie refactor hassle
17:31:36 [plh]
... we focused on MVP
17:31:47 [plh]
... there are some ideas for the future
17:32:02 [plh]
... we wouldn't add more features unless people ask for them
17:33:04 [plh]
?: if you register a constructor in the registry, what do you get if we're not supporting constructors?
17:33:18 [plh]
Justin: you get the same constructor you passed to it
17:34:11 [plh]
... in order to preserve backward compatibility, constructors are assumed to only work if they are in the global registry
17:34:31 [plh]
... in the future, if we want it to work, it would have to be a different API call
17:34:38 [plh]
... we want some kind of consistent behavior here
17:34:53 [plh]
Caridy: yes, many holes there.
17:35:08 [plh]
Ryosuke: in terms of solving this problem later.
17:35:25 [plh]
... each constructor can only be register with at most one registry
17:35:38 [plh]
... if we don't do that, we would forever close a path
17:36:16 [justinfagnani]
q+
17:36:17 [plh]
... one way to have a proxy
17:37:00 [plh]
Ryosuke: if we allow same object with multiple registration, I'm afraid of having an API that would prevent future solutions
17:37:19 [plh]
Brian: so prevent the same constructor to be registered in different scope?
17:37:23 [plh]
Ryosuke: right
17:38:02 [plh]
... we could have a proxy or prevent multiple registrations
17:38:09 [plh]
Brian: agreed
17:38:37 [caridy]
Brian->Caridy
17:38:47 [plh]
Justin: s/Brian:/Caridy/
17:40:09 [plh]
Justin: image you have a class that is not registering in the global registry. if it gets used in an other package, the registry could fail
17:40:55 [plh]
?: it feels like adopting Ryosuke doesn't prevent from changing in the future
17:41:10 [plh]
... we can continue to work on this problem
17:41:16 [Rob]
q+
17:41:24 [plh]
Justin: the multi-actor hazard might be an issue
17:41:37 [plh]
... this solves the constructor problem however
17:41:53 [Westbrook]
q+
17:41:57 [plh]
... I'm inclined to say that, with this solution, we would get callable constructor
17:42:30 [plh]
?: I'm happy with that. we need to agree on the definition of MVP. is it small enough to try it out?
17:43:42 [plh]
Ryosuke: let's wait on what to do until we get more feedback?
17:44:14 [plh]
Justin: fine with this. newable constructor aren't that important. most things are created by createElement
17:44:23 [plh]
... I'm fine with this restriction for now
17:45:04 [plh]
Proposal: registration only accepts constructors that haven't been registered yet?
17:45:25 [bkardell_]
bkardell_ has joined #components
17:45:25 [plh]
Justin: we should be able to merge my PR
17:45:28 [Rob]
I don't think this combination of constraints is going to work.
17:45:39 [plh]
... and we can track that one decision separately
17:45:40 [Rob]
I'
17:45:46 [Rob]
I'd like to get a word in real quick.
17:46:44 [plh]
Rob: if we have a situation with no inheritance and single registry, you'll have problems when you can't use the component
17:46:50 [plh]
... eg a button
17:47:06 [plh]
?: you would subclass the button
17:47:13 [plh]
?: or proxy
17:47:19 [plh]
q- Rob
17:47:34 [Westbrook]
This means that is any of the code in your page uses scope then likely ALL code in your page needs to do this, which feels really forceful.
17:47:59 [Westbrook]
q-
17:48:02 [plh]
Justin: without subclassing, we don't have a way to introduce a constructor in a particular scope
17:48:09 [plh]
... I'll open an issue on this
17:49:29 [plh]
Ryosuke: can we agreement on restricting the registry?
17:49:30 [plh]
Justin: fine with me
17:49:44 [plh]
Topic: Next meeting
17:50:07 [plh]
RyosukeL CSS script modules will be discussed in a different session
17:50:23 [plh]
Caridi: what about the ordering issue?
17:50:31 [plh]
Ryosuke: let's take it offline
17:50:39 [plh]
Justin: I'll open separate issues
17:50:50 [plh]
... after we merge the PR
17:50:50 [Rob]
We need to keep discussing this issue of ctors in multiple registries.
17:51:29 [plh]
zakim, end meeting
17:51:29 [Zakim]
As of this point the attendees have been bkardell, boazsender, masonfreed, plh, smaug, Hidde, Leo, Balter, Rob, Eisenberg, Dan, Clark, zcorpan, jensimmons, kenchris, Jemma,
17:51:32 [Zakim]
... justinfagnani, Joshue, slightlyoff
17:51:32 [Zakim]
RRSAgent, please draft minutes v2
17:51:32 [RRSAgent]
I have made the request to generate https://www.w3.org/2020/10/26-components-minutes.html Zakim
17:51:34 [Zakim]
I am happy to have been of service, plh; please remember to excuse RRSAgent. Goodbye
17:51:38 [Zakim]
Zakim has left #components
18:15:26 [CaioGondim]
CaioGondim has joined #components
19:23:39 [danyao]
danyao has joined #components
20:32:16 [bicknellr]
bicknellr has joined #components
20:35:36 [bicknellr]
bicknellr has left #components
20:36:56 [bicknellr]
bicknellr has joined #components
20:42:20 [bicknellr]
/window/next
20:43:05 [bicknellr]
oops, trying to learn irrsi
21:10:24 [JohnRiv]
JohnRiv has joined #components
21:22:26 [bicknellr]
bicknellr has joined #components
22:06:12 [JohnRiv]
JohnRiv has joined #components
23:08:35 [bicknellr]
bicknellr has joined #components
23:09:42 [JohnRiv]
JohnRiv has joined #components