WECG Meetings 2022, Public Notes, Oct 13

Agenda: github issues

The meeting will start at 3 minutes after the hour.

Carry-over from previous meetings
- None
Other new issues
- Issue 287: Allow workers & shared workers to be created within an extension service worker
- Issue 288: Inconsistency: Host permissions in Manifest V3
Open discussion queue (add yourself at the bottom)
- The deliverables of the WECG and its meetings
Check-in on ongoing issues
- Old housekeeping issues (Added by Carlos, in an attempt to reduce the amount of open issues)
  - Issue 1: What to do with the older CG?
  - Issue 9: New Issue Template
  - Issue 29: Update contributing guide with info on current communication channels
  - Issue 30: Set up Bikeshed spec pipeline
- Issue 282: Proposal: declaring background scripts in a neutral way
- Issue 281: Inconsistency: runtime.getURL()

Issue 287: Allow workers & shared workers to be created within an extension service worker

[simeon] In short, SW does not allow nested workers to be created. On the Chrome side we are supportive of this.
[rob] I think that we support nested workers in Firefox, but in any case we are supportive of this capability.
[timothy] Also supportive. May also make sense to have this be available on the open web.
[simeon] I was specifically talking about extensions, but agree that this would be a useful capability on the open web.
[frederic] What are the next steps?
[simeon] A representative of the group to reach out to the groups related to workers, e.g. Timothy or I.

Issue 288: Inconsistency: Host permissions in Manifest V3

[timothy] Looks like a UI issue between Chrome and Edge. Out of scope for this group.
[simeon] Out of curiosity, how should we deal with UI differences, despite them being out of scope?
[timothy] There are always going to be UI differences, but I don't think that we can resolve them.
[simeon] How much utility is there in issues that seem interesting, but not part of the spec? Close and allow inactive discussion to continue?
[timothy] I'm in favor of closing it.
[rob] Also support closing it. Browser-specific (UI) issues should be reported on the browser-specific issue tracker.
[mukul] Please add Edge specific issue at https://github.com/microsoft/MicrosoftEdge-Extensions/

Old housekeeping issues (Added by Carlos, in an attempt to reduce the amount of open issues)

Issue 1: What to do with the older CG?
- [simeon] I've met with Florian, it's now on me to decommission the old repo etc.
Issue 9: New Issue Template
- [rob] We could link issue 155 about external issue trackers. I'll create a template.
Issue 29: Update contributing guide with info on current communication channels
- [rob] I'll update this too.
Issue 30: Set up Bikeshed spec pipeline
- [mukul] I think that we've completed this already.
- [simeon] I'll close the issue.

Issue 282: Proposal: declaring background scripts in a neutral way

[simeon] Unfortunately, due to travel and schedule conflicts I have not been able to follow up yet. I plan to sync with the team on this and other issues next week.
[carlos] limited event pages; Could type:”module” be supported there too?
[rob] Yes. I had already created a Firefox bug for that: https://bugzilla.mozilla.org/show_bug.cgi?id=1775574
[timothy] I think that it makes sense to type:”module” for event pages. Not sure if we currently support that for Service workers, but I would be in favor of that for event pages.
[carlos] Should we file a separate WECG issue to track support for type:”module”?
[rob] Yes.
[carlos] Done: w3c#289

The deliverables of the WECG and its meetings

[simeon] During TPAC we discussed allocating meeting time to working on the specification.
[oliver] Since we're usually running out of time, a separate meeting would make sense.
[timothy] I am supportive of another meeting if we keep filing this meeting.
[simeon] Historically, we have used all of our time for discussions. If we half the time for the usual discussions, when we would lose time for the open discussion.
[rob] So far most of the meetings have been filled with useful discussions. So far we have a lot of issues where we have loose consensus, but have not followed through on and driven to completion. Suggest triaging old issues to identify such items.
[simeon] Sounds valuable, even if just to clean up our backlog of open issues. Identifying items where we are all aligned is a good use of organizer's time.
[timothy] At TPAC we came up with the idea of gathering WebIDL files. I think it makes sense to create a directory in the WECG github and collect IDL files. I would be willing to do that and submit PRs with Safari's IDL files.
[rob] Sounds useful. With the IDL files present it'll be easier for other browser vendors to verify interoperability in terms of API surface.
[simeon] We have just identified two large tasks: going through the backlog and submitting IDL files. We can divide these tasks - I will go through the backlog and Timothy to curate the IDL files.

Issue 281: Inconsistency: runtime.getURL()

[carlos] Safari has filed issues for inconsistencies 1, 2 and 11. What's the status?
[timothy] Have filed the issues but no progress on that yet.
[timothy] Some inconsistencies are due to canonicalization of URLs, which Firefox and Safari do, but Chrome doesn't.
[rob] Inconsistencies in the string representation of the URL are fine, as long as the resolved URL maps to a valid resource when it should.
- [simeon] +1. Inconsistencies resulting in a material difference between browsers would be of interest.
- [timothy] +1
Is there any practical impact of the different outcomes among the reported inconsistencies?
[carlos] Inconsistency 3, where the return value of getURL() can return a non-extension URL, can potentially be a security issue.
[rob] Agreed with that.
[simeon] Could these examples be added to a test suite?