- From: Elad Alon via GitHub <sysbot+gh@w3.org>
- Date: Tue, 29 Mar 2022 20:25:20 +0000
- To: public-webrtc-logs@w3.org
> Given GC should not be observable, whether weak or not is not a hard requirement, it is a clarification. We're in agreement here. > If you do not like weak My concern is mostly with "reference". > I am fine reverting to the original PR: no mention of weak but a GC note. To ensure we're discussing the same thing, could you please link me to the revision you're referring to? > @youennf: As of objects referencing other objects that are potentially in other process, this is an existing pattern that proved to work well, let's use it. > > @jan-ivar: [HTML](https://html.spec.whatwg.org/#offscreencanvas-placeholder) uses "An [OffscreenCanvas](https://html.spec.whatwg.org/#offscreencanvas) object may hold a weak reference to a placeholder canvas element," Thank you for the reference, @jan-ivar. Please allow me some time to dig into these more closely. > "Weak reference" is also the term the TAG uses in their design principles in § 5.3. Don’t expose garbage collection. That example seems to refer to "weak reference" precisely in the interpretation that I want to avoid, is it not? "This means that you shouldn’t expose any API that acts as a weak reference, e.g. with a property that becomes null once garbage collection runs." It is **this** interpretation of "weak reference" that I worry would be confusing. -- GitHub Notification of comment by eladalon1983 Please view or discuss this issue at https://github.com/w3c/mediacapture-region/pull/24#issuecomment-1082340220 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Tuesday, 29 March 2022 21:54:22 UTC