| 01:43 | <GPHemsley> | why did wpt-pr-bot set me (and only me) as the assignee on this PR? https://github.com/web-platform-tests/wpt/pull/25010 |
| 01:45 | <GPHemsley> | I guess I'm ASCIIbetically first? |
| 01:45 | <GPHemsley> | https://github.com/web-platform-tests/wpt/blob/master/url/META.yml |
| 05:21 | <Domenic> | GPHemsley: it picks a random reviewer from the list of potential reviewers, and makes them the "assignee", so as to avoid all the reviewers thinking that some other reviewer will do it |
| 14:32 | <JakeA> | https://html.spec.whatwg.org/multipage/browsers.html#fully-active - this can't be used in synchronous code right? Stepping up through documents would be cross thread/process |
| 14:33 | <JakeA> | (it is used in synchronous code, but I'm trying to figure out if it's valid) |
| 14:33 | <Domenic> | JakeA: that feels like one of the things that's proactively pushed out to each process |
| 14:33 | <Domenic> | It's used all over the place as a precondition for sync stuff |
| 14:34 | <JakeA> | Domenic: should that be done in the spec (via posting tasks) or is it ok to hand-wave it? |
| 14:34 | <Domenic> | JakeA: OK to hand-wave... at least, it would be very far down on my list of priorities to make all the syncing implementations explicit |
| 14:35 | <JakeA> | cool |
| 19:36 | <Mek> | What is the latest thinking/recommendations on API namespacing; i.e. webidl namespaces (hardly used? not currently supported by blink bindings) vs global singletons (i.e. self.caches, self.indexedDB) vs properties of Navigator (i.e. navigator.storage) vs just exposing some methods directly on the global with no namespacing? |
| 19:38 | <Domenic> | Mek: I like exposing directly on the global. |
| 19:38 | <Domenic> | I dislike Navigator |
| 19:38 | <Domenic> | Namespaces seem good when you have a bunch of functions or constants but they can get overstretched pretty fast... |