| 13:54 | <Michael Ficarra> | honestly, MDN |
| 16:56 | <shu> | i agree how-we-work doesn't feel appropriate |
| 16:57 | <shu> | it also doesn't appropriate for MDN as general web developer-facing documentation. are there MDN articles that are tagged like "internals, intended for implementers, tooling authors, polyfill authors, etc"? |
| 17:02 | <Rob Palmer> | Shall we create tc39/docs repo? |
| 17:05 | <shu> | i feel like no, actually, i agree that MDN is a better fit than our spinning up a new repo. i'd just like it to not muddle with the general developer-facing docs on there |
| 17:26 | <Michael Ficarra> | @shu developers implement (both consume and produce) these protocols, there's nothing inappropriate about it living on MDN IMO |
| 17:28 | <bakkot> | I think this is too much detail for MDN personally; very few developers should be thinking about this stuff |
| 17:29 | <shu> | @shu developers implement (both consume and produce) these protocols, there's nothing inappropriate about it living on MDN IMO |
| 17:29 | <shu> | like the only way they would produce this protocol is if they are writing a VM, or a downlevel transform, right? |
| 17:30 | <shu> | but i think we're splitting hairs: i think MDN is fine. i don't want it to be misconstrued as "general knowledge" |
| 17:31 | <bakkot> | here's MDN's page currently (which josh-cena has done a fair bit of work on recently) https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Iteration_protocols |
| 17:32 | <bakkot> | which I think covers everything developers should know |
| 17:32 | <bakkot> | (and I guess is partially based on my doc) |
| 17:32 | <bakkot> | I guess I mostly feel weird about MDN being a place to live because I don't treat MDN as authoritative about the language, and I don't expect other implementers to do so either |
| 17:33 | <bakkot> | it's describing things at the level of detail you need to know as a developer, which can elide certain things which are important only to implementers |
| 17:34 | <Michael Ficarra> | like the only way they would produce this protocol is if they are writing a VM, or a downlevel transform, right? |
| 17:34 | <shu> | that is really not true for this case |
| 17:36 | <shu> | kevin's document would be inappropriate level of detail for someone implementing an object that satisfies the protocol |
| 17:42 | <Michael Ficarra> | 🤷♂️ it's the level of detail I'd want, but maybe that's just me |
| 17:43 | <shu> | i can say with some certainty you are grossly miscalibrated if the calibration we're going for is "all sorts of developers" |
| 17:43 | <Michael Ficarra> | lmao kaomoji are impossible on matrix |
| 20:06 | <ljharb> | ¯\_(ツ)_/¯ |