| 16:38 | <smaug____> | huh, the spec terminology about microtasks has changed a lot since I read it last time |
| 16:46 | <annevk> | Really? Hmm |
| 16:52 | <smaug____> | what is " execute a compound microtask subtask" about? |
| 16:52 | <smaug____> | and what is 'solitary callback microtasks'. the spec says "This specification only has solitary callback microtasks." |
| 16:52 | <smaug____> | but nothing seems to use them |
| 16:53 | <smaug____> | and nothing explains what "solitary callback microtasks" |
| 16:53 | <smaug____> | are |
| 16:55 | <smaug____> | or how they are used |
| 16:55 | <smaug____> | I guess it is just a name for something, but I don't know when it is used |
| 16:58 | <annevk> | smaug____: compound is for mutation observers |
| 16:58 | <smaug____> | sure, that is clear |
| 16:58 | <annevk> | smaug____: to make sure they do not get interleaved with promises and such |
| 16:59 | <annevk> | smaug____: forgot where HTML queues microtasks, used to be for table sorting but that is out |
| 16:59 | <smaug____> | but what is "solitary callback microtasks'" ? |
| 17:00 | <smaug____> | and what is " execute a compound microtask subtask" about, why is it needed? |
| 17:00 | <annevk> | smaug____: I guess the counterpart to a compound one? |
| 17:00 | <smaug____> | ok, and when is it used? |
| 17:01 | <smaug____> | I guess with Promises, but where is that defined? |
| 17:02 | <annevk> | smaug____: JS integration I guess? |
| 17:02 | <smaug____> | and where is that ? :) |
| 17:02 | <annevk> | smaug____: sorry, not in a good place to research |
| 17:02 | <smaug____> | ok |
| 17:02 | <annevk> | smaug____: just have a phone |
| 17:03 | smaug____ | is in a train so network may die at any time |
| 17:03 | <smaug____> | I really don't understand the comment "This specification only has solitary callback microtasks." when nothing in the spec seems to use such microtasks |
| 17:03 | <annevk> | smaug____: https://html.spec.whatwg.org/multipage/webappapis.html#integration-with-the-javascript-job-queue |
| 17:05 | <annevk> | smaug____: compound almost reads like a nested task |
| 17:06 | <annevk> | smaug____: I wonder what that does to timing of Promise callbacks |
| 17:07 | <annevk> | smaug____: promise callbacks queued during mutation callbacks run directly after or some such? |
| 17:07 | <annevk> | Anyway, I once again recommend reaching out to aklein |
| 17:08 | <annevk> | I have the feeling every other month someone from Moz is confused by this 😊 |
| 17:08 | <smaug____> | I thought Domenic had written this stuff. Ok, thanks |
| 17:08 | <annevk> | Domenic wrote the JS integration stuff |
| 17:09 | <annevk> | Compound microtasks are way older |
| 17:09 | <annevk> | Basically existed since mutation observers |
| 17:09 | <smaug____> | annevk: well, I'm confused. I invented microtasks and then every couple of months more complications are added to the specs dealing with them, and somewhat amusingly no one dares to ask me about the changes :p |
| 17:11 | <annevk> | smaug____: you should talk to aklein |
| 17:11 | <annevk> | smaug____: also, nothing changed since forever |
| 17:11 | <smaug____> | well, all the slots stuff is quite new |
| 17:12 | <annevk> | smaug____: could maybe look at git log of DOM and find out when it started using compound microtask |
| 17:12 | <annevk> | smaug____: oh yes, that is |
| 17:12 | <annevk> | smaug____: we discussed that I think |
| 17:13 | <annevk> | smaug____: that is not a microtask change though |
| 21:09 | <tobie> | TabAtkins: struggling with how to markup some of the WebIDL. For example: what is the type of void? dfn? typedef? |
| 21:10 | <tobie> | http://heycam.github.io/webidl/#idl-void |