| 06:09 | <gsnedders> | trying to add |
| 08:30 | <gsnedders> | idea for StackOverflow bot: comment on any answer using TR links (cc Domenic) |
| 12:14 | <annevk> | Domenic: one or more of your comments on https://github.com/whatwg/html/pull/3038 wasn't on the final diff |
| 12:14 | <annevk> | Domenic: which makes them hidden |
| 12:14 | <Domenic> | Oh, oops, sorry |
| 12:55 | <annevk> | Should we start archiving blame files for whatwg/html again? |
| 12:56 | <annevk> | Even on desktop it takes quite a while to compute them, downloading them would likely be faster |
| 12:56 | <annevk> | I guess I could just keep a local archive |
| 13:24 | <annevk> | JakeA: Domenic: put up a commit message at https://github.com/whatwg/html/pull/3038#issuecomment-331154972 |
| 13:24 | <annevk> | JakeA: Domenic: I suggest we squash land this and don't try anything more fancy |
| 13:26 | <Domenic> | lgtm |
| 13:26 | <JakeA> | same |
| 13:33 | <annevk> | JakeA: will you follow-up with service workers on this? |
| 13:33 | <annevk> | JakeA: to potentially redo the job stuff? |
| 13:33 | <annevk> | JakeA: there was a lot of talk about that in the original issue I just realized that we didn't quite touch on yet |
| 13:37 | <JakeA> | annevk: Created https://github.com/w3c/ServiceWorker/issues/1197 for it. It isn't a small job, but yeah I think it'd clean stuff up a lot. |
| 13:38 | <JakeA> | annevk: added it to me todo list, but want to get the abortable fetch + service worker stuff done first |
| 13:38 | <JakeA> | s/me/my I didn't mean to sound like Oliver Twist |
| 13:49 | <wanderview> | JakeA: are "parallel queues" equivalent to SW job queues, though? |
| 13:50 | <wanderview> | I'm not sure I want to rewrite our job queue code again |
| 13:51 | <JakeA> | wanderview: the change should be editorial |
| 13:51 | <wanderview> | ok |
| 13:51 | <JakeA> | it's a pita in the spec |
| 13:52 | <wanderview> | JakeA: I've been wondering how many PITA spec stuff you are going to do before you decide you want to travel giving talks again :-) |
| 13:53 | <JakeA> | wanderview: haha well umm I may have booked myself into a Bangalore and Singapore trip in December and January |
| 13:53 | <JakeA> | Going to keep a balance though |
| 13:53 | <JakeA> | hopefully |
| 13:53 | <wanderview> | cool |
| 13:53 | <JakeA> | A family holiday coincides with Chrome Dev Summit this year, which got me off the hook for all that work |
| 14:11 | <nox> | JakeA: How dare you prioritise family over dev summits!!1! |
| 14:13 | <JakeA> | nox: It is in-laws, so it's not all fun. I mean, they're lovely but there's always that extra layer of filtering that needs to be active to avoid saying anything wrong. |
| 14:14 | <nox> | JakeA: DO YOU MEAN CSS FILTERS |
| 14:14 | <nox> | :) |
| 14:14 | <JakeA> | Well, in CSS terms it's .liberal-opinions { display: none; } unfortunately |
| 14:15 | <nox> | JakeA: Ah ah. |
| 14:16 | <JakeA> | They're not that bad, just a bit Daily Mail. |
| 14:16 | <annevk> | JakeA: but you enjoy that kind of stuff all the time on Twitter |
| 14:16 | <nox> | JakeA: I always have to remember where the people I talk to come from, |
| 14:16 | <nox> | because to the French me, "liberal" is usually a swear word. |
| 14:17 | <JakeA> | annevk: yeah, but I don't have to have Christmas dinner with the people on Twitter :D |
| 14:18 | <tantek> | JakeA, I bet you're good at asking the right questions tho |
| 14:19 | <JakeA> | tantek: I dunno, questions sound dangerous |
| 14:21 | <wanderview> | JakeA: wait, do they display the English flag? |
| 14:21 | <annevk> | I like pushing my in-laws a little bit |
| 14:21 | <Ms2ger> | English or UK? :) |
| 14:21 | <JakeA> | wanderview: nah they're not racist |
| 14:22 | <wanderview> | Ms2ger: English... apparently its become the flag of extreme right in UK |
| 14:22 | <wanderview> | JakeA: well, thats something you can talk about then |
| 14:22 | <annevk> | They came around on Sinterklaas being racist after all |
| 14:23 | <nox> | Latest far right drama around here is food-related, an exec from the Front National took a selfie eating a couscous instead of choucroute while being in Strasbourg. |
| 14:24 | <annevk> | Sounds pretty serious |
| 14:24 | <nox> | Part of his own party's electoral base was so disappointed he called them idiots publicly, amazing. |
| 14:26 | <Ms2ger> | Couscous trucks on every corner? |
| 14:26 | <nox> | Hah. |
| 14:30 | <JakeA> | haha |
| 14:31 | <wanderview> | JakeA: I guess Cache.add() should just work with abortable fetch if the signal is passed through correctly? |
| 14:32 | <wanderview> | I wonder if we have a test for that yet |
| 14:33 | wanderview | sees a typo in his tweet. |
| 14:33 | <JakeA> | wanderview: hmm, I don't think we do. I'll add that when I do the SW tests |
| 14:33 | <JakeA> | wanderview: When is 57 beta? |
| 14:33 | <wanderview> | JakeA: dev-edition is the same binary as beta... official beta release is next week |
| 14:34 | <wanderview> | JakeA: maybe I shouldn't point people at dev-edition 57 yet... but I don't care :-) |
| 14:34 | <JakeA> | wanderview: niiiice! I'll get an abortable fetch post for https://developers.google.com/web/updates/ together tomorrow |
| 14:34 | <wanderview> | thanks! |
| 14:35 | <JakeA> | wanderview: Pointing at dev is totally fine, (I think) we just try and limit /web/updates/ stuff to things heading for stable |
| 14:36 | <JakeA> | Usually Chrome stable of course, but y'know, whatever |
| 14:36 | <wanderview> | JakeA: here is the dev-platform mail where we indicated we are going to ship: https://groups.google.com/d/msg/mozilla.dev.platform/jk6XtUch4GQ/RnlbpugIAQAJ |
| 14:36 | <JakeA> | ta! |
| 14:39 | <wanderview> | wish we could have gotten streams in for 57 |
| 15:36 | <JakeA> | Domenic: if a method returns a promise that resolves once an action completes, and the completion of that action also causes an event, which order should they be queued? https://github.com/whatwg/html/pull/1936 is promise-first, which is what I've gone with in similar code. Is that "the way"? |
| 15:49 | <annevk> | JakeA: look at fullscreen? |
| 15:49 | <annevk> | JakeA: we should abstract this pattern… |
| 15:49 | <JakeA> | annevk: aha, cheers! |
| 15:59 | <annevk> | JakeA: you generally need a task to resolve a promise so might as well dispatch at that point |
| 16:00 | <annevk> | JakeA: otherwise you get two tasks or an additional microtask |
| 16:00 | <annevk> | JakeA: I hope that is what fullscreen does anyway |
| 16:09 | <JakeA> | annevk: full screen is different as the full screen events happen as part of the render steps of the event loop, whereas the promise resolves sooner. |
| 16:09 | <JakeA> | So the promise resolves before the element is actually fullscreen, I think |
| 16:11 | <JakeA> | https://github.com/whatwg/html/pull/1936/ queues two tasks, one for the promise and one for the event. |
| 17:08 | <luxio> | is whatwg like w3c? |
| 19:25 | <leclerc> | luxio: it's what w3c hopes to be when they grow up one day. |
| 19:31 | <wanderview> | JakeA: it seems to me the ready promise should fire here... what do you think? https://sw-ready-resurrection.glitch.me/ |
| 19:33 | <wanderview> | JakeA: both chrome and firefox fail to ever resolve the ready promise there even though there is a live registration.active service worker for the current scope |
| 19:41 | <annevk> | JakeA: hmm I’ll look tomorrow then |
| 19:50 | <wanderview> | JakeA: I filed it as an issue: https://github.com/w3c/ServiceWorker/issues/1198 |
| 20:44 | <luxio> | leclerc: what would convince browser developers to follow whatwg standards instead of w3c? |
| 20:44 | <luxio> | w3c already has microsoft, google, apple, netflix, etc on their side |
| 20:44 | <luxio> | the fact that they have mozilla and google is practically enough |
| 20:52 | <TabAtkins> | luxio: leclerc was trolling. |
| 20:53 | <TabAtkins> | WHATWG is a standards organization, similar to W3C. It has some slightly different practices, and some interesting history related to the HTML spec, but is mostly interchangeable with W3C. |
| 20:54 | <TabAtkins> | When a spec exists both in WHATWG and W3C, it's almost always the case that the WHATWG version is the maintained one, and the W3C version is an outdated fork. Browser devs already generally follow the WHATWG versions of specs, for this reason. (There are arcane patent-related reasons for the W3C versions to exist, but they're still generally maintained badly, which sucks.) |
| 20:56 | <luxio> | so why's everyone going crazy about EME? |
| 21:11 | <TimothyGu> | luxio: well some people don't think it should exist in the Web platform, and some people within that group are more vocal than others |
| 21:11 | <luxio> | but if people follow whatwg rather than w3c, does w3c standardizing DRM really mean anything? |
| 21:12 | <TimothyGu> | well, TabAtkins only specified the situation when both a WHATWG version and a W3C version of the same spec exist |
| 21:12 | <TimothyGu> | EME is a W3C-only spec |
| 21:12 | <TimothyGu> | heck, we even link to W3C specs from WHATWG specs |
| 21:12 | <TimothyGu> | and vice versa too |
| 21:13 | <TimothyGu> | luxio: ^^ |
| 21:14 | <TabAtkins> | Yeah, WHATWG only maintains a dozen or so specs. Most web specs are done in the W3C. |
| 21:29 | <leclerc> | TabAtkins: I wasn't "trolling", I was "joking". What you said about W3C is quite more uncomfortable. WHATWG also has some weaknesses that make IETF and W3C a necessary source. |