2018-03-01 [21:48:45.0000] Without multiple implementations it’s not really a standard [23:36:54.0000] and competition among browser projects has been a big factor in driving innovations and improvements [23:37:17.0000] concrete example is the case of JavaScript engines [05:53:05.0000] how often do things that "need implementer interest" become actual features [05:54:21.0000] I would imagine if a feature doesn't start out with one of the browsers pushing for it it's already dead [05:59:05.0000] devsnek: There are plenty of DOM methods which were introduced after libs mentioned they would be useful for their code. [05:59:10.0000] (AFAIK.) [06:00:02.0000] I'm just noticing that these issues on the html repo go back to 2015 [06:00:21.0000] im wondering how I can make sure that doesn't happen with mine :) [06:00:39.0000] devsnek: it varies, I don't think implementers actively crawl them, so reaching out to folks can help [06:00:58.0000] Convince people or try to write an initial implementation yourself, I suppose [06:01:00.0000] devsnek: https://whatwg.org/faq#adding-new-features [06:01:11.0000] devsnek: it's a long process too; 3 years wouldn't be too bad [06:01:38.0000] yea I realize that [06:01:58.0000] 3 years sounds nice if stuff is happening [06:05:02.0000] I assumed more of the whatwg people would actually be working for the implementers [06:12:30.0000] Quite a few do, but they don't necessarily decide what to work on 2018-03-02 [05:06:56.0000] No existing tests for "pixelated"; quelle surprise [05:14:42.0000] I am working on the following issue https://github.com/w3c/web-platform-tests/issues/8616 and I found that i have to add amaplike declaration at the line highlighted in the document https://github.com/w3c/web-platform-tests/blob/master/resources/idlharness.js#L2197 .So I thought I should add a code like this https://pastebin.mozilla.org/9078942 I s it right? [05:14:42.0000] And is it necessary to be readonly? [05:14:42.0000] if it is what should be the key values I should use? [05:14:42.0000] It would be great if some one helps me. [05:14:42.0000] thank you [05:28:58.0000] shannu: I think it will need a lot more code than that. You can look at https://www.npmjs.com/package/webidl2#iterable--legacyiterable--maplike--setlike-declarations to see how the parser represents setlike and maplike types, and then work out how to test those interfaces. [05:29:39.0000] ricea: Thank you [05:32:18.0000] shannu: It might be easier to start by looking for interfaces that are implicitly iterable as described in that issue, and then call this.add_iterable_members on them. [05:34:05.0000] Ok [06:16:17.0000] https://drafts.csswg.org/css-lists/#counter-functions [06:16:36.0000] Isn't it a bit weird for the first argument of counter() to be ? Does it make sense to be able to name a counter 'inherit'? [06:18:59.0000] AFAIK counter-increment and friends want a , [06:19:06.0000] so you can't actually use a counter named 'inherit' in them. [06:31:39.0000] JakeA: so http://software.hixie.ch/utilities/js/live-dom-viewer/saved/5807 Chrome doesn't seem to resize the SVG [06:31:53.0000] JakeA: Safari does resize it, but cannot draw it (or ends up drawing nothing) [06:32:18.0000] JakeA: unclear where it fails in Firefox [06:35:26.0000] annevk: ugh, yeah. It should absolutely be 100x100 afterwards. I guess this stuff is relatively new. [06:36:32.0000] JakeA: if I remove 1,1,1,1 btw nobody still draws anything [06:37:02.0000] annevk: does it work if you give the svg an explicit width & height? [06:37:50.0000] JakeA: no, but that has made me think whether that's required here [06:38:29.0000] JakeA: can you crop when there's no intrinsic dimensions, I guess you should be able to since for SVG we resize first and then crop, but we only specced that in the PR [06:38:54.0000] annevk: I thought took an intrinsic height from the viewbox [06:39:26.0000] JakeA: this differs between Chrome and Safari too, that's why I added an explicit width/height [06:39:41.0000] JakeA: I think Safari might actually do that correctly, where it takes up the maximum width possible [06:39:50.0000] JakeA: per replaced element sizing in CSS 2.1 [06:40:19.0000] I can't believe even that is still non-interoperable in 2018 [06:40:44.0000] SVG has no love from browsers [06:41:00.0000] It's probably the biggest divide between developer interest and browser interest [06:41:38.0000] Tell foolip [06:42:32.0000] I'm hoping we can improve SVG a bit by leaning on the things that are common between CSS, , and SVG, and give them common underpinnings [06:42:52.0000] But that's only part of the issue I suppose [06:43:12.0000] You know, sometimes I don't mind that moratorium idea so much [06:43:18.0000] There's a lot of technical debt [06:43:20.0000] A lot [06:43:47.0000] :D [06:44:22.0000] Yeah, I'd like SVG to stop being a special thing. Like, maybe allow etc outside svg [06:44:44.0000] has 's d="" attribute [06:44:51.0000] And make some kind of separate viewboxing thing. [06:45:10.0000] But *what working group* would spec that? [06:45:26.0000] heh [06:45:42.0000] (Minor nit: emphasis solely goes on what, iirc.) [07:02:51.0000] https://github.com/w3c/csswg-drafts/pull/2377 [07:03:09.0000] JakeA: We would love you if you implemented SVG in Servo. :P [07:04:11.0000] I'm the ideas guy not the science guy :D [07:04:57.0000] Ah ah. [07:05:05.0000] nox: Is the feeling that SVG is just too complicated? Because I don't think developers would be against something else a little more GPU-aligned. [07:05:22.0000] JakeA: Nah, the feeling is that days are still only 24 hours long even for Servo devs. :) [07:07:02.0000] nox: super excited about servo. Great seeing bits of it going into Firefox too. [07:07:11.0000] Thanks! [07:12:20.0000] annevk, http://ln.hixie.ch/?start=1086387609&count=1 appears to back me :) [07:21:25.0000] Ah I see I came to the right place for an argument about Hixie's use of typography 14 years ago [07:23:54.0000] nox: could you open an issue/PR on counter()/? [07:25:22.0000] astearns: I don't understand what that means. [07:25:26.0000] I did a PR. [07:25:37.0000] oh sorry, I didn't check [07:25:39.0000] I linked to it earlier. https://github.com/w3c/csswg-drafts/pull/2377 [07:25:53.0000] astearns: No problem. :) [07:26:40.0000] thanks for the PR - now I'll go bug TabAtkins :) [07:27:14.0000] astearns: Cool. Forgot to mention that the test linked in that PR already passes in Safari and Chrome, [07:27:26.0000] even better [07:27:34.0000] astearns: and I suspect it was passing too in pre-Stylo Firefox, but I'm too lazy to actually confirm that. [07:27:45.0000] ecobos may know how to check that easily. [07:28:30.0000] nox: danke [07:28:34.0000] nox: layout.css.servo.enabled=false should still work, though not for long [07:28:59.0000] ecobos: Oh right, somehow I almost went and read nsCSSParser.cpp, [07:29:01.0000] thanks for saving me. [07:29:30.0000] ecobos: Do I need to restart Firefox? [07:29:40.0000] nox: nah, just reload [07:29:51.0000] ecobos: It was already broken, then. [07:30:31.0000] TabAtkins: while you're around, what about https://github.com/w3c/csswg-drafts/pull/2168? It's trivial and has been sitting around for months. Blink & Firefox don't implement the special behavior of display: contents for `` because well, it's no longer a special element in any way. [07:30:33.0000] TabAtkins: You're welcome, thanks for the fast merge. :) [07:31:56.0000] ecobos: huh, so applet just didn't do anything any more? [07:32:09.0000] TabAtkins: nope [07:32:32.0000] Cool, merged [07:32:41.0000] TabAtkins: thanks! [07:32:51.0000] TabAtkins: (and sorry for nagging :P) [07:33:13.0000] Nah, is necessary, thanks [07:33:28.0000] I don't keep up with the PRs like I should [07:33:55.0000] TabAtkins: you and fantasai do a great work, I don't know how you keep up tbh [07:34:35.0000] The secret is we don't, but we're still productive enough that it's not too noticeable [07:35:11.0000] hah [07:45:35.0000] Ms2ger: touché [07:48:32.0000] TabAtkins: can I prod you to review https://github.com/w3c/web-platform-tests/pull/8822? [10:29:11.0000] if there's no objection I can submit a pull request for https://github.com/whatwg/whatwg.org/issues/187 [10:30:02.0000] mobile already has consistent padding from what I've seen, this change only affects the desktop version [10:35:29.0000] railgun: seems fine to me, it'd be great to have before after screenshots in the PR if not too much trouble [10:35:33.0000] railgun: makes reviewing easier [10:35:41.0000] Domenic: ^^ [10:36:44.0000] I looked in to this at some point and found weird things where we had some padding but it got overridden or something. Good to fix, but check the nearby code to check if there might be leftovers or weird things. [10:44:35.0000] will do (as a side note using a css preprocessor with variables / mixins for padding may have prevented this issue from popping up in the first place) [10:45:20.0000] railgun: a lot of this is over a decade old [10:45:35.0000] railgun: we could maybe add some CSS variables here and there, now we started using them on the website [11:17:01.0000] heh it's been a while [11:17:15.0000] I remember when you couldn't use css variables since not all browsers supported it [11:17:39.0000] now it seems pretty universal https://caniuse.com/#feat=css-variables [11:20:39.0000] actually looking back at it there was little to no browser support when I learned about css variables 2018-03-03 [00:08:06.0000] TabAtkins: I'm not gonna grant private access for that Octobox thing [00:08:27.0000] TabAtkins: if that's required please start a whatwg/meta discussion about [00:34:06.0000] I see that /css/css-counter-styles/README points to https://www.w3.org/International/tests/ but it seems many tests from there are missing from wpt. Is that known and ntended? [00:34:08.0000] intended* [01:29:43.0000] nox: seems like a bug [01:30:04.0000] annevk: And can I add tests in the root there? [01:32:05.0000] nox: dunno [01:32:16.0000] nox: I hear /css is weird [03:16:02.0000] nox: gsnedders is the best person to ask [04:00:55.0000] I am working on the following issue https://github.com/w3c/web-platform-tests/issues/8616 and I found that i have to add amaplike declaration at the line highlighted in the document https://github.com/w3c/web-platform-tests/blob/master/resources/idlharness.js#L2197 . Can I know what is use of mapike? and how to use that in an interface? [04:00:55.0000] It would be great if some one helps me. [04:00:55.0000] thank you [04:11:21.0000] shannu: I think https://drafts.css-houdini.org/css-typed-om/ uses it for some things, but generally I don't think it has much adoption thus far [04:12:37.0000] annevk: Can I know what all things i should include in the isuue above? [09:05:22.0000] is the weird spacing here intentional? this occurs in other sections as well iirc https://i.imgur.com/72wnp7o.jpg [09:12:57.0000] I also think that note sections should always be on their own separate line, see https://i.imgur.com/s0E3ouT.jpg [09:13:33.0000] taken from https://dom.spec.whatwg.org/#document-trees [09:16:25.0000] annevk, shannu: Not using maplike anymore in typed om. But there is a setlike in [09:18:03.0000] Ok thank you [09:19:25.0000] TabAtkins: could you help me in writing maplike in the above issues? [09:19:40.0000] Or setlike? [09:19:46.0000] I can't, sorry. I don't know anything about the idlharness code. [09:44:24.0000] Can I know where or how I could get help? 2018-03-04 [23:34:24.0000] railgun: it's intentional that some notes are inline and some are their own paragraphs and some are their own sections. [23:34:46.0000] I'm not sure what's weird about the spacing in the first screenshot. [01:01:38.0000] Regarding the issue, I am working on, I am required to add a maplike and setlike handling inside a switch. However I am bit confused regarding the type of the arguments required. Can someone please assist me? 🙂 [04:04:54.0000] shannu: maybe ping @jgraham and @tobie in the issue? (Make sure to give enough context around the problem.) [04:38:50.0000] annevk: when i pinged @jgraham i was lead to this channel [05:28:51.0000] Domenic: yeah just realized that span.note has its own css (inline notes look a bit weird though since the note part gets cut off) [05:29:08.0000] Oh heh I just noticed the cutoff, fascinating. [05:29:18.0000] Maybe that should un-rotate [05:29:27.0000] (We talked about un-rotating all of them, actually.) [05:30:14.0000] are there any set rules for when to use each? (inline, paragraph, section) [05:30:27.0000] hmm [05:31:49.0000] I think it'd be easier just to not use inline since it looks like they cause more problems than solve [05:32:45.0000] this is bothering me https://i.imgur.com/KmlGtbU.jpg [05:32:59.0000] I mean, it's like English. Sometimes you say "Note, X" without starting a whole new paragraph. [05:33:59.0000] maybe the style could be different? [05:34:30.0000] inline notes aren't as sectioned out as other notes [05:34:49.0000] Suggestions always welcome, although we'd want them to be visually connected to whole-paragraph notes. [05:36:49.0000] whatwg specs don't feel the same with rotate removed though [05:38:03.0000] I'm probably biasedd [05:38:15.0000] biased* [05:39:26.0000] :) [05:50:30.0000] seems like the csswg uses inline notes as well https://i.imgur.com/3pyDRxS.jpg [06:03:03.0000] here is one suggestion for inline notes https://i.imgur.com/1q10YrM.jpg (original https://i.imgur.com/s0E3ouT.jpg) [06:06:55.0000] (could also bold note to separate it from the content) [06:44:50.0000] Domenic: from your review I get Error: missing for topic "attr-form-method" even after a rebase [07:51:09.0000] Domenic: I see that Takeshi's name on Chromium-Review now has status "Leaves the project on 2018/03/02", is he no longer working on Chrome? [14:42:59.0000] railgun: maybe that isn't the exact ID [14:43:20.0000] railgun: That inline note doesn't feel similar enough to other notes, IMO. [14:43:34.0000] domfarolino: sadly yes :( [15:07:42.0000] Domenic: Oh wow, what’s he moving on to? [15:22:15.0000] domfarolino: a smaller company 2018-03-05 [17:34:39.0000] railgun: https://github.com/whatwg/html/pull/3484/files doesn't have the x/y updates? [17:35:04.0000] I guess no changes are uploaded yet :) [18:27:24.0000] god the source of http://oldnavy.gap.com/ is a freak show [18:27:36.0000] and no https [18:27:59.0000] and same for all *.gap.com sites [18:33:50.0000] MikeSmith, they're not bankrupt yet? [18:34:26.0000] what is with all those line feeds? [18:35:41.0000] at least the signon page is https [18:35:47.0000] a-ja: I guess they’re staving off bankruptcy by having their CFO write all the code for their sites [18:35:56.0000] heh [18:37:28.0000] astearns: I’m guessing those line feeds are artifacts of some servere-side templating/pre-processing parts left over after the other bits got parsed/stripped out? [01:51:47.0000] Maybe we should have some sort of constitution-like doc where we could write: No changes to the HTML parsing algorithm anymore. [04:23:50.0000] hsivonen: :) [05:59:53.0000] I mean, you could definitely write that on your blog, but there's no cross-browser consensus on it. 2018-03-06 [19:43:54.0000] annevk: review/rubber stamp https://github.com/whatwg/html/pull/3472 ? [21:34:58.0000] Domenic: test commit says "first batch"? [21:35:19.0000] annevk: yeah, I think good enough for spec merge. Most of the rest of the tests will be manual. [21:35:26.0000] Ooh that's a big change, I'm not gonna get to that today I think :/ [21:35:42.0000] Fair enough, I guess we didn't get much editorial review from the other reviewers. [21:35:52.0000] Each commit is pretty reasonable FWIW. [12:38:43.0000] TabAtkins: When should I use `...` over `{{...}}` (or vice-versa) when linking to something that is defined in `
` as type `constructor`?

[12:39:00.0000] 
My understanding is that that both will link to IDL types (constructor being one)

[12:40:00.0000] 
Yeah, there's really no reason to ever use  when linking to any IDL type.

[12:42:23.0000] 
TabAtkins: Sorry, I was trying to finish my question but then a professor walked in ...it seems that they produce different styles though, is `{{...}}` just generally preferred though, and `` is just used for style-less linking?

[12:43:25.0000] 
{{...}} happens to wrap itself in a  in the output, that's the styling difference you're seeing.

[12:44:05.0000] 
(Or maybe it just adds a class=idl-code, which the default stylesheet targets. I forget.)

[12:45:14.0000] 
TabAtkins: Ah, iirc I think it does wrap in a . Ok thanks, didn't know if it is preferred to use one over the other, but I guess that's context-dependent

[12:45:15.0000] 
 is just the generic, always-works-for-all-types linking syntax.  If there's a shorthand, it's usually better to use it.  (And if you need to specify something that the shorthand can't accommodate, like the `spec` attribute, you should use {{something}}, rather than breaking it back down fully into an .)

[12:50:25.0000] 
Ok that makes sense, thank you

[13:12:19.0000] 
I've recently updated the clipboardData that Chrome sets when a user copies an image.  There's some inconsistency between browsers right now regarding what types (html/text/png) are copied, and also inconsistency between paste events and drop events (in dataTransfer).  who knows this area and could answer a few more questions of mine?

[13:17:46.0000] 
Domenic: Can we get away with `Call`ing the %String% constructor in https://console.spec.whatwg.org/#formatter as opposed to `Construct`ing from it?

[13:18:11.0000] 
Seems like there's no good reason to call Construct over Call in that case (since newTarget is irrelevant)

[13:53:22.0000] 
annevk: Are there wpts for https://github.com/whatwg/fetch/pull/585#issuecomment-329121865 ?

[15:07:25.0000] 
domfarolino: definitely Call; Construct would incorrectly create a wrapper.

[15:11:58.0000] 
Domenic: Ok, could you possibly explain how this wrapper gets created? I'm just curious, reading https://tc39.github.io/ecma262/#sec-ecmascript-function-objects-construct-argumentslist-newtarget

[15:15:29.0000] 
domfarolino: Construct sets NewTarget. From there, check the definition of the String constructor.

[15:15:51.0000] 
Domenic: Ah, I'm assuming it is because PrepareForOrdinaryCall uses newTarget instead of undefined, in [[Construct]]

[15:16:04.0000] 
Ok I'll check that out

[15:17:45.0000] 
Oh ok, thanks


2018-03-07
[16:18:15.0000] 
domfarolino: don’t think so, I think the main problem is updating existing tests;

[16:18:42.0000] 
domfarolino: because of that it would be useful to have a browser with the fix

[16:19:13.0000] 
domfarolino: but maybe affected tests can be located with a couple greps

[17:07:41.0000] 
annevk: Ah that makes sense. I can take a look sometime this week assuming a patch to change default credentials behavior is fairly trivial

[17:08:27.0000] 
domfarolino: cool

[18:09:03.0000] 
JakeA: I suspect that for https://github.com/whatwg/fetch/issues/145 you'll have to find a new networking person given that tyoshino left

[20:16:14.0000] 
Domenic: happy to help figure out things for this API

[20:16:20.0000] 
\o/

[20:16:21.0000] 
Domenic: imperative slotting, that is

[20:16:42.0000] 
Domenic: I'm mainly wondering what developers would want to happen in various places

[20:17:07.0000] 
Ah yeah. I was kind of hoping they would never mix styles, but developers sometimes surprise us.

[20:17:21.0000] 
Domenic: e.g., if you modify the children of a host element, is it fine that everything gets reset?

[20:17:59.0000] 
Domenic: should this allow elements that are not in a tree? (what if they get removed?)

[20:18:18.0000] 
Domenic: not connected I mean (everything is in a tree)

[20:18:22.0000] 
Interesting, why would it get reset?

[20:19:23.0000] 
Domenic: that's how we redistribute; if something changes we assign things anew

[20:19:41.0000] 
Domenic: e.g., step 10 of https://dom.spec.whatwg.org/#concept-node-remove

[20:19:42.0000] 
Maybe I am misunderstanding what you mean by "host element"?

[20:19:54.0000] 
Right OK, an assigned element

[20:20:26.0000] 
Not the whole shadow host

[20:20:35.0000] 
Hmm, I guess it's not quite connected either, but we require it to be a child of a host element atm

[20:21:32.0000] 
Blah I should refresh my memory of this stuff

[20:21:40.0000] 
It'd be nice to get some examples into the spec

[20:21:45.0000] 
Some are trapped in the old spec

[20:23:11.0000] 
FWIW, I guess rniwa probably knew about this when he proposed this

[20:23:33.0000] 
So I suspect his stance here is that if you mutate the tree you'll have to deal with the consequences

[20:23:49.0000] 
I am hopeful that is fine with developers

[20:23:51.0000] 
But would be good to make that clear with some examples around the API maybe

[20:23:55.0000] 
Yeah exactly

[20:35:12.0000] 
Domenic: I think the main question we should answer is what happens if they are already assigned, and how do we update those other slot elements

[20:35:23.0000] 
Domenic: presumably we do want to fire slotchange at least

[23:11:40.0000] 
Domenic: it'd be more productive if you proposed something you have in mind in that issue

[23:12:05.0000] 
annevk: I am not in favor of the feature in general, so I am not inclined to spend time on developing it. I am happy to give my constraints though.

[23:12:15.0000] 
Domenic: it just came to mind that built-ins don't have is="", but do have namespaces to consider, so their registry is different anyway

[23:13:19.0000] 
Domenic: it just seems so weird to suddenly insist built-ins be treated the same while there's no such consistency enforced elsewhere

[23:13:31.0000] 
Domenic: it appears you're just being arbitrary to stall things

[23:13:45.0000] 
The problem is the same for both: the browser has a registry, and some frameworks can't deal with just constructors, so need some sort of side-table.

[23:14:26.0000] 
I don't think it's arbitrary to insist we expose the primitives.

[23:14:43.0000] 
But as I've shown it's not the same primitive and it doesn't live in the same place

[23:15:00.0000] 
The primitive for custom elements lives on the custom element registry

[23:15:10.0000] 
The other primitive exists, but it's not layered in the same way at all

[23:15:20.0000] 
So insisting on the same API or solving them at the same time is arbitrary

[23:16:17.0000] 
We also don't have something.get(localname, namespace) that returns a constructor for built-ins

[23:16:35.0000] 
Why didn't you add that? Where is the consistency?

[23:19:58.0000] 
Indeed, I'm saying we should add that

[23:20:06.0000] 
The problem space needs to be considered holistically.

[23:20:18.0000] 
If we're to address these frameworks which can't cope

[23:20:22.0000] 
But you haven't done that

[23:20:30.0000] 
Indeed, I have not. I've only solved the important use cases so far.

[23:20:45.0000] 
Right, what's the problem with continuing to do so?

[23:21:09.0000] 
The problem posed in the OP is the same for built-in elements and custom ones

[23:21:09.0000] 
I've looked at it holistically and I've outlined how all this could work, but we don't need to add all at once

[23:21:18.0000] 
So we should not solve only half the problem

[23:23:56.0000] 
Right and as pointed out in the meeting and even by that person, it's not a problem for built-ins (also as shown by other issues)

[23:24:12.0000] 
I feel like we're going in circles...

[23:24:39.0000] 
I don't agree it's not a problem for built-ins. For built-ins people keep a large table.

[23:24:45.0000] 
They can do the same for custom elements

[23:24:47.0000] 
But they don't want to

[23:27:00.0000] 
Domenic: okay, I guess I can try to solve this more generically; it is a little weird that ElementRegistry will have a different API shape from CustomElementRegistry

[09:12:14.0000] 
can whatwg definitions be used in the mdn glossary? in particular "insertion point" is a dead link but the whatwg definition is here https://html.spec.whatwg.org/multipage/parsing.html#insertion-point

[09:15:59.0000] 
mdn uses "(CC-BY-SA), v2.5 or any later version" so that should be compatible with CC-BY-4.0 (if they didn't say "or any later version" it wouldn't be compatible?)


2018-03-08
[19:41:19.0000] 
Domenic: FWIW, I think https://github.com/w3c/webcomponents/issues/468#issuecomment-370672895 makes it a little more complicated and definitely somewhat related to shadow root infrastructure

[19:41:56.0000] 
annevk: I think it'll be that way in the spec. I'm not sure it'll be observable to web devs.

[19:42:05.0000] 
I guess :hsot

[19:42:07.0000] 
:host

[19:42:31.0000] 
Yeah, which is nice to have here

[19:42:50.0000] 
I'll edit to add "with an interpretation of :host as the custom element being customized"

[22:35:01.0000] 
kochi: seems style sheets for built-ins needs a more concrete proposal

[22:35:08.0000] 
kochi: for custom elements*

[22:35:44.0000] 
annevk: for some time we can ignore for built-ins?

[22:36:30.0000] 
kochi: sorry, I meant a more concrete proposal for https://github.com/w3c/webcomponents/issues/468

[22:36:42.0000] 
kochi: and how it works together :host et al

[22:38:50.0000] 
annevk: sure, for speccing, but I don't know there is real use case for styling *some* children of a custom element without shadow root, like my example.

[22:41:26.0000] 
kochi: built-ins can do it

[22:42:23.0000] 
annevk: (looking at ua stylesheet...) I found 'select[disabled]>option { color: GrayText }', yeah

[23:24:23.0000] 
Seems like the Company Type is wrong at https://www.crunchbase.com/organization/whatwg lol. I’ve never heard of crunchbase tho

[23:24:55.0000] 
annevk: actually I’m curious, where does WHATWG get funds for hosting and domains etc

[23:25:56.0000] 
for profit of the world

[23:26:22.0000] 
👍

[23:26:51.0000] 
Whenever I try searching something that I expect it would end up in an MDN page, it goes to Japanese translated version of the wiki (due to ranking of search according to my language preference), then I switch the locale of the page, MDN asks me "Do you want to view this site with en-US"?, so I answer "Yes," but the search results keeps returning Japanese page, and MDN doesn't redirect :(

[23:30:24.0000] 
crunchbase seems to be a database of tech companies run by TechCrunch?

[23:30:40.0000] 
domfarolino: Ian used to pay it out of pocket, now some of us pay for it and some is paid for by the host

[23:31:10.0000] 
kochi: hmm maybe file a bug against mdn?

[23:32:15.0000] 
kochi: not sure where though, maybe https://github.com/mdn/mdn

[23:32:16.0000] 
annevk: I don't know it's expected or not - if I directly follow the URL pointing to a Japanese page, MDN is expected to redirect according to the cookie-set preferred locale?

[23:32:39.0000] 
kochi: dunno, but maybe they're open to enabling that

[23:37:00.0000] 
Ah yeah correct. Didn’t do too much digging around on their mobile site.

[23:37:22.0000] 
annevk: ok cool, thanks

[23:41:13.0000] 
annevk: https://github.com/mdn/mdn/issues/25 filed anyway

[00:45:07.0000] 
annevk: I'm a bit confused on dependencies of relatedTarget changes.  Which one I should look at first?

[00:46:24.0000] 
kochi: studying just https://github.com/whatwg/dom/pull/585 is probably fine, maybe I should close the other one since I'm not sure addressing these issues separately is worth the effort

[00:47:39.0000] 
kochi: added a clarifying comment to the earlier PR

[00:48:02.0000] 
annevk: thanks, so dom#585 is the spec change and wpt#9919 is the corresponding test change, am I correct?

[00:49:42.0000] 
kochi: yeah, though I'll need to write more tests (or someone else should)

[00:50:07.0000] 
kochi: the tests only cover a small part so far

[01:51:24.0000] 
annevk: can you tell me why we need to copy initiator and destination for CORS preflight fetch?

[01:52:25.0000] 
annevk: https://fetch.spec.whatwg.org/#cors-preflight-fetch

[03:54:42.0000] 
yhirano__: don’t recall offhand

[08:45:50.0000] 
kochi: annevk: the front-end for mdn is actually managed in this repository https://github.com/mozilla/kuma and issues are reported via bugzilla

[08:50:34.0000] 
seems like this is a duplicate of https://bugzilla.mozilla.org/show_bug.cgi?id=926963


2018-03-09
[16:04:02.0000] 
ah, thanks

[18:16:28.0000] 
railgun: thanks, i'll follow those bugs in the bugzilla.

[18:32:17.0000] 
is the loader spec just waiting on realms to finalise now?

[18:33:36.0000] 
devsnek: I don’t think loader will happen as-is; need something that builds on service workers instead

[18:33:47.0000] 
yea thats what the warning says

[18:34:29.0000] 
all the workers are fully spec'd now, realms are still stage 1 so i assumed thats why the spec isn't advancing

[18:35:13.0000] 
Don’t think realms matter here, but dunno

[18:35:58.0000] 
i don't know much about any of this hehe, thats why i came by to ask :)

[19:34:28.0000] 
annevk: they look unnecessary to me. if you agree i'd like to remove the copying steps.

[19:47:35.0000] 
TIL, xhr terminates an ongoing xhr when opening the next ; https://xhr.spec.whatwg.org/#the-open()-method

[19:47:39.0000] 
interesting

[19:47:57.0000] 
i guess due to some of the fields being static?

[19:50:51.0000] 
or.. this is only about the same instance being reused, right. I don't even...

[20:28:15.0000] 
Don't use XHR :)

[21:20:39.0000] 
^


2018-03-12
[00:08:43.0000] 
annevk: If you click "Details" in the PR's participation check you'll get to https://participate.whatwg.org/agreement-status?user=yoavweiss&repo=fetch which is a real-time view with a form at the bottom for updating GitHub. It indicates Yoav is still not quite set up; my guess is his org membership is not public.

[00:09:03.0000] 
See also https://github.com/whatwg/participate.whatwg.org/issues/21 for reducing one step

[00:09:24.0000] 
checking on my end

[00:10:14.0000] 
yup, was private

[00:10:22.0000] 
should be good now

[00:11:11.0000] 
Perfect. I'll do the tell-GitHub-about it step, although anyone can do it by going to that URL and pasting https://github.com/whatwg/fetch/pull/659 in the box

[00:11:57.0000] 
Domenic: thanks

[00:12:12.0000] 
All green now

[00:17:54.0000] 
yoav: all I meant with follow-up issue was filing one, no need to have it solved straight away

[00:59:55.0000] 
annevk: oh, OK :) I'll file one

[01:00:31.0000] 
/me doesn't yet have an opinion on imperative API. Gut feeling is that non-direct-child-nodes shouldn't be supported, but not sure.

[01:06:46.0000] 
I guess we made some progress on it in the sense that we have an updated version of the problem space

[01:18:25.0000] 
annevk: so, basically you want me to file a Fetch equivalent of https://github.com/w3c/webappsec-csp/issues/282?

[01:23:45.0000] 
yoav: yeah I suppose

[01:24:02.0000] 
yoav: I wasn't aware we already track it there, but tracking it on both sides seems good

[01:24:22.0000] 
cool, will do

[01:29:04.0000] 
annevk: https://github.com/whatwg/fetch/issues/683

[02:29:03.0000] 
yoav: I think the only thing remaining is impl bugs

[02:29:13.0000] 
yoav: they weren't filed as part of the CSP change it seems

[02:29:34.0000] 
mkwst: maybe make that a policy? To file impl bugs for WebAppSec normative spec changes?

[04:27:52.0000] 
smaug____: I'd appreciate review of https://github.com/whatwg/dom/pull/585

[04:30:41.0000] 
annevk: let me have lunch first

[04:37:23.0000] 
Hmm, Chrome exposes Touch / TouchEvent on desktop, but nobody else does?

[04:37:30.0000] 
foolip: known interop issue? ^^

[04:37:50.0000] 
Not sure how to test relatedTargets then...

[06:21:59.0000] 
annevk: so is the diff in the first comment of https://github.com/whatwg/dom/pull/585 the right one to look at?

[06:22:45.0000] 
(it isn't clear to me from the UI what the 'diff' refers to)

[06:35:44.0000] 
smaug____: yes

[06:36:13.0000] 
smaug____: it’s a human readable diff of the whole PR

[08:47:40.0000] 
Anyone around who can explain why a ServiceWorkerRegistration needs {installing,waiting,active} attributes? Why is one not enough?

[08:48:56.0000] 
Ms2ger: when you install a new one, you might also have a different active one serving content at the moment

[09:16:46.0000] 
annevk: that sounds familiar, dtapuska would know the background


2018-03-13
[21:45:34.0000] 
dbaron: https://github.com/mozilla/standards-positions/blob/master/CONTRIBUTING.md#which-specifications seems a little exclusionary :)

[21:51:23.0000] 
Domenic: created https://github.com/mozilla/standards-positions/pull/71

[21:51:44.0000] 
\o/

[21:52:12.0000] 
Pretty sure it's an oversight since we list them elsewhere

[21:52:32.0000] 
I hope at some point we'll also include Khronos, Unicode, CA/B, et al

[21:52:44.0000] 
Yeah fair point

[21:53:03.0000] 
Although not sure how many specs the CA/B produces

[21:54:40.0000] 
Well, they have a set of guidelines CAs need to adhere to that gets change proposals

[21:54:59.0000] 
So they produce at least one "Living Standard" of sorts that I know of

[21:55:00.0000] 
Ah yeah, that fits

[21:59:11.0000] 
annevk: if you're working on DOM bugs today https://github.com/whatwg/dom/issues/371 still really bugs me

[22:00:58.0000] 
Domenic: I'd like all the shadow stuff to land first

[22:01:16.0000] 
Domenic: also, it's starting to get to me a bit that I'm always cleaning up other folks their trash

[22:02:20.0000] 
:-/

[22:02:48.0000] 
For shadow DOM that's an issue with how the incubation proceeded I think

[22:03:00.0000] 
Not sure what happened with the service worker thing, seems like it needed more review going in?

[22:04:26.0000] 
Yeah, it didn't have enough architectural scrutiny and it took a long time for the lifetime to become somewhat defined

[22:05:01.0000] 
jungkees is doing a lot of the cleanup there, which is great :)

[22:05:19.0000] 
Coupled with WPT upstreaming

[22:06:41.0000] 
Yeah, that certainly helps

[23:01:15.0000] 
Maybe I can tackle that a bit as it does not touch dispatch

[23:23:28.0000] 
Domenic: merged

[23:42:33.0000] 
oops, I'm not used to using GitHub review tool for making comments...

[00:25:47.0000] 
TabAtkins: https://github.com/whatwg/dom/issues/185

[01:24:41.0000] 
Domenic: JakeA: to what extent do we still want https://github.com/whatwg/dom/issues/438?

[01:25:18.0000] 
Domenic: JakeA: maybe we should file a dedicated issue on serialize/transfer for AbortController and close that one?

[01:27:05.0000] 
Hmm I can't remember why I wanted that :/

[01:40:45.0000] 
That seems like a good reason to close it 😃

[01:50:03.0000] 
annevk: Domenic: It's interesting, but transferring promises/streams would be higher on my list

[01:50:35.0000] 
Yeah it seems like transferring streams is coming up soon

[02:05:10.0000] 
Domenic: what's the solution for deciding which objects should be transferred, and which should be cloned?

[02:05:17.0000] 
That's the open question

[02:05:30.0000] 
Although at least for readable byte streams we can just easily decide they're all transferred

[02:05:35.0000] 
yeah

[02:06:11.0000] 
controller.enqueue(whatever, transferables) is the best I can think of

[02:06:39.0000] 
For tasklets surma / iank___ contemplated just always transferring all the transferrable things

[02:08:10.0000] 
Sure. Might be weird that it's different to postMessage though. Or stream.pipeThrough(new TransferAll())

[02:08:56.0000] 
Which could be explained using controller.enqueue(whatever, transferrables)

[02:09:36.0000] 
Yeah c.e(w, t) seems like probably the right primitive? But maybe it should be easier?

[02:09:52.0000] 
I was thinking new ReadableStream({ autoTransfer: true, ... }) maybe

[02:10:38.0000] 
The pipeThrough idea works but feels a bit strange at the moment, hmm

[02:10:43.0000] 
I guess it depends on who we want to be in control

[02:11:53.0000] 
const stream = new ReadableStream({ autoTransfer: true, … }).pipeThrough(new TransformStream()); // what happens if I transfer stream? Has the transform stream unmarked the objects as transferrable?

[02:12:18.0000] 
hmm

[02:14:02.0000] 
Hmm yeah

[02:14:13.0000] 
There are several levels of transferrability here, I guess autoTransfer is not the interesting one

[02:15:07.0000] 
Hmm

[02:15:27.0000] 
Maybe x.postMessage(stream, [ReadableStream.transferChunks])???

[02:16:13.0000] 
Kinda magic

[02:16:26.0000] 
Orrr even just x.postMessage(stream, []) vs. x.postMessage(stream, [stream])

[02:16:43.0000] 
I guess the question is, what does the first one do.

[02:16:53.0000] 
Does it tee stream or does it lock it forever and clone the chunks over

[02:17:15.0000] 
s/tee/clone I guess, which we still haven't really done

[02:17:16.0000] 
I figured it'd tee, or throw

[02:17:57.0000] 
yeah I guess cloning is the right thing

[02:18:34.0000] 
But I think the cloning/transferring of the stream is separate to the chunks

[02:19:07.0000] 
I was kinda hoping it wasn't, then we could just use that distinction

[02:19:11.0000] 
At least a transform stream lets you do it on a chunk by chunk basis, but having to make sure it's the very last transform in the sequence might be a gotcha

[02:20:03.0000] 
I'm no longer sure how that would work, except by some kind of magical marking of chunks as transferrable

[02:20:11.0000] 
Because the transfer-or-not decision needs to be made inside the postMessage code

[02:20:54.0000] 
Yeah I guess the stream would have a weakmap of chunks to transferrables

[02:21:02.0000] 
but yeah, I see your point

[02:21:28.0000] 
Auto-transferring seems nice...

[02:21:34.0000] 
We could also add an options bag to all the postMessage APIs...

[02:22:20.0000] 
obj[Symbol.transferrables] 😈

[02:26:17.0000] 
Let the games begin https://twitter.com/domenic/status/973490376731328513

[02:31:55.0000] 
annevk: Around?

[02:34:39.0000] 
nox: yeah

[02:35:04.0000] 
annevk: https://github.com/immersive-web/webxr/issues/25#issuecomment-372602299

[02:35:21.0000] 
annevk: If the answer is kill, I'm removing the test from a Servo PR and it should land in WPT soon afterwards.

[02:35:31.0000] 
annevk: It's just mentioned in WebVR's idlharness tests.

[02:36:26.0000] 
nox: sorry, don't know about webxr

[02:36:33.0000] 
Fair enough.

[02:36:48.0000] 
nox: it does seem that if allowvr is implemented that needs upstreaming to HTML /  Feature Policy

[02:37:05.0000] 
annevk: 'allow="vr"' is a thing,

[02:37:14.0000] 
so probably just something that we forgot to remove.

[02:37:14.0000] 
nox: that's feature policy

[02:37:29.0000] 
nox: allow= is here to stay

[02:37:38.0000] 
Yeah that I know.

[02:38:18.0000] 
Domenic: hmm, how do you count?

[02:38:26.0000] 
Domenic: are Worker and MessageChannel separate?

[02:38:43.0000] 
annevk: IDL method declarations

[02:40:27.0000] 
Can I kill an arrow function in a test?

[02:40:33.0000] 
Servo still doesn't support them.

[02:40:43.0000] 
nox: I'd rather you not

[02:40:59.0000] 
Well I guess we won't run those tests then. What could go wrong?

[02:41:03.0000] 
nox: I land stuff with them all the time and they're actually more readable

[02:41:21.0000] 
I don't see how "more readable" enters the picture here.

[02:41:36.0000] 
Now I need to maintain a separate fork of tests.

[02:41:43.0000] 
nox: in that I likely have to read that test again in the future

[02:42:03.0000] 
nox: there's a root problem here and the arrow functions are not it

[02:42:50.0000] 
annevk: Ok.

[02:43:31.0000] 
Domenic: I wonder if that's really the upper bound, given AudioWorklet

[02:43:43.0000] 
annevk: I checked, AudioWorklet uses MessagePort

[02:44:31.0000] 
Servo should just give up and add Babel at this point

[02:44:33.0000] 
Domenic: hurray

[02:44:45.0000] 
No, we should get our SM bump,

[02:44:55.0000] 
but SM's build system is… Well it exists.

[02:45:07.0000] 
May have brainfarted anyway, I think the arrow function itself is borken.

[02:45:36.0000] 
Is `r => r.text()` supposed to work?

[02:45:41.0000] 
  │   → invalid arrow-function arguments (parentheses around the arrow-function may help)

[02:46:08.0000] 
yeah it should...

[02:46:15.0000] 
Need to be more ambitious, re-introduce IronMonkey but now with the meaning of being a JS impl in Rust

[02:46:43.0000] 
So we are on a SM version that choke on some arrow functions or something, wonderful.

[02:47:07.0000] 
That seems really weird

[02:48:30.0000] 
annevk: Oh no. The expression on which it chokes is in `async`,

[02:48:41.0000] 
which is probably what we are actually missing.

[02:53:14.0000] 
That makes more sense

[03:02:19.0000] 
annevk: Is -manual a special test name suffix?

[03:04:30.0000] 
nox: yeah, it means it cannot be automated

[03:04:36.0000] 
nox: well, that it's not automated

[03:04:41.0000] 
:)

[03:19:33.0000] 
w3c-test.org timing out for everyone or just for me?

[03:20:04.0000] 
hsivonen: wfm

[03:20:09.0000] 
annevk: thanks

[03:20:31.0000] 
it worked for me too a short while ago

[03:23:53.0000] 
ooh. now it came back

[03:56:16.0000] 
I just realized that if you use event handler attributes you leak an event listener until the element is GC'd

[04:57:25.0000] 
yes that seems accurate

[04:57:52.0000] 
Well it's the same as event listeners though?

[04:57:58.0000] 
In both cases you can remove manually, but if you don't, they leak

[05:16:26.0000] 
Domenic: you cannot remove an event handler

[05:16:35.0000] 
annevk: el.onfoo = null?

[05:16:51.0000] 
Domenic: does not remove the listener

[05:17:03.0000] 
sure, but now the listener does nothing, so I can't imagine engines keep around much if any memory for it

[05:17:18.0000] 
Sure, but still

[05:17:24.0000] 
Not sure I'd call that a leak

[05:22:47.0000] 
what is supposed to happen when there is a shadow DOM and element(s)FromPoint is called on that area?

[05:23:12.0000] 
do we always need to check the shadowRoot exists and use that?

[05:23:31.0000] 
what happens if its a closed shadowRoot?

[05:24:26.0000] 
AutomatedTester: https://github.com/w3c/csswg-drafts/issues/556

[05:26:06.0000] 
annevk: thanks!

[16:00:20.0000] 
annevk/hsivonen: Regarding the worries about CBOR in https://github.com/mozilla/standards-positions/issues/29#issuecomment-334376573, have you gone through WebAuthn to make sure its use of CBOR and CDDL is fine with you? I'm working on improving the CBORbis spec precision (https://github.com/cbor-wg/CBORbis/compare/master...jyasskin:editorial-improvements), but WebAuthn also depends on CDDL which is even less precise.


2018-03-14
[20:47:48.0000] 
jyasskin: sigh, I only looked at bits of WebAuthn

[20:48:03.0000] 
jyasskin: can you file an issue?

[20:49:22.0000] 
Folks are way too technical debt happy

[05:39:49.0000] 
annevk: How does the DOM handle: "class Foo extends Event {}; new Foo('hi')"? The spec says the event constructor "Returns a new event"

[05:40:11.0000] 
Safari seems to follow that to the letter, and returns an object with the Event constructor

[05:40:27.0000] 
Chrome/Firefox return an object with the Foo constructor

[05:44:09.0000] 
Edge behaves the same as Safari

[05:52:21.0000] 
JakeA: I don't really understand how that setup works; presumably it follows from IDL, but it's a little unclear to me

[05:52:52.0000] 
JakeA: worth filing an issue on, if only to get test coverage (and maybe an example)

[05:53:15.0000] 
JakeA: I think we do have such a test for EventTarget btw, but maybe not all browsers implement that yet

[05:53:31.0000] 
annevk: https://github.com/whatwg/dom/issues/598

[05:54:22.0000] 
JakeA: and dispatchEvent() works fine with it too, I guess?

[05:54:31.0000] 
/me wonders what else can be surprised here

[05:55:01.0000] 
annevk: It works in all browsers I've tested, but in Edge/Safari you just get an Event object

[06:17:13.0000] 
annevk: The EventTarget thing is a separate issue, right?

[06:25:07.0000] 
JakeA: if there's an issue, I guess

[06:25:19.0000] 
JakeA: just mentioning it to figure out what we should be saying in general

[06:25:30.0000] 
gotcha

[06:27:29.0000] 
Sebmaster: you around?

[06:27:40.0000] 
What's up?

[06:28:10.0000] 
Sebmaster: Unicode put up a new draft of UTS 46 with updated tests

[06:28:24.0000] 
Sebmaster: http://www.unicode.org/draft/Public/idna/11.0.0/IdnaTest.txt

[06:28:44.0000] 
hmm

[06:28:54.0000] 
Maybe I should be pinging TimothyGu instead ^^

[06:29:22.0000] 
Did they change anything about the actual spec?

[06:30:08.0000] 
I could probably poke tr46.js sometime tonight, haven't done any oss work recently

[06:30:40.0000] 
Sebmaster: http://www.unicode.org/draft/reports/tr46/tr46.html has the latest diff, doesn't seem substantial

[06:31:22.0000] 
Ah, pretty sure we have this in the code already due to test necessity

[07:06:31.0000] 
It looks like they added instructions on how to actually use the test file

[07:06:41.0000] 
And changed the format to be more usable for our purposes

[07:06:53.0000] 
It would be nice to verify that the updates were aligned with what we were hoping for, and if so, tell them

[07:07:20.0000] 
Domenic: yeah, I updated the relevant whatwg/url issue, Timothy will have a look next week hopefully

[07:08:12.0000] 
Domenic: I created https://github.com/w3c/web-platform-tests/pull/4504 once, but I never really figured out what the underlying issues were

[07:08:49.0000] 
I think TimothyGu did some magic to either the implementation or the test runner to make tr46.js pass all the tests

[07:09:01.0000] 
We should figure out how to port that to WPTs

[07:09:51.0000] 
The main problem here is lack of browser interest to fix anything

[07:10:07.0000] 
E.g., ToASCII of ASCII vs non-ASCII input

[07:10:10.0000] 
Yeah

[07:10:16.0000] 
I dunno, maybe WebKit cares

[07:10:37.0000] 
Still nice to have tests in case any of them wake up

[07:10:38.0000] 
E.g., getting Chrome and Edge to deal with "nontransitional"

[07:11:05.0000] 
Domenic: even Safari has weird bugs, where they flip the nontransitional bit depending on which API you use...

[07:11:19.0000] 
Maybe worth filing those

[07:11:26.0000] 
These are all filed

[07:11:29.0000] 
Ah OK

[07:11:46.0000] 
They're just not getting any attention

[16:54:17.0000] 
annevk: are you aware of where the new idnatest.txt is? It's not here yet: https://unicode.org/Public/idna/


2018-03-15
[19:18:34.0000] 
Sebmaster: I tweeted at the spec author and CCed you

[19:18:39.0000] 
Seems like a pretty bad oversight, lol

[21:45:16.0000] 
Sebmaster: apparently it was meant to be private still and I messed up

[22:32:23.0000] 
oh dear

[22:54:55.0000] 
THIS IS A FREENODE BREAKING NEWS ALERT!! Hitechcg AND opal ARE GOING AT IT RIGHT NOW WITH A LOT OF FIGHTING AND ARGUING WOW YOU DON'T WANT TO MISS THIS!! TYPE /JOIN ## TO SEE THE ACTION...AGAIN TYPE /JOIN ## TO SEE THE ACTION!!

[22:54:55.0000] 
THIS IS A FREENODE BREAKING NEWS ALERT!! Hitechcg AND opal ARE GOING AT IT RIGHT NOW WITH A LOT OF FIGHTING AND ARGUING WOW YOU DON'T WANT TO MISS THIS!! TYPE /JOIN ## TO SEE THE ACTION...AGAIN TYPE /JOIN ## TO SEE THE ACTION!!

[22:54:55.0000] 
THIS IS A FREENODE BREAKING NEWS ALERT!! Hitechcg AND opal ARE GOING AT IT RIGHT NOW WITH A LOT OF FIGHTING AND ARGUING WOW YOU DON'T WANT TO MISS THIS!! TYPE /JOIN ## TO SEE THE ACTION...AGAIN TYPE /JOIN ## TO SEE THE ACTION!!

[22:54:55.0000] 
THIS IS A FREENODE BREAKING NEWS ALERT!! Hitechcg AND opal ARE GOING AT IT RIGHT NOW WITH A LOT OF FIGHTING AND ARGUING WOW YOU DON'T WANT TO MISS THIS!! TYPE /JOIN ## TO SEE THE ACTION...AGAIN TYPE /JOIN ## TO SEE THE ACTION!!

[22:54:59.0000] 
rego a-ja dboehmer howdoi plutoniix eric_carlson justJanne hendry karlcow dmiles dgrogan jsbell espadrine Garbee stalled kbrosnan Manishearth mpt yoav gavin__ hober XhmikosR globbot moo-_- scott_gonzalez felixsanz MikeSmith robertkowalski wakaba tomaw manu ricea beowulf MrMaxMeranda daurnimator Domenic jdescottes hayato aklein birtles lerc nox annevk miketaylr twisted` timeless Krinkle jschoi SimonSapin NavidZ Dashiva wcpan Bakkot cgbrews

[22:57:15.0000] 
They always leave before I can kick them...

[22:57:56.0000] 
cowards!

[23:08:29.0000] 
/me was watching The Terminal minutes before and heard Gupta say “You are a coward! You come *this* close to America.. Coward!”

[00:38:21.0000] 
nox: https://github.com/whatwg/dom/pull/585#pullrequestreview-103789697

[01:15:41.0000] 
annevk: What about it?

[01:18:33.0000] 
nox: it's the thing you wrote a PR for once about resetting more state; question now is whether this other stuff should be reset too

[01:21:49.0000] 
Oh. Seems to me like everything should be.

[02:06:18.0000] 
Is w3c-test.org down?

[02:08:44.0000] 
It's here

[02:09:43.0000] 
Hi, in the Participant Agreement what does "a person who does not work in the field of web technologies" mean? If I work for a SaaS company that builds HTML widgets but doesn't contribute to specs or browsers, do I meet the definition of not working in "field of web technologies"?

[02:11:22.0000] 
beowulf: I think so

[02:11:56.0000] 
annevk: thanks

[02:19:44.0000] 
annevk: Ah damn. Doesn't work for me.

[02:20:14.0000] 
nox: I was agreeing

[02:25:37.0000] 
annevk: Oh. :)

[02:36:59.0000] 
annevk: https://github.com/whatwg/html/pull/3472 would be nice to get reviewed and merged

[03:03:11.0000] 
Domenic: looking

[04:03:41.0000] 
Domenic: are you waiting with reviewing the StaticRange thing until impl have made the change?

[04:03:51.0000] 
Domenic: or should someone else review it?

[04:04:59.0000] 
annevk: oh, hmm, I think I forgot. I guess since nobody else is stepping up, I should... Tomorrow, hopefully.

[04:05:26.0000] 
Domenic: I could ping that email thread with folks asking for it

[04:05:52.0000] 
I mean, might be good to get more eyes/involvement.

[04:08:13.0000] 
I pinged Florian, Johannes, Gary, and Chaals who I had been emailing with to some extent about this

[04:08:24.0000] 
We'll see what happens

[04:09:14.0000] 
Maybe I should ping Chong too, looking at the blink-dev thread

[04:14:35.0000] 
wptserve is running on the http://w3c-test.org/ host so I think the problem might be with the W3C firewall not allowing requests to get through to it

[04:15:03.0000] 
which I can’t fix on my own, so will ping the systems team

[04:15:23.0000] 
nox: ⬆

[04:15:28.0000] 
Pinged them too (and some others on the blink-dev thread)

[04:15:37.0000] 
MikeSmith: Ack, thanks.

[04:16:52.0000] 
Somehow I have the feeling that all these emails are typically in vain

[06:13:42.0000] 
nox: http://w3c-test.org/ is back

[06:13:58.0000] 
MikeSmith: You rock.

[06:15:28.0000] 
/me leans gets up from his rocking chair to pat himself on the back

[07:56:16.0000] 
annevk: so is the issue about the return value of dispatch algorithm or whether to return early? the initial comment hints about latter, but your latest comment former

[07:57:27.0000] 
smaug____: the return value of the returning early condition

[08:00:43.0000] 
I wonder if instead of returning early I should just invert the conditional and nest a whole bunch of steps

[08:00:52.0000] 
That would actually be a fair bit cleaner

[08:01:59.0000] 
I think return value should be false, if the event was already canceled before dispatching

[08:02:11.0000] 
did you test Gecko?

[08:02:26.0000] 
smaug____: Gecko throws on attachShadow

[08:02:38.0000] 
what has this to do with attachShadow

[08:03:01.0000] 
oh, retargeting

[08:03:29.0000] 
Anyway, I think I agree the return value should still branch on the canceled flag

[08:03:42.0000] 
you could enable dom.webcomponents.shadowdom.enabled

[08:04:14.0000] 
smaug____: thanks, I didn't know we had that flag, I have custom elements enabled

[08:04:38.0000] 
although it seems that's enabled by default these days

[08:04:55.0000] 
custom elements pref is enabled in Nighlies

[08:06:23.0000] 
smaug____: so yeah, Gecko returns false (doesn't clear target though if relatedTarget was in a shadow tree)

[08:06:59.0000] 
annevk: right, the target clearing happens per (old) spec

[08:08:16.0000] 
The inverse of "If target is relatedTarget and target is not event’s relatedTarget" is "If target is not relatedTarget or target is event's relatedTarget", right?

[08:25:18.0000] 
Went with that

[15:57:36.0000] 
annevk: from the ABNF at https://fetch.spec.whatwg.org/#http-new-header-syntax it currently seems not at all clear that the * wildcard is an allowed value for Access-Control-Allow-Methods and Access-Control-Allow-Headers


2018-03-16
[22:35:35.0000] 
MikeSmith: guess we should add a note

[22:36:07.0000] 
MikeSmith: the header and method production allow for *

[22:39:26.0000] 
yeah to be clear, it’s not a problem for implementors

[22:39:35.0000] 
instead just for developers/authors reading the spec

[22:40:43.0000] 
the root problem for developers/authors is that from reading the spec, it’s not clear they can use * in Access-Control-Allow-Headers to mean “any header name”

[22:41:42.0000] 
anyway I will make time to write up a PR that adds a note

[22:49:56.0000] 
The ABNF is mostly for implementers but sure

[22:51:30.0000] 
yeah

[22:55:52.0000] 
maybe making it clear to developers/authors is as simple as changing “For `Access-Control-Expose-Headers`, `Access-Control-Allow-Methods`, and `Access-Control-Allow-Headers` response headers the value `*` counts as a wildcard” to “For `Access-Control-Expose-Headers`, `Access-Control-Allow-Methods`, and `Access-Control-Allow-Headers` response headers, you can use the value `*` as a wildcard

[22:55:58.0000] 
(to mean "any header name" or "any method name")”

[23:00:39.0000] 
I’d have to read the CORS section again

[00:05:16.0000] 
foolip: https://github.com/whatwg/dom/issues/410

[00:14:30.0000] 
annevk: I have it pinned in my inbox, but won't be back at work until Tuesday

[00:15:08.0000] 
foolip: oh sorry, I did see you were away yesterday and forgot again

[00:15:13.0000] 
foolip: enjoy things

[01:12:11.0000] 
Domenic: can you review https://github.com/w3c/web-platform-tests/pull/9967 too?

[01:12:36.0000] 
foolip are IDL updates automated yet?

[01:12:58.0000] 
I don't think so, but note that it does more than just updating the IDL

[01:13:16.0000] 
Also, he's on vacation

[01:14:08.0000] 
Fair

[01:14:19.0000] 
Domenic: PRs aren't auto-created yet, but lukebjerring is working through all of interfaces/ to get them in sync to enable just that

[01:14:36.0000] 
Domenic: the ones with some boilerplate at the top are verbatim copies now

[01:14:45.0000] 
Ah OK cool

[02:21:35.0000] 
annevk: heading on vacation in like 20 minutes, but will try to get through some reviews and bug filing and stuff...

[02:21:58.0000] 
Domenic: aah enjoy all the great food there

[02:22:30.0000] 
Domenic: you got a week or a couple of weeks?

[02:22:33.0000] 
Two

[02:24:40.0000] 
Domenic: I think WebKit/Chrome have contenteditable=plaintext

[02:24:44.0000] 
Domenic: at least internally

[02:25:09.0000] 
Huh, I thought that was a proposal from the editing folks that browsers never implemented.

[02:25:46.0000] 
Hmm, now I'm starting to doubt, oh well, in any event, we shouldn't have multiple cursor/selection models

[02:27:44.0000] 
Nah I was right: https://cs.chromium.org/chromium/src/third_party/WebKit/Source/core/html/HTMLElement.cpp?type=cs&q=contenteditable+plaintext&sq=package:chromium&l=1232

[02:34:17.0000] 
I mean, we might have one model, but with the Range/Selection API just upconverting a cursor into an empty selection.

[02:36:14.0000] 
Domenic: but the