09:46
<annevk>
Anyone here understand compound microtasks? https://www.w3.org/Bugs/Public/show_bug.cgi?id=24810
09:47
annevk
was hoping to align mutation observers with the new world order
09:51
<annevk>
Oh beautiful
09:51
<annevk>
var x = document.createAttribute("test:test", "test"); w(x.localName)
09:54
<annevk>
Ooh, createAttribute does not take a second argument, what the actual fuck?
10:05
<annevk>
Teehee, x = XMLHttpRequest() now throws in Nightly
10:16
<annevk>
If you read http://www.w3.org/TR/DOM-Level-3-Core/core.html#ID-887236154 what would you expect to be returned when an attribute replaces itself?
10:26
<annevk>
foolip_: does my plan make sense to you?
10:27
<annevk>
foolip_: if you feel like we could get rid of createAttribute() / setAttributeNode() I could leave those out for now, but it seems like a lost cause
12:55
<zcorpan_>
<details> and <summary> aren't especially searchable names
13:10
<GPHemsley>
annevk, dglazkov_: Can you clarify for me what HTML Imports are intended to be used for?
13:11
<annevk>
Importing HTML fragments without the overhead of a browsing context
13:12
<smaug____>
xinclude ;)
13:12
<wilhelm_>
Is this in any browsers yet?
13:13
<GPHemsley>
annevk: To answer your unrelated question from earlier: Given "Replacing an attribute node by itself has no effect." and "If the newAttr attribute replaces an existing attribute, the replaced Attr node is returned, otherwise null is returned.", I would expect null to be returned.
13:14
<GPHemsley>
annevk, smaug____: How does one use it? (The one example in the spec doesn't really show the results.)
13:14
<smaug____>
(and no, it is not xinclude. But similar ideas come and go.... Shadow DOM ~ XBL1)
13:15
<GPHemsley>
Is it only accessible via scripting?
13:29
<GPHemsley>
annevk: Alright, well, I've given my preliminary assessment; feel free to correct any misunderstandings I may have.
13:46
<zcorpan_>
Hixie: "For example "</video>" causes the video element to start fetching video content" https://www.w3.org/Bugs/Public/show_bug.cgi?id=24798 - that is not true iirc. fetching can start earlier. but playback is blocked because there might be <track>s that should block
13:49
<annevk>
GPHemsley: null is what I expected as well, but of course browsers do the other thing (consistently though, so there's that)
13:50
<annevk>
GPHemsley: you'd use HTML imports through scripting, yes
14:58
<GPHemsley>
annevk: I suppose the other behavior is also a possible interpretation of those quotes. I guess "replacing" is a cause, not an effect, so the replacement does actually happen, returning the new Attr node (which is the same as the old Attr node).
14:58
<GPHemsley>
annevk: Seems like the wording to be improved to clarify what "no effect" means (because apparently returning the Attr node instead of null is not an effect).
15:28
<dglazkov>
good morning, Whatwg!
15:32
<annevk>
GPHemsley: well that spec is dead :-)
15:33
<annevk>
the new one will have a crystal clear algorithm
16:17
<GPHemsley>
alrighty
16:26
SamB
wonders why http://www.w3.org/Style/CSS/Test/Fonts/Ahem/Ahem.ps is in the old Mac format, or, heck, has a .ps extension ...
16:28
<tantek>
perhaps ask Todd Fahrner?
17:12
<SamB>
tantek: meh, don't care that much ...
17:13
<annevk>
SamB: prolly because it was created back then
17:46
<annevk>
foolip_: sorry about that
17:46
<annevk>
foolip_: really thought ownerElement was widely deployed
17:47
annevk
wonders if foolip_ actually ever checks IRC
18:00
<Hixie>
annevk: can you elaborate on https://www.w3.org/Bugs/Public/show_bug.cgi?id=25089 ?
18:00
<Hixie>
(microtasks in workers question)
18:01
<Hixie>
why would workers have microtasks?
18:01
<Hixie>
i thought all microtask things were dom-based right now
18:01
<Hixie>
table sorting, mutation observers, web components, etc
18:02
<Hixie>
i guess it could be made more generic
18:02
<Ms2ger>
Object.observe?
18:05
<Hixie>
Object.observe interfaces with HTML's event loop?
18:06
<Hixie>
i have yet to hear anything form TC39 about anything i'm supposed to do about their event loop thing
18:06
<Hixie>
from
18:10
<annevk>
Hixie: promises and Object.observe() are microtasks
18:10
<annevk>
Hixie: ES6 calls them "tasks" at the moment, confusingly enough, they're considering changing the name
18:10
<Domenic___>
despite allen's insistence that ES tasks are different :(
18:10
<Hixie>
i've updated the spec to make microtasks happen on workers too
18:10
<Domenic___>
wow how do i keep growing underscores on my name
18:10
<Hixie>
right now though, ES thingies, whatever they are called, aren't invoked by HTML's microtasks logic.
18:11
<Hixie>
so they ain't microtasks yet.
18:11
<annevk>
Hixie: fair enough, at some point we need to study the "host integration" I guess
18:11
<annevk>
Hixie: hopefully Allen fixes the bug around global != this too
18:11
<annevk>
Hixie: I filed that a while ago now...
18:11
<Hixie>
that'd be nice
18:34
<annevk>
web-apps-tracker keeps breaking down :(
19:07
<annevk>
Might be a DreamHost issue, I filed a ticket
19:08
<annevk>
It seems https://github.com/whatwg/html-mirror/commits can be used meanwhile although it lags a little bit
19:11
<Domenic_>
TabAtkins: how's the CSS variables thread going
19:31
<TabAtkins>
Domenic_: OH GOD SAVE ME
19:31
<TabAtkins>
In other words, it's done. We'll discuss tomorrow and make a decision.
19:31
<TabAtkins>
Nothing going on in the thread is worth anything anymore.
19:32
<Domenic_>
yeah, that's what i figured. (yet you keep replying... :P)
19:32
<TabAtkins>
I'm a bit of a masochist in some realms.
19:32
<Domenic_>
where are people standing, generally?
19:33
<TabAtkins>
Looks like the use of -- as a prefix will work out well. It's a small parser change.
19:33
<Domenic_>
interesting!
19:33
<Domenic_>
nice, that seems like a win. and you think it can be extended to other custom thingies?
19:33
<TabAtkins>
Yup, I'll just change over the current proposal from a _ prefix to a -- prefix.
19:33
<Domenic_>
that's awesome
19:43
<annevk>
what's wrong with _?
19:45
<TabAtkins>
It's ugly.
19:46
<Ms2ger>
Like --
19:46
<Domenic_>
LIKE YOUR FACE
19:47
<TabAtkins>
People seem to consider dashes less ugly than underscores, shrug.
19:51
<jamesr_>
my shift key thanks you
19:52
<TabAtkins>
That's probably at least partially the reason, yes.
19:52
<TabAtkins>
(That's one of the reasons I like Lisp better than most languages - I can just use "-" rather than having to hit Shift to get _ or camelCase.)
19:57
<hober>
yeah, when i suggested prefix--foo i was thinking of the emacs lisp convention for package-private symbols
20:16
<TabAtkins>
Ah. Well, people can always use more -- on the inside, if they want.
20:43
<Ms2ger>
annevk, getAttributeNode() is coming back?
20:44
<annevk>
Ms2ger, yeah it might
20:57
<SamB>
one actual problem I've had with underscores is that sometimes they get buried in link underlines ...
21:05
<Ms2ger>
TabAtkins, so you're just doing Formal Day now?
21:05
<TabAtkins>
Just trying to class it up more regularly. Still only doing full suits on Friday.
21:06
<TabAtkins>
Also, I just bought a sweater, so I wanted to try it out.
21:07
<Ms2ger>
... Slowly moving into management
21:07
<TabAtkins>
I am, technically, a manager at the moment, since I'm employing the CSSWG scribe.
21:07
<TabAtkins>
...also, I'm now paying fantasai.
21:07
<Ms2ger>
Ha
21:10
<TabAtkins>
Though I do dress better than 90% of Google management (at least) on my classy days.
21:14
<Hixie>
TabAtkins: you confused me when i ran into you earlier, because it wasn't friday
21:14
<TabAtkins>
Hahahaha
21:14
<Hixie>
TabAtkins: i was busy trying to do two things at once on my phone and listen to a podcast and then there's this guy in a suit waving at me and i'm like "aaah, what's going on"
21:14
<Hixie>
"it's not friday!"
21:17
<TabAtkins>
I'm not even close to being in a suit.
21:17
<TabAtkins>
Jeans and bare feet, I'm just wearing a nice shirt, a sweater, and a tie.
21:18
<Ms2ger>
That sounds like an interesting combination
21:18
<Ms2ger>
Like something a news anchor could wear if he's behind a desk
21:18
<TabAtkins>
They just don't wear pants at all.
21:19
<Hixie>
TabAtkins: well i dunno, i wasn't looking where i was going, let alone what you were wearing :-)
21:21
<Hixie>
as far as i'm concerned, if you're wearing a tie, it's asuit
21:21
<Hixie>
even if it's printed on a t-shirt, that's probably close enough
21:21
<TabAtkins>
Oh god.
21:21
<Hixie>
:-P
21:21
<TabAtkins>
That's terrible.
21:21
<TabAtkins>
Does anyone know what the offset is for the hours in Gavin's stats? They kind look like west coast times.
21:22
<TabAtkins>
Since I have no blue at all, which would correspond to roughly midnight-6am.
21:22
<Ms2ger>
Sounds right
21:23
<TabAtkins>
Tangentially related, I like that Divya holds both first and second place for most all-caps messages.
21:23
<TabAtkins>
Or rather, highest percentage.