07:05
<annevk>
JakeA: thanks for https://github.com/w3c/webappsec-mixed-content/issues/7
07:05
<annevk>
JakeA: was thinking about raising that issue this morning in that exact repository
07:11
<JakeA>
No problem! Still worried about the practicality of it, but I get where you're coming from
07:18
<annevk>
Yeah, well, folks should just move to HTTPS really
07:19
<annevk>
That we present HTTP as more favorable than mixed content is part of the problem, they're sorta the same
07:21
<JakeA>
We may have made that change in Chrome, or we've certainly been moving in that direction
07:25
<annevk>
Hmm, https://github.com/whatwg/dom/commit/9e3ce67c7927d6642646a3d0c84fa6d8f7926cfa no longer points to the correct PR
07:25
<annevk>
So removing the PR annotation might be problematic?
07:27
<Domenic>
Wasn't that in the pre-green-merged-button days?
07:28
<Domenic>
It might also just be a GitHub bug.... no idea why tab's fork took over the pointer
07:33
<annevk>
Yeah could be, though April seems pretty recent
08:11
<annevk>
Domenic: accessible nodes have their own tree as well with focusing and such?
08:11
<annevk>
Domenic: so now we have shadow trees, accessible trees, and normal trees?
08:11
<Domenic>
annevk: apparently. I am not super-jazzed about them having parallel concepts, *especially* event dispatch.
08:11
<annevk>
Domenic: that sounds very complicated
08:12
<Domenic>
annevk: part of the idea, I think, is to allow adding a "virtual" ax tree underneath a canvas
08:12
<Domenic>
It *is* apparently exposing the primitives that browsers already implement
08:12
<annevk>
Domenic: all browsers?
08:12
<annevk>
Domenic: I doubt we have something like that, unless they mean XBL, which is shadow trees
08:13
<Domenic>
annevk: hmm i'm not sure. but i believe all the implementers are involved...
08:13
<annevk>
Domenic: but nobody with a background in API design is on the list of editors
08:16
<Ms2ger>
https://lists.w3.org/Archives/Public/public-w3process/2016Aug/0017.html
08:18
<annevk>
Domenic: also, I thought <canvas> was fine with just having an actual subtree
08:19
<Domenic>
annevk: yeah, that's why they asked for a review...
08:19
<Domenic>
Not sure what's the problem with canvas subtrees
08:19
<Domenic>
Ms2ger: this seems boring, why did you send me a boring thing
08:20
<Ms2ger>
Domenic, maybe this is better: https://www.youtube.com/watch?v=Y6EAWMscDvI
08:20
<Domenic>
awwwww
08:21
<mkwst>
annevk: Do you happen to know whether I need to poke at you, TabAtkins, or tobie to get Fetch into Bikeshed's anchor data list?
08:21
<Ms2ger>
Ah, mkwst, I was looking for you earlier
08:21
<annevk>
Ms2ger: oh joy, more formatting requirements on drafts and charters to get patent protection
08:21
<annevk>
Ms2ger: what could go wrong
08:22
<mkwst>
Is it a data problem with the spec's markup, or a specref/bikeshed issue?
08:22
<annevk>
mkwst: TabAtkins
08:22
<mkwst>
Ms2ger: What can I do for you?
08:22
<annevk>
mkwst: though he might say that I should convert Fetch to bikeshed
08:22
<Ms2ger>
mkwst, https://github.com/whatwg/html/issues/1142
08:23
<Domenic>
specifically https://github.com/whatwg/html/issues/1142#issuecomment-216990289 i would imagine
08:23
<mkwst>
He probably will. *shrug* I'd help if that's the direction you want to do. Having a separate build system just for Fetch is annoying. :)
08:26
<annevk>
mkwst: it's not used just for Fetch and I don't disagree, but conversion is error prone and takes a lot of time
08:26
<mkwst>
annevk: Well, on _my_ computer, it's just Fetch. :)
08:26
<annevk>
It took months to fix regressions in URL and DOM
08:27
<annevk>
And they were only fixed thanks to careful reviewers, of which Fetch doesn't have too many
08:27
<Domenic>
DOM still has lots of formatting regressions, I noticed yesterday :(
08:27
<mkwst>
Fair. I suspect Fetch is contained enough that we'd be able to do it reasonably well, but I would have said the same about URL.
08:27
<Domenic>
Half the methods are defined as bold code, the other half as bold
08:28
<Domenic>
We could do what Tobie's doing for Web IDL and work on an XSLT transform :O
08:31
<Domenic>
ohoho fighting words ^_^ https://github.com/tabatkins/bikeshed/issues/772#issuecomment-237487886
08:33
<Domenic>
Hmm except for scheduling meetings Tokyo is a pretty great timezone. You get to watch the Europe people waking up and answering your issues, then the US people all send you email while you sleep which you can batch-process in the morning.
08:34
<tobie>
I know you're joking, but frankly, XSLT is a pain. The only reason it made sense for webidl is because webidl was already using it to convert from XML and I could hack on a pre-existing sheet.
08:40
<annevk>
Domenic: Europe has similar advantages with London being the best I think, but Tokyo et al is sure nice too
08:56
<tobie>
Basically, any place where you get to batch-process emails coming from the US is good.
08:58
<tobie>
It would be interesting to compare actual output of remote engineers based in the vicinity of their HQ's timezone.
08:58
<tobie>
I'm ready to bet those with very little work hour overlap have an edge.
08:59
<tobie>
On the other, their network is probably poorer, and so they might have less leverage.
09:20
<stennowork>
i don't know if this is the right channel to ask, but is there any sensible explanation why 'contenteditable' is an 'enumerated' attribute and not a boolean attribute?
09:21
<stennowork>
given that its enumerated values are exactly 'true' and 'false'
09:28
<Ms2ger>
Legacy
09:28
<stennowork>
legacy? wasn't it added in html5?
09:34
<stennowork>
hmm i see that html4 doesn't even define a boolean datatype
09:34
<Ms2ger>
It existed long before it had a spec
09:35
<stennowork>
i see, fair
09:35
<stennowork>
thanks for the explanation
09:36
<stennowork>
so i could send a proposal to change it to boolean?
09:37
<Ms2ger>
You could, but it won't be accepted
09:37
<Ms2ger>
That would break existing sites
09:37
<stennowork>
hmm aw
09:39
<Ms2ger>
Also, isn't there an implicit third state "inherit"?
09:41
<stennowork>
hmm thats right
09:41
<stennowork>
i must've overread that
09:41
<stennowork>
my bad
10:04
<botie>
hsivonen, at 2016-07-28 03:02 UTC, MikeSmith said: multiple reports about validator.nu redirecting to hsivonen.fi https://github.com/validator/validator/issues/325
10:05
<annevk>
Domenic: other things to consider if you're still in Tokyo, some kind of PromiseResponse subclass that fetch() returns on which we expose 1xx responses and push promises
10:05
<annevk>
Domenic: both of which are effectively async arrays
10:06
<annevk>
Domenic: if you're still in the office in Tokyo*
10:34
<smaug____>
does anyone know if there is some better code indexing tool for webkit than trac.webkit.org ?
10:37
<beverloo>
I usually look at the GitHub mirror - https://github.com/WebKit/webkit
10:41
<smaug____>
I asked for better tool ;)
10:41
<smaug____>
I wouldn't consider github being good with anything
10:44
<smaug____>
something like searchfox.org or mxr or dxr or google code search?
10:49
<smaug____>
anyone with webkit?
10:49
<smaug____>
beverloo ?
10:49
<smaug____>
What does data:text/html,<script>document.write("onwebkitTransitionEnd" in document.documentElement); </script> say?
10:50
<smaug____>
and what if that is onwebkittransitionend
10:51
<beverloo>
smaug____, my main use is finding specific bits of code, but I used to be fairly well acquainted with WebKit code anyway. GitHub works fine for that
10:51
<beverloo>
Not sure if there's anything around that really indexes the code
10:52
smaug____
is so used to mxr and now searchfox.org. searchfox.org particularly has rather good UI for blame.
10:52
<smaug____>
beverloo: right. anyhow, I assume you have webkit there?
10:52
<smaug____>
could you test data:text/html,<script>document.write("onwebkitTransitionEnd" in document.documentElement); </script>
10:52
<smaug____>
and then data:text/html,<script>document.write("onwebkittransitionend" in document.documentElement); </script>
10:53
<beverloo>
No, sorry, I don't
10:53
<smaug____>
ah, k
11:35
<smaug____>
hsivonen: ping
11:36
<smaug____>
hsivonen: ejpbruel in moznet#developers has some questions to you
11:57
<MikeSmith>
smaug____: false in webkit for data:text/html,<script>document.write("onwebkitTransitionEnd" in document.documentElement); </script>
11:58
<MikeSmith>
smaug____: true in webkit for data:text/html,<script>document.write("onwebkittransitionend" in document.documentElement); </script>
11:58
<smaug____>
MikeSmith: thanks (though I got the answer elsewhere)
11:58
<annevk>
I forgot to comment on that bug
11:58
<annevk>
The suggested capitalization of event handlers seemed extremely suspect
11:59
<annevk>
Glad testing reveals WebKit is somewhat reasonable
11:59
<smaug____>
I wish there was some website where one could run scripts in various browser engines
11:59
<smaug____>
I don't need UI usually
11:59
<annevk>
smaug____: browserscope?
12:00
<annevk>
smaug____: it's not ideal I guess for just running scripts, but maybe someone has written something on top...
12:02
<smaug____>
haven't used that for ages
12:02
<smaug____>
it was a bit heavy and wasn't it using Flash
12:02
<smaug____>
but perhaps it has changed
12:02
<annevk>
I don't think I have Flash enabled, but it does require an extension to be installed
12:04
<smaug____>
ahaa
12:04
<smaug____>
and google account to log in
12:04
smaug____
creates yet another random google account
12:06
<smaug____>
oh, it is very different what it used to be
12:08
<smaug____>
I thought it used to let one to load random url and it showed the UI of the browser
12:10
<annevk>
Nah, you can interact with the browser, it's quite nice
12:10
<annevk>
Although the debugging tools on Edge 14 make it crash
12:10
<annevk>
At least the last time I tried, but Edge 13 works fine
12:11
<smaug____>
hmm, how do I get the UI?
12:11
<smaug____>
I must be missing something
12:16
<annevk>
smaug____: apologies, I meant browserstack
12:16
<smaug____>
ahaa
12:16
<annevk>
smaug____: jst can give you credentials if you don't have any, we have some kind of deal with them
12:16
<smaug____>
right, this is the MS' site
13:27
<annevk>
JakeA: nice summary about the service worker meeting
13:28
<JakeA>
Cheers!
16:31
<benydc>
Hi All
16:40
<benydc>
I have a question about fetch http client
16:41
<benydc>
I want to make repeated requests but without overflowing the memory, how can I close the connections and clear the memory?
16:50
<annevk>
benydc: memory?
16:51
<benydc>
I have screenshot
18:13
<TabAtkins>
mkwst, annevk: While I can put Fetch into Bikeshed, as it's not a Bikeshedded spec and doesn't independently follow Bikeshed's dfn-markup rules, the definitions won't be exposed in any useful manner. They'll all be "dfn" type and unexported.
18:23
<mkwst>
TabAtkins: What's the minimum that needs to happen to make the import useful?
18:23
<TabAtkins>
mkwst: Commenting in the Bikeshed bug that I just saw right now.
18:23
<mkwst>
s/<dfn/<dfn data-export/g ? :)
18:24
<mkwst>
Ok. If you add it to the automatic importing thing, I'll send annevk patches that do whatever you're about to tell me to do. Then it'll Just Work(tm), I'm sure.
18:25
<TabAtkins>
Correct.
18:25
<TabAtkins>
Whenever it gets fetched by Shepherd, which in the worst case is overnight.
18:32
<mkwst>
Wunderbar. I'll poke at that at some point, thanks!
18:33
<mkwst>
I'm sure annevk is going to love a million data attributes cluttering his pristine HTML. :)
18:33
<TabAtkins>
He already handwrites a bunch of stuff that would have been omitted by a decent processor, so he's bringing it on himself. ^_^
21:07
<wanderview>
JakeA: are you awake?
21:12
<wanderview>
JakeA: I wrote my question up as an issue: https://github.com/slightlyoff/ServiceWorker/issues/943
21:49
<JakeA>
wanderview: will read properly tomorrow, but your idea seems right at a glance
23:01
<gsnedders>
where does @apply come frm?