07:38
<hober>
MikeSmith: I figured out that jsdom issue
07:38
<MikeSmith>
oh?
07:38
<MikeSmith>
cool
07:38
<MikeSmith>
what was the problem?
07:39
<hober>
If you require('jsdom/level1/core') instead of require('jsdom/level2/core'), it works
07:39
MikeSmith
tries it
07:39
<hober>
var DOM = browser.browserAugmentation(require('jsdom/level1/core').dom.level1.core)
07:39
<hober>
(is the replacement for the relevant line)
07:40
<MikeSmith>
seet
07:40
<MikeSmith>
sweet
07:40
<MikeSmith>
I guess the level2 stuff is pre-alpha
07:40
<hober>
yeah
07:40
<hober>
html->atom is much nicer with things like createElementNS
07:40
<hober>
but I guess I can fake it for now
07:41
<MikeSmith>
yeah
07:41
<MikeSmith>
so after that change is htmlparser working now for you? (for what you were using it for)
07:42
<MikeSmith>
and should I post a follow-up comment to the bug you pointed me to before?
07:43
<hober>
Oh. I haven't been using node-htmlparser at all; I've been using aredridel/html5
07:43
<hober>
which is essentially html5lib in js
07:44
<MikeSmith>
hober: yeah, that's the one I meant
07:44
<hober>
aredridel/html5 has been working for me. at least, last I remember screwing with this stuff it seemed to work.
07:45
<hober>
but creating new documents from whole cloth was failing, and now I know it's due to my premature desire to use level 2 DOM methods.
07:45
<MikeSmith>
among other things, I'm wondering how its performance is relative to html5lib (python) or to whatever non-HTML5 parsers you might have tried (e.g., the HTML parser in libxml2)
07:45
<hober>
for my current purposes (offline, static site generation), I'm not terribly concerned about performance
07:48
<MikeSmith>
ok
07:50
<hober>
(that said, I imagine I'm the outlier, Node.js-wise, insofar as performance concerns go.)
07:51
<hober>
I'm working on a publication workflow in which the "pretty this HTML up for publication" logic can happen browser-side, during composition, and server-side, when the publish button is pressed
07:51
<hober>
(the prettyifying code being exactly the same)
07:55
<MikeSmith>
yeah, that's a really nice advantage to have
07:55
<MikeSmith>
(being able to use the same code server side and client both)
07:56
<hober>
The longer-term idea I'm working on, in a somewhat cryptic tweet: http://twitter.com/hober/status/15739328281
08:37
<MikeSmith>
hober: interesting
09:22
<gsnedders>
jgraham: Listening to Rasputina again, I again conclude they're not my sort of music.
10:29
<annevk>
Hixie, <details> changes the DOM based on user interaction
11:13
<annevk>
I wish I had paid closer attention to DOM Events
11:35
<MikeSmithX>
annevk: you mean because you think it has problems that should have been fixed before now?
11:40
<annevk>
I think it might have and I think it would have helped to point them out at early stages
11:41
<MikeSmithX>
I see
11:41
<MikeSmithX>
I guess that's true in general of any spec
11:41
<annevk>
Yep
11:42
<annevk>
See e.g. all the RDF nonsense to describe something as simple as WAI-ARIA
11:42
<MikeSmithX>
but some are more important than others and it doesn't harm as much to have ignored them
11:42
<annevk>
makes it completely unreadable, but it's not going to change for this edition
11:42
<MikeSmithX>
oh
11:43
<MikeSmithX>
I can see that's a problem
11:43
<MikeSmithX>
but not quite as serious as I thought you might have meant
11:44
<annevk>
well, real serious ends up getting ignored at the implementation stage
11:44
<MikeSmithX>
yeah
11:44
<annevk>
e.g. various things that were once in SVG 1.2
11:44
<MikeSmith>
yeah, i know
11:44
<MikeSmith>
as far as browsers go at least
11:45
<MikeSmith>
which is what's really important of course
11:45
<annevk>
semi-serious things -- like how the CSS syntax is insanely complicated if you look at the details while it did not have to be at all -- need to be tackled sooner rather than later
11:46
<MikeSmith>
um, what's exactly the RDF part you're referring to?
11:46
<annevk>
there's no reason that the character sequence "@namespace" can be written in near-infinite ways
11:47
<annevk>
http://www.w3.org/TR/wai-aria/roles is full of it
11:47
<annevk>
"superclassing" "subclassing" etc.
11:48
<annevk>
all kinds of complexity that a) authors are not interested in and b) implementations are not either
12:00
<MikeSmith>
on the face of it, that seems like the same kind of overengineering that's a misfeature of a lot of specs
12:01
<MikeSmith>
but maybe it could be useful in some complex apps
12:01
<MikeSmith>
I mean, I don't myself know all what's involved in adding ARIA to a complex app like Gmail or whatever
12:04
<MikeSmith>
hmm, I think browsing with Disable Caches set in Webkit/Safari seems to possibly be a great way to cause the fan on your laptop to run at high speeds continuously
12:05
<MikeSmith>
I think it might be causing "Safari Webpage Preview Fetcher" to keep launching and re-launching to fetch previews of pages that then don't get cached anywhere, and so it fetches them again and again
12:05
<MikeSmith>
maybe
12:06
<MikeSmith>
"defaults write com.apple.Safari DebugSnapshotsUpdatePolicy -int 2" seems to possibly be the way to stop it
12:08
<MikeSmith>
sitepoint.com pop-ups are really cheesy
12:09
<MikeSmith>
I thought the people behind Sitepoint were better than that
12:15
<hsivonen>
the inconsistency of the format support table at http://en.wikipedia.org/wiki/HTML5_video annoys me again. It's even more bogus thatn before.
12:15
<hsivonen>
I vented on the talk page, since the page itself is guarded by a user with an intimidatingly long user page
12:16
<hsivonen>
people with *huge* user pages always have the time to out-revert you, so why even bother
16:36
<Dashiva>
So there's no window in workers, only self?
16:41
<annevk>
yup
17:56
<MikeSmith>
hsivonen: dunno how much time if any you've had lately to think about migrating the v.nu sources to a git or mercurial hosting service, but I wanted to say I'd be willing to do the work to make it happen
17:57
<MikeSmith>
that is, dealing with actually getting them migrated over, as well as updating the build script and whatever else would be needed
17:57
<MikeSmith>
github seems like it might best place to consider moving to
17:59
<MikeSmith>
for among other reasons the network effects of the fact that there are so many very active developers there
18:01
<hober>
github supports svn, btw
18:02
<hober>
(see http://github.com/blog/644-subversion-write-support for more)
18:05
<MikeSmith>
hober: part of the motivation is to get away from svn, actually
18:05
<MikeSmith>
at least on my part
18:06
<MikeSmith>
I've never been a big subversion fan
18:06
<MikeSmith>
I really that github makes it dead simple to fork repos
18:06
<hober>
github's svn support is actually git-backed. it lets people who aren't yet comfortable with git use something familiar while the other collaborators forge on ahead with git
18:06
<MikeSmith>
ah, I see
18:06
<MikeSmith>
well that's good then
18:07
<hober>
e.g. svn checkout http://svn.github.com/sideshowbarker/jsblog.git
18:08
<hober>
(use https for a read/write checkout)
18:11
<MikeSmith>
cool
18:12
<MikeSmith>
there more I use github, the more I like it
18:12
<MikeSmith>
it ain't perfect but it was clearly designed by people with clue
18:13
<MikeSmith>
and in general it just seems to work the way a system like that ought to
18:13
<hober>
yeah, those guys are rad
18:14
<MikeSmith>
the ease of forking and the culture really encourage experimentation
18:15
<MikeSmith>
I think getting the validator.nu sources available there might help a lot toward getting some code contributions for yet-to-be-implemented features and bugs too
18:15
<MikeSmith>
it'd raise more awareness about it and hopeflly get some more good devs involved
18:21
<MikeSmith>
heh
18:21
<MikeSmith>
http://www.xtranormal.com/watch/6995033/ is pretty good
18:21
<MikeSmith>
the first couple of minutes at least
18:39
<micheil>
MikeSmith: still working on your KO?
18:41
<MikeSmith>
well, I'm just working on a personal KO
18:42
<MikeSmith>
I don't want to embarrass myself by bringing much attention to it yet
18:42
<MikeSmith>
but so far I've messed around a bit with express, Connect, and the ext.js stuff
18:43
<MikeSmith>
and it's all really quite nice
18:44
<MikeSmith>
and of course the fork I made of brianleroux's wtfs stuff
18:45
<MikeSmith>
gotta love that dude
18:45
<MikeSmith>
his comments in the "templating" part of that About page for that site are great
18:47
<MikeSmith>
[[
18:47
<MikeSmith>
Nothing is perfect. And sometimes shit is just plain wack. The ExpressJS framework is seriously awesome but it has a hardon for HAML which solves problems I do not have. I opted, out of lazyness, for EJS which is not beautiful; tho it is functional.
18:47
<MikeSmith>
]]
18:47
<MikeSmith>
that part
18:47
<MikeSmith>
http://wtfjs.com/about
18:50
<MikeSmith>
micheil: what'd you got cooked up for KO?
18:51
<micheil>
well, I'm judging, but I am working on a canvas game thuing
18:52
<micheil>
MikeSmith: y'know, I could always track down the repo :P
18:54
<MikeSmith>
I'm just at the rubbing-two-sticks-together-to-make-fire stage
18:54
<MikeSmith>
but then again, that's pretty much the stage at which all the code I write remains
18:55
<MikeSmith>
regardless of what language it's written in
18:55
<MikeSmith>
a one-man coding WTF
19:12
<micheil>
MikeSmith: deployed anything yet?
19:14
<MikeSmith>
nope
19:26
<micheil>
MikeSmith: what was your team name?
19:26
<MikeSmith>
no team, man
19:27
<micheil>
do you wish to have a team?
19:27
<micheil>
that would make you officially part of KO
19:27
<MikeSmith>
yeah, in general
19:28
<MikeSmith>
I prefer always to have other people to do my work for me when I can
19:28
<micheil>
okay, shoot an email to all⊙nc asking if you could have a team created for yourself
19:28
<MikeSmith>
hai
19:28
<MikeSmith>
oh!
19:29
<MikeSmith>
somebody from japan showing up on the map finally
19:29
<MikeSmith>
connection
19:29
<MikeSmith>
walter-sobchak
19:31
<hober>
I thought he didn't code on shabbas
19:32
<MikeSmith>
heh
19:32
<MikeSmith>
they have done some impressive work so far -
19:32
<MikeSmith>
http://walter-sobchak.no.de/
19:33
<MikeSmith>
bold, minimalist approach
19:33
<micheil>
hmm.. does unsaing-daiva keep coming up on the map?
19:34
<micheil>
*unsaving
19:43
<MikeSmith>
ah, my Webkit nightly seems to have a turned itself into a torture-testing tool for the fan on the laptop
19:43
<MikeSmith>
to see how long it can continuously run at full speed until it quits completely
19:48
<MikeSmith>
hmm, or maybe it's just that tab I had up for the xtranormal movie I started making 3 hours ago and forgot about til now...
19:58
<Dashiva>
How did noscript get all the way to issue status?
20:03
<Philip`>
It was "raised for Shelley Powers" whose "purpose in wanting to file change proposals isn't that I expected any of them to succeed", so presumably the point is just to waste time and/or to make a point
20:05
<Dashiva>
So there's no filtering at all on issues now, any bug that gets closed has a free promotion card?
20:41
<MikeSmith>
hey, it's the voice of micheil
20:41
<MikeSmith>
http://thechangelog.com/post/927103350/episode-0-3-1-websockets
20:41
<micheil>
it is
20:41
<micheil>
didn't I post that?
20:42
<micheil>
I mean, didn't I post a link to that in here?
20:46
<MikeSmith>
micheil: probably I missed it
20:46
<micheil>
okay
20:46
<micheil>
well, enjoy :)
21:04
<MikeSmith>
micheil: you didn't say so much on that podcast
21:04
<MikeSmith>
those other dudes talk too much
21:04
<micheil>
not really
21:05
<MikeSmith>
it was actually pretty interesting though
21:05
<micheil>
I'm meant to be a host, so, it's focusing on the guests
21:06
<micheil>
it gets' kinda awkward though when you interview guests and can answer the other's questions
21:08
<micheil>
hmm.. http://unsaving-daiva.no.de/