00:42
Lachy
waves from Australia
00:43
<Lachy>
It's good to be back in a normal country with Vegemite, Cottees Cordial and Tim Tams :-)
00:47
<eric_carlson>
Hixie: ping
00:47
<Hixie>
hi
00:47
<eric_carlson>
Hey. I have a question about firing media events
00:48
<eric_carlson>
the spec clearly says that some should fire synchronously, eg. "must synchronously fire a progress event called loadstart" (4.8.10.4)
00:48
<eric_carlson>
and others are clearly asynchronous, eg. "must queue a task to fire a simple event called play" (4.8.10.7)
00:48
<eric_carlson>
but the descriptions in the section on ready states (4.8.10.6) are somewhat ambiguous
00:48
<eric_carlson>
eg. "and fire a simple event called play"
00:49
<eric_carlson>
(even though "canshowcurrentframe" and "play" are explicitly asynchronous elsewhere)
00:49
<eric_carlson>
so are the ready state change events fired differently from others?
00:50
<Hixie>
"fire a simple event" is synchronous, but that doesn't preclude an error in the spec :-)
00:50
<Hixie>
i should probably remove th word "synchronously" from the spec so that it doesn
00:50
<Hixie>
't confuse people when i omit it
00:51
<eric_carlson>
OK, thanks.
00:52
<Hixie>
wait, hold on
00:52
<Hixie>
it has to be asynchronous for those
00:52
<Hixie>
there's nothing to be synchronous with
00:52
<eric_carlson>
good, that's what I was about to argue.
00:52
<eric_carlson>
now I don't need to bother ;)
00:53
<Hixie>
i'll file a bug saying that these should be fixed to say that a task should be queued to fire the events
01:01
<eric_carlson>
that was quick. thanks!
01:03
<Lachy>
LOL, someone pretending to be Hixie is posting spam to the whatwg blog :-)
01:04
<Lachy>
(unless Hixie really was posting about eugenics and nazis, but I doubt it)
01:12
<Hixie>
ooo
01:12
<Hixie>
where
01:12
<Hixie>
where
01:15
<Lachy>
Hixie, it was in the moderation queue, I already marked it as spam
01:15
<Lachy>
but you wrote: "The biggest problem I see with supporting eugenics is the bad name it has aquired, through being associated with the Nazi movement, for instance. I'll leave that problem up to the PR folk to solve." -- ian hickson http://ln.hixie.ch/?start=1023585606&amp;count=1
01:16
<Lachy>
Looks like it's a quote from that blog entry
01:16
<Lachy>
so you *really* were writing about eugenics and nazis :-)
01:18
<Hixie>
yup that's an accurate quote
01:19
<annevk2>
someone posted that on my blog too iirc
01:19
annevk2
let it through though
01:19
<Hixie>
yeah doesn't seem like spam
01:19
<annevk2>
http://annevankesteren.nl/2008/09/re-two-thousand-twenty-two#comment-6629
01:20
<Hixie>
i wonder what the relevance of that quote is
01:21
<Lachy>
considering it was posted twice and whoever posted it didn't use Hixie's email address, I figured it had to be spam of some kind.
01:21
<annevk2>
maybe about letting PR folks solve the problem, just like with 2022?
01:23
<Lachy>
hmm, I'm trying to figure out what I should go do this weekend while I'm on the Gold Coast
01:23
<Hixie>
annevk2: heh
01:23
<Hixie>
Lachy: check e-mail :-)
01:25
Lachy
checks...
01:25
<Lachy>
nothing yet.
01:26
<Lachy>
Hixie, did you send me something, or is that a suggestino for what I should spend my weekend doing?
01:26
<Hixie>
the latter
01:26
<Lachy>
oh, LOL
01:35
<annevk2>
checking e-mail during the weekend is sort of a waste of time I found out
01:35
<annevk2>
there isn't many
01:35
annevk2
suggests organizing parties
01:39
<Hixie>
heh
01:48
<takkaria>
my life as a vegan got much better today: I found a spread that I like, and I found that the local supermarket sell vegan icecream
01:49
<takkaria>
I'm not sure why this warrants mentioning here of all places, but so it goes :)
02:07
<eric_carlson>
Hixie: a minor typo in "4.8.10.8. Seeking"
02:08
<eric_carlson>
step 9 says: "The user agent must queue a task to queue a task to..."
02:08
<eric_carlson>
probably don't need both of those "queue a task"
02:09
<Hixie>
thanks
02:09
<eric_carlson>
sure
02:18
Philip`
discovers that it is quite cold out at 2am, but streetlights make the mist look pretty
02:46
Hixie
tries to get postMessage working in IE in a spec-compliant way
02:48
<othermaciej>
good luck!
02:54
Hixie
tries to get postMessage working in any browser in a spec-compliant way
03:01
<Hixie>
i wish webkit wouldn't complain about favicon.ico files being sent as text/html
03:01
<Hixie>
i didn't ask you to fetch favicon.ico
03:01
<Hixie>
there is no favicon.ico
03:01
<Hixie>
what you got back was a 404 and it had the right mime type
03:01
<othermaciej>
where does it complain?
03:01
<Hixie>
console
03:01
<Hixie>
Resource interpreted as image but transferred with MIME type text/html. favicon.ico
03:02
<othermaciej>
it probably should not complain in the case where it just asked for the default icon url and got back a 404
03:03
<Hixie>
well the funny thing is that if you do send the 404 as an image, it doesn't even use it
03:03
<Hixie>
so the message, not only is it a bug, it's also a lie
03:03
<Hixie>
the resource wasn't interpreted as an image
03:04
<Hixie>
hey how do you inspect the inside of an iframe?
03:05
<othermaciej>
in the latest WebKit you can see the DOM for the iframe under the frame element
03:05
<othermaciej>
or at least I seem to recall that being done
03:06
<Hixie>
doesn't seem to be the case
03:06
<Hixie>
and i believe i'm using last night's nightly
03:07
<othermaciej>
let me check
03:07
<othermaciej>
actually, let me ask the expert
03:08
<othermaciej>
Hixie: hmmm on tivofaq.com I see the document DOM under the <frame> element
03:08
<othermaciej>
(document node not represented, but the doctype and then the <html> element are there)
03:51
<Hixie>
opera doesn't support e.origin?
03:53
<Hixie>
oh it still has e.uri
03:53
<Hixie>
ok well i guess i won't demo opera for this part
07:34
Hixie
finds more bugs in IE's implementation of HTML5 stuff
07:34
<Hixie>
localStorage.getItem(...) return undefined instead of null for unknown stuff
07:41
<Hixie>
this is going well, i think i might pull this off.
07:41
Hixie
is trying to give a presentation about html5 made of purely demos with no future tense anywhere in the presentation
11:14
<Hixie>
http://www.whatwg.org/demos/2008-sept/color/color.html is trippy, especially in webkit
11:14
<Hixie>
(url will be dead in a few minute)
11:24
<annevk2>
cool demo
11:24
annevk2
makes a copy
11:31
<hsivonen>
Did Richard Maher's email contain a technical argument why Web Socket is "bollocks"?
11:34
<Hixie>
no idea
11:34
<Hixie>
nn
11:37
<jruderman>
darn, i missed the demo
13:19
<jgraham>
hsivonen: AFAICT the closest to a technical argument was "it's not unfettered access to sockets"
13:21
<hsivonen>
jgraham: ok. but that's the whole point of the design
13:21
<jgraham>
hsivonen: Possibly explaining that to him would work. Possibly not
13:22
<Philip`>
How does the "existing Socket support from SUN Java, Adobe Flex, and Microsoft Silverlight" handle security?
13:22
<hsivonen>
jgraham: the email didn't make it clear why it's bad to restrict stuff like this
13:22
<Philip`>
hsivonen: It seems obvious that it's bad because it's incompatible with how the rest of the world does sockets
13:23
<jgraham>
Philip`: OTOH he didn't present any use cases that require real sockets rather than WebSockets
13:23
<gsnedders>
jgraham: I want to write an HTTP client.
13:23
<gsnedders>
(that would be kinda awesome to do)
13:24
<jgraham>
gsnedders: that would be kinda stupid
13:24
<Philip`>
jgraham: That's because they're obvious :-)
13:24
<hsivonen>
Philip`: according to Sun's FAQ, it seems that Java applets can open sockets to the host of origin
13:24
<gsnedders>
Then you know exactly how the HTTP client behaves :P
13:24
<hsivonen>
which is *bad* in the context of virtual hosts
13:25
<hsivonen>
I didn't actually test anything
13:25
<Philip`>
("I have an existing socket-based service that does <var>X</var>, and I want to access it directly over the web" for any value of X)
13:25
<Philip`>
s/over the web/from a web browser/
13:26
<jgraham>
Philip`: As you well know there is a problem when X is "sends email" and similar
13:26
<gsnedders>
Time to try re-writing my personal statement
13:26
gsnedders
sighs
13:34
<Philip`>
jgraham: Indeed - it doesn't seem like that general class of use cases is solvable
13:34
<jgraham>
Do any browsers have WebSockets implementations yet?
13:35
jgraham
is composing a brief reply
13:37
<gsnedders>
jgraham: no
13:54
<gsnedders>
Writing a personal statement is ahrd.
13:54
<gsnedders>
*hard
13:57
<gsnedders>
"I am a sixth year pupil at Madras College, currently doing four Advanced Highers: computing, English, mathematics, and physics. I would like to study either computer science or physics because…" — does that seem a sane start?
14:02
<jgraham>
gsnedders: It sounds saner.
14:02
jgraham
can't remember what he put on his personal statement
14:03
<jgraham>
I guess there might be a copy of it somewhere but I have no idea where that would be. Probably an old floppy disk that has long since died
17:27
<jgraham>
gsnedders - well if it's executing circular motion mv^2/r = GMm/r**2 so E_k = GMm/(2r)
17:29
gsnedders
wonders where that is in his notes
17:29
<gsnedders>
(probably answer: somewhere)
17:42
<jgraham>
gsnedders: I would hope its under the heading "circular motion" or somesuch
17:42
<gsnedders>
jgraham: Angular Motion, somewhere, I expect
17:49
<gsnedders>
OK, I just did a rather bad drawing of the gravitational fields of the Earth and Moon
18:28
<Philip`>
jgraham: To hyphenate a single instance of a word, it's easier to write "foo\-bar" to give a hyphenation hint
18:38
<jgraham>
Philip`: I guess in this case there's no harm in having Brehmsstrahlung hyphenated in other places too
18:44
<jgraham>
(but thanks)
19:07
Philip`
sees that Silverlight has its own clientaccesspolicy.xml file to enable socket communication with the host
19:08
<Philip`>
(It supports Flash's crossdomain.xml for some features but not sockets)
19:08
<Philip`>
(according to http://msdn.microsoft.com/en-us/library/cc645032(VS.95).aspx )
19:18
<Hixie>
policy files are such a bad design
19:18
<Hixie>
crossdomain.xml showed that
19:29
<Philip`>
Presumably it didn't show it to a sufficient extent to stop people from thinking it's a good idea
19:41
<Hixie>
i wish there was a way to pause vmware without suspending it (which takes a lot of time)
19:43
<gsnedders>
I wish there was a way to not do homework (which takes a lot of time)
19:44
<gsnedders>
But it does have the use of keeping my mind off what would otherwise worry me
19:52
Hixie
sends a sig 17 to the vmware process
19:52
<Hixie>
that worked
20:07
jgraham
suggests adding the text "semantic in html have largely failed. If you find yourself having an argument about element semantics, it is possible that you should consider instead getting a hobby" to the HTML 5 text
20:08
<Hixie>
heh
20:08
<Hixie>
but this IS my hobby!
20:09
<gsnedders>
jgraham: Can doing homework please _NOT_ be my hobby? *puppy-eyes*
20:12
<jgraham>
Hixie: Sadly it seems to be my hobby too :(
20:14
<Hixie>
holy crap, ie8 beta 2 actually managed to _break_ the rendering of acid3's _reference rendering_
20:14
<jgraham>
But I don't endorse long lilliputian arguments about whether it is proper to use <var> for variables that are constant or just variables that are variable
20:14
<jcranmer>
Hixie: it's so that it can match the reference rendering ;-)
20:15
<Hixie>
it actually makes the reference rendering look even less like the test
20:16
<jgraham>
Impressive application of gun to foot there
20:29
<gsnedders>
jgraham: Ah, it's given as $\frac{1}{2}I\omega^2$
20:30
<gsnedders>
jgraham: (i.e., relative to the moment of inertia)
20:32
<jgraham>
gsnedders: FWIW the result that in a 1/r**2 potential E_k = -E_p/2 is called the virial theroem
20:33
<jgraham>
http://en.wikipedia.org/wiki/Virial_theorem
20:33
gsnedders
is probably too tired to comprehend this
20:36
gsnedders
gives up on the one question he can't do
20:36
gsnedders
hopes he hasn't got any other homework due tomorrow
22:13
<gsnedders>
http://www.flickr.com/photos/gsnedders/2876109667/in/set-72157607417932255/
22:13
<gsnedders>
Kinda impressive.
22:20
<jgraham>
gsnedders: Not as impressive as invisible green hair :)
22:25
<annevk2>
lol: http://webkit.org/blog-files/animation-demo.svg
22:29
<jgraham>
annevk2: Hmm Opera seems to use quite a lot of CPU running that
22:33
<Hixie>
there's irony to using that logo without any script
22:34
<hsivonen>
oh joy. Sun adds crossdomain.xml support to Java applet networking
22:41
<aboodman>
Hixie - any updates on workers?
22:42
<aboodman>
i didn't see any repsonse to that last thread, but maybe in person?
22:47
<Hixie>
aboodman: i'm deep in web forms 2 right now
22:47
<aboodman>
i will ping jonas
22:48
<Hixie>
aboodman: but workers is on the list for the next thing i need to work on after wf2
23:14
<Hixie>
anyone got a video i could use to demo <video>?
23:15
<Hixie>
i tried using something from my media library but i can't get it to work, even in safari
23:17
<hendry>
Hixie: i have lots of video OGGs on http://video.natalian.org/ you're welcome to take
23:18
<hendry>
Hixie: though the test one I use http://video.natalian.org/test/sample.ogg is pretty boring
23:19
<roc>
Chris's page has some
23:21
<Hixie>
cool
23:23
<roc>
http://www.double.co.nz/video_test/
23:34
<annevk2>
http://annevankesteren.nl/2008/new-computer
23:35
<annevk2>
(stole from howcome, not sure where the original is)
23:37
<annevk2>
Hixie, btw, will you reinstate those demos once the presentation has been given?
23:51
<Hixie>
my plan is to write them as i'm giving the presentation
23:52
<takkaria>
where is the presentation?
23:53
<Hixie>
it's an internal google thing
23:53
<Hixie>
annevk2: is <video> in any released version of opera?
23:55
<annevk2>
just experimental builds so far: http://labs.opera.com/downloads/
23:56
<annevk2>
Hixie, well, I meant the one you removed earlier today (regarding <canvas>)
23:57
<Hixie>
right
23:57
<Hixie>
my plan is to write it during the presentation
23:57
<annevk2>
doh
23:57
<Hixie>
the presentation is in less than 24 hours, so they'll be up soon :-P