00:05
<TabAtkins>
Dammit, <fieldset>. Quit having special arcane rendering rules!
00:05
TabAtkins
will be very happy when <legend> and <fieldset> stop being magic.
00:06
<cying>
http://www.satine.org/archives/2009/10/16/wuh-working-group/
00:07
<TabAtkins>
Hehe.
00:20
<Dashiva>
I agree with TabAtkins
00:20
<Dashiva>
Acronyms are much better when pronounced as words
00:22
<TabAtkins>
/high-five
01:35
<ilhan>
Is this allowed in datime? 2009-06-03T-07:00 meaning time zone is "-7"?
01:38
<Lachy>
ilhan, no. You need at least an hour and minute for the time
01:39
<Lachy>
so you could do 2009-06-03T00:00-07:00
01:39
<ilhan>
ok, thanks a lot
04:51
<TabAtkins>
All right, McEvoy is on ignore now. >_<
06:35
<Hixie>
does this play for anyone in the uk? http://www.channel4.com/programmes/bremner-bird-and-fortune/4od#2973181
07:34
<annevk42>
w3.org down?
08:16
<jgraham>
annevk42: WFM now
08:16
<annevk42>
yup
08:16
<annevk42>
ta
09:54
annevk42
is in the train to Stockholm
10:33
<Hixie>
really? <textarea> doesn't need magic?
10:33
Hixie
is skeptical
10:49
<zcorpan_>
i guess we should research all the (r)cdata elements using the current parsing rules and look at pages that contain </foo where foo is the element's tag name
11:03
gsnedders
finally emails jgraham
11:16
<gsnedders>
http://code.google.com/p/html5lib/source/detail?r=937a421712e2edc2dbf498040f5ad3d6e13d395b
11:17
<zcorpan_>
the pronunciation discussion makes me think of http://www.youtube.com/watch?v=xqpd7WcBmjM
12:18
<gsnedders>
Down to one test failure in html5lib!
12:18
<Philip`>
Delete the test and then you'll be down to zero!
12:20
<jgraham>
gsnedders: That last checkin totally needs tests
12:20
<jgraham>
I mean apart from the one test we already have
12:22
gsnedders
finds he's broken it in another way
12:22
<jgraham>
Since the one existing test didn't catch whatever issues previously existed
12:22
<gsnedders>
I've now got the html element always in no namespace instead of always in it.
12:23
gsnedders
swears that didn't happen when he last ran he test script
12:23
<jgraham>
gsnedders: I don't remember what code there already is, but does it work correctly with things that require you to look for all the elements in certain categories
12:23
<jgraham>
Since those lists use (namespace, name) pairs
12:24
gsnedders
grumbles at the html element being special cased all over the place
12:25
<jgraham>
(so you have to check that the .nameTuple property always returns the right thing regardless of whether the element is actually in a namespace or not
12:25
<jgraham>
)
12:27
gsnedders
will get it working today, dammit
12:29
<jgraham>
Totally different question: anybody have any good advice about how to structure a socket-based program that may get multiple incoming messages in a go or multiple outgoing messages wih no incoming message
12:29
<gsnedders>
(That's the I-will-not-lose-against-jgraham's-code statement.)
12:31
<jgraham>
(the problem being that afaik while True: self.socket.recieve(); self.send_queued_messages(); won't ever process the queued messages unless it recieves a message first)
12:31
<jgraham>
(assuming the socket is blocking and so on)
12:31
<Philip`>
jgraham: Use Twisted instead of synchronous sockets?
12:32
<jgraham>
(and I'm not horribly misunderstanding the problem space)
12:32
<jgraham>
(which is possible since I have never written anything more complex than an echoing server with sockets)
12:32
<Philip`>
jgraham: Use 'accept' (with a timeout) to detect whether there's anything to receive, before doing a blocking receive?
12:33
<jgraham>
Philip`: I'm avoiding twisted since this is partially just a learning excercise
12:33
<jgraham>
maybe accept will work
12:34
jgraham
guesses he could also try using non-blocking sockets and select
12:34
<Philip`>
Oh
12:34
<Philip`>
I meant select, not accept
12:34
<Philip`>
I think
12:35
<Philip`>
Ah, yes, I did mean select
12:35
<Philip`>
You shouldn't need non-blocking sockets, because select will let you detect and avoid the situations that would block
12:35
<jgraham>
True
12:36
<Philip`>
unless I'm wrong
12:37
<Philip`>
But if it's a learning exercise, you ought to learn about Twisted instead of raw sockets :-)
12:38
<jgraham>
Philip`: Doesn't help if the goal is to learn about raw sockets :)
12:38
<jgraham>
(I would also like to learn bout twisted but maybe a different day)
12:44
<mikekelly>
I think there's going to be over-lap between this conneg thread and the resource thread
12:44
<mikekelly>
hopefully the concept of resources/representations is made clearer by the example I gave involving cache invalidation
12:45
<mikekelly>
Hixie: did that help explain the distinction?
12:46
<mikekelly>
and/or the practical benefits of that constraint for intermediaries
13:59
<lazni>
is this an appropriate use of <br>? http://lazny.tang.la/jbo/cukta/jbobaucku/jbobaucku.html
13:59
<lazni>
bible-like sentence numbering
14:00
<webben>
I would suggest span to wrap your sentences rather than line break
14:01
<lazni>
I wrap the sentence separator to easily insert it with a yas snippet, <br><span>.i</span>
14:01
<webben>
if people are supposed to refer to the numbers, I probably wouldn't use CSS.
14:02
<lazni>
so that I can write sentences more or less normally
14:02
<lazni>
good point
14:02
<webben>
i think that's an example of not using good enough authoring tools and letting bad authoring process dictate your deliverable.
15:33
<jgraham>
gsnedders: 18:15
15:33
<gsnedders>
jgraham: When you leave or when you get here?
15:34
<jgraham>
gsnedders: When we pick you up
15:34
<gsnedders>
jgraham: Where I suggested?
15:34
<jgraham>
Where did you suggest?
15:35
<gsnedders>
jgraham: Djurgårdsgatan
15:36
<jgraham>
gsnedders: How about the dorner of there and Barnhemsgatan?
15:36
<jgraham>
*corner
15:57
gsnedders
adds, as promised, "No, you're just high." to his about page
16:42
<Philip`>
gsnedders: You appear to keep accidentally modifying html5lib without adding tests for the changed behaviour :-)
16:43
<gsnedders>
Philip`: I'm not sure how to get all the tests checking the namespaces on the created nodes
16:44
<Philip`>
gsnedders: Maybe it would be adequate to have just a .py file doing some custom tests, rather than going through the testdata/*.dat files
16:45
<gsnedders>
Philip`: The reason why I made another commit was because my prior one had broken stuff with namespacing both enabled and disabled, this seems to work
17:12
<mikekelly>
why do I get the impression this Smylers guy isn't actually interested in considering any of the points I make
17:12
<mikekelly>
he keeps asking for stuff I've already provided
17:20
<webben>
mikekelly: Stuff like what?
18:07
<eighty4>
would it be semanticly ok to use both a <header> for the head of the page and a <header> in every <article>, containing meta info about a blog post
18:56
<webben>
eighty4: if the "meta info" constitutes a "group of introductory or navigational aids", then yes.
18:58
<eighty4>
webben: h* for the post, date, comment count and permalink to the post.
18:58
<webben>
sounds okay to me.
19:00
<eighty4>
that was my thought, but always good to check here :)
22:47
<cardona507>
I am trying to get <video> to work on firefox and my server doesn't support .ogv - I want to change this with an .htaccess file but am having trouble - this is what i have in my .htaccess file: AddType video/ogg .ogv - I am new to .htaccess - can someone help out? I have the .htaccess file in the same directory as the video and index.html
22:51
<cardona507>
/join #web
23:26
<eighty4>
cardona507: sounds more like a apache question
23:26
<cardona507>
eighty4 - I got it working
23:28
<cardona507>
is there a solution to getting html5 pages that have the chrome frame meta tag to validate?
23:28
<eighty4>
nice
23:29
<cardona507>
I was told to serve up the file with this mime type application/chromeframe - but that makes firefox want to download the file