00:03
<hober>
404
00:36
<Hixie>
http://www.ietf.org/internet-drafts/draft-hixie-thewebsocketprotocol-00.txt
01:33
<maodun>
Maybe this isn't a big deal, but unless I can't read html5lib doesn't seem to conform to the spec here: http://rafb.net/p/3AqRUj30.html . Is this intentional?
01:34
<takkaria>
the spec has probably changed under html5lib in that case
01:35
<maodun>
so the spec is more modern?
01:35
<takkaria>
yeah, either that or there's a bug
01:35
<maodun>
ok, thank you
01:35
<Hixie>
i didn't think that had changed... do you have a test case demonstrating the problem?
01:36
<takkaria>
either that, or the code does do what the spec says, anyway
01:36
<Philip`>
How is html5lib differing from the spec there?
01:37
<Philip`>
(It's a bit different because it outputs the previously-consumed characters directly, rather than unconsuming them, but the effect should be identical)
01:37
<maodun>
Philip - ah, you're right, my mistake. Sorry.
01:38
<Philip`>
(It does that to make sure there's only ever one unconsumed character at a time, for minor efficiency gains)
01:38
<Hixie>
ok, now that's i've split out the protocol to an rfc, i guess teh next step is splitting the api to a w3c draft
01:38
<Hixie>
s/rfc/id/
08:02
<zcorpan___>
hsivonen: http://validator.nu/?doc=http%3A%2F%2Fdret.net%2Fnetdret%2Fdocs%2Fwilde-cacm2008-xml-fever.html
08:02
<zcorpan___>
hsivonen: since us-ascii is to be treated as windows-1252, why is reparsing required?
08:02
<hsivonen>
zcorpan___: hmm. maybe us-ascii has been aliased to windows-1252 after I last reimplemented aliases
08:04
<zcorpan___>
hsivonen: btw, doesn't validator.nu support changing encoding on the fly?
08:08
<hsivonen>
zcorpan: it doesn't.
08:08
<hsivonen>
zcorpan: with natural buffering, it doesn't work
08:08
<hsivonen>
zcorpan: and bookkeeping of when you can switch is pointlessly complex
08:08
<zcorpan>
hsivonen: ok
08:09
<hsivonen>
zcorpan: OTOH, hindering buffering to enable on-the-fly change would suck, too
08:10
<zcorpan>
did it support on the fly encoding change before?
08:13
<hsivonen>
zcorpan: yes, with natural buffering plus bookkeeping
08:13
<hsivonen>
zcorpan: which was pointless, because nearly all the time the bookking indicated that you can no longer switch
08:14
<hsivonen>
zcorpan: consider having non-ASCII in <title> and the decode buffer not only containing meta but the title, too
08:16
<zcorpan>
hsivonen: makes sense. i was just wondering if i was remembering correctly or not :)
09:10
<hsivonen>
did Palm just migrate to WebKit as their app runtime? is the old Palm OS still under it?
09:13
<billyjackass>
hsivonen: yeah, they see to have moved to using WebKit as the Web engine for their new "Palm WebOS", but the old Palm OS is not under it
09:13
<billyjackass>
as far as I can tell
09:13
<hsivonen>
is WebOS Linux plus proprietary user land?
09:14
<billyjackass>
hsivonen: seems so
09:14
<hsivonen>
Wikipedia says Closed source and Linux
09:14
<billyjackass>
but there are not a lot of details
09:14
<billyjackass>
Palm has not developed a Palm operating system of their own for several years
09:14
<billyjackass>
they sold the code to Access several years back
09:15
<billyjackass>
but I don't think that WebOS is based on any of that old code at all
09:15
<hsivonen>
how many of all the companies that have their own port of WebKit contribute to the platform-independent core feature-wise?
09:15
<billyjackass>
hsivonen: very few, I think
09:16
<billyjackass>
in the particular of Palm, I don't think they have contribued any code back to the trunk at all yet
09:16
<billyjackass>
but maybe they will not that it's public
09:17
<hsivonen>
I see
09:17
<billyjackass>
Torch Mobile is an exception
09:17
<hsivonen>
Did Palm just throw away the BeOS code? did they buy it for the staff?
09:17
<billyjackass>
but that's kind of a special case since George Staikos is also one of the main developers of the Qt port of WebKit
09:18
<billyjackass>
I don't know that they did with the BeOS code. That's part of what Access owns now. Not sure what if anything they did with it
09:24
<jgraham>
Philip`: Does that mean that you are resposible for some of the \ line continuation characters in html5lib?
09:25
jgraham
has been stripping those out whenever he encounters them
09:25
<jgraham>
(but I thought anne had put them all in)
09:27
<Philip`>
jgraham: I don't think I've intentionally added any, except when copying existing code that has them
09:28
<Hixie>
http://dev.w3.org/html5/websockets/ http://www.ietf.org/internet-drafts/draft-hixie-thewebsocketprotocol-00.txt
09:28
<Hixie>
both are autogenerated from http://www.whatwg.org/specs/web-apps/current-work/multipage/comms.html#network
09:28
<Hixie>
(which i'll eventually remove, once they're stable and so on)
09:29
<jgraham>
Philip`: OK :) It ould be nice if they disappeared over time :)
09:29
<Philip`>
jgraham: I'm happy to agree with that :-)
09:36
<MikeSmith>
Hixie: cool
09:37
<MikeSmith>
Hixie: btw, do you know why Adam published the content-type-sniffing draft at his own space instead of at the IETF site?
09:38
<Hixie>
same reason http://www.whatwg.org/specs/web-apps/current-work/.ietf-websocket-protocol/draft-hixie-thewebsocketprotocol-01.prepare exists
09:38
<Hixie>
he's still editing it
09:44
<MikeSmith>
OK
10:07
<gsnedders>
Hixie: How are you creating the RFC format doc in the end?
10:08
<gsnedders>
Hixie: "_This section only applies to user agents, not to servers._" — that goes against RFC rules
10:10
<Hixie>
html-[script]->xml-[xml2rfc]->rfc
10:10
<Hixie>
what goes against RFC rules?
10:10
<gsnedders>
Hixie: underlining
10:11
<Hixie>
that's not underlining in the sense they mean in the rules
10:11
<Hixie>
they mean literally overstriking an underbar
10:11
<Hixie>
it's a holdover from the 70s
10:13
gsnedders
is probably just missing something
10:15
gsnedders
can't see how you can overstrike (in what he thinks of as overstriking) in a plain US-ASCII file
10:15
<Hixie>
that's because you're not using a teletype :-)
10:16
<Hixie>
in the olden days, when RFCs were new, computer screens weren't the norm
10:16
<Hixie>
and teletypes would literally print output
10:16
<gsnedders>
Ah, yeah
10:16
<gsnedders>
Teletypes would allow such things, true
10:16
<Hixie>
so you could underline an X by outputting 0x58 0x08 0x5F
10:16
<gsnedders>
Yeah, I can see.
10:16
<Hixie>
that's what's not allowed
10:17
<Hixie>
(because 0x08 isn't allowed)
10:17
<Hixie>
anyway :-)
10:17
<gsnedders>
http://www.rfc-editor.org/rfc-editor/instructions2authors.txt also more generally states, though "No overstriking (or underlining) is allowed."
10:17
<Hixie>
right, that's what they mean though
10:18
<gsnedders>
I guess we can find out for sure upon RFC Ed. review :)
10:19
<Hixie>
i wonder how to handle terms like "task source" in http://dev.w3.org/html5/websockets/
10:19
<Hixie>
link them to WHATWG? list them in an appendix with references to HTML5?
10:19
<Hixie>
if anyone has any bright ideas, let me know
10:21
<Hixie>
nn
10:21
<gsnedders>
Hixie: they defined in HTML 5?
10:22
<Hixie>
yes
10:23
<Hixie>
but i must sleep
11:13
<hsivonen>
Am I the only one who finds it amusing that Shelley Powers complains about multiple venues of discussion in the commenting system of her blog thereby launching yet another venue?
11:15
<MikeSmith>
hsivonen: I suggest we all complain/agree about that on our blogs, and get further discussions going there
11:16
<hsivonen>
also, the notion that the WHATWG list needs to be shut down implies that those who demand it don't trust in the HTML WG list getting enough traction on its own merit
11:19
<Philip`>
Maybe they do trust in that, and just don't think the administrative cost of having multiple lists (and hence having to subscribe to them all, and deciding which one to post to, and failing to reach people who are only on one list) is justified by the benefits of keeping both lists
11:22
<hsivonen>
as it happens, the reasons why the W3C list exists in the first place are the reasons why some people who are on the WHATWG list don't subscribe to the W3C list
11:23
<hsivonen>
1) the patent policy shuts out non-appointed employees of Members and 2) getting input from a wider community gets too much input
11:23
<hsivonen>
although #2 has gotten a lot better lately
11:29
<webben_>
hsivonen: Rather amusingly, 1) basically shuts out employees of ex-members too.
11:30
<MikeSmith>
well, perhaps the best solution to this is to shut down both lists and just have discussions on blogs
11:30
<webben_>
or Twitter ;)
11:30
<MikeSmith>
or on a Wiki
11:30
<MikeSmith>
or multiple Wikis
11:30
<hsivonen>
whoa. Yahoo! is not on http://www.w3.org/Consortium/Member/List when did that happen?
11:30
<MikeSmith>
webben_: now you're talking :)
11:30
<jgraham>
Decentralisation FTW
11:30
<webben_>
hsivonen: And that would be the cause of my last comment ;)
11:30
<MikeSmith>
hsivonen: recently
11:31
jgraham
wonders if there is a plot of #Members against time
11:36
<hsivonen>
I guess Yahoo! is really serious about its need to cut costs.
11:36
<webben_>
AFAIK we weren't actually participating in any WGs.
11:36
<hsivonen>
I get really weird errors from Jing when trying to integrate MikeSmith's patches
11:36
<hsivonen>
"Loaded balanced string"
11:36
<hsivonen>
Datatype library “http://www.w3.org/2001/XMLSchema-datatypes” not recognized.
11:36
<MikeSmith>
I'm good a breaking stuff
11:38
<hsivonen>
hmm. Jing is loading the datatype sample...
11:40
<hsivonen>
ok. the sample datatype was confusing Jing
11:43
<MikeSmith>
hsivonen: btw, have you yet integrated any of your Jing changes into the upstream code?
11:43
<hsivonen>
MikeSmith: onto the branch
11:44
<hsivonen>
MikeSmith: now it's up to James Clark to pick stuff onto the trunk
11:46
<MikeSmith>
OK
11:46
<hsivonen>
MikeSmith: if the trunk gains all the functionality I need, I'll switch to the trunk
11:46
<MikeSmith>
cool
11:46
<MikeSmith>
James seems to have gone quiet again
12:03
<beowulf>
hello, is there an html5 help room, or is this it?
12:04
<MikeSmith>
beowulf: hey there. Long time no hear from you
12:04
<MikeSmith>
this is the right place
12:04
<MikeSmith>
or #html-wg on the W3C server
12:04
<MikeSmith>
or your own blog, if you prefer
12:04
<beowulf>
sweet :)
12:05
<MikeSmith>
if it's a complaint, though, the preference is definitely that you complain on your own blog or someone else's
12:05
<MikeSmith>
or on multiple blogs
12:05
<beowulf>
excellent
12:06
<annevk>
IRC ranting is acceptable too I think
12:06
<MikeSmith>
as long as you link to it from a blog
12:06
<annevk>
and easier than writing a whole post :)
12:06
<annevk>
heh
12:07
<hsivonen>
MikeSmith: linking to IRC rants from a blog is automatically outsourced
12:08
<annevk>
mr last week ^^
12:08
<MikeSmith>
yep, yeay for that. nobody could complain that we aren't collectively doing everything we can to enable them to complain more easily
12:08
<MikeSmith>
one big happy family
12:08
<beowulf>
:)
12:09
<MikeSmith>
beowulf: btw, wonder if you have heard any get details about the WebKit version on the new Palm WebOS thingey
12:09
<MikeSmith>
beowulf: also, if you have something you really need help with or info about, please don't let my nonsense distract you
12:10
<beowulf>
MikeSmith: I have left the world of dotmobi behind, so i'm a little out of touch with the mobile web biznez
12:10
<MikeSmith>
ah, OK
12:11
<MikeSmith>
you landed somewhere else already, or have you become a gentleman of leisure?
12:11
<beowulf>
have landed in a wee web company, 'tis fun
12:12
<MikeSmith>
cool
12:13
<beowulf>
though if i was still with dotmobi i'd be getting paid in euro, not sterling...
12:13
<beowulf>
:(
12:14
<annevk>
ooh, sterling is bad these days
12:14
annevk
is happy Opera now pays him in EUR instead of NOK
12:14
<MikeSmith>
sterling, what is that? Pirate money, like doubloons?
12:16
<beowulf>
doubloons would be good, her majesty's nuggets not so much
12:23
<Lachy>
LOL! I just noticed the poll on http://lastweekinhtml5.blogspot.com/ - "Which WHAT WG teamster is the biggest Smeghead" :-D
12:24
jgraham
notices that zcorpan is never mentioned on last week
12:26
<takkaria>
heh
12:26
<takkaria>
this could turn into Werewolf quicklt
12:41
<annevk>
ah good, my name is wrong, so I'm not on the list
12:43
<beowulf>
i voted you all as smegheads
12:43
<beowulf>
seems only fair
12:43
<Lachy>
annevk, the only thing I see wrong with your name on the list is that "kesteren" has been spelt with a lowercase k.
12:43
<Lachy>
I voted for myself
12:43
<annevk>
Lachy, must be a different person then
12:43
<beowulf>
a narcissistic smeg head
12:43
annevk
wonders why Ben Millard has so many votes
12:43
<Lachy>
I assume Mr Last Week has rigged the vote
12:43
<takkaria>
such accusations
12:43
<annevk>
meanwhile RB has posted something called "HTML4All HTML"
12:43
<annevk>
good times
12:43
<jgraham>
annevk: Where?
12:43
<annevk>
html4all public list?
12:48
<jgraham>
I like this from Josh: "[it's not worth bothering with this] Unless you come up with some really hot ideas that the ivory tower of
12:48
<jgraham>
the WHATWG will absorb, and their operational modus is capricious at best.
12:48
<jgraham>
"
12:48
<jgraham>
Nice to see that we so thoroughly convinced him of the benefits of doing research rather than believing hearsay</sarcasm>
12:48
<zcorpan>
http://simon.html5.org/tools/js/coords-parsing.html
12:49
<zcorpan>
for comparison with ie
12:50
<zcorpan>
dunno how to get the parsed value from other browsers with js
12:52
<zcorpan>
i noticed that some characters aren't "fatal" in ie but are in html5, like linebreaks or ´
12:53
<zcorpan>
there are some characters that ie treat like the spec treats "."
13:00
<zcorpan>
http://software.hixie.ch/utilities/js/live-dom-viewer/?%3C!DOCTYPE%20html%3E%0D%0A%3Cmap%3E%3Carea%20shape%3D%22circle%22%3E%3C%2Fmap%3E%0D%0A%3Cscript%3E%0D%0Avar%20a%20%3D%20document.getElementsByTagName('area')%5B0%5D%3B%0D%0Afor%20(var%20i%20%3D%201%3B%20i%20%3C%200x3ff%3B%20%2B%2Bi)%20%7B%0D%0A%20%20a.coords%20%3D%20'1'%20%2B%20String.fromCharCode(i)%20%2B%20'2%2C3'%3B%0D%0A%20%20w('0x'%20%2B%20i.toString(16)%20%2B%20'%20('%20%2B%20String.fr
13:00
<zcorpan>
omCharCode(i)%20%2B%20')%3A%20'%20%2B%20a.coords)%3B%0D%0A%7D%0D%0A%3C%2Fscript%3E
13:00
<zcorpan>
or http://tinyurl.com/9s7mb3
13:04
<takkaria>
I was wondering why RB was so silent for so long
13:06
<takkaria>
I also really don't understand why he thinks implementors' needs get seen to more than users
13:06
<takkaria>
cos all implementers actaully need is a clear spec to implement
13:06
annevk
wonders what mailing list to use to comment on http://tools.ietf.org/html/draft-hixie-thewebsocketprotocol-00
13:09
<annevk>
http://www.w3.org/TR/2008/WD-jlreq-20081015/ might be useful to figure out Japanese ruby usage
13:09
<Philip`>
annevk: Have you not been paying attention recently? Just comment on a random person's blog
13:10
<annevk>
on a random blog entry?
13:10
<Philip`>
Sure
13:10
<Dashiva>
Comment to your own blog
13:11
<Philip`>
Bonus points if your comment gets deleted as spam, and then you can complain in a year's time that the spec editor just ignored your feedback
13:13
<jgraham>
Commnent on MySpace
13:14
<takkaria>
quite an impressive amount of work RB has done
13:14
<zcorpan>
Philip`: could you maybe look for pages that have something other than u+0020, u+002c, u+002d, u+002e, 0-9, u+003b, A-Z and a-z in a coords attribute?
13:15
annevk
whistles "wot"
13:20
<Philip`>
zcorpan: Looking for simply (?i)coords="[^"]*[^0-9,\x20"-] I find none so far but will leave it a bit longer
13:22
<Philip`>
Ah, there's a few unusual ones
13:22
<Philip`>
Including <area shape="rect" coords="157,5,233,20' href=" href="http://president.psu.edu/gallery/"; alt="Photo Gallery"#">
13:24
<zcorpan>
"157,5,233,20' href=" gives different results in html5 and ie (html5 drops the 20)
13:26
<annevk>
http://intertwingly.net/blog/2009/01/09/Mime-Sniff :)
13:26
<annevk>
sverrej!
13:35
<Philip`>
<area alt="UNESCO" title="UNESCO" coords="rect:0,0,110,90" href="http://www.unesco.org">;
13:44
<Philip`>
zcorpan: http://philip.html5.org/data/coords-with-unusual-chars.txt
13:53
<zcorpan>
Philip`: thanks!
14:01
<Philip`>
(I hope you don't care about non-double-quoted attributes)
14:03
<zcorpan>
not really, although the data contains some noise (e.g. i don't care about 89.0,19.0,89.0...)
14:04
<zcorpan>
"615, 0, 768, 40 " (with tab at the end) parses differently
14:05
<zcorpan>
"438,110,496,1ß2,496,156,438,159" parses the same
14:05
<zcorpan>
"2,20', 87,38'" differently
14:07
<Philip`>
I left that noise in since it was rare enough to be easy to ignore when reading the output, and since I wanted to see how common non-integers were
14:07
<zcorpan>
ok
14:19
<zcorpan>
so 3 pages break with the html5 algorithm, out of 130k
14:20
<zcorpan>
i guess that's acceptable, but otoh those pages work in today's browsers and it's easy to change the algorithm
14:21
<jmb>
what are these pages that break?
14:22
<zcorpan>
http://www.psu.edu/ur/GSpanier/gallery/
14:22
<zcorpan>
http://www.motorsportforbundet.no/
14:22
<zcorpan>
http://www.kipwmi.com/
14:24
hallvors
makes Opera pretend certain <a href="javascript:;"> links in hotmail have "aNewWindow='true'" attributes.
14:38
hallvors
admires the obsession with detail in the "requirements of Japanese typography" document
14:38
<hallvors>
"Japanese composition is based on the full width design of characters, but strictly following the full width based composition sometimes produces an unbalanced appearance. In such exceptional cases, the appearance as a result of the composition must be given higher priority than the full width design principle. When and how to invoke such exceptional procedures has a direct bearing on the quality of the text layout. In other words, it is
14:38
<hallvors>
of how to resolve the conflicts between the principle and the products of it."
14:39
<hallvors>
nice. "resolving conflicts between the principle and the products of it"?!
14:45
<takkaria>
hmm, I want to go read that document now
15:15
zcorpan
looks at http://realtech.burningbird.net/semantic-web/semantic-markup/oh-look-its-not-just-us-semantic-web-dweebs-who-noticed and wonders why a semanticist uses inline svg for decorative background images
15:17
<annevk>
SVG clutter is hip
15:19
annevk
is a bit confused by http://alistapart.com/comments/semanticsinhtml5?page=7#65
15:20
<annevk>
e.g. "I’d predict the chances of its widespread adoption happening anytime soon is pretty much non-existent. You only have to look to XHTML2 for a very near parallel example."
15:20
<hsivonen>
zcorpan: curiously, the bg image doesn't zoom in Firefox. So SVG is not Scalable? :-)
15:21
<hsivonen>
XHTML2 totally hasn't crossed The Chasm
15:21
<hsivonen>
(HTML5 hasn't either)
15:23
<annevk>
But why XHTML2 is a parallel example is beyond me. (Although it seems that the XHTML2 WG does want to make it more in line in some way. E.g. by using the same namespace and by allowing people to use text/html as media type. (Not sure if the XHTML2 WG decided to do that, but various WG members have suggested as much.))
15:24
<webben_>
annevk: I think the parallel is authors won't adopt XHTML2 because it doesn't work in IE; he's saying the new elements mean HTML5 won't work in IE and that therefore authors won't use HTML5 either.
15:25
<webben_>
same with CSS.
15:27
<annevk>
that I can agree with
15:27
<annevk>
although some set of authors have been pushing IE by including scripted workarounds, etc.
15:27
<webben_>
Yep. That set of authors should be fine with HTML5 too.
15:28
<hsivonen>
what we need is authors using CSS features that IE doesn't support and then we need people to oversee how sites looks in Other Browsers
15:28
<webben_>
oversee?
15:28
<webben_>
you mean over people's shoulders?
15:28
<annevk>
and the theory is that if we use attributes at least the things that do not have much behavior but just layout effects could be made to work in IE?
15:28
<hsivonen>
webben_: yes
15:29
<webben_>
annevk: Yes.
15:29
<webben_>
annevk: Well, IE7.
15:29
<hsivonen>
webben_: because putting a "best viewed" conditional comment taunt would be annoying
15:29
<Lachy>
I think I'm going to need to write a detailed response to that ALA article to explain why we're introducing new elements and how we came up with the set we did, if I can find the time to do it.
15:30
<Lachy>
and also explain why using attributes like John proposed is a bad idea
15:30
<hsivonen>
Lachy: and perhaps publish in on ALA to generate a twitter storm in the other direction :-)
15:30
<Lachy>
yep
15:30
<webben_>
since div[section] or div[role=section] or whatever will only work in IE7
15:31
<webben_>
I guess you could use <div section class="section"> for IE6 support.
15:31
<webben_>
though that's getting pretty verbose!
15:31
<Lachy>
Does anyone else find it ironic that he complains about the new elements not working in IE6 without a JavaScript based hack, and then proposes an alternative solution using attributes and attribute selectors that also doesn't work in IE6 without a much more complicated JS hack (like Dean Edward's IE7 script)
15:34
<webben_>
Yes.
15:35
<webben_>
I guess it depends on how you assess the importance of IE6 support.
15:36
<webben_>
well, IE6 -JS support.
15:37
<zcorpan>
in the comment annevk pointed to he says "My argument is that we simply can’t ignore IE6 and backwards compatibility more generally..."
15:38
<annevk>
yeah, which is weird, since it has always been our goal that features are implementable in IE
15:38
<annevk>
though more recently some advanced stuff has been added because the IE Team is back in the game
15:39
<webben_>
anybody know off-hand how IE8 RC1 handles <section> ... can we style it?
15:40
<annevk>
webben_, if you use the createElement hack, yes
15:40
<annevk>
webben_, from what I've heard anyway
15:41
<zcorpan>
yep
15:41
<webben_>
Anyone chatted to them about fixing that so it works without JS?
15:41
<hsivonen>
I take it that Windows 7 beta includes a newer IE than IE8b2?
15:42
<hsivonen>
it seems that the blog where the download is supposed to get announced has melted
15:42
<gsnedders>
hsivonen: yes
15:42
<zcorpan>
hsivonen: i think it includes the beta that, um, "beta testers" got a while back
15:42
<gsnedders>
hsivonen: The link was meant to be on <http://www.microsoft.com/windows/windows-7/default.aspx>;
15:42
<jgraham>
webben_: My informal understanding is that it is too late in the IE8 release cycle for substantial changes at this stage
15:43
<gsnedders>
hsivonen: Nothing there has changed
15:43
<gsnedders>
webben_, jgraham: Yeah, cwilso was saying as much at TPAC w.r.t other things
15:43
<hsivonen>
I wonder how many people will get trojaned when the download quota of Windows 7 beta is up those who didn't get it seek illicit copies
15:44
<gsnedders>
hsivonen: It was meant to be released afternoon PST
15:44
<gsnedders>
hsivonen: So it isn't even the right time yet
15:44
<jgraham>
gsnedders: BTW I think BenMillard was at home but has somewhat offset hours to the rest of the population
15:44
<gsnedders>
jgraham: Sounds possible.
15:47
rubys
can't wait to see how his blog looks like on the new ie8
15:48
<jgraham>
hsivonen: What is "The Chasm"?
15:48
<zcorpan>
rubys: want a screenshot?
15:48
<hsivonen>
jgraham: http://en.wikipedia.org/wiki/Crossing_the_Chasm
15:49
<rubys>
zcorpan: which ie8?
15:49
<zcorpan>
rubys: RC1
15:49
<zcorpan>
rubys: i think it's the same as in windows 7
15:49
<zcorpan>
windows 7 beta
15:49
<rubys>
sure, I'd like to see a screenshot
15:50
<zcorpan>
of http://intertwingly.net/blog/ ?
15:50
gsnedders
was under the impression there wasn't a RC1 yet, that the current builds are just from the RC1 branch
15:50
<rubys>
sure
15:50
<jgraham>
hsivonen: "Adoption of continuous innovations (that do not force a significant change of behavior by the customer) are still best described by the original Technology adoption lifecycle"
15:51
<smedero>
gsnedders: they shipped RC1 to a small group of microsoft partners.
15:51
<jgraham>
That is a HTML5 goal
15:51
<smedero>
not through the more public beta group
15:51
<jgraham>
So if we have succeeded we shouldn't ned to cross the chasm
15:51
<gsnedders>
smedero: My impression was that that was not actually RC1 but just from the RC1 branch
15:51
<zcorpan>
rubys: http://simon.html5.org/dump/sam-ruby-ie8rc1.PNG
15:52
<hsivonen>
jgraham: true. let's see if authors think it's continuous
15:52
<rubys>
that looks much improved. Significant omission being the nav bar.
15:53
<rubys>
can't wait to play with it myself to see if the javascript workaround for unknown elements is still required and to see if I can get the navbar back.
15:53
<hsivonen>
oh. nice. span is now allowed on both col and colgroup
15:53
<gsnedders>
rubys: It is
15:54
<rubys>
odd that only the first <hr> is shown, and that the number of comments isn't right aligned.
15:54
<zcorpan>
rubys: the workaround is still required
15:54
<rubys>
the workaround didn't work for ie8 beta2
15:54
<zcorpan>
it works now
15:55
<Philip`>
The workaround worked for parsing, just not for CSS
15:55
<gsnedders>
rubys: It works now
15:55
<Philip`>
but now it works for CSS too, so that's okay
16:00
<hsivonen>
validator.nu goes down for kernel update
16:01
<zcorpan>
rubys: depending on the user's settings however your page could be as broken as it is in ie7
16:02
<hsivonen>
oh great. the validator.nu process didn't restart properly...
16:02
<rubys>
if my page is valid and you can actually read the text and navigate in IE, I consider that a victory.
16:04
<hsivonen>
it seems that my clever deployment automation broke automatic recovery from reboot...
16:04
<gsnedders>
I had all my content disappear in IE7 adding document.createElement("header"). It's quite funny.
16:05
<hsivonen>
back up
16:06
rubys
takes a step backwards and awaits further instruction
16:12
<gsnedders>
It appears the only way I can get my website to look how I want it without bloating the markup is css3-layout. Yay.
16:14
<Philip`>
Is bloating the stylesheets any better than bloating the markup?
16:16
<hsivonen>
html5.validator.nu goes down for kernel update
16:16
<gsnedders>
Philip`: It means I'm not bloating semantic markup with stuff purely for presentational purposes
16:17
<Philip`>
gsnedders: Why is that good?
16:17
<jgraham>
gsnedders: Who is consuming the semantics of your web page
16:17
<gsnedders>
Nobody :P
16:17
<Philip`>
So you're just wasting your time? :-)
16:17
<hsivonen>
Philip`: it presentation of structure and bloat
16:17
<gsnedders>
Of course :)
16:18
<hsivonen>
html5.validator.nu is taking distressingly long to boot...
16:18
<Philip`>
hsivonen: I can't work out what that sentence means
16:18
<hsivonen>
oops
16:18
<hsivonen>
Philip`: s/presentation/separation/
16:19
<hsivonen>
so many buzzwords
16:19
<Philip`>
Ah
16:23
<jgraham>
gsnedders: pragmatically CSS is harder to understand than HTML so you are better off having complexity in the HTML than in the CSS</devils advocate>
16:25
<jgraham>
gsnedders: See the thing I posted in the gap
16:26
<gsnedders>
jgraham: But CSS is cooler than HTML.
16:26
<jgraham>
Wow, you have a screwed up definition of cool
16:27
<gsnedders>
:P
16:28
<gsnedders>
masochism ftw! :P
16:28
<gsnedders>
(therefore, CSS ftw! :P)
16:33
<hsivonen>
hmm. looks like html5.validator.nu isn't coming back up
16:35
<hsivonen>
still trying some things before changing the DNS to plan B, because reversing that solution will take time
16:35
<hsivonen>
actually, I guess I'd better do that immediately anyway
16:36
<hsivonen>
ok. DNS for html5.validator.nu is being updated to point to the other server
16:38
<hsivonen>
it could be that Gandi is relocating Xen images upon reboot and doing it *really* slowly this time
16:41
<Philip`>
http://www.opera.com/docs/specs/presto211/canvas/index.dml - "<canvas width="150" height="200"> <canvas> element not supported.</canvas>"
16:41
<Philip`>
That's not going to work so well
16:49
<hsivonen>
I can't figure anything more that I could do to bring html5.validator.nu faster than it takes to update DNS :-(
16:49
<hsivonen>
sorry about the inconvenience
16:51
hsivonen
doesn't the feeling of powerlessness when a VM in a datacenter far away breaks
16:51
<hsivonen>
*doesn't like
16:53
<hsivonen>
whoa. now the server is up but the management UI thinks it isn't
16:53
<jgraham>
the figures in http://blog.whatwg.org/supporting-new-elements-in-ie look really ugly in Opera and Firefox
16:57
<hsivonen>
whee! html5.validator.nu is back up
16:57
<hsivonen>
I wonder if the UI is still trying to stop it...
17:06
<hsivonen>
oh. great. now Gandi stopped the VM
17:07
<hsivonen>
whoa. they estimate a boot time of 9 minutes!
17:38
<gsnedders>
is there anyway to float around absolutely positioned element?
17:38
<gsnedders>
s/around/around an/
17:41
<Dashiva>
Use relative positioning instead?
17:42
<gsnedders>
My whole design relies upon it being absolute.
17:42
gsnedders
hacks away at it some more to remove the need
17:44
<hsivonen>
37 minutes to boot the Xen VM at Gandi
17:44
<hsivonen>
previously 1 minute...
17:45
<hsivonen>
anyway, html5.validator.nu is now back
17:45
<hsivonen>
sorry about the inconvenience
17:45
<hsivonen>
deployment of new code has to wait to another day
18:04
<gsnedders>
Damned CSS being designed to be able to be applied iterating over a tree.
18:05
gsnedders
gives in and adds a class attribute
18:15
<gsnedders>
I've just said the stupidest thing ever on #css! That div and span ideally should never be used without attributes that have semantic meaning!
18:23
<gsnedders>
Heh. Now I'm getting ignored. Fun.
18:33
gsnedders
is just about holding back from just trolling now
18:39
<jcranmer>
gsnedders: well, it could be worse
18:39
<jcranmer>
you could be advocating for use of <font>
18:46
<jcranmer>
mmm, I probably shouldn't be discussing stuff on the W3C CSS mailing list here
18:46
<jcranmer>
but Mr. Schmid's proposals are just... o_O
18:46
<gsnedders>
jcranmer: We discuss everything here, though :P
18:47
<jcranmer>
"Don't use rem, use em of root"
18:47
<gsnedders>
no, "em of :root"
18:47
<jcranmer>
because we all know that's wonderful to write a CSS parser for..
18:47
<jcranmer>
oh right, it's a selector
18:48
<jcranmer>
ambiguous parses FTW!
18:48
<gsnedders>
Obviously.
18:51
<jcranmer>
font: 5 em of .water p
18:52
<jcranmer>
is it 5 "em of .water p" or 5 "em of .water" with a font-family of p?
18:53
<gsnedders>
:D
18:54
<gsnedders>
Win7 Beta is available if you manage to not get "Server Not Available"
20:32
<Hixie>
annevk: appropriate mailing list right now is whatwg or public-html; that i-d is literally generated automatically from the html5 source document
21:18
<Hixie>
http://www.ietf.org/internet-drafts/draft-abarth-mime-sniff-00.txt
21:18
<Hixie>
http://www.ietf.org/internet-drafts/draft-hixie-thewebsocketprotocol-01.txt
21:27
<Hixie>
holy crap editing an RFC is expensive
21:27
<Hixie>
https://www.ietf.org/registration/ietf74/ietfreg.py
21:27
<Hixie>
that's crazy talk
21:28
<Hixie>
and people say the _w3c_ is pay-to-play!
21:31
<takkaria>
this is to be an editor?
21:33
<Hixie>
well to form a working group
21:33
<Hixie>
that can only happen at one of these face-to-face meetings apparently
21:35
<takkaria>
nice
22:36
<Hixie>
http://www.w3.org/mid/d375f00f0901091127m362cad08vc544bb9c6a8f4538⊙mgc
22:36
<Hixie>
^ people are _still_ claiming they need the profile="" attribute, despite not being able to explain why at all
22:38
<jcranmer>
the user is always right
22:39
<Dashiva>
Hixie: It's in the spec, of course they need it!
22:39
<Dashiva>
The spec is always right
23:00
<gsnedders>
http://windowsteamblog.com/blogs/windows7/archive/2009/01/09/update-on-windows-7-beta-availability.aspx
23:00
<gsnedders>
(i.e., they've pulled it for now)
23:33
<gsnedders>
Someone give me a hug.
23:33
<smedero>
bear hug or just you know, one of those ever-so-manly "buddy" hugs?
23:34
dave_levin
hugs gsnedders and has no idea why.
23:34
<gsnedders>
smedero: a "he's crying and needs a hug" hug.
23:34
<smedero>
oh, yuck. something happen with your school admissions stuff?
23:34
<gsnedders>
Nope
23:35
<gsnedders>
Just in general.
23:35
<gsnedders>
Ever unfulfilled hope, even when what I hope for I ought to get.
23:35
<gsnedders>
dave_levin: Follow me on Twitter, you'll get the idea :P
23:38
gsnedders
shouldn't turn #whatwg into #gsneddersbitchchannel
23:41
<gsnedders>
Is it bad whenever I see "flicker" I think it is spelt wrong?