00:01
<Hixie>
arg, karl is again implying (not very subtly) that browser vendors are somehow hostile to the desires of web developers
00:01
<jgraham_>
Hixie: where?
00:01
<Hixie>
public-html
00:02
jgraham_
hasn't got that yet
00:03
<jgraham_>
Oh. What an odd email.
00:06
<Dashiva>
The vast browser-wing conspiracy lives on
00:08
<jgraham_>
I wonder why browser makers would want to implement features that "a big majority" of web designers and developers were opposed to.
00:09
<Hixie>
i think the history of whatwg has shown that should that happen, the spec changes pretty fast to match the desires of said developers
00:25
<othermaciej>
I think most browser vendors would like to give web designers what they need and want, on the engine side
00:26
<Hixie>
othermaciej: of course, that's the easiest way to get web developer mindshare, which is a good way to get more market share
00:26
<othermaciej>
certainly Apple has no nefarious plan to force in features that web designers actively are hostile to
00:26
<Hixie>
othermaciej: i'm baffled as to why karl believes otherwise
00:26
<othermaciej>
and I can't imagine what motive we'd have to do so
00:27
<othermaciej>
not only that, but I am not sure there is such a thing as a feature that a majority of web developers hate so much that they don't want to see it implemented
00:27
<Hixie>
yeah, they'd just ignore it if they didn't like it
00:27
<othermaciej>
that they would be actively upset to see it
00:28
<jgraham_>
I assume he feels that browsers are over-represented in the group and other communities are under represented and therefore concludes that we might do something browser friendly that is hostile to those other communities. But I don't think the conclusion follows in general (for the reason mentioned)
00:29
<jruderman>
what feature is he unhappy about?
00:29
<jgraham_>
although in can in specific cases e.g. features that are hard to use in a WYSIWYG editor
00:29
<othermaciej>
browser vendors are highly motivated by economic incentives to do what web developers and end users want
00:29
<gavin>
jruderman: I don't think he's pointed to a specific feature
00:30
<jgraham_>
othermaciej: agreed
00:31
<jgraham_>
I guess s/for the reason mentioned/for the reasons mentioned by othermaciej and Hixie/
00:32
<webben>
I wonder if " some features of the specification" could actually just mean "some aspects of the specification"
00:33
<webben>
(especially given the immediate context was a discussion of whether things should be namespaced, which is more of an implementation detail than a feature)
00:33
<othermaciej>
it is highly unlikely that a majority of web developers want specific features to be in a foreign namespace
00:34
<othermaciej>
it is highly unlikely that a majority of web developers would even know what that means, let alone have an opinion
00:35
<Philip`>
Is a foreign namespace one that's written in German?
00:36
<webben>
othermaciej: Well, I guess the relevant datum is developers already namespacing stuff in text/html content.
00:37
<webben>
They don't really need to know what they're doing to have existing practices.
00:37
<othermaciej>
webben: would that be evidence that web developers want built-in elements to be namespaced, or that web developers don't understand XML namespaces?
00:38
<webben>
Neither.
00:38
<webben>
But I suspect those devs don't want their content to break.
00:39
<webben>
And would not be perturbed by new features that were namespaced.
16:03
gsnedders
replies on ietf-http-wg
16:05
<zcorpan>
pointer?
16:07
<gsnedders>
http://lists.w3.org/Archives/Public/ietf-http-wg/2007OctDec/0233.html
18:38
<Hixie>
wtf happened in public-html
18:38
<Hixie>
we had weeks of nice constructive discussion and suddenly a massive thread about nothing
18:40
<hober>
Yeah, I threw exactly one post in there to try to steer it toward specific changes to spec text: http://lists.w3.org/Archives/Public/public-html/2007Nov/0319.html
18:40
<hober>
I doubt it actually helped though
18:41
<hsivonen>
Hixie: looks like I was wrong to stir the "Web pro syntax" thing
18:41
<hsivonen>
sorry
18:42
<Hixie>
i love the people demanding that we remove the option to omit quotes on attribute values
18:43
<Hixie>
especially given that there are significant groups demanding that we keep that feature
18:46
<Hixie>
i have to say though
18:46
<Hixie>
i don't understand what all that thread was about
18:46
<Hixie>
if people want to write docs, they should do so
18:46
<Hixie>
as anne pointed out, the only people who have written intro docus so far are him and (to a lesser extent) me
18:47
<Hixie>
so if people want more docs, they should stop whining and just write
18:50
<hsivonen>
well, I whined what I *don't* want written *within* the WG and it went downhill from there
18:51
<hsivonen>
I guess I should have waited for someone to write something before whining about the wrong thing getting written
18:51
<Lachy>
I suppose I could just get started and write something
18:55
<Hixie>
Lachy: if you write something we won't have to worry about the wrong thing being written :-)
18:56
<Hixie>
hsivonen: yeah, i think in retrospect when an area has a history if being all talk and no action, whining about the potential action is counterproductive
18:56
<Hixie>
hsivonen: i guess we learn something every day :-)
18:58
<Lachy>
Chaals indicated he's happy for me to work on it, should I check with DanC or Karl if their happy for me to be an editor first?
19:14
<Hixie>
Lachy: i see no reason _not_ to check, but they'd be pretty silly to say no :-)
19:14
<Hixie>
Lachy: anyway you can always publish it elsewhere if the w3c doesn't want it
19:14
<Hixie>
btw karl is no longer our staff contact
19:14
<Hixie>
that's mike smith now
19:17
<MikeSmith>
but karl is the one taking initiative on any work related to separate docs for authors
19:18
<Lachy>
yeah, that's why I thought I should check with him
19:18
<MikeSmith>
as far as I can see, it's not necessary and probably not useful for me to involve myself
19:19
<Lachy>
Do I have write access to the html5 folder in CVS? I have it for webapi and waf, do I need to get it enabled for html5 too?
19:19
<MikeSmith>
Lachy - probably maybe
19:19
<Lachy>
I suppose I could just try it out when I have something to commit. If not, do I just mail sysreq?
19:20
<MikeSmith>
Lachy - please try a commit
19:20
<MikeSmith>
if it doesn't work, I can ping system team to set it up
19:21
<Lachy>
alright, I'll create an html-author folder for this
19:21
<Lachy>
it worked
19:21
<MikeSmith>
Lachy - maybe just htm5/authoring ?
19:23
<Lachy>
I figured html-author would be a good short name for when it eventually goes to /TR/html-author/
19:23
<Lachy>
but then html5-author would probably be better
19:23
Lachy
will fix it
19:27
<MikeSmith>
Lachy - name in the repository doesn't need to be the same as shortname for TR space
19:27
<Lachy>
ok
19:27
<Lachy>
so you want me to rename it to authoring?
19:28
<MikeSmith>
Lachy - it doesn't matter really
19:28
<MikeSmith>
If you've already created it
19:29
<hsivonen>
hmm. canvas hit the html4all list
19:29
<Lachy>
ok, I'll leave it. It's easier than trying to mess around renaming stuff in CVS (I've had bad experience with CVS in the past, I didn't want to risk it :-))
19:32
<Hixie>
as far as i can tell dev.w3.org doesn't have per directory access
19:33
Lachy
notes that Dreamweaver cannot handle opening and editing a 2MB file!
19:34
<hsivonen>
Lachy: using the spec as a stress test case?
19:35
<Lachy>
I was using the current spec as a starting point, and was just going to rewrite the intro and strip out all stuff aimed at implementers
19:35
<Lachy>
I'll have to find a better editor that can handle large files
19:36
<Philip`>
The spec needs reorganisation to be readable, not just stripping things out
19:37
<Lachy>
Philip`, yeah, but it's got all the element definitions in there
19:37
<Hixie>
if all you want to do is make a version of the spec for authors, i'd wait for the spec itself to have that annotated
19:37
<Hixie>
i thought you wanted to write a totally separate document
19:38
<Hixie>
keeping a document in sync with the html5 spec is going to be a gigantic amount of work for you
19:38
<Lachy>
Hixie, it will be more than than that. But it will include the element element definitions (i.e. the contnet models and dom apis) is a starting point
19:39
<Hixie>
ok, but fyi, the content models are very much in flux (i haven't yet looked at much of henri's feedback)
19:39
<Lachy>
I realise that
19:39
<Hixie>
ok
19:40
<Hixie>
just don't want you getting snowed in with make-work :-)
19:40
<Hixie>
afk, going to work
20:12
<virtuelv>
Lachy: on documents that large, the only thing that works is either VI or emacs
20:19
<Lachy>
I don't like vi or emacs, they're too complicated
20:20
<virtuelv>
OpenKomodo also handles some of the stuff I do ok-ish
20:20
<virtuelv>
(but it pretty much chokes on the html5 spec
20:34
<hsivonen>
Lachy: you are on Mac, TextWrangler works
20:39
<Lachy>
I do my editing on the PC cause it has a bigger screen and better keyboard
20:39
<Lachy>
I just use this macbook for IRC, IM clients, email, browsing, etc.
20:40
<Lachy>
I also need to find a nice CVS and SVN client for the mac that gives me a simple GUI like TortoiseSVN/CVS do on Windows
20:42
hsivonen
uses command line and Eclipse
20:42
<Hixie>
command line and emacs baby
20:42
<Hixie>
(there's no other real solution that works under ssh)
20:43
<Lachy>
I like to avoid command line where possible, since it's requires too much typing
20:43
<hsivonen>
I just rediscovered that HTML5 floats and WF2 floats are different
20:43
<Hixie>
that's what scripts are for
20:45
<hsivonen>
hmm. when I'm done with HTML5 (including WF2) datatypes, the only XSD thing left is ID or NCName
20:45
<Lachy>
hmm. OSX doesn't include the cvs command by default
20:45
<hsivonen>
Lachy: it does if you install everything that came in the box
20:45
<hsivonen>
Lachy: Leopard includes svn, too
20:45
<Hixie>
yeah you might have to install the dev tools
20:45
<hsivonen>
(everything includes the Developer Tools)
20:46
<Lachy>
I installed the XCode tools earlier today
20:46
<hsivonen>
I need to come up with nice names for WF2 floats
20:47
<hsivonen>
nice names for code and nice names for UI
20:47
<hsivonen>
floating point number with optional exponent?
20:47
<hsivonen>
float-exp
20:47
<Hixie>
yeah, sorry about the mess there
20:47
<Hixie>
not sure exactly what we should do with those
20:48
<Hixie>
hopefully the forms task force will finish soon and we'll be able to finally merge WF2 with HTML5
20:48
<Hixie>
(hahaha)
20:49
<Hixie>
oh hey, only two more days til the "forms tf internal charter review period" ends
20:50
<hsivonen>
oops email, mimetype-list and charset-list have fallen through the cracks
20:50
<hsivonen>
more entries for the wiki page
20:50
<hsivonen>
more code to write
20:50
<Hixie>
anyone want to lay odds on the review period being extended because everyone forgot to review the charter?
20:50
<Hixie>
i give it about even odds
20:51
<hsivonen>
Hixie: saying it here is tampering with the odds
20:51
<Hixie>
i guess so
20:51
<Hixie>
though i don't know in which direction :-)
20:52
<hsivonen>
I feel tempted to restrict charset-list to charsets that are implemented and sane
20:52
<hsivonen>
but I guess in theory I should scrape the IANA registry
20:53
<hsivonen>
I expect unexpected craziness in the email datatype
20:53
<Hixie>
i think i scoped out the craziness of the e-mail datatype
20:54
<Hixie>
by picking a strategic entry point for you
20:54
<hsivonen>
ah. excellent. thanks
20:55
<hsivonen>
oh. cute. dependency on IDN and, therefore, Punycode
20:56
<Hixie>
ah yeah, can't help you avoid that, sorry.
20:58
<hsivonen>
minor nit: I think you should say "addr-spec production" instead of "addr-spec token"
21:01
<Hixie>
yeah, good point. can you send mail? i'll totally forget it by the time i get back to wf2.
21:01
<hsivonen>
Hixie: email sent
21:01
<Hixie>
ta
21:03
<Lachy>
I haven't seen any comments on public-forms-tf about the charter since the telcon at the end of october. I don't think anyone has reviewed it :-)
21:04
<Hixie>
Lachy: yeah, exactly
21:04
<Hixie>
it's not really clear to me that anyone on the task force actually really cares, to be honest
21:05
<Lachy>
nice to know the task force working as expected :-)
21:05
<Hixie>
no comment
21:08
<hsivonen>
Hixie: is it intentional that 1. is a valid HTML5 float
21:08
<hsivonen>
?
21:08
<hsivonen>
it isn't a valid WF2 float
21:09
<hsivonen>
hmm. .1 is also not a valid WF2 float
21:12
<Hixie>
i doubt any of the differences are intentional
21:12
<Hixie>
they were mostly based on whatever i found when reverse engineering stuff
21:12
<Hixie>
it's likely the wf2 stuff is based on what i thought would be good, vs html5's stuff which is based on legacy implementations
21:13
<hsivonen>
makes sense since WF2 is a submission format
21:13
<Hixie>
yeah
21:13
<hsivonen>
however, for consistency, .1 and 1. could be made non-conforming on the non-WF2 side
21:13
<Hixie>
i'd rather make them conforming on the other side, if they work
21:13
<Hixie>
but yeah
21:13
<Hixie>
(especially .1)
21:13
<hsivonen>
I'm not sure if that would help authors, though
21:14
<hsivonen>
yeah
21:14
<hsivonen>
I guess they are now as good as they are going to get
21:14
<hsivonen>
and authors just have to deal with the discrepancy with leading/trailing zero omission
21:14
<Hixie>
what are? authors? :-)
21:14
<Hixie>
oh, the formats
21:14
<Hixie>
i'd like them the same if we can
21:14
<Hixie>
but i'm not sure we can
21:17
<hsivonen>
the wiki list of microsyntaxes keeps growing
21:18
<hsivonen>
31 already. I'm going to add 2 more
21:20
<hsivonen>
and that's ignoring the inputmode attribute, because magically, it is actually possible to construct with RELAX NG list{}
21:23
<Hixie>
heh
21:25
<gsnedders>
Hixie: http://lists.w3.org/Archives/Public/ietf-http-wg/2007OctDec/0241.html — more mailing list fun
21:25
<hsivonen>
in due course, I should write a by-reference spec for these 33 microsyntaxes
21:26
<Hixie>
gsnedders: actually it would make a lot of sense for ABNF to support namespaces
21:26
<Hixie>
gsnedders: since it's a programming format
21:26
<gsnedders>
hsivonen: if you want oddities in the microsyntaxes in HTML 5, take a look at my emails on public-html (just search the subject for something like detailed spec review of common microsyntaxes)
21:27
<gsnedders>
Hixie: agreed
21:27
<Hixie>
the biggest problem with ABNF is that it is REALLY freaking hard to write ABNF in a way that defines error handling
21:27
<Hixie>
that's the main reason i don't use it much
21:27
<hsivonen>
gsnedders: I don't want oddities. I just think that the RELAX NG datatype library for HTML5 should have a spec
21:27
<Hixie>
(i used it for the event-source stuff, and even there, where i was TRYING to define error handling, i still made mistakes, and ended up having to duplicate the work in prose anyway)
21:28
<hsivonen>
precise enough to let someone write an interoperable implementation in another programming language without looking at my code
21:28
<Hixie>
i've found some sites that use an "adx.js" library
21:28
<Hixie>
anyone know anything about it?
21:28
<gsnedders>
Hixie: in reply to the one on the same list yesterday regarding error handling, I got told basically that I could do it in another spec, and they wouldn't do it
21:28
<Hixie>
gsnedders: cool, a license to write a replacement http spec from the http group
21:28
<Hixie>
gsnedders: sounds like a plan!
21:29
<hsivonen>
RFC 26165
21:29
<gsnedders>
RFC2616bisbis
21:30
<Hixie>
oh maybe this is openads
21:31
<hsivonen>
do mailto URL allow something crazier than what WF2 allows for the address?
22:17
<hsivonen>
bah. mediawiki does not like <samp>
22:35
<Hixie>
http://www.losingfight.com/blog/2006/08/11/the-sordid-tale-of-mm_menufw_menujs/ is funny
22:45
<Dashiva>
I've been in that "expert" position way too often. Who am I going to hate now? :P
22:47
<Hixie>
hehe
22:47
<Hixie>
anyone able to work out what tmv11.js is from?
22:50
<hsivonen>
Hixie: are you mining the most common scripts for declarative features to add?
22:50
<Hixie>
i'm mining the most common scripts, yeah
22:50
<Hixie>
for various reasons, one of which is finding out what is needed for html5
22:52
<Hixie>
it's surprising how many scripts have pretty unique filenames
22:52
<Hixie>
(a strong argument against namespacing being needed, imho)
22:53
<Hixie>
and that's in a space with no need for unique names, given that we have domains and so forth to disambiguate
22:54
<Lachy>
I've made a start on the document http://lachy.id.au/temp/html5-authors.html
22:56
<Hixie>
"and un XHTML examples"
22:56
<Hixie>
s/un/in/
22:56
<Lachy>
fixed in local copy
22:57
<Hixie>
you can probably cut a lot out of the Terminology section
22:57
<Lachy>
yeah, I know. I cut out a bit already
22:57
<Lachy>
going to rewrite some of it to be more author-friendly
22:58
<Hixie>
in fact you should probably assume that nobody will read that section, they'll just jump to the subsection they care about, and thus the language should be unambiguous even without the terminology section
22:58
<Lachy>
ok, good point
22:58
<Hixie>
you have the advantage that you don't need it to be perfectly defined, since this isn't normative
22:58
<Lachy>
ok
23:18
<Hixie>
wtf is this tmv11.js thing
23:18
<Hixie>
i can find lots of sites using tmv11.js, but they're all different
23:18
<Hixie>
yet the name seems unlikely to be coincidentally chosen by multiple different groups
23:18
<Hixie>
http://www.postimees.ee/lib/tmv11.js is one
23:19
<Hixie>
i wanna see avenue q again.
23:19
<Hixie>
er, wrong channel.
23:19
<Hixie>
http://www.diskusjon.no/tmv11.js
23:19
<hsivonen>
looks like a statistics tracker script
23:20
<Hixie>
yeah but normally i can track these down to someone who made it
23:20
<Hixie>
oh hey they both point to statistik-gallup.net
23:20
<Hixie>
'course that returns a 404
23:22
<Dashiva>
To the web archive-mobile
23:23
<Dashiva>
Or not, robots.txt'd
23:24
<Hixie>
i'll just call it the statistik-gallup.net tracking script
23:43
<Dashiva>
Heh, I forgot all the heading elements in that HTML element quiz
23:45
<Hixie>
i wanna see the results for that quiz
23:45
<Hixie>
which elements do people remember?
23:46
<Dashiva>
Well, even though I forgot option, I still remember to use it :)
23:49
<Hixie>
30 seconds left
23:49
<Hixie>
18 elements remain
23:50
<Hixie>
hmmm
23:50
<Philip`>
They should make it like The Typing of the Dead, so you have to type in HTML element names to fend off hordes of zombies
23:50
<Hixie>
i forgot BASE, BASEFONT, BDO, DFN, DIR, KBD, LABEL, MENU, NOFRAMES, NOSCRIPT, OPTGROUP, PRE, Q, SAMP, SUB, SUP, TITLE
23:51
<Dashiva>
I got some of the obscure ones like bdo just because I remember them as obscure :)
23:51
<Hixie>
heh
23:51
<Hixie>
i'm just bad at remembering things like this
23:51
<Hixie>
i'm amazed i got 74
23:52
<Hixie>
wtf is http://cpro.baidu.com/cpro/ui/cp.js
23:55
<Dashiva>
I'd guess it's for embedding flash in an iframe based on that big function
23:57
<Dashiva>
Assembles lots of vars, then pushes them as query string to a PHP script used as src for the iframe...
23:59
<Philip`>
The variables like cpro_client and cpro_h look like inputs the script, so presumably they're set before this script is called
23:59
<Philip`>
*inputs to the script