10:05
<annevk>
yay
10:05
<Charl>
ok this time the fault was my server, not whatbo
10:05
<Charl>
not the logbot system
10:06
<krijnh>
MS still didn't join the WG?
10:06
<annevk>
nope
10:09
<krijnh>
Are they going to?
10:10
<annevk>
well, one of the Microsofties is co-chairing the group
10:11
<annevk>
I assume Chris is fighting some internal lawyer to get him (and others) in
10:12
<krijnh>
Hmm
10:16
<karlUshi>
annevk: ?
10:17
<annevk>
karlUshi, as logbot for this channel
10:19
<Charl>
i obviously missed the discussion earlier on
10:19
<Charl>
but we really need to get a better host for the logbot
10:19
<karlUshi>
annevk: http://dev.w3.org/cvsweb/2001/sw/logger/
10:20
<Charl>
at the moment whatbot is running on a server sitting beneath my desk
10:21
<annevk>
well, hosting it too
10:21
<annevk>
if I could host a logbot I would've done so
10:21
<annevk>
but my host doesn't allow it :(
10:21
<annevk>
Charl :)
10:21
<krijnh>
What's needed for it to host it?
10:21
<annevk>
krijnh, a server that allows IRC connections I suppose
10:21
<krijnh>
Makes sense ;]
10:21
<krijnh>
I'll ask my host
10:22
<Charl>
you see the problem is this
10:22
<Charl>
one can host it anywhere
10:22
<Charl>
but you have some funny PHP hack
10:22
<Charl>
that you need to run on some webserver
10:22
<Charl>
together with the logs
10:22
<krijnh>
Ah
10:22
<karlUshi>
what's wrond with whatbot?
10:22
<Charl>
in other words, your host needs to have a public static IP address
10:22
<krijnh>
Charl: we've hacked php for your site as well ;)
10:22
<karlUshi>
s/wrond/wrong/
10:23
<annevk>
karlUshi, unstable
10:23
<Charl>
krijnh: that original cms i had going was a hack and nothing but a hack :)
10:23
<krijnh>
Charl: You mean wordpress? :)
10:23
<krijnh>
With the mysqli hack
10:23
<Charl>
krijnh: oh no that was before i even installed WP :)
10:24
<krijnh>
Ah
10:24
<Charl>
krijnh: now i am pwned because there is an update to the WP DB file
10:24
<Charl>
so hacking all over agaIN
10:24
<krijnh>
You're using wp 2 now right?
10:24
<Charl>
yeah 2.0 still
10:24
<krijnh>
Which reminds me, I still have to send you a bill
10:24
<krijnh>
;)
10:24
<karlUshi>
why guys you haven't tried chump?
10:25
<karlUshi>
http://usefulinc.com/chump/
10:25
<karlUshi>
example of use
10:25
<karlUshi>
http://swig.xmlhack.com/
10:25
<Charl>
krijnh: cool bill me any time, i'll pay you in south african rands, they're not worth anything in europe anyway :)
10:26
<krijnh>
Hehe
10:42
<krijnh>
Charl: How is Fidelis going btw?
10:43
<karlUshi>
I'm off for today.
10:44
<Charl>
krijnh: sorry was away from the keyboard quickly
10:44
<Charl>
krijnh: am busy getting started on a very simple blogging system
10:44
<Charl>
krijnh: can't stand WP any longer
10:45
<krijnh>
(Use QMS)
10:45
krijnh
(n=krijnhoe⊙kxn) has left #whatwg
10:45
<krijnh>
Whoops
10:54
annevk
(n=annevk⊙8972) Quit (Read error: 110 (Connection timed out))
10:55
<krijnh>
http://blog.whatwg.org/html5lib-09-released#comment-2345
11:06
<annevk>
seems like we might get Animated PNG soonish
11:06
<annevk>
in Mozilla that is
11:06
<annevk>
otoh, that has been going on for some time now as well...
11:38
<Lachy>
krijnh, thanks, spam is removed
11:38
<krijnh>
Np
11:39
<krijnh>
Lachy: Are you still working on Fidelis?
11:39
<Lachy>
when I have time
12:05
Charl
(n=charlvn⊙nmcz) Quit ("Leaving")
12:05
annevk
wonders if people considered that Hixie works for the company that does Youtube
12:06
<krijnh>
I hadn't
12:07
Lachy
wonders if Hixie's actually had a chance to discuss <video> with the people that work on youtube
12:08
<Hemebond>
<video>?
12:08
<Lachy>
Hemebond, yes
12:08
<Hemebond>
Surely there's not to be an element <video>...
12:09
<annevk>
surely there is
12:09
<Lachy>
http://whatwg.org/specs/web-apps/current-work/#video
12:09
<Hemebond>
*saigh*
12:09
<Hemebond>
er
12:09
<Hemebond>
*sigh*
12:10
<annevk>
see topic
12:10
<krijnh>
annevk: I only considered you're working for the company that's already testing with <video> :)
12:10
<Lachy>
why?
12:10
<annevk>
s/testing/playing/
12:10
<Hemebond>
"Please leave your sense of logic at the door, thanks!"
12:10
<krijnh>
annevk: Same to me
12:10
<Hemebond>
That's what I forgot to do. I only hung up my coat.
12:10
<annevk>
krijnh, fair enough
12:11
<Lachy>
Hemebond, why don't you like the idea of <video>
12:11
<Hemebond>
Because it's a step backwards.
12:11
<Lachy>
how so?
12:11
<annevk>
Hemebond, you did read the proposal, did you?
12:11
<Hemebond>
I would have liked to see more generalisation.
12:11
<Hemebond>
No.
12:11
<Hemebond>
Wait a second...
12:11
<Hemebond>
Who are you Anne?
12:12
<Hemebond>
Not http://annevankesteren.nl/2007/03/public-html ...
12:12
<Lachy>
<video> offers a specific a API for working with video using script
12:12
<annevk>
Hemebond, http://www.google.com/search?q=anne
12:12
<Hemebond>
Heh. It was your latest post that led me here.
12:12
<annevk>
Hemebond, yes, I wrote that
12:12
<annevk>
cool
12:12
<annevk>
welcome
12:13
<Hemebond>
Thanks.
12:13
<Lachy>
Hemebond, if you don't like video, you might not like the idea of <audio> either, which is currently being discussed :-)
12:14
<Hemebond>
Heh. Doesn't appeal to me, no.
12:14
<krijnh>
I think there's more use for <video> than for <audio>
12:15
<Lachy>
I think we need <audio> or at least improve the Audio() API
12:15
<krijnh>
<video role="just-an-audio-wrapper" src="music.mp3">
12:16
<Lachy>
there is no role attr in HTML5
12:16
<krijnh>
There's kidding in my line though
12:16
<Lachy>
<audio> would be cool if browsers could provide their own UI or even visualisations
12:16
<virtuelv>
Hemebond: why exactly don't you like <video>?
12:17
<Hemebond>
I prefer generalisation.
12:17
<Hemebond>
<object>
12:18
<annevk>
why is generalization good?
12:18
<virtuelv>
Hemebond: besides the API complexity annevk has pointed at, there is one more issue: author simplicity
12:18
<annevk>
i mean, "I prefer <tag>"
12:18
<Hemebond>
Heh
12:18
<annevk>
pretty meaningless
12:19
<virtuelv>
you can (mostly) use <object> in place of <img> today, and have richer fallbacks, or multiple fallbacks, yet virtually noone uses it
12:19
<Hemebond>
I do.
12:19
<Hemebond>
I love object.
12:21
<annevk>
that's about the only advantage of <object> over <img>
12:21
<annevk>
and generally you don't need richer fallback so it's not much of a problem
12:22
<virtuelv>
annevk: my point exactly
12:23
<Hemebond>
Wasn't the whole point of <object> to get away from the "tag for everything" problem?
12:23
<Hemebond>
Or "issue"?
12:24
<annevk>
I think so
12:24
<annevk>
since then we found out that it doesn't work
12:24
<Hemebond>
It doesn't?
12:24
<annevk>
too complex for implementors/authors, too complex to make sophisticated APIs for specialized content
12:25
<krijnh>
People can still use object of course
12:25
<annevk>
not really semantic
12:25
<krijnh>
If they want to use a not-so-open format and a browser plugin
12:25
<annevk>
yeah, although I think <embed> is prolly more compatible for that
12:25
<krijnh>
Probably
12:27
<krijnh>
<object> is for external plugins (basically), <img>, <video> and <audio> are cross platform, cross UA elements
12:29
<annevk>
<object> is also ok for SVG, and HTML documents although I suppose you might as well use <iframe>
12:29
<Lachy>
I assume <video> doesn't have to mean native support, does it? Surely whether or not some format is supported natively or with a plugin is just an implementation detail?
12:29
<annevk>
Lachy, I think that's the plan although not everyone agrees
12:29
<krijnh>
Hmm, then I don't think <video> makes sense
12:29
<annevk>
howcome for instance, thinks it should imply native support
12:29
<krijnh>
Or are the methods just passed on to the plugin?
12:29
<annevk>
krijnh, why? It's about the API and semantics, not about content
12:30
<Hemebond>
Couldn't semantics be implied by the content-type?
12:30
<annevk>
Hemebond, too late imo
12:30
<Lachy>
as long as the browser can pass it's API calls onto the plugin, there shouldn't be a problem
12:30
<Hemebond>
Too late?
12:30
<annevk>
you want the semantics to be clear before fetching data
12:31
<Hemebond>
What about the type attribute?
12:31
<Lachy>
though, if the system as a codec for some format, then the browser should just be able to use that and run it natively
12:31
<krijnh>
I agree
12:31
<annevk>
Hemebond, that's only relevant when data= isn't set
12:31
<annevk>
Hemebond, if data= is set it's just ignored
12:31
<Hemebond>
Oh is it? Hmm, I've always set both.
12:31
<krijnh>
I think you'll get the same issues as with Real, QuickTime, etc else
12:32
<Hemebond>
Oh I see. HTML 5 has that.
12:32
<annevk>
Hemebond, browsers too
12:32
<annevk>
Hemebond, you're allowed to set both though
12:32
<Hemebond>
Oh.
12:32
<annevk>
topic ;)
12:32
<Lachy>
the major problem I have with making it only native support is that most digital cameras available today only capture in patented formats, so that would make <video> useless to all those who don't know how/want to convert
12:33
<Hemebond>
Topic? Is there a specific topic I should be keeping to?
12:33
<annevk>
Hemebond, don't assume things are logical
12:33
<Lachy>
Hemebond, no.
12:33
<Hemebond>
Oh I see.
12:33
<Hemebond>
But I want logiiiiiiic *whine*
12:33
<Lachy>
converstaions often get chaotic in here, and I often just say random things
12:33
<krijnh>
Same here :)
12:33
<krijnh>
And half of what I say makes no sense anyway
12:34
<Hemebond>
Are you a spec?
12:34
<krijnh>
Who?
12:34
<Hemebond>
NVM
12:36
<Hemebond>
I guess I'll stick to learning XHTML2
12:36
<krijnh>
I think the openness of the web is more expressed in <videmus> than in <video> btw :p
12:37
<annevk>
Hemebond, XHTML2? And use it where?
12:38
<Hemebond>
I dunno.
12:38
<Hemebond>
I just like it, is all.
12:38
<Hemebond>
I could use XSL to tranform it to regular XHTML1.1
12:39
<krijnh>
Client side XSL?
12:39
<mpt>
Esperanto!
12:39
<Hemebond>
Could do.
12:39
<krijnh>
Why would you want that?
12:39
<Hemebond>
*shrug*
12:40
<Hemebond>
I don't make websites for a living anymore.
12:40
<Lachy>
what do you do for a living?
12:40
<krijnh>
Are you going to again? :)
12:41
<Hemebond>
krijnh: I spend a lot of time reading about it. Playing with it. But then I think of the customers, the incompatibilities, IE, etc...
12:41
<Hemebond>
And I'm not sure I could be bothered with it.
12:41
<Hemebond>
Lachy: Network technician.
12:43
<krijnh>
Hemebond: Then why stick to learning xhtml2?
12:43
<Hemebond>
I think it's a nice evolution of XHTML
12:44
<Hemebond>
And HTML in general.
12:44
<Lachy>
trust us, it's not
12:44
<Lachy>
it has too many flaws
12:44
<Hemebond>
It does?
12:44
<annevk>
So which parts exactly do you like?
12:44
<Hemebond>
I admit I've only looked at it breifly.
12:45
<Hemebond>
I like the idea of <section><h>
12:45
<annevk>
I mean, we tried to incorperate all good ideas into HTML5
12:45
<annevk>
That'd be one of those ideas :)
12:45
<Lachy>
e.g. href on any element: browser vendors have said that's very difficult to implement
12:45
<krijnh>
:)
12:45
<Lachy>
same with src="", and many other global attributes
12:46
<krijnh>
That's the generalisation part Hemebond likes, I think
12:46
<Lachy>
the spec itself leaves many things totally undefined
12:46
ROBOd
(n=robod⊙8321) Quit ("http://www.robodesign.ro";)
12:47
<Lachy>
from an authoring perspective, global attributes look nice on the surface. There was a time when I thought the idea was great, till I looked at the reality of the situation
12:47
ravenn
(n=ravenn⊙2dina) has left #whatwg
12:48
<krijnh>
Is there a thread about the alternatives?
12:48
<krijnh>
Like, not href on every element, but on some?
12:48
<krijnh>
<img href>
12:49
<Lachy>
still a problem with backwards compat
12:49
<krijnh>
True
12:50
<annevk>
note that <a> is not just href=
12:50
<annevk>
<a> is quite complex
12:50
<annevk>
the proposal effectively makes every element that much more complex
12:50
<annevk>
for not that much benefit (besides the fact that it breaks backwards compatibility)
12:50
<krijnh>
And every piece of software which checks "links" as well
12:50
<Lachy>
yes, if href goes on any element, we'd also need target, ping, rel, media ,hreflang and type
12:51
<annevk>
and the API
12:51
<annevk>
and we need to define what happens for elements which now have multiple activation behaviors
12:51
<annevk>
such as <input>
12:51
<Lachy>
and type="" would clash with other uses of type on other elements
12:51
<annevk>
i should blog about this too i think
12:52
<krijnh>
You should :)
12:52
<Lachy>
I should blog about something too. it's been about a month since I've posted :-)
12:52
<krijnh>
And say that <a> should just allow block level elements
12:52
<Lachy>
I even forgot to post on my b'day!
12:52
<krijnh>
Which solves the problem mostly, I guess
12:52
<krijnh>
And is backwards compatible
12:53
Lachy
has a draft post about the problems with Outlook to post soon, plus some more on XBL
12:53
annevk
wants an impl of XBL
12:54
<Hemebond>
Mozilla not enough?
12:54
<krijnh>
Lachy: You were too busy licking Tim ;)
12:54
<annevk>
it's like a new box of tricks
12:54
<annevk>
Hemebond, I meant XBL2
12:54
<Hemebond>
Ah.
12:54
<Lachy>
XBL2, Moz only has XBL1
12:54
<Lachy>
did someone post a photo of me on timwouldlickti?
12:55
<Lachy>
the other Lachlan's been licked so many times, I feel like I'm missing out ;-)
12:59
<krijnh>
Aaw, poor you :)
13:00
krijnh
(n=krijnhoe⊙kxn) Quit ()
13:04
virtuelv
(n=virtuelv⊙poc) Quit ("Leaving")
13:42
<annevk>
hmm, do implementations also check <?xml encoding="x"?> for character encoding sniffing in text/html?
13:43
<Lachy>
I don't think so
13:48
<annevk>
according to a WebKit dev they do
13:48
<Lachy>
maybe web kit does, but I'm fairly sure IE and FF don't
13:49
Lachy
makes a test case...
13:49
<annevk>
IE is what matters i suppose
13:49
<annevk>
XHR encoding detection is a mess too
13:50
<annevk>
it doesn't actually follow any rules or something
13:50
<Lachy>
the test showed that FF does, IE doesn't
13:51
<Lachy>
I'm sure Hixie would have tested that when he wrote the encoding sniffing algorithm
13:51
<Lachy>
does XHR follow XML rules for XML types?
13:52
<annevk>
well yes
13:52
<annevk>
but there's also responseText
13:52
<Lachy>
oh good
13:52
<Lachy>
yeah, I expected there to be problems with that
13:52
<annevk>
at some point the fallback has to be UTF-8
13:52
<annevk>
for when no content-type is provided
13:52
<annevk>
for when it's text/plain
13:53
<Lachy>
I think you can probably adopt the HTML5 algorithm for text/html
13:53
<annevk>
implementors say no
13:53
<Lachy>
really?
13:53
<Lachy>
why is text/html for XHR any different from other text/html?
13:53
<annevk>
well XHR doesn't follow text/css either atm
13:54
<annevk>
simplified:
13:54
<annevk>
1) if the response Content-Type specifies a charset, use it;
13:54
<annevk>
2) otherwise, if the response is XML follow XML rules;
13:54
<annevk>
3) otherwise, if Content-Type is not specified or empty, also follow XML rules;
13:54
<annevk>
4) otherwise, use utf-8, and do not look inside the response data.
13:55
<annevk>
not entirely clear to me how you arrive at 4 or how 3 would work
13:55
<Lachy>
does IE fallback to UTF-8 like that?
13:56
<annevk>
it seems so, although Bjoern discovered some other incompatibilities iirc
13:56
peepo
(n=Jay⊙hrbc) Quit ("later")
14:02
BenWard
(n=BenWard⊙cccnc) Quit ()
14:11
icaaq
(i=icaaaq⊙c2cbs) has left #whatwg
14:38
Ashley_
(n=ash⊙plbion) Quit (Read error: 104 (Connection reset by peer))
14:49
<annevk>
the part on <script> should prolly deal with <!-- and --> in script blocks...
14:49
<annevk>
or does ECMAScript deal with that?
14:50
<virtuelv>
annevk: ecma-262-3 or 4 doesn't really deal with angle brackets at all
14:51
<virtuelv>
I would be fine if it was specified that browsers just ignore it
14:51
<virtuelv>
the problem is when you have constructs such as <script><!-- if (foo-->bar) --></script>
14:51
<virtuelv>
where does that comment end?
14:52
<annevk>
I suppose HTML would have to define it then
14:52
<annevk>
prolly makes more sense too
14:54
<virtuelv>
the bizarre thing is that the comments have been fluff since Netscape 2.0, or so
14:55
<krijnh>
Yet 'everybody' still adds them
14:55
<virtuelv>
krijnh: sadly, yes
14:55
<virtuelv>
I've advocated that people drop them for years
14:56
<Lachy>
sadly, there's actually a "modern" browser that requires the comments :-( Luckly, it can mostly be ignored
14:56
<krijnh>
Which one?
14:56
<virtuelv>
Lachy: which?
14:56
<Lachy>
the Motorolla v3 phone browser
14:57
<virtuelv>
in that case, ignore it
14:57
<Lachy>
it sucks big time
14:57
<Lachy>
I have to deal with it for the mobile sites I build at work
14:57
<krijnh>
In any case, external scripts
14:57
<Lachy>
yeah, external scripts should be used anyway
14:58
<Lachy>
though, not in all cases
14:58
annevk
was mainly asking this as implementor / testcase writer
14:59
<Lachy>
/--> gets ignored by the script engine because it's commented out, it causes an error if it's not commented
14:59
<Lachy>
<!-- should be ignroed by the script engine
15:00
<Lachy>
I know in CSS, the CSS parser specifically deals with <!-- and -->
15:00
<krijnh>
With inline css?
15:01
<Lachy>
within <style><!-- --></style> and in external style sheets. I don't think within style="" is allowed
15:01
<Lachy>
check CSS2.1 syntax for an accurate answer
15:03
<krijnh>
http://www.w3.org/TR/CSS21/syndata.html#comments
15:03
<krijnh>
Doh, makes sense
15:03
<virtuelv>
annevk: mostly, //.*--> gets ignored by the script parser
15:03
<virtuelv>
<!-- gets ignored
15:04
<virtuelv>
--> does not get ignored by everyone at least
15:04
<virtuelv>
constructing a test case is a bit hard, though
15:04
<Lachy>
--> causes syntax errors if it's not commented out with // or /* */
15:05
<virtuelv>
Lachy: again 'if (foo-->bar){ doSomething() } ' is, IIRC; perfectly legal ECMA syntax
15:05
<Lachy>
I think if <!-- appears in an external script, where it isn't touched by the HTML parser, the script engine handles it fine
15:05
<virtuelv>
ecmascript*
15:06
<annevk>
Lachy, for some value of "fine" I suppose?
15:06
<annevk>
in CSS that wouldn't work for instance
15:06
<Lachy>
what wouldn't work in CSS?
15:07
<annevk>
<!-- body { background:lime } in an external file
15:07
<annevk>
where
15:07
<annevk>
data:text/html,<style>%0a<!--body { background:lime }</style><body>test
15:07
<annevk>
does work
15:08
<Lachy>
it should do, the CSS sytnax rules looks like CDO and CDC are allowed to appear there
15:09
<Lachy>
stylesheet : [ CDO | CDC | S | statement ]*;
15:15
<Lachy>
good night
15:16
PaiTrakt
(n=opera⊙8) Quit (Client Quit)
15:16
<annevk>
oh, maybe i was mistaken
15:16
<annevk>
i suppose i confused it with people having <style> in their external style sheet
15:20
peepo
(n=Jay⊙hrbc) Quit ("later")
16:12
hasather_
(n=hasather⊙8ttc) Quit (Read error: 110 (Connection timed out))
16:23
om_sleep
(n=mjs⊙dsdsn) Quit ()
16:27
om_sleep
(n=mjs⊙dsdsn) Quit (Client Quit)
16:45
annevk
(n=annevk⊙8972) Quit (Read error: 60 (Operation timed out))
16:54
mw22_______
(n=chatzill⊙hdsn) Quit (Read error: 110 (Connection timed out))
16:56
BenWard
(n=BenWard⊙cccnc) Quit ()
17:15
sayrer_
(n=chatzill⊙ucmc) Quit (Read error: 110 (Connection timed out))
17:19
<krijnh>
annevk: Do you know if Mark Wubben is on irc somewhere?
17:20
<hasather>
krijnh: I can ask him to join here
17:20
<krijnh>
There should be a #sifr channel on this server, but there isn't one :)
17:21
<hasather>
hey markwubben
17:21
<annevk>
magic
17:21
<krijnh>
Magic indeed
17:21
<krijnh>
:)
17:21
<markwubben>
krijnh: two people started asking me to join!
17:21
<markwubben>
(namely david and anne)
17:21
<krijnh>
Hehe, hooray for IM
17:22
<krijnh>
markwubben: where is #sifr ?
17:22
<krijnh>
Or can I just bug you with a question? :)
17:22
<markwubben>
perhaps in #osflash, nobody ever came to the sifr channel so i stopped hanging out
17:22
<markwubben>
sure
17:23
<krijnh>
Using 2.0.2, IE uses <object> in stead of <embed>, which it did in 2.0.1
17:24
<krijnh>
Is that a bug or correct behaviour?
17:24
<markwubben>
correct
17:24
<krijnh>
K
17:24
<markwubben>
IE screwed up after the Eolas patch
17:24
<krijnh>
Cause the flashvars are added as a query string to the swf
17:24
<markwubben>
using embed
17:24
<markwubben>
on some machines
17:24
<markwubben>
correct, and that's beginning to bite
17:24
<krijnh>
Yeah, pretty irritating
17:24
<markwubben>
reason is, if you don't and the innerhtml of an ancestor is changed, ie will forget the vars
17:25
<markwubben>
i didn't quite realize back then that it would cause other problems
17:25
<markwubben>
but you should use sifr 3 anyway
17:25
<markwubben>
;)
17:25
<krijnh>
I should
17:25
<krijnh>
:)
17:25
<krijnh>
Ready for public?
17:28
<markwubben>
the latest nigthly is pretty good
17:29
<krijnh>
Pretty good as in it can be used on nfu.nl ?
17:29
<markwubben>
one issue with IE that could cause issues, but i haven't been able to look further into it
17:29
<markwubben>
it's a lot better than 2.0.2
17:30
<markwubben>
the IE issue appears to be minor, and only occurs if you have an innerhtml dom manipulation right inside the page (so no onload events etc)
17:30
<markwubben>
so yea, use it
17:30
<krijnh>
Okay, I will
17:30
<markwubben>
just make sure to stay up to date with newer versions
17:30
<krijnh>
That's the problem
17:30
<markwubben>
not all, but important fixes and releases
17:30
<markwubben>
yea i know :)
17:30
<krijnh>
I don't have control over this server :/
17:32
<krijnh>
And I hope TYPO3 won't screw up
17:32
<krijnh>
Anyway
17:32
<krijnh>
Thanks :)
17:32
<markwubben>
well, let them decide then
17:32
<markwubben>
sIFR 3 will save them a lot of bandwith
17:32
<markwubben>
cause it fixes IE and Safari's problems with fetching the flash movies
17:33
<krijnh>
I should convince them to use my cms and server then ;p
17:38
tantek
(n=tantek⊙adspn) Quit ()
17:40
<krijnh>
markwubben: thanks, je mag weer gaan ;o)
17:40
<markwubben>
lol
17:40
<markwubben>
i'll stick around until colloquy restarts
17:40
<krijnh>
:)
17:58
zcorpan_
(n=zcorpan⊙8sts) Quit (Read error: 104 (Connection reset by peer))
17:59
<zcorpan_>
Charl: i've made some changes to the status script
18:00
<Charl>
cool, can i re-upload for you?
18:00
<zcorpan_>
sure
18:00
<Charl>
ok hold on
18:00
<Charl>
my firefox is crashing as usual :)
18:01
<zcorpan_>
using trunk?
18:01
<Charl>
yeah
18:04
markwubben
(n=mark⊙1dcn) Quit ()
18:05
bewest
(n=ben@httpcraft/bewest) Quit (Connection timed out)
18:10
j00p
(n=zeno⊙adxn) Quit (Read error: 110 (Connection timed out))
18:10
gsnedders
(n=gsnedder⊙hrbc) Quit ()
18:11
<Charl>
zcorpan_: sorry my checked out copy is at work and i need to go home, my mom is shouting already ;)
18:11
<Charl>
i'll look at it first thing tomorrow morning for you
18:11
<Charl>
cheers all
18:11
Charl
(n=Charl⊙snaz) Quit ()
18:41
icaaq
(i=icaaaq⊙c2cbs) has left #whatwg
18:57
othermaciej
(i=mjs@nat/apple/x-23ed88dabce42f39) Quit (Read error: 104 (Connection reset by peer))
19:02
othermaciej
(i=mjs@nat/apple/x-02e803889fbdbe6e) Quit (Nick collision from services.)
19:12
<annevk>
http://www.w3.org/2007/03/19-tagmem-minutes.html#item04
19:13
<annevk>
hmm
19:13
<annevk>
don't read that
19:13
<annevk>
it's not interesting
19:13
<sayrer>
sure isn't ;)
19:14
<annevk>
and those are the people making up the web architecture...
19:14
<annevk>
right
19:14
<sayrer>
well, it is politically difficult for them to say things we would like to hear
19:15
<sayrer>
so one can never tell if they get it or not from public records
19:17
<Hixie>
wrt the earlier comment, yes, i spoke to both the youtube people and the google video people
19:24
<Hixie>
i love all the discussions about whether we should allow native ui
19:24
<Hixie>
i've already specced that feature, it's just commented out...
19:24
<annevk>
the discussion is about having it by default
19:25
<annevk>
fwiw
19:25
<sayrer>
Hixie, why is it commented out?
19:25
<annevk>
(well, part of it anyway)
19:25
<annevk>
sayrer, complexity of the intitial impl
19:25
<Hixie>
it's commented out as part of the many simpliciations for v1
19:25
<Hixie>
if we require v1 to have ui, then people will criticise it when the ui sucks
19:25
<sayrer>
I see
19:25
<Hixie>
as, e.g., they have with wf2 and opera's ui
19:26
<Hixie>
and that will give a lot of strength to the argument that we should drop video altogether
19:26
<Hixie>
(not really a valid argument, but that rarely matters in wg discussions, sadly)
19:26
<Hixie>
(i'm talking about it getting removed if this gets to the html wg, btw)
19:27
<sayrer>
I agree with your parenthetical reasoning
19:27
<annevk>
if the UI really needs to be there by default it will be in the end anyway
19:28
<Hixie>
i'm also a little confused by the "either we use <object> or we have an element per media type!" arguments
19:28
<sayrer>
my only concern is dimensions
19:28
<Hixie>
why can't we use <object> for the uncommon cases and have elements for the important ones...
19:28
<Hixie>
sayrer: of?
19:28
<annevk>
Hixie, that's not logical (apparently)
19:28
<sayrer>
of <video> given presence of a native UI
19:28
Hixie
points to the topic again
19:29
<Hixie>
sayrer: ah yes
19:29
annevk
only points out what's being said
19:29
<Hixie>
sayrer: yeah that's one of the tough ones
19:29
<Hixie>
sayrer: should it be inside the box, outside the box, etc
19:41
othermaciej
(i=mjs@nat/apple/x-99683985d5cae1cf) Quit (Nick collision from services.)
19:52
csarven
(n=nevrasc⊙m1mvc) Quit (Read error: 104 (Connection reset by peer))
19:55
<Hixie>
no controversial e-mails in the various lists i watch
19:55
<Hixie>
how sad
19:55
<Hixie>
ok bbl
19:55
annevk
wonders what counts as controversial these days
19:56
<annevk>
<xxx src>
20:05
<ROBOd>
annevk: that's not controversial :)
20:05
<annevk>
<nigger>?
20:05
<ROBOd>
ok, *that* might be
20:06
<ROBOd>
:)
20:06
<annevk>
yeah, might...
20:06
<annevk>
oh well, got to go
20:06
<ROBOd>
depends who's on the mailing list
20:09
briansuda
(n=briansud⊙brhi) Quit ("off to find food")
20:09
<zcorpan_>
we still get spammers to the forum :(
20:11
<ROBOd>
zcorpan_: phpbb, that's what i call spam
20:11
<ROBOd>
for a forum i host on my server i made some patches, to fight against spammers
20:12
<zcorpan_>
what patches?
20:12
<ROBOd>
first of all, i configured phpbb so that registration requires permission from moderators
20:12
<ROBOd>
but that's not enough...
20:12
<zcorpan_>
how do you do that?
20:13
<ROBOd>
in the administration module
20:13
<ROBOd>
however, that's not enough, because spammers will, anyway, signup, and until you delete their account
20:13
<ROBOd>
their account profile will be public
20:13
<ROBOd>
url, occupation, email and all the values they can fill into the profile
20:13
<ROBOd>
all spam keywords... they are all visible
20:13
<ROBOd>
so, my patches deal with that
20:14
<zcorpan_>
would be great if you could send your patches to hixie
20:14
<ROBOd>
inactive accounts don't show in "The newest registered user is ..."
20:14
<ROBOd>
simple users can't see the profile of inactive accounts
20:15
<zcorpan_>
"Enable account activation: Admin"?
20:15
<ROBOd>
yes
20:16
<ROBOd>
and in the members list, no other value (such as location), except the username, is displayed in the page
20:16
<ROBOd>
for normal users, again
20:16
<ROBOd>
admins see no difference
20:16
<ROBOd>
zcorpan_: shall i send the patches to Hixie?
20:17
<zcorpan_>
yes please
20:17
<ROBOd>
i will also make them public, on my server
20:17
<zcorpan_>
ok
20:23
BenWard
(n=BenWard⊙cccnc) Quit (Read error: 110 (Connection timed out))
20:26
annevk
(n=annevk⊙5ccn) Quit (Read error: 110 (Connection timed out))
20:31
<Hixie>
if we require moderator permissions, we'd better have active moderators :-)
20:42
<Dashiva>
How about just routing registration through IRC?
20:44
<Lachy>
hey kingryan, yt?
20:47
<Lachy>
kingryan, I wanted to know if it would be possible to set up a copy of 'mfbot' that you use of #microformats, but for announcing changes to the whatwg wiki.
20:53
<ROBOd>
Hixie: other changes can't be made quickly, to phpbb
20:53
<ROBOd>
another forum is needed, but any system is susceptible to spam like that
20:54
<ROBOd>
in my case i tried to make it irrelevant for me if the spammer tries to signup
20:54
<ROBOd>
and, on the forum i host, i am not the moderator who deals with the account activations...
20:56
<zcorpan_>
obviously it would be better if spammers didn't get though the registration in the first place
20:56
<zcorpan_>
or even find the registration page
20:56
Dashiva
repeats IRC suggestion
20:57
<ROBOd>
zcorpan_: correct, change the phpbb system
20:57
<ROBOd>
i also dislike it spammers can signup
20:58
<zcorpan_>
ROBOd: i did a modification, but it wasn't enough
20:58
<ROBOd>
zcorpan_: preparing the patches
20:59
<ROBOd>
zcorpan_: i didn't do any modification to the actual signup process
20:59
<ROBOd>
i didn't want to bother too much :)
20:59
<ROBOd>
what did you change?
20:59
<zcorpan_>
the name of the captcha field
20:59
<Dashiva>
What if you rotated the field names? :)
21:00
<zcorpan_>
that could work
21:00
<zcorpan_>
use name="email" for something else, and the something else for the email
21:02
<zcorpan_>
or use NCRs
21:02
<hasather>
Dashiva: great idea, name="password" for password and name="repeatpassword" for e-mail or something probably confuses the bots
21:02
<hasather>
or what zcorpan_ said
21:11
<ROBOd>
done, will send the patches now to Hixie
21:11
<zcorpan_>
ROBOd: wait
21:11
<ROBOd>
waiting now
21:11
<zcorpan_>
perhaps swithing name=""s is enough
21:12
<zcorpan_>
if it isn't then you can send it to hixie later
21:12
<ROBOd>
they are very simple patches...
21:12
<ROBOd>
only 3 files, and a few lines
21:12
sayrer
(n=chatzill⊙ucmc) Quit (Read error: 110 (Connection timed out))
21:12
<zcorpan_>
sure, but we don't want admin registration :)
21:13
<ROBOd>
admin confirmation
21:13
<ROBOd>
ok then ... i will publish the patches on my site, anyway :)
21:13
<zcorpan_>
sure, thanks anyway :)
21:14
<ROBOd>
no problem
21:17
<Dashiva>
If it's going to be phpBB demand a strike-through mod
21:19
<ROBOd>
a moderator will still be needed for the forums
21:19
<ROBOd>
to clean inappropriate messages of users with no logic in their brains :)
21:20
icaaq
(i=icaaaq⊙c2cbs) Quit (Read error: 110 (Connection timed out))
21:20
<ROBOd>
(e.g. obscene messages, or otherwise vulgar messages)
21:33
<kingryan>
Lachy: hi, I'm here now
21:34
<kingryan>
mfbot is actually just a shell script tying together a few pieces that are built into mediawiki
21:35
<kingryan>
Lachy: here's mfbot in its entirety:
21:35
<kingryan>
#!/bin/sh
21:35
<kingryan>
cd /var/www/wiki/irc
21:35
<kingryan>
php rcdumper.php -m \#microformats | ./mxircecho.py mfbot irc.freenode.net
21:39
<ROBOd>
done, now i uploaded the patches on my site
21:39
<ROBOd>
http://www.robodesign.ro/mihai/blog/phpbb-2-fighting-spam
21:39
<ROBOd>
if they are needed, they are now public
21:41
<Lachy>
kingryan, thanks. How and where would I need to install it and set it up?
21:42
<ROBOd>
gotta go now
21:42
<ROBOd>
good night everyone
21:42
ROBOd
(n=robod⊙8321) Quit ("http://www.robodesign.ro";)
21:42
<kingryan>
Lachy: for us it's installed on the same web host as the wiki (it hits the db directly)
21:43
<Lachy>
I can install it on that host
21:43
krijnh|afk
(n=krijnhoe⊙kxn) Quit (sterling.freenode.net irc.freenode.net)
21:43
karlUshi
(n=karl⊙1rhnj) Quit (sterling.freenode.net irc.freenode.net)
21:43
Dashiva
(i=Dashiva⊙vsnn) Quit (sterling.freenode.net irc.freenode.net)
21:44
<kingryan>
Lachy: we're also using an older version of mediawiki, so YMMV
21:45
<Lachy>
I gotta get ready for work, could you write and send me some step-by-step instructions for installing it?
21:46
<kingryan>
there aren't really any steps, just run that shell script
21:46
<kingryan>
though you may need to change some of the details
21:46
<Lachy>
where do the files rcdumber.php and mxircecho.py come from?
21:47
<Lachy>
s/rcdumber/rcdumper/
21:47
<kingryan>
from mediawiki
21:47
<kingryan>
at least they were in our version of it
21:47
<Lachy>
ok, I'll give it a try after work today
21:50
<Lachy>
we don't seem to have those files, but they're probably available somewhere online
21:52
<Dashiva>
Very newest MW version, nice
21:52
<Lachy>
oh, I just remembered dreamhost apparently doesn't allow IRC, I probably wouldn't be allowed to set it up anyway :-(
22:06
icaaq
(i=icaaaq⊙c2cbs) Quit (Read error: 145 (Connection timed out))
22:06
<Hixie>
zcorpan_: done (what does it do?)
22:07
<Hixie>
gotta love some of the mails to whatwg
22:07
<Hixie>
"i'm not an expert in any of the relevant topics, but here's my opinion anyway"
22:07
<Hixie>
especially amusing given that people who _are_ experts have already given their opinion.
22:07
<zcorpan_>
Hixie: it changes the captcha's name to "email" and the email's name to "sneaky" (NRC escaped to confuse bots even more)
22:08
<Hixie>
ah so it has to be a real human who posts basically
22:08
<zcorpan_>
Hixie: well, it's good they say they're not experts so i don't have to read it ;)
22:09
<zcorpan_>
Hixie: hopefully, yes
22:09
<Hixie>
let me know how that goes
22:09
<zcorpan_>
sure
22:11
csarven
(n=nevrasc⊙m1mvc) Quit (Read error: 104 (Connection reset by peer))
22:12
<Dashiva>
I've been wondering, is Chris Wilson joining the html wg "enough"?
22:13
<Hixie>
enough for what?
22:13
<Dashiva>
To get meaningful participation from microsoft
22:13
<Hixie>
what would you consider meaningful?
22:14
<Dashiva>
Resulting changes in the browser
22:16
BenWard
(n=BenWard⊙cccnc) Quit ()
22:16
<Hixie>
i don't think that's up to anyone but management at microsoft
22:17
<zcorpan_>
Dashiva: probably a more effective way to get changes in the browser is to submit bug reports
22:27
icaaq_
(i=icaaaq⊙c2cbs) Quit (Read error: 110 (Connection timed out))
22:51
othermaciej
(i=mjs@nat/apple/x-55e1df219b966991) Quit (Read error: 110 (Connection timed out))
22:55
<othermaciej_>
Dashiva: I think more people from MS than him will join
22:55
<othermaciej_>
Dashiva: and I also think that while IE implementing the spec would be better, we can't consider it a showstopper for progress
22:57
<Dashiva>
Oh, I wasn't really thinking like that. More if him joining was a sign of intended action, or more of a personal thing on his behalf
22:58
<Hixie>
probably a mixture -- after all, his personal beliefs presumably have a big impact on IE
23:01
<othermaciej>
he is the IE Platform Architect, so one would assume his opinions carry some weight there
23:06
karlUshi
(n=karl⊙1rhnj) Quit ("Where dwelt Ymir, or wherein did he find sustenance?")
23:11
Hixie
wonders how the people who support MPEG4 plan on letting Firefox and other free browsers pay for the MPEG4 license fees
23:13
<othermaciej>
it's not a per-unit royalty
23:13
<Hixie>
what is it "per" then?
23:13
<othermaciej>
MPEG-LA (I think) licenses based on approximate size of mpeg-related business you are doing
23:13
<othermaciej>
so MoCo/MoFo would presumably owe some fixed amount
23:13
<Hixie>
so how would that work for, say, linux distros?
23:14
<othermaciej>
otoh such code could not be under MPL/LGPL/GPL triple license, would need to be something else
23:14
<Hixie>
that's rather a blocker
23:14
<othermaciej>
well, it's fine to link a seperate library licensed under, say, BSD license, to an MPL/LGPL/GPL library
23:14
<othermaciej>
I don't know the details
23:15
<othermaciej>
and I don't mean to advocate adding an MPEG4 requirement
23:15
<Hixie>
sure
23:15
Hixie
doesn't really care about the codec himself, it's just that he has to take into account the needs of people who are distributing software without any financial backing
23:16
<Hixie>
which rather puts any fees of any kind out of the window
23:16
markwubben
(n=mark⊙1dcn) Quit ()
23:17
<othermaciej>
I don't know the details of when the fees apply, but it might not preclude that as long as licenses the patent
23:19
<Hixie>
as long as who licenses the patent?
23:19
<othermaciej>
er, sorry
23:19
<othermaciej>
I don't know if I want to make claims about this b/c I have not studied the terms
23:20
<othermaciej>
but I think for example that it is possible to bundle the QuickTime plugin without paying the patent license because Apple has paid for it
23:20
<othermaciej>
so analogous condition could apply for other vendors
23:20
<othermaciej>
but like I said, don't take my word for it
23:20
<othermaciej>
just don't assume it is unworkable either
23:20
<Hixie>
seems to me that if it was that simple, someone would have paid the license fee already and linux distros would be happily shipping the codec
23:22
<tantek>
or it could just be transactional friction
23:22
<tantek>
which prevent many things "that if it was that simple" from ever happening.
23:24
<Hixie>
what do you mean by transactional friction?
23:26
Hixie
begins replying to the 206 <video> element e-mails
23:28
<Hixie>
50 people sent messages to this thread
23:28
<Hixie>
that's an average of 4 mails per person
23:31
zcorpan_
has sent one email on <video>
23:32
<Hixie>
that's great though, means that we have at least 10% active participants on the whatwg list
23:32
<Hixie>
that's high
23:33
<zcorpan_>
i know some who are subscribed but i haven't seen any mails from them to the list
23:33
<Hixie>
oh there are hundreds of lurkers, sure
23:33
<Hixie>
that's pretty normal
23:33
<zcorpan_>
yup
23:34
othermaciej
(i=mjs@nat/apple/x-9c192c7c979df1e1) Quit (Read error: 145 (Connection timed out))
23:35
<zcorpan_>
i've seen it being suggested that in forums, 1% post new threads, 10% reply to them, and the rest are lurkers
23:36
<zcorpan_>
i guess the same figures could be applied to wg mailing lists
23:43
<Dashiva>
whatwg is less scary to post to than public-html, at least :)
23:44
<tantek>
Hixie, it means that nothing is as easy as simply "doing the work", that there is communication involved, that there are other things that are higher priority (opportunity costs) etc. all kinds of often dismissed costs of doing anything, however simple.
23:44
<Hixie>
tantek: seems like the cost of not being able to play back any modern video content would offset that
23:45
<tantek>
no because that is an unrealized gain, rather than a cost
23:45
<tantek>
the two are different
23:45
<Dashiva>
mplayer probably eats up most of the gain from video content
23:45
<Hixie>
fair enough
23:45
<tantek>
people dismiss unrealized gains ALL the time without even knowing it
23:46
<tantek>
in fact, you and I are probably dimissing many unrealized gains right now, if we had time to actually think about them
23:46
<Hixie>
wow, there's a majority in favour of the htmlwg having a f2f meeting
23:46
<Hixie>
tantek: yeah
23:46
<othermaciej_>
a lot of modern video is Flash or Windows Media, those patents might not be freely licensable
23:46
<Hixie>
i wonder if the majority is because only people who actually want to go to a f2f bothered to reply
23:47
<Hixie>
othermaciej_: true
23:47
<Dashiva>
Hixie: How many votes?
23:48
<Hixie>
13 yes, 8 no. http://www.w3.org/2002/09/wbs/40318/ftf07/results
23:48
<Hixie>
if you're an html wg member, fill the survey in: http://www.w3.org/2002/09/wbs/40318/ftf07/
23:49
<Dashiva>
"Hey Surveyman, don't forget to fill out those surveys"
23:49
<Dashiva>
Those ads always amused me
23:50
<Hixie>
of the people here, hasather, hendry, jgraham, hsivonen, citoyen, othermaciej, and probably others whom i haven't spotted in the list haven't filled in http://www.w3.org/2002/09/wbs/40318/ftf07/
23:50
<othermaciej>
I will fill it in
23:51
<othermaciej>
but I have to ask my corporate masters if I can offer to host on Apple's behalf
23:51
<Hixie>
you want to host a meeting with over a 100 people?
23:51
Hixie
is very skeptical that such a meeting is even sensible
23:51
<othermaciej>
I doubt 100 will actually show up
23:52
<othermaciej>
I imagine the ratio of people showing up to nominal WG members will be similar to the ratio of people posting on the list to nominal list members
23:52
<Dashiva>
Most of the new members coming via the whatwg invitation call are probably on private funding, so less likely to travel
23:53
zcorpan_
has now filled in the survey
23:54
Dashiva
is trying to phrase the comments
23:54
<hasather>
filled it in too
23:54
<Hixie>
uh. the numbers just changed more towards the no.
23:54
<Hixie>
i wonder who changed their minds.
23:56
<Hixie>
oh maybe i was looking at the wrong results, nm.
23:57
<Dashiva>
Need to get a job, become rich and wealthy...
23:58
<hasather>
Dashiva: working at Opera this year?
23:58
<Dashiva>
If they'll have me, that's the plan
23:59
<Dashiva>
You?
23:59
<hasather>
hopefully
23:59
<Dashiva>
"Skal vi kjäka?"