00:29
<Hixie>
can anyone find a better 16x16 image on a microsoft-owned domain to represent IE than http://www.microsoft.com/favicon.ico ?
00:32
<jruderman>
it needs to be hosted on a microsoft domain?
00:38
<jruderman>
does exploiting an XSS hole on a microsoft domain count?
00:38
<Hixie>
i'd like it to be hosted on a microsoft domain
00:38
<Hixie>
what's your uri? :-)
00:39
<gavin>
does it have to be exactly 16x16?
00:39
<Hixie>
if possible
00:41
<gavin>
closes I've found so far is http://www.microsoft.com/enable/images/icons/windowsvista/ie7.jpg
00:41
<jruderman>
jpg wtf
00:43
<Philip`>
http://msdn2.microsoft.com/en-us/library/ms535205.new(en-us,VS.85).gif
00:43
<Hixie>
oooh, that might work
00:43
<Philip`>
http://msdn2.microsoft.com/en-us/library/ms531207.ATTR_msext(en-us,VS.85).gif
00:46
<Hixie>
sweet, that first one is perfect
00:46
<Hixie>
thanks
00:47
<Hixie>
how about something for safari that's better than http://www.webkit.org/favicon.ico ?
00:48
<Hixie>
the problem with apple is that their "icons" have become photo-realistic multi-megapixel images at this point
00:48
<Hixie>
(ideally something on apple.com or webkit.org)
00:51
<Philip`>
"Your search - imagedoesnotexist - did not match any documents. Results have been omitted at your request. If you'd like, you can <a href=/images?q=imagedoesnotexist&svnum=50&um=1&hl=en&safe=off&client=opera&rls=en&sa=G>repeat your search with the omitted results included</a>." [as the displayed text] - hmm, looks like someone escaped their HTML one time too many
00:52
<Hixie>
uri?
00:53
<Philip`>
http://images.google.com/images?imgsz=icon&q=imagedoesnotexist
00:55
<Philip`>
I guess http://images.apple.com/safari/images/performance_safari20070611.gif is a bit too big
00:55
<Hixie>
yeah
00:56
<Philip`>
Use <canvas> to resize it :-)
00:57
<Hixie>
i'm using it from css
00:58
<Philip`>
Use <canvas> to resize it and toDataURL it into the CSS
00:58
<Philip`>
(except that doesn't work because of the security restrictions :-( )
00:59
<Hixie>
(i reported the overescaped html issue)
00:59
<Hixie>
(thanks)
01:04
Hixie
continues working on http://damowmow.com/playground/spec-annotation/001.html
01:04
<Hixie>
(which is what i need those icons for, btw)
01:10
<othermaciej>
cute
02:37
<jruderman>
Hixie: i think http://www.hixie.ch/tests/evil/page-loading/incremental/001.cgi?mime=text%2Fxml&delay=1&repeats=10 is busted
03:26
<Hixie>
jruderman: could be
03:27
<Hixie>
jruderman: not sure how to fix it
03:44
<jruderman>
Hixie: :/
03:44
<jruderman>
not having that test makes me sad, because it was the testcase for https://bugzilla.mozilla.org/show_bug.cgi?id=368909
03:45
<Hixie>
can you tell what's going on?
03:45
<Hixie>
is it getting gzipped or something?
03:45
<Hixie>
i don't really even know where to begin looking for a fix
03:45
<jruderman>
dunno. i tried using telnet last week but didn't get very far.
03:47
<jruderman>
ok, yeah, it's getting gzipped
03:47
<Hixie>
that's really weird
03:48
<Hixie>
gzip is disabled on this server in theory
03:48
<Hixie>
maybe i'm going through some sort of proxy or something
06:33
<Dan->
hello
06:35
<Dan->
any iphone software developers here?
06:49
<othermaciej>
Dan-: in what sense?
06:50
<Dan->
i want to start a new company
06:51
<Dan->
and need people to program web apps for the iphone
07:15
<mitsuhiko>
andrioid ftw!
07:30
<othermaciej>
Dan-: you can probably look for anyone who is a good web developer then
08:32
<Hixie>
http://damowmow.com/playground/spec-annotation/001.html
08:32
<Hixie>
if anyone wants to fill in the XXX parts in http://damowmow.com/playground/spec-annotation/status.js -- be my guest :-)
08:33
<Hixie>
(in particular, the network object at the bottom)
08:33
<Hixie>
(see also http://damowmow.com/playground/spec-annotation/PROTOCOL )
08:34
<OmegaJunior>
Interesting. Using Javascript to automate compliance testing?
08:35
<OmegaJunior>
No, it does something else
08:38
<zcorpan>
Hixie: nice
08:39
<OmegaJunior>
I happen to prefer using Opera... Does Opera not support the feature or does your feature only work in MSIE?
08:39
<Hixie>
i doubt it works in IE
08:39
<jacobolus>
Hixie: what's it supposed to do?
08:40
<OmegaJunior>
It tells me it only works in MSIE though...
08:40
<Hixie>
jacobolus: i hope to evneutally develop it to allow us to annotate sections of the spec
08:40
<Hixie>
OmegaJunior: seems to work fine in opera 9.x
08:40
<OmegaJunior>
Wouldn't that require a javascripted write action to the server document?>
08:41
<jacobolus>
Hixie: so the javascript is supposed to do what then?
08:41
<OmegaJunior>
OK, I'll try again and disregard the status message.
08:41
<jacobolus>
i.e. what does SpecStatusNetworkConnection do?
08:41
<Hixie>
jacobolus: the XXX bits?
08:41
<Hixie>
jacobolus: it's the API between the UI and the server
08:41
<Hixie>
jacobolus: it'll do XHR calls to the cgi scritps (which are also yet to be written)
08:42
<jacobolus>
gotcha
08:42
<Hixie>
basically it implements what that PROTOCOL file describes
08:42
<OmegaJunior>
Cute
08:42
<Hixie>
anyway, unless someone beats me to it, i'll work on those XXX bits tomorrow :-)
08:42
<jacobolus>
Hixie: so those XXX bits are pretty simple then. they just do a cgi call and then update the UI?
08:42
<OmegaJunior>
It's like Google Maps, using Javascript to add material to a server-based file
08:42
<Hixie>
then it's the server side, unless mike smith beats me to it
08:43
<Hixie>
jacobolus: just do a cgi call and then invoke the callback, yeah
08:43
<jacobolus>
aha. so why are they all separate functions? seems like it'll just be repeated code
08:43
<OmegaJunior>
Hixie: you\re right, the feature does work in Opera 9. The Implementation Status tells us otherwise.
08:44
<Hixie>
OmegaJunior: you may be confusing the test data with some sort of report on the code itself
08:44
<OmegaJunior>
Possible
08:44
<Hixie>
jacobolus: look at the PROTOCOL file, they each have different things they have to send
08:44
<OmegaJunior>
:)
08:45
<OmegaJunior>
Oh, I see. Yes, I did misinterpret that Implementation Status.
09:03
<hsivonen>
Hixie: very cool.
09:22
<zcorpan>
hmm, ie doesn't seem to interpret us-ascii as windows-1252
09:22
<zcorpan>
it seems to convert high bytes to some ascii character
10:51
<hsivonen>
I need Open Source (non-viral) author readable succinct descriptions of the expected syntax for each attribute value microsyntax in HTML5
10:51
<hsivonen>
should I write my own?
10:52
<OmegaJunior>
Delegate to working group?
10:53
<hsivonen>
which one? the WHATWG? W3C document output is not Open Source in general
10:54
<OmegaJunior>
Hmm.
10:54
<OmegaJunior>
Start a draft on WikiPedia, select open source license, have others add on and correct.
10:55
<hsivonen>
hmm. I guess I'll start a draft on the WHATWG wiki as its licensing policy suits my needs
10:56
<OmegaJunior>
Suggested WikiPedia for the larger audience.
10:56
<OmegaJunior>
Granted, WhatWG wiki will reach target group.
11:25
<mpt>
OmegaJunior, Wikipedia is licensed under the FDL, which is share-alike aka viral
11:26
<mpt>
And you'd likely have people crying "WP:NOT", too
11:26
<OmegaJunior>
Hmm. MediaWiki allows several license options. I had hoped WikiPedia would allow that as well.
11:27
<OmegaJunior>
People will cry that about any kind of social encyclopedia, but also about any non-open source too
11:27
<OmegaJunior>
They'll cry because they like to cry.
11:27
<mpt>
particularly "Wikipedia is not an indiscriminate collection of information"
11:27
<OmegaJunior>
Is anything?
11:28
<mpt>
No, but that's not relevant to this question :-)
11:28
<OmegaJunior>
:D
11:29
<hsivonen>
It seems to me there's quite a bit of original research on Wikipedia and people complain semirandomly when they don't like the results
11:30
<OmegaJunior>
So they should do their own research and publish their own results.
11:31
<mpt>
There's quite a bit of lots of things that are against Wikipedia policy on Wikipedia, and some people will remove it for political reasons, but others will remove it because it's against Wikipedia policy
12:30
<jacobolus>
hsivonen: they mostly complain when they know that some bit of it is BS
12:30
<jacobolus>
(which is quite common :)
12:30
<jacobolus>
some "original research" is pretty useful, and sticks around
12:31
<jacobolus>
hsivonen: what's an attribute value microsyntax?
12:31
<hsivonen>
jacobolus: e.g. floating point number, media query, MIME type, IRI reference
12:32
<jacobolus>
why do you need a non-viral license?
12:33
<jacobolus>
or you want to incorporate it into some non-free derivative work?
12:35
<jacobolus>
hsivonen: how many of those are there? is it a major undertaking to write such a document?
12:37
<hsivonen>
http://wiki.whatwg.org/wiki/MicrosyntaxDescriptions
12:38
<hsivonen>
jacobolus: I want the software as a whole to be embeddable as widely as possible and I want it to be packageable for Debian which is likely to invoke GPLv3 for some parts and I can't have competing strong copyleft licenses in there
12:39
<jacobolus>
gotcha
12:39
<jacobolus>
yeah, that is pretty many of them :)
12:39
<jacobolus>
shouldn't be too hard to fill in if you put out a blog post or something
12:48
<hsivonen>
jacobolus: I think I'll fill in a sample entry and then announce the project on the wiki
13:14
<hsivonen>
hmm. a couple of Polish link spammers have found the WHATWG blog
13:16
<hsivonen>
zapped the moderation queue
13:24
<hsivonen>
I wrote sample content for language tags: http://wiki.whatwg.org/wiki/MicrosyntaxDescriptions#Language
13:25
<hsivonen>
does it look author-readable?
13:25
<hsivonen>
that is, is it a good sample?
13:27
<hsivonen>
can anyone think of a politically correct non-contrived example of a language tag that has language, non-default script and a region?
13:35
<hsivonen>
Lachy, Hixie: blog.whatwg.org warns that WordPress is out of date
13:35
<zcorpan>
hsivonen: it looks ok to me
13:35
<zcorpan>
(the language tag text, that is)
13:36
<Lachy>
hsivonen, yes, I'm aware. It's just a minor revision update that I need to do later
13:36
<hsivonen>
zcorpan: thanks.
13:36
<hsivonen>
Lachy: ok
13:38
<hsivonen>
posted announcement
13:43
zcorpan
gets an index of/ at blog.whatwg.org
13:45
<hsivonen>
zcorpan: weird. works for me
13:46
<Lachy>
I just tried upgrading, but will have to do it later
13:46
<Lachy>
things got messed up. it's restored now
13:55
<zcorpan>
works now, yep
13:56
<Lachy>
blog is now upgraded
13:56
<Lachy>
I should probably update the wiki to the latest version of media wiki later too
13:58
<dglazkov>
hi all
14:35
<zcorpan>
Lachy: links point to 404s in the blog
14:35
<zcorpan>
e.g. http://blog.whatwg.org/microsyntax-descriptions
14:53
<Lachy>
zcorpan, hsivonen, fixed. I just forgot to copy .htaccess
14:57
<hsivonen>
Lachy: thanks
16:00
<zcorpan>
http://simon.html5.org/test/xml/xml-stylesheet/
16:00
<zcorpan>
time to start drafting a spec
16:01
<zcorpan>
any opinions off-hand?
16:02
<hsivonen>
zcorpan: do you mean you are going to write style sheet PI 5?
16:02
<zcorpan>
yes
16:02
<hsivonen>
nice
16:02
<hsivonen>
so many specs in need of the 5 treament
16:02
<zcorpan>
indeed
16:08
<zcorpan>
btw, does someone know enough xslt to know how http://simon.html5.org/test/xml/xml-stylesheet/support/transform.xml can be modified such that the <title> survives the transform?
16:08
zcorpan
knows close to nothing about xslt
17:46
<gsnedders>
hmmm… my site rather breaks in IE5.01. sorry 0.25% of my site's visitors. the 0.2% using IE5.5 are fine, though
17:48
<Philip`>
My recent sites break in IE6 and IE7 too, and it was only intentional in one of them
17:49
<Philip`>
(*one of the sites)
17:56
<zcorpan>
interesting. ie parses entities twice when you feed it xml with a css stylesheet -- once with the xml parser and then again with the html parser
17:58
<zcorpan>
test case: <?xml-stylesheet type='text/css'?><a>&amp;amp;</a>
17:58
<zcorpan>
(actual result: "&", expected result: "&amp;")
17:59
<zcorpan>
you can also use &amp;auml; or so to confirm that html entities work
18:17
<zcorpan>
http://simon.html5.org/specs/xml-stylesheet5
19:50
<Hixie>
well. the http group is a whole new level of wow.
19:51
<hsivonen>
Hixie: ?
19:54
<gsnedders>
Hixie: multi-content-type?
19:55
<gsnedders>
Hixie: or the RFC 2119 discussion?
20:14
<Hixie>
gsnedders: all mail to that list in the last 24 hours
20:16
<gsnedders>
Hixie: well, over 50% of that is RFC 2119 :(
20:17
gsnedders
wonders if he actually has any homework that isn't overdue
20:50
<gsnedders>
why the hell does the UCAS pre-registration form require you to choose six subject areas you may be interested in studying at higher education!?
20:51
<Philip`>
Pre-registration?
20:52
<Philip`>
(I guess this isn't the same as the thing where you choose six courses/universities to apply to)
20:52
<gsnedders>
Philip`: nowadays you're meant to get a UCAS card in Year 12/S5
20:53
<gsnedders>
this form is badly marked up and badly worded and illogical.
20:53
<Philip`>
Oh, okay
20:53
<gsnedders>
do not…receive…by mail
20:53
<gsnedders>
do…receive…by email
20:53
<gsnedders>
do…receive…by text messaging
20:53
<gsnedders>
can we not be consistent!?
20:54
gsnedders
wonders if he can get away with putting just four subjects down
20:54
<Philip`>
It's annoying when they take a mostly-understandable paper form and then migrate to an online system which doesn't work and doesn't make sense and is full of errors
20:54
<gsnedders>
(as, quite frankly, if I didn't get in for any of them, I wouldn't go to uni)
20:54
<gsnedders>
it seems to have worked
20:54
<Philip`>
and then charge you £20 to apply onlnie instead of on paper
20:54
<gsnedders>
heh
20:55
<Philip`>
*online
20:55
<gsnedders>
Philip`: you're at Cam, aren't you?
20:55
<Philip`>
I am
20:55
<gsnedders>
which college?
20:55
<Philip`>
King's
20:55
<gsnedders>
nice
20:56
<Philip`>
I think it is :-)
20:56
<gsnedders>
my sister went to Sewlyn to do law
20:56
<gsnedders>
however you spell it :)
20:56
<gsnedders>
admittedly, my sister is far older than me (i.e., old enough to legally be my mother!)
20:57
<Philip`>
*Selwyn :-p
20:57
<gsnedders>
yeah, phonetically that makes more sense
20:57
<gsnedders>
:)
20:58
<gsnedders>
(
20:58
gsnedders
facedesks
20:58
<gsnedders>
typing what you want to say before pressing return helps.
20:59
<gsnedders>
*(I'm planning on doing computational physics, having talked to various people, to a master's degree: neither of oxbridge offer it)
21:00
<Philip`>
I walk past Selwyn most days, but don't really know much about it except that I don't particularly like the name
21:00
<gsnedders>
the actual quad is nice, not so the rest of the site
21:01
<gsnedders>
Question 2 of my Computing homework (given to 16/17 y/os): "What is a _user interface_?"
21:02
<gsnedders>
now, time to try and get back to doing my homework :)
21:04
<Philip`>
(My old physics teacher was quite happy that Cambridge forces you do to Natural Sciences for a year instead of something specialised like physics, since he originally went there to do chemistry and then changed his mind)
21:06
gsnedders
writes date in ISO8601 form, just too confuse most of his teachers and prove he will not be forced to break specs :P
21:06
<gsnedders>
(in the basic form, and therefore without hyphens)
21:07
<gsnedders>
Philip`: I haven't done any other sciences since before the two-year Standard Grade (GCSE equiv.) course started
21:08
<gsnedders>
Philip`: ironically, I am actually better than some people who have done chemistry/biology Standard Grade
21:09
<gsnedders>
I occasionally wish I had done chemistry, but I've never liked biology at all really
21:10
<Philip`>
There are a few options in the NatSci first year that don't require any previous knowledge, like, uh, geology
21:12
<gsnedders>
I will most likely go to Edi, and go straight into second year (as traditionally in Scotland you go to uni after S5 — I'm staying on to S6 and will have done A-Level equivs. (some people just take two years to get to A-S level equivs.))
21:13
<gsnedders>
and going straight into second year limits the options for outside subjects, partly as you have extra teaching of the one or two things you'll be missing from the slightly more specialist course at uni
21:16
<gsnedders>
How much of the algorithm is actually designed by the programmers in large software development companies?
21:16
gsnedders
has his doubts about quite how true what he is being taught is
21:17
<Philip`>
"the algorithm"?
21:23
<gsnedders>
i.e., a formal design such as that within HTML 5's UA requirements
21:36
<gsnedders>
as is taught, every tiny smallest stage is defined
22:13
jgraham_
points out that he did Cellular Biology in his first year of Natural Sciences despite not having done Biology since GCSE
22:14
<jgraham_>
(although in retrospect the computer science option would have been much more useful, although Cells was fun)
22:16
<gsnedders>
jgraham: I stopped the two before GCSE equiv :P
22:21
<jgraham_>
hmm my keyboard just stopped working entirely.
22:21
<Philip`>
(I'm not sure the first year of CS is especially useful, particularly the 25% version - most of the interesting bits come in later years)
22:21
<jgraham_>
Philip`: The same is true of almost all of Nat. Sci.
22:22
<jgraham_>
I guess
22:26
jgraham_
finds it slightly odd that he works ~5 minutes from Philip` yet they have never knowingly met
22:27
gsnedders
won't have the excuse in a month's time of having a grand-mother in Cambridge
22:28
<jgraham_>
gsnedders: the excuse for what?
22:28
<gsnedders>
jgraham_: going to Cambridge
22:54
<Philip`>
jgraham_: It seems surprisingly hard to meet people accidentally - there are a few people in Cambridge from my previous school who I've only encountered by chance about twice in three years
22:55
<Philip`>
(or maybe that's just because I don't hang around the colleges much)