01:35
<apucacao>
with HTML5 I can use the audio element to put a song on my website. I still need to obtain the rights to do so, correct? What about putting 30 second clips? (not sure if this is the right place to ask though...)
01:36
<aho>
I still need to obtain the rights to do so, correct? <- yes
01:36
<EclipseGc>
apucacao: this would be more of a technical howto less of a legal "allowed to" channel I suspect
01:37
<aho>
What about putting 30 second clips? <- same thing (in most countries)
01:37
<aho>
so basically... don't do it unless you got the rights (e.g. your own - self produced - music)
01:38
<aho>
(i'm not a lawyer) :>
01:38
<EclipseGc>
yeah, aho that totally ignores fair use (which the music industry would like to ignore exists)
01:41
<aho>
well, where i'm living i'd even have to pay fees if i were to use a single 44.1khz sample (even if that doesn't make any sense whatsoever) :>
01:42
<apucacao>
ok
01:43
<aho>
well, if you're living in the us and make some show about music where you play very short segments and talk about the songs, the album, the artists, etc... it would be *probably* fine
01:43
<aho>
(as long as you dont talk about stuff prince made) :>
01:45
<aho>
someone of thatguywiththeglasses.com makes some show like that
01:45
<aho>
http://thatguywiththeglasses.com/videolinks/teamt/paw/full-circle/3690-full-circle-session-1
01:52
<roc>
apucacao: actully no, HTML5 abolishes the copyright system, see section 4.8.11
01:54
<roc>
:-)
01:55
<EclipseGc>
roc: if only
07:39
<annevk>
heh, drclue is back on list
07:39
<annevk>
"I have suggested in the past that we implement XML and XSL attributes, but somewhere in the excitement of other fascinating topics this humble but incredible suggestion got lost in the winds."
07:41
<othermaciej>
I still don't understand that suggestion
07:42
<Hixie>
i think i have his suggestion on my pile two or three times
07:42
<ksemeks>
Is firefox really slower than chrome when trying to save string/objects in localStorage?
07:42
<ksemeks>
or is it, cause im doing something wrong?
07:45
<annevk>
hmm, one day IETF costs USD 350
07:54
<Rik`>
ksemeks: https://bugzilla.mozilla.org/show_bug.cgi?id=536544
07:57
<micheil>
annevk: hmm?
07:58
<micheil>
Rik` / annevk: I'm co-hosting an episode on a popular open-source podcast, and the topic in focus is WebSockets, is there anything either of you two would like mentioned? (I'll try my best to mention it, if relevant)
07:59
<annevk>
micheil, http://www.ietf.org/meeting/78/index.html
07:59
<annevk>
micheil, not really, thanks :)
07:59
<micheil>
okay
08:00
<micheil>
that is really weird.. the time 15:15 has come up for different things about 20 times in the pasts two days
08:03
<micheil>
Rik`: websockets in firefox are only ~4.0, right?
08:10
<Rik`>
micheil: yes, since beta1 iirc
08:11
<micheil>
but it's not currently in a stable / released version?
08:13
<Rik`>
micheil: exactly
08:15
<micheil>
okay, /I thought that was the case
08:28
<ksemeks>
where do i find the max value for localStorage? i read every browser has different. where do i find ti?
08:28
<ksemeks>
it*
08:31
<ksemeks>
and what is the max key size?
08:37
<annevk>
http://www.w3.org/blog/CSS/2010/07/21/resolutions_117 -- hmm, they want a reference to HTML5? I think that's a first for most WGs I'm involved in...
09:54
<annevk>
lol, first cookies, now cakes
10:00
<Workshiva>
I thought we wanted to send less traffic with every request, not more
10:00
<Workshiva>
And cakes are usually much bigger than cookies
10:01
<hsivonen>
cakes?
10:07
<jgraham>
There is cake?
10:07
<ksemeks>
xD
10:08
annevk
tries to find a URL
10:08
<annevk>
http://www.ietf.org/mail-archive/web/http-state/current/msg00893.html
10:09
<annevk>
(has a proposal for a cake header that handles session management)
10:11
<hsivonen>
I don't like writing test cases that need server-side scripting dealing with request timing
10:12
<annevk>
you're doing that now?
10:12
<hsivonen>
annevk: I'm doing it in the sense of trying to figure out how to do it. I'm not doing it in the sense of writing the test just yet.
11:02
<annevk>
Hixie, fwiw, I argued against sub-protocol negotiation on the basis that content negotiation failed
11:02
<annevk>
Hixie, and because imo it was needless complexity for v1
11:12
<hsivonen>
http://www.paulgraham.com/top.html seems relevant to the HTML WG
11:27
<nessy>
hsivonen: why? it's an interesting read but I fail to see how it relates to the HTML WG....
11:28
<hsivonen>
nessy: the "disputes" part
11:40
<Rik`>
anyone know why http://demos.hacks.mozilla.org/openweb/videos/transitions-demos.medium.ogv is not playing on Opera (10.6 Mac)
11:40
<othermaciej>
hsivonen: imagine how much fun it is to be responsible for resolving a lengthy series of disputes
11:41
<othermaciej>
not good for my coding output :-(
11:46
<annevk>
is anyone going to the hybi meeting in Maastricht this Friday?
11:46
<annevk>
I could probably get approval for the day fee but I am somewhat doubtful it is worth it
11:47
<annevk>
especially having read up on some of the discussions
11:49
<jgraham>
annevk: Seems unlikely on the whole :)
11:51
<annevk>
quite nice that they renamed their work "HTML Media Capture" so it gets some more substantial review than before
12:05
<jgraham>
othermaciej: I believe the current protocol meets the requirement that it is easilly implementable in scripting langauge's stdlib
12:06
<othermaciej>
jgraham: "easily" and "efficiently" are definitely not the same standard
12:06
<jgraham>
I would expect them to have functions for generating hashes, but not necessarily for setting up full TLS connections, for examplke
12:06
<hsivonen>
is Content-Type: text/html; charset="utf-8" legal in MIME?
12:06
<hsivonen>
note the quotes around "utf-8"
12:06
<annevk>
hsivonen, yeah
12:07
<annevk>
hsivonen, Gecko has/had a long standing bug on that
12:07
<annevk>
not sure which of the two
12:07
<hsivonen>
annevk: ah. thanks
12:07
annevk
-> food
12:07
<jgraham>
othermaciej: My definition of "efficiently" was supposed to be "don't have to write significant amounts of code not directly related to WebSockets"
12:07
<othermaciej>
I don't believe MD5 is part of the perl standard library
12:07
<othermaciej>
ah, you meant "efficiently" in the sense of programmer time, rather than software performance?
12:08
<jgraham>
Yes
12:08
<othermaciej>
I misunderstood
12:08
<othermaciej>
anyway, I am performing my "Python QQP" experiment with some real protocols
12:09
<othermaciej>
I have yet to find a protocol where the socket tutorial is on the first page of Google hits
12:09
<jgraham>
othermaciej: I don't know what QQP is, but I am not that surprised
12:10
<Philip`>
http://perldoc.perl.org/perlmodlib.html#Standard-Modules includes Digest::MD5
12:10
<othermaciej>
so far I have tried: SMTP, IMAP, IRC, Jabber, ftp, http
12:10
<othermaciej>
jgraham: I just mean that I searched for "Python X" for many values of X, where X is a network protocol
12:10
<jgraham>
For HTTP, I think the server is basically generic; there is little coupling between the server and the application. So it is not surprising that people use preexisting servers in most cases
12:11
<othermaciej>
in all cases the top hits seem to be a library, not a sockets tutorial that tells you how to write a server from scratch
12:12
<othermaciej>
also tried: tls, caldav, pop3
12:12
<jgraham>
I don't think that's surprising
12:12
<jgraham>
But isn't websockets lower-level in some sense
12:12
<jgraham>
FTP is a protocol for transferring files
12:13
<othermaciej>
is WebSocket going to be the first protocol where that's not the case?
12:13
<jgraham>
WebSockets is a substrate for writing your own protocols
12:13
<jgraham>
That talk to servers
12:13
<jgraham>
from web browsers
12:13
<othermaciej>
so is TCP, but people don't implement TCP themselves on top of IP whenever they need to implement an application protocol on top of TCP
12:14
<jgraham>
So a simple websocket library would look more like a simple socket library than a FTP library
12:14
<hsivonen>
I expect the Web Socket layer to become part of the libs in due course
12:14
<Rik`>
WebSockets is also new, so there's not a lot of frameworks to build on top of
12:14
<othermaciej>
for the record, searching for "Python WebSocket" currently finds a mix of tutorials with socket code, and premade libraries/frameworks
12:14
<jgraham>
So I think the comparison to ordinary sockets is more relevant than the caomparison to FTP or HTTP or whatever
12:15
<othermaciej>
I expect the ratio to move in favor of libraries as WebSocket becomes more popular and less experimental
12:15
<jgraham>
Sure, I don't disagree that people will use libraries
12:16
<jgraham>
But I would expect e.g. python to bundle a low-level libarry that looks like the socket library
12:16
<othermaciej>
I expect the vast majority of people will use libraries
12:16
<jgraham>
rather than something that looks like twisted
12:16
<jgraham>
Like I said, that makes some complexity acceptable
12:16
<jgraham>
e.g. in the handshake
12:17
<othermaciej>
hopefully the library takes care of message framing for you too
12:17
<othermaciej>
it would be perverse if it didn't
12:17
<jgraham>
Yeah.
12:18
<othermaciej>
and that's pretty much all there is to the protocol
12:18
<jgraham>
Indeed. It is simple :)
12:18
<Philip`>
Could you make a server API that looks very much like the browser client API?
12:18
<othermaciej>
I guess there is also the shutdown sequence (not sure if that is in the draft yet)
12:18
<othermaciej>
yes, I would expect a server API to be about as high-level as the client API
12:18
<jgraham>
Also, I didn't mention Safari 5 because I think that one browser shipping something is much less relevant than 4 browsers shipping something
12:18
<Philip`>
Seems good to aim for that symmetry if possible
12:19
<othermaciej>
the server API has to be a little different, because for a nontrivial service you need to create a new instance of something-or-other for each client connection
12:19
<jgraham>
One browser basically means people can write demos.
12:20
<jgraham>
Multiple browsers means people will write real applications
12:20
<othermaciej>
so there has to be at least an extra object that represents the listening service, as opposed to a specific client connection, and it makes client connections appear
12:20
<othermaciej>
I think both Safari and Chrome have WebSocket, but I think not the same protocol version :-/
12:20
<jgraham>
Yes
12:20
<othermaciej>
I would agree that in a year or two it may be too late for incompatible changes
12:20
<jgraham>
I think a year or two is optimistic
12:21
<jgraham>
Firefox 4 plans to ship by the year end
12:21
<jgraham>
I think
12:21
<othermaciej>
anyway, I think a server API that looks much like the client API would be *way* easier to use than any protocol you have to implement from scratch, so solutions of that form will dominate
12:21
<jgraham>
Chrome has an aggressive release schedule
12:21
<Rik`>
othermaciej: like micheil's API ?
12:21
<othermaciej>
they've already shipped WebSocket, probably more than once, and have made changes along the way
12:22
<othermaciej>
Rik`: I'm not familiar with that one - pointer?
12:22
<Rik`>
http://github.com/miksago/node-websocket-server/
12:22
<jgraham>
Yes, but like I said it is different if there is > 1 implementor
12:23
<jgraham>
Or at least enough that people rely on it
12:23
<othermaciej>
Rik`: from my quick scan, that looks pretty neat
12:23
<othermaciej>
jgraham: I think services deployed using it is the more significant metric
12:23
<othermaciej>
though it's true, most server operators will not use it til it hits some threshold of deployed client market share
12:24
<Rik`>
jgraham: there are already some libraries and servers that use WebSockets, fallback on Flash or long polling
12:24
<jgraham>
If you want substantial changes, I would push for something concrete as soon as possible, and try to explicitly keep exiting implementors in the loop
12:25
<othermaciej>
fortunately implementors read the list :-)
12:25
<jgraham>
The HyBi list? I would be surprised if everyone keeps up
12:26
<othermaciej>
I don't think I personally would insist on any changes other than to change the handshake if someone can come up with an option that is more secure and otherwise acceptable fairly soon
12:27
<othermaciej>
I'd also like to see support for binary frames added once there is a reasonable way to handle that in JS
12:27
<jgraham>
That has always been the plan for binary AIUI
12:27
othermaciej
wonders how the ECMA committee is doing on typed arrays
12:28
<gsnedders>
othermaciej: Nothing that on public list about it, at least
12:28
<remysharp>
any reason why Webkit might be **excluding** the referrer in the request header for video requests?
12:28
<othermaciej>
I know it was discussed in a recent f2f
12:36
<micheil>
Rik` / othermaciej ?
12:39
<micheil>
othermaciej: to be fair, I've not yet seen a doesn't tutorial telling anyone "this is how you implement the websocket protocol" — But I rely on the spec for that information
12:41
<micheil>
othermaciej: fwiw, I'm actually implementing a routing api addition to my websocket server, so you can essentially have different actions based on the different resources you can connect to.
12:41
<othermaciej>
micheil: neat!
12:42
<micheil>
Philip`: having server and browser use the similar API doesn't quite work, as the server is more evented, by I've tried to stick to something similar, while still following the node.js http/net module api's
12:43
<micheil>
othermaciej: for a list of protocol version support, see: http://wiki.github.com/miksago/node-websocket-server/browser-support/
12:45
<micheil>
othermaciej: as for other things in browser support, there may be more data out in the next week to two weeks, from what I've heard from various people.
12:47
<micheil>
I'm currently wanting to work with other developers of websocket libraries so that we have just one, and if you have something that want's to give a different api, or provide a different fallback, then it's just an extension on a standard websocket library
12:48
<micheil>
currently I'm working on ways that you can include parts into a websocket server, but not need them in the core of the code (which is more an issue surrounding environment over protocol)
12:49
<micheil>
jgraham: we could use me as a scale of implementability :P (I am after all just another javascript programmer)
12:58
<hsivonen>
Hixie: is it intentional that HTML5 doesn't URL-unescape the value of the name attribute on the <a> element?
13:17
<gsnedders>
hsivonen: yes
13:18
<remysharp>
 sorry chaps, dropped my connection - any word on video referrer headers?
13:19
<remysharp>
do we know if it's on purpose?
13:19
<Peter`>
remysharp: you could also try #chromium
13:19
<remysharp>
it affects Firefox too - which is why I wonder if it's the spec
13:19
<remysharp>
though I wouldn't have thought so - but wanted to check first
13:26
<annevk>
what is in the spec?
13:27
<Peter`>
http://www.whatwg.org/specs/web-apps/current-work/multipage/urls.html#fetch
13:38
<hsivonen>
gsnedders: thanks
13:39
<hsivonen>
gsnedders: I was reading Gecko's old sink code and noticed special treatment of one attribute... (<a name>)
13:40
<annevk>
ooh crap
13:41
<hsivonen>
annevk: ?
13:41
<annevk>
two days left to reply to their reply to my reply to their reply to my comments on WAI-ARIA
13:41
<annevk>
there's just no productive discussion at all
13:41
<annevk>
I give some input
13:42
<annevk>
and months later I get some committee response that is near useless
13:42
<annevk>
maybe I should just give up
13:43
<annevk>
maybe there'll be an abarth that fixes WAI-ARIA half a decade down the road and all be fine
13:43
<hsivonen>
annevk: aargh. I haven't responded, either. Bad email filtering.
13:44
<hsivonen>
Well, today, I'm going to go forward with parser feature work.
13:44
<hsivonen>
I'll never get software written if I just deal with email
13:45
<annevk>
what are you going to do with the XML parser btw?
13:45
<hsivonen>
annevk: moving it to use the machinery that the HTML parser uses for off-the-main-thread stuff
13:45
<hsivonen>
to unify the way the two parser manage incrementalism
13:46
<hsivonen>
I want to write an XML5 parser
13:46
<hsivonen>
but since Firefox uses DTDs for localization, it's not practical to simply zap DTDs
13:47
<jgraham>
I thought XML5 had DTDs?
13:47
<hsivonen>
and rewriting DTD handling would be quite a distraction
13:47
<annevk>
jgraham, not external
13:47
<jgraham>
Ah
13:47
<hsivonen>
jgraham: well, if it weren't for localization, I'd just argue for killing DTDs in any shape or form
13:48
<hsivonen>
if you have a bad optional feature, someone will do something useful with it, and it becomes harder to kill the feature :-(
13:48
<hsivonen>
just say NO to bad optional features before anyone uses them
13:49
<hsivonen>
also, I haven't in general yet figured out what to do about XUL and XBL
13:49
<hsivonen>
they don't build normal DOMs
13:49
<hsivonen>
they have dedicated tree builders in Gecko
13:49
<hsivonen>
(not the usual XML to DOM builder)
13:50
<hsivonen>
but that's post-Firefox 4.0 stuff
13:50
<hsivonen>
my current concerns are View Source and the sanitizer
13:50
<annevk>
XBL should get fixed by sicking's work right? if that is still ongoing that is
13:51
<hsivonen>
annevk: yeah
13:51
<annevk>
btw, HTTP WG meeting starts in 20min
13:51
<hsivonen>
I don't know exactly what the plan for killing off XBL1 from the *content* process is
13:51
<annevk>
http://tools.ietf.org/wg/httpbis/agenda has details
13:52
<hsivonen>
doh the *chrome* process
13:57
<annevk>
back to 800
13:57
<annevk>
emails
13:58
<annevk>
yay
13:58
<annevk>
also readied CORS for publication
13:58
<annevk>
now Media Queries
13:58
<nessy>
nice!
15:43
<MikeSmith>
https://labs.ericsson.com/blog/beyond-html5-implementing-device-and-stream-management-webkit
15:46
<hsivonen>
MikeSmith: do Ericsson products use WebKit GTK?
15:46
hsivonen
thought Ericsson was moving Android
17:13
<jgraham>
If you believe wikipedia, Hixie is slightly blurry at the edges
17:14
<Workshiva>
Only the edges?
17:15
<jgraham>
Well the rest could just be down to nice skin
17:15
<Workshiva>
He's also Stalin, according to one editor
17:19
<jgraham>
Ah, I see it is an enlargement from a group photo. Someone should really find one where a) it is actually a photo of him and b) he doesn't have the horizon growing out of his ears
17:19
<gsnedders>
Do you not think it's a good look on him?
17:20
<jgraham>
It makes him look egocentric
17:20
<jgraham>
"look at me, the whole world is spawed from my ears"
17:20
<jgraham>
*spawned
17:21
<jgraham>
(also there is an island or something which is hard to distingusih from the sea, so it looks like the horizon jumps as it goes through his head. That just freaks me out)
17:22
<gsnedders>
A headland of the bay, I think, from the angle
17:34
<dandaman>
<div id="backgrounddiv" name="backgrounddiv" style="border: thin solid #d9d9d9;border-width:2px;background:#f2f5fa;display:table" >
17:34
<dandaman>
that div tag covers everything in the body
17:34
<dandaman>
yet the table that is in it has a little margin on the page
17:34
<dandaman>
looks like 5 pixels from the left
17:34
<dandaman>
and like 4 from the top
17:35
<dandaman>
how can i get it so it fits 100% on the page
17:35
<TabAtkins>
<body> starts with some margin.
17:35
<dandaman>
ahh
17:35
<dandaman>
is there a way to get rid of it?
17:35
<TabAtkins>
The way you'd expect.
17:35
dandaman
is embarrassed
17:35
<dandaman>
i forgot the margin tag ;\
17:36
<dandaman>
:\
17:36
<dandaman>
margin: 0 auto?
17:36
<TabAtkins>
Sure, or just margin:0, depending on what you want.
17:36
<dandaman>
worked
17:36
<dandaman>
thanks man
18:12
<dandaman>
so i basically am using this code http://www.zurb.com/article/266/super-awesome-buttons-with-css3-and-rgba to make a cool button, i was wondering if anyone could tell me what the style name is to make that button fill an entire <td> cell would be
18:12
<dandaman>
css is kind of dead right now :\
18:12
<dandaman>
sorry that i keep pestering the channel
18:12
<TabAtkins>
height:100%;width:100%?
18:13
<dandaman>
ignorance on my part, thank you
18:17
<hsivonen>
hmm. so the new YouTube iframe has hard-coded width and height...
18:24
<hsivonen>
are there other known-dangerous CSS properties than -moz-binding, binding and anything whose value has expression()?
18:24
<TabAtkins>
In terms of exposing arbitrary scripting, or does it count to kill a page's performance to unusability?
18:25
<hsivonen>
I was just thinking about scripting
18:25
<TabAtkins>
In that case, I don't *think* so. I don't believe anyone runs scripts in SVG linked through CSS.
18:25
<hsivonen>
TabAtkins: are there properties that are so perf-sensitive that they should be cosidered dangerous?
18:26
<TabAtkins>
Sure. Shadows can kill a page with 30+ second scroll times.
18:28
<TabAtkins>
I suspect it's possible to attack a page with the ~ combinator, too.
18:29
<Philip`>
hsivonen: Do you care about things being dangerous to IE6 users?
18:29
<Philip`>
Apparently you can do things like list-style-image:url("javascript:alert('XSS')"); there
18:34
<Hixie>
jgraham: belies doesn't mean what you think it means
18:35
<hsivonen>
Philip`: thanks
18:38
<Hixie>
hsivonen: yeah, the weird non-escaping behaviour is intentional
18:38
<hsivonen>
Hixie: ok
18:39
<hsivonen>
(I think the unescaping behavior was weird)
18:40
<Hixie>
no kidding
18:51
<AryehGregor>
hsivonen, MediaWiki blocks filter:, accelerator:, and url(). That's meant to prevent information leaks as well as scripting.
18:51
<TabAtkins>
accelerator? What's that for?
18:51
<AryehGregor>
I have absolutely no idea.
18:52
<TabAtkins>
http://www.css3.com/css-accelerator/
18:52
AryehGregor
finds http://apiblog.youtube.com/2010/07/new-way-to-embed-youtube-videos.html, yay!
18:52
AryehGregor
subscribes to that blog
18:52
<TabAtkins>
Yus, very nice.
18:54
<AryehGregor>
I really have to commend Apple for forcing people to switch to HTML5 video. Awesome move on their part. Too bad Google doesn't take the same stand. Also too bad Apple is completely and irredeemably evil in other ways.
18:54
<jgraham>
Hixie: What do I think it means? i.e. how did I use it?
18:55
AryehGregor
switches his forum to use the new YouTube embed style
18:56
<AryehGregor>
hsivonen, the current YouTube embed code has hardcoded width and height too.
18:57
<AryehGregor>
Although a *different* hardcoded width and height, I think?
19:01
<Philip`>
AryehGregor: Surely nobody can be irredeemably evil
19:01
<Philip`>
Not even Darth Vader was
19:02
<AryehGregor>
Well, okay.
19:02
<AryehGregor>
I exaggerate.
19:07
<TabAtkins>
I love how Darth Vader is the Hitler of nerd-dom.
19:08
<jgraham>
TabAtkins: If I invoke Godwin's law at this point, what happens?
19:08
<TabAtkins>
That just declares an argument to be over. We aren't arguing yet, so it's a noop.
19:09
<jgraham>
I was hoping we could declare your position "I love how..." to be wrong
19:09
<Philip`>
Hitler would have been a poor example to pick, since as far as I'm aware he was not redeemed before his death, so he would have contradicted my argument
19:09
<Philip`>
Palpatine would have been an equally poor example
19:11
<AryehGregor>
Hitler committed suicide, does that count? Probably not.
19:12
<Philip`>
No
19:16
<Hixie>
jgraham: you used it to mean "reveals" but it really means "disguises"
19:18
<AryehGregor>
Hixie, where did he use it?
19:18
<Hixie>
hybi
19:19
<Hixie>
i only mention it because i keep trying to use the word but can't find a context in which it's correct to use it :-)
19:20
<Philip`>
"Dogs like it when you scratch their belies"
19:21
<AryehGregor>
"I think the current discussion on this list belies a false belief that we are still at a stage where the protocol can be trivially changed."
19:21
<AryehGregor>
Yeah, that's not right.
19:21
<AryehGregor>
It means the opposite.
19:21
<Hixie>
Philip`: lol
19:29
<hsivonen>
annevk: why is the intrinsic size of captured images of more interest than the size in terms of file pixels?
19:29
<jgraham>
Ah
19:29
<hsivonen>
annevk: and when would they differ?
19:31
<jgraham>
Hixie: I suspect I do that sort of thing quite often; anytime my general impression of where a word sounds good doesn't match its real meaning
19:32
<jgraham>
I probably meant to say "betrays"
19:35
<jgraham>
http://jeffkreeftmeijer.com/2010/experimenting-with-node-js/ is fun
19:36
<annevk>
hsivonen, high resolution displays
19:36
<annevk>
hsivonen, though actually I suppose if you haven't given them a size...
19:37
<hsivonen>
I'd expect the site to care about the file pixels when the user is about to upload an image
19:37
<TabAtkins>
jgraham: Haha, that's awesome.
19:38
<annevk>
hsivonen, yeah, I guess there is no difference, not sure what I was thinking
19:39
<annevk>
hsivonen, well, I looked at what HTMLImageElement and HTMLVideoElement did, but they are not too relevant
19:40
<Hixie>
jgraham: yeha like i said, i have the same problem :-)
19:42
hsivonen
had to look up what belie actually means. I bet "misrepresent" or "contradict" (depending on usage) are both better understood
19:45
<hsivonen>
google failed, btw, and offered me definitions of "bely" in Dutch when I searched for define:belies
20:08
<Philip`>
Hooray for PDF
20:08
<Philip`>
I'm looking at an ISO specification which says "... End system or level reachable address prefix ..."
20:08
<Philip`>
I assume there's meant to be either a 1 or a 2 in that gap
20:09
<Philip`>
The IETF's text version says "2" so presumably that's what it should be
22:03
<gsnedders>
jgraham: You were talking about ORMs coping cleanly with cases where some conditions were optional, got any code of one doing so?
22:03
gsnedders
can't see anything like that
22:24
<Philip`>
https://cybersecuritychallenge.org.uk/news.html - ooh, an HTML5 page
22:25
<Philip`>
and it looks like they might be using the sectioning elements sanely
22:28
<TabAtkins>
Yup, that outline looks pretty sane.
22:29
<Philip`>
https://cybersecuritychallenge.org.uk/conformance.html - and they made themselves a nice tick for the validator
22:30
<Philip`>
http://validator.nu/?doc=https://cybersecuritychallenge.org.uk/candidates/registration.html - too bad the site's not actually valid
23:13
<dandaman>
having a little trouble right now, so i'm testing this site on the android emulator(i have disabled scrolling because everything is supposed to fit on the screen at once). problem is whenever you click into a text field it brings up the keyboard on screen and pushes the site up and when the keyboard goes back down it stays pushed up
23:14
<dandaman>
anyone know how i can make the scroll go back to it's original height?
23:14
<TabAtkins>
Absolutely no clue. Phone browsers are voodoo magic.
23:16
<dandaman>
damn, alright, thanks anyway