03:23
<othermaciej>
TabAtkins: yeah I couldn't resist replying to that one, even though I should have
06:37
<lazni>
https://gist.github.com/f7d3e10233d436eb16dc comment on HTML6, I can't distill it down in one IRC line
07:52
<boblet>
krijnh: thanks for upgrading your sewing machine ;-) logs fully loading again
07:57
<JonathanNeal>
Hello all
07:58
<hsivonen>
hi
07:59
<JonathanNeal>
what's the haps, how goes html5?
08:06
<virtuelv>
huh?
08:06
<virtuelv>
http://www.w3.org/html/wg/wiki/HTML5_Authoring_Conformance_Study - why are there fewer, and different conformance errors on google.de compared to other googles?
08:06
<virtuelv>
s/, and different//
08:22
<hsivonen>
I need to switch to an OpenID provider whose authentication page doesn't break in Minefield.
08:24
<JonathanNeal>
So Wikipedia, Mozilla, and WHATWG get it right, virtuelv.
08:29
<virtuelv>
no, many sites gets it right
08:29
<virtuelv>
with regards to their declared doctypes and intended goals
08:32
<JonathanNeal>
well yes :)
08:34
<virtuelv>
the thing I find some comfort in, is that for the sites with many errors, the error count goes down when validated against HTML5
08:39
<boblet>
virtuelv: the very same reason will no doubt make many people agitated ;-)
08:41
<boblet>
hsivonen: are you planning to add an outliner to validator.nu?
08:46
<JonathanNeal>
virtuelv, how is that a good thing?
08:46
<JonathanNeal>
You think they'll be encouraged to fix their bugs if they have 10 instead of 20?
08:46
<JonathanNeal>
The most common validation error I see on corporate sites is improperly escaped / unescaped text.
09:08
<hsivonen>
boblet: yes, in some far future after a Firefox release has been shipped with the HTML5 parser
09:08
<boblet>
hehe
09:09
<boblet>
hsivonen: here’s hoping that far future isn’t too far, eg for block-level links etc
09:13
<hsivonen>
boblet: I thought we were going to enable the HTML5 parser on the trunk by the end of this month. However, now it seems that that's not going to happen.
09:14
<hsivonen>
boblet: I don't really know what the time scales are
09:14
<boblet>
hsivonen: that sucks ;( did something big come up to stop it from happening?
09:15
<hsivonen>
boblet: we don't have continuous integration test coverage for speculative parsing
09:15
<boblet>
hsivonen: well … that_sounds_ big… :? hehe
09:16
<boblet>
or at least impressive
12:26
<Dashiva>
This seems to be somehwat overstating the abilities of CSS transforms, or am I wrong? "Actually, because CSS3 implements the same functionality as the canvas element (which makes me believe that the canvas element becomes redundant then)"
12:42
<annevk>
presumably you can implement <canvas> with <div> and CSS3
12:42
<annevk>
but then you can implement CSS3 with <font> and DOM
12:42
<annevk>
and sufficiently small font sizes
12:47
<jgraham>
So we conclude that you can implement <canvas> with <font> and DOM and sufficiently small font sizes
12:47
<annevk>
indeed, same for SVG btw
12:47
<annevk>
I wonder why we bothered with all this stuff in the first place
12:48
<annevk>
anyway, time to go
12:48
<annevk>
new passport + presentation coming up
13:03
<hsivonen>
with <font> going through DirectWrite, you even get GPU acceleration if you implement stuff on top of tiny fonts!
13:17
virtuelv
goes back to backlog JonathanNeal / boblet : my point is that the HTML5 spec is closer to specifying real authoring practices
13:17
<virtuelv>
if everyone ceased from creating new documents on the web, HTML5 UAs should eventually become more compatible with that content
13:18
<virtuelv>
*interoperable
13:19
<boblet>
virtuelv: is this re: the wiki authoring conformance study?
13:19
<virtuelv>
boblet: yes
13:20
<virtuelv>
of course, people will debate whether a specification that actually takes common legacy implementations into account is jumping the shark or not
13:21
<boblet>
my take was that those who e.g. advocate draconian error handling will see this as a watering down, a move in the wrong direction
13:22
<virtuelv>
I don't see how anyone can advocate draconian error handling
13:22
<Philip`>
Dashiva: I expect doing e.g. dynamic edge filtering of videos would be quite tricky with CSS transforms
13:22
<virtuelv>
(but that's a different matter)
13:23
<virtuelv>
draconian error handling is only possible on the syntactic level when you abstract away the syntax and let tools serialize your documents
13:24
<boblet>
or when you start from scratch and decree it
13:25
<Philip`>
boblet: Like with RSS?
13:26
virtuelv
has his name in RFC 4287
13:26
<virtuelv>
rss is still not gone :/
13:26
<boblet>
had TBL made draconian error handling a basic tenet of HTML from the outset it’s possible that the web would be using it
13:26
<virtuelv>
boblet:
13:26
<virtuelv>
the short answer is "absolutely not"
13:27
<Philip`>
boblet: I expect the browser developers would have just ignored him
13:27
<boblet>
but I completely certain the web would be a shadow of its current state
13:27
<virtuelv>
my thesis is that had TBL made error handling draconian, the web wouldn't have taken off
13:27
<Philip`>
since they could make their products better for authors and users by violating the spec
13:27
<boblet>
virtuelv: yep, that’s what I think
13:28
<boblet>
Philip`: I don’t think there would have been a need, as browsers would only have been made/bought by academia and hobbyists
13:28
<Philip`>
http://www.walterzorn.com/jsgraphics/jsgraphics_e.htm - <canvas>-style graphics using <div>
13:28
<Philip`>
so it's not merely possible, it's already implemented
13:30
<virtuelv>
Philip`: I seem to recall digg at some point using that library
14:34
Philip`
continues to think that HTML5 shouldn't defer to RFC2119 for definitions of "must" etc, because RFC2119 doesn't make sense in the context of authoring requirements
14:35
<MikeSmith>
hmm, yeah, true
14:36
<Philip`>
(It's written about vendors implementing behaviours)
14:37
<MikeSmith>
Philip`: is it really, though?
14:37
<MikeSmith>
I'm reading it now and I don't immediately see anything that makes it bound so closely to implementations
14:38
<MikeSmith>
to being about requirements for implementors
14:38
<Philip`>
"the full implications should be understood and the case carefully weighed before implementing any behavior described with this label"
14:38
<MikeSmith>
ohg
14:38
<Philip`>
That's about implementing behaviours
14:38
<MikeSmith>
yeah, and "enhances the product while another vendor may omit the same item"
14:39
<Philip`>
The "MAY" definition is all about vendors and implementations
14:39
<MikeSmith>
right
14:39
<MikeSmith>
I see that now
14:39
<MikeSmith>
the Security Considerations section also
14:40
<Philip`>
Some of the definitions are more generic, but if the goal was for RFC2119 to apply to non-implementations requirements too then presumably the whole thing would have been written in a more generic way
14:40
<MikeSmith>
yes
14:40
<MikeSmith>
though perhaps it doesn't really matter what the original intent was
14:41
<Philip`>
It matters if people start trying to selectively interpret the intent of the original phrasing
14:42
<jgraham>
FWIW I agree with Philip` to the extent that RFC 2119 is clearly not intended to deal with authoring requirments when the client is liberal
14:42
<jgraham>
(or perhaps in general)
14:42
<MikeSmith>
Philip`: it is a mistake for them to do that
14:42
<jgraham>
I don't think this is a big problem except insofaras people try to spac lawyer as if it was
14:42
<Philip`>
I think the use of must/should/etc for authoring requirements is useful and sensible and good, but specs should be clear that it's a wilful violation of RFC2119
14:43
<Philip`>
so that people can't validly claim that the spec is silently inconsistent with the other specs it normatively references
14:45
jgraham
would prefer people just stopped that because it is obviously a trivial inconsistency, hasn't been a problem for other specs, and only matters to spec lawyers anyway
15:04
Philip`
wonders if 'it should be feasible to explain to authors exactly how to write valid HTML' is one of the criteria that Sam is missing
15:06
<Philip`>
(where 'to write an attribute value containing a '&', you can write it escaped or you can write '&' unless it's followed by one of these 106 strings of characters in unescaped form' is not considered a feasible explanation)
15:07
<Philip`>
(...'unless those strings are themselves followed by an alphanumeric character' or whatever the rules are now)
15:28
<MikeSmith>
interesting that an engineer from Microsoft Research is implementing a conformant HTML5 parser
15:28
<MikeSmith>
http://lists.w3.org/Archives/Public/public-html-comments/2010Mar/0014.html
15:28
<MikeSmith>
I wonder what language he's coding it in
15:31
<Philip`>
http://langnetsymposium.com/2009/speakers.aspx - "Prior to joining MSR, Herman worked in the Developer Division, initially on JScript .NET. After completing the JScript .NET project Herman worked in collaboration with MSR on Comega and Spec#. Since joining MSR on a full time basis, Herman has worked on Spec#, the Verifying C compiler, CCI version 2 and lately on JScript again (initially in the context of the Gazelle browser project)."
15:32
<Philip`>
Sounds like he largely works on stuff around C#, so maybe that
15:33
<MikeSmith>
that would make sense, I guess
15:33
<Philip`>
Gazelle uses C# too
15:34
<Philip`>
(but reusing parts of Trident, like the parser, in the version in the published paper)
15:38
<MikeSmith>
I seem to vaguely remember previously hearing about someone already working on a C# HTML5 parser
15:38
<MikeSmith>
but maybe I imagined it
15:38
<MikeSmith>
anyway, it would seem to be nice to have one
15:39
<Philip`>
http://code.google.com/p/twintsam/
15:41
<MikeSmith>
ah yeah
15:41
<MikeSmith>
I rememeber that name now
15:42
<MikeSmith>
Thomas Broyer
15:42
<MikeSmith>
me tries to remember what "twinstam" means
15:45
<Philip`>
The Web Is Not Tag Soup Any More
15:46
<Dashiva>
Soup Tags?
15:50
<lazni>
<water><rice/><meat state=minced/></water>
15:51
<Dashiva>
I'd use a generic <minced> wrapper element to allow mincing arbitrary ingredients
16:06
<MikeSmith>
heh
16:06
<MikeSmith>
I want a <mike> tag
16:07
<MikeSmith>
we have that tag named after Mark Pilgrim already
16:07
<MikeSmith>
so it would only seem fair
16:08
<beowulf>
if you write enough documents with a <mike> tag doesn't it have to be in html6?
16:08
<Dashiva>
Not if it doesn't have special behavior
16:08
<beowulf>
i could wave my hands around, would that get it in?
16:09
MikeSmith
ponders what the behavior of the <mike> tag should be
16:10
<lazni>
<device type=microphone>
16:22
<Wolfman2000>
Morning/afternoon. As a reminder, is this the primary HTML5 channel on IRC?
16:23
lazni
looks at the channels on the log site in /topic
16:23
<Philip`>
It's where most discussion about the spec takes place
16:24
<Dashiva>
http://masinter.blogspot.com/2010/03/browsers-are-rails-web-sites-are-trains.html
16:24
<Wolfman2000>
...then perhaps you guys can be of assistance. I'm still a bit confused by the purpose of the <canvas> element. Is it supposed to allow for drawing whatever you want on it?
16:25
<Dashiva>
It seems rail/train doesn't really match how HTML works. It seems more like ocean/anything-that-floats
16:25
<Dashiva>
(It seems I like saying "It seems")
16:25
<Philip`>
Wolfman2000: Yes
16:25
<lazni>
programmatically
16:25
<Philip`>
Yes
16:26
<lazni>
no SVG importing?
16:26
<Wolfman2000>
Can this canvas be given a width and height, but with a height "cap" of sorts to force scrolling in one direction or another?
16:26
<Dashiva>
lazni: You can drawImage svg onto a canvas
16:26
Philip`
was interpreting "you" as meaning a page developer, not a user
16:26
<Philip`>
Dashiva: Only in Opera
16:26
<Philip`>
(I think)
16:26
<Dashiva>
Probably
16:26
<Philip`>
and not in the spec
16:26
<Philip`>
(except via <img src=foo.svg>)
16:26
Wolfman2000
is more a web developer than a user these days.
16:26
<Dashiva>
But I expect it will spread
16:27
<Philip`>
Wolfman2000: It has a width and a height, just like an <img>
16:27
<lazni>
Wolfman2000: you want something like http://yuiblog.com/assets/slicing/vr_test.html?
16:27
<lazni>
blog post: http://www.yuiblog.com/blog/2008/06/23/slicing/
16:27
<Wolfman2000>
hmm...unsure the best way to phrase this.
16:28
<Philip`>
Wolfman2000: It's just a rectangle on the page (of the given width/height), and scrolls like any other static object on a page
16:29
<Wolfman2000>
I want to turn a java application that I have and make it use the power of HTML5. http://beta.pumpproedits.com/java/PPEdits.zip <-- that's the app and the libraries. It's meant for creating machine dance step data for an arcade game, and the charts scroll downward.
16:30
Philip`
can't easily run Java applications
16:30
<Wolfman2000>
I'll get you a screenshot then if I can.
16:31
<Philip`>
If you want some kind of scrolling within the content of the <canvas> (rather than having it be like a large <img> where the whole page can scroll to reveal parts of it), then you'll just have to write code to detect user input and to offset all your drawing functions
16:31
<Wolfman2000>
http://tinypic.com/3ia31x0i Not just scrolling. Adding elements.
16:32
<Wolfman2000>
Either way, I can "possibly" do this with <div>s and lots of <img>s. I just want to try to take advantage of what HTML5 has to offer.
16:33
<Philip`>
Might be easier to do it with SVG than with canvas
16:33
<JonathanNeal>
heyo
16:34
<Wolfman2000>
Philip`: javascript and SVG? Well, the charts that I generate are SVG already.
16:35
<Wolfman2000>
http://beta.pumpproedits.com/chart/quick/143/rhythm <-- one of the recent ones, showing off a new game mode. (Firefox views it best)
16:35
<Wolfman2000>
Still...I could have thought plain <svg> files couldn't have a <script> or <title> tag.
16:35
<Wolfman2000>
How exactly can I combine SVG and Javascript?
16:38
<Philip`>
Wolfman2000: You can use <script>s in SVG files, and also you can embed SVG inside an XHTML page (sent as application/xhtml+xml, not text/html yet) in order to use normal HTML form inputs and scripts and so on
16:39
<AryehGregor>
You can embed SVG in text/html in Firefox with html5.enable and in IE9, and probably in other browsers at some point.
16:39
<Wolfman2000>
html5.enable is at true right now
16:40
<Philip`>
Wolfman2000: Not by default
16:40
<Wolfman2000>
...wonder how I turned it on then
16:40
<AryehGregor>
Then you can stick <svg><circle r=100 full=red></circle></svg> in your HTML documents and get SVG as part of the DOM. (Except I don't know SVG, so that might not work.)
16:40
<AryehGregor>
(Do void elements auto-close in SVG-in-text/html?)
16:41
<Philip`>
Restricting yourself to IE9 + Firefox-with-non-standard-options is probably a bit excessively anti-user
16:41
<AryehGregor>
Maybe.
16:41
<Wolfman2000>
Then please suggest a good alternative
16:42
<Philip`>
Doing it in application/xhtml+xml will work in every browser except IE8 (and older IEs)
16:42
<Philip`>
AryehGregor: You have to say <circle></circle> or <circle/>
16:42
<Philip`>
The parser doesn't know the details of all SVG elements
16:43
<AryehGregor>
Ah, so <circle /> does work.
16:43
<AryehGregor>
This works in Firefox with html5.enable: data:text/html,<!doctype html><svg width=200 height=200><circle cx=100 cy=100 r=100 fill=red /></svg>
16:43
<AryehGregor>
Hurrah.
16:43
<Wolfman2000>
Philip`: since I'm using HTML5 on my website primarily, what requires changing for it to work as application/xhtml+xml?
16:43
<AryehGregor>
Wolfman2000, IE < 9 doesn't work with application/xhtml+xml.
16:43
<AryehGregor>
Nor does it support SVG at all, in fact.
16:43
<Wolfman2000>
AryehGregor: besides IE
16:43
<Philip`>
AryehGregor: and circle isn't an empty element in SVG anyway
16:43
<AryehGregor>
Oh, then you just have to make sure it's well-formed, or else it will die with a fatal error.
16:44
<AryehGregor>
So it's kind of inconvenient.
16:44
<AryehGregor>
Probably easier to use <object> for now. You can still have <script> inside the SVG file itself, as I understand it.
16:45
<Wolfman2000>
so HTML5 and XHTML use the same thing?
16:46
<Wolfman2000>
...I'll rephrase better. All I have to do is make sure my HTML5 is well formed and pass the right mime type, and it magically works?
16:47
<Philip`>
Yes, except it's not magic because you're following all the steps needed to make it work :-)
16:48
<Wolfman2000>
I wasn't expecting magic
16:48
<Wolfman2000>
I think most of my code is well formed as is at least.
16:48
<Philip`>
It's generally very difficult to guarantee well-formedness, so you should probably avoid it as much as possible
16:48
<Philip`>
and only use XHTML for pages that really require it
16:49
<Philip`>
(There's also quite a few subtle behavioural differences, which can break things)
16:49
<Wolfman2000>
Philip`: the only pages that require XML are the generated charts...and hopefully the new javascript chart editor I want to get working.
16:49
<Wolfman2000>
so...what's the best validator to use in this case? W3 doesn't offer XHTML5
16:50
<Philip`>
AryehGregor might be sensible in suggesting using <object> instead, so most of your page is plain old HTML and it just embeds a minimal amount of SVG XML in the <object>
16:50
<Philip`>
and you can put scripts in the outer HTML page, and have them manipulate the inner SVG page's DOM
16:51
<Philip`>
http://html5.validator.nu/ should work
16:54
<lazni>
Wolfman2000: also be carefull with user input
16:54
<lazni>
unicode encoding and such
16:55
<Wolfman2000>
...no more summary attribute? huh
16:56
<lazni>
data-summary and provide your own behaviour
16:56
<Wolfman2000>
I meant for <table>
16:56
<Wolfman2000>
(yes, I have a valid use for table)
16:57
<lazni>
I meant as a workaround
16:57
<Philip`>
http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html#attr-table-summary
16:58
<AryehGregor>
summary is allowed, although it's controversial.
16:58
<Philip`>
The bit above that has lots of suggested better ways of expressing summary-style text
17:02
<Wolfman2000>
Lots to consider...if nothing else, thanks for the updates
19:34
<JonathanNeal>
Hello World
19:37
<lazni>
XML parsing error: missing XML prolog
19:38
<JonathanNeal>
Sounds like my Hello World test worked better than yours, lazni :P
20:17
<gsnedders>
Hello world!
20:27
<annevk>
when is Ubuntu finally going to work with projectors normally?
20:27
<lazni>
maybe with KMS
20:28
<lazni>
I think Lucid will use KMS with nouveau and intel
20:28
<annevk>
I had to actually remove xorg.conf, log out, and log in again to fix issues after letting Ubuntu write to it to make it recognize my normal resolution again
20:28
AryehGregor
guesses it will be a nonissue in less than five years
20:28
<annevk>
also, several icons are now in the wrong place
20:28
<AryehGregor>
GNOME panels always randomly reorganize themselves when you log out and back in.
20:28
<AryehGregor>
It's infuriating.
20:28
<AryehGregor>
Xfce panels are far superior in all respects.
20:29
<annevk>
it's mostly weird they haven't fixed this
20:29
<annevk>
you'd think fundamental problems like this would have been fixed ages ago
20:29
<AryehGregor>
That's not the kind of remark I'd expect from a software developer.
20:29
<annevk>
using a projector is sort of an important use case
20:30
<AryehGregor>
I'd have thought you'd realize that things that seem obviously important to you might be less important or more difficult to fix than you'd think.
20:30
<jgraham>
Argh, now I've missed the chance to make the "more functional hello world" joke
20:30
<AryehGregor>
Most people don't use projectors ever. You give talks all the time, so you use projectors regularly.
20:30
<AryehGregor>
What use does a normal person have for a projector?
20:30
<AryehGregor>
I don't think I've ever used one in my life.
20:31
<annevk>
lots of people give talks
20:31
<jgraham>
Which would of course be a fibonacci sequence generator...
20:31
<jgraham>
AryehGregor: Lucky you
20:31
<AryehGregor>
A tiny minority of people give talks.
20:31
<annevk>
www-tag email is now classified as spam
20:31
<annevk>
way to go Opera Mail
20:31
<jgraham>
AryehGregor: I'm not sure that is true
20:32
<AryehGregor>
annevk, wait, so is that a false positive or not?
20:32
gsnedders
has got rejeced by Edinburgh uni
20:32
<jgraham>
Presentations are common in business and in academia
20:32
<AryehGregor>
jgraham, I don't give talks, therefore nobody does. Very simple.
20:32
<AryehGregor>
Yes, yes, I know.
20:32
<gsnedders>
*rejected
20:32
<annevk>
meetings and presentations are all the rage in corporate culture
20:32
<annevk>
also in university
20:32
<AryehGregor>
Most people are not in business or academia, or if they're in business they're too low on the pecking order to give talks.
20:32
<jgraham>
gsnedders: Oh. Their loss really
20:32
<annevk>
also what jgraham said
20:32
<AryehGregor>
gsnedders, :(
20:32
<AryehGregor>
Anyway, obviously Ubuntu hasn't received too many complaints about projectors not working, so that favors my hypothesis.
20:32
<annevk>
projectors work
20:33
<gsnedders>
jgraham: I told you I should've applied for English ;P
20:33
<AryehGregor>
Not working well. Whatever.
20:33
<jgraham>
gsnedders: Don't make me come over there
20:34
<annevk>
it's just that a) it requires Ubuntu writing to xorg.conf b) because of that it requires logging out and logging in c) because of the xorg.conf changes my original resolution can no longer be selected (which is somewhat weird admittedly, 16:10) d) it therefore requires me to sudo rm xorg.conf and log out and log in again...
20:34
<annevk>
none of this should be needed, of course
20:35
<gsnedders>
jgraham: I guess I'll have to follow the example of others on the Carakan team and go to Glasgow :P
20:35
<AryehGregor>
Why does it need to write to xorg.conf?
20:35
<AryehGregor>
Doesn't X autoconfigure these days?
20:37
AryehGregor
has written a 566-word post in response to the authoring conformance thread: trim down or post?
20:37
<AryehGregor>
566 words isn't *that* long.
20:38
<gsnedders>
Shorter than what I end up writing without meaning too...
20:38
AryehGregor
trims it down some
20:38
<Dashiva>
AryehGregor: Do you have a blog?
20:38
<AryehGregor>
Dashiva, no.
20:38
<AryehGregor>
I've been meaning to start posting blog-like stuff on aryeh.name, but never got around to setting it up.
20:38
<AryehGregor>
I'm too elitist to use Wordpress or something, you see.
20:38
<AryehGregor>
So I'd have to write it.
20:38
<Dashiva>
You could write a summary and paste the full entry below it
20:39
<jgraham>
post === email?
20:39
lazni
writes an atom feed, doesn't need html
20:40
<annevk>
so based on the IP address google.com.hk is either in simplified or traditional Chinese?
20:40
annevk
didn't quite understand that part of the announcement
20:40
<Dashiva>
The IP address...
20:41
<annevk>
and why the difference?
20:42
AryehGregor
gets it down to 528 words and gives up
20:43
jgraham
wonders if he should reply to Sam
20:48
<AryehGregor>
Which post?
20:49
<AryehGregor>
"You seem to suggest that the policy is: scream loudly enough and you too can get your favorite pet peeve disallowed"
20:49
<AryehGregor>
Wait, that isn't official policy?
20:49
<Dashiva>
Sure seems to me
20:53
<JonathanNeal>
I like that policy.
20:53
<JonathanNeal>
no ...
20:53
<JonathanNeal>
I LOVE THAT POLICY!
20:55
<Dashiva>
You need more exclamation points
20:57
<gsnedders>
Maybe I should just stay in Sweden...
21:03
<JonathanNeal>
exclamation points work like <h1-6> but opposite
21:03
<annevk>
hmm, per broadband.gov the US has more ambitious goals than what I have now
21:03
<JonathanNeal>
! (is of most importance) ! and !! (! is of secondardy importance, !! is of most importance)
21:04
<TabAtkins>
Theoretically, or practically? I find that !!!!!!!!! is generally less important than a *period*.
21:04
<JonathanNeal>
WOW THAT'S REALLY IMPORTANT!!!!!!!!!!!!!!!!!!!!!
21:04
<annevk>
100/50 mbits is pretty good
21:04
<MikeSmith>
AryehGregor: well stated
21:04
<MikeSmith>
(your message on the authoring-conformance thread)
21:05
<Dashiva>
Little-endian exclamation points
21:05
<MikeSmith>
gsnedders: who went to Glasgow?
21:05
<gsnedders>
MikeSmith: Sigbjorn Finne
21:05
<MikeSmith>
ah
21:05
<MikeSmith>
I don't think I know him
21:06
<MikeSmith>
is beer and other alcohol as expensive in Sweden as it is in Norway
21:06
<gsnedders>
Less expensive
21:06
<MikeSmith>
do they have a beer in Sweden that's as bad as Rignes?
21:10
<Dashiva>
I like your message, AryehGregor
21:10
<Dashiva>
It didn't even seem that long
21:10
<jgraham>
MikeSmith: I think he is relatively new
21:11
<jgraham>
gsnedders: Presumably if you were planning on following grand Opera traditions you would go to Linköping univeristy
21:11
<gsnedders>
jgraham: And fall in love with LysKOM?
21:12
<jgraham>
AryehGregor: You mean which post from Sam? The one in the "bug 7034" thread
21:12
<jgraham>
gsnedders: I doubt it ould brain damage you that much
21:12
<MikeSmith>
ah, LysKOM
21:12
<jgraham>
*could
21:12
<gsnedders>
jgraham: What are you trying to imply!?
21:13
<MikeSmith>
gsnedders, gsnedders - have you come to realize the brilliance of LysKOM yet?
21:13
<gsnedders>
MikeSmith: Don't tell me you're a kommie!
21:13
<MikeSmith>
it's the greatest
21:13
<MikeSmith>
what do we need IRC for, when we have LysKOM?
21:13
<gsnedders>
I thought you were fucked up before, but now...
21:13
<jgraham>
What could be btter than a messanging system that allows for threads to form arbitary directed graphs?
21:14
<jgraham>
It makes designing good UI so easy
21:14
<MikeSmith>
or even, what do we need e-mail for? or the Web?
21:14
<gsnedders>
jgraham: Email does that to
21:14
<gsnedders>
jgraham: You just have a In-Reply-To header with an arbitrary list of message ids
21:14
<jgraham>
gsnedders: Curiously all email clients just pretend like that is impossible
21:15
<Dashiva>
_You_ try designing a good graph-based UI
21:15
<jgraham>
Dashiva: Well I might design the protocol in such a way that it didn't exclude good UI in the first place
21:16
<gsnedders>
jgraham: So do semi-sane KOM clients
21:16
<Dashiva>
jgraham: How would you design a graph-based protocol so as to not require graph-based UI?
21:17
<jgraham>
Dashiva: I would not design a graph-based protocol
21:17
<jgraham>
Or rather I would limit it to trees
21:19
<jgraham>
(I certianly wouldn't tout the graphiness as a desirable feature)
21:20
<jgraham>
gsnedders: There are semi-sane KOM clients?
21:20
<gsnedders>
jgraham: Maybe
21:23
<Dashiva>
jgraham: In that protocol that is email + ignoring references that would create cycles, that is already the case
21:23
<Dashiva>
So it seems like a solved problem
21:46
<gsnedders>
jgraham: http://www.computerra.ru/terralab/softerra/516485/ --- interview with jl
21:46
<gsnedders>
(apparently)
21:54
<jgraham>
google translate ftw
21:56
<jgraham>
"On the singularities of the new interpreter Javascript "Computerra-line" said Jens Lindström (Jens Lindström), a leading developer Carakan."
21:56
jgraham
would like to make it clear that contrary to popular belief we are not actually harnessing the power of black holes to make Carakan faster
21:59
<Dashiva>
If you did, I imagine the technology race would get a lot more exciting
22:11
<jgraham>
Well I guess it would at least explain the ALIEN- bugs in the changelogs :)
22:14
<Dashiva>
"Safari reclaims the throne on sunspider using actual sun to power their javascript engine"
22:28
Hixie
returns
22:29
<jgraham>
Oh well, there goes the neighborhood
22:29
JonathanNeal
is swept away.
22:29
<JonathanNeal>
Here crawls back the neighborhood.
22:29
JonathanNeal
crawls back.
22:30
<jgraham>
By which I mean of course "welcome back. Have you been doing anything fun?"
22:33
<gsnedders>
jgraham: shhh. you're not meant to comment on impl detail like our black-hole magic!
22:33
<gsnedders>
Hixie: Anolis 2 is close ;)
22:34
<Hixie>
jgraham: spent thursday-saturday reffing this year's silicon valley regional of the FIRST robotics competition
22:34
<Hixie>
and sunday sleeping
22:34
<Hixie>
i fear my inbox
22:34
<gsnedders>
Inboxes are overrated
22:35
<Philip`>
Select all, mark as read
22:35
<Philip`>
Anything important should be in Bugzilla anyway
22:36
<boblet>
I’d fear Hixie’s inbox every day
22:36
<gsnedders>
s/'s inbox//
22:37
<othermaciej>
hello Hixie
22:37
gsnedders
remembers he was going to ask othermaciej something, but forgets what.
22:37
<gsnedders>
Why can't people just be around when I want them? :P
22:38
<othermaciej>
Hixie: there are some bugzilla bugs where I'd like to request expedited handling if possible, but I will give you a day or two to deal with the inbox
22:38
<othermaciej>
gsnedders: sorry!
22:38
<Hixie>
othermaciej: can you mark them P1?
22:39
<othermaciej>
Hixie: sure, I was going to ask you if that was a good way to do it
22:39
<Hixie>
yes please
22:39
<jgraham>
gsnedders: About webkit bugs?
22:41
jgraham
doubts it but it is the only conversation he had with gsnedders today where othermaciej would be an obvious source of information
22:42
<gsnedders>
jgraham: It wasn't today I wanted him.
22:45
<jgraham>
gsnedders: Oh well in that case it doesn't matter that you can't remember today :p
22:47
gsnedders
is somewhat disturbed by the fact he thinks he knows almost everything in the first year CS course at Glasgow
22:50
<annevk>
hmm, talk about bad timing: http://html5.org/tools/web-apps-tracker?from=4857&to=4858
22:51
<Hixie>
bad timing?
22:52
<jgraham>
gsnedders: So try going to some more challenging classes like extra Maths or whatever
22:52
<Hixie>
annevk: ?
22:53
<annevk>
I posted something on my blog just before
22:53
annevk
updates now before sleeping
22:53
Hixie
looks
22:53
<Hixie>
oh heh
22:53
<gsnedders>
jgraham: Can't with a C at AH
22:53
<Hixie>
you had me worried that i'd posted some edit after someone was complaining that the worst thing i could do would be to work on this or something :-)
22:54
<gsnedders>
jgraham: My AH grades prohibit me from doing anything like that, although I know the vast majority of the first year material. It sucks.
22:54
<annevk>
Hixie, ah, no worries :)
22:54
<jgraham>
gsnedders: You can't take the exam. I guess you can go to the lectures or something
22:54
<jgraham>
(or, er, ask nicely. It may be that the rules aren't as rigid as you suppose)
22:55
<gsnedders>
jgraham: I got the impression before that they really didn't like breaking them
22:56
<jgraham>
gsnedders: Even so
22:56
<jgraham>
Where is the actual first year content listed?
22:57
<gsnedders>
In random PDFs
22:57
gsnedders
won't be able to find the URLs again
22:57
<gsnedders>
Ooooh! Awesome! I'd get to learn Javascript!
22:57
<gsnedders>
At last!
22:58
<gsnedders>
I might finally get my head around that language!
22:59
<annevk>
sounds like it's bedtime gsnedders
22:59
<annevk>
though come to think of it you always act like this :p
22:59
<jgraham>
gsnedders: Seriously, urls
22:59
<jgraham>
Also: why "computing science"?
22:59
annevk
goes to bed himself
22:59
<annevk>
nn
23:00
gsnedders
can't sleep
23:00
<gsnedders>
jgraham: For no real reason, as far as I can tell
23:01
gsnedders
wishes he had any idea about what he wanted to do
23:01
<jgraham>
I don't think it even makes grammatical sense
23:01
<jgraham>
(it sounds like an attempt to calculate science)
23:01
<jgraham>
Anyway it was bedtime a while ago
23:02
<gsnedders>
Also, some of my information comes from what was said and given out at the open day
23:02
<gsnedders>
I can't find all of it online
23:04
gsnedders
should blatantly have applied for Scandavian Studies at Edinburgh
23:07
<gsnedders>
Or Gaelic.
23:10
<Hixie>
wow, help@ got a lot of traffic
23:33
<TabAtkins>
Hixie: help@?
23:37
<gsnedders>
help⊙wo