search of the:   for: 
code logs -> 2010 -> Thu, 03 Jun 2010< code.20100602.log - code.20100604.log >
--- Log opened Thu Jun 03 00:00:22 2010
00:04 You're now known as TheWatcher[zZzZ]
00:15 Attilla [Attilla@Nightstar-dc2d92b4.threembb.co.uk] has quit [Client closed the connection]
00:17 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: Beam them up, Scotty!]
00:19 AbuDhabi [annodomini@Nightstar-2075dedf.adsl.tpnet.pl] has quit [[NS] Quit: ZZZzzz...]
01:04 Derakon[AFK] is now known as Derakon
01:05 RichardBarrell [mycatverbs@Nightstar-58acb782.cable.virginmedia.com] has quit [Connection closed]
01:45 Serah [Z@26ECB6.A4B64C.298B52.D80DA0] has quit [Ping timeout: 121 seconds]
02:21 Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code
03:06 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code
03:20 gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?]
03:30 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code
04:29
< celticminstrel>
Well, restarting using execvp is great. Now I just need to rig it to not attempt restart if a syntax error is introduced.
04:30
< celticminstrel>
Can't be too hard, right?
04:45
< celticminstrel>
...I guess compile(open("xxx.py").read(),"xxx.py","exec") will have to do...
04:47
<@ToxicFrog>
o.O
04:47
< celticminstrel>
?
04:57 Rhamphoryncus [rhamph@Nightstar-bbc709c4.abhsia.telus.net] has joined #code
04:59
< celticminstrel>
I haven't yet had any lag since upgrading the RAM, so it seems that likely was the source of all my troubles.
05:02
<@ToxicFrog>
celticminstrel: so...your python main manually loads and compiles another python script and, if it errors, calls execvp() unless it was a syntax error?
05:03
< Derakon>
He's trying to find a way to remotely restart his script so he can reload changes to the program.
05:04
<@ToxicFrog>
Aah.
05:04
< celticminstrel>
Precisely.
05:04
< celticminstrel>
And it works, by the way.
05:07
< celticminstrel>
(Replacing "xxx.py" with the name of the file that contains the above code.)
05:07
< celticminstrel>
On an unrelated note, is there some document that explains IRC formatting? Particularly colours?
05:10
< Namegduf>
"Google and a certain willingness to interpret terribly written documents aimed at people writing bots in mIRC script"
05:10
< Namegduf>
I don't remember them all.
05:11
<@Vornicus>
Colors in IRC: ^K11,6 will give text color 11 and background color 6. I don't remember what those two colors are.
05:11
<@Vornicus>
%K11,6Let's find out!
05:11
<@Vornicus>
that didn't work.
05:11
< Namegduf>
^C is the character
05:12
< celticminstrel>
Colour is \x03
05:12
< Namegduf>
And "good" IRC clients won't interpret %<letter> as anything because it's non-standard, but some might.
05:12
< Namegduf>
"Ignore them" is the rule, it only matters for colour stripping.
05:12
< Namegduf>
They *shouldn't* emit it.
05:13
< Namegduf>
If they do, it'll show as garbage for other people.
05:13
< celticminstrel>
The main thing I'm wondering about is leading zeros on the colour numbers; mIRC seems to be sending them, but my client doesn't.
05:13
<@Vornicus>
Well, no, I was trying to convince xchat to emit
05:13
< Namegduf>
celticminstrel: Ah.
05:13
< Namegduf>
Easy explanation there.
05:13
< Namegduf>
They don't matter persay...
05:13
< Namegduf>
BUT if you want to send a number directly after a colour code
05:14
<@Vornicus>
%B is the bold tag in xchat, for instance, and it translates in sending.
05:14
< Namegduf>
And said code is single-digit
05:14
< Namegduf>
You need to include the 0.
05:14
< celticminstrel>
Ah, so it's probably wise to include them then.
05:14
< Namegduf>
^C410,000 <-- wrong.
05:14
< Namegduf>
0,000
05:14
< Namegduf>
Is how it shows, colour might vary between clients.
05:14
< Namegduf>
^C0410,0000 <-- right.
05:15
< Namegduf>
10,000
05:15
< Namegduf>
Which should be a red (4 being red) 10,000 in properly working clients.
05:15
< celticminstrel>
So, is technically any number up to 99 a valid colour?
05:15
< Namegduf>
No, 0 to 16
05:15
< Namegduf>
Or 15
05:15
< Namegduf>
But clients may interpret others as they feel like
05:16
< Namegduf>
And will generally consume up to two digits always even if that makes a bad colour.
05:16
< celticminstrel>
?
05:16
< Namegduf>
Might be standardised somewhere, but it's certainly defacto.
05:16
< celticminstrel>
That question mark had colour #17.
05:16
< Namegduf>
I didn't see it at all.
05:16
< Namegduf>
Maybe my client made it black.
05:16
< celticminstrel>
Likely.
05:17
< celticminstrel>
My client seems to interpret invalid colours as black.
05:17 Derakon [Derakon@Nightstar-5213d778.ca.comcast.net] has left #code []
05:17 Derakon [Derakon@Nightstar-5213d778.ca.comcast.net] has joined #code
05:17 mode/#code [+o Derakon] by Reiver
05:17
< celticminstrel>
Oh, one other question. What is \x0f for?
05:17
< Namegduf>
Hmm.
05:17
< Namegduf>
15
05:17
< Namegduf>
Don't remember.
05:17
<@Vornicus>
^O clears formatting
05:17
< Namegduf>
Oh, yes.
05:17
< celticminstrel>
So, it's not actually necessary if your entire line is formatted.
05:17
< Namegduf>
No.
05:18
<@Vornicus>
bold bold underline nothing
05:18
<@ToxicFrog>
No. And in general it's better to match up your formatting codes
05:18
<@ToxicFrog>
Eg, use ^B bold text ^B rather than ^B bold text ^O
05:18
< Namegduf>
If you're styling a whole line it doesn't matter
05:18
<@ToxicFrog>
test
05:19
< Namegduf>
If you're styling a sub-bitof a line that is more maintainable if you add later changes that're suppoed to contain the inner one.
05:19
< celticminstrel>
If I want bold followed by bold underline, do I need to clear the formatting before the underlined part?
05:19
<@Vornicus>
IRC clients will automatically clear formatting after each line.
05:19
< Namegduf>
No.
05:19
<@ToxicFrog>
celticminstrel: no. They nest.
05:19
<@Vornicus>
%Bbold %Ubold underline%O nothing
05:19
< Namegduf>
You can turn colour codes off by including ^C with no numbers after it.
05:19
<@Vornicus>
Is what I wrote up there.
05:19
< Namegduf>
For a revert to default.
05:20
<@ToxicFrog>
Quick rundown: ^B - toggle bold. ^Cff,bb - set foreground and background colours, 0-15. ^_ - toggle underline. ^V - toggle inverse video. ^O - clear all formatting.
05:20
< celticminstrel>
Oh, so \x03 without the following numbers reverts to default colours without cancelling any other formatting?
05:20
<@ToxicFrog>
foobar
05:20
< Namegduf>
Yes.
05:20
<@ToxicFrog>
Yes.
05:20
< celticminstrel>
Y'know, now that I see them as ^-codes, the choice of characters makes a lot of sense.
05:21
<@Derakon>
Incidentally, you'll irritate most people if your bot uses anything except bold.
05:21
<@ToxicFrog>
^I is used by a tiny minority of (broken) clients for italics, everything else uses tab.
05:21
<@ToxicFrog>
Er, everything else understand ^I as tab or as generic whitespace.
05:21
< celticminstrel>
Ah.
05:22
< celticminstrel>
Mine interprets ^V as italics for some reason.
05:22
< celticminstrel>
Derakon: I generally prefer to use underline rather than bold.
05:22
<@ToxicFrog>
And yeah, in rough order of annoyance - bold, underline, readable colours, inverse video, rainbows, unreadable colours, combinations of the above
05:22
<@Derakon>
In general, the less subtle your emphasis, the less people will like you.
05:23
<@Derakon>
Bold is about as subtle as IRC gets.
05:23
<@ToxicFrog>
Note that "readable colours" varies depending on client colour settings, so your best bet is to use bold/underline/inverse.
05:23
< celticminstrel>
I prefer underline because it's considered roughly equivalent to italics, and italics is not available (unless I know the person receiving uses the same client as me).
05:23
<@ToxicFrog>
And colours only for local output, if at all.
05:23
<@Vornicus>
But I /like/ fake italics
05:23
< celticminstrel>
I do that too sometimes, but only where no formatting is available.
05:23
<@ToxicFrog>
Since when is underline considered at all eqv to italics?
05:23
<@Derakon>
You can always use *fake* _font_ /setting/.
05:24
<@Derakon>
Underline, IMO, is for titles. Bold is for emphasis.
05:24
< celticminstrel>
ToxicFrog: Underline is the handwritten italics.
05:24
<@ToxicFrog>
No, italics is the handwritten italics.
05:24 * ToxicFrog slep
05:24
< celticminstrel>
No, because you can't really do an effect italics in handwriting.
05:25
<@Derakon>
With decent control of your handwriting you can.
05:25
< celticminstrel>
Underline is easier.
05:25
< Namegduf>
They're right
05:25
<@Derakon>
Anyway, IRC isn't handwriting.
05:25 * Namegduf was always taught that instead of italicising titles in handwritten stuff, he should underline them
05:26
< celticminstrel>
Wikipedia: "underlining was originally used in hand-written or typewritten documents to emphasise text."
05:26
< celticminstrel>
I learned the same as Namegduf, I guess.
05:26
< celticminstrel>
It's true that IRC is not handwriting, but I just don't really like bold. It's too subtle for me.
05:29
<@Derakon>
If you need to be less subtle, then use ALL CAPS.
05:30
< celticminstrel>
I hope _ is not a reserved name in Python. <_<
05:30
<@Vornicus>
It sort of is
05:30
< celticminstrel>
...only sort of?
05:30
<@Vornicus>
in the interactive interpreter, _ is used as "the result of the previous command"
05:31
< celticminstrel>
Ah.
05:31
< celticminstrel>
So, as long as it's not a global scope, conflict will not occur.
05:31
<@Vornicus>
>>> 2*2 # 4
05:31
<@Vornicus>
>>> _*_ # 16
05:32
<@Vornicus>
Why the hell are you calling it that /anyway/
05:32
< celticminstrel>
Because it's the underline character, which is ^_
05:32
< celticminstrel>
It has a descriptive name too, though.
05:33
< celticminstrel>
I'm sort of following the example of the regex module, which gives single-letter names to its attributes (eg re.I)
05:55 Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds]
06:08
< celticminstrel>
...wait. I have a regex that matches all formatting characters. Why don't I just do pat.sub(...) instead of looping through to find them?
06:10
<@Vornicus>
dunno.
06:12
< celticminstrel>
Probably because I wrote the function before the pattern. Still, it'd be neater or something.
06:13 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: *hums* Can't stay now!]
06:24 cpux is now known as shade_of_cpux
06:36 Reiv[Graduate] [orthianz@Nightstar-9641d5e4.xnet.co.nz] has joined #code
08:10 AnnoDomini [annodomini@Nightstar-2075dedf.adsl.tpnet.pl] has joined #code
08:10 mode/#code [+o AnnoDomini] by Reiver
08:16 Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has quit [Ping timeout: 121 seconds]
08:16 Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has joined #code
08:37 cpux [Moo@Nightstar-20a84089.dyn.optonline.net] has joined #code
08:40 shade_of_cpux [Moo@Nightstar-20a84089.dyn.optonline.net] has quit [Ping timeout: 121 seconds]
08:41 cpux is now known as shade_of_cpux
08:49 You're now known as TheWatcher
08:52 Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has quit [Ping timeout: 121 seconds]
08:57 Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has joined #code
10:13 Reiv[Graduate] [orthianz@Nightstar-9641d5e4.xnet.co.nz] has quit [Connection reset by peer]
10:19 Reiv[Graduate] [orthianz@Nightstar-9641d5e4.xnet.co.nz] has joined #code
10:24 Reiv[Graduate] [orthianz@Nightstar-9641d5e4.xnet.co.nz] has quit [Connection reset by peer]
10:46 Reiv[Graduate] [orthianz@Nightstar-9641d5e4.xnet.co.nz] has joined #code
11:26 Vornicus is now known as Vornicus-Latens
11:59 Orth [orthianz@Nightstar-02971c14.xnet.co.nz] has joined #code
12:01 Reiv[Graduate] [orthianz@Nightstar-9641d5e4.xnet.co.nz] has quit [Ping timeout: 121 seconds]
13:06 gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code
13:31 Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has quit [[NS] Quit: brb]
13:33 Alek [omegaboot@Nightstar-c5f3565b.il.comcast.net] has joined #code
13:36 Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code
13:38 Orth [orthianz@Nightstar-02971c14.xnet.co.nz] has quit [[NS] Quit: routerboot]
13:39 Reiv[Graduate] [orthianz@Nightstar-fd5ecaf2.xnet.co.nz] has joined #code
13:51 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code
13:53 Rhamphoryncus [rhamph@Nightstar-bbc709c4.abhsia.telus.net] has quit [Client exited]
13:58 Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [[NS] Quit: If the world didn't suck, we'd all fall off.]
14:00 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed]
14:00 Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code
15:21 Reiv[Graduate] [orthianz@Nightstar-fd5ecaf2.xnet.co.nz] has quit [Client closed the connection]
15:22 Reiv[Graduate] [orthianz@Nightstar-fd5ecaf2.xnet.co.nz] has joined #code
15:39 Tarinaky [Tarinaky@Nightstar-5c655275.adsl.virginmedia.net] has quit [Operation timed out]
15:54 Tarinaky [Tarinaky@Nightstar-0a8e7649.adsl.virginmedia.net] has joined #code
16:16 Serah [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds]
16:57 Tarinaky [Tarinaky@Nightstar-0a8e7649.adsl.virginmedia.net] has quit [Ping timeout: 121 seconds]
16:57 Serah [Z@26ECB6.A4B64C.298B52.D80DA0] has joined #code
17:11 Tarinaky [Tarinaky@Nightstar-7bce9701.adsl.virginmedia.net] has joined #code
17:25 Attilla [Attilla@Nightstar-418116f1.threembb.co.uk] has joined #code
17:25 mode/#code [+o Attilla] by Reiver
18:03
< celticminstrel>
...so, if I format dates as yyyy-mm-dd, I can compare lexicographically. Convenient.
18:04
< Namegduf>
Yeah.
18:07
<@Vornicus-Latens>
Oh, celmin: I think you should use U, not _. A hell of a lot clearer.
18:08
< celticminstrel>
Well, I currently have both, and I'm using U internally...
18:10 AnnoDomini [annodomini@Nightstar-2075dedf.adsl.tpnet.pl] has quit [Ping timeout: 121 seconds]
18:12 AnnoDomini [annodomini@Nightstar-16d1b861.adsl.tpnet.pl] has joined #code
18:12 mode/#code [+o AnnoDomini] by Reiver
18:13 PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
18:15
<@Vornicus-Latens>
Without a compelling reason, you should only give a thing one name.
18:15
< Namegduf>
Unless it's some kind of demon lord.
18:16
<@jerith>
celticminstrel: This is one of the awesome features of ISO 8601 time formats.
18:16
< Namegduf>
I've used that format for writing dates since I was a child.
18:16
< Namegduf>
YYYY-MM-DD
18:23 PinkFreud [WhyNot@NetworkAdministrator.Nightstar.Net] has joined #code
18:31
< gnolam>
It's the only sane date format.
18:31
< gnolam>
And as usual, the US standard is the most illogical. :P
18:33
< Reiv[Graduate]>
I've always either dealt with YYYY-MM-DD or DD-MM-YY.
18:33
< Reiv[Graduate]>
The latter being more useful when hand-sorting items of close chronological age - eg, lecture notes.
18:34
< Reiv[Graduate]>
Given the first bit you read is the day, and that's all you care about half the time.
19:56 * Namegduf generally just switches to reading right-to-left in that kind of case
19:59
<@ToxicFrog>
Oh god, partial application of operators, yes
20:00
<@ToxicFrog>
scala> List(4.0, 5.0, 6.0).map(_ / 2)
20:00
<@ToxicFrog>
res13: List[Double] = List(2.0, 2.5, 3.0)
20:26
<@Vornicus-Latens>
cute
20:27
<@ToxicFrog>
Yes.
20:27 * ToxicFrog rubs his face all over Scala's partial application like a cat
20:29
<@ToxicFrog>
It's more verbose than Haskell's, but you aren't limited to applying in order
20:29 Vornicus-Latens is now known as Vornicus
20:38
<@ToxicFrog>
Seriously, the more I noodle around with Scala the more impressed I am.
21:11
<@McMartin>
10:31 < gnolam> And as usual, the US standard is the most illogical. :P
21:11
<@McMartin>
I'm not sure which came first, the format or the speech pattern, but the American dialect says dates differently than the others.
21:11
<@McMartin>
We say "November 5th", not "5 November".
21:12
< Namegduf>
I say "5th of November"
21:13
<@McMartin>
"4th of July" is the only standard one I can think of.
21:15
< celticminstrel>
That looks a lot like Boost.Bind, though that only works for functions.
21:16
<@ToxicFrog>
celticminstrel: what does?
21:17
<@Vornicus>
Partials are a very common idiom.
21:18
<@Vornicus>
Some languages - particularly functional ones like Haskell - include syntax support for it; others are more roundabout about it.
21:18
< celticminstrel>
[3:00pm] ToxicFrog: scala> List(4.0, 5.0, 6.0).map(_ / 2)
21:19
< celticminstrel>
I'm talking about the use of the underscore as a placeholder.
21:19
<@ToxicFrog>
Underscore as a placeholder is common; underscore as a placeholder in partial application is less so.
21:19
<@ToxicFrog>
Ooo, I wonder if I can partially apply constructors
21:19
< celticminstrel>
Is "partial" different from "currying"?
21:20
<@ToxicFrog>
scala> (List(1, _:Int, 3))(2)
21:20
<@ToxicFrog>
res33: List[Int] = List(1, 2, 3)
21:20
<@ToxicFrog>
Partial application is the use case for currying.
21:20
< celticminstrel>
Ah.
21:20
< celticminstrel>
The use case? The only use case?
21:20
<@McMartin>
Hm, that looks a lot like a properly integrated version of boost::bind
21:20
< celticminstrel>
...that's what I just said!
21:21
<@McMartin>
So you did. I skipped over it because C++ things are all :: in my hindbrain.
21:21
<@McMartin>
And yeah, the only use case because all the "other" use cases are use cases for partial evaluation
21:21
< celticminstrel>
Right, I was using Boost documentation format rather than Boost code format.
21:24
<@ToxicFrog>
I do wish the type inference system was a bit smarter
21:24
<@ToxicFrog>
List(1,2,3).map(_:(Int) => Int)
21:24
<@ToxicFrog>
Should be writable as 'List(1,2,3).map _' IMO
21:25
<@ToxicFrog>
Oh wait, no it shouldn't, foo _ is for parenlesses.
21:28
<@ToxicFrog>
But yeah, so far my biggest gripe about Scala is that the type inference system doesn't go far enough
21:33
<@Vornicus>
http://blog.wolfire.com/2010/06/Aquaria-goes-open-source <--- for... anybody who's not in #fleet too, I guess
21:48
<@McMartin>
TF: I think there's a pretty sharp limit to how far you can get with inference in a mixed-paradigm language
21:48
<@McMartin>
I suspect ML is the theoretical limit and it's a lot weaker than Haskell's.
21:48
< celticminstrel>
ML?
21:49
< celticminstrel>
Was Aquaria one of the games in that "Humble Indie Gamer Bundle"?
21:49
<@McMartin>
ML is a family of impure functional languages that is one of the major inspirations for Haskell.
21:49
<@McMartin>
Yes, it was
21:49
<@McMartin>
That's why it opened its engine up
21:50
<@McMartin>
The OCaML dialect of ML is, I believe, presently the highest-performance functional language extant, with generated code on mathematical computation coming in between C and C++/STL.
21:51
<@McMartin>
However, OCaml is horribly ugly and while I used to use it for occasional things, I've never gone back since learning Haskell.
21:51
<@McMartin>
Both had the practicality problems.
21:52
<@McMartin>
Anyway, it sounds like Scala needs to be on my List.
21:53
<@McMartin>
Especially since Jambi is also an option.
21:56 Rhamphoryncus [rhamph@Nightstar-bbc709c4.abhsia.telus.net] has joined #code
21:57
<@ToxicFrog>
The Qt Java binding?
21:58
<@ToxicFrog>
I've been looking at org.gnome.* myself, because I like me some Glade
22:01
<@McMartin>
Yeah, Qt
22:02
<@McMartin>
I find Qt UI Designer superior to Glade, but we've been around this before
22:02
<@McMartin>
I basically migrated wx -> GTK -> Swing -> Qt4 and found it a step up each time
22:03
<@McMartin>
Swing and Qt4 are the only ones with what I consider acceptable Mac support, though GTK may have improved some since
22:03 * Vornicus couldn't get swing to work at all.
22:03
<@Vornicus>
It did literally nothing I expected it to.
22:03
<@McMartin>
Swing's programming model is worse than GTK+Glade - I'm talking about end results here
22:04
<@ToxicFrog>
End-results-wise, all of my experiences with Qt have been through KDE apps, and have left me rather non-enamored of it.
22:05
<@ToxicFrog>
Swing produces better, or at least more consistent, results than GTK, but is such an incredible pain in the ass to develop for that these days I dismiss it automatically.
22:05
<@ToxicFrog>
I may check out Qt Designer next time I'm working on a graphical app, but at present I already have the Glade files, so
22:05
<@Vornicus>
(to be specific, my first and only attempt in Swing was a simple calculator! But I couldn't get the window to hold the buttons.
22:06
<@McMartin>
TF: Yeah, a lot of it is taste in architecture
22:06
<@McMartin>
If your experience with Qt has been via KDE integration, that was Qt3, which was inferior to GTK by every conceivable metric
22:06
< celticminstrel>
I managed to make a Swing calculator.
22:06
< celticminstrel>
Moreover, it was a graphing calculator.
22:06
<@ToxicFrog>
Don't recent (KDE4) apps use Qt4?
22:06
<@McMartin>
Qt4 is the only cross-platform native GUI library that's using native components on Windows, Linux, and Mac.
22:06
<@McMartin>
And is actually single-source
22:07
< celticminstrel>
It was one of the assignments for my course in OO programming.
22:07
<@ToxicFrog>
What does "native components" mean on Linux?
22:07
<@McMartin>
TF: Possibly, but they may be using the Qt3 parts or doing it in ugly ways.
22:07
<@Vornicus>
I probably could have pulled that off too, celmin -- but my difficulty was the fact that I couldn't freaking get the buttons to sit where I thought I told them to.
22:07
<@McMartin>
Direct calls to X instead of something like a frame buffer.
22:07
<@McMartin>
"The application will look like all your other applications."
22:07
< celticminstrel>
I think I used some kind of layout manager to hold the buttons, and nested managers as necessary to get the layout I wanted.
22:08
<@McMartin>
"And it will do so because it's asking the Windowing system to draw stuff for you, unlike Swing, which draws what it guesses your other apps look like"
22:08
<@Vornicus>
Oh. By the way, celmin, if you haven't seen it already: do "import this" in the python interactive terp
22:08
<@ToxicFrog>
Ok, but all of my other applications can't agree, because most of them are using GTK, some are using Qt3, and one or two are using Motif~
22:08
<@McMartin>
My experience on Linux is that GTK and Qt4 look indistinguishable under GNOME.
22:08
< celticminstrel>
O_O
22:08
<@ToxicFrog>
Ok, in that case the KDE apps I'm bitching about are definitely not Qt4
22:08
<@McMartin>
Unless stylesheets are involved, in which case all bets are off
22:09
<@McMartin>
(Qt4 has this CSS-like thing that you can use to tweak what basically everything looks like)
22:09
<@McMartin>
Once I get FotH in shape for beta I'm going to try to port Blorple to C++/Qt4
22:09
<@McMartin>
Since I'm unhappy with the way Swing is handling it on Windows.
22:10
<@McMartin>
(My job involves working with Qt4 too, which helps for keeping skills sharp - it's worth noting that we make heavy use of stylesheets though to make the application alien everywhere, but the same alien. The goal is a uniform experience regardless of OS, there.)
22:11
<@McMartin>
(Blorple would have the opposite goal.)
22:11
<@McMartin>
TBH though my primary interest is looking like Aqua on Mac and looking like Forms on Windows; Linux support is a bonus AFAIAC.
22:14 * ToxicFrog nods
22:14
<@ToxicFrog>
My current plan is to do Spellcast in Scala + GTK/Glade
22:15
<@ToxicFrog>
After that, I might put together an alternate Qt interface to test my MVC-fu, or work on something else entirely
22:29 Derakon [Derakon@Nightstar-5213d778.ca.comcast.net] has quit [Ping timeout: 121 seconds]
22:33 You're now known as TheWatcher[T-2]
22:36 * Vornicus finds himself once again considering his RPS ruleset generator.
22:41 You're now known as TheWatcher[zZzZ]
22:41
<@Vornicus>
One of the problems I previously had was that generating the canonical version of any possible RPS took for/ever/, and even worse than that, each of the canonical versions was generated a number of times equal to the number of permutations available.
22:42
<@Vornicus>
And that each of those times I had to generate all of those permutations.
22:42
<@Vornicus>
So it was something like O(n!^2
22:42
<@Vornicus>
Which is in fact entirely batshit.
22:43 ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Connection reset by peer]
23:45 Reiv[Graduate] [orthianz@Nightstar-fd5ecaf2.xnet.co.nz] has quit [Ping timeout: 121 seconds]
23:49 ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code
23:49 mode/#code [+o ToxicFrog] by Reiver
23:51 SmithKurosaki [Smith@Nightstar-0a57481d.dsl.teksavvy.com] has joined #code
23:56
<@Vornicus>
What if I found a way to find the canonical version in fewer steps? It may be possible to, at each step in the permuter, only use items which are beaten by the several previous items.
23:57
<@Vornicus>
Which can drastically reduce the number of permutations to try. Dunno how much improvement that gives though!
--- Log closed Fri Jun 04 00:00:23 2010
code logs -> 2010 -> Thu, 03 Jun 2010< code.20100602.log - code.20100604.log >
 search of the:   for: