00:11 (quit) dnolen: Ping timeout: 255 seconds
00:21 (join) veer
00:32 (join) yoklov
00:39 (quit) yoklov: Quit: Leaving.
00:57 (quit) veer: Quit: Leaving
01:35 (quit) jonrafkind: Ping timeout: 240 seconds
01:37 (join) jonrafkind
03:10 (quit) jonrafkind: Read error: Operation timed out
03:43 (join) Shvillr_
03:44 (quit) Shviller: Disconnected by services
03:44 (nick) Shvillr_ -> Shviller
04:13 (join) sindoc
05:45 (quit) Shvillr: Ping timeout: 248 seconds
05:45 (join) Shvillr
06:00 (quit) AlbireoX: Read error: Operation timed out
06:20 (join) veer
06:22 (quit) veer: Client Quit
06:50 (quit) realitygrill: Quit: realitygrill
07:06 (join) masm
07:48 (quit) masm: *.net *.split
07:55 (join) masm
07:57 (join) MayDaniel
08:08 (join) Lajla
08:10 (quit) MayDaniel: Read error: Connection reset by peer
08:53 (join) metadave
08:53 (part) metadave
09:12 aidy: Hi, we have to make a game in r5rs for class, and I was wondering if making it as purely functional as possible would make it noticably slower.
09:13 aidy: E.g. I never user set-car! or set-cdr! so maybe I could prevent mcons being used instead of cons?
09:46 (nick) chaozzbubi -> ChaozZBubi
10:12 (quit) snorble: Read error: Connection reset by peer
10:14 (join) jeapostrophe
10:27 bremner: eli: I'm trying to run the handin server on a headless box with (debian packages of) racket 5.2. It complains 'Gtk initialization failed for display ":0"'. Does it make any sense that the handin server should be trying to do something graphical?
10:47 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/P-zGTA
10:47 RacketCommitBot: [racket/master] Adjusted rewrite-side-condition/check-errs so that it normalizes the internal - Robby Findler
10:48 (join) freakazoid
11:01 (join) yoklov
11:02 (join) anRch
11:08 (join) karswell
11:12 (quit) karswell: Read error: Operation timed out
11:18 (join) karswell
11:37 (nick) Lajla -> novato_nl
11:54 (join) mceier
11:56 (join) jrslepak
12:00 (quit) jeapostrophe: Ping timeout: 248 seconds
12:13 (quit) anRch: Quit: anRch
12:13 (quit) karswell: Ping timeout: 240 seconds
12:19 (quit) freakazoid: Quit: Computer has gone to sleep.
12:20 (join) dnolen
12:28 (join) anRch
12:34 (quit) anRch: Ping timeout: 240 seconds
12:36 (join) anRch
12:46 (quit) yoklov: Quit: Leaving.
12:50 (quit) anRch: Quit: anRch
12:52 (join) jonrafkind
12:54 (join) freakazoid
13:13 eli: bremner: Usually it does -- since most uses of it is with ugrad courses where they're using graphics.  Because of that I've always running in an X session (via vnc), and so did everyone else who uses it.
13:13 bremner: bummer
13:13 eli: But it should be possible to run it without X now, so something might still be wrong.
13:14 eli: s/might/is probably/
13:14 eli: Are you just running it with the racket executable?
13:14 bremner: yes
13:14 bremner: racket -l- handin-server
13:15 bremner: PLTCOLLECTS needs to be set
13:16 bremner: it might be relevant that I installed the pl .plt into that users ~/.racket
13:16 eli: I forgot -- yeah, it doesn't matter if you start it with `racket', since it can start the gui code too.
13:17 eli: bremner: OK, I see the problem.
13:17 eli: Look at the top of "handin-server/main.rkt" --
13:18 eli: it requires `racket/gui/base', and if you comment that out it will work for you.
13:18 eli: And there's a comment next to it explaining why it is doing that.
13:18 bremner: oops. I should have caught that.
13:18 eli: If you never use gui stuff in your student's code, then removing it should be fine.
13:19 eli: bremner: You should also remove the require from other files that use it.
13:20 bremner: ok, that gets me started. DISPLAY='' racket -l- handin-server   seems copacetic
13:20 (join) karswell
13:20 bremner: thanks!
13:20 eli: bremner: I'm not done...
13:20 bremner: ok?
13:21 eli: bremner: The two other files that use it are "checker.rkt" and "utils.rkt".
13:22 eli: bremner: they both need it, to translate the submission files (which are transferred as "binary" WXME media files) to text.
13:23 eli: Actually only "checker.rkt" does that, "utils.rkt" does something a little different.
13:23 eli: So if you remove it from them they probably won't work.
13:24 eli: So the conclusion is that you should probably just use it with X, either via vnc or via an ssh forward.
13:24 bremner: ok, a bit sad, but what can you do.
13:24 eli: That's my guess -- that the real problem is that something in `racket/gui' is using gtk in a too-eager way.
13:25 eli: bremner: I'll submit a bug report and CC you?
13:25 bremner: thanks
13:33 eli: bremner: Done, with your email, so you'll get the replies.
13:33 eli: (It's a bad way to get this, we should really dump gnats...)
13:35 (join) realitygrill
13:38 bremner: ok, I got the reply to the mail you forged in my name ;)
13:39 eli: bremner: Yeah, switching to github will improve things a little.
13:39 eli: The main problem is keeping it possible for anyone to submit bugs, rather than forcing them to use github.
13:40 eli: But we didn't find anything better.
13:41 bremner: well, I've been meaning to investigate http://xpra.org
13:41 bremner: not for bug tracking, for detaching x sessions
13:42 eli: Yeah, just saw it.
13:42 eli: (I was hoping for the latter, since we really searched a lot for a solution.)
13:42 eli: Anyway, why not vnc?
13:43 bremner: it does a whole screen, rather than a rootless window, iirc
13:43 bremner: I don't actually know much about vnc.
13:44 eli: Yes, it does a whole session, but that shouldn't be a problem.
13:44 eli: You can use an old-fashined xclients file to specify what comes up, which you could do with a single terminal winodw.
13:44 bremner: sure, I'm sure it will work.
13:45 bremner: if xpra turns out to be not-ready-for-prime-time, I'll figure out vnc
13:49 (join) yoklov
13:53 eli: bremner: I can only recommend vnc -- which I've used for a long time now.  It's been my main mode of work for probably more than 6 years.
13:54 bremner: ah, well, I have X11 on every desktop. So it seems slightly silly to use another protocol. But that doesn't mean it isn't a good solution in practice.
13:54 eli: I started with that mindset, then I discovered how convenient it is to use the same running desktop everywhere.
13:55 eli: It's also a reason why I stopped caring about getting laptops that run linux fine -- we get windows machines, and use mostly use them as vnc terminals...
13:55 eli: ("we" = me and my wife.)
14:02 bremner: welcome to the cloud ;)
14:02 (join) AlbireoX
14:44 bremner: handin-server seems to be running ok in xpra (detached)
14:44 (join) jao
14:47 bremner: eli: any reason not to use allow-new-users? It seems like the simplest thing to do. I guess the students will screw up their id numbers and so on.
14:49 eli: bremner: re the cloud -- not exactly yet, since you get a real proper desktop.  With the current meaning of "the cloud" we could kind of get the same productivity if we conformed to crappy tools like the browser's editor, and things like google docs.  They're fine tools for what they are, but they're still very far from beating a desktop.
14:50 (quit) jonrafkind: Ping timeout: 252 seconds
14:50 eli: And re the new users -- I set it on at the beginning of the semester, and then I turn it off once everybody's on.
14:51 (quit) realitygrill: Quit: realitygrill
14:51 eli: I was also worried at some point about random people getting on (and, for example, get to see the solutions) but that's not a problem since I made up the hook thing that sends me emails when people are added etc.
14:51 bremner: oh right. I should double check that address. I know it doesn't go to you, but it might go to example@example.com ;)
14:54 eli: bremner: :) That's what I thought happened when I asked you about it...
15:35 (quit) Fulax: Read error: Connection reset by peer
15:36 (quit) dnolen: Quit: Page closed
16:13 (join) MayDaniel
16:20 (join) jeapostrophe
16:22 (quit) sindoc: Quit: Leaving.
16:31 (quit) karswell: Ping timeout: 240 seconds
16:33 (quit) jeapostrophe: Ping timeout: 240 seconds
16:43 (quit) yoklov: Quit: Leaving.
16:51 (join) karswell
17:04 (quit) MayDaniel: Read error: Connection reset by peer
17:10 (join) anRch
17:11 (join) jonrafkind
17:38 (join) snorble
17:42 (join) Fulax
17:48 (join) jyc
17:54 (quit) mceier: Quit: leaving
18:13 (quit) anRch: Quit: anRch
18:55 (quit) karswell: Ping timeout: 268 seconds
19:02 (join) karswell
19:07 (join) neilv
19:08 (join) Antonio_
19:10 (quit) Antonio_: Client Quit
19:24 (quit) jyc: Ping timeout: 252 seconds
19:27 (join) jyc
19:46 (quit) masm: Quit: Leaving.
19:47 bremner: I have a beginner typed racket question.  http://paste.debian.net/150436/  does not typecheck, I guess because the typechecker doesn't understand the if. Is there some annotation I should make here?
19:53 (quit) karswell: Ping timeout: 240 seconds
19:54 jonrafkind: Natural is nonnegative right, dont you mean Integer?
19:55 bremner: well, if x > 1, both subtractions are non-negative
19:56 jonrafkind: ah, well I guess typed racket can't determine that
19:56 jonrafkind: it just knows that x is still natural
19:56 bremner: yeah, that's fair enough
19:56 bremner: I guess maybe a seperate Natural subtraction is needed
19:57 bremner runs away for a bit.
20:04 (join) karswell
20:22 (quit) karswell: Ping timeout: 240 seconds
20:29 (quit) jrslepak: Quit: Leaving
20:32 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/TaQ8tw
20:32 RacketCommitBot: [racket/master] fix doc typo - Matthew Flatt
20:32 RacketCommitBot: [racket/master] fix locale sensitivity of inexact-number parsing - Matthew Flatt
21:01 (quit) jyc: Ping timeout: 252 seconds
21:04 neilv: i hope the planet server doesn't mind that my tool makes several rapid package download requests and drops the connection before reading the http content
21:10 RacketCommitBot: [racket] plt pushed 11 new commits to master: http://git.io/sJwavA
21:10 RacketCommitBot: [racket/master] Avoid bash regexps - Eli Barzilay
21:10 RacketCommitBot: [racket/master] Clarify a comment for future work; a bunch of new internal definition uses. - Eli Barzilay
21:10 RacketCommitBot: [racket/master] Switch `delay's to `lazy's. - Eli Barzilay
21:38 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/kFBI7w
21:38 RacketCommitBot: [racket/master] fix cross-module inlining for multi-arity functions - Matthew Flatt
21:45 (join) samth
21:45 (join) yoklov
21:55 (quit) freakazoid: Quit: Computer has gone to sleep.
22:15 (join) jeapostrophe
22:40 (join) freakazoid
22:45 (nick) ChaozZBubi -> chaozzbubi
22:46 (quit) samth: Ping timeout: 252 seconds
23:03 (quit) neilv: Quit: Leaving
23:04 (quit) freakazoid: Quit: Computer has gone to sleep.
23:46 (quit) dsantiago: Quit: Leaving...
23:55 (join) dsantiago