01:38 (quit) jonrafkind: Read error: Operation timed out 03:14 (join) noel_ 03:20 (quit) noel_: Quit: Page closed 04:47 (join) masm 04:57 (join) mije 04:57 mije: oy ! 04:58 eli: yo. 05:02 mije: eli: you are a guy behind racket ? 05:02 eli: mije: Yes, one of them. 05:03 mije: i just wanted to say that what you do with racket kicks ass 05:03 eli: :) 05:05 mije: i'm using it as the language for the katas i organize in my company 05:06 mije: after a few of them, people using java as their daily language begin to say their job would be much funnier with racket :) 05:06 eli: "katas" as in http://en.wikipedia.org/wiki/Kata ? 05:06 mije: yeah sort of :) 05:07 eli: So that's really kicking ass... 05:07 eli: What do you do with it? 05:07 mije: more like this kind of katas http://codekata.pragprog.com/2007/01/code_kata_backg.html#more 05:08 mije: we have sessions for a few hours, we take a reocurring problem and try to find more clever way to solve it, without the pressure of the productive work 05:09 eli: That sounds like fun. (And more geeky than the karate ones...) 05:09 mije: i'd say that people i work with find very interesting to introduce functionalities their language doesn't have in order to solve their problems 05:10 mije: like "ok, i have hard time doing it in an OO way in java, let's try in functional" 05:11 eli: That's certainly helpful, and will probably be easier to adapt to new functionality in Java. (Eg, when it gets to have first class functions, like C#'s delegates.) 05:11 mije: or with contracts, or using traits and mixins, or creating ad-hoc macros 05:13 mije: what amazes me the most is that some people who were just "i code because it's my job" begin to be personnaly interested in programming, as they where when they were in college 05:13 mije: all the possibilities of the languages makes programming more challenging 05:13 mije: and more interesting 05:14 eli: Yes, if you have people switch to being interested in programming, then I'm sure you really enjoy it. 05:15 eli: I once did something similar in a company -- we had scheme session and other kinds of general code quality things, and I really liked doing that. 05:15 mije: i guess that sooner or later the guys i work with will want to program with racket on a daily basis 05:16 eli: :) 05:16 eli: That would be nice too, of course. 05:16 eli: (But just learning it can be a lot of fun in itself...) 05:16 mije: yeah, learning is my work :) 05:17 mije: are you confident that racket can be used to create real world heavy duty software ? 05:23 eli: mije: For most purposes -- yes. But if you have something specific in mind, then you should ask on the mailing list. 05:59 (quit) sizur: Remote host closed the connection 07:34 mije: does rackunit provides a way to test contracts ? 07:35 bremner: what is the difference between "unix source" and "windows source"; do they both contain subsets of the git repo? 07:50 eli: bremner: They have mostly the same contents, but each has the relevant gui sources for its own platform only. 07:51 eli: mije: If it's a simple predicate than you can test it as usual. 07:51 eli: Otherwise you can just write some sample functions with contracts and test the error messages. 07:57 mije: eli: ok thanks 07:58 bremner: eli: thanks. Is is documented somehow what revision the tarballs are based on? 08:00 eli: bremner: What do you mean? The 5.0 downloads are for v5.0... 08:00 eli: There's a "v5.0" tag in the git repo, if that's what you're looking for. 08:01 bremner: eli: thanks, I missed the tag somehow, that is exactly what I was looking for 08:01 bremner: I always forget git fetch --tags 08:03 eli: bremner: But why do you need the tag if you're using the repo? 08:04 bremner: eli: I'm trying to have an upstream branch that matches the release tarballs (for debian packaging). 08:06 eli: Ah -- you're taking over the debian package? 08:06 bremner: eli: I'm just getting my feet wet, but maybe 08:06 eli: Did you get in touch with Ari about it? 08:06 bremner: yes 08:07 eli: He's probably happy to get it off his hands... 08:07 eli: Does that cover the Ubuntu package too, or are they separate 08:07 eli: ? 08:07 bremner: so I understand. I'm just figuring out if I can do a good job maintaining it. That partly depends how much I end up using it for $work 08:07 bremner: Ubuntu usually (75% of the time) just copies the debian package 08:08 eli: OK, that would be nice to have two birds in one stone. 08:08 eli: In any case, if you have any questions, feel free to ask me. 08:08 eli: There were a number of problems with the old setup. 08:08 eli: And in general it should be much easier now. 08:08 bremner: eli: great, thanks for the offer 08:09 eli: I'm just trying to get to a state where everything works fine and there's no need to deal with per-distro patches. 08:09 bremner: the only patch in debian currently is for sparc, but I think that is obsolete now. 08:09 eli: ...except for start menu entries and such -- I was told that there's no way for us to do it, and that it's best to let the distros do their thing. 08:20 (join) jao 08:45 (quit) askhader: Ping timeout: 260 seconds 10:23 (quit) sstrickl: Quit: sstrickl 10:31 (join) askhader 10:41 (join) stamourv 11:19 (join) sstrickl 11:23 chandler: eli: My understanding is that there is in fact a standard for application menu icons these days which allows them to be created either system-wide or in a user's home directory. 11:24 chandler: eli: I believe that the standard is promulgated by the freedesktop.org group. 11:26 chandler: eli: Ah, see http://www.freedesktop.org/wiki/Specifications/menu-spec 11:36 eli: chandler: I'm just repeating what the Fedora guy told me about a year ago -- so I'm assuming that relying on these standards is still premature... 11:39 eli: chandler: Ah, it's xdg stuff, and IIRC, things like `xdg-open' were not installed by default on my F12 machine. 11:39 chandler: Ah, OK. 11:40 chandler: I have seen a few non-package system application installers create menu icons (VMware Player, for instance), but I don't know how they're doing that. 11:42 eli: It probably wouldn't be hard to figure out what system you're using and do the right thing -- and I don't remember the details, but this was my idea, and this was what he explicitly said would not be a good idea. 11:50 (join) jonrafkind 12:03 (quit) mije: Ping timeout: 252 seconds 12:36 (quit) jao: Ping timeout: 265 seconds 13:22 (join) jao 15:12 (join) carleastlund 15:53 (join) mije 16:35 (quit) mije: Ping timeout: 252 seconds 16:36 (quit) sstrickl: Quit: sstrickl 17:39 (join) sstrickl 18:00 (join) ben_m 18:01 (quit) masm: Quit: Leaving. 19:09 (part) stamourv: "ERC Version 5.2 (IRC client for Emacs)" 19:12 (quit) ben_m: Quit: sooosdfjsf 19:36 (quit) jao: Ping timeout: 265 seconds 20:07 jonrafkind: im getting this error compile: access from an uncertified context to unexported variable from module: "/home/jon/bin/plt/collects/racket/private/modbeg.rkt" in: print-values 20:08 jonrafkind: how can I know when im going to use syntax-case or whatever on an uncertified syntax object? 20:27 (part) carleastlund 20:48 derrida: eli: i think it's looking pretty good as far as distro agnostic goes. i've now had no problems compiling on 2 linuxes and osx. 20:59 askhader: Has anyone here built drracket on Ubuntu 10.04 x86_64? 21:02 jonrafkind: i dunno, but supposedly the ubuntu jaunty stuff works fine on 10.04 x86_64 21:02 askhader: Did you read this somewhere? 21:03 jonrafkind: someone came into this channel and claimed it to be so 21:03 askhader: Ah I see 21:03 jonrafkind: You folks might want to update the website - worked just fine with Ubuntu 10.04 x64 21:03 jonrafkind: i asked eli to add that explicit message to the downloads page, but we eventually compromisde with the current text "a build might work on your machine" 21:05 askhader: Interesting- I will try this 21:12 (quit) jonrafkind: Ping timeout: 258 seconds 21:18 derrida: askhader: were you asking because you had a problem? 21:19 askhader: No, I ask because I was considering using the debian i686 build 21:19 eli: askhader: You should see that message when you choose any linux download. In general, there are too many of them, and they change too frequently, to maintain any kind of compatibility table. 21:20 eli: So I left it with a generic message saying that with linux you should always try other installers if there's none for your particular configuration. 21:20 askhader: Fair enough. 21:20 askhader: I assume you maintain the plt racket web presence? 21:20 eli: derrida: What was your "distro agnostic goes" referring to? 21:21 derrida: eli: the source distribution 21:21 derrida: i think it's best to support *no* distros, but that's just me :D 21:21 eli: You mean distribute sources only for linux? 21:22 derrida: i appreciated the simplicity of the build process 21:22 eli: Well, as simple as it is (which I wholeheartedly agree with), some people will always want a binary and nothing else. 21:23 derrida: i'm sure true 21:25 chandler: The alternative is probably that people will install an old build from their distribution's repository, then ask a lot of questions or file a lot of bug reports pertaining to the old version. 21:25 derrida: distribution is a tricky problem 21:28 derrida: eli: i tried out geiser mode for emacs btw. seems to work really well. 21:44 eli: derrida: I meant to try it for a while now... 21:52 askhader: Do any of you folks write scheme applications that read/write from a MYsql database? 21:54 askhader: I guess my real question would be, what do you use to interface with a mysql db?