00:01 (quit) superjudge: Quit: superjudge
00:08 (quit) janne: Ping timeout: 252 seconds
00:08 (join) janne
00:14 (quit) tewk: *.net *.split
00:14 (quit) stride: *.net *.split
00:14 (quit) fds: *.net *.split
00:38 (quit) bremner_: Ping timeout: 246 seconds
00:39 (quit) realitygrill: Quit: realitygrill
00:42 (join) PLT_Notify
00:42 PLT_Notify: racket: master Asumu Takikawa * 3e3e736 (1 files in 1 dirs): Fix bug in toolbar class in guibuilder
00:42 PLT_Notify: racket: master Asumu Takikawa * 9922062 (1 files in 1 dirs): Make guibuilder preview work and Racketize
00:42 PLT_Notify: racket: master Asumu Takikawa * 7868909 (1 files in 1 dirs): Fix docs on editor-snip% init arg
00:42 PLT_Notify: racket: master Asumu Takikawa * 7f143f0 (3 files in 3 dirs): Add impersonator-contract? along with docs/tests. ...
00:42 PLT_Notify: racket: master Asumu Takikawa * f4aab53 (3 files in 3 dirs): Added option/c for possibly #f values in unstable.
00:42 PLT_Notify: racket: master commits b548bc5...f4aab53 - http://bit.ly/iypNXo
00:42 (part) PLT_Notify
00:43 (join) bremner_
00:44 (quit) mario-goulart: Read error: Connection reset by peer
00:44 (join) mario-go`
00:45 (join) stride
00:45 (join) fds
00:45 (join) tewk
00:49 (join) ideasman_42
00:49 ideasman_42: Hi, anyone here?
00:49 ideasman_42: just listened to the podcast :)
00:52 jonrafkind: yea
00:53 ideasman_42: jonrafkind, can you embed racket in a C applicaton?
00:53 jonrafkind: yea
00:53 ideasman_42: does it have an embeddable C api like - tinyscheme, cpython, lua
00:53 jonrafkind: yea
00:54 jonrafkind: http://docs.racket-lang.org/inside/overview.html?q=embed#(part._3m_.Embedding)
00:54 ideasman_42: cheers
00:54 ideasman_42: (I work on embedded python in blender3d)
00:54 RandalSchwartz: ideasman_42 - I knew the podcast would bring more people here. :)
00:55 ideasman_42: hey, RandalSchwartz, I was on your sho a while ago!
00:55 RandalSchwartz: oh hey!
00:55 ideasman_42: the blender one ofcourse
00:55 RandalSchwartz: yeah - I suspected as much
00:55 ideasman_42: recall it was 3am and using my laptop mic
00:55 jonrafkind: embedded python is nice. embedded ruby is a nightmare
00:55 jonrafkind: to be honest, ive played with the C api for racket and couldnt figure it out
00:55 jonrafkind: other people have gotten it to work though
00:56 ideasman_42: RandalSchwartz, interestingly people use blender in a similar way to racket
00:56 RandalSchwartz: as a teaching platform?
00:56 jonrafkind: also you have to play nicely with the racket garbage collector which uses precise garbage collection. in particular you have to explicitly mark which stack objects can contain pointers
00:56 ideasman_42: more as a way of learning python/3d
00:57 ideasman_42: we have a 3D environment and interactive command line which is very easy to get into
00:57 RandalSchwartz: ahh
00:57 RandalSchwartz: yeah
00:57 ideasman_42: you can for example display python console vars in the 3d view
00:57 (join) neilv
00:57 ideasman_42: http://wiki.blender.org/index.php/File:Add-ons_MathVis.png
00:57 ideasman_42: ack
00:57 ideasman_42: http://wiki.blender.org/uploads/c/c2/Add-ons_MathVis.png
01:00 (quit) jeapostrophe: Quit: jeapostrophe
01:01 ideasman_42: since we have most of our access through a generic type system now Im keeping an eye out on other embeddable languages
01:04 neilv: i embedded racket many years ago. it was definitely doable.  if i were to do it nowadays, i might instead put it in its own process, and embed only a little communication adapter that talks through a socket
01:04 jonrafkind: neilv, you should attempt to fix the embedded C examples we have
01:04 ideasman_42: yeah, think thats wise in general
01:04 neilv: if it's good enough for postgresql... :)
01:04 jonrafkind: i tried a year ago and failed
01:05 ideasman_42: problem with how we embed python its it locks the whole application while python runs
01:06 ideasman_42: kindof horrible, but in practice its not tooo bad. just means writing an interactive debugger is not really an option
01:06 jonrafkind: collects/mzscheme/examples
01:08 (join) dnolen
01:12 (join) jeapostrophe
01:13 (quit) mithos28: Quit: mithos28
01:15 neilv: with racket's more mainstream exposure on merlyn's podcast, i'm feeling like a hipster
01:15 neilv: i knew racket when they were underground, mannn
01:25 RandalSchwartz: heh
01:38 (quit) dnolen: Quit: dnolen
01:46 (quit) jeapostrophe: Quit: jeapostrophe
01:53 (quit) jonrafkind: Ping timeout: 252 seconds
02:26 (join) vu3rdd
02:26 (quit) vu3rdd: Changing host
02:26 (join) vu3rdd
02:30 (join) samth
02:33 (part) samth
02:33 (join) samth
02:54 ohwow: Hello, anybody up?
03:08 neilv: everyone is listening to drscheme v103 on vinyl
03:24 ohwow: http://www.youtube.com/watch?v=CAYQNb8v-9w
03:24 ohwow: that's why i love twit
03:30 (join) hkBst
03:37 (join) leo2007
03:44 (quit) alcuadrado: Ping timeout: 246 seconds
03:44 (join) MayDaniel
03:46 (quit) sstrickl: Ping timeout: 246 seconds
03:46 (quit) MayDaniel: Read error: Connection reset by peer
03:49 (join) sstrickl
04:05 (join) guestNaN
04:24 (quit) samth: Ping timeout: 246 seconds
04:29 (join) misterncw
04:48 (quit) neilv: Quit: Leaving
04:59 (quit) leo2007: Ping timeout: 260 seconds
05:03 (join) masm
05:11 (join) leo2007
05:31 (join) RpR`
06:37 (join) lucian
06:43 (join) lucian_
06:46 (quit) lucian: Ping timeout: 260 seconds
06:57 (quit) vu3rdd: Remote host closed the connection
07:04 (nick) mario-go` -> mario-goulart
07:07 (join) ysangkok
07:07 ysangkok: hello?
07:09 ysangkok: how do i do (apply or (list #t #f #f)) ?
07:11 DT``: ysangkok, you can't, or is a macro.
07:12 ysangkok: yeah i found out
07:12 ysangkok: thanks so much for answering DT``! :)
07:12 DT``: np.
07:12 DT``: btw, there's ormap.
07:12 ysangkok: yeah i know but ormap takes a function
07:13 DT``: rudybot, (ormap values (list #t #f #f))
07:13 rudybot: DT``: your sandbox is ready
07:13 rudybot: DT``: ; Value: #t
07:13 ysangkok: ah
07:13 ysangkok: i was thinking about doing (lambda (x) x)
07:13 ysangkok: what is preferrable?
07:13 DT``: it's the same (with one argument), I prefer values.
07:13 ysangkok: okay
07:14 ysangkok: it's better than   (foldl (lambda (i j) (or i j)) false list1) which is what i do now
07:15 ysangkok: ah! turns out i can't use "values" since i'm restricted to use HtDP languages
07:15 ysangkok: so i'll do the NOOP lambda instead
07:16 DT``: didn't it had `identity'?
07:17 ysangkok: yes it does :) thanks a lot
07:20 (join) spdegabrielle
07:51 (nick) lucian_ -> lucian
07:53 (quit) spdegabrielle: Quit: Page closed
07:53 ysangkok: what's the best way of getting the index of value in a list?
07:54 ysangkok: currently i use: (- (length (memf (lambda (x) (equal? needle x)) (reverse haystack))) 1)
07:55 ysangkok: rudybot, (- (length (memf (lambda (x) (equal? 'c x)) (reverse '(a b c)))) 1)
07:55 rudybot: ysangkok: ; Value: 2
07:58 DT``: ysangkok, why do you need that?
08:05 ysangkok: because i'm trying to find out which of two lists that contains an element from a third list first
08:06 ysangkok: i'm at the uni now and i have to go home. i'll be back in ~15 min !
08:06 ysangkok: but you can answer now if you want
08:06 ysangkok: i'll check the logs
08:06 (part) ysangkok
08:10 (quit) lucian: Remote host closed the connection
08:12 (quit) leo2007: Quit: rcirc on GNU Emacs 23.3.50.1
08:21 (quit) shofetim: Ping timeout: 255 seconds
08:26 hkBst: ysangkok: search both lists simultaneously...
08:28 hkBst: otherwise (let loop ((n 0)) (if (equal? needle (car haystack)) n (loop (+ n 1))))
08:36 (quit) janne: Read error: Operation timed out
08:38 (join) janne
08:40 (join) ysangkok
08:43 (join) leo2007
09:00 ysangkok: hkBst: thanks a lot for the suggestion, but i am restrcited to htdp-intermediate-lambda :/
09:01 hkBst: ysangkok: I am not familiar with its restrictions
09:05 ysangkok: hkBst: i get: let: expected a parenthesized sequence of local name definitions after `let', but found something else
09:06 DT``: ysangkok, try this: (define (index-of x xs) (((lambda (x) (x x)) (lambda (loop) (lambda (xs n) (if (equal? (car xs) x) n ((loop loop) (cdr xs) (+ n 1)))))) xs 0))
09:06 (join) vu3rdd
09:06 (quit) vu3rdd: Changing host
09:06 (join) vu3rdd
09:06 DT``: basically, use a loop.
09:07 hkBst: ysangkok: the named let is syntactic sugar for a letrec... if you don't have that, then maybe use the Y combinator like DT`` suggests...
09:10 ysangkok: hkBst: i have let. doc says: "same as Intermediate’s local, letrec, let, and let*."
09:15 hkBst: ysangkok: in that case something like (let ((loop (lambda (needle haystack n) (if (equal? needle (car haystack)) .... might work.
09:17 ysangkok: DT``: thanks a lot. i was hoping to do it in a cleaner functional way though. also when benchmarking, my version is >50% faster: https://gist.github.com/993116
09:17 DT``: ysangkok, there's nothing more functional than ω!
09:17 DT``: It's also meant to be slow.
09:18 DT``: since you have local, use an internal define to define the loop: (define loop (lambda (xs n) (if (equal? (car xs) x) n ((loop loop) (cdr xs) (+ n 1]
09:18 DT``: *(loop (cdr xs) (+ n 1]
09:19 ysangkok: ω? i don't know if i used the right term, i meant like higher-order-functions vs loops
09:20 DT``: sometimes being smarter is worse, though.
09:20 DT``: (ω is (λ (x) (x x)))
09:20 ysangkok: ah okay
09:21 hkBst: ysangkok: intermediate with lambda has letrec in my racket...
09:21 ysangkok: hkBst: yes it has it, i cited the docs too. they said that too. the error is about something else
09:22 ysangkok: DT``: could you paste the full function? i have difficulties figuring out what you tried fixing with the "*"
09:23 DT``: sure, wait a sec.
09:24 DT``: http://paste.lisp.org/display/122279
09:24 (join) dnolen
09:26 ysangkok: hkBst: this function (let ((loop (lambda (needle haystack n) (if (equal? needle (car haystack)))))))
09:26 ysangkok: hkBst: gives me this error: let: expected an expression after the name-defining sequence, but nothing's there
09:27 hkBst: ysangkok: yeah, you quoted some text which said `let' acts (among other things) like `letrec', but inter+lam has `letrec', so I don't think that quote is correct.
09:29 (join) PLT_Notify
09:29 PLT_Notify: racket: master Matthew Flatt * c0811dc (1 files in 1 dirs): cocoa: fix `get-y' in `frame%' - http://bit.ly/kCvhko
09:29 (part) PLT_Notify
09:31 (quit) sstrickl: Quit: sstrickl
09:31 ysangkok: hkBst: quote is from here: http://goo.gl/DhouH
09:32 (quit) dnolen: Quit: dnolen
09:33 ysangkok: DT``: your second version is faster, but still pretty slow. benchmark: 2*10^6+1 elements. my: 281. dt-ver1: 4415. dt-ver2: 2854
09:33 hkBst: ysangkok: the way you quoted it it seemed like you were saying `let' acted like all four of those forms... but that's not what the doc says.
09:33 (join) dnolen
09:34 DT``: ysangkok, for some reason, the htdp language is really slow, it's ~2000 ms here.
09:34 DT``: in #lang racket, yours is still faster, though (46 vs 50)
09:35 ysangkok: DT``: btw i had to modify your function like this: http://paste.lisp.org/display/122279#2
09:36 ysangkok: DT``: but it kinda boggles me. i thought (reverse ...) would be really expensive. but i guess that it's fast since memf does all the finding...
09:36 DT``: ysangkok, it looks the same to me (look the one above)
09:37 DT``: ysangkok, beats me too, I thought it would have been slow, but it's faster than an explicit loop.
09:37 DT``: is it the JIT doing miracles?
09:37 ysangkok: hmm yeah i must have made an error copypasting
09:41 ysangkok: hkBst: ah yes i'm sorry. i misinterpreted it. i still can't get it to run though
09:43 ysangkok: hkBst: the reason is that i searched for "let", then clicked the function name for the correct language. i didn't look at the function definition syntax since i thought it was irrelevant. i then read "same as ..." and thought that since i clicked on "let", this means that this function is the same as these other functions
09:50 ysangkok: hkBst: I even get an error when running (define (index-hkbst needle haystack) (let ((loop (lambda (needle haystack n) (if (equal? needle (car haystack))))))))
09:50 ysangkok: hkBst: in normal racket
09:50 ysangkok: hkBst: let: bad syntax in: (let ...
09:51 DT``: it's missing a body.
09:51 DT``: (I don't know what hkBst wanted to do)
09:53 hkBst: ysangkok: you need to use letrec instead of let and fill in the ... blanks
09:54 ysangkok: hkBst: you mean closing parenthesises by ..., right?
09:55 ysangkok: hkBst: this is my definition: (define (index-hkbst needle haystack) (letrec ((loop (lambda (needle haystack n) (if (equal? needle (car haystack))))))))
09:55 ysangkok: hkBst: and i get the same "expected an expression" error (with inter+lam now)
09:56 ysangkok: aaah hkBst okay i guess you didn't want me to just close them parantheses then...
09:57 ysangkok: DT``: thanks for the hint
09:58 hkBst: ysangkok: http://paste.pocoo.org/show/395584/ for example
10:01 (join) MayDaniel
10:02 (join) shofetim
10:03 ysangkok: hkBst: yes now it works, thanks
10:11 (part) shofetim: "ERC Version 5.3 (IRC client for Emacs)"
10:13 (quit) dnolen: Quit: dnolen
10:39 (join) bugQ
10:53 (join) samth
11:02 (join) sstrickl
11:04 (quit) misterncw: 
11:06 (join) dnolen
11:19 (quit) guestNaN: Remote host closed the connection
11:20 (join) ckrailo
11:30 (quit) samth: Ping timeout: 260 seconds
11:33 (quit) bugQ: Read error: Operation timed out
11:56 (join) jonrafkind
12:00 (quit) MayDaniel: Read error: Connection reset by peer
12:02 (quit) leo2007: Ping timeout: 260 seconds
12:16 (quit) DT``: Quit: EOF
12:19 (join) DT``
12:22 (quit) hkBst: Remote host closed the connection
12:23 ohwow: http://farm3.static.flickr.com/2676/5755280417_307c972448_b.jpg
12:25 ysangkok: meh, he got the font wrong
12:26 ysangkok: this one's better: http://www.geekologie.com/2007/05/24/mazda-pi.jpg
12:27 (quit) masm: Ping timeout: 250 seconds
12:27 ohwow: omg
12:30 eli: The MA guy has a better license plate number too.
12:32 (join) lucian
12:37 ysangkok: i saw it with e=2.71... too but i can't seem to find it
12:40 jonrafkind: zero=0.0000000000000000
12:41 ohwow: heh
12:43 ysangkok: i meant the irrational number e. approximation is there so that it didn't look like a typo or something
12:54 eli: 1=0.999999999999999999...
13:00 (join) leo2007
13:06 (join) masm
13:20 (join) samth
13:24 (quit) ysangkok: Quit: Leaving.
13:24 (quit) tonyg: *.net *.split
13:24 (quit) ohwow: *.net *.split
13:25 (join) tonyg
13:25 (join) ohwow
13:30 (join) bugQ
13:31 bugQ: aaaah plai doesn't have fold!
13:33 ohwow: huh?
13:34 ohwow: can't you load srfi1?
13:38 bugQ: ok ok yes
13:42 bugQ: but I mean come on, that request is 11.6 years old!! hasn't it been proofed enough yet?
13:47 bugQ: well ok, it looks like some stuff has been added bit by bit
13:48 bugQ: how long until fold gets in? it's a pretty essential iterator
13:48 RandalSchwartz: by the way, the videos of yesterday's interview are now up at twit.tv/floss
13:50 ohwow: yeah
13:50 ohwow: twit is horrible tho
13:51 ohwow: i mean the had ads for literally more than 3 minutes
13:51 samth: bugQ: use `foldl'
13:51 samth: or `foldr'
13:51 ohwow: "OH GOD I LOVE NETFLIX ITS SO GOOD"
13:51 bugQ: samth: now there's an answer that makes sense
13:51 RandalSchwartz: 3 minutes out of 60
13:51 RandalSchwartz: that's less than broadcast tv
13:51 ohwow: It's not a tv show, it's a podcast, jeez
13:52 RandalSchwartz: and yeah, I got a bit carried away, and read more than the copy
13:52 ohwow: and it's about floss
13:52 RandalSchwartz: ohwow - if you open your pocketbook for the $500 or so it costs to produce each show, I'll be happy to take that instead of the ad money
13:52 RandalSchwartz: seriously - there *are* costs
13:52 ohwow: duh
13:52 ohwow: it's a hobby, not a job
13:53 ohwow: of course there are cost
13:53 RandalSchwartz: it's not a hobby for leo
13:53 RandalSchwartz: it *is* a full time job for the 7 people at the cottage
13:53 ohwow: huh
13:54 RandalSchwartz: so leo either puts ads on floss, or he has to subsidize it from his pocket
13:54 ohwow: well they should quit their `job' then and do something else
13:54 (join) PLT_Notify
13:54 PLT_Notify: racket: master Vincent St-Amour * c3da0ba (2 files in 2 dirs): sub-in -> multi-in
13:54 PLT_Notify: racket: master Vincent St-Amour * 3210aa6 (1 files in 1 dirs): Extend multi-in to deal with more general trees.
13:54 PLT_Notify: racket: master Vincent St-Amour * 48b5725 (1 files in 1 dirs): Document extension to multi-in.
13:54 PLT_Notify: racket: master Vincent St-Amour * 2c470ee (1 files in 1 dirs): Tests for multi-in.
13:54 PLT_Notify: racket: master commits c0811dc...2c470ee - http://bit.ly/jMfT7f
13:54 (part) PLT_Notify
13:54 ohwow: probably read some marx too
13:54 RandalSchwartz: ohwow... at this point, you've lost any interest from me.
13:54 ohwow: full-time job as a podcaster, what a joke
13:54 RandalSchwartz: you're just someone who wants everything free all the time.
13:54 ohwow: I am sorry you feel that way
13:55 ohwow: and yes, I want everything  to be free for everybody
13:55 RandalSchwartz: twit.tv is a network of high quality shows, and there are costs in producing that
13:55 ohwow: sorry for not being a republican
13:55 eli: RandalSchwartz: We have a link on the blog post that was made (blog.racket-lang.org), but the embedded video is from youtube -- just because it was easier to plop that in...
13:55 RandalSchwartz: I'm not a republican either.  no need to apologize
13:55 RandalSchwartz: but I do believe in free market.
13:55 RandalSchwartz: yeah - the youtube one is fine
13:55 offby1: Gentlemen, you can't fight here.  This is the War Room.
13:56 ohwow: yeah i do believe in free market too.. i believe that it causes poverty and destruction
13:56 eli: (FWIW, I have no problems with the ads -- the direct benefit to OSS here is very direct.)
13:56 RandalSchwartz: ohwow - let me know what the color of the sky is on the planet you're from then. :)
13:56 ohwow: Racket is a serious project and many people *really* work on it doing a lot of usefull stuff. I don't see any ads on racket-lang.org
13:57 RandalSchwartz: that's because the costs are buried in PhD salaries and grad student sweat
13:57 RandalSchwartz: and apparently, that's sufficient
13:58 RandalSchwartz: but the guy who professionally edits my show doesn't have any motivation to produce FLOSS.  he's just a hired guy.
13:58 RandalSchwartz: similarly, the electric company doesn't care about open source... they just provide electricity, and a bill at the end of the month
13:58 RandalSchwartz: so for FLOSS Weekly to exist, someone has to pony up, about $500 or so per week
13:59 (join) superjudge
13:59 RandalSchwartz: and from that, you get a website, audio, video, hosts, guests, etc.
13:59 ohwow: you pay your hosts and your guests?
13:59 RandalSchwartz: nope.
13:59 RandalSchwartz: just saying, that's the result
13:59 ohwow: also lol @ hiring someone who doesnt give a shit about podcast
13:59 samth: ohwow: this isn't really an appropriate way to behave on #racket
13:59 RandalSchwartz: ohwow - I'm guessing you're young, or naive, or both.
14:00 RandalSchwartz: perhaps never had to live away from school or your parents.
14:00 samth: there's really nothing morally wrong with advertising supported media
14:00 eli: ohwow: RandalSchwartz is absolutely right about the funds that go into developing racket.
14:01 RandalSchwartz: probably believes it's ok to copy copyrighted songs too, "because it helps the performers get more exposure"
14:01 eli: They're disguised in the form of various academic projects, by overall the costs are definitely there and they are pretty big.
14:01 ohwow: eli: why can't twit get donations or other forms of support?
14:01 RandalSchwartz: and we also did try to run floss weekly strictly on contributions
14:02 RandalSchwartz: it's a small piece, but people don't pony up
14:02 ohwow: Plus, I am not really familiar with Racket development cycle, but a lot of OSS projects receive support for volonteers
14:02 RandalSchwartz: how much would *you* be willing to contribute each week for floss weekly, and if so, why haven't you started?
14:02 eli: ohwow: see what RandalSchwartz said -- and that doesn't durprise me.
14:02 eli: s/dur/sur/
14:02 ohwow: RandalSchwartz: I haven't started because that's the first episode I have watched
14:03 ohwow: never heard of that podcast before
14:03 RandalSchwartz: cool.  and now that you've seen it, would you contribute $1 per week, and help encourage 499 of your friends to do so?
14:03 RandalSchwartz: if so, we'll stop the ads. :)
14:03 ohwow: and I dont listen to artists, who are against sharing music
14:04 RandalSchwartz: you now sound more bitter than happy
14:04 RandalSchwartz: Sad... in that sense.
14:04 ohwow: I might as well send a donation, but unfortunately I dont have 499 friends, not even on facebook.
14:04 RandalSchwartz: "the world must conform to my rules!"
14:04 eli: ohwow: What RandalSchwartz is telling you is that people in general would not commit in a form of a payed account.  (Eg, see offby1's earlier post.)  Otherwise everyone would be happy to have no-ads pay-to-subscribe service.
14:04 ohwow: I just believe that copyright system is wrong
14:04 RandalSchwartz: That's not how it works.  The world conforms to a consensus.
14:04 ohwow: Can you please link me to offby1's post?
14:05 RandalSchwartz: eli - very true
14:05 RandalSchwartz: it was tried
14:05 (join) carleastlund
14:05 RandalSchwartz: at the moment, the production costs are paid by sponsor ads.  the bandwidth is paid by network sponsors (and tiny ads).  Leo's salary is the only thing that comes from the contributions, and even that's below his normal paygrade.
14:06 RandalSchwartz: However, we could keep floss weekly ad-free if enough contributions were earmarked that way.
14:06 ohwow: "That's not how it works.  The world conforms to a consensus." <-- also a lot of people are starving to death in this world
14:06 RandalSchwartz: And a lot of people are also having a great time.
14:07 eli: ohwow: Here -- samth: offby1: see lwn.net; offby1: I used to read them!  They had/have a great editor.; offby1: But I am also the problem: I think I stopped reading once they charged $ :)
14:07 RandalSchwartz: If you want to live in an ideal world, build a spaceship, and go there.
14:07 RandalSchwartz: I'll continue to live on Planet Earth in the meanwhile.
14:07 eli: My guess is that he won't mind some small payment if it was painless to do so.
14:08 RandalSchwartz: yeah - a paywall for floss weekly would quickly kill the circulation
14:08 ohwow: Why can't we have change on this planet?
14:08 RandalSchwartz: maybe make it 5% of what it was
14:08 eli: But as someone who needs to pay N bills to N places, I know that it is impractical in general to expect people to do so.
14:08 RandalSchwartz: ohwow - you get change by first acknowledging what there is.
14:08 RandalSchwartz: you can't live in a fantasy world.  it unenrolls others from caring about what you say.
14:09 eli: ohwow: And such things (small payments by what you actually read) have been suggested and tried *many* times.  I don't think that anything ever came close to being practical.
14:09 eli: Except for advertising.
14:09 RandalSchwartz: Yeah - repeated failures.  ads are about the only consistent model
14:09 eli: (Which is unfortunate, because I hate some of the more offensive forms of advertising.)
14:09 ohwow: I am not talking about small payments, I am talking about donations
14:09 RandalSchwartz: I'm very careful to advertise only things that I personally believe in
14:09 ohwow: I have been on a webforum which relies solely on donations
14:10 RandalSchwartz: Leo insists on that from his hosts.
14:10 ohwow: and they manage to raise $400 every month
14:10 eli: And in that sense, I was actually happy for the netflix ads that RandalSchwartz did -- they're a step back to personal radio advertising.
14:10 eli: IE, this way it's very unlikely that you'll try to hear a racket interview and get some viagra ad...
14:13 (join) MayDaniel
14:15 (join) stchang
14:17 ohwow: As for copying songs -- I recommend reading "Against Intellectual Property" and Stallman's book about freedom, I forgot it's name. Sorry for hijacking the topic, but this is a serious issue for me.
14:17 RandalSchwartz: Until that is a majority view, it's not relevant to me.  I live in a world that has heavily entrenched ideas about ownership of creative acts.
14:18 ohwow: RandalSchwartz: I would like to talk to you more, but without spreading offtopic flames in this channel. May I pm you later?
14:18 RandalSchwartz: I *can* suggest that I prefer much shorter copyright lenghts.
14:18 RandalSchwartz: like 14 years for *everything*.  non-renewable.
14:18 RandalSchwartz: I'd be willing to put all of my works under such a clause.
14:18 RandalSchwartz: if I can't monetize my work in 14 years, it wasn't really time for it anyway. :)
14:19 ohwow: But Twit is released under CC?
14:19 RandalSchwartz: sadly, we seem to be slipping ever further into "protect mickey mouse at all costs" territory
14:19 RandalSchwartz: leo's shows are all CC, yes.
14:19 RandalSchwartz: I forget the specific version
14:19 ohwow: and what about your show?
14:20 RandalSchwartz: by-nc-sa
14:20 RandalSchwartz: it's leo's show, not mine.  he's the exec producer, I'm just the producer and host.
14:20 ohwow: oh, sorry
14:20 RandalSchwartz: I have no copyright stake at all in floss
14:20 (join) anRch
14:24 RandalSchwartz: And no, I'm not interested in a private debate.  I have other things to do. :)
14:24 RandalSchwartz: If I liked to debate on this topic, I'd join EFF-ish and FSF-ish mailing lists.
14:24 RandalSchwartz: instead, I just live in the real world.  simpler. :)
14:24 ohwow: farewell then
14:25 samth: lets not be expecting people to leave *this* community b/c they disagree about whether advertising is bad
14:26 samth: ohwow: you should also note that RMS doesn't agree with the positions you've been taking in this discussion
14:27 ohwow: About ads? Well, it's hard to find a person who 100% agrees with all of your views, isn't it?
14:28 (quit) jonrafkind: Read error: Operation timed out
14:30 (join) asumu
14:48 (quit) MayDaniel: Read error: Connection reset by peer
14:59 (quit) vu3rdd: Remote host closed the connection
15:00 (part) RpR`: "ERC Version 5.3 (IRC client for Emacs)"
15:08 (quit) anRch: Quit: anRch
15:29 (quit) ideasman_42: Remote host closed the connection
15:40 (quit) masm: Quit: Leaving.
15:59 (quit) superjudge: Quit: superjudge
16:17 (quit) tonyg: Quit: leaving
16:35 (quit) bugQ: Ping timeout: 276 seconds
16:58 (part) fds
17:14 (join) jeapostrophe
17:15 (quit) jeapostrophe: Client Quit
17:22 (quit) leo2007: Quit: rcirc on GNU Emacs 23.3.50.1
17:24 (join) jonrafkind
17:29 (join) jeapostrophe
17:31 (quit) jeapostrophe: Client Quit
17:42 (join) shofetim
17:52 (join) masm
17:55 (quit) DT``: Ping timeout: 246 seconds
18:03 (quit) dnolen: Ping timeout: 252 seconds
18:09 (quit) sstrickl: Quit: sstrickl
18:53 (join) dnolen
19:02 (join) Demosthenes
19:03 (quit) lucian: Remote host closed the connection
19:04 (quit) asumu: Read error: Operation timed out
19:05 Demosthenes: so, what's everyone's preferred logging solution?
19:06 (quit) carleastlund: Quit: carleastlund
19:08 jonrafkind: (define-syntax (log str x ...) (when verbose? (printf str x ...))
19:08 samth: Demosthenes: have you seen the built in logging system?
19:12 offby1: samth: I found it surprisingly hard to use
19:12 offby1: especially given that I didn't really care about its main feature (dispatching based on the type of the message)
19:12 samth: offby1: i haven't used it much
19:14 (quit) ckrailo: Quit: Leaving...
19:14 Demosthenes: samth: yeah, it blows ;]
19:14 Demosthenes: see offby1's comments ;]
19:16 Demosthenes: i got a code snippet that allows me to bump the loglevel on the fly, but learning threads, sync, and events just to spawn log messages... its overboard
19:17 (join) ckrailo
19:17 offby1: yep
19:18 Demosthenes: neat video off the ml too
19:18 offby1: could probably be fixed by adding some helper functions
19:18 (join) lucian
19:18 offby1: I really only played with it when it was new; I haven't tried to use it.
19:18 offby1: e.g. I don't think I ever integrated it into rudybot
19:18 samth: yeah, it would be nice to have a library on top of it
19:19 Demosthenes: i've built a big, dispatching packet parser, and the ability to specify log levels in different portions of the program, to a file or stdout, on the fly...
19:19 Demosthenes: would rock
19:21 Demosthenes: really, i find that i want to trace stuff in a single function, or a step through debugger... the lack of them is annoying
19:28 offby1: I've heard you can do that with DrRacket's "stepper", but it never seems to work for me
19:28 bremner_: the macro stepper or the debugger stepper?
19:29 bremner_: I guess the debugger, from context
19:30 Demosthenes: i'm in geiser, or cli
19:30 Demosthenes: no dr racket here
19:32 offby1: not even sure what the difference is, actually.
19:32 (quit) samth: Ping timeout: 240 seconds
19:33 Demosthenes: i don't use drracket... geiser is the emacs repl
19:42 (quit) qha: Quit: Leaving.
19:47 offby1: I use that too.  But I don't know any way to step via geiser.
19:47 offby1: M-x geiser-db RET would be sweet.
19:48 Demosthenes: yep
19:48 Demosthenes: or just racket-debug on the cli, drop me to a prompt
19:49 (quit) lucian: Ping timeout: 240 seconds
19:53 (join) lucian
19:54 (join) qha
19:54 Demosthenes: mr. flatt had a great interview!
19:59 offby1: I swear that guy looks like his 24 years old
19:59 offby1: must be the hair gel
19:59 offby1: s/his/he's/
20:04 Demosthenes: *laugh* well, still gave a great interview
20:04 Demosthenes: i need to go read up on contracts...
20:09 (quit) lucian: Read error: Operation timed out
20:10 RandalSchwartz: I hate to add another bin to my already excessive $PATH.  on OSX, can I just symlink the few things in that I might invoke directly, and it'll find the rest of them relatively, or do I need to add it to my PATH?
20:12 shofetim: RandalSchwartz: I think the symlinking would work. Its pretty smart about paths. But never tried it on a Mac.
20:12 (part) shofetim: "off to LUG meeting"
20:13 Demosthenes: (i think i was just watching his interview!)
20:14 eli: RandalSchwartz: It works fine like that by design.  (On linux too, which makes it very easy to just put the directory somewhere and remove when you're done or when you update.  Much easier than the FHS scatter.)
20:15 RandalSchwartz: cool
20:15 RandalSchwartz: it must do a hard pwd then internally?
20:16 RandalSchwartz: I mean if I do "ln -s ~/Projects/Racket/bin/raco ~/bin"
20:16 RandalSchwartz: then I'm saying "raco foo"
20:16 RandalSchwartz: which maps to $HOME/bin/raco
20:16 RandalSchwartz: but the pieces are really over in ~/Projects/Racket/lib/*
20:17 eli: Yeah, `raco foo' does its own dispatch over whatever library registered a `foo' verb.
20:17 RandalSchwartz: but it finds it ok given the above scenario?
20:18 eli: Yes, you shouldn't have any problems.
20:18 RandalSchwartz: probably doing absolute path.
20:18 Demosthenes: hrm, anyone know how to tell regexp to debug verbosely?
20:18 eli: RandalSchwartz: No, it's all relative to the racket binary.  (After chasing symlinks.)
20:18 Demosthenes: perl has a way to tell regexp to dump internal state...
20:19 RandalSchwartz: perl has a way to do everything. :)
20:19 eli: Demosthenes: Racket's `regexp-replace' will accept a function, which you might use to do some debugging.
20:20 eli: But I'm not surprised that Perl has much more support for debugging regexps...  They're used much more...
20:20 RandalSchwartz: I don't think it's compiled in by default
20:20 RandalSchwartz: I think you have to build a special Perl
20:20 Demosthenes: http://perl.active-venture.com/pod/perldebguts-debugregexp.html
20:21 Demosthenes: i've used it before, needed something similar for racket ;]
20:21 RandalSchwartz: you could just post it to a pastebin... and wait for me to look it over. :)
20:23 RandalSchwartz: I've been hacking regex for 30+ years
20:23 Demosthenes: oh no, thats not the issue
20:23 Demosthenes: when capturing more than one item, against a multi-line input string... how else can you debug other than disassembling?
20:24 RandalSchwartz: look at the result? :)
20:24 Demosthenes: when it all fails ;]
20:24 RandalSchwartz: and in Perl, use side-effects to store the vars.
20:24 (join) lucian
20:24 RandalSchwartz: no - even when it fails, I can run it in my head to figure out what's wrong
20:24 Demosthenes: named capture in perl rocks, something racket doesn't have yet
20:24 RandalSchwartz: seriously... I just "grok" this stuff
20:25 Demosthenes: i wrote a multiline extractor because i was sick of making event driven parsers over and over against just so i could multiline extract AND be able to debug / trace.
20:25 Demosthenes: perl's re you can trace.
20:28 RandalSchwartz: even though I don't.  I just "grok" them :)
20:28 Demosthenes: i love re's ;]
20:29 Demosthenes: http://bazaar.launchpad.net/~rladams/+junk/RacketParser/files
20:29 RandalSchwartz: can you have code run on the finalization of an object?
20:29 RandalSchwartz: I'm looking at the "More: ..." docs, and the (define stop (serve 8081)) looks a bit silly
20:29 Demosthenes: eli: i don't see where regexp-replace takes a function in the ref
20:31 (quit) ckrailo: Quit: Computer has gone to sleep.
20:33 RandalSchwartz: whoa.  datalog
20:35 eli: Demosthenes: http://docs.racket-lang.org/reference/regexp.html#(def._((quote._~23~25kernel)._regexp-replace))
20:35 eli: "The insert argument can be either a (byte) string, or a function that returns a (byte) string."
20:36 eli: But looking at your file, it seems that you're implementing a parser, do you know about parser-tools?
20:36 eli: (BTW, using (regexp foo) at runtime means that you pay for the regexp compliation at runtime.)
20:36 Demosthenes: eli: nope
20:37 Demosthenes: but i'm not doing computer languages, i'm doing HL7 and other vendor records :P
20:37 (join) ckrailo
20:38 eli: Demosthenes: As long as it's a proper formal language, it's the same thing...
20:38 RandalSchwartz: perl66 is going to be great at that
20:38 RandalSchwartz: errr.. .perl6
20:39 eli: At parsing?  Aren't there 2000 libraries now?
20:39 Demosthenes: nope. imagine mixes of k/v pairs and such
20:39 Demosthenes: ugly
20:39 Demosthenes: and they tend to diverge and change over time, sometimes i can at least get a version number
20:40 (quit) ckrailo: Client Quit
20:40 Demosthenes: regexp has all the power i need, when multi-line and some sort of debug mode
20:41 (quit) jonrafkind: Ping timeout: 252 seconds
20:44 Demosthenes: eli: lexers look close, but i run better off examples...
20:46 (join) PLT_Notify
20:46 PLT_Notify: racket: master John Clements * 53cfbd2 (1 files in 1 dirs): fix to string to fulfill contract on empty input
20:46 PLT_Notify: racket: master John Clements * 10cb0e6 (1 files in 1 dirs): oops, no match available here
20:46 PLT_Notify: racket: master commits 2c470ee...10cb0e6 - http://bit.ly/jKjJNF
20:46 (part) PLT_Notify
20:46 eli: You should really look at it -- IIRC, the lexer part is very easy, and I think that it can use regexps for the specifications.
20:46 Demosthenes: if you're on linux, try "cat /proc/zoneinfo" and that's almost as nasty ;]
20:47 eli: That looks very easy.
20:47 eli `find-file's instead of `cat'ting, of course...
20:50 Demosthenes: i'm essentially reverse printf'ing, and think of it like text extracting from mainframe reports.
20:50 Demosthenes: the TXR project does similar
20:51 Demosthenes: funny, i do this in racket because i felt i could do the control code i wnated in perl ;]
20:52 (join) jeapostrophe
20:53 Demosthenes: i'm extracting data into a hash, indexing on keys and typecasting or converting
20:54 (quit) jeapostrophe: Client Quit
20:56 (quit) dnolen: Quit: dnolen
21:03 (quit) lucian: Quit: Ex-Chat
21:06 (join) lucian
21:08 eli: Demosthenes: Well, it's similar to the zoneinfo thing, it doesn't sound hard.
21:10 Demosthenes: i've got lines with multiple k/v pairs to extract, inconsistent formatting, all kinds of crazy
21:10 Demosthenes: i wish it were just "k: v" :P
21:10 Demosthenes: lexers sound close, certainly similar to what i was doing
21:11 (quit) masm: Ping timeout: 240 seconds
21:11 (join) jeapostrophe
21:15 Demosthenes: bbiab
21:20 (quit) Demosthenes: Ping timeout: 240 seconds
21:29 (join) Demosthenes
21:29 Demosthenes: right, back ]
21:31 (quit) martinhex: Remote host closed the connection
21:32 (part) jeapostrophe
21:33 (join) jeapostrophe
21:33 (quit) jeapostrophe: Client Quit
21:34 (quit) Demosthenes: Ping timeout: 264 seconds
21:42 (join) tauntaun
21:45 (join) Demosthenes
21:45 Demosthenes: bah, stupid hotel net ;]
21:47 (join) martinhex
21:50 (quit) martinhex: Read error: Connection reset by peer
21:51 offby1 hates Hates HATES hotel internet
21:51 offby1: Demosthenes: lemme guess -- you're in a kinda expensive hotel.
21:51 offby1: If you were in a cheap fleabag, the Internet would be free and reasonably reliable
21:52 offby1: /proc/zoneinfo -- I was expecting timezones
21:52 (quit) Demosthenes: Ping timeout: 240 seconds
21:55 (join) martinhex
22:04 (quit) lucian: Read error: Operation timed out
22:10 (join) asumu
22:22 (join) Demosthenes
22:29 (quit) Demosthenes: Ping timeout: 276 seconds
22:35 (join) jeapostrophe
22:37 (quit) martinhex: Remote host closed the connection
22:45 (quit) jeapostrophe: Quit: jeapostrophe
22:48 (join) Demosthenes
22:49 (join) samth
22:49 Demosthenes: gah. lemme get you a sample record
22:52 (join) dnolen
22:52 Demosthenes: http://pastebin.com/G436icZW
22:53 Demosthenes: with an event driver parser, i just capture the controller detail, then parse each k: v pair as needed. with that lexer-like paragraph parser i can specify the keys and pre/post processing (ie: convert to number or date)
22:53 Demosthenes: the issue is treating a pattern to match like a sliding window going down a file
22:53 Demosthenes: i'm just saving to hashes.
22:54 (join) Rodolfo
22:55 Rodolfo: hey, is there any way to copy stuff from DrRacket into somewhere else (Blogger.com) and keep colors?
22:56 Rodolfo: I would like to write on a blog post some results from my interactions area
22:57 Rodolfo: (maybe there's somewhere else a syntax coloring facility outputting HTML?)
22:58 samth: Rodolfo: the syntax coloring isn't part of the copy-and-paste, it's just text
22:59 samth: you can try gist.github.com for someplace w/ syntax coloring
22:59 (join) martinhex
23:00 (join) ZachBeta
23:01 Rodolfo: thanks
23:01 Rodolfo: I tried this solution, and actually found it on an old thread in the list
23:02 Rodolfo: but I'd like to keep the default colors in my post, just like "a picture to my reader"
23:02 Rodolfo: Is there any path using http://docs.racket-lang.org/syntax-color/index.html ?
23:03 (join) neilv
23:03 (join) jeapostrophe
23:03 Rodolfo: well, Slideshow can colorize code, and also save as HTML... so I guess there's something in this process that I could use
23:05 (join) leo2007
23:05 (quit) leo2007: Client Quit
23:06 samth: slideshow doesn't produce html, that's scribble
23:08 (quit) jeapostrophe: Quit: jeapostrophe
23:11 Demosthenes: hrm, so i'm still trying to read on lexing, it looks like the lex layer provides no dependencies between items. ie: you can say in "k: v" k is a key, v is a value, but then you have to pass it to a parser to understand they are related or BOTH are required to make a valid pair.
23:11 (quit) neilv: Quit: Leaving
23:15 Rodolfo: samth: yeah, that's right, my mistake
23:16 Rodolfo: anyway, is there any chance to produce the same coloring from DrRacket in HTML (in a sane way)?
23:16 samth: scribble is the best way
23:17 samth: see the `scribble/manual' library
23:18 (nick) ec|detached -> ec|irssi
23:19 Rodolfo: thanks
23:23 (nick) ec|irssi -> ec|detached
23:25 Demosthenes: hrm, and i have to match every k/v pair in a multiline record for the record to "count", i guess i'm lexing and parsing at the same time
23:25 (join) sstrickl
23:31 (join) neilv
23:34 (quit) samth: Ping timeout: 244 seconds
23:41 (part) Rodolfo
23:43 (quit) tauntaun: Quit: Ex-Chat
23:53 (join) jeapostrophe