00:20 (quit) dnolen: Quit: dnolen
00:21 (join) dnolen
00:46 (quit) otzi: Ping timeout: 260 seconds
01:07 (quit) realitygrill: Quit: realitygrill
01:12 (join) orivej
01:18 (quit) janne: Quit: puf
01:20 (join) janne
01:45 (join) MayDaniel
01:49 (join) masm
01:50 (quit) neilv: Quit: Leaving
01:52 (quit) orivej: Ping timeout: 260 seconds
01:52 (join) superjudge
01:56 (quit) superjudge: Client Quit
02:00 (quit) feor: Ping timeout: 260 seconds
02:07 (join) hkBst
02:11 (quit) MayDaniel: Read error: Connection reset by peer
02:17 (quit) jonrafkind: Ping timeout: 258 seconds
02:29 (join) orivej
02:32 (quit) dnolen: Quit: dnolen
02:52 (quit) mithos28: Quit: mithos28
02:59 (quit) orivej: Ping timeout: 240 seconds
03:17 (quit) spewn: Ping timeout: 258 seconds
03:32 (join) noelw
03:45 (join) orivej
03:50 (join) RacketCommitBot
03:50 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/nkp68w
03:50 RacketCommitBot: [racket/master] New Racket version 5.1.3.8. - Eli Barzilay
03:50 (part) RacketCommitBot
04:25 (nick) Sterkedrankh -> qfr_master_of_Is
04:25 (nick) qfr_master_of_Is -> qfrslaam
04:27 (quit) dmac: Ping timeout: 260 seconds
04:31 (quit) orivej: Ping timeout: 258 seconds
04:55 (join) orivej
05:09 (quit) orivej: Ping timeout: 260 seconds
05:12 (join) orivej
05:14 (part) orivej
05:17 (join) mceier
05:37 (quit) ve: *.net *.split
05:37 (quit) bfulgham: *.net *.split
05:37 (quit) bremner: *.net *.split
05:37 (quit) shachaf: *.net *.split
05:38 (join) ve
05:38 (join) bremner
05:38 (join) bfulgham
05:38 (join) shachaf
07:22 (join) jeapostrophe
07:46 (join) littlebobby
08:14 (quit) jeapostrophe: Read error: Connection reset by peer
08:15 (join) jeapostrophe
08:26 (quit) torche: Quit: torche
08:36 (join) dnolen
09:39 (join) ianj_neu
09:40 (quit) jeapostrophe: Quit: jeapostrophe
10:00 (join) jeapostrophe
10:07 (join) realitygrill
10:20 (join) realitygrill_
10:20 (quit) realitygrill: Read error: Connection reset by peer
10:20 (nick) realitygrill_ -> realitygrill
10:21 (join) torche
10:21 (quit) ozzloy: Ping timeout: 260 seconds
10:21 (join) ozzloy
10:35 (quit) realitygrill: Read error: Connection reset by peer
10:40 (join) realitygrill
10:52 (quit) dnolen: Quit: dnolen
10:57 (quit) realitygrill: Quit: realitygrill
10:58 (join) RacketCommitBot
10:58 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/jtwZMg
10:58 RacketCommitBot: [racket/master] Fixing the handling of request bodies on non-POST methods - Jay McCarthy
10:58 (part) RacketCommitBot
11:01 (quit) hkBst: Read error: Connection reset by peer
11:07 (join) anRch
11:09 (join) realitygrill
11:16 (join) pcavs
11:19 gf3: is there a quicker way to "zip" two lists than: → (for*/list ([letter '('a 'b 'c)] [num '(1 2 3)]) (list letter number))
11:20 stamourv: (map list '('a 'b 'c) '(1 2 3))
11:20 stamourv: `map' can take multiple lists, and as long as the function argument has the right arity, everything works fine.
11:21 gf3: stamourv: amazing, thank you!
11:22 (join) RacketCommitBot
11:22 RacketCommitBot: [racket] plt pushed 4 new commits to master: http://git.io/OI3lGw
11:22 RacketCommitBot: [racket/master] Fix whitespace. - Vincent St-Amour
11:22 RacketCommitBot: [racket/master] Add a field to base types to indicate if they are numeric types or not. - Vincent St-Amour
11:22 RacketCommitBot: [racket/master] Optimize subtyping of numeric types. - Vincent St-Amour
11:22 (part) RacketCommitBot
11:24 (quit) noelw: Quit: noelw
11:41 (join) jonrafkind
11:48 (join) egnarts-ms
11:50 egnarts-ms: racket structs - are they built-in types, like vectors
11:51 egnarts-ms: or it's something that's built over other primitive data structures ?
11:51 samth_away: egnarts-ms: each struct is a new type
11:51 (nick) samth_away -> samth
11:51 jonrafkind: they are built in
11:54 egnarts-ms: jonrafkind: ok, good, they are built-in. But things like define-struct or struct are macros that accept many different struct and field options
11:54 jonrafkind: yea
11:54 egnarts-ms: obviously, those macros (define-struct) are not built-i
11:55 (quit) si14: Remote host closed the connection
11:55 egnarts-ms: this means that any racketeer is able to create some other macros like define-struct or define-struct/contract
11:56 ohwow: egnarts-ms: why do you say that those macros arent built-in?
11:57 egnarts-ms: i mean that those macros are not special forms like "let" or "lambda"
11:57 jonrafkind: let is not special
11:57 egnarts-ms: sorry, i'm a newcomer )
11:57 egnarts-ms: but this doesn't matter
11:58 (join) dmac
11:58 egnarts-ms: so, to put it more clear, i just cannot understand where is the bound between something that is built-in and something i can implement myself
11:59 egnarts-ms: where is that boundary
12:02 samth: egnarts-ms: you can implement just about anything yourself
12:02 samth: i would advise not worrying about deep questions about boundaries, and instead asking "how can i implement this thing that I want"
12:03 egnarts-ms: samth: oh, this is completely inacceptable for me )))
12:04 egnarts-ms: just another example...
12:04 egnarts-ms: in Emacs Lisp, structures are absent at all, as far as i know
12:05 ohwow: samth: hm, I don't think you can implement 'lambda, I might be mistaking tho
12:05 samth: egnarts-ms: structures in elisp: http://www.gnu.org/software/emacs/manual/html_node/cl/Structures.html
12:05 samth: ohwow: that depends what you mean by `lambda'
12:06 samth: the binding of `lambda' in `racket/base' isn't a built-in form
12:06 samth: and keyword arguments (which are an important feature of lambda) aren't built-in
12:07 (quit) dmac: Ping timeout: 260 seconds
12:07 samth: egnarts-ms: what do you mean by "built-in"?
12:08 egnarts-ms: samth: look at this citation from what you have just sent me :
12:08 ohwow: hm
12:08 ohwow: strange
12:08 egnarts-ms: Since the underlying Emacs Lisp system provides no way to create new distinct types, this package implements structures as vectors (or lists upon request) with a special “tag” symbol to identify them.
12:08 ohwow: on top of what do you build lambda then?
12:08 egnarts-ms: this shows that actually structs are NOT built-in in elisp
12:09 samth: egnarts-ms: what that shows is that you can't create new distinct types
12:10 samth: ohwow: it's a combination of macros, procedural structs, the `case-lambda' form (which is provided by the core) and the core `#%plain-lambda' form
12:10 ohwow: ah
12:10 egnarts-ms: samth: no no, please wait some more, i'm just trying to express this thought
12:10 samth: egnarts-ms: i don't know what you mean by "built-in"
12:10 ohwow: so at the end of the day it's still lambda calculus
12:11 samth: egnarts-ms: i'm here all day :)
12:11 (quit) anRch: Quit: anRch
12:11 samth: ohwow: i wouldn't say that -- the call-by-value lambda calculus is a model of some programs written in a subset of racket
12:12 samth: but racket isn't the lambda calculus
12:15 egnarts-ms: by "built-in" i mean some abstraction of the lowest level. that means that 1) all abstractions that you can ever build are eventually based on this built-ins (they are built upon them, may be through many layers); 2) there cannot be anything more basic, more simple that those built-ins
12:15 egnarts-ms: a good example is cons cell, in all Lisp dialects
12:16 egnarts-ms: you cannot do anything with the car of a cons cell, except read and write. you cannot, for example, attach a property list to th car itself
12:16 egnarts-ms: because pairs are built-in, and i understand this
12:16 egnarts-ms: the same thing about vectors, for example
12:17 egnarts-ms: and it appears that structs in racket are also built-in
12:17 egnarts-ms: am i right ?
12:19 (join) anRch
12:22 (join) dmac
12:24 (quit) anRch: Ping timeout: 260 seconds
12:28 (join) MayDaniel
12:31 (join) anRch
12:32 (join) feor
12:33 jonrafkind: is there an xpath implemenation anywhere for racket?
12:33 jonrafkind: seems like it should be near the xml library
12:36 samth: jonrafkind: there's sxpath on planet
12:36 jonrafkind: webscrapehelper??
12:36 jonrafkind: maybe sxml-path
12:36 jonrafkind: i mean sxml-match
12:37 samth: egnarts-ms: structs are built-in in the sense that they're implemented by a part of the runtime written in C, rather than implemented in Racket
12:38 jonrafkind: im confused by our xml terminology.. theres "xml", "xexpr", and "sxml"
12:39 jonrafkind: "sxml" seems to be a pointless name, its just s-expressions
12:39 egnarts-ms: samth: great, that's clear
12:40 egnarts-ms: but what's still unclear, is how define-struct is connected with this C runtime ?
12:41 egnarts-ms: it's natural to assume that any programmer is able to writ something like "define-struct" himself
12:42 egnarts-ms: and in that case, how to do this ?
12:45 (join) tonyg
12:46 tonyg: Is it guaranteed that (eq? (cons 1 1) (cons 1 1)) == #f ?
12:46 tonyg: I'm looking for a nice way of getting a unique ID
12:46 jonrafkind: egnarts-ms, look at racket/collects/racket/private/define-struct.rkt
12:47 (join) si14
12:47 egnarts-ms: jonrafkind: i did it already. but with no result.. ok, may be i have to delve more into racket source
12:50 jonrafkind: I think 'make-struct-type' is the primitive thing that you can build structs out of
12:50 (join) markmillikan
12:51 egnarts-ms: yeah, i also found this routine after a little reading of Racket sources
12:51 (quit) anRch: Ping timeout: 260 seconds
12:51 (part) markmillikan
12:51 egnarts-ms: anyway, thanks you all guys
12:52 (join) dbushenko
12:53 egnarts-ms: Racket seems to be a very powerful Lisp and Scheme dialect. but at the same time it's so rich that one may easily loose himself ))
12:54 egnarts-ms: * lose *
12:54 egnarts-ms: while trying to find out, what is implemented on top of what
12:55 egnarts-ms: and what layer of abstraction to begin with, to build own abstractions
12:55 samth: tonyg: that's guaranteed
12:56 samth: egnarts-ms: what i'm trying to say is that you don't need to know what is implemented on top of what to build your own abstractions
12:56 tonyg: samth: thanks
12:57 samth: egnarts-ms: you can build abstractions on top of `struct', and you don't have to care about `make-struct-type'
12:58 (join) anRch
12:59 egnarts-ms: samth: your point is clear, but i cannot develop decent software without a solid understanding of how my tools and instruments work
13:00 samth: egnarts-ms: i totally agree with that in one sense -- reading the reference for `struct' is important for building on it (although not all the details are equally important)
13:00 samth: but you don't need to know how `struct' is implemented to have a solid understanding of how it works
13:00 samth: that's why it's an abstraction
13:03 egnarts-ms: ok, put it another way. the tools and abstractions that you use (in the widest sense - any library function is a tool) don't belong to the same layer. they form what's called a pyramid of abstraction layers
13:04 egnarts-ms: some of them are higher in this pyramid, some reside on lower levels
13:04 egnarts-ms: my point is that one must know their relative positioning in this pyramid
13:04 jonrafkind: now you're just generating heat
13:04 jonrafkind: dont bother
13:05 egnarts-ms: sorry, i'm not a native speaker :) what means "to generate heat" in this context ?
13:05 samth: egnarts-ms: i disagree that you need to know that, and i think you don't know that about almost any system you use
13:06 samth: for example, how many layers of caching are there for a file write on your machine?
13:06 jonrafkind: egnarts-ms, it means arguing your point is worthless
13:08 egnarts-ms: samth: you are right, i don't know that and i don't bother with it. but i don't bother because it's sufficient for me to know how to interact with the highest layer only
13:09 (join) shofetim
13:09 egnarts-ms: i build all the software on top of th whole pyramid
13:09 egnarts-ms: of caching
13:09 samth: egnarts-ms: and the same is true for whether the `lambda' is a "core" feature of racket, or built on top of something else
13:09 egnarts-ms: of caching layers
13:10 egnarts-ms: and in Racket, i build the software and interact both with lower level (cons cells) and higher level (a class or a struct)
13:11 (join) jrslepak
13:12 (quit) anRch: Ping timeout: 260 seconds
13:14 (join) anRch
13:15 (quit) realitygrill: Quit: realitygrill
13:21 (join) markmillikan
13:23 (quit) markmillikan: Client Quit
13:48 (join) mithos28
13:52 (quit) egnarts-ms: Quit: Page closed
13:55 (quit) anRch: Quit: anRch
14:18 (quit) littlebobby: Quit: Ex-Chat
14:25 (join) realitygrill
14:31 (quit) dbushenko: Quit: Ex-Chat
14:32 samth: stamourv: do you have a tree with that patch?
14:33 stamourv: The one I just sent you? Yeah.
14:34 stamourv: Too lazy to build?
14:34 (join) dnolen
14:36 jonrafkind: clklein, redex/tests seems to take a significant amount of time for raco setup
14:37 jonrafkind: all of scheme, scribble, and scribblings was setup before redex/tests finished
14:37 samth: stamourv: no, i just prefer using git trees to using patchs
14:37 samth: jonrafkind, clklein: this is why tests belong in tests/
14:38 jonrafkind: well that and when a test is broken it wont break the build :p
14:38 jonrafkind: as i found out multiple times
14:39 samth: i'm very frustrated that certain people don't put their tests in collects/tests/
14:39 stamourv: samth: Well, I have a tree with that on my computer.
14:40 (quit) mithos28: Quit: mithos28
14:42 samth: ah, ok
14:43 jonrafkind: wait raco setup will compile all the stuff in tests/ too
14:44 jonrafkind: or at least some stuff
14:45 samth: jonrafkind: most of it is excluded by `compile-omit-files'
14:45 jonrafkind: oh ok
14:55 (quit) DT``: Quit: 
14:56 (join) RacketCommitBot
14:56 RacketCommitBot: [racket] plt pushed 4 new commits to master: http://git.io/uLet6A
14:56 RacketCommitBot: [racket/master] [honu] function application respects precedence so that dot can bind higher - Jon Rafkind
14:56 RacketCommitBot: [racket/master] [honu] add simple class form. share the function parsing syntax class - Jon Rafkind
14:56 RacketCommitBot: [racket/master] [honu] dont produce an extra void expression in a block. add a for form to iterate over a list. add an example of linq with xml - Jon Rafkind
14:56 (part) RacketCommitBot
15:02 (quit) dmac: Ping timeout: 240 seconds
15:03 samth: stamourv: looks good to me
15:03 samth: we still need to fix the formatting of function types
15:04 stamourv: Right.
15:04 stamourv: I'll look into that.
15:04 stamourv: But just to make sure, do you mean good in terms of code review, or good in terms of UI.
15:07 samth: good in terms of ui
15:07 stamourv: Byte the way, minor annoyance. Since we moved everything to `typed-racket', git log is useless to list changes that happened before that.
15:08 samth: actually, you should either use [] or () consistently for repl info
15:08 samth: use git log --follow
15:08 stamourv: Example: I know I changed printing of case-lamdas at some point
15:08 samth: b/c git is stupid
15:08 stamourv: Oh, I didn't know that.
15:08 stamourv: Is there any reason not to turn it on by default?
15:08 samth: i think it does a content search to find the previous revision
15:09 samth: so it's much much slower
15:09 stamourv: Oh, I see.
15:09 stamourv: No alias for me, then.
15:10 (quit) realitygrill: Ping timeout: 240 seconds
15:10 samth: as of the () vs []: it's this that i'm concerned about:
15:10 samth: > 0
15:10 samth: - : Integer (generalized from Zero)
15:10 samth: 0
15:10 samth: > +
15:10 samth: - : (Number * -> Number) ... [Use (:print-type <expr>) to see more.]
15:10 samth: #<procedure:+>
15:10 stamourv: Yeah, I understood that. I fixed it already.
15:11 samth: ok, great
15:11 stamourv: I just didn't think about it.
15:11 stamourv: And case-lambda printing is fixed, trying it out now.
15:11 (join) john___
15:11 samth: did you measure any speedups from the numeric union changes?
15:11 (nick) john___ -> sir_lewk
15:11 stamourv: Haven't tried benchmarking.
15:11 stamourv: I guess looking at DrDr timing will tell us.
15:13 sir_lewk: while using racket interactively from the command line, is there a way to make a (display ..) in another thread flush?
15:13 sir_lewk: I'm not seeing it actually print until I press enter or whatever
15:14 samth: sir_lewk: see `flush-output'
15:14 samth: rudybot: doc flush-output
15:14 rudybot: samth: http://docs.racket-lang.org/reference/port-buffers.html#(def._((quote._~23~25kernel)._flush-output))
15:14 stamourv: samth: Ok, case-lambda type printing fixed. Now, should we switch to printing `(case-> ...)' instead of `(case-lambda ...)'
15:14 stamourv: ?
15:14 samth: stamourv: yes, definetely
15:15 stamourv: Ok.
15:15 sir_lewk: perfect, thanks
15:17 stamourv: samth: According to DrDr, the numeric subtyping stuff shaved ~10 seconds off running the TR tests. But that could be noise.
15:18 samth: yeah, if you look at the history, it's pretty noisy
15:18 samth: i wish we had better historical benchmarking
15:18 stamourv: Exactly.
15:18 stamourv: What would make it better?
15:19 samth: stamourv: should we change `random' to only expect an `Integer'?
15:19 samth: stamourv: not running 11 other racket processes on the machine
15:21 stamourv: samth: I'll fix `random'.
15:21 stamourv: And I see your point wrt DrDr.
15:21 samth: so, you think we should change it?
15:22 samth: it does seem inconsistent currently
15:22 stamourv: Yes.
15:22 stamourv: Everything else that only works on Nat accepts Int.
15:22 samth: sure
15:22 stamourv: And we can't get more precise information about the return type anyway.
15:26 samth: indeed
15:26 samth: that would be hard :)
15:27 stamourv: Running the tests, them I'm pushing all that.
15:29 samth: great
15:33 (join) dmac
15:34 (quit) dnolen: Ping timeout: 252 seconds
15:34 (quit) pcavs: Read error: Connection reset by peer
15:34 (join) pcavs1
15:35 (nick) pcavs1 -> pcavs
15:41 (join) jao
15:42 (join) RacketCommitBot
15:42 RacketCommitBot: [racket] plt pushed 4 new commits to master: http://git.io/YIU6kg
15:42 RacketCommitBot: [racket/master] Generalize types shown at the REPL, to keep types simple. - Vincent St-Amour
15:42 RacketCommitBot: [racket/master] Switch to square brackets for REPL type display. - Vincent St-Amour
15:42 RacketCommitBot: [racket/master] Fix case-lambda type printing. - Vincent St-Amour
15:42 (part) RacketCommitBot
15:49 (quit) MayDaniel: Read error: Connection reset by peer
15:49 (join) superjudge
15:59 (join) dnolen
16:00 (join) RacketCommitBot
16:00 RacketCommitBot: [racket] plt pushed 3 new commits to master: http://git.io/j6_azA
16:00 RacketCommitBot: [racket/master] adjust check syntax to use the new capability in free-identifier=? - Robby Findler
16:00 RacketCommitBot: [racket/master] add a test so that the current version of files with collapsed snips in them - Robby Findler
16:00 RacketCommitBot: [racket/master] fix the collapsed snipclass so that it cooperates with the wxme library and - Robby Findler
16:00 (part) RacketCommitBot
16:14 stamourv: samth: Do you think Matthew's changes to `free-identifier=?' could help us with out identifier forgery at TR compile time?
16:17 samth: stamourv: maybe
16:17 samth: i don't fully understand it
16:18 stamourv: Neither do I.
16:19 (join) anRch
16:19 samth: from robby's change based on it, it's just free-id=?/phases in the core
16:19 stamourv: Oh.
16:26 (join) realitygrill
16:27 (join) edv
16:49 (quit) superjudge: Quit: superjudge
17:01 (quit) edv: Quit: Page closed
17:02 (join) MayDaniel
17:16 (quit) MayDaniel: Read error: Connection reset by peer
17:30 (join) RacketCommitBot
17:30 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/szpcDg
17:30 RacketCommitBot: [racket/master] Fix xrepl tests. - Vincent St-Amour
17:30 (part) RacketCommitBot
17:42 (quit) anRch: Quit: anRch
17:51 (quit) dnolen: Ping timeout: 252 seconds
18:00 (quit) torche: Ping timeout: 260 seconds
18:10 (join) torche
18:11 (join) mithos28
18:14 (quit) bfulgham: Quit: bfulgham
18:19 (join) lisp_panda
18:21 (quit) flying_rhino: Ping timeout: 260 seconds
18:25 (quit) lisp_panda: 
18:25 (join) flying_rhino
18:26 flying_rhino: I was wondering if anyone has read Steve Yegge's rant http://steve-yegge.blogspot.com/2007/01/pinocchio-problem.html
18:27 offby1: probably, I've read most of his stuff
18:27 flying_rhino: it is long and somewhat rambling, but he makes couple of good points about nature of programming. You don't have to agree but it is an interesting view
18:29 flying_rhino: it is about nature of living sysems vs dead systems and how we should try building the former (instead of latter as we too often do)
18:29 offby1: _all_ his stuff is long and rambling
18:30 samth: i usually find myself to be unimpressed by yegge
18:31 flying_rhino: well, I disagree with him on some stuff. But he makes some good points along with bad points. And he writes in such a way that even if he is wrong about some stuff it doesn't diminish his overall case much (in this case that living system have important advantages over dead systems)
18:34 samth: that essay basically says "systems should be programmable"
18:35 samth: plus that he likes some systems, and doesn't like others, but they don't seem to be related
18:36 flying_rhino: ara you arguing that systems shouldn't be programmable, then?
18:36 flying_rhino: :P
18:37 (quit) feor: Ping timeout: 260 seconds
18:37 (quit) jeapostrophe: Quit: jeapostrophe
18:41 samth: no, not at all
18:41 samth: just that it's not the world's most thrilling argument
18:41 samth: as demonstrated by the part where he concludes that common lisp, a language mostly designed in the 70s, is pretty great
18:43 jonrafkind: rebooting is a symptom of the problem. hes only asking for preserving data/meta-data
18:43 jonrafkind: which in essence shouldn't be super hard to achieve
18:43 flying_rhino: yeah but since most people are content with building system that aren't programmable, and you rarely have many tools that can help you, his point is worth hammering
18:45 flying_rhino: common lisp isn't the greates lang in the world but it has a lot of tools that make it easier to build systems that are programmable, which is significant for his point.
18:47 (quit) pcavs: Quit: Leaving.
18:50 (quit) jao: Ping timeout: 252 seconds
19:05 (nick) samth -> samth_away
19:11 (quit) mceier: Quit: leaving
19:25 (join) realitygrill_
19:27 (quit) realitygrill: Ping timeout: 260 seconds
19:27 (nick) realitygrill_ -> realitygrill
19:32 (quit) masm: Quit: Leaving.
20:08 (quit) jonrafkind: Ping timeout: 252 seconds
20:10 (quit) realitygrill: Ping timeout: 252 seconds
20:10 (join) realitygrill
20:14 (join) dnolen
20:15 (join) otzi
20:24 (join) jonrafkind
20:37 (join) jeapostrophe
20:45 (quit) jeapostrophe: Quit: jeapostrophe
20:49 (quit) jonrafkind: Ping timeout: 260 seconds
20:50 (quit) realitygrill: Ping timeout: 252 seconds
20:53 (join) realitygrill
21:10 mithos28: how experimental are refinement types in TR?
21:27 (quit) dnolen: Quit: dnolen
21:30 (join) neilv
21:46 (join) RacketCommitBot
21:46 RacketCommitBot: [racket] plt pushed 2 new commits to master: http://git.io/QEvM_w
21:46 RacketCommitBot: [racket/master] docs for new read-snip-from-port function in the wxme library - Robby Findler
21:46 RacketCommitBot: [racket/master] add some more snips to the test suite - Robby Findler
21:46 (part) RacketCommitBot
21:47 asumu: Anyone else getting a lot of seg faults with DrRacket today?
21:48 rapacity: I am
21:48 neilv: are you using 5.1.3?
21:49 rapacity: I'm on 5.1.3.7
21:50 neilv: so a nightly is broken? submit a bug report or post on the email list.  get a stack trace from a core dump if you can
21:52 rapacity: http://bugs.racket-lang.org/query/?cmd=view%20audit-trail&pr=12144
21:53 neilv: nice. thanks
21:53 rapacity: it's been crashing with me randomly for a while now
21:53 neilv: do you have core dumps enabled? a stack trace might help
21:55 rapacity: how do I enable it ?
21:58 neilv: depends which distro you're using
21:58 rapacity: hmm so it's not some compilation flag ?
21:58 neilv: http://www.randombugs.com/linux/core-dumps-linux.html
21:58 neilv: that gives some of it
21:59 rapacity: thanks, I'll do that and follow up on my report next time it crashes
21:59 neilv: linux has really made core dumps a pain in the butt. distros tend to default to disabled in user limits, plus there are kernel options to set
22:09 (quit) neilv: Quit: Leaving
22:33 (join) neilv
22:35 (join) dnolen
22:44 offby1 tries to recall the last time he saw a core dump
22:45 neilv: after unix workstations stopped coming with a compiler, it was only a matter of time before core dumps were disabled by default :)
22:49 (part) shofetim: "ERC Version 5.3 (IRC client for Emacs)"
22:56 (quit) dmac: Ping timeout: 260 seconds
22:57 (join) jeapostrophe
23:04 neilv: should the default documentation file for packages be doc.scrbl ?
23:04 (join) RacketCommitBot
23:04 RacketCommitBot: [racket] plt pushed 1 new commit to master: http://git.io/vGDGKQ
23:04 RacketCommitBot: [racket/master] fix errortrace (required API changes), add `syntax-shift-phase-level' - Matthew Flatt
23:04 (part) RacketCommitBot
23:11 offby1: neilv: I dunno.  As opposed to what?
23:11 neilv: "manual.scrbl"?
23:11 neilv: i previously had it "${PACKAGE_NAME}.scrbl" in my old documentation thingy
23:12 neilv: i think i'll just use doc.scrbl
23:14 offby1: can't see what difference it'd make, really.
23:14 (quit) otzi: Ping timeout: 260 seconds
23:22 (quit) dnolen: Quit: dnolen
23:43 neilv: ah, here is where i saw it before: scribble/how-to-doc.html uses the example of manual.scrbl
23:43 (join) dmac
23:52 offby1: ok
23:52 offby1: who ya gonna believe -- me, or some shoddy PLT documentation?!