00:00:12 <boily> speaking of flowery stuff, what is that one? → https://www.dropbox.com/s/yiki722p5addshq/2014-05-30%2018.05.16.jpg
00:00:45 <kmc> douglass_: ^ do you know this flower
00:01:32 <boily> douglass_ knows flowers? I'll remember that :)
00:02:12 -!- tertu has joined.
00:02:20 <douglass_> the flowers look very generic rosaceae but could be something else
00:02:58 <douglass_> those notched leaves are probably more distinctive
00:03:33 <oerjan> boily: an expert in all plantal and fungal matters, clearly
00:03:43 <boily> the flowers are very small, about ¼” Ø.
00:03:49 <shachaf> Bicyclidine: ([97,98,99], chr)
00:04:46 <boily> oerjan: I can't type the real Unicode Diameter symbol on my keyboard, so I defaulted to Ø.
00:05:25 <oerjan> `unicode DIAMETER SYMBOL
00:05:39 <oerjan> `unicode [DIAMETER SYMBOL]
00:05:40 <HackEgo> U+0000 <control> \ UTF-8: 00 UTF-16BE: 0000 Decimal: � \ . \ Category: Cc (Other, Control) \ Bidi: BN (Boundary Neutral) \ \ U+0001 <control> \ UTF-8: 01 UTF-16BE: 0001 Decimal:  \ . \ Category: Cc (Other, Control) \ Bidi: BN (Boundary Neutral) \ \ U+0002 <control> \ UTF-8: 02 UTF-16BE: 0002 Decimal:  \ \ Category: Cc (Other, C
00:05:41 -!- tertu3 has quit (Ping timeout: 264 seconds).
00:05:46 -!- mhi^ has quit (Quit: Lost terminal).
00:06:23 -!- Bicyclidine has quit (Ping timeout: 252 seconds).
00:06:25 <boily> `unicode DIAMETER SIGN
00:06:48 <boily> oh, right. I'm using urxvt, therefore I can Ctrl-Shift-Hexcode.
00:07:26 <douglass_> I can look harder at it when I'm not at work
00:08:48 <HackEgo> [U+2300 DIAMETER SIGN] [U+23D1 METRICAL BREVE] [U+2300 DIAMETER SIGN]
00:15:02 <boily> all-hands meeting on a Friday afternoon? why not? it's vile, it's horrendous, it makes God do Bad Things to kittens!
00:16:46 -!- ter2 has joined.
00:16:46 -!- tertu has quit (Disconnected by services).
00:18:36 -!- Zyprexa has joined.
00:26:16 -!- tertu3 has joined.
00:28:22 -!- tertu3 has quit (Read error: Connection reset by peer).
00:28:44 -!- tertu3 has joined.
00:29:20 -!- ter2 has quit (Ping timeout: 255 seconds).
00:38:48 -!- Patashu_ has quit (Quit: Soundcloud (Famitracker Chiptunes): http://www.soundcloud.com/patashu MSN: Patashu@hotmail.com , AIM: Patashu0 , YIM: patashu2 , Skype: patashu0 .).
00:41:52 -!- tertu3 has quit (Read error: Connection reset by peer).
00:43:18 -!- ter2 has joined.
00:49:17 -!- tertu3 has joined.
00:52:33 -!- ter2 has quit (Ping timeout: 265 seconds).
00:54:19 -!- ter2 has joined.
00:56:00 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
00:56:08 -!- Patashu has joined.
00:56:30 -!- augur_ has joined.
00:57:17 -!- augur has quit (Ping timeout: 264 seconds).
00:57:54 -!- tertu has joined.
00:58:19 -!- tertu3 has quit (Ping timeout: 260 seconds).
00:59:21 -!- tertu3 has joined.
00:59:21 -!- tertu has quit (Disconnected by services).
01:00:11 -!- ter2 has quit (Ping timeout: 260 seconds).
01:02:00 -!- ter2 has joined.
01:04:22 -!- ter2 has quit (Read error: Connection reset by peer).
01:04:44 -!- ter2 has joined.
01:05:07 -!- tertu3 has quit (Ping timeout: 265 seconds).
01:05:27 <kmc> developing for rustc is a world of rebase pain... between the time you write a patch and when you get it merged, somebody will have renamed a std lib function and literally 2,500 occurrences of it within the rustc codebase
01:05:43 <kmc> servo is meant to build on a specific rustc version that only updates every month or two, but rustc is always meant to be built by the latest version of itself, AND by the "stage0" snapshot binaries, which can be significantly further behind
01:07:17 <kmc> ah well it's all part of the fun really
01:07:23 <kmc> I need a faster machine at home to do this stuff
01:07:42 -!- idris-bot has quit (Quit: Terminated).
01:08:10 <kmc> at least the std lib is split into more compilation units now so it can build more in parallel
01:08:50 -!- tertu3 has joined.
01:12:01 -!- ter2 has quit (Ping timeout: 252 seconds).
01:12:31 <oerjan> i shouldn't really be complaining about getting 55 SO rep just because GHC doesn't have -fwarn-tabs on by default and stackoverflow reformats tabs to 4 spaces so no one else realizes why the code is wrong, but it's a little stupid.
01:12:59 -!- tertu3 has quit (Ping timeout: 240 seconds).
01:19:48 <shachaf> so rep can be redeemed for so many good jams
01:19:58 -!- rottytooth has quit (Quit: HydraIRC -> http://www.hydrairc.com <- s0 d4Mn l33t |t'z 5c4rY!).
01:21:02 <oerjan> i already have jam, whats the problem
01:21:18 <kmc> how is this done shachaf
01:21:45 <shachaf> kmc: i haven't collected enough so rep to find out
01:21:57 <kmc> how do you know it can be done
01:22:09 <shachaf> what else would people be collecting it for
01:23:36 <shachaf> oerjan: what kind of jam do you have
01:23:47 <shachaf> is it Balsamic Caramelized Onion Spread
01:24:07 <Taneb> shachaf, that's more of a chutney...
01:24:29 <shachaf> is it a 15-card booster pack of Magic: The Gathering cards
01:25:17 <oerjan> haven't had that in a while.
01:27:10 <shachaf> apparently this one berry has multeple names
01:31:27 <boily> I had a cruchon of lingonberry jam from ikea. it was good.
01:31:37 <boily> (how do you say fr:cruchon in English?)
01:31:55 <kmc> mcpherrin: dare I go this far into the macro rabbit hole? https://gist.github.com/kmcallister/94de6ddc5098c43fd3ff
01:32:29 <oerjan> itt kmc manages to get macros banned from rust
01:32:41 <kmc> i'll fork the language if that happens
01:33:19 -!- boily has quit (Quit: GLASS CHICKEN).
01:33:23 -!- metasepia has quit (Remote host closed the connection).
01:33:55 <mcpherrin> kmc: I once wrote a program that turned the XQuery spec into an xquery compiler
01:34:13 <kmc> mcpherrin: nice
01:34:15 <kmc> 'night Taneb
01:34:19 <mcpherrin> kmc: so you should use that as your DSL
01:34:43 <mcpherrin> surely there's a grammar for HTML5 you can transform
01:34:59 <kmc> have you seen the HTML5 syntax spec.
01:35:18 <kmc> http://www.whatwg.org/specs/web-apps/current-work/multipage/tree-construction.html#tree-construction
01:36:06 <kmc> it's like spaceteam. 'Set the frameset-ok flag to "not ok"!' 'Reconstruct the active formatting elements!'
01:36:36 <kmc> the tokenizer is a bit more structured; you could maybe translate it, although it would be better for WHATWG to maintain a machine-readable spec that is translated to HTML
01:36:39 <kmc> http://www.whatwg.org/specs/web-apps/current-work/multipage/tokenization.html
01:36:40 <pikhq> I suppose technically it counts as a "grammar"?
01:36:54 <kmc> well it's not actually undecidable, so yeah
01:36:56 <pikhq> Though that's about as much a grammar as is the output of Yacc.
01:37:46 <oerjan> grammars can be undecidable hth
01:38:15 <oerjan> @tell boily http://en.wiktionary.org/wiki/cruchon hth
01:40:42 <kmc> mcpherrin: note that the existing code has shit like start!(mut t) if match_atom!(t.name { ... 30 LINE LONG GUARD ... }) => Done,
01:46:26 <kmc> still really pleased that LLVM can compile match_atom!() into tests on an immediate bitmask
01:53:02 -!- Bike has quit (Ping timeout: 276 seconds).
01:54:03 -!- Bike has joined.
01:58:08 -!- Sprocklem has joined.
02:26:49 <Sgeo> Why don't there seem to be many offline HTML5 validators?
02:27:03 <kmc> i was wondering that once
02:27:25 <kmc> I failed at setting up a local instance of the popular one, forget what it's called
02:27:57 -!- conehead has quit (Quit: Computer has gone to sleep).
02:28:16 <Sgeo> Incidentally, do any of them allow settings to allow for one specific kind of invalidity to pass? >.>
02:28:26 <kmc> don't know
02:28:32 <kmc> which kind?
02:29:01 <Sgeo> Unknown non data- attributes. I blame Angular.
02:31:33 <Bike> kmc: helluva macro
02:32:45 <kmc> only problem is it doesn't exist yet
02:34:02 <Sgeo> I wish ads were better targetted. Don't show me ads for Christian dating sites, and don't show me ads for a game that I'm already addicted to (although ads for fansites still make sense)
02:34:24 <kmc> are you willing to accept more invasive personal tracking to make that happen
02:34:30 <maurer> kmc: they shouldn't need it right?
02:34:37 <maurer> kmc: Like, if they are going to show me an eve ad
02:34:49 <maurer> they can say "He logged on to eve form that IP 30 minutes ago"
02:34:53 <maurer> "instead show him an ad for PLEX"
02:36:39 <kmc> grr, people using historical life-expectancy-at-birth as an indicator of when average adults would die
02:38:04 <kmc> regarding Christian dating: "Look, obviously God wanted us to enjoy ourselves, that's why he invented pills and clubs and lube and hardcore. But he also wanted us to give something back, and that's why he created the homeless, the lepers and the oil spills."
02:41:02 <HackEgo> colacoin unacoin revergcoin mempcoin pohtcoin reszekcoin hesecoin excollicoin luillineigateuropricoin tunimalcoin rocontinucoin homeofthesiscoin cordcoin libacoin sebourgcoin skynurtecoin timebrackcoin numgestcoin minimaltllycoin patrecoin
02:41:25 -!- augur has joined.
02:43:29 -!- augur_ has quit (Ping timeout: 264 seconds).
02:44:47 -!- zzo38 has joined.
02:45:14 -!- fungot has quit (Ping timeout: 252 seconds).
02:46:44 <zzo38> How can you link a program compile with GNU C compiler with a program compile with Visual Basic 6?
02:49:27 <elliott> I'm worried about whatever plan you have, zzo38.
02:50:02 <mcpherrin> VB6 isn't quite gone away enough to make it cool and retro
02:50:58 <pikhq> Nah, go for IBM COBOL.
02:52:40 <zzo38> None of that is the point...
03:06:12 <kmc> pointless cobol
03:13:27 <pikhq> Have you considered telling some of my company's clients?
03:13:42 <Bike> http://en.wikisource.org/wiki/Subjective_Effects_of_Nitrous_Oxide read plz
03:13:54 <pikhq> I'd like to not finish a cobol struct parser kthx
03:14:48 <kmc> oh that guy
03:17:13 <oerjan> pikhq: ooh you actually do cobol-related programming?
03:18:08 <pikhq> Presently? Yes, actually.
03:18:38 <kmc> zzo38: iirc VB 6 can load .dlls
03:18:59 <pikhq> Turns out that some places will pay ludicrous amounts of money for a product that lets you take in random mainframe data and spit out something actually sane.
03:20:14 <zzo38> kmc: Yes I know, but I want it to be static linking
03:20:26 <kmc> then i don't know
03:21:25 <pikhq> It's not exactly *hard*, just irritating, because it's practically a computer archaeology project.
03:21:57 <zzo38> I want to make a Windows GUI program which can be static linked with cross-platform C programs
03:22:57 <kmc> the moves-per-second indicator I drunk-coded for 2048 shows NaN until you use the "restart game" feature at least once
03:23:00 <kmc> ah, javascript
03:23:14 <Sprocklem> zzo38: There's got to be better ways of doing that
03:24:23 <pikhq> This incidentally is *not* the fun part of my job. :P
03:24:43 <zzo38> Sprocklem: Maybe, but, what is it going to be?
03:28:33 <madbr> or js fills missing numerical variables with nan? :O
03:29:00 <kmc> I think because undefined * 0 = NaN
03:29:02 <kmc> but who knows
03:30:28 <madbr> how does js compare to, like, python, lua and perl?
03:30:40 <madbr> which are the other languages in that "category"
03:31:01 <mcpherrin> I suppose it depends how you're trying to classify...
03:31:19 <mcpherrin> it's kinda like python, but with a more C-like syntax, but without python's big std lib
03:31:24 <madbr> my classification depends on the major preformance tradeoffs
03:31:43 <mcpherrin> JS has probably one of the fastests JITs around
03:31:52 <madbr> no garbage collection = C, C++, asm, some others like fortran
03:32:09 <madbr> garbage collection but no dynamic typing = java, C#
03:32:21 <mcpherrin> JS is garbage collected, interpteted/JITted, dynamically typed (but secretly not for perf)
03:32:35 <madbr> dynamic typing = JS, lua, perl, python...
03:32:54 <kmc> madbr: itym C, C++, asm, Rust!
03:33:11 <Sprocklem> madbr: JS is also very weakly typed
03:33:12 <mcpherrin> JS uses closures pretty heavily, including as "classes"
03:33:43 <mcpherrin> JS has a lot of weird type system interactions too, especially with things like implicit conversion to strings and numbers
03:33:54 <Sprocklem> madbr: It's got prototype base object orientation
03:34:21 <madbr> object orientation doesn't really have an impact on performance
03:34:24 <kmc> my record time so far is 220 seconds, which I guess is pretty good, although I'm playing a variant of 2048 which is easier than the official one
03:34:43 <zzo38> I think SQLite also uses a kind of dynamic typing? (This is different from some other SQL implementations which are statically typed)
03:35:19 <kmc> mcpherrin: the biggest difference between JS / asm.js and NaCl is not language but platform
03:35:21 <madbr> inversely dynamic typing does... you can go through all sorts of JIT compilation hoops to try to remove the impact but that's hard
03:35:25 <kmc> asm.js uses standard web APIs
03:35:34 <kmc> NaCl uses things that are chrome-only for all practical purposes
03:35:39 <kmc> you can guess which one mozilla favors...
03:35:40 <mcpherrin> kmc: NaCl with standards APIs would be <3
03:35:52 <kmc> but you can guess which one Google favors, too
03:36:01 <mcpherrin> kmc: I got drunk with Eich the day Google announced Dart; that was funny :p
03:36:04 <madbr> does anything important use nacl yet?
03:36:55 <kmc> there is a list of Google web properties allowed to use NaCl hardcoded into chrome
03:36:59 <kmc> \rainbow{vertical integration}
03:38:23 <mcpherrin> kmc: rust needs emscripten support :p
03:38:37 <kmc> not my problem
03:38:50 <mcpherrin> kmc: but running servo inside firefox :-o
03:39:45 <kmc> we could use WebGL for compositing, and native <canvas> for rendering
03:39:58 <madbr> only thing I've implemented on nacl was some sound support stuff
03:40:05 <kmc> I was actually wondering about implementations of CSS / web layout in JS the other day
03:40:09 -!- conehead has joined.
03:40:13 <kmc> because of some discussion of whether JIT would help you
03:40:17 <madbr> incidentally that's an area where html5 is broken but the nacl stuff isn't
03:40:27 <kmc> Blink already has a JIT for CSS selector matching, but I mean for layout
03:40:42 <kmc> compile the flow tree to native code so you can redo layout extra fast when parts of it change
03:40:48 <kmc> i don't know if it makes any sense, but it sounds pretty cool!
03:41:58 <kmc> pcwalton also wants to store our CSS selectors as *only* machine code and decompile them to implement CSSOM
03:42:01 <kmc> I think I mentioned that here
03:43:11 <madbr> like, sound is the kind of stuff that only really work well in compiled languages tbh
03:43:37 <kmc> but JS running in a modern JS engine is compiled
03:43:53 <kmc> and you have unboxed arrays even
03:44:00 <kmc> I agree it's not realtime, though
03:44:03 <kmc> but neither are a lot of "compiled languages"
03:44:08 <madbr> it's more of a problem of
03:44:19 <kmc> really every implementation out there is somewhere on a spectrum between interpreted and compiled
03:44:27 <madbr> if the garbage collector runs, is that going to take more than 10ms?
03:44:40 <Bike> reminds me, firefox has https://twitter.com/alex_gibson/status/467280491260166146/photo/1 now apparently
03:44:44 <madbr> also sound is like 100% always threaded
03:44:45 <kmc> but this is also a problem in Haskell and OCaml, which are typically native compiled
03:44:57 <kmc> ahead-of-time even
03:45:06 <newsham> way-ahead-of-time compilation
03:45:19 <kmc> it's weird that there aren't more mainstream high level languages which are generally AOT native-compiled
03:45:49 <mcpherrin> kmc: yeah I was talking with pcwalton about that, I think decompiling sounds silly though
03:45:58 <newsham> http://research.microsoft.com/pubs/68190/coots97.pdf
03:46:09 -!- esowiki has joined.
03:46:10 -!- glogbot has joined.
03:46:10 -!- douglass_ has joined.
03:46:13 -!- esowiki has joined.
03:46:13 -!- esowiki has joined.
03:46:14 -!- EgoBot has joined.
03:46:28 <madbr> even java is kinda dubious for sound applications
03:48:36 <madbr> the usual equation is "no garbage collector" => "only languages with explicit deallocation" => "C++ only"
03:49:17 <madbr> what does that mean?
03:49:38 <newsham> its a type system that tracks resources. each resource must be consumed only once, etc..
03:49:39 <Bike> some nerd shit
03:49:51 <kmc> that's Rust's approach yeah
03:49:53 <mcpherrin> http://en.wikipedia.org/wiki/Substructural_type_system
03:50:20 <kmc> deallocation is sort of explicit in that it's deterministic, but it's done in a statically memory safe way
03:50:30 <kmc> it's like C++11 move semantics except actually fully checked at compile time
03:50:58 <kmc> if you have let x = ...; f(x) and x is not something implicitly copyable like a primitive numeric type, it will forbid you from using x after that function call
03:51:05 <kmc> you've transferred ownership of it
03:51:34 <madbr> never tried that kinda stuff
03:51:47 <kmc> same for sending things between threads
03:52:16 <kmc> you can have datastructures shared between threads, but you have to write them in the unsafe dialect of the language, and then assert that they're safe to share
03:53:05 <kmc> madbr: the other major unique thing in Rust is references / "borrowed pointers" with lifetime checking
03:53:22 <kmc> so you can write f(&x) and not transfer ownership into f, but if f tries to stash that pointer away somewhere the compiler will complain
03:53:56 <newsham> "you can peek at my x, but ITS MINE!@#"
03:54:19 <kmc> you can have any number of immutable borrowings of a value, or exactly one mutable borring
03:54:22 <kmc> that's statically checked too
03:54:27 <kmc> prevents iterator invalidation and related issues
03:54:34 <madbr> if that kindof stuff lets the compiler do real anti aliasing, then eventually it might become faster than C++ and then rust will see much more adoption
03:54:56 <kmc> that's the hope
03:55:09 <madbr> it's kindof dependent on cpu architectures tho
03:55:24 <kmc> although I think approximately as fast as C++, with no segfaults or use-after-free vulnerabilities is also a pretty compelling value proposition
03:56:12 <kmc> and if it's not as fast, you can do whatever you'd do in C or C++ within the unsafe dialect
03:56:16 <kmc> or link to actual C
03:56:30 <mcpherrin> yeah zero overhead to call C is pretty compelling
03:57:08 <mcpherrin> more ffi support for c++ would be nice though
03:57:10 <Bike> you can tell i'm not cut out for mathematics because i'm still staring at something oerjan said five hours ago
03:57:14 <kmc> C++ is a pit of snakes though
03:57:33 <kmc> I mean first class C++ support would include things like using C++ RAII smart pointers from Rust
03:57:36 <kmc> that's Hard
03:57:38 <kmc> but would be really cool
03:57:41 <pikhq> C FFI is pretty reasonable. I mean, C ain't the nicest around, but it's at least fairly limited.
03:57:42 <madbr> it's got the original sin (everything is pointers)
03:57:56 <pikhq> C++ is A Lot Of Stuff.
03:57:57 <mcpherrin> Teh c++ people are talking about standardizing on an aBI
03:57:58 <kmc> madbr: hm? C++ is one of the few languages where objects *aren't* all pointers
03:58:03 <kmc> mcpherrin: lol
03:58:21 <madbr> kmc: true but arrays sortof are
03:58:26 <madbr> except std::vector
03:58:27 <newsham> [17:54] < Bike> you can tell i'm not cut out for mathematics because i'm still staring at something oerjan said five hours ago
03:58:40 <newsham> isnt the ability to stare at somethign for 5hrs pretty much the main requirement?
03:58:51 <Bike> a necessary but not sufficient condition
03:58:54 <Bike> c.f. video games
03:59:03 <kmc> in Rust there are slices, which are a pointer + length into a string or array owned by someone else
03:59:13 <madbr> the default object for more than one of anything is just an offset with no real size
03:59:13 <kmc> also fixed-size arrays, which are passed by value (unless you take a reference type of them, of course)
03:59:25 <kmc> and heap-allocated growable Vec<T>
03:59:25 <Bike> can you do fortran-style submatrices
03:59:26 <newsham> "you can tell i'm not cut out for mathematics because i stared at something oerjan said 5hrs ago and gave up after 30seconds" is prob a more realistic statement
03:59:42 <Sgeo> Hmm... so Elixir has consistent distinction between 'size' and 'length'... but it would still take time getting used to Elixir to keep it in mind
03:59:53 <Sgeo> (size for O(1) length for needing computation)
04:00:12 <Sgeo> s/needing computation/slower than O(1)/
04:00:29 <Sgeo> "the function should be named size if the operation is in constant time (i.e. the value is pre-calculated) or length if the operation requires explicit counting."
04:00:37 <kmc> Bike: multidimensionally? i haven't seen that, but it should be implementable
04:00:55 <Bike> that that is is silly
04:00:56 <zzo38> I like to use Csound for sound related stuff in computer
04:01:14 <mcpherrin> kmc: O(question mark in a box) :-(
04:01:20 <Bike> kmc: yeah you store stride and stuff
04:01:26 <HackEgo> [U+A66E CYRILLIC LETTER MULTIOCULAR O]
04:01:44 <Bike> i think you can even do ones with less locality, like make a vector out of a diagonal or some shit
04:02:03 <zzo38> Can't you use Csound for audio programming?
04:02:15 <Bike> granted, but you'r eon fire
04:02:29 <kmc> mcpherrin: https://en.wikipedia.org/wiki/Multiocular_O
04:02:44 <Sgeo> Well, getting a bit harder to remember: and/or/not take booleans only, &&/||/! use truthiness
04:02:49 <kmc> a 15th century scribal doodle that made it into the Universal Character Set
04:03:14 <Sgeo> Consistent but hard to remember patterns
04:03:18 <Sgeo> Still better than PHP
04:03:28 * oerjan declares that he has just eaten pizza, then stares at Sgeo while intoning "best before march 24"
04:03:41 <madbr> zzo38 : not really
04:04:07 <zzo38> madbr: Why not really?
04:04:16 <Bike> someone tell me why volterra series were explained to me as "taylor series, but with time dependence" instead of "taylor series, but on functionals"
04:04:26 <Sgeo> Did I say something weird on March 24th?
04:04:28 <Bike> shut the fuck up about 'memory' assholes
04:06:11 <madbr> zzo38 : I'm not well versed in how csound integrates with other stuff so I'm not sure how to answer
04:06:18 <madbr> but it sounds like a bad plan
04:07:13 <zzo38> Well, I find Csound is good for synthesizing audio and writing music, at least. It does work with Open Sound Control, MIDI, Wii remotes, and extension libraries can be written in C.
04:07:19 <madbr> something like "is it still well behaved once you have to compile it into a dll for use in a multithread environment along with a gui"
04:07:37 <madbr> "in a way that's not going to take more code/effort than just doing everything in C++ as usual"
04:07:56 <zzo38> madbr: There are GUI programs for Csound, although I have not used them so I don't know. Some of these GUI programs are even written in Java
04:09:02 <zzo38> I do not particularly like the standard score format in Csound for the purpose of writing music, so I wrote a MML compiler targeting Csound, to make it easier to write music.
04:09:10 <madbr> osc I haven't been impressed with yet
04:09:22 <madbr> it's got a nice type specification etc system
04:09:45 <madbr> but nobody has put on his pants and specified a real standard for note and controller data
04:09:55 <madbr> which is like 75% of the feature set of MIDI
04:10:04 <oerjan> <Bike> you can tell i'm not cut out for mathematics because i'm still staring at something oerjan said five hours ago <-- hm?
04:10:59 <madbr> "you can listen for CC1 because you know the user's m-audio midi controller has a mod wheel and that unless the user has reprogrammed it for whatever reason, it sends CC1"
04:11:12 -!- MoALTz has quit (Ping timeout: 265 seconds).
04:13:18 <madbr> my criticism of those systems is that they're typically not correctly set up for writing music
04:14:02 <zzo38> I haven't used those kind of MIDI devices for writing music, so I don't know
04:14:34 <madbr> writing music requires a good piano roll editor
04:14:44 <madbr> DAWs are basically the only programs that have that
04:14:51 <zzo38> Although I don't prefer to write music "live" anyways (although I know people who do)
04:14:59 <madbr> DAWs can load up VST plugin synthesizer plugins
04:15:10 <madbr> result: everybody uses DAW with VST plugins
04:15:14 <zzo38> I don't see why you need a piano roll editor though. It can be one way, or you can use MML; I prefer to use MML.
04:15:23 <zzo38> I also don't like VST; I prefer to use Csound.
04:15:27 <madbr> I've never gotten into MML
04:15:40 <zzo38> So I use MML for the score and Csound for the instruments.
04:16:02 <zzo38> Since I have found this works better for me than using piano roll editors/VST
04:16:16 <madbr> people that write lots of music use piano roll like >99% of the time
04:17:14 <madbr> newsham : that involves live musicians
04:17:18 <oerjan> <Sgeo> Did I say something weird on March 24th? <-- no. *sigh* i'm being too subtle again... you're just worrying about the health of food all the time so i was joking about the fact i ate a pizza that was > 2 months overdue hth
04:17:20 <madbr> so I guess making them parts
04:17:25 <zzo38> Yes you can use piano roll if you prefer; different people use different ways. Some people like to write on music notation paper.
04:17:33 <madbr> which I guess is notation, rather than piano roll
04:17:40 <Bike> i just poke actual holes in an actual piano roll
04:18:25 <madbr> also, making nice music notation is hard
04:18:33 <zzo38> I do know someone who only writes music live, never writing it down.
04:19:21 <Bike> it'll never catch on.
04:19:35 <zzo38> He then connects the audio output port of a digital piano to a computer to record it onto a CD.
04:19:43 <newsham> next thing you'll have people having impromptu discussions without rehersing
04:20:12 <madbr> zzo38 : afaik that's how you're supposed to use piano rolls anyways
04:20:23 <madbr> record live, use piano roll to fix mistakes
04:20:48 <zzo38> However I myself don't like writing music live, DAW, piano roll, VST, tracker music, and that stuff.
04:21:27 <zzo38> madbr: Actually I think piano roll notation is the kind of thing he wanted to use for this purpose, although he didn't. It would be possible, since this digital piano does have a MIDI output port too.
04:21:29 <kmc> record drunk, piano roll sober
04:21:29 <newsham> what would richard james do?
04:21:33 <Bike> oerjan: the lambda calculus thing.
04:21:54 <kmc> newsham: cocaine's a hell of a drug
04:22:32 <kmc> newsham: close enough?
04:22:47 <newsham> http://en.wikipedia.org/wiki/Richard_James_(musician)
04:23:09 <newsham> oops i meant this one http://en.wikipedia.org/wiki/Aphex_Twin
04:23:09 <Bike> now listening to Die Fantastischen Vier -- Krieger (Aphex Twin Baldhu Mix)
04:24:39 <kmc> newsham: DMT's a hell of a drug
04:24:46 <zzo38> I don't like VST; it requires a GUI, isn't an open standard, and seems for writing live music (which isn't what I do). Perhaps VST is OK for live music writers who don't care if it is usable in anything other than Windows, I suppose.
04:24:53 <oerjan> Bike: ok let me detail it a bit: you write an interpreter for TM's in lambda calculus which works fairly normally, except that (1) it keeps a count of steps such that it never repeats as long as the TM is running (2) if the TM _halts_, instead of halting itself it just restarts from the beginning, resetting the count. then whether that LC interpreter loops <=> whether the TM halts.
04:24:53 <kmc> DMT's a hell of a drukq?
04:25:38 <zzo38> To do effects and instrument sound and so on writing music, I found Csound works much better than anything else I have used.
04:25:48 <newsham> hah, i'm listening to drukqs right now
04:25:51 <Bike> well, that's a bit disappointing, but not unexpected
04:25:58 <zzo38> And if you do like to use Csound and VST, you can do it, in both directions!
04:26:20 <Bike> for my next trick, i shall disprove hilbert's 16th
04:26:32 <Bike> or rather, say something about it here and wait for oerjan to prove it for me
04:26:42 <zzo38> What is Hilbert's sixteenth?
04:27:16 <Bike> http://en.wikipedia.org/wiki/Hilbert%27s_sixteenth_problem
04:34:08 <oerjan> i think you need to work more on your tricks
04:38:19 -!- Zerker has joined.
04:39:39 -!- augur has quit (Ping timeout: 252 seconds).
04:46:25 <zzo38> Do you know how to use SQL for editing music which is already compiled (into a .MOD, .S3M, .MID, etc)?
04:48:18 <zzo38> For working with audio files, SoX works OK (although I think something combining ImageMagick with SoX, allowing the same effects using with sounds and pictures, would be much better), although for working with editing and conversion of music compiled into MOD/S3M/IT/XM/MID files, that a SQL-based system would work best.
04:49:08 -!- augur has joined.
04:49:30 -!- augur has quit (Remote host closed the connection).
04:49:43 -!- augur has joined.
04:50:05 <madbr> What does sox do that you wouldn't normally do in audacity/soundforge/etc?
04:51:30 <zzo38> Command-line functioning
04:51:59 <pikhq> sox is quite nice for scripting.
04:52:36 <pikhq> For instance, it was pretty nice when converting my inexplicable 88.2 kHz 24-bit copy of Distant Worlds 2 here.
04:52:40 <zzo38> I have used Audacity and those things, but I find SoX much better to work with, at least.
04:52:42 <pikhq> (WHY WOULD YOU DISTRIBUTE THAT)
04:53:02 <zzo38> Even if it isn't for scripting.
04:53:21 <madbr> pikhq : that takes about 10 seconds to do in audacity
04:53:37 <pikhq> And it took about that long in sox.
04:53:44 <Bike> valuable seconds i could have spent jamming!
04:53:52 <pikhq> sox is the imagemagick of audio.
04:53:58 <pikhq> This is both its strength and its weakness.
04:55:24 <zzo38> I also find ImageMagick good for pictures, though. However, sometimes I find I want to use some of ImageMagick effects on audio too, and SoX doesn't do that, so I will use SoX to convert to raw, ImageMagick then loads and saves it as a single-row pictures and then use SoX to convert back to other format. It wouldn't be quite best way, but it can work.
04:56:03 <zzo38> Program with both kind of feature, I think is going to work better.
04:57:51 <zzo38> I have written a program for using ImageMagick and METAFONT together, so that it can be used to make up new drawings in a reasonable way, too.
04:58:18 <fowl> are there any programming languages based on mirc script
04:58:55 <zzo38> fowl: I don't know of any other than mirc script itself (and I don't know how mirc script works either)
04:59:39 <fowl> it would make a good base for an esolang
04:59:50 <zzo38> You can try to make such a thing if you like to do so
05:00:02 <fowl> not me, i'm just a poor farm girl
05:00:34 <zzo38> Not me, I don't use mirc script
05:04:34 -!- augur has quit (Ping timeout: 272 seconds).
05:06:21 <zzo38> However I did make esolang based on the classical sequent calculus, where the two negation rules correspond to creating and consuming continuations.
05:06:42 <zzo38> It has first class continuations, but doesn't have any first class functions.
05:11:34 <zzo38> Maybe someone can write esolang based on astronomy and/or astrology, too.
05:13:53 <zzo38> Can you make a esolang where instruction sets are frequency-modulated?
05:14:49 <oerjan> fowl: does myndzi count \o| \o/ |o/
05:18:17 <madbr> zzo38 : considering FM is essentially table lookup, probably
05:18:37 <madbr> provided that you find some way to express the infitely large tape
05:22:16 -!- Zerker has quit (Quit: Colloquy for iPad - Timeout (10 minutes)).
05:27:48 -!- conehead has quit (Read error: Connection reset by peer).
05:29:18 <zzo38> In the Dungeons&Dragons I played today, I was paying a shopkeeper to fix a bag, but then a dwarf came in, and said that the shopkeeper was about to accidentally poke his finger, which he did, and then said "duck!" and stole one of the tools on the wall and left. The shopkeeper then died (he was killed by a crossbow bolt), so I used his tools to fix the bag myself and took back the coins I gave to him.
05:33:12 -!- Zyprexa has quit.
05:39:46 <zzo38> I think I may need to fabricate a duplicate of this situation (but fake, using illusions and invisible walls) in another town.
05:40:16 <kmc> invisible walls!
05:40:19 <kmc> what is the duplicate for?
05:40:46 <zzo38> In order to trap the crossbowmen, which are probably all over the place.
05:41:16 <Bike> did you fix the bag while being shot at
05:42:02 <zzo38> Bike: No, I was hiding underneath the counter. The counter was on the separation between the inside and outside of the shop; normally the customer stand on the outside and the shopkeeper stands inside.
05:43:50 <zzo38> The walls are so that neither the shopkeeper doesn't get killed; the illusions will duck and so on just the same.
05:44:27 <kmc> mcpherrin: I tried to play D&D exactly once and then there was a tornado (in real life, not in the game)
05:44:51 <zzo38> The tool that the dwarf stole was not needed to fix this bag, so that was not a problem.
05:46:24 <kmc> how many crossbowmen do you think there are
05:46:32 <zzo38> It is a sign with "DO NOT READ THIS SIGN" written on it, almost.
05:47:50 <zzo38> kmc: I don't know, but probably there are many all over the place, they were near the docks, on both ends of the river, and in between, and even on the way to this river from a larger boat, and in this North River Village, so probably there would be some in South River Village too.
05:47:52 <Bike> time to take a look at the biomedical program i'll be working on. HOW EXCITING all the globals will be
05:49:43 <oerjan> go into anthropology instead, then you can meet exciting locals
05:49:55 <kmc> hot singles in your area?
05:50:15 <Bike> hot singles in an area
05:50:42 <kmc> what if i prefer to date people who are already in relationships
05:51:02 <shachaf> i played d&d in third grade in hebrew
05:51:14 <Bike> "Representation of the vowel /eh/ in normal and impaired auditory nerve fibers: Model predictions of responses in cats" i feel like fizzie
05:51:54 <shachaf> zzo38: do you play this in person with people or over the internet
05:51:59 <shachaf> the descriptions are v. detailed
05:52:02 <zzo38> shachaf: In person.
05:52:51 <zzo38> We do not use a game board or anything like that; just paper, pencil, books, dice, and thinking.
05:53:32 <Bike> is there a standard license that's "open source, but if you use this and publish something about it, cite so and so"?
05:53:50 <Bike> «#include <math.h> /* Added for MS Visual C++ compatability, by Ian Bruce, 1999 */» get hype
05:54:25 <zzo38> Bike: I think there may be something like that, but I don't know.
05:54:41 <Bike> wowwwww i didn't even know you could declare functions within a function
05:54:44 <zzo38> But I thought you meant citing it in a bibliography or something like that.
05:54:45 <Bike> i'm learning so much!!
05:54:52 <Bike> that's what i meant.
05:55:21 <zzo38> Maybe there is something like that.
05:55:30 <zzo38> Why did you want to know of it?
05:56:01 <zzo38> I do not know the proper way to cite non-literate computer programs in a bibliography, though.
05:56:39 <Bike> zzo38: because this code says "if you use this, please cite x paper" in a comment.
05:56:43 <Bike> they want the paper cited, not the code.
05:57:04 <Bike> * COMPLEX.HPP header file * use for complex arithmetic in C
05:57:10 <zzo38> O, so the code is separate from the paper.
05:57:36 <Bike> yeah. code in a paper doesn't happen much outside CS.
05:58:02 <zzo38> In that case you could easily cite the paper, but where and how?
05:58:29 <zzo38> In a literate program you can cite it in the bibliography, but in other case I don't know how?
05:59:12 <Bike> If you are yourself writing a paper.
05:59:32 <zzo38> Yes, if you are yourself writing a paper, it works.
05:59:41 <zzo38> But what do they mean by "if you use this" in this case?
05:59:57 <zzo38> Using it doesn't seems to necessarily imply that you are writing a paper?
06:00:51 <Bike> well, w hat it actually says is "Please cite these papers if you publish any research results obtained with this code or any modified versions of this code."
06:01:24 <zzo38> O, that's what it says. That works then.
06:01:41 <zzo38> But is that even a license?
06:02:35 <Bike> No. It's an informal request. Which is why I was curious as to whether there was a formal license. In service of this curiosity, I asked several minutes ago whether there was such a license, and then this conversation happened, culminating for the moment in this message you are reading.
06:03:55 <zzo38> I do not know of such a thing, although such an informal request will work. If it is included in a comment in the source file, probably people who do modify it, will keep it there (especially since it says "or any modified versions of this code").
06:07:26 <Bike> struct __COMPLEX { double x,y; }; typedef struct __COMPLEX COMPLEX;
06:08:07 <Bike> n probably but i'm bitter
06:08:40 <elliott> Bike: no because __COMPLEX isn't a name you're allowed to use
06:09:14 <Bike> maybe i should just start cold emailing people to tell them they're out of conformity and should feel bad
06:09:24 <elliott> Bike: unless you are the implementation
06:10:11 <elliott> Bike: probably the most silly common identifier-related nonconformance is naming a type *_t under posix
06:10:28 <Bike> yes i'm just confused by this one though
06:10:49 <Bike> it's just defining a simple complex type (which is redundant in C99 but whatever), and the name __COMPLEX isn't actually used outside of the typedef
06:10:53 <kmc> doesn't C99 have efb
06:11:10 <elliott> oh this is someone else's code
06:11:24 <Bike> do you think so little of me
06:11:25 -!- drdanmaku has quit (Quit: Connection closed for inactivity).
06:11:29 <elliott> Bike: presumably it's to stop people saying struct __COMPLEX because it's an implementation detail or whatever
06:11:42 <zzo38> You can try to write typedef struct { double x,y; } COMPLEX; does it work OK?
06:12:01 <elliott> some people always do the typedef separately. one reason to do that is uniformity: if you have a cycle in struct inclusions, then you cannot just use "typedef struct { ... } name;".
06:12:10 <elliott> so sometimes you must name structs.
06:12:10 <zzo38> (It might not work if you need forward references, though)
06:12:19 <elliott> right, forward refs in general.
06:12:45 <Bike> maybe i'll just replace the code with _Complex and see how it goes
06:13:52 <elliott> Bike: #include <complex.h> and you can say "double complex" etc.
06:14:16 <Bike> right, _Complex is the C99 type isn't it
06:14:25 <Bike> oh, wait, i see.
06:14:27 <elliott> Bike: C99 can't define complex outright because it'd clash with existing stuff
06:14:31 <elliott> so they define _Complex and #define it in the header
06:14:39 <elliott> same situation with stdbool
06:15:04 <Bike> also, this entire program is written in C, but you're supposed to use it as a matlab program. i don't get this field
06:18:12 <fizzie> typedef struct { double re; double im; } zomplex; hth
06:18:39 <fizzie> (That's from the SGI FFT library on IRIX.)
06:20:12 <Bike> maybe i should rewrite this in rust. i bet i could squeeze it into a one line macro invocation
06:20:22 <Bike> ho'ws rust's matlab ffi capabilities,
06:22:15 -!- password2 has joined.
06:23:12 <fizzie> https://dl.dropboxusercontent.com/u/113389132/Misc/20140531-cat_wagon.jpg <- from Kyoto
06:24:27 <fizzie> kmc: https://dl.dropboxusercontent.com/u/113389132/Misc/20140531-boss.jpg
06:24:38 <kmc> now you believe me
06:25:17 <fizzie> I thought you were just being silly at first, but then I saw it.
06:26:13 <kmc> do you see the fnords
06:27:30 -!- MoALTz has joined.
06:27:52 -!- fungot has joined.
06:27:57 <fungot> fizzie: but it seems to be latin1 last i checked
06:28:11 <fungot> kmc: and i modified code affecting the initial image, right?)
06:28:49 <fizzie> That probably explains the absence.
06:30:27 <kmc> does befunge support self-modifying code?
06:31:05 <Bike> matlab has static (in the internal-to-function sense) variables, huh
06:31:28 <fizzie> kmc: Yes, though the fungot code doesn't use very much of that. The ^reload command possibly counts.
06:31:28 <fungot> fizzie: which is? tuples?!" readings. it's much easier
06:33:49 <fizzie> In fact, my half-finished (at most) static AOT Funge-98 compiler was enough to run fungot, sans support for ^reload, IIRC.
06:33:49 <fungot> fizzie: and is an indy much faster than we do today ( eg. i eat steak.
06:34:05 <fizzie> fungot: You can't eat steak, you're a program.
06:34:05 <fungot> fizzie: no it didn't. not even by riastradh!
06:38:35 <shachaf> kmc: imo you should see _Rosencrantz and Guildestern Are Dead_
06:42:10 <shachaf> it is a play that was also conveniently turned made into a movie
06:43:05 <fizzie> fowl: It's an owner-only command, sorry. :/
06:43:16 <fizzie> I'd do it but it'd probably break, it's not used much.
06:43:26 <fungot> fizzie: i'd call them smug." jesse fnord'
06:43:45 <fizzie> Not much to see, really.
06:44:12 <fizzie> Right side of lines 1-2 of https://github.com/fis/fungot/blob/master/fungot.b98
06:45:18 <fizzie> (It's up top so that if the layout of the rest of the code changes, it hopefully won't move around.)
06:46:49 <fizzie> Oh, and also because it needs to blank the old code before loading the new one ('i' does not overwrite cells where the input file has spaces, IIRC) so it must be more or less out of the way.
07:17:27 -!- Bike has quit (Quit: shutdown).
07:18:30 <HackEgo> danddreclist 53: shachaf nooodl boily \ http://zzo38computer.org/dnd/recording/level20.tex
07:18:35 -!- madbr has left.
07:25:07 <zzo38> One day I made a diagram, consisting of four concentric circles. The innermost circle is labeled "Universe". Around that one is a circle labeled "Multiverse". Around that is a circle labeled "Mathematics". Around that is the outermost, labeled "GOD".
07:25:33 <shachaf> what's the label in the space outside the outermost circle
07:26:18 <zzo38> The space outside of the outermost circle isn't part of the diagram, but if it were it would also have to be "GOD". Such a diagram is impossible draw, however.
07:30:39 <zzo38> (It seems possible as written here, but that is because it is too difficult (perhaps impossible) to expression the proper way, so this is the best approximation I can have.)
07:35:33 <zzo38> Is good that you try to ask such question, but, such question wouldn't be sensible any more than trying to figure out the position of omega in the English alphabet, or the page number of the shelf that stores the books.
08:14:56 -!- greg79 has joined.
08:15:31 -!- greg79 has left.
08:41:45 -!- ais523 has joined.
08:58:10 -!- idris-bot has joined.
09:11:35 <Taneb> Making corruption art is fun :)
09:11:49 <Taneb> http://i.imgur.com/zOSp2gp.jpg (made by blanking random bytes)
09:12:10 -!- password2 has quit (Ping timeout: 240 seconds).
09:12:23 -!- nooodl has quit (Ping timeout: 245 seconds).
09:14:24 <mcpherrin> Taneb: http://www.reddit.com/r/glitch_art/top/ is people into that kind of stuff, if you want some upboats for your pictures :p
09:15:44 <Taneb> mcpherrin, I'd rather get it really good first
09:15:49 <Taneb> And, you know, not Nigel Farage
09:17:58 <oerjan> what better image to use for corruption art
09:19:08 <Taneb> oerjan, I don't think Nigel Farage is actually corrupt. Just, you know, hideously wrong on most political issues.
09:22:05 <Taneb> So, other than bit-flipping and byte-blanking, what other effects are easy to produce>
09:22:47 -!- nooodl has joined.
09:23:48 <Taneb> JPEG compression does fun things too
09:27:41 <Taneb> oerjan, can you elaborate on that?
09:28:40 <oerjan> basically set some arbitrary bytes to constants?
09:29:43 <oerjan> sprinkle nigel with "666"s, that sort of thing.
09:40:14 <FreeFull> https://gist.github.com/FreeFull/6059a9be7500a2d6f6c5/c194838b6ad50f65aa02831e2f745bf56046383b Suggestions on improving this quine?
09:40:27 <FreeFull> Other than removing whitespace, which is trivial
09:41:27 <Taneb> Well, it's obviously Rust
09:41:45 <FreeFull> I don't know of any other language with print!()
09:42:04 -!- Gracenotes_ has joined.
09:42:32 <Taneb> @type let (!) = ($) in print!()
09:43:25 <FreeFull> And doesn't have the same behaviour =P
09:43:49 <FreeFull> > let (!) = ($) in print!("{}","Meow")
09:44:02 <FreeFull> Oh, damn, forgot lambdabot doesn't do IO =P
09:44:15 -!- Gracenotes has quit (Ping timeout: 265 seconds).
09:44:16 -!- EgoBot has quit (Ping timeout: 265 seconds).
09:44:28 -!- EgoBot has joined.
09:44:29 <Taneb> Yeah, the behaviour is horribly different
09:44:56 <Taneb> !haskell main = let (!) = ($) in print!("{}","Meow")
09:45:08 <EgoBot> help: General commands: !help, !info, !bf_txtgen. See also !help languages, !help userinterps. You can get help on some commands by typing !help <command>.
09:45:24 <Taneb> Yeah, that's horribly different
09:46:29 -!- oerjan has quit (Quit: leaving).
09:58:51 <Taneb> fizzie, any ideas for making glitches?
09:59:26 <Taneb> I'd prefer ones that could conceivably occur on actual crappy lines.
10:09:46 <fizzie> Try to make errors not detectable by typical ECCs, perhaps? Nobody'll notice, but you'll know.
10:11:27 <mcpherrin> Taneb: well, you could get a crappy line! :p
10:12:31 <fizzie> Remove most significant bit, that happens every now and then.
10:12:41 <fizzie> Or set it according to the parity of the other seven bits.
10:12:48 <fizzie> Maybe not so often these days.
10:13:07 <fizzie> ("Remove" == clear, in this case.)
10:13:42 <mcpherrin> I just had this legit glitch out on me :p
10:19:47 -!- Phantom_Hoover has joined.
10:31:24 <olsner> ooh, my coffee is done
10:52:53 -!- MindlessDrone has joined.
11:00:07 -!- impomatic has joined.
11:04:40 <HackEgo> [wiki] [[User:Rdebath]] http://esolangs.org/w/index.php?diff=39730&oldid=39726 * Rdebath * (+133) /* Interpreter List */
11:08:35 -!- impomatic has quit (Ping timeout: 240 seconds).
11:11:41 -!- impomatic has joined.
11:17:31 -!- impomatic_ has joined.
11:18:58 -!- impomatic has quit (Ping timeout: 240 seconds).
11:37:23 -!- impomatic_ has quit (Ping timeout: 245 seconds).
11:46:16 -!- impomatic_ has joined.
11:58:35 -!- impomatic_ has quit (Remote host closed the connection).
12:04:18 -!- impomatic_ has joined.
12:07:58 -!- boily has joined.
12:10:28 -!- boily has quit (Client Quit).
12:10:59 -!- boily has joined.
12:23:32 -!- yorick has joined.
12:28:25 -!- Phantom__Hoover has joined.
12:29:15 -!- Phantom_Hoover has quit (Ping timeout: 252 seconds).
12:29:27 -!- impomatic_ has quit (Ping timeout: 260 seconds).
12:29:50 -!- Sorella has joined.
12:30:25 -!- Sorella has quit (Max SendQ exceeded).
12:31:47 -!- Sorella has joined.
12:55:48 -!- impomatic_ has joined.
13:04:05 -!- MindlessDrone has quit (Ping timeout: 252 seconds).
13:07:30 -!- impomatic_ has quit (Ping timeout: 252 seconds).
13:08:36 -!- impomatic_ has joined.
13:09:34 <impomatic_> mroman: the first tournament results are in https://twitter.com/xcorewar/status/472725852976062464
13:11:40 -!- MindlessDrone has joined.
13:13:55 <boily> aaaaaah... sweeeeet druuuugs...
13:14:21 <lambdabot> oerjan said 11h 36m 5s ago: http://en.wiktionary.org/wiki/cruchon hth
13:15:37 <boily> it's not quite right. a cruchon here is something that can be as small as a Mason jar, or as big as those fatso pickle jars.
13:16:07 <boily> (hm. I guess I just answered my own question here. qc:cruchon → en:"glass jar".)
13:16:34 <boily> @tell oerjan here cruchon is a glass jar.
13:17:33 <Taneb> impomatic_, are you entering?
13:18:59 -!- impomatic_ has quit (Ping timeout: 240 seconds).
13:22:25 -!- impomatic_ has joined.
13:27:23 -!- impomatic_ has quit (Ping timeout: 245 seconds).
13:28:54 -!- impomatic_ has joined.
13:46:16 -!- impomatic__ has joined.
13:48:38 -!- impomatic_ has quit (Ping timeout: 245 seconds).
13:57:32 -!- impomatic__ has quit (Ping timeout: 255 seconds).
14:00:41 -!- impomatic__ has joined.
14:02:02 -!- Patashu has quit (Ping timeout: 255 seconds).
14:08:15 -!- impomatic__ has quit (Ping timeout: 252 seconds).
14:25:19 -!- MoALTz has quit (Quit: Leaving).
14:27:55 -!- sebbu has quit (Ping timeout: 240 seconds).
15:02:19 -!- impomatic__ has joined.
15:04:57 -!- tertu has joined.
15:19:02 -!- impomatic__ has quit (Ping timeout: 276 seconds).
15:21:13 -!- conehead has joined.
15:21:53 -!- impomatic__ has joined.
15:37:32 -!- impomatic__ has quit (Ping timeout: 265 seconds).
15:38:25 -!- ^v has quit (Quit: Leaving).
15:40:13 -!- boily has quit (Quit: Poulet!).
15:42:56 -!- impomatic__ has joined.
15:47:59 -!- impomatic__ has quit (Ping timeout: 252 seconds).
15:53:04 -!- ^v has joined.
15:53:40 -!- impomatic__ has joined.
15:58:35 -!- impomatic__ has left.
16:21:55 -!- ^v has quit (Ping timeout: 240 seconds).
16:28:39 -!- Bicyclidine has joined.
16:28:51 -!- sebbu has joined.
16:29:24 -!- sebbu has quit (Changing host).
16:29:24 -!- sebbu has joined.
16:30:31 -!- impomatic__ has joined.
17:09:15 -!- AnotherTest has joined.
17:11:22 <HackEgo> [wiki] [[5command]] http://esolangs.org/w/index.php?diff=39731&oldid=39704 * Rdebath * (+118) /* Interpreters */
17:12:47 -!- ^v has joined.
17:21:04 -!- augur has joined.
17:22:38 -!- ais523 has quit.
17:27:17 -!- boily has joined.
17:27:35 <HackEgo> [wiki] [[5command]] http://esolangs.org/w/index.php?diff=39732&oldid=39731 * Rdebath * (+97) Add source
17:38:11 <Taneb> I think I'm writing an esoteric data structure help
17:38:20 <Taneb> It's sort of a heap and a binary search tree simultaneously
17:38:58 <Taneb> No, it's heap sorted on one key and tree sorted on another key
17:40:08 <boily> so it's a keyed treap?
17:41:01 <Taneb> olsner, when the two keys correlate it degenerates into a linked list! :)
17:41:08 <boily> I should try to worm something named "KeyedTreap" in our codebase at work and have it pass code review :P
17:42:00 <myname> Taneb: xor linked list, please
17:48:24 <olsner> hm, so my high-level language rewrite of thousands of lines of assembly code was easy to write but came out buggy ... unfortunately debugging it doesn't seem any easier
17:50:19 -!- ^v has quit (Ping timeout: 240 seconds).
17:50:39 <olsner> yes, once you get to debugging something you have already lost, the only recourse is to write it correctly
17:55:10 <Taneb> http://lpaste.net/104902
17:57:17 <Taneb> I have not tested this at all, btw
18:01:43 -!- ais523 has joined.
18:16:56 <douglass_> http://www.thisiscolossal.com/2014/05/fantastic-fungi-steve-axford/
18:17:05 <kmc> ewige einhornkraft
18:31:29 -!- HackEgo has quit (Ping timeout: 264 seconds).
18:39:57 -!- Bicyclidine has quit (Ping timeout: 252 seconds).
18:41:57 -!- Bicyclidine has joined.
18:49:37 -!- drlemon has joined.
18:54:45 <Melvar> ( :t \f, a => (x ** (f x, x -> a))
18:54:46 <idris-bot> (input):1:14:When elaborating argument x to constructor Ex_intro:
18:55:04 <Melvar> ( :t \f, a => (x : _ ** (f x, x -> a))
18:55:04 <idris-bot> (input):1:23:When elaborating argument x to constructor Ex_intro:
18:56:07 <Melvar> ( :t \f, a => the Type (x ** (f x, x -> a))
18:56:07 <idris-bot> \f => \a => the Type (x ** (f x, x -> a)) : (Type -> Type) -> Type -> Type
18:56:10 <impomatic__> Photo from the CW tournament http://codu.org/logs/log/_corewars/2014-05-31
18:56:35 <Melvar> < boily> the flowers are very small, about ¼” Ø. – Also that should be ¼″
18:56:36 <impomatic__> No wait... Photo from the CW tournament https://twitter.com/john_metcalf/status/472810336501133314 :-)
18:57:00 <Melvar> < mcpherrin> kmc: O(question mark in a box) :-( – ⍰
19:05:43 <impomatic__> Phantom__Hoover: evolved program green, hand coded program red. The hand coded program won! :-) +1 for humanity.
19:09:51 <Bicyclidine> +1 for evolved writer over writing evolver
19:10:10 <kmc> i have a small quadcopter now!
19:10:49 <zzo38> It would be expected that evolving would be very slow (but not completely impossible) compared to doing it directly.
19:16:36 <impomatic__> zzo38: for small coresizes (like 80, 800) it's possible to evolve something competitive overnight.
19:17:28 <zzo38> Yes if it is small I would think it can work faster
19:17:37 <zzo38> Same thing with most algorithms really
19:28:44 <kmc> Bicyclidine: too small
19:31:21 -!- ais523 has quit (Read error: Connection reset by peer).
19:31:23 -!- callforjudgement has joined.
19:31:33 <kmc> have to not fly it up my skirt :X
19:35:59 <kmc> no i don't
19:36:32 <kmc> my friend brought his much larger quadcopter over
19:36:43 <kmc> he says you're supposed to charge the battery within a special fireproof sack
19:36:46 <kmc> but he lost the sack
19:37:12 <kmc> now they're off to the north bay to fly said quadcopter and take photos for a wedding invite
19:37:39 <fizzie> Much larger compared to what?
19:39:28 <fizzie> I guess larger than said friend.
19:41:25 <kmc> larger compared to my little one that can't even carry a burrito
19:41:42 <kmc> the bigger quadcopter has already injured an intern
19:41:44 <kmc> it's tasted human blood
19:42:25 <fizzie> Our days as the dominant species are probably soon over.
19:43:23 <kmc> once they learn how to reproduce yeah
19:43:58 <fizzie> I have here the most boring video of a deer imaginable.
19:44:20 -!- Bike has joined.
19:44:33 <fizzie> It's from http://en.wikipedia.org/wiki/Nara_Park and it's just doing boring things.
19:44:51 <Bike> i'm telling you, i have high standards
19:45:18 <olsner> I imagine as boring as imaginable is a lot less boring than possible
19:45:50 <kmc> i saw a tourist chasing a deer who had stolen her map in nara
19:45:55 <kmc> it was pretty fucking funny
19:46:05 <kmc> fizzie: are you going anywhere else besides kyoto and nara?
19:46:08 <olsner> that's probably not the video fizzie's watching
19:46:24 <fizzie> We're actually home already, but we only did a day trip to Tokyo.
19:46:34 <fizzie> (In addition to those two places.)
19:46:42 <kmc> welcome home
19:46:43 <fizzie> (It was a very efficient vacation.)
19:46:45 <Bike> also, re quadcopters, did you know that claude shannon once said "I visualize a time when we will be to robots what dogs are to humans. And I am rooting for the machines."
19:46:53 <kmc> did he actually say that
19:47:44 <Bike> according to wikiquote, at least
19:49:35 <fizzie> https://dl.dropboxusercontent.com/u/113389132/Misc/20140531-nara.jpg no mention about map-stealing
19:50:21 <kmc> Deer used Knock down! It's super effective!!
19:50:25 -!- drdanmaku has joined.
19:50:32 -!- callforjudgement has changed nick to ais523.
19:50:36 <fizzie> Also note that the "attack people, so please be careful" part is a sticker. One wonders what was under it.
19:55:17 <boily> Melvar: why? I like «”».
20:00:15 <kmc> Board of Equalization is such an Orwellian name for a state government body
20:00:38 <boily> which government is it from?
20:00:56 -!- HackEgo has joined.
20:01:46 <Bike> http://bayesianbodybuilding.com/wp-content/uploads/2013/11/BAYESIAN-BODYBUILDING-v5.png
20:06:19 -!- hambubger has joined.
20:22:48 -!- boily has quit (Quit: Poulet!).
20:27:11 <zzo38> If you have a multiplayer pinball game without enough delay before next player, then it is possible any spinner which is spinning, will continue spinning on the next player's turn.
20:33:10 -!- hambubger has left ("Leaving").
20:35:19 <mroman> impomatic__: well... at least i'm not last
20:35:31 -!- lollo64it has joined.
20:39:28 <impomatic__> mroman: no it's not too bad :-) The next tournament will be in October so hopefully you'll be a bit higher...
20:42:51 <mroman> I'm pretty sure one of my beginner's hill wariour would have scored better :)
20:42:59 <mroman> NotAScanner doesn't even get on the beginner's hill
20:59:41 -!- AnotherTest has quit (Ping timeout: 264 seconds).
21:01:12 -!- MoALTz has joined.
21:13:36 -!- lollo64it has quit (Quit: Leaving).
21:17:36 -!- lollo64it has joined.
21:28:02 -!- adu has joined.
21:50:45 <Sgeo> zzo38: unless you make the pinball game have the spinners magically stop spinning
21:50:50 <Sgeo> (I'm not even sure what a spinner is)
21:57:56 <zzo38> Sgeo: Microsoft seems to call them "flags" instead of "spinners"
21:58:42 <zzo38> Would you use magnets to somehow stop them spinning?
21:59:38 <nooodl> "Polynomial functions are a class of functions having many important properties. They are all continuous, smooth, entire, computable, etc.[citation needed]"
22:00:24 <Bike> i'm not sure i could find one source claiming all of those. another encyclopedia, maybe.
22:02:08 <zzo38> How are you supposed to check overflow converting float->int in a C program?
22:02:51 -!- boily has joined.
22:03:21 <ais523> compare the float to (float)INT_MAX and (float)INT_MIN?
22:04:02 -!- ais523 has quit.
22:04:10 <zzo38> Which header files do I use?
22:04:42 <zzo38> And what is best way checking arithmetic errors when doing calculation with floating points?
22:05:17 <HackEgo> [wiki] [[Spleenmap]] http://esolangs.org/w/index.php?diff=39734&oldid=34648 * 68.203.11.82 * (-1) make "it's" into the correct pronoun
22:08:02 -!- Patashu has joined.
22:09:49 <zzo38> Actually I am using my own int max/min anyways
22:10:27 <nooodl> are they really computable? what if your polynomial just: contains non-computable coefficients
22:11:05 <Bike> i don't think i've ever dealt with a polynomial with transcendent coefficients.
22:15:19 <Bike> you can have polynomials with computable roots but uncomputable coefficients. heh.
22:21:09 -!- aretecode has joined.
22:28:19 <int-e> as long as at least one of the roots is uncomputable ...
22:29:09 <Bike> sure, if you're a COWARD
22:29:10 <int-e> I want monic polynomials. Thanks.
22:29:38 -!- adu has quit (Quit: adu).
22:29:59 <Bike> x² + ax - a - 1 then.
22:30:26 <Bike> i guess i don't know what the other root is
22:30:31 <boily> I don't do monopolynomials. I had a friend who went that way and now, he can't even.
22:31:05 <int-e> that has roots -a-1 and 1, one of which is not computable.
22:31:24 <int-e> anyway, never mind.
22:31:35 <int-e> since I clearly missed your point :)
22:31:56 <Bike> i didn't really have a point, i'm just saying silly things
22:32:11 <Bike> kind of a shame there's no conjugate property like for irrationals and complexes, though
22:32:15 -!- Patashu has quit (Disconnected by services).
22:32:15 -!- Patashu_ has joined.
22:32:18 <Bike> probably you don't have that for transcendents either...
22:32:45 * boily wonders if there's a greek prefix for something between mono- and poly-...
22:32:49 <int-e> there is a difficulty in the opposite direction, too. the roots of x^2 - a, with computable a, may not be computable *real* numbers (namely, if it's undecidable whether a < 0).
22:33:46 <int-e> but that difficulty disappears for complex numbers, as far as I can see. If you allow non-monic polynomials though, then ax - 1 = 0 has the same problem.
22:34:38 <Bike> at least they're still entire.
22:34:40 <int-e> or are those not greek? I'm weak.
22:36:32 <boily> sometimes, not the brightest questions drift through my mind.
22:37:23 <boily> int-e: bi- is latin, di- is greek.
22:37:25 <nooodl> un- bi- tri- quad- is latin, mono- di- tri- tetr- is greek
22:38:54 -!- oerjan has joined.
22:39:14 <boily> apparently, there is such a thing as “dodrant-”.
22:40:03 <nooodl> base 3/4 aka dodrantary
22:41:27 <lambdabot> boily said 9h 24m 53s ago: here cruchon is a glass jar.
22:42:48 <int-e> oh wikipedia. "A real number is computable if and only if the set of natural numbers it represents (when written in binary and viewed as a characteristic function) is computable." not.
22:46:32 <Bike> is that not true? "it represents" is a bit rude but it still seems okay to me
22:46:37 <Bike> well the binary thing is also pointless
22:46:56 <Bike> oh, and you need an ordering, i guess...
22:47:19 <int-e> Bike: the trouble is that sometimes you cannot decide whether the next digits should be 01111....1 or 10000....0.
22:52:01 -!- quintopia has changed nick to Quintopia.
22:53:26 <int-e> For example given a Turing machine T, you can define a = 1 + (-1/2)^n if T halts in n steps, and a = 1 otherwise. That's a computable number but you'll have a hard time figuring out whether it's 1.000... or 0.111...
22:55:33 <oerjan> int-e: it's nevertheless true, as for those ambiguous real numbers _both_ representations are obviously computable.
22:56:03 <int-e> oerjan: if T halts then there is no ambiguity at all.
22:57:11 <oerjan> int-e: the thing is, being computable is a property of the real number, not of the TM used to define it.
22:58:01 <oerjan> so it doesn't _matter_ whether you cannot compute the digits from knowing only the TM
22:58:34 <oerjan> and the only numbers where you cannot compute the digits are numbers which are also represented by halting machines.
22:58:53 <elliott> int-e: um, but 0.111... = 1?
22:59:43 <oerjan> elliott: the point is that you cannot prove in finite time that it's exactly equal to 1, so you don't know that you're not forced to choose 0 or 1 at some point
23:00:34 <oerjan> in a sense, int-e is correct if you are thinking in constructive logic, but not if you are assuming excluded middle for the existence of the digit representations
23:01:43 <oerjan> i suppose in constructive logic you end up with not being able to prove that the constructive real number _has_ a digit representation.
23:01:44 <nooodl> computability theory sounds like it's very easy to say wrong things about!
23:01:59 <Bike> i can assure you that it's very easy to say wrong things in general
23:02:31 <Bike> see, there you go
23:02:50 <Bike> how presumptuous
23:02:51 <M28> i has a solution
23:03:00 <M28> let's define 2 = 1
23:03:06 <M28> then the mistery is solved!
23:03:14 <M28> brb gonna get my nobel
23:03:30 <int-e> oerjan: Subtle. Hmm. So the point is that you can give some advice to the program producing the binary representation.
23:04:06 <Quintopia> i really want to believe that story about nobel's wife cheating with a mathematician is true
23:04:11 <oerjan> M28: you mean your abel hth
23:04:32 <int-e> (and the only kind of advice you really need is that "this is a rational number of shape a/2^n, and here is n.")
23:05:09 <oerjan> Quintopia: you know why that isn't true, right
23:05:28 <oerjan> Quintopia: no, but the abel prize is a real prize
23:05:45 <oerjan> intended precisely to close the nobel gap
23:05:47 <Bike> the nobels are all fairly random. i mean, "physiology or medicine"?
23:07:06 <shachaf> my father says of any unsolvable problem "if you solve that, you'll get the nobel prize in mathematics"
23:07:28 <int-e> oerjan: it's weird because there is no computable function from computable reals of the first kind (producing approximations of precision 1/2^n) to the second kind (producing binary expansions)
23:07:41 <oerjan> shachaf: is that just a subtle way of using reductio ad absurdum to say you won't be able to solve it?
23:07:45 <Taneb> Wow, Hackers was such a 90's movie
23:07:56 <shachaf> oerjan: only moderately subtle hth
23:08:00 <boily> Quintopia: QUINTOPIAAAAAAAAAAAAAAAAAAAAAAAIAIAIAIAIAIAIAIAIIIIIIIIIIIIIIIAAAAAAAH!
23:08:32 <oerjan> old cthulhu had a farm
23:09:29 <int-e> (annoyingly I should know this stuff. oh well.)
23:09:44 <Bike> well rounding is uncomputable in general
23:10:25 <Quintopia> oerjan: i just looked it up. the reason it isn't true (in principle, where wife can be substituted by lover as needed) is that it simply happens not to be true of the girlfriends we know about...
23:10:44 * ion computes some well rounded numbers
23:10:45 <shachaf> oerjan: now that i've looked up the lyrics to Old MacDonald Had a Farm, i see the origin of certain lyrics from Chicago
23:11:14 <boily> Chicago and R'lyeh are related?
23:11:17 <oerjan> Quintopia: bah you're overcomplicated it
23:11:50 <oerjan> shachaf: i don't know chicago :(
23:12:06 <Quintopia> oerjan: but it coooooould still be true and history just forgot >.>
23:12:24 <shachaf> oerjan: http://www.metrolyrics.com/roxie-lyrics-chicago-the-musical.html
23:12:49 <shachaf> anyway now i know the origin of "here a * / there a * / everywhere a *"
23:13:07 <Quintopia> wait...it's not from old macdonald?
23:13:53 <Quintopia> oh. wait... what made you think it came from anywhere else?
23:13:57 <shachaf> i guess i'd never heard the full song in english before
23:14:18 <shachaf> i think i've heard it in hebrew
23:14:36 <shachaf> ("uncle moshe had a farm")
23:15:04 -!- MindlessDrone has quit (Ping timeout: 252 seconds).
23:15:20 <Quintopia> well now i've also learned something...
23:15:52 <shachaf> maybe i've heard it in english too and just forgotten
23:16:06 <oerjan> boily: the only problem is they're non-euclidean antipodes, so you won't find r'lyeh just by changing coordinate signs hth
23:17:03 <Quintopia> actually, i happen to know that many parts of chicago are very close to Fairy
23:17:32 <Quintopia> there are probably parts very close to R'lyeh also
23:17:57 <Quintopia> how would you like to live in that house?
23:18:16 <boily> oerjan: I don't change signs anymore now that I work with dental prosthesis. last time I did that... well. it was... I don't think the result would have fit in a three-dimensional mouth.
23:18:40 <oerjan> boily: did you just denture cthulhu?
23:19:07 <boily> once we expand our clientèle over to the Outer Planes.
23:19:34 <oerjan> (actually changing signs isn't right for longitude anyway)
23:20:16 <boily> axes in dentistry are fun! we have: buccal/lingual, occlusal/cervical and mesial/distal!
23:20:22 <boily> and they're context dependent!
23:27:19 <Taneb> Is it possible to send graphics via telnet?
23:28:03 <Taneb> In such a way that common telnet clients do the write thing.
23:28:11 <Taneb> And I'm not counting ASCII art
23:28:35 <ion> Aww, i was about to link telnet://nyancat.dakko.us
23:28:44 <boily> https://github.com/asciimoo/drawille ?
23:29:29 -!- MindlessDrone has joined.
23:30:19 <Taneb> boily, Telnet isn't 8-bit clan
23:30:41 <zzo38> Doesn't telnet have a way to specify terminal capabilities?
23:52:04 <oerjan> * boily wonders if there's a greek prefix for something between mono- and poly-... <-- oligo- ?
23:52:38 <oerjan> although that's really a subset of poly-, i guess.
23:52:38 <FireFly> So I know what a monopoly is, and what an oligopoly is, but what is a polypoly?
23:57:04 <boily> at least you identified the Fourth Kind of polies. mono, oli, roly, and poly.