00:00:00 <nys> apparently arbitrary limits can be constructed from products and equalizers
00:00:07 <nys> so i'd just need to figure out how to do equalizers
00:01:08 <shachaf> Sure. But Haskell doesn't have equalizers, for instance.
00:01:31 <nys> yeah, maybe some types could be.. equalizable?
00:01:41 <shachaf> Easier with dependent types. :-)
00:09:39 -!- GeekDude has joined.
00:11:44 -!- ProofTechnique has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
00:32:39 <zzo38> When parsing a multipart/form-data block is it necessary to consider base64 and quoted-printable encodings?
00:36:07 <pikhq> Yes, it explicitly may be encoded as such.
00:37:53 -!- trout has changed nick to constant.
00:38:01 <zzo38> At least cURL seems to only use binary encodings when sending data
00:38:42 <zzo38> But, I can try to detect it anyways, if I can write a C code to decode base64 encodings.
00:39:12 <pikhq> It will be stated in the headers of the multipart/form-data block.
00:39:18 <pikhq> (content-transfer-encoding)
00:39:26 <zzo38> Yes I know it is mentioned there
00:40:20 <zzo38> Before reading the data, it is necessary to figure out the transfer encoding, field name, and if applicable the MIME type.
00:48:41 -!- constant has quit (Quit: 1 found in /dev/zero).
00:57:17 -!- variable has joined.
01:53:23 -!- Wright_ has joined.
01:53:23 -!- Wright has quit (Read error: Connection reset by peer).
02:00:43 -!- boily has quit (Quit: TRIBUTARY CHICKEN).
02:08:53 -!- Phantom_Hoover has quit (Read error: Connection reset by peer).
02:10:46 -!- nys has quit (Quit: quit).
02:19:28 -!- oerjan has joined.
02:25:25 -!- Wallacoloo has joined.
02:38:46 <HackEgo> Piet is a really colorful programming language.
02:43:25 <HackEgo> bash: line 0: type: 0: not found
02:43:55 <HackEgo> [wiki] [[Special:Log/delete]] delete * Oerjan * deleted "[[Don't bother clicking here.]]": Vandalism: content was: "I did actually fucking bother. #firstworldanarchism" (and the only contributor was "[[Special:Contributions/85.0.153.1|85.0.153.1]]")
02:45:16 <HackEgo> [wiki] [[Special:Log/block]] block * Oerjan * blocked [[User:85.0.153.1]] with an expiry time of 1 month (anonymous users only, account creation disabled): Why isn't "Vandalism" listed as a block reason?
02:45:35 <HackEgo> /hackenv/bin/?: POSIX shell script, UTF-8 Unicode text executable
02:45:42 <HackEgo> #!/bin/sh \ topic=$(echo "$1" | lowercase | sed "s/noo\+dl/nooodl/;s/ *$//") \ topic1=$(echo "$topic" | sed "s/s$//") \ cd wisdom \ if [ \( "$topic1" = "ngevd" \) -a \( -e ngevd \) ]; \ then cat /dev/urandom; \ elif [ -e "$topic" ]; \ then cat "$topic" | rnooodl; \ elif [ -e "$topic1" ]; \ then cat "$topic1" | rnooodl; \
02:46:20 <izabera> can it export to a pastebin?
02:46:56 <fowl> the original esoteric language
02:46:59 <HackEgo> #!/bin/sh \ echo http://codu.org/projects/hackbot/fshg/index.cgi/file/tip/wisdom/
02:48:00 <oerjan> `` sed -i '2curl wisdom' bin/pastewisdom
02:48:07 <HackEgo> http://codu.org/projects/hackbot/fshg/index.cgi/file/tip/wisdom
02:49:03 <oerjan> izabera: see also wisdom.pdf in topic
02:52:18 <oerjan> `` ls wisdom/unlambda*
03:00:55 <tswett> Today's interesting neural net quote:
03:00:59 <tswett> 19:39:12: <zemhill> mroman_.(the post the the other the esoteric and international out out around on wiki: <http://esolangs.org/>. (For the other th other and design esoteric information, the international hub wiki: <http://esolangs.org/> (watching ", expected prefixible for type crapped and probably obviously)" index. <= <-- you could many bit on the point of expression
03:00:59 <tswett> of scope: (general vocale constructor script to the top of what the manipulate "know" and better for lambdabot ...)) \ command \ std
03:02:57 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
03:08:48 -!- bb010g has joined.
03:20:35 <augur> tswett: what was your nick before tsweet? i forget
03:21:06 <tswett> ihope. But dang, that was a long time ago.
03:22:07 <tswett> The corresponding greeting is, of course, "ihellope".
03:22:57 -!- password2 has joined.
03:25:33 <HackEgo> footnote 8/Isn't it fun reading through all the footnotes?
03:25:36 <HackEgo> right/Right is not two wrongs but three lefts.
03:28:36 <tswett> `` ls wisdom/footnote*
03:28:53 <HackEgo> kmc/kmc ran the International Devious Code Contest of 2013
03:29:00 <HackEgo> / is a space, unless you're hackego and don't understand wide characters.
03:29:04 <HackEgo> color/Color is a phenomenon from outer space designed to drive humanity insane and bring forth the new age of Cthulh
03:29:09 <HackEgo> tdh/tdh is the past tense of a successful hth. hth.
03:29:42 <oerjan> <shachaf> It's a little broken due to canary trouble lately. <-- i think possibly `revert has trouble with reverting file creation? and that it may be unrelated to the canary.
03:30:04 <shachaf> Did we have canary errors before that fateful day?
03:30:56 <shachaf> you're either wisdom or againsdom
03:31:27 <oerjan> the error is pretty consistent with someone doing rm canary.orig and not expecting that some fool may have turned it into a directory.
03:32:39 <HackEgo> olist/Update notification for the webcomic Order of the Stick.
03:33:25 <tswett> `` echo uuu >> wisdom/color
03:33:30 <HackEgo> Color is a phenomenon from outer space designed to drive humanity insane and bring forth the new age of Cthulhu.
03:33:44 <HackEgo> Color is a phenomenon from outer space designed to drive humanity insane and bring forth the new age of Cthulhu.
03:33:54 <tswett> Did my echo do anything at all?
03:34:13 <oerjan> tswett: (1) we do _not_ >> into wisdoms. (2) it's actually just a cut off line.
03:34:24 <HackEgo> rm: cannot remove `/home/hackbot/hackbot.hg/multibot_cmds/env/.hg/store/data/canary.orig': Is a directory \ Done.
03:34:39 <oerjan> `` tail -c +10 wisdom/color
03:34:40 <HackEgo> 0olor is a phenomenon from outer space designed to drive humanity insane and bring forth the new age of Cthulhu.
03:34:56 <shachaf> oerjan: why do you hate newlines twh
03:34:59 <HackEgo> Newlines are le/rn's \ biggest weakness.
03:35:04 <oerjan> `` tail -c 10 wisdom/color
03:35:23 <oerjan> shachaf: because they make wisdoms ugly.
03:36:01 <shachaf> do you hate wisdom/newline
03:36:29 <tswett> Does >> add a newline at the beginning of the appended stuff?
03:36:42 <oerjan> tswett: no, there's already a newline at the end
03:37:13 <shachaf> maybe wisdoms shouldn't have a newline at the end hth
03:38:41 <shachaf> i'm tempted to alter learn and le/rn and so on to use echo -n and remove newlines from all the wisdoms
03:39:02 <shachaf> but that would modify the culprits list
03:40:07 <oerjan> shachaf: the thing is, it will _still_ not help with people doing >> because _they_ aren't going to know about removing newlines from what they add.
03:40:30 <shachaf> sure, but other people can use echo -n >>
03:41:02 <shachaf> rather than learn_append which maybe doesn't work for entries with spaces in their names and so on
03:41:22 <HackEgo> bin/learn_append \ bin/learn_append2 \ bin/print_args_or_input
03:41:48 <shachaf> `culprits bin/learn_append2
03:46:49 -!- Wright_ has quit (Ping timeout: 252 seconds).
03:48:29 -!- idris-bot has quit (Quit: Terminated).
04:00:07 <shachaf> copumpkin: whoa, that picture is great
04:00:28 <zzo38> We can make RDF file which is categorizing the HackEgo wisdom files by date, user, history, arbitrary tags, etc. RDF requires strings to consist only of Unicode codepoints (with a few exceptions), but it does have a way to represent arbitrary binary data so you can use that if the file contains invalid or non-canonical UTF-8 sequences or null bytes.
04:01:25 <zzo38> However, the other problem is if the filename contains invalid UTF-8 sequences. I have a way around that too though. URIs in a RDF document that contain non-ASCII characters are *not* supposed to be percent-encoded or Punycode-encoded. If the filename contains invalid UTF-8 sequences, I suggest percent-encoding them anyways.
04:02:57 <zzo38> (My own RDF parser doesn't care about invalid UTF-8 sequences but it is possible that others will, so for compatibility reasons it must be considered.)
04:04:57 -!- Wright has joined.
04:05:37 <shachaf> The picture in question is http://i.imgur.com/6ocuQsZ.jpg
04:06:55 -!- ChanServ has set channel mode: +o oerjan.
04:07:07 -!- oerjan has kicked shachaf Linking NSFL pictures.
04:07:11 -!- oerjan has set channel mode: -o oerjan.
04:08:43 -!- shachaf has joined.
04:09:18 <shachaf> oerjan: I can't tell if that's serious or not.
04:09:32 <shachaf> But it messed up my irssi windows, since I have it set up to reuse windows and I got a /msg
04:09:49 <shachaf> zzo38: http://www.reddit.com/r/MachineLearning/comments/3a1ebc/image_generated_by_a_convolutional_network/a
04:10:31 <shachaf> I mean http://www.reddit.com/r/MachineLearning/comments/3a1ebc/image_generated_by_a_convolutional_network/
04:10:39 <zzo38> O, it is a convolutional network. OK
04:11:22 <zzo38> What is a convolutional network?
04:11:45 <shachaf> It's an artificial neural network that has convolution layers, I think.
04:11:45 <copumpkin> shachaf: yeah! I can't stop staring
04:12:44 <shachaf> A convolution layer is a layer where instead of having an affine map where you can change every weight separately, you have a small convolution matrix thing applied everywhere.
04:13:11 <oerjan> shachaf: i think this pretty much sums it up http://www.reddit.com/r/MachineLearning/comments/3a1ebc/image_generated_by_a_convolutional_network/cs8hsei
04:13:20 <shachaf> So it's like training a whole bunch of the network at once with the same values, I guess.
04:14:45 <shachaf> zzo38: Do you know about artificial neural network and other things like that?
04:31:20 -!- Herbalist has quit (Ping timeout: 240 seconds).
04:41:27 -!- j-bot has joined.
04:53:49 -!- Wright has quit (Ping timeout: 255 seconds).
05:33:58 -!- ProofTechnique has joined.
05:37:58 -!- password2 has quit (Ping timeout: 265 seconds).
06:02:57 <oren\n> That's some scary stuff!
06:03:26 <Sgeo_> https://twitter.com/fuckeveryword is a thing that exists
06:46:17 -!- ProofTechnique has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
07:01:34 -!- J_A_Work has joined.
07:16:44 -!- Wallacoloo has quit (Quit: Leaving.).
07:41:25 -!- variable has quit (Ping timeout: 250 seconds).
07:48:05 <zzo38> How do you directly build a frequency-domain of a decaying tone?
07:59:58 <Jafet> Hmm, wikipedia has disabled HTTP access
08:00:40 <Jafet> That's going to make it difficult to cache
08:01:36 <zzo38> Yes, I told them that too
08:06:45 <zzo38> Make the proxy server that uses HTTP with the client but HTTPS with the Wikipedia server
08:18:16 -!- rdococ has joined.
08:19:20 <zzo38> But there is another way. They said they didn't want problems with accessing insecure connections and then someone tamper with it or steal the setting of cookie or whatever. This can be solved by using a different domain name for insecure connections.
08:20:40 <rdococ> so I heard about this "nullity" thing someone thought up 9 years ago... don't get me wrong, it's total BS, but it makes me think... I wonder what an esoteric language with it built in would be like...?
08:22:47 <rdococ> wait... are you shocked? do you think I like the idea? or do you like the idea and are offended?
08:23:04 <zzo38> I do not quite know the answer though.
08:25:19 -!- hppavilion1 has joined.
08:26:18 <oerjan> we discussed nullity way back and i think we decided it was crap. although here's a more serious way https://en.wikipedia.org/wiki/Wheel_theory
08:26:21 <hppavilion1> I always start with hello worlds when I join something new
08:26:21 <rdococ> oh!!! I know what you were doing...
08:26:32 -!- Patashu has joined.
08:26:45 <rdococ> its okay... its kinda funny actually... but for a second I didn't know what you were talking about...
08:27:06 <hppavilion1> Wait, are you talking to me? I doubt it, but just checking
08:27:15 <HackEgo> hppavilion1: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: <http://esolangs.org/>. (For the other kind of esoterica, try #esoteric on EFnet or DALnet.)
08:28:48 <zzo38> Is nullity meaning a name for zero divided by zero (even though of course you cannot divide by zero)? I think I heard "nullity" meaning it once?
08:29:46 <rdococ> http://www.bbc.co.uk/berkshire/content/articles/2006/12/06/divide_zero_feature.shtml ... don't get me wrong, 0 / 0 would equal nullity, but that's because 0 / 0 would equal anything if you let it...
08:29:56 <zzo38> I suppose it can be useful to have a name for it in some cases, even though it is not actually a possible operation.
08:29:59 <hppavilion1> I just spent the last few hours writing a Befunge Interpreter in python
08:30:09 <oerjan> there's already an esolang including nullity it's called IEEE 754 hth
08:30:35 <rdococ> a / b = c means c * b = a. applying that to 0/0, we get 0 / 0 = x means x * 0 = 0... and every number satisfies that, including nullity itself
08:30:52 <zzo38> Yes, I can see that
08:31:36 <zzo38> It isn't what I meant though; what I meant is you are explicitly dividing zero by zero. For example to figure out the percentage of questions you answered correctly on a test even though there are no questions on the test.
08:32:10 <oerjan> i'm not sure the befunge experts are awake at the moment
08:32:18 <zzo38> It isn't a number, but rather some kind of convenient shortcut in some cases.
08:32:24 <oerjan> fungot: except you, of course!
08:32:24 <fungot> oerjan: fnord is a fetish, now is. anyway, obviously tusho's dialect forbids " got" it is? :) where would you like
08:32:45 <zzo38> You can't do any proper mathematical operations with it.
08:33:32 <rdococ> IEEE 754 might have been vandalised... there's nothing on the article
08:34:10 <rdococ> you said there was an esolang called IEEE 754
08:34:31 <oerjan> https://en.wikipedia.org/wiki/IEEE_floating_point
08:35:21 <rdococ> for most purposes, I would say 0 / 0 = 1.
08:35:51 <hppavilion1> I think that the Befunge interpreter is the longest thing I've ever written. At <500 lines. Which is sad.
08:36:02 <oerjan> hppavilion1: i think the standard thing to do with a new befunge interpreter around here is to get it to pass Deewiant's mycology test suite
08:36:11 <rdococ> that would be a new record for me
08:36:13 <zzo38> rdococ: I would not say any one for "most purposes"
08:36:40 <rdococ> where are you likely to see 0/0 in maths where it's not just an error on your part?
08:37:00 <rdococ> then we'll get the value we should assign to 0/0 for those purposes.
08:37:10 <rdococ> it could equal anything, so any value will do.
08:37:59 <zzo38> Whether or not anything should be assigned at all depends on the situation.
08:38:29 <rdococ> Anderson provides a proof that 0^0 = 0/0. usually, 0^0 = 1 for most purposes
08:38:32 <oerjan> rdococ: l'Hopital's rule hth
08:38:44 <zzo38> Mathematically there is no proper answer that you can use
08:39:24 <zzo38> Although I do think 0^0=1 is valid. But if you have a proof that 0^0 = 0/0 that doesn't seem valid because the "0/0" part is the not-valid part.
08:40:03 -!- hppavilion1 has quit (Ping timeout: 246 seconds).
08:40:45 <rdococ> a long while ago I stumbled upon a question, "in a foreign country, 5 / 2 = 3. based on the same proportions, what is 10 / 3?" I immediately thought, "what foreign country would do this and why?"
08:41:15 <rdococ> I wonder if every country did that before the invention/discovery/whatever of zero.
08:41:19 <zzo38> It doesn't make sense.
08:42:52 <rdococ> it does to me... simply put, such a country denies the existence of zero in their mathematical teachings. in modern countries, x / 2 is the number in the middle of 0 and x. but when and where zero doesn't exist, x / 2 would be the number in the middle of 1 and x instead.
08:46:15 <zzo38> I was making different assumptions, in an attempt to be mathematically correct; it has nothing to do with foreign countries or whether or not zero is invented
08:53:48 <rdococ> well, of course 5 / 2 does not equal 3. I wasn't saying it does.
08:54:16 <rdococ> but in a world without 0, or in a world without fractions, division or numbers might take a different meaning
08:54:43 <rdococ> for example, someone might say 5 objects / 2 objects = 3 objects, even if the third object is only half as big as the other two
08:54:56 <rdococ> but we would say 5 / 2 = 2.5
08:54:59 <zzo38> Maybe, I don't know. But, the current mathematical way would still be possible too.
08:55:43 <rdococ> even if there's no concept of the number zero, it has no effect on the applications of division and sharing
08:56:30 -!- AnotherTest has joined.
08:56:36 <rdococ> imagine if we could create an esolang with this alternate division
08:57:45 -!- oerjan has quit (Quit: Later).
09:48:16 -!- FreeFull has quit (Ping timeout: 245 seconds).
09:49:15 -!- FreeFull has joined.
10:05:58 -!- Phantom_Hoover has joined.
10:17:18 -!- boily has joined.
10:47:01 -!- J_A_Work has quit (Quit: J_A_Work).
10:49:32 -!- idris-bot has joined.
11:20:34 -!- FreeFull has quit (Quit: bbs).
11:22:40 -!- ProofTechnique has joined.
11:23:30 -!- Patashu has quit (Ping timeout: 255 seconds).
11:24:25 -!- Patashu has joined.
11:29:46 -!- boily has quit (Quit: PRODLLY CHICKEN).
11:34:43 -!- FreeFull has joined.
11:43:56 <mroman> @tell oerjan http://codepad.org/LA8kRBCM
12:01:26 <Taneb> Does anyone know enough about http://www.eclipse.org/epsilon/ to give me a tl;dr
12:02:16 <int-e> . o O ( I can do the ;dr part just fine... )
12:03:08 <Taneb> Phantom_Hoover, evidently not
12:12:52 <rdococ> epsilon = 1/infinity = 1/(1/0) = 0...
12:14:05 <Taneb> Hmm, the ebook documenting it is written by 4 people from this uni
12:14:12 <rdococ> omg... I have an idea...
12:15:19 <Taneb> `learn epsilon stands for Extensible Platform of Integrated Languages for Model Management (EPILMM)
12:15:22 <HackEgo> Learned 'epsilon': epsilon stands for Extensible Platform of Integrated Languages for Model Management (EPILMM)
12:15:42 -!- ProofTechnique has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
12:16:13 <rdococ> you know numbers, 1, 2, 3...? when you mutiply a number x by some number y and add some number z, you get a linear function with a constant derivative...? imagine a number δ you can multiply by some other number y and add another number z, where you don't get a linear function with a constant derivative...?
12:17:09 <Taneb> I do not think such language exists
12:17:23 <rdococ> but imagine if there was... that'd be... interesting
12:17:35 <rdococ> more interesting than that nullity bull
12:18:33 <rdococ> also... another idea... there are minus numbers and plus numbers, why can't there be multiplicative numbers or divisional numbers? -3, +4, *2, and /6
12:19:04 <Taneb> That sounds interesting...
12:19:11 <Taneb> It would make arithmetic weird
12:19:20 <mroman> What's the system called where an operation is carried out if enough arguments are available
12:19:21 <Taneb> What is 3 times *4 for example?
12:19:47 <mroman> meaning that 5 + 3, + 3 5 and 5 3 + are actually equivalent and are both the same expressions in that system
12:19:53 <Taneb> This makes multiplication non-commutative and non-associative
12:20:00 <Taneb> Maybe addition too
12:20:56 -!- ProofTechnique has joined.
12:21:10 <mroman> (they are semantically identical)
12:21:34 <rdococ> and, multiplicative numbers would act like δ as I said - the second derivative of δx+b isn't constant zero
12:22:31 -!- Patashu has quit (Ping timeout: 276 seconds).
12:22:37 <Taneb> rdococ, would there be exponential numbers?
12:22:50 -!- ais523 has joined.
12:24:56 <rdococ> 2 - ^2 = square root of 2
12:27:29 <Taneb> What about ^3 ^ ^2?
12:29:41 <ais523> is this some sort of logarithm-based arithmetic?
12:29:58 <Taneb> ais523, it is where numbers carry operators
12:30:05 <ais523> and + is concatenation?
12:30:42 -!- TieSoul has quit (Quit: No Ping reply in 180 seconds.).
12:32:09 -!- TieSoul has joined.
12:35:24 <Taneb> Anyway, I need to figure out what Epsilon is by Friday afternoon
12:36:58 <ais523> @message mauris you can omit the parens in a method call in Lua? you seem to do to that inconsistently, I think I'm missing what the parse rules are for it
12:36:58 <lambdabot> Maybe you meant: messages messages-loud messages?
12:37:10 <ais523> @ask mauris you can omit the parens in a method call in Lua? you seem to do to that inconsistently, I think I'm missing what the parse rules are for it
12:38:13 -!- GeekDude has joined.
12:38:20 <b_jonas> ais523: you can omit them if the argument is a string literal or table constructor iirc
12:38:23 -!- J_Arcane has quit (Quit: ChatZilla 0.9.91-rdmsoft [XULRunner 32.0.3/20140923175406]).
12:38:39 <mroman> Does C11 have generics yet?
12:38:48 <ais523> mroman: it has _Generic, so no
12:38:55 <ais523> (_Generic is sort-of the opposite of generics, in a way)
12:39:07 <b_jonas> ais523: http://www.lua.org/manual/5.3/manual.html#3.4.10
12:39:43 <ais523> huh, I've never actually read the Lua spec
12:39:51 <ais523> I guess there'll be some nice bedtime reading for me :-)
12:39:59 <mroman> instead of having node_t with a void* data; pointer
12:40:12 <mroman> you could have node_<type> with a type* data; pointer
12:40:19 <b_jonas> ais523: asking here is also ok
12:40:42 <mroman> I mean I could do stuff like uhm
12:40:52 <ais523> b_jonas: that requires being oline
12:41:08 <mroman> #define list_int list_t
12:41:14 <mroman> to annotate that this list points to integers
12:41:20 -!- mitchs has quit (Ping timeout: 272 seconds).
12:41:35 <rdococ> imagine an infinite-dimensional polynomial number plane, with units like 1, x, x^2, x^3, x^4, x^5...
12:41:38 <b_jonas> ais523: true. whereas the docs are downloadable
12:41:59 <b_jonas> and the implementation too
12:42:15 <mroman> or just name it foo(list_t int_list);
12:42:19 <ais523> I have a standalone Lua impl already but I rarely use it
12:42:27 <ais523> I use it for testing the algorithmic parts of Enigma levels sometimes
12:42:43 <rdococ> a Polynomial datatype, with a number for each exponentiation... 2 + 3x + 4x^2, is stored as the tuple (2, 3, 4)...
12:42:52 <b_jonas> even though I usually have internet connection, I have a lot of things downloaded that I'm not using right now but could be useful
12:43:00 <mroman> I think C++ has some stuff like that
12:43:02 <b_jonas> I think as you're less often online you probably do that more
12:43:28 <b_jonas> but I have used lua to write at least one signature-length obfu
12:43:47 <mroman> template <typename T> struct node { T* data; node* next; };?
12:44:25 <mroman> I'll guess I'll use a C++ compiler then *only* for the template feature :D
12:44:27 <b_jonas> `lua -e a="for b=2,26 do c=0;for d,e in ipairs({a:byte(1,-1)})do c=(b*c+e)%127 end;io.write(string.char(c))end--$S`U$-}OPX41,@aYH\3\26Q2\23*|>"; loadstring(a) {ambrus}
12:44:36 <b_jonas> but something close to that should work
12:44:48 <b_jonas> `lua -e a="for b=2,26 do c=0;for d,e in ipairs({a:byte(1,-1)})do c=(b*c+e)%127end;io.write(string.char(c))end--$S`U$-}OPX41,@aYH\3\26Q2\23*|>"; loadstring(a) {ambrus}
12:44:49 <HackEgo> lua: (command line):1: attempt to call a nil value \ stack traceback: \ (command line):1: in main chunk \ [C]: in ?
12:44:58 <b_jonas> `lua -e a="for b=2,26 do c=0;for d,e in ipairs({a:byte(1,-1)})do c=(b*c+e)%127end;io.write(string.char(c))\nend--$S`U$-}OPX41,@aYH\3\26Q2\23*|>"; loadstring(a) {ambrus}
12:44:58 <HackEgo> lua: (command line):1: attempt to call a nil value \ stack traceback: \ (command line):1: in main chunk \ [C]: in ?
12:45:26 <b_jonas> `lua -e a="for b=2,26 do c=0;for d,e in ipairs({a:byte(1,-1)})do c=(b*c+e)%127\nend;io.write(string.char(c))end--$S`U$-}OPX41,@aYH\3\26Q2\23*|>"; loadstring(a) {ambrus}
12:45:44 <b_jonas> it's easier when it has an actual newline there
12:46:39 <b_jonas> this one sucks, I should rewrite it
12:49:33 -!- ethiraric has quit (Ping timeout: 272 seconds).
12:52:37 -!- mitchs has joined.
12:53:03 -!- J_Arcane has joined.
13:00:19 -!- staffehn has quit (Quit: No Ping reply in 180 seconds.).
13:01:28 -!- staffehn has joined.
13:05:15 -!- ProofTechnique has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
13:11:52 -!- ethiraric has joined.
13:35:14 -!- `^_^v has joined.
14:01:52 -!- ProofTechnique has joined.
14:27:39 -!- Sprocklem has quit (Ping timeout: 240 seconds).
14:55:48 -!- idris-bot has quit (Quit: Terminated).
14:56:08 -!- idris-bot has joined.
15:19:59 -!- Vorpal has quit (Read error: Connection reset by peer).
15:21:49 -!- ProofTechnique has quit (Ping timeout: 252 seconds).
15:22:56 -!- myndzi\ has joined.
15:47:42 -!- SopaXT has joined.
15:48:15 -!- gniourf has quit (Quit: Leaving).
15:49:31 -!- gniourf has joined.
15:56:15 -!- gniourf has quit (Quit: Leaving).
15:57:33 -!- gniourf has joined.
15:57:50 -!- FreeFull has quit (Read error: Connection reset by peer).
15:58:00 -!- FreeFull has joined.
16:04:09 <ais523> b_jonas: is the trailing comma traditional in any language but Perl?
16:04:21 -!- password2 has joined.
16:04:25 <ais523> (actually I don't know why it's there in Perl either, I'm assuming historical reasons; maybe a typo or optimization in the very first JAPH)
16:08:02 <ais523> Taneb: the standard output of a JAPH is "Just another Perl hacker,\n"
16:08:30 <ais523> b_jonas's JALH has a trailing comma despite not being in Perl
16:08:43 <Taneb> Just A Lisp Hacker?
16:08:44 <ais523> (and the goal of a JAPH is to print that string in as bizarre a way as possible)
16:09:27 <Taneb> What would be a bizarre language to write a JA*H for
16:10:08 <ais523> Python, although that's probably been done
16:10:22 <ais523> basically languages which aren't meant to have weird corners
16:10:24 <Taneb> Two languages I do not know very well
16:10:28 <ais523> Prolog would be a good one too, but for a different reason
16:10:45 <Taneb> Saying that I do not know many languages very well
16:11:13 <ais523> oh, not Burlesque, the more obvious ways have to be readable
16:11:28 <ais523> there's not much point of doing a JAMH in Malbolge because it doesn't look visually different from any other Malbolge program
16:12:04 <ais523> (Whitespace is the extreme example of that principle, I guess…)
16:13:36 -!- variable has joined.
16:31:03 <Taneb> I am bored I think that I will install a new OS on my laptop
16:32:12 -!- myndzi\ has quit.
16:32:38 <ais523> wait, people get bored?
16:32:59 <ais523> sometimes I get drained of mental energy and can't do anything interesting, but there's always tons of stuff I want to do and don't have time to
16:33:30 <Taneb> ais523, I forget to write those down
16:33:54 <Taneb> And then forget what they are
16:34:45 <Taneb> Actually, I will work on my natural numbers representation
16:37:15 <Taneb> newtype Nat = Nat [Nat]
16:38:31 -!- `^_^v has quit (Quit: This computer has gone to sleep).
16:40:12 <ais523> Taneb: with the second Nat there being a constructor?
16:40:22 <ais523> that seems more general than the naturals, offhand
16:40:37 <Taneb> ais523, with the restriction that the lists are finite and finitely deep
16:41:15 -!- `^_^v has joined.
16:41:16 -!- mitchs_ has joined.
16:41:33 -!- Melvar` has joined.
16:43:13 -!- mitchs has quit (*.net *.split).
16:43:14 -!- rdococ has quit (*.net *.split).
16:43:14 -!- sebbu has quit (*.net *.split).
16:43:14 -!- Melvar has quit (*.net *.split).
16:43:14 -!- olsner has quit (*.net *.split).
16:47:09 -!- olsner has joined.
16:48:17 -!- ProofTechnique has joined.
16:51:16 -!- `^_^v has quit (Quit: This computer has gone to sleep).
16:55:39 -!- `^_^v has joined.
17:10:31 -!- Melvar` has changed nick to Melvar.
17:10:39 <Phantom_Hoover> Taneb, i still don't get how this representation is meant to work
17:11:32 -!- SopaXT has quit (Remote host closed the connection).
17:14:04 <Taneb> Phantom_Hoover, https://gist.github.com/Taneb/f2021eab65ba59aa3693
17:14:39 <Taneb> Basically, it's a list of the offsets between the set digits, starting from least significant
17:25:16 <FreeFull> I imagine doing JA*H would work well in J
17:26:24 <ais523> now I'm trying to remember which humour article it was that claimed that C was a recursive acronym
17:27:33 -!- Sprocklem has joined.
17:27:46 <shachaf> http://codu.org/logs/_esoteric/2008-02-01.txt 17:22:56: <ais523> like the entry from the IAQ: "C is a recursive acronym. It stands for 'C'."
17:28:18 <ais523> presumably I thought this was amusing even back in 2008
17:28:37 <HackEgo> Thanks, past ais523. Thast ais523.
17:28:46 <shachaf> http://www.seebs.net/faqs/c-iaq.html
17:29:00 <ais523> it's basically a parody FAQ for C
17:30:18 <HackEgo> C is the language of��V�>WIד�.��Segmentation fault
17:33:51 <pikhq> Those are some delightful answers.
17:34:13 <HackEgo> indentity function/indentity function is the function that measures how indented source code is.
17:39:33 <J_Arcane> fn main(){for x in 1..101{match(x%3,x%5){(0,0)=>println!("FizzBuzz"),(0,_)=>println!("Fizz"),(_,0)=>println!("Buzz"),_=>println!("{}",x),}}}
17:40:08 <fowl> Rust is esoteric yes
17:42:22 <pikhq> Now do it in valid C without headers.
17:42:34 <HackEgo> gazpacho/You like Gazpacho and I like Gaspacho. Let's call the whole thing off!
17:43:08 <ais523> pikhq: is there some trick I'm missing to that (C fizzbuzz without headers)? printf seems like the hardest part but that can be declared manually
17:43:23 -!- bb010g has quit (Quit: Connection closed for inactivity).
17:43:38 <pikhq> ais523: Nah, just that most people don't realize it's legal to declare functions without using headers if you can declare them without reference to types in headers.
17:44:07 <pikhq> At least, it seems like some toolchain devs don't realize it. :P
17:44:10 <ais523> you do have to get the declaration right (unless you're using gcc, which will tell you if you've got it wrong and ignore it, IIRC)
17:44:37 <ais523> one thing many people don't realise is that you can declare things like printf /inside/ functions
17:44:41 <pikhq> Yes, but note that int foo(); is a valid declaration.
17:44:56 <ais523> actually, can't you do something like "extern int x, printf(const char *, ...);"?
17:45:26 <pikhq> int printf(); is a full declaration of printf.
17:45:27 -!- gamemanj has joined.
17:46:23 <ais523> pikhq: without the extern?
17:46:52 <ais523> hmm, now I need to try to figure out what extern actually does on functions, if anything
17:47:04 <pikhq> This behavior, BTW, means glibc's strerror_r is nonconforming. :)
17:48:12 <pikhq> (as is mingw's printf, but mingw only resembles C if you squint at it anyways)
17:48:59 <ais523> mingw's printf uses 64 to print long longs :-P
17:49:06 <ais523> *mingw's printf uses %I64 to print long longs :-P
17:49:21 <ais523> sorry, forgot that %I in Konversation is an escape for tab, which it renders as "toggle italics"
17:53:51 -!- quietello has quit (Ping timeout: 256 seconds).
17:53:55 -!- Phantom_Hoover has quit (Ping timeout: 246 seconds).
18:00:06 -!- idris-bot has quit (Quit: Terminated).
18:00:42 -!- idris-bot has joined.
18:08:07 -!- idris-bot has quit (Quit: Terminated).
18:08:33 -!- idris-bot has joined.
18:10:57 -!- Herbalist has joined.
18:10:58 -!- Herbalist has quit (Changing host).
18:10:58 -!- Herbalist has joined.
18:17:10 <HackEgo> pietbot/Pietbot is the only thing that can defeat fungot.
18:17:23 <HackEgo> oerjan_/oerjan_ is oerjan and ørjan's chimæric clone. he shows up on irc when the network is having trouble.
18:21:03 <zzo38> If your program uses SQLite, you can still use SQLite's printf functions to print long long numbers in the way compatible with other computers too
18:21:46 -!- password2 has quit (Ping timeout: 256 seconds).
18:22:22 <zzo38> (Another way is you can use #ifdef to check)
18:23:59 <shachaf> zzo38: One time I was playing Magic: The Gathering, and I equipped Executioner's Hood on Chromanticore, thinking it was a good move.
18:24:50 -!- Sprocklem has quit (Ping timeout: 264 seconds).
18:26:36 -!- Anarchist has joined.
18:27:00 -!- Anarchist has changed nick to Guest26937.
18:27:24 <ais523> shachaf: hey, it's still unblockable by the eldrazi titans!
18:27:41 <ais523> (also it has flying anyway…)
18:28:49 -!- TieSoul has quit (Ping timeout: 252 seconds).
18:29:10 <shachaf> There were no Eldrazi in the game.
18:29:35 <ais523> still not a 100% useless move unless you /knew/ there were no Eldrazi in the game :-P
18:29:57 <ais523> (also, Ugin's Herald was recently printed, which is colourless because it couldn't be an artifact for flavour reasons)
18:30:50 -!- Guest26937 has quit (Ping timeout: 246 seconds).
18:31:00 <shachaf> The format was "Innistrad and onward".
18:31:29 <ais523> Ugin's Herald is in Dragons of Tarkir, which is the most recent major set
18:31:41 <ais523> also, I didn't realise people still played Extended
18:32:45 <ais523> trying to work out if there are any potential colourless nonartifact creatures under those restrictions
18:33:52 <ais523> ah right, I must be remembering the wrong name from somewhere
18:34:35 <J_Arcane> So this seems like a place where people are likely to know; I seem to be confused (or perhaps not and others are) on the distinction between "closure" and "lambda/anonymous function". I've been reading about Rust and Lua a lot lately, and they repeatedly refer to an anonymous function as a "closure," but my understanding was that a closure was rather a specific use of an anonymous function...
18:34:36 <J_Arcane> ...that captures some local state.
18:34:59 <shachaf> A closure isn't necessarily a function at all.
18:35:41 <shachaf> But anyway a closure is a way to implement things that "lexically close" over their environment.
18:35:45 <ais523> J_Arcane: a closure is basically some way of capturing your current local variables in scope
18:35:49 <J_Arcane> Indeed, the terms as explained to me were overlapping but not exclusive: a non-anonymous function can be a closure, and an anonymous function doesn't have to be a closure either.
18:35:49 <ais523> so that you can refer to them later
18:35:57 <shachaf> It can be anonymous or not, and in some cases it doesn't have to be a function at all.
18:36:09 <ais523> they're mostly only useful for nested and anonymous function definitions; I don't think it'd /technically/ have to be for a function but it nearly always is
18:36:44 <ais523> and a "lexical closure" captures lexical scope; I guess you could have a dynamic closure which captures dynamically-scoped variables
18:36:53 <ais523> but that's probably no different to making a copy
18:36:56 <shachaf> In Haskell closures are useful for non-functions.
18:38:38 <zzo38> shachaf: Whether something is a good move usually depends on the situation.
18:39:53 -!- sebbu has joined.
18:40:32 -!- sebbu has quit (Changing host).
18:40:32 -!- sebbu has joined.
18:41:33 <zzo38> For example unblockable by colorless eldrazi cards as mentioned above, or if you just want to unequip it from your other creature
18:42:28 <ais523> most of the cases I can think of where I'd want a creature unequipped - and I wanted to equip it in the first place - it's because something has happened to the equipment to make it hurt the creature it's on
18:42:46 <ais523> in which case, it seems unlikely that I'd have something more valuable than a chromanticore to save
19:01:19 -!- impomatic_ has joined.
19:01:45 <ais523> (OK, so it only costs 5, but it's the most difficult 5)
19:06:27 <shachaf> The context was that I had an artifact deck with all colorless creatures, so Executioner's Hood was great.
19:06:50 <shachaf> And since all my creatures were colorless, I had land colors evenly distributed.
19:07:15 <shachaf> And since I had all sorts of land colors, I put in a Chromanticore.
19:07:52 <shachaf> Under the reasoning that it wasn't colorless but at least it was color-balanced.
19:08:42 <ais523> shachaf: fwiw, those decks which are full of colourless cards normally run colourless lands because they have nice side effects
19:08:45 <ais523> basics are cheaper, though
19:09:20 <ais523> (in Commander, you actually /have/ to run colourless lands in a colourless deck; there are actually enough even for a 100-card deck, but it can be a pain finding them all)
19:10:39 <pikhq> I wish they'd print Cave.
19:11:40 <ais523> pikhq: a basic land that generates {1}?
19:12:01 <Taneb> I need to get more into Magic, but I am refusing to spend money on it
19:12:03 <ais523> pikhq: apparently they've been trying for literally years but it breaks too many things to add another basic land
19:12:04 <pikhq> They were *this* close at one point to printing "Cave. Basic Land - Cave."
19:12:11 <ais523> Taneb: continue not spending money on it
19:12:17 <Taneb> ais523, it is working well so far
19:12:21 <pikhq> ais523: Yeah, it's a rules headache to introduce.
19:12:22 <ais523> you can do what I do, which is to read a lot of articles and not actually play
19:12:22 <Taneb> I have a few friends who lend me decks
19:12:44 <ais523> they've been trying ever since Domain was invented, which is a very long time ago in Magic terms
19:15:19 <shachaf> Taneb: I can give you a bunch of cards when you come visit.
19:15:52 <Taneb> shachaf, I may one day take you up on this offer
19:17:31 <pikhq> Grrawr, impatience.
19:17:51 <zzo38> But the other reason might be if you want to force opponent to block one of your other cards then you would unequip Executioner's Hood, therefore equip to something else instead.
19:17:56 <Taneb> shachaf, if you can pay for my flights, sure
19:19:59 <pikhq> Flights from Europe? to SFO? are probably pretty pricy.
19:20:22 <ais523> I'd guess in the hundreds of pounds range
19:20:31 <ais523> (also you'd likely need to fly back the other way too)
19:20:46 <ais523> hmm, I wonder how flights compare to ferries, price-wise, over that kind of distance
19:20:54 <ais523> that's the only other viable option and it's much slower
19:21:36 <shachaf> You could fly to the east coast and then hitchhike here.
19:22:03 <ais523> shachaf: hmm, I wouldn't actually expect costs to the east coast and to the west coast to be all that different; I'd expect popular destination / unpopular destination to be more of a factor
19:22:16 <ais523> the US is very wide, and so there'll be fuel and staffing costs going from one side to the other
19:22:23 <shachaf> As I remember it it's somewhat cheaper.
19:22:36 <ais523> but the Atlantic is also wide, and overheads in flying internationally are going to be relevant
19:22:44 <ais523> plus, of course, the standard cost of flying generally
19:22:55 <ais523> anyway, you can get cheaper flights if you're willing to be advertised at constantly en route
19:23:07 <shachaf> Usually transatlantic flights to here consist of two legs.
19:23:15 -!- AnotherTest has quit (Ping timeout: 245 seconds).
19:23:23 <b_jonas> ais523: yes, also depends on when you buy the tickets and stuff like that
19:23:30 <pikhq> SFO-London is a well-established direct flight though.
19:23:31 <shachaf> But maybe that's less true for .uk than for .il.
19:23:37 <b_jonas> like, whether they guess it's a popular time or not
19:23:57 <ais523> shachaf: I would be very surprised if there wasn't a direct flight from Heathrow
19:24:03 -!- w00tles has joined.
19:24:09 <ais523> it's got to be one of the most connected airports in the world
19:24:32 <pikhq> ais523: There is, it's one of the top 10 most heavily flown flights between Europe and North America I believe.
19:24:39 <ais523> (this reminds me of Birmingham New Street, actually, the most connected train station in the UK; London doesn't have a single major train station, just lots of small ones)
19:25:17 <ais523> strangely, not all traffic to central Birmingham goes through New Street; some uses Moor Street instead, which is maybe about 200m away from it but on different physical train lines
19:25:24 <ais523> probably just due to not being able to fit all the lines into one station
19:25:30 -!- w00tles has quit (Remote host closed the connection).
19:25:34 <pikhq> I don't think Heathrow flies direct to that *many* US airports, but SFO is a big damned airport.
19:25:40 <shachaf> Sounds like a conversation for #trains.
19:25:45 <Taneb> OK, 800 there and back
19:25:52 -!- w00tles has joined.
19:26:02 <Taneb> In late September, on British Airways
19:26:08 -!- AnotherTest has joined.
19:26:10 <shachaf> You want me to pay for a flight back, too?
19:26:26 <Taneb> Well, BA there, AA back, there direct, back via Dallas
19:26:34 <Taneb> shachaf, I did say my flights
19:26:35 <pikhq> Well, customs gets pissy when you don't have a visa and don't have a return flight.
19:26:45 -!- w00tles has quit (Remote host closed the connection).
19:26:51 <b_jonas> ais523: yes, but mind you, some of that stuff was probably built when the city was smaller and is very hard to change now
19:27:08 -!- w00tles has joined.
19:27:13 <b_jonas> like, in Budapest, trains can enter from three directions, somewhat corresponding to the three main railway station endpoints but not completely,
19:27:21 <ais523> b_jonas: actually, what happened with Birmingham Moor Street was, they had a reasonably large station there
19:27:39 <ais523> then decided they needed a through line, which they built a bit too far from the station, and so they made a new Moor Street station and closed the old one
19:27:49 <ais523> and then it grew enough that nowadays they use both stations
19:27:54 <ais523> and made them into one continuous building
19:28:35 <b_jonas> and this results in two big cuts in Pest cutting it to three pieces, where there's only train tracks in the cuts, and other vehicles can't cross at all, because no road can cross the trains, and even pedestrians can't go through
19:29:00 <b_jonas> so if you look at the car map of Budapest, there's basically the Danube which you can cross only on bridges, plus a smaller cut from the two mountains, plus these two big cuts
19:29:11 <b_jonas> and it's even worse by bus
19:29:45 <b_jonas> so for any place in Pest that's not in the city, the route there is obvious:
19:30:20 <b_jonas> if it's near Újpest, to the north of Rákosrendező, you have to take metro 3 to the north then change to a bus;
19:30:37 <b_jonas> if it's in the middle between Rákosrendező and Keleti, take metro 2 and change to a bus;
19:30:50 <b_jonas> if it's on the south, take metro 3 to the south and then take a bus.
19:31:27 <b_jonas> all three metros don't reach far enough, so a large part of the outer town is covered by buses only from the terminus of the metro or two or three other metro stations close to the terminus
19:31:37 <ais523> b_jonas: is the metro a sort of local train line? and does it run underground?
19:31:49 <ais523> (the word "metro" is used for a huge number of different things in English)
19:31:53 <b_jonas> yes, the metro is mostly undreground,
19:32:04 <b_jonas> it does come up to the surface near some of the ends
19:32:36 <b_jonas> there's now three of them, metro 3 is the biggest and covers north and south, and is underground everywhere except the very end at south
19:33:29 <b_jonas> metro 2 covers east and comes up to above ground near the end for the last two stations,
19:33:54 <ais523> London has way more than 3 underground lines
19:34:19 <b_jonas> metro 2 also goes a bit to north-Buda, which is traditionally considered the rich part of Budapest, so most people have cars, but for those who don't, the tram lines are getting somewhat improved in the future hopefully,
19:34:22 <ais523> whereas Birmingham doesn't have an underground system, but has an above-ground train system that links to pretty much all areas of the city, also one tram
19:34:31 <b_jonas> and metro 4 is new and covers south-Buda
19:35:00 <Taneb> Newcastle has a weird undergroundy overgroundy thing with poor coverage of the west of the city
19:35:03 <ais523> (some of the lines go underground very near the centre, but they're all /mostly/ above-ground)
19:35:06 <Taneb> York has nothing of the sort
19:35:27 <b_jonas> technically there's also metro line 1 which is different, it covers only the city, and isn't deep underground, but immediately below roads
19:35:54 <ais523> London's underground lines get deeper and deeper the newer they are
19:36:07 <ais523> the really old ones, they just cut the ground open, built the line, then built a roof over it
19:36:22 <ais523> the newest ones they're boring through rock
19:36:35 <b_jonas> there are also HÉVs, which are trains that _should_ ideally be continuation of the metros that cover the farther parts of the town and the agglomeration,
19:36:40 <ais523> and sometimes it takes longer to walk down to the train line, then back up again at the destination, than it takes the train to get from station to statin
19:36:47 <b_jonas> and ideally should be just the continuation of the metros,
19:36:58 <b_jonas> but there's two problems with that idea,
19:37:21 <b_jonas> one is that some of them are placed inconveniently and don't match the metros, but only trams,
19:37:46 <b_jonas> and the other is that for all kinds of stupid policy reasons they just can't be converted to get connected with the metro even where it could work.
19:40:20 <b_jonas> and they all go above ground, but they're not metro because, and they're also not train which anyone can tell from how you need different tickets, there's no toilets on the carriages, the rules for bicycles is different, and the tracks have designated directions, namely right hand side on most lines, but left hand side on the eastbound lines.
19:40:43 <b_jonas> so basically, there's metros and HÉV and trams and they're unconnectible and don't ever form a sane system.
19:43:19 <b_jonas> the trams are actually different
19:43:39 <b_jonas> trams work like road traffic, they often have to wait for traffic lights together with cars, or even wait for cars to pass,
19:44:29 <b_jonas> whereas road traffic always has to yield to trains and HÉV at level crossings because those can't stop.
19:44:46 <b_jonas> but the metro and HÉV situation is just stupid and traditional
19:46:19 -!- w00tles has quit (Remote host closed the connection).
19:46:51 -!- w00tles has joined.
19:54:42 -!- quietello has joined.
19:59:33 -!- nys has joined.
19:59:44 <FireFly> In Stockholm we have two relatively shallow lines, one deeper line behind those two, and then they're building a commuter rail tunnel that's going to go below /that/ line
20:00:51 <shachaf> I thought you were in .fi.
20:00:57 <shachaf> I might be confusing you with someone else.
20:01:22 <shachaf> I was in Stockholm once. Twice.
20:01:54 <shachaf> I was 3 years old the first time.
20:02:05 <FireFly> I think it's a pretty boring city, but I guess that's because I'm just too used to it
20:02:42 <shachaf> The second time we went to one or more museums.
20:02:55 <shachaf> I must've been 15 years old.
20:03:01 <shachaf> Museums aren't usually my thing.
20:03:08 <FireFly> Perhaps you saw our excellent example of engineering; a ship that broke after a couple km
20:03:35 <FireFly> It's probably the most popular museum here
20:03:54 <shachaf> I also met some friends and/or relations.
20:05:34 <olsner> I was also surprised about FireFly being in stockholm, but I think I was confusing you with Vorpal
20:06:21 <FireFly> I remember Vorpal being in the middle of the forest somewhere
20:06:29 <FireFly> Maybe in Värmland or something? I don't know
20:07:02 <olsner> I think he told me some time, but I don't remember where
20:13:19 -!- FreeFull has quit (Quit: Bbs).
20:18:36 -!- Wallacoloo has joined.
20:24:42 -!- TieSoul has joined.
20:30:22 -!- TieSoul has quit (Ping timeout: 252 seconds).
20:35:48 -!- TieSoul has joined.
20:42:12 -!- mitchs_ has quit (Read error: Connection reset by peer).
20:42:21 -!- mitchs_ has joined.
20:47:06 -!- shachaf has quit (Remote host closed the connection).
20:47:14 -!- shachaf has joined.
20:51:51 -!- w00tles has quit (Quit: quit).
21:03:42 -!- Patashu has joined.
21:03:43 -!- `^_^v has quit (Quit: This computer has gone to sleep).
21:04:26 <ais523> huh, xkcd was in an esolangy mood recently: http://xkcd.com/1537/
21:04:50 <ais523> I thought those examples might be from an actual language for a while, but clearly they aren't
21:04:56 <ais523> thus, must definitely be in esolang territory
21:10:46 <nys> i think it might be a take on that javascript wat thing
21:10:50 -!- Herbalist has quit (Read error: Connection reset by peer).
21:11:17 <ais523> many languages do that sort of thing, but not so extreme
21:11:23 -!- ProofTechnique has quit (Quit: My Mac has gone to sleep. ZZZzzz…).
21:11:50 -!- Herbalist has joined.
21:13:12 <Taneb> ais523, we had a discussion trying to figure it out
21:13:42 <ais523> I don't think there /is/ a pattern, and I think that's the point
21:14:03 <HackEgo> recursion/You might expect a reference to recursion here, but to make it interesting you'll actuallSTACK OVERFLOW
21:15:07 <Taneb> ais523, [1], [2], [4], are just weird casting, [13] is just an unusual floor function
21:15:19 <Taneb> [5] I think is oddly greedy quote marks
21:15:34 <HackEgo> lie/Lies are even easier than monoids. They form groups, known as Lie groups.
21:15:44 <Taneb> [10] is working out the line number + 2
21:15:51 -!- variable has changed nick to constant.
21:15:55 -!- TieSoul has quit (Remote host closed the connection).
21:16:11 <Taneb> [11] add 2 to the number 2, so whenever there is a 2 in the future there is a 4 instead (see [12]^W[14])
21:16:11 <ais523> I can't explain [2] with weird cast rules though
21:16:22 <ais523> Taneb: oh, 11 is Forte-style? I missed that
21:16:58 <Taneb> ais523, I think it's casting "2" to an int, then to a list of ints, then appending the two lists, then turning the whole thing into a string
21:17:16 <ais523> Taneb: exactly, there's no reason to turn the whole thing into a string at the end
21:17:24 <ais523> other than that a string was involved to start with
21:17:30 <Taneb> I did specify weird casting rules, right?
21:17:45 <ais523> or, maybe this is an everything-is-a-string language like Perl is when it doesn't leak implementation details?
21:18:12 <Taneb> I don't really know what [6] through [8] are doing, though
21:18:21 <ais523> IMO, philosophically, in Perl, if you write (say) 10 that's really just sugar for the string "10", and the fact that it's stored internally as an int is an implementation detail
21:18:43 <Taneb> Well, [8] is making fun of IEEE floating point
21:18:46 <ais523> Taneb: in [8], the idea is that the denominator is approximately 0
21:18:54 <ais523> thus, dividing 2 by it gives you a NaN with a rounding error
21:19:08 <ais523> (although that specific example doesn't work because it's using powers of 2)
21:20:45 -!- AnotherTest has quit (Ping timeout: 246 seconds).
21:23:33 -!- AnotherTest has joined.
21:40:55 -!- Patashu has quit (Ping timeout: 272 seconds).
21:52:00 -!- gamemanj has quit (Ping timeout: 256 seconds).
22:02:29 -!- AnotherTest has quit (Ping timeout: 276 seconds).
22:08:18 -!- ProofTechnique has joined.
22:10:50 -!- Sprocklem has joined.
22:22:39 -!- FreeFull has joined.
22:25:08 -!- Wright has joined.
22:30:55 <HackEgo> [wiki] [[Special:Log/newusers]] create * Hppavilion1 * New user account
22:32:02 -!- GeekDude has quit (Quit: {{{}}{{{}}{{}}}{{}}} (www.adiirc.com)).
22:33:11 -!- aretecode has joined.
22:35:27 -!- polytone has quit (Quit: "I can hear myself... I think I'm a bit afraid.").
22:36:16 <HackEgo> [wiki] [[Damarok]] N http://esolangs.org/w/index.php?oldid=43214 * Hppavilion1 * (+243) Created Article (I'm so sorry)
22:36:46 -!- hppavilion1 has joined.
22:48:00 <HackEgo> off by two/An off by two error is what happens when you expect an off by one error but compensate in the wrong direction
22:48:16 <shachaf> `culprits wisdom/off by two
22:49:16 <HackEgo> mroman_/mroman_ is probably mroman but you can never be sure. (NSFW)
22:49:22 -!- monotone has joined.
22:50:07 <HackEgo> badger badger badger badger badger badger badger badger mushroom mushroom
22:51:39 <HackEgo> [wiki] [[Esolang talk:Categorization]] http://esolangs.org/w/index.php?diff=43215&oldid=42034 * Hppavilion1 * (+403) /* Category: Libraries */ new section
22:51:47 <pikhq> It feels a bit like magic sometimes, cooking.
22:52:20 <pikhq> I mean, somehow some flour, salt, baking powder, butter, milk, and sausage turned into biscuits and gravy.
22:58:18 <hppavilion1> They better have programmed that in modified BrainFuck or something
22:59:00 <pikhq> HackEgo is weird, but in a way different from what you expect.
22:59:21 <pikhq> Each command is a Linux program. It spawns a Linux kernel, runs that program, and outputs the result.
22:59:38 <pikhq> Now, if you want an esolang bot, you need look no further than fungot.
22:59:38 <fungot> pikhq: depends on the order i described it roughly earlier. have a happy fnord i am
22:59:41 <pikhq> fungot is in Befunge.
22:59:45 -!- Wright has quit (Ping timeout: 264 seconds).
23:00:51 -!- Herbalist has quit (Ping timeout: 246 seconds).
23:00:57 <pikhq> It's a pretty neat language.
23:01:31 <ais523> hppavilion1: -93 or -98?
23:01:32 <hppavilion1> I promised anyone reading my commits I'd get it working today...
23:01:36 <ais523> anyway, if you haven't seen it yet
23:01:40 <ais523> get hold of a copy of Mycology
23:01:46 <ais523> it'll catch a ton of common mistakes
23:01:56 <ais523> why do we not have actually useful information in the learndb? :-(
23:02:45 <ais523> `learn mycology is a Befunge-98 (also -93 to some extent) testsuite that can be found at https://deewiant.iki.fi/projects/mycology/
23:02:47 <HackEgo> Learned 'mycology': mycology is a Befunge-98 (also -93 to some extent) testsuite that can be found at https://deewiant.iki.fi/projects/mycology/
23:02:55 <pikhq> Befunge-98 is kinda hard to implement.
23:02:56 <ais523> hppavilion1: definitely read that, it'll probably tell you what your interp is doing wrong
23:03:09 <hppavilion1> And stumled across an idea for a language called Darmok
23:03:13 <pikhq> Befunge-93 is not very hard, but it's also not that interesting to have an implementation for. :)
23:03:34 <hppavilion1> So I created a page for it. I apologized in the notes
23:03:44 <ais523> Befunge-98 is a great language to impl, though
23:03:51 <ais523> there are several Mycology-passing interps now I think
23:03:56 <hppavilion1> Darmok, if you aren't aware, is an episode of ST:TNG. Apparently
23:04:37 <hppavilion1> In Darmok, they encouter a species that communicates completely through metaphors and references to local mythology
23:04:54 <pikhq> True. It's hard to implement, but hard to implement in interesting ways.
23:05:05 <pikhq> And a Befunge-98 implementation can actually do neat things.
23:05:55 <hppavilion1> "Terminate Sibling" for example, would be CAIN AND ABEL IN THE GARDEN (if "terminate sibling" is even a real command)
23:06:24 <ais523> it's actually an operation that can be done on threads that makes sense, and might potentially be slightly useful on occasion
23:06:25 <pikhq> What, though, does DARMOK AND JILAD AT TENAGRA mean?
23:06:45 <ais523> e.g. if you're doing some sort of threaded exhaustive search and only need one result
23:06:53 <pikhq> Or, indeed, GILGAMESH AND ENKIDU AT URUK?
23:07:05 <hppavilion1> I think that Damarok the language would also only reference real mythology
23:07:16 <hppavilion1> I don't remember how the epic of gilgamesh goes
23:07:31 * pikhq was just referencing the episode; don't mind me
23:07:35 <ais523> the problem isn't so much finding something for your commands to do, though, but finding some way to express the commands you need
23:07:48 <ais523> or, well, you can just add a ton of commands and hope they add up to something TC
23:07:53 <ais523> (see: initial Snowflake)
23:07:56 <hppavilion1> The problem would be that there would ALWAYS be libraries needed
23:08:04 <hppavilion1> And everyone would implement them differently
23:08:06 <ais523> actually that was awkward because I needed a pretty exact amount of computational power
23:08:41 <hppavilion1> So they'd need to implement the entirety of some libraries on their own
23:09:13 <ais523> hppavilion1: that's not so dissimilar from INTERCAL, we solved the problem by encouraging people to submit them to a centralised repo
23:09:14 <hppavilion1> I think to terminate a sibling you would type "CAIN AND ABEL IN THE GARDEN, WHERE ABEL IS <sibling>"
23:09:24 <ais523> so they could use each other's libraries
23:09:41 <ais523> (even more a problem in INTERCAL because all line numbers are global, so you need to avoid clashes between different libaries)
23:10:24 <hppavilion1> So I think that every statement would have two clauses (well, one main clause and an arguments clause), separated by the comma
23:10:48 <hppavilion1> The command clause is the reference to mythology
23:11:21 <hppavilion1> And the WHERE clause is substitutions (which are needed for formality)
23:11:44 <hppavilion1> So "CAIN AND ABEL IN THE GARDEN, WHERE ABEL IS <sibling>" means to kill a process of name <sibling>
23:12:13 <hppavilion1> "CAIN AND ABEL IN THE GARDEN" alone would... I don't know... Kill the next sibling sequentially?
23:12:21 <ais523> <Deewiant> mycology.b98 tests every fingerprint that I am aware of, apart from FNGR, SGNL, and WIND. ← it doesn't test IFFI :-(
23:13:17 <hppavilion1> Of course that doesn't account for complex statements...
23:13:23 <ais523> I think IFFI postdates Mycology, though
23:14:16 <hppavilion1> Name a famous decision from popular folklore/mythology
23:14:29 -!- Wallacoloo has quit (Ping timeout: 255 seconds).
23:15:29 <ais523> crossing the Rubicon? although that's more historical rather than mythological
23:17:27 <ais523> Julius Caesar taking his part of the Roman Army into Italy (which was totally illegal), thus sparking a civil war
23:17:34 <hppavilion1> CEASER CROSSING THE RUBICON, WHERE CONDITION IS <condition> AND CROSSING IS:\n\t<stuff to do>
23:17:55 <ais523> hppavilion1: note that you don't really want to just design an imperative language with the keywords swapped out
23:18:07 <ais523> because it isn't really interesting
23:18:33 <hppavilion1> But I'm currently just creating some basic ideas
23:19:03 -!- Phantom_Hoover has joined.
23:20:10 <ais523> huh, there's an interesting debate about what exactly Caesar said when crossing the river
23:20:25 <ais523> although it's widely agreed how it's translated into English, there's some debate about the word order of the original and whether it was in Latin or Greek
23:24:15 -!- hppavilion1_ has joined.
23:24:50 -!- hppavilion1 has quit (Ping timeout: 246 seconds).
23:27:04 <hppavilion1_> What can we do to make it less keyword-swapout-y
23:28:07 -!- Wright has joined.
23:28:15 -!- Herbalist has joined.
23:28:40 <hppavilion1_> Ideally it'd be able to take a set of metaphors and interpret what they mean, but that'd require a prize-winning AI that probably would lag normal computers to hell
23:30:09 <shachaf> `wisdom doesn't work that way hth
23:30:10 <HackEgo> shiasdayviaerqjjjjjjjj/shiasdayviaerqjjjjjjjj is the reason why the USA don't use the metric system.
23:30:19 <HackEgo> relevant info/The large-eyed mouse lemur is a nocturnal tree-dweller.
23:30:19 <shachaf> Should `wisdom be merged with `? ?
23:30:37 <HackEgo> fiora/Fiora is half JRPG fangirl, half SIMD dork, and all sucrose. She's a sous-chef who shushes sushi.
23:31:04 <FireFly> I'm pretty sure Fiora isn't shachaf's mum
23:31:10 <ais523> coppro: you probably meant `?
23:31:14 <HackEgo> twh/twh would help, but is an hth derivative. hth. twh. hand.
23:31:23 <FireFly> No, it doesn't take any parameter
23:31:27 <ais523> hppavilion1_: hint: `wisdom ignores any arguments
23:31:35 <HackEgo> wisdom is always factually accurate, except for this entry, and uh that other one? it started with like, an ø?
23:31:38 <HackEgo> INTERCAL has excellent features for modular program for the enterprise market.
23:31:47 <pikhq> Huh, it's been a while since Fiora's been here, hasn't it?
23:31:57 <HackEgo> Along with C, C++ is a language for smart people.
23:32:01 <FireFly> pikhq: yeah, I think she left around the time when kmc left
23:32:13 <HackEgo> zork/Zork is like York, except for the first letter.
23:32:15 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
23:32:19 <pikhq> kmc left too? Nooooo
23:32:29 <pikhq> What's next, no ehird?
23:32:30 <coppro> `learn java is a programming-language shaped collection of misfeatures
23:32:35 <HackEgo> Learned 'java': java is a programming-language shaped collection of misfeatures
23:32:40 <HackEgo> /hackenv/bin/learn: line 3: wisdom/: Is a directory \ Learned '':
23:33:03 <ais523> even I've been known to leave this channel for long periods of time, because it wasn't ontopic enough
23:33:06 <ais523> although it's been fine recently
23:33:17 <HackEgo> Learned 'yo': yo mama so fat, insert joke here
23:33:35 <HackEgo> rm: cannot remove `/home/hackbot/hackbot.hg/multibot_cmds/env/.hg/store/data/canary.orig': Is a directory \ Done.
23:33:35 <ais523> shachaf: I was about to `revert but I think we might have reverted each other
23:33:54 <ais523> we /still/ have a broken canary in the system somewhere?
23:33:56 <HackEgo> java is a programming-language shaped collection of misfeatures
23:34:05 <FireFly> Yeah, it seems it messed up the history somehow
23:34:12 <FireFly> or rather, it's messed up because of the history, somehow
23:34:38 <hppavilion1_> `learn haskell is the most beautiful language ever invented
23:35:02 * FireFly wonders if hppavilion1_'s client chose a nick based on the computer's name, or if it's an actual nickname
23:35:24 <hppavilion1_> I chose the nick when I was 7 and didn't want to forget my username
23:35:33 <HackEgo> Unbound implicit parameter (?haskell::Wisdom) \ arising from a use of implicit parameter `?haskell'
23:35:35 <hppavilion1_> It's usually hppavilion1, but my browser timed out
23:35:50 -!- hppavilion1_ has changed nick to hppavilion1.
23:36:03 <FireFly> It annoys me that my brain lexes it as hpp|avilion|1
23:37:22 <HackEgo> Runs arbitrary code in GNU/Linux. Type "`<command>", or "`run <command>" for full shell commands. "`fetch <URL>" downloads files. Files saved to $PWD are persistent, and $PWD/bin is in $PATH. $PWD is a mercurial repository, "`revert <rev>" can be used to revert to a revision. See http://codu.org/projects/hackbot/fshg/
23:37:34 <shachaf> `` hg log --removed wisdom/yo
23:37:36 <HackEgo> changeset: 5605:a4afa9eaa9ff \ tag: tip \ user: HackBot \ date: Wed Jun 17 23:35:10 2015 +0000 \ summary: <hppavilion1_> learn yo mama so fat, insert joke here
23:37:42 <HackEgo> cat: wisdom/yo: No such file or directory
23:37:46 <HackEgo> yo mama so fat, insert joke here
23:38:17 <shachaf> It seems that revert isn't actually working.
23:38:29 <shachaf> Or it isn't modifying the hg repository, even if it's modifying the working directory?
23:38:46 <shachaf> fizzie: what's going on twh
23:39:49 <hppavilion1> It uses references to mythology and metaphor
23:40:08 <hppavilion1> I have no clue how to implement it short of brute force regexes
23:40:28 <FireFly> That sounds like an inefficient implementation
23:43:14 <hppavilion1> And I suppose it would only sort #included libs
23:44:39 <shachaf> FireFly: also you should set up /hilight on ireFly\b
23:45:03 <hppavilion1> It's more of a thought experiment language and something to use for mindbendingness, not for implementing stuff
23:45:20 <FireFly> shachaf: I probably should, just for you. What channel did I miss it in now?
23:46:16 <hppavilion1> I just use 1 because it's soooo much easier
23:50:52 <tswett> I'm not sure that my newer SLOBOL language can be run faster than double exponential time.
23:51:40 * hppavilion1 DDGs double exponential time, though he's pretty sure he knows what it means
23:52:52 <shachaf> double exponential time is exponential time approximated with double-precision floating point arithmetic
23:53:08 <HackEgo> macabre/The Macabres have been the hereditary rulers of Lochaber for 3 centuries.
23:53:31 -!- Wright has quit (Ping timeout: 244 seconds).
23:54:33 <HackEgo> /home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: walrus: not found
23:54:45 <HackEgo> \ Usage: \ kill [options] <pid> [...] \ \ Options: \ <pid> [...] send signal to every <pid> listed \ -<signal>, -s, --signal <signal> \ specify the <signal> to be sent \ -l, --list=[<signal>] list all signal names, or convert one to a name \ -L, --table list all signal names in a nice table \
23:54:51 <tswett> If a SLOBOL program contains n points, then the easiest upper bound on the number of ways of executing the program is (3^n)!.
23:55:29 <hppavilion1> I must now open wolfram alpha and see that graphed
23:56:40 <tswett> For 0, it's 1; for 1, it's 6; for 2, it's 362880; for 3, it's 10888869450418352160768000000; for 4, it's 5797126020747367985879734231578109105412357244731625958745865049716390179693892056256184534249745940480000000000000000000; and so on.
23:57:19 <tswett> This is just for a naive implementation, though.
23:57:28 <tswett> I haven't actually implemented it.
23:57:47 <tswett> I'm not planning to implement it, either.
23:58:34 <shachaf> higgledy piggledy / hp pavilion / doesn't like jokes that are / written in text; // uncontroversially, / one in a million is / roughly the chance they won't / leave them perplexed
23:58:55 <shachaf> hm, in retrospect that could be seen as slightly rude tdnh