←2018-04-26 2018-04-27 2018-04-28→ ↑2018 ↑all
00:21:03 -!- doesthiswork has joined.
00:28:27 -!- brandonson has quit (Ping timeout: 240 seconds).
00:30:36 -!- brandonson has joined.
00:32:46 <doesthiswork> So I came up with a "softmedian" function to match the "softmax" and "softmin" functions
00:34:37 <shachaf> What is it?
00:38:09 <doesthiswork> for all elements x in the collection sum (e^(x-a) -1)/(e^(x-a)+1) with "a" such that the sum equal zero
00:38:41 <shachaf> What is softmin?
00:39:12 <shachaf> Oh, I was mixing up softmax and another thing.
00:39:35 <doesthiswork> I need to solve for a to make this actually useful
00:40:35 <doesthiswork> but for collections of only two elements it simplifies down to the mean of the two elements
00:51:46 -!- friendlyGoat has joined.
00:52:12 -!- friendlyGoat has quit (Client Quit).
01:55:40 <zzo38> What is the maximum length of the real name field in IRC?
02:00:15 <imode> the RFC doesn't seem to have anything about specific maximums.
02:00:19 <imode> so I think it's up to the server.
02:01:27 <alercah> and the maximum length of 510 for a single message
02:24:25 -!- MDead has joined.
02:25:47 -!- MDude has quit (Ping timeout: 256 seconds).
02:25:51 -!- MDead has changed nick to MDude.
03:43:55 -!- newsham_ has changed nick to newsham.
03:45:34 -!- moony has changed nick to x86-64.
04:04:42 <shachaf> newsham: hewsham
04:04:54 <shachaf> `5 w
04:04:57 <HackEso> 1/2:ppntat//Pen Pineapple Nutmeg Tamarind Apple Tangerine \ mockingbird//mockingbird is watching you.. closely! Is it mocking you? Probably. \ minsky//to Minsky on : /mɪnskiː/ To act as a Minsky machine on; of a program or programming language, to encode its entire state into the object as a single integer. \ cricket//Cricket is one team is in and one team is out, and the team who is out tries to get the team who is in to be out, and then the tea
04:05:03 <shachaf> `n
04:05:04 <HackEso> 2/2:m who was previously out can be in. Whoever earns more points wins, unless you run out of time, in which case nobody wins. \ dosh//The doshes are what the gostak distims.
04:17:41 <x86-64> o/ shachaf
05:04:14 <esowiki> [[User:Nobody]] https://esolangs.org/w/index.php?diff=54921&oldid=54847 * Nobody * (-43)
05:06:57 <esowiki> [[Brainfuck--]] M https://esolangs.org/w/index.php?diff=54922&oldid=37901 * Nobody * (+0) typo fixed
05:26:42 -!- SopaXorzTaker has joined.
06:01:54 -!- sprocklem has quit (Quit: [).
06:02:19 -!- variable has quit (Quit: /dev/null is full).
06:10:03 -!- sleffy has quit (Ping timeout: 268 seconds).
06:16:43 -!- variable has joined.
06:23:16 -!- sleffy has joined.
06:28:27 -!- brandonson has quit (Ping timeout: 240 seconds).
06:30:33 -!- brandonson has joined.
06:48:15 -!- sprocklem has joined.
06:51:22 -!- SopaXorzTaker has quit (Ping timeout: 268 seconds).
06:57:42 -!- SopaXorzTaker has joined.
07:07:16 -!- variable has quit (Quit: /dev/null is full).
07:10:39 -!- variable has joined.
07:19:29 -!- sleffy has quit (Ping timeout: 248 seconds).
07:21:05 -!- AnotherTest has joined.
07:22:02 <esowiki> [[Special:Log/newusers]] create * DevHammed * New user account
07:25:58 <esowiki> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=54923&oldid=54920 * DevHammed * (+81) /* Introductions */
07:32:13 <esowiki> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=54924&oldid=54923 * DevHammed * (+7) /* Introductions */
07:48:57 <shachaf> Taneb: So if I write a' = a, can indices of a be used for indexing a'?
07:49:27 <shachaf> Can I store indices of a somewhere where a isn't in scope, and use them later?
07:52:52 <shachaf> Is there such a thing as a type of values that can only be used as arguments for a specific function?
08:11:31 -!- AnotherTest_ has joined.
08:12:33 -!- variable has quit (Quit: /dev/null is full).
08:12:37 -!- AnotherTest has quit (Ping timeout: 260 seconds).
08:12:37 -!- AnotherTest_ has changed nick to AnotherTest.
08:34:09 -!- AnotherTest has quit (Ping timeout: 248 seconds).
08:38:14 -!- doesthiswork has quit (Quit: Leaving.).
08:47:28 <Taneb> shachaf: fdepends on the semantics of the language. They're different types, in something like C that could work but in Haskell you might need fromIntegral
08:47:59 <shachaf> What could work?
08:48:02 <Taneb> I don't see why you wouldn't be able to store indices somewhere where the array isn't available
08:48:49 <Taneb> I don't think a type of values that can only be used as arguments for a specific function makes sense to me, as it prevents composition
08:49:09 <Taneb> I'm answering your questions, starting from the one at 8:48 BST, in order
08:49:47 <shachaf> Well, if a' is equal to a, I imagine an index of a should be usable for a'?
08:49:58 <shachaf> Depending on whether you have a notion of object identity and what equality means.
08:50:25 <Taneb> Oh! I misunderstood!
08:50:26 <shachaf> Taneb: But an array is just a function from indices to addresses.
08:51:43 <Taneb> Yes, I'd imagine an index of a would be usable as an index of a' if a' is the same array
08:51:50 <Taneb> Possibly even if a' is a slice of a?
08:52:26 -!- SopaXorzTaker has quit (Remote host closed the connection).
08:58:00 <shachaf> Taneb: This is sounding like a complicated feature.
08:58:18 <shachaf> Wouldn't you get a lot of the benefit from index types tied to array types but not to specific arrays?
09:02:37 -!- xkapastel has quit (Quit: Connection closed for inactivity).
09:12:49 <Taneb> That wouldn't have prevented the error that inspired this
09:12:54 <Taneb> I'll do some thinking over the weekend
09:12:58 <shachaf> Why not?
09:13:44 <Taneb> Because I was indexing into an array of indexes
09:19:17 <shachaf> Right. So if your array of indices and array of values had different types, it could've helped.
09:21:26 -!- SopaXorzTaker has joined.
10:06:57 -!- imode has quit (Ping timeout: 240 seconds).
10:10:27 -!- brandonson has quit (Ping timeout: 240 seconds).
10:12:14 -!- brandonson has joined.
10:33:06 <esowiki> [[Brainfuck extensions]] https://esolangs.org/w/index.php?diff=54925&oldid=52755 * Chronos * (+19)
10:38:10 <esowiki> [[Brainfuck extensions]] https://esolangs.org/w/index.php?diff=54926&oldid=54925 * Chronos * (+39) added Grawlix
10:50:24 -!- ais523 has joined.
10:50:45 <ais523> is there a standard formalization for computable partial functions?
10:50:52 <ais523> e.g. based on a Turing machine with some specific I/O mechanism?
10:51:20 <ais523> we have Turing machines as a formalism for halt/no-halt programs, and brainfuck as a formalism for interactive I/O
10:51:27 <ais523> but the case of non-interactive I/O is a bit less clear to me
10:53:03 <int-e> there are partial recursive functions...
10:54:08 <ais523> hmm, the µ-recursive functions seem to be a standard definition of what I'm looking for
10:54:20 <int-e> or you could use Brainfuck without , with some fixed convention for reading the input to tape. (say, >[,>]<[<]> ... i.e., the input starts from the second tape position; the first byte is kept at 0)
10:54:28 <ais523> (it's basically primitive recursive + a minimum operator)
10:54:43 <ais523> right, I'm hoping for a mathematical definition really
10:54:59 <int-e> then I'd use that one
10:55:04 <ais523> I'm working on a language that aims to be a "standard language for noninteractive I/O" which you implement to prove your language can do it
10:55:19 <ais523> so I'd like to prove my language equivalent to the mathematical definition
10:57:46 <int-e> the formalization I've recently worked with cheated a bit... a partial recursive function is a partial function whose graph (modulo an encoding of pairs) is a recursively enumerable set; a set X is recursively enumerable if there is a function f such that x in X iff there is a y with f(x,y) = 0.
10:58:36 <int-e> (This is equivalent to adding a minimisation operator... thanks the the normal form theorem, essentially ... but the formalization doesn't prove that.)
10:59:20 -!- ais523 has quit (Quit: sorry for my connection).
10:59:33 -!- ais523 has joined.
10:59:47 <esowiki> [[Brainfuck extensions]] https://esolangs.org/w/index.php?diff=54927&oldid=54926 * Chronos * (+48)
11:15:45 -!- brandonson has quit (Ping timeout: 248 seconds).
11:17:20 -!- brandonson has joined.
11:30:45 -!- wob_jonas has joined.
11:31:09 <wob_jonas> "<ais523> is there a standard formalization for computable partial functions?" => that language Amycus was intended to be before I made an error is a pretty good formulation for that
11:31:37 <wob_jonas> the functions take integers as input and give integers as output, but lists and the code of the functions are also encoded as integers
11:31:45 <wob_jonas> actually natural numbers, not integers
11:32:04 <wob_jonas> no interactive IO mechanism, just an input parameter and an output parameter
11:32:43 <ais523> what I'm learning from this is that the standard formalisations are much more complicated than my typical tarpit :-)
11:33:01 <wob_jonas> no!
11:33:04 <wob_jonas> (0) isn't complicated
11:33:06 <wob_jonas> it's quite simple
11:33:10 <wob_jonas> no wait
11:33:16 <wob_jonas> I mean Amycus isn't complicated
11:33:19 <wob_jonas> (0) is, don't use that
11:33:25 <wob_jonas> use the language Amycus was meant to be
11:34:08 <wob_jonas> it only has function calls, successor for integers, making pairs, constant zero, and comparison or whatever the primitives were
11:34:10 <wob_jonas> let me check
11:37:16 <wob_jonas> the primitives are: 0: identity function, 1: any constant, 2: successor, 3: tuple indexing, 4: equality, 5: composition, 6: eval
11:37:22 <wob_jonas> darn, yes, that's not the simple language you want
11:37:50 <wob_jonas> no wait
11:37:51 <wob_jonas> 6 isn't eval
11:38:06 <wob_jonas> it's eval in my botched Amycus version
11:38:09 <wob_jonas> how the heck did this work
11:38:33 <wob_jonas> it was something more natural in the original Amycus
11:40:51 <wob_jonas> ais523: in the mainstream places that use Turing-machines, they use multi-tape machines to define things like this, I think.
11:41:04 <wob_jonas> as in, with separate tapes just for input and output
11:42:38 <wob_jonas> ... how the heck did the original Amycus work? I never really wrote it down properly
11:44:01 <wob_jonas> ah I see
11:44:06 <wob_jonas> those weren't how the primitives worked
11:44:53 <wob_jonas> the trick is that you have to consider everything as multi-argument functions, taking not a single integer, but a list of which it will access only the first few (constant number of) elements
11:45:41 <ais523> hmm, that's basically a syntax for µ-recursion at this point I think
11:45:57 <ais523> or, maybe not?
11:46:59 <ais523> µ-recursion has integer successor, +1, identity, composition/substitution, primitive recursion, minimisation
11:47:26 <wob_jonas> rule 1 is constant, rule 2 is successor, rule 3 accesses an input argument (at a given constant index), rule 4 is equality conditional, rule 5 is composition, and rule 6 is sort of eval, but 5 and 6 together can be used to emulate the S from a lambda expression, that is, calling variable functions
11:48:39 <wob_jonas> ais523: another thing you could do is lambda calculus where every value is either a function or a natural number, and if you try to call a natural number or do arithmetic on a function, you get a runtime error. this is somewhat similar to the original Amycus
11:49:27 <wob_jonas> only the original Amycus doesn't give runtime errors when you mix types, and also uses some de-Bruin or whatever encoding of the lambda terms
11:52:16 <wob_jonas> and the syntax is common-lisp-like in that the ordinary syntax lets you call only constant functions, and you need a special primitive 6 to call the result of a computation
11:52:45 <wob_jonas> and you need that for Turing-completeness, because just like in lambda-calculus, calling the result of a computation is the only way to loop
11:55:23 <wob_jonas> but yes, lambda calculus is probably not the easiest to simulate if you want to prove lots of languages T-c
11:57:33 <wob_jonas> hmm
11:58:14 <wob_jonas> the model I would prefer is pointer machines with read-only heap, no call stack only gotos. these can take input and give output in a register.
11:58:40 <wob_jonas> but that's also not one you'd want to implement.
11:58:45 <ais523> what's a read-only heap? do you mean that objects are read-only after construction?
11:59:32 <wob_jonas> yes, you can create conses (or possibly other algebraic structures too, declared at compile time, but here let's stick to conses), you have registers, each value is either a null or a pointer to a cons,
12:00:23 <wob_jonas> the operations are determining if a register contains cons and conditional jumping (there's no comparison), taking the car or cdr of a register and putting it to another register, copying a register to another register (this is actually optional), and putting null to a register.
12:01:07 <wob_jonas> also allocating a new cell on the heap by consing two registers and putting a pointer to the result in a register.
12:01:23 <wob_jonas> each state of the program tells which of these are done, and one or two next states.
12:01:30 <wob_jonas> and the registers they're done on
12:01:40 <ais523> hmm, you could almost do this using bignums and mingle from INTERCAL (together with a corresponding demingle)
12:01:48 -!- AnotherTest has joined.
12:01:59 <ais523> actually, given how easy demingle is to write in INTERCAL, this is basically bignum INTERCAL full stop :-)
12:03:50 <wob_jonas> ais523: perhaps, but actually representing trees with integers is not very natural, they tend to grow exponentially
12:04:11 <wob_jonas> anyway, that's sort of what Amycus does, but with a different representation as integers that grows even worse
12:04:28 <ais523> wob_jonas: well these integers aren't really interpreted as integers
12:04:31 <wob_jonas> so in an implementation, you'll actually store lists, not integers
12:04:39 <ais523> they're just a Gödel encoding of nested conses
12:07:02 <wob_jonas> anyway, if you want easy proof of T-c, then you don't even want to store trees,
12:07:08 <wob_jonas> you want some simpler language that works only with integers
12:07:09 <wob_jonas> easier to emulate
12:07:41 <wob_jonas> dunno, like Minsky machines
12:07:50 <ais523> I'm working on one now
12:07:53 <wob_jonas> those could take input and output in a register, right?
12:08:01 <ais523> I just wanted to prove it /equivalent/ to the existing examples
12:08:08 <ais523> and yes, Minsky machines are non-interactive-IO complete with 3 counters
12:08:18 <ais523> but I'm not sure if there's a formal proof of that anywhere
12:09:45 <wob_jonas> TAOCP uses a string-replacement esolang (Chomsky style) to define computations in the abstract sense, but I'm not sure what IO that has
12:10:00 <wob_jonas> do we even have a wiki entry for that by the way?
12:10:53 <wob_jonas> no we don't. I'll have to create one
12:14:24 <ais523> wob_jonas: that's what I was working on :-)
12:14:29 <ais523> although probably with slightly different semantics
12:14:42 <ais523> you can define noninteractive I/O for it fairly easily
12:15:16 <esowiki> [[User:B jonas]] https://esolangs.org/w/index.php?diff=54928&oldid=53643 * B jonas * (+218)
12:19:30 <ais523> hmm, it seems that we have a copy of _The Theory of Algorithms_ in our university somewhere
12:19:34 <ais523> I might look for it
12:20:21 <wob_jonas> ais523: Knuth also lists a newer version
12:20:39 <wob_jonas> and be careful, there are so many different books named like that
12:20:53 <wob_jonas> it's a pretty popular title
12:21:14 <wob_jonas> no wait
12:21:24 <wob_jonas> maybe the popular title was "Introduction to the theory of algorithms"
12:21:33 <wob_jonas> hmm
12:21:37 <ais523> wob_jonas: I checked to make sure the author matched too
12:21:43 <ais523> although it seems that the original wasn't in English, this one is a translation
12:21:59 <ais523> (I assume it wouldn't have been translated to English if it was already in English)
12:27:05 <wob_jonas> ais523: yes, and Knuth gives a reference to a translation
12:27:36 <ais523> I've had interesting experiences in the past with following citations
12:27:42 <wob_jonas> ais523: by the way, there's an interesting tradeoff here. I'm not talking about the one how Minsky machines are very easy to implement but not very powerful, but among the more powerful ones:
12:28:12 <ais523> once I discoverde that everyone was citing a particular book, but they actually just copied a citation from a well-known paper and the thing cited isn't actually there
12:28:28 <ais523> and not only that, but nobody else had checked at least at my University (it took them a while to find the book in question, and it hadn't been checked out for decades)
12:28:33 <wob_jonas> pointer machines let you easily define the runtime of a program, at least up to a polylog factor, so you cal tell if an algorithm runs in O(n**(1+epsilon)) time or O(n**(2+epsilon)) time for input of length n;
12:28:55 <ais523> err, that statement about pointer machines seems wrong
12:28:58 <ais523> it's circular
12:29:10 <wob_jonas> whereas string replacement machines let you easily define the memory use of a program, but not so much the runtime.
12:29:13 <ais523> you're saying "it's easy to define the runtime" whereas what you mean is "it's easy to define the runtime /on a pointer machine/"
12:29:28 <ais523> you're using pointer machines as the reference for runtime implicitly, because your computer is one of those
12:29:51 <wob_jonas> ais523: of course it is, but I mean it's a useful definition of runtime, as in, one that will correspond nicely (up to a polylog factor) to other more complicated realistic computable models,
12:30:28 <wob_jonas> and no, pointer machines still aren't the best choice for this, but among the simpler models they are a pretty good choice I think
12:31:46 <wob_jonas> you could measure runtime on a Turing-machine or a string-replacement machine or a Minsky-machine, but that's a less useful thing to measure, because the best runtime you get is still too slow for reality
12:32:24 <wob_jonas> Turing-machines let you measure the runtime up to polynomials, so you can define the class of polynomial runtime algorithms, as in P for decision problems
12:37:47 <wob_jonas> So what's the name of this esolang in Knuth 1.1?
12:38:45 <ais523> hmm, I suddenly realised that the most Knuth content I've read is when he sent me an email
12:38:52 <ais523> which is probably not a good thing
12:39:02 <wob_jonas> what?
12:39:14 <ais523> he wanted an updated version of the INTERCAL interpreter
12:39:33 <wob_jonas> yes, but I mean how have you not read a lot of TAOCP or Concrete Mathematics?
12:39:44 <wob_jonas> those books are thick, they're probably thicker than the email he sent you
12:39:54 <wob_jonas> and they're popular and good
12:39:57 <wob_jonas> you should read them
12:40:03 <ais523> a book being large doesn't necessarily mean I've read it
12:40:06 <ais523> also it was a short email
12:40:06 <wob_jonas> I'm not saying they're cheap, but they're worth the money
12:40:13 <ais523> the "not a good thing" was about not having read TAOCP
12:40:21 <wob_jonas> sure, I don't expect that you've read all of it
12:40:28 <wob_jonas> I just expect that you've read some of it
12:40:38 <ais523> hmm, how would I know which bits to read?
12:40:58 <wob_jonas> there's a procedure described for that in the introductino of TAOCP
12:41:07 <wob_jonas> that's for telling which part of TAOCP to read
12:41:14 <wob_jonas> I'm not sure for Concrete Mathematics
12:41:41 <wob_jonas> s/introductino/preface/
12:42:12 <wob_jonas> it's actually sort of a separate section after the preface, called "Procedure for Reading This Set of Books"
12:44:12 <wob_jonas> Also, if you can, read the third edition of volume 1 and 2, and the second edition of volume 3, as opposed to older editions; except obviously read the ultimate edition instead when it is published like twenty years from now
12:51:22 <wob_jonas> Hmm... I'll have to try to understand how rusts's CoerceUnsized works, because currently it sounds like heavily compiler-supported black magic to me that shouldn't have a right to exist, but it probably isn't that
12:56:37 <wob_jonas> `? yak
12:56:38 <HackEso> yak? ¯\(°​_o)/¯
12:58:20 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
13:00:34 -!- doesthiswork has joined.
13:01:10 -!- variable has joined.
13:01:18 -!- variable has quit (Client Quit).
13:04:35 -!- variable has joined.
13:29:37 -!- brandonson has quit (Ping timeout: 248 seconds).
13:31:34 -!- brandonson has joined.
13:43:58 -!- arseniiv has joined.
13:47:55 -!- variable has quit (Quit: /dev/null is full).
13:48:20 -!- SopaXorzTaker has quit (Remote host closed the connection).
14:11:28 -!- doesthiswork has quit (Quit: Leaving.).
14:33:12 -!- imode has joined.
14:38:21 -!- imode has quit (Ping timeout: 240 seconds).
14:56:42 -!- sebbu2 has joined.
14:58:39 -!- sebbu has quit (Ping timeout: 246 seconds).
15:07:56 <arseniiv> is there a known esolang based on https://en.wikipedia.org/wiki/%CE%9C-recursive_function ? Got one with binary strings instead of naturals, tell me if you want the description!
15:09:05 <arseniiv> it’s pretty old, I haven’t thought to dump it to the wiki and I hadn’t even known about this channel
15:18:15 <int-e> putting it on the wiki sounds like a good plan to me (provided you are okay with the CC0 license)
15:19:23 <arseniiv> totally okay
15:20:03 <arseniiv> is something like MathJax/KaTeX supported there?
15:23:04 <ais523> I don't think so
15:23:12 <ais523> it might be useful, not sure how easy mathjax is to install
15:23:22 -!- xkapastel has joined.
15:26:43 <arseniiv> what code has fungot for Befunge? I have stuck with the captcha :D
15:26:43 <fungot> arseniiv: so it should be: ( format " fnord" on my site if you want, but it's usually preferred to not have a standard fingerprint implementation
15:27:06 <arseniiv> ^be 9386882626>\#+:#*9-#\_$.@
15:27:15 <arseniiv> ^bf 9386882626>\#+:#*9-#\_$.@
15:27:22 <arseniiv> ^b 9386882626>\#+:#*9-#\_$.@
15:27:27 <arseniiv> well
15:27:39 <Taneb> I am fairly sure fungot cannot interpret befunge
15:27:40 <fungot> Taneb: but we'll see. it's still the same as picking the regular search replace in editors. she worked for a contract was done with plt. pgsql or something like the system() function in that program, guess not
15:27:55 <arseniiv> Taneb: ow((
15:28:15 <arseniiv> then I’m going to read its page, it seems
15:34:59 <esowiki> [[Special:Log/newusers]] create * Arseniiv * New user account
15:35:22 <arseniiv> ha ha ha I found js interpreter
15:35:50 <arseniiv> (at first I proceeded manually, but then I got that it would be a long story)
15:45:58 <esowiki> [[Esolang:Introduce yourself]] M https://esolangs.org/w/index.php?diff=54929&oldid=54924 * Arseniiv * (+234) introduced myself
15:50:40 -!- sebbu2 has changed nick to sebbu.
15:56:45 -!- LKoen has joined.
16:13:38 -!- SopaXorzTaker has joined.
16:16:20 -!- sleffy has joined.
16:21:47 -!- x86-64 has changed nick to risc-v.
16:32:08 -!- oerjan has joined.
16:52:48 -!- erkin has joined.
17:10:38 -!- imode has joined.
17:28:18 -!- SopaXorzTaker has quit (Remote host closed the connection).
17:33:01 -!- xkapastel has quit (Quit: Connection closed for inactivity).
17:33:21 -!- sleffy has quit (Ping timeout: 264 seconds).
17:40:01 -!- sleffy has joined.
17:53:02 -!- xkapastel has joined.
18:05:34 <oerjan> `! befunge 9386882626>\#+:#*9-#\_$.@
18:05:35 <HackEso> 271211889
18:05:44 <oerjan> arseniiv: ^
18:07:38 <arseniiv> thank you!
18:07:49 <arseniiv> (but it’s too late)
18:07:52 <oerjan> HackE{g,s}o was originally conceived to be an esolang interpreting bot like the old EgoBot, but more flexible. alas it got out of hand.
18:08:14 <oerjan> it doesn't help that the actual port of EgoBot's langs doesn't fit well into the rest.
18:08:46 <arseniiv> alas it got out of hand.> aren’t they all? :D
18:09:01 <oerjan> i mean it got used for even sillier things instead.
18:10:16 <arseniiv> I rethought the language I’m going to post, and its name will be YEOOIIOOIOA. Maybe someone can deduce what it means, while I’ll be writing the description!
18:10:43 <oerjan> arseniiv: the reason fungot cannot interpret befunge despite being written in it is because the obvious way of doing it would allow you to take over the bot
18:10:43 <fungot> oerjan: eiku joskus. sillon kun ei oo fnord toi on fnord
18:10:56 <oerjan> so only fizzie is allowed to run that command.
18:10:59 <arseniiv> (that one with binary strings, not ugh… Nora)
18:11:51 <arseniiv> oerjan: what, fungot is written on Befunge?? :o :o
18:11:51 <fungot> arseniiv: why? it would be a good idea; how else are you going to use an interpreter: interpreter program note: program can be the value of baz
18:11:56 <oerjan> ^source
18:11:57 <fungot> https://github.com/fis/fungot/blob/master/fungot.b98
18:12:47 <arseniiv> <I have no words>
18:13:01 <oerjan> oh binary string...
18:14:30 <oerjan> OOIIOOIO could be L, or possibly 2.
18:14:53 <arseniiv> I’ll guess I’ll be writing the description for several hours, my own notes are quite sparse
18:15:29 <arseniiv> yeah, it's 2, or more specifically "2", as there is a mechanism to interpret it as a number, and it will be 0x132
18:23:11 <arseniiv> am I format inline code by using <tt>...</tt> or some other thing?
18:23:21 <arseniiv> s/am/do
18:27:31 <oerjan> arseniiv: <code>...</code> is the usual way
18:28:44 <arseniiv> isn’t there a distinction between inline code and a block? ok
18:29:00 <oerjan> yes, blocks can be done in two different ways
18:29:24 <oerjan> either indent all lines with 1 space, or use <pre>...</pre>
18:30:18 <oerjan> <code> is inline, not a block
18:31:20 <arseniiv> got it
18:39:25 <oerjan> also see Esolang:Help
18:39:40 -!- SopaXorzTaker has joined.
18:41:55 <arseniiv> thanks!
18:43:08 -!- MildSpecter has joined.
18:51:10 <izabera> Taneb: why did jay-z start selling ice cream?
19:03:02 -!- risc-v has changed nick to firepaw.
19:03:10 -!- firepaw has changed nick to risc-v.
19:05:15 <zzo38> I think I fixed ff-composite.c now, did anyone make tests?
19:09:32 -!- SopaXorzTaker has changed nick to semiconductor.
19:10:30 -!- semiconductor has left ("Leaving").
19:15:24 -!- MildSpecter has left ("Quit").
19:16:53 <zzo38> Do you like my idea of Scientific Role Playing System? There are difference from GURPS, such as in GURPS I think the points are too coarse, so SciRPS has less coarse. Also, instead of zero level being human level as in GURPS, the zero is nothing (although the human template is included in the book). There is powers and skills.
19:18:48 <zzo38> Skills will have both the potential level and the actual level, and some skills may be worth different number of points. Attributes are also skills. Powers can have modifiers with a "layer system"; there can be many different kind of powers, such as normal powers (vision, flying, claws, poison bite, walking, etc) and special powers (such as magical and psychic powers).
19:24:35 -!- sleffy has quit (Ping timeout: 240 seconds).
19:26:21 -!- AnotherTest has quit (Read error: Connection reset by peer).
19:28:36 -!- LKoen has quit (Remote host closed the connection).
19:28:54 -!- AnotherTest has joined.
19:39:55 -!- sleffy has joined.
19:54:04 <Taneb> izabera, ice cream is good
20:04:35 -!- sleffy has quit (Ping timeout: 240 seconds).
20:11:13 -!- AnotherTest has quit (Read error: Connection reset by peer).
20:13:03 -!- xkapastel has quit (Quit: Connection closed for inactivity).
20:15:51 -!- AnotherTest has joined.
20:16:04 -!- risc-v has changed nick to x86-64.
20:34:04 <arseniiv> ah, I’m now come to the heart of the issue and am writing about expression semantics
20:34:36 <arseniiv> have came*
20:48:22 -!- LKoen has joined.
21:08:15 <izabera> Taneb: he sells ice cream because he has 99 problems
21:08:30 <Taneb> I see
21:13:58 <shachaf> fizzie: happy vappu
21:16:13 -!- wob_jonas has joined.
21:16:16 <wob_jonas> zzo38: great, I'll test it
21:18:48 -!- impomatic has quit (Ping timeout: 256 seconds).
21:32:50 <fizzie> shachaf: Glada vappen.
21:35:23 -!- oerjan has quit (Quit: Nite).
21:35:50 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
21:36:43 <esowiki> [[Special:Log/newusers]] create * E * New user account
21:37:12 -!- wob_jonas has joined.
21:38:12 -!- wob_jonas has quit (Client Quit).
21:38:59 -!- wob_jonas has joined.
21:44:15 <wob_jonas> zzo38: why are ffpbm and pbmff not named ffpnm and pnmff instead?
21:53:57 -!- LKoen has quit (Remote host closed the connection).
21:57:21 <esowiki> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=54930&oldid=54929 * E * (+379) me
21:57:50 <esowiki> [[User:E]] N https://esolangs.org/w/index.php?oldid=54931 * E * (+5) Created page with "fnord"
22:03:48 -!- erkin has quit (Quit: Ouch! Got SIGIRL, dying...).
22:10:25 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
22:13:32 <zzo38> wob_jonas: I don't know; that is just what I did at first. Maybe ffpnm/pnmff might have been better, but I won't change it now.
22:15:57 -!- LKoen has joined.
22:17:23 <zzo38> (Possibly, because they called it Netpbm, even though there are other Netpbm formats too, is probably why I did that.)
22:20:15 -!- sleffy has joined.
22:29:49 -!- wob_jonas has joined.
22:31:49 -!- boily has joined.
22:34:30 <zzo38> I tried to answer your question
22:35:40 <wob_jonas> ok
22:36:01 -!- x86-64 has changed nick to moony.
22:40:10 <wob_jonas> zzo38: gcc warns that the chan global variable is unused in ff-back.c and ff-matrix.c
22:41:10 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
22:44:16 <zzo38> You are correct; I removed that global variables
22:45:47 * boily suffers from today's alt texkcdt puns. ow.
22:49:34 -!- AnotherTest has quit (Ping timeout: 264 seconds).
22:52:13 <int-e> tbh cutting a pizza into a spiral sounds tedious
22:53:24 <int-e> unless, perhaps, you use an industrial strength record player
22:56:57 -!- xkapastel has joined.
22:59:46 <esowiki> [[YEOOIIOOIOA]] N https://esolangs.org/w/index.php?oldid=54932 * Arseniiv * (+10156) at last
23:01:37 <esowiki> [[YEOOIIOOIOA]] M https://esolangs.org/w/index.php?diff=54933&oldid=54932 * Arseniiv * (-2) link to Wikipedia is now fixed
23:02:58 -!- sparr__ has changed nick to sparr.
23:11:40 <esowiki> [[User:Arseniiv]] N https://esolangs.org/w/index.php?oldid=54934 * Arseniiv * (+63) let there be a page
23:23:38 <esowiki> [[YEOOIIOOIOA]] M https://esolangs.org/w/index.php?diff=54935&oldid=54933 * Arseniiv * (-6) typos and minor things
23:24:40 <arseniiv> should I add to https://esolangs.org/wiki/Language_list or categorization is enough?
23:31:33 -!- sleffy has quit (Ping timeout: 264 seconds).
23:43:07 <zzo38> It mentions Unicode, although Unicode seems irrelevant to the specification, since they are just byte strings, whether they represent Unicode characters or something else, it seem like
23:48:38 <arseniiv> yeah, it’s all byte strings, but for I/O I think it’s useful to treat them like UTF-8-encoded somethings
23:49:41 <arseniiv> or ASCII-encoded; Hello, world example utilizes at least that
23:51:15 <esowiki> [[YEOOIIOOIOA]] M https://esolangs.org/w/index.php?diff=54936&oldid=54935 * Arseniiv * (+1) there was ANSI instead of ASCII, what have I thought?
23:52:05 <zzo38> Yes, although you can leave the encoding up to the user; it might not even be text, but rather a picture (in farbfeld or Sixel format, perhaps), or audio, or non-Unicode text. If it is UTF-8 then of course it should not contain any invalid UTF-8 characters, but if it isn't, then it can be anything. Due to this, possibly also, an empty bit string becomes empty output, too.
23:52:24 <zzo38> (Note that ASCII is a subset of Unicode, so I am not counting it as "non-Unicode text" for the purpose of the above.)
23:54:43 <arseniiv> you’re right in that I overspecified what I/O there could be, I’ll edit
←2018-04-26 2018-04-27 2018-04-28→ ↑2018 ↑all