00:00:24 -!- moon_ has quit (Ping timeout: 250 seconds). 00:00:31 :t f 00:00:32 FromExpr a => a 00:00:39 hm weird 00:00:44 :t \x -> f x 00:00:45 (Show t, FromExpr r) => t -> r 00:01:33 > f 00:01:34 No instance for (Show a0) 00:01:34 arising from a use of ‘show_M461523684287504534910579’ 00:01:34 The type variable ‘a0’ is ambiguous 00:01:51 And I'm not sure I'll have the patience to actually read it. But the Goldbach program is relatively short in Laconic, and it really doesn't look like it should require 4888 states 00:02:35 "link to the paper"? 00:02:59 ? https://esolangs.org/wiki/Laconic 00:03:20 no, i mean, did you really mean "to" there 00:03:25 oh 00:03:57 * int-e is probably blind to an alternative way of parsing this sentence right now. 00:04:21 "the paper on"... oh. 00:04:36 that would be confusing :) 00:09:56 -!- moon has joined. 00:10:07 [wiki] [[Laconic]] https://esolangs.org/w/index.php?diff=46922&oldid=46921 * Oerjan * (+14) Fmt 00:10:14 int-e: the almost 5000 states they need has a roughly 4000 state interpreter for locanic 00:10:19 -!- moon has changed nick to Guest80724. 00:10:26 Laconic 00:10:39 int-e: well mostly it was just that i didn't realize someone made that page 00:11:19 -!- Guest80724 has changed nick to moon__. 00:11:33 tromp: I see. 00:12:19 * oerjan thinks formatting refs on esolang is confusing after he started using Cite templates on Wikipedia. 00:14:52 * moon__ is downloading freebsd to run on a virtual machine 00:19:05 -!- `^_^v has quit (Read error: Connection reset by peer). 00:19:23 [wiki] [[Var'aq]] https://esolangs.org/w/index.php?diff=46923&oldid=46920 * Oerjan * (-233) /* External resources */ merge and template 00:20:30 -!- Xe has changed nick to _8ade2af05f26. 00:20:53 -!- _8ade2af05f26 has changed nick to Xe. 00:21:57 Can we substitute the complexes for something weirder to get weird quantum physics? :P <-- istr that discussed and the answer is essentially "no" 00:22:14 at least quaternions give nothing more 00:22:38 -!- ybden has quit (Quit: leaving). 00:22:43 oerjan: Yeah, but what about the duals or split-complexes? 00:23:31 i don't remember what those are. 00:24:52 anyway, another idea is that quantum physics is based on a different idea of probability, and somehow only ordinary probability and quantum probability work without breaking the light speed limit for information. 00:25:01 oerjan: the dual numbers are what you get if you start with the numbers and reverse all the arrows hth 00:25:21 shachaf: meh 00:25:34 ouch 00:25:58 please try better than that if you want a swatting hth 00:26:12 that hurt more than any swatting hth 00:26:17 i know 00:27:05 well i had some good ones yesterday 00:27:12 (and not breaking the light speed limit is important because it's so easy to make up something "weird" that fails to do that) 00:27:15 true. 00:28:04 so the really weird thing about quantum mechanics isn't just how weird it is, but how it manages _not_ to give certain powers. 00:28:24 weird, but not always weird enough. 00:30:12 dan piponi wrote about something like that 00:31:03 mhm 00:31:59 freebsd is quite close to unix, right? 00:32:10 the image of dorian gray <-- . o O ( this file never changes, although it's SHA256 hash does... ) 00:32:18 *its 00:32:35 spooky 00:33:00 no one ever talks about that kind of attack on a hash function 00:33:03 second image attack 00:35:34 freebsd is quite close to unix, right? 00:36:11 ah, http://blog.sigfpe.com/2013/10/distributed-computing-with-alien.html 00:36:33 "It suggests that quantum mechanics is right at the edge of the space of possible physics that make life difficult for us. If quantum mechanics were to be tweaked the tiniest amount to make correlations any stronger, large numbers of difficult distributed computing problems would suddenly collapse to become trivial. If the conjecture is true it means that nature looks a bit like a conspiracy to keep comp 00:36:39 uter scientists in work." 00:36:47 so close to unix, so far from plan 9 00:37:18 shachaf: hm that might be where i got it from too 00:38:17 good 00:38:34 that means running a virtual machine of it and treating it like a terminal is not a bad idea (lol) 00:41:16 * oerjan think moon__ may have got whooshed 00:41:23 actually 00:41:30 i can be batshit crazy at times 00:41:56 i think they're both POSIX systems, no? 00:43:34 oh wait 00:43:53 * oerjan misread unix as linux 00:44:10 anyway, it's all a big happy family. 00:44:34 don't mind the strange sounds in the attic. 00:46:34 vi woont work, saying the disk is readonly, but i have a empty disk mounted, help? 00:47:25 what does df say 00:47:50 hm wait 00:48:22 something about dwarves 00:49:13 lol damnit 00:49:27 nice one sha 00:49:42 it says i have two drives mounted (i guess?) rootfs and devfs 00:50:36 sorry, that command doesn't say whether it's readonly. i've forgotten what does. 00:50:52 * oerjan doesn't do sysadmining himself. 00:51:39 'mount' generally does. 00:51:39 mount 00:52:14 fizzie: how's pooches? 00:52:35 if you have root mounted as readonly, then mount will usually give incorrect information, because it tries to read the list of mounted disks from /etc/mtab which is read-only so previous mounts couldn't write it. 00:52:50 -!- moon has joined. 00:53:00 connection died 00:53:14 -!- moon has changed nick to Guest90291. 00:53:23 ok, i just had problems getting it to show only one mount. 00:53:28 You can try reading /proc/self/mounts instead to ask the kernel to give you a list of mounts (sadly it's formatted to human-readable irreversibly) 00:53:31 -!- Guest90291 has changed nick to moon___. 00:53:44 b_jonas: At least my mount just reads from /proc. I don't know about FreeBSD's, though. 00:54:00 so i did df . first to get the mount point, then mount | grep ... 00:54:08 -!- moon__ has quit (Ping timeout: 250 seconds). 00:54:15 nope 00:54:52 ? 00:54:56 what oerjan? 00:54:57 (And my /etc/mtab is a symlink to ../proc/self/mounts, apparently.) 00:55:02 *is linux shit* 00:55:15 moon___: fizzie said you might have trouble if root is readonly 00:55:20 moon___: three guesses: (1) try vim, (2) mount a writable disk (possibly a tmpfs) and use the :set directory command of vi to put the temp files there, (3) ask #vim on this network 00:55:20 I didn't. 00:55:24 b_jonas did. 00:55:28 oh. 00:55:41 yea, i used the virtual disk supplyed by freebsd 00:55:44 shachaf: I don't know whether that's an idiom or reference. 00:55:55 b_jonas: please stop resembling fizzie twh 00:56:12 (4) if all else fails, just use sed, it doesn't need to create temp files 00:56:30 rootfs is read only 00:56:35 trouble, darnit 00:56:42 `? fizzie 00:56:55 fizzie is not fnord with a monad but the king of #esoteric, see http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/src/fizziecoin.jpg 00:57:06 moon___: also, um, this might be dumb but, any chance you can just remount the root fs to read-write? 00:57:26 Any chance you can just have a normal installation of a thing. 00:57:40 like, how? 00:57:50 how b_jonas 00:58:42 moon___: like, first find out why it was mounted read-only, if it was mounted read-only because there are serious file system or disk io errors, then it might be dangerous to try to remount so don't do it, but if it is mounted ro only at boot and something happened and the boot couldn't proceed then it MIGHT be safe to remount, 00:58:51 moon___: then mount -oremount,rw / 00:59:07 its a virtual machine 00:59:12 cant hurt anything for now 00:59:13 (if it's on a read-only disk, then it won't work) 00:59:22 * oerjan vaguely assumes moon___ has never used a unix-alike system before. 00:59:31 * moon___ notes that oerjan is right 00:59:39 I thought he's just never adminned one 00:59:50 which means this advice is probably a bit over his head 01:00:03 (it's pretty close to over mine, too) 01:00:32 b_jonas: This would be relatively normal experience to have as a *knowledgable* user of a typical Unix system. 01:00:33 ill try just using a installation disk 01:00:43 oerjan: well, learning unix by managing an installation at home and trying to fix stuff I break did work for me. 01:00:44 It's "mount -u -o rw /" on FreeBSD, I believe. But, I mean, I distinctly remember a stock FreeBSD installation having writable storage. 01:01:00 Though certainly not, say, an OS X system, or a slightly less knowledgable user of a Unix system. 01:01:12 thx fizzie 01:01:13 done 01:01:50 pikhq: dunno, the mount might be but the vi magic part probably isn't. some unixy people can live their lives without knowing much of vi 01:02:05 -!- Phantom_Hoover has quit (Remote host closed the connection). 01:02:11 `? vim 01:02:12 `? vi 01:02:16 `? 5 01:02:25 `? ex 01:02:41 Okay, yes, knowing vi in detail is mostly gonna be admins or certain serious programmers. 01:02:55 Classically, you need to know enough of vi to get out of it. 01:03:01 No output. 01:03:04 vi is in a relationship with emacs. 01:03:05 ex? ¯\(°_o)/¯ 01:03:05 5? ¯\(°_o)/¯ 01:03:28 fizzie: my opinion is that you need to know enough of vi to modify some makefiles of another editor you compile 01:04:11 On systems people *actually use*, knowing nano exists is plenty. 01:04:32 i just realized my system thinks it doesnt have mk, im worried it might not 01:04:34 [wiki] [[Confusion]] M https://esolangs.org/w/index.php?diff=46924&oldid=46918 * H3LL * (+1) 01:04:41 http://smbc-comics.com/index.php?id=718 01:04:51 moon___: mk is a HackEgo command we made ourselves. 01:04:56 oh 01:05:06 *sigh* 01:05:15 `cat bin/mk 01:05:16 [[ "$1" == ?*//* ]] || { echo usage: "mk[x]" file//contents >&2; exit 1; }; key="${1%%//*}"; value="${1#*//}"; echo "$value" > "$(echo-p "$key")" && echo "$key" 01:05:23 pikhq: maybe you also have to know what handful of configuration you have to set to make programs invoke your editor of choice as the default editor, because not everything respects $VISUAL or $EDITOR 01:05:29 -!- centrinia has joined. 01:05:31 moon___: because we don't have proper terminal access for a normal editor. 01:05:42 i dnt get vi at all 01:05:55 moon___: do you have nano, as pikhq suggested? 01:06:09 nope 01:06:34 Sure. 01:06:43 oerjan: we could implement a small stateful line editor though. probably something like the editors in basic or APL systems that don't renumber lines until you explicitly ask but let you insert lines between already existing ones. 01:07:23 b_jonas: we don't like too much state, it clutters up the repository history 01:08:01 oerjan: yeah... isn't there a stateful directory (or other state) that's not versioned though, so we use that? 01:08:05 nope 01:08:07 No. 01:08:18 Not even as an accidental feature or bug? 01:08:31 Not as a known one. 01:08:33 moon___: what are you hoping to edit with vim anyway if your file systems are read-only? 01:08:51 i fixed that 01:08:54 fizzie 01:09:11 fizzie: hmm... note to self, find some such bug 01:09:39 It would likely involve a bug in the uml sandbox. 01:10:37 freebsd lacks vim 01:10:39 it has vi 01:10:44 and ee (easy editor) 01:10:48 fizzie: I don't think it needs to. those sandboxes aren't really built to be completely deterministic, so we can use some side-channel to store state. 01:10:50 im going to use ee to learn 01:10:59 hppavilion[1]: Ew. 01:11:18 oerjan: Correct 01:11:30 moon________________________________________! 01:11:48 b_jonas: Each execution is a fresh boot of UML, and the only thing bound read-write from outside is the repository directory. 01:12:08 fizzie: yes, but I said side-channels. non-deterministic stuff. 01:12:26 not actualy fs directories 01:12:41 moon___: btw you can always replace mk with echo '...' >filename , although you might need to escape 's inside 01:12:44 I'll believe that once I see it. 01:13:40 there might not be a practically usable one of course 01:16:17 Having an easy-to-use stateful editor would also be a loss in the sense that editing files would be less of a puzzle game every time. 01:17:50 `cat bin/sedlast 01:17:51 sed -i "$1" "$(lastfiles)" 01:18:08 `ls fetch 01:18:09 ls: cannot access fetch: No such file or directory 01:18:14 `cat bin/fetch 01:18:15 cat: bin/fetch: No such file or directory 01:18:20 `cat fetch 01:18:21 cat: fetch: No such file or directory 01:18:24 hmm 01:18:48 moon___: fetch is a builtin of HackEgo that's outside the sandbox precisely so it can access the web freely 01:18:55 oh 01:19:12 fizzie: well, it's a win-win because if I don't find such a state, then I can just say termbot is superior 01:19:21 (for it has persistent state) 01:19:30 `` ls bin/*sed* 01:19:31 bin/sedlast 01:19:32 (and it's not version controlled) 01:20:40 that does not sound superior hth 01:20:59 what would be a good way to put a file (like ciol intepreter src) onto freebsd? 01:21:18 moon___: normally i'd suggest curl or wget but i don't know if those work inside your VM. 01:21:27 well 01:21:31 mine is set up ith connection 01:21:34 no no, first check if you can install from ports 01:22:07 b_jonas: ciol is his own esolang 01:22:37 * oerjan never used a VM so should shut up at this point 01:23:22 but there's probably a standard way to transfer files between inside and outside 01:23:39 do we even know what kind of vm it is? 01:23:47 good point 01:24:26 also, it's never "a standard" 01:24:55 well standard for the VM 01:24:57 there's probably multiple standard ways 01:24:58 its virtualbox 01:25:39 moon___: well, if it already has a network connection to the outside, then you can probably transfer the file through that 01:25:39 . o O ( wednesday, so two extra mezzacotta comics ) 01:25:42 shachaf: the heck is a comonoid? Is it... something where... nah, I'm stumped. 01:25:59 `? comonoid 01:26:00 tswett: it's like a monoid but with the arrows reversed hth 01:26:00 comonoid? ¯\(°_o)/¯ 01:26:09 wait, we don't have an entry on this one? 01:26:15 strange 01:26:29 tswett: this time shachaf speaks truth 01:26:30 `? comonad 01:26:32 Comonads are just monads in the dual category. 01:26:52 Is a comonoid in a category a monoid in the opposite category? 01:26:58 Yes. 01:27:29 b_jonas: they're uninteresting in haskell because _every_ type is a comonoid with the operations \x -> () and \x -> (x,x) (iirc) 01:27:59 oerjan: For (,)-comonoids, yes. 01:28:10 You can also talk about comonoids with Either, I guess. 01:28:20 are those interesting? 01:28:23 No. 01:28:36 You have counit : M -> Void and comult : M -> Either M M 01:28:38 hat does pkg mean when 'Error fetching : no package record? 01:28:46 So only Void is an instance. 01:29:16 You can also talk about monoids with Either -- unit : Void -> A and mult : Either A A -> A -- and those are also uninteresting. 01:29:28 oerjan: um, just because there's one definition that works, there might also be more interesting instances you can define. 01:29:35 b_jonas: There aren't. 01:29:58 On the other hand, if you're talking about monoids in the category of endofunctors you get all sorts of interesting options. 01:30:12 You can get Applicative, Alternative, and Monad. 01:30:20 Alternative too? 01:30:31 Yes, you can get it two different ways. 01:30:51 The simpler one is with (:*:). 01:31:14 unit : Proxy ~> F; mult : (F :*: F) ~> F 01:31:16 is there a category of monoid concepts in the category of endofunctors, and does it have initial and final elements 01:31:30 what's a monoid concept 01:31:49 those options you mention 01:32:01 so s/concept/option/ 01:32:20 Well, the "options" here are choices of tensor products for the category of endofunctors, I guess. 01:32:32 what does pkg mean when 'Error fetching : no package record'? 01:32:41 im trying to have it install itself 01:32:48 it doesnt come preinstalled 01:32:59 I'm not sure what a morphism from one choice of tensor product to another is? 01:33:36 I guess a tensor product is a bifunctor : CxC -> C 01:33:40 shachaf: the category is defined in the natural way hth 01:33:43 So you can talk about natural transformations between them. 01:34:05 b_jonas: so what's the answer twh 01:34:31 fizzie: can we have a graph of the rise of hth along with its friends twh tdh tdnh etc. twh 01:35:03 shachaf: you'd imagine the Applicative => Monad relationship to give an arrow somewhere? 01:35:24 Maybe? 01:35:41 Also the category is probably more complicated because you have the identity and the isomorphisms. 01:36:38 > f :: Expr -> Expr 01:36:39 Expr> 01:36:51 (no relation) 01:38:10 (i've just been thinking about the awkward fact that Expr stuff needs too much type annotation) 01:38:59 > f (g x) :: Expr 01:39:01 No instance for (Show t0) arising from a use of ‘f’ 01:39:01 The type variable ‘t0’ is ambiguous 01:39:01 Note: there are several potential instances: 01:39:29 :t \g -> f (g x) :: Expr 01:39:30 Show t => (Expr -> t) -> Expr 01:40:14 hm lambdabot doesn't do ExtendedDefaulting? 01:40:53 :t f (g x) :: Expr 01:40:54 No instance for (Show t0) arising from a use of ‘f’ 01:40:54 The type variable ‘t0’ is ambiguous 01:40:54 Note: there are several potential instances: 01:41:01 :t f (g x) 01:41:02 Could not deduce (Show t0) arising from a use of ‘f’ 01:41:02 from the context (FromExpr t) 01:41:02 bound by the inferred type of it :: FromExpr t => t 01:44:32 So I got to thinking today—using pretty much only just existing software, how could you make a Linux distribution which contains the smallest possible number of non-.NET components? 01:45:14 Let's say that in order to be considered a "Linux distribution", it has to be capable of compiling itself, and it also has to be minimally usable. 01:45:56 -!- J_Arcane has quit (Read error: Connection reset by peer). 01:45:56 By definition, the distro will have to include Linux. If I'm not mistaken, GCC is also mandatory, because nothing besides GCC can compile Linux. 01:46:18 Without binutils, you're in a pretty bad spot. 01:46:23 -!- J_Arcane has joined. 01:46:31 I think technically you could add Busybox and call it done. 01:46:53 But that's not an answer that's in the spirit of the question. 01:47:08 So, add in Mono as well. 01:47:26 @ask int-e Would anything important go wrong if you added Expr to the end of lambdabot's default list? (i can think of one thing: expressions confusing Integrals and Fractionals might inexplicably start giving Exprs instead of errors.) 01:47:26 Consider it noted. 01:47:27 tswett: you also need a libc besides the busybox 01:47:46 Oh yeah, libc is pretty important. 01:47:52 tswett: as well as a make, a couple of libraries for gcc, and some shell utilities to make the kernel build work 01:48:10 Do you really need make at all? 01:49:01 tswett: for compiling the kernel, sure, it's driven by makefiles 01:49:10 * tswett nods. 01:49:22 So either use make, or replace all the makefiles with something else. 01:49:24 oh, and you need libc headers too, to be able to compiler gcc 01:49:41 You could use fsharpi as a shell. 01:49:52 It would make a pretty crappy standard Linux shell. 01:50:04 tswett: what you could do is to not include gcc directly, but instead include a smaller compiler and use that to compile a gcc 01:50:07 freebsd is causing problems, wont let me run files 01:50:37 although with recent gcc versions that's pretty difficult because you actually need a c++ compiler, and there's not many working ones other than gcc and clang 01:50:47 any help? 01:50:54 i need to run rc but i cant 01:50:59 moon___: um, be more specific please 01:51:08 permission denied 01:51:21 Sounds like you might need to chmod +x the file you're trying to run. 01:51:26 tswett: you need a _real_ shell and a couple of utils to run those makefiles 01:51:43 -!- hppavilion[1] has quit (Ping timeout: 252 seconds). 01:52:00 So the minimal self-compiling Linux distro actually includes, like, several components. 01:52:49 tswett: sure. and in a real distro, you'd also need an init and a modutils, though if you ONLY want to self-compile, not do anything else, you can probably omit them 01:53:14 Surely init could be written in .NET. 01:53:29 Or, like, 01:53:39 ln -s fsharpi /bin/init 01:53:41 tswett: ugh... no. and traditional init is a small program, so you don't have to. 01:53:58 it's easier to just have an init 01:54:25 tswett: wait wat 01:54:27 -!- jaboja has joined. 01:54:35 tswett: the biggest problem isn't even gcc 01:54:47 tswett: in "self-compiling", do you mean also compiling libc? 01:54:58 Probably. 01:55:04 because compiling libc is pretty much impossible (this is a known bug or known feature, depending on who you ask) 01:55:24 so you can't really have a completely self-compiling linux distro with gnu libc at all 01:55:37 you'll need to use an alternate libc, and that makes a lot of things harder 01:55:47 glibc is nearly impossible to compile? 01:56:06 tswett: yes, or it was last I checked, some years ago. there's even a bug ticket about it which says so. 01:56:31 tswett: have you ever compiled one? I did, but that was a decade ago, so I know it's *changed* to become impossible to compile at some point. 01:56:39 It's not just that I'm stupid and that's why I can't compile it. 01:56:43 It grew and became worse. 01:56:44 I don't think I have. 01:57:26 More like 15 years ago actually. How time flies. 02:00:43 But failing to build libc was also a long time ago, so eventually I should try building libc again, because it's annoying that I can't. I know libc has changed a lot since I failed to build it. 02:01:14 i dont get rc.conf at all, and i cant enable networking *sigh* 02:01:29 I should probably try to build it around the time I try contact lenses the third time, for the situation is somewhat analogous, with me, contact lenses, and libc all changing a lot during the yeras. 02:07:31 > last (takeWhile (> 0)) (iterate (/ 2) 1) 02:07:32 Couldn't match expected type ‘[[Double] -> t]’ 02:07:32 with actual type ‘[Integer] -> [Integer]’ 02:07:32 Probable cause: ‘takeWhile’ is applied to too few arguments 02:07:47 > last (takeWhile (> 0) (iterate (/ 2) 1)) 02:07:49 5.0e-324 02:14:26 oh well, im going to have to go to my ip configuration to make address space for freebsd 02:24:42 -!- Akaibu has quit. 02:29:28 -!- moon___ has quit (Ping timeout: 250 seconds). 02:31:54 `` echo -e 'foo\nbar\rPRIVMSG #esoteric :hi\r\n' #just double checking fizzie's claim 02:32:16 foo \ bar 02:33:00 maybe it fails because it doesn't output \r\n, only \r 02:37:34 `` echo -e 'foo\nbar\rQUIT' 02:37:37 foo \ bar 02:41:06 > 5.0e-324 02:41:08 5.0e-324 02:41:10 > 5.0e-324 / 2 02:41:12 0.0 02:41:40 > 324 * logBase 2 10 02:41:42 1076.3047027435055 02:42:12 > length (takeWhile (> 0) (iterate (/ 2) 1)) 02:42:13 1075 02:42:31 > subtract 1 . last . takeWhile (> 1) . map (+1) . iterate (/ 2) $ 1 02:42:32 2.220446049250313e-16 02:43:09 > (1 + 2.220446049250313e-16, 1 + 2.220446049250313e-16 / 2, 1 + 2.220446049250313e-16 / 3, 1 + 2 * 2.220446049250313e-16 / 3) 02:43:11 (1.0000000000000002,1.0,1.0,1.0000000000000002) 02:43:33 What rounding rule is this thing using? 02:43:40 Ties go towards zero? 02:50:26 -!- Moon__ has joined. 02:50:32 On my tablet now 02:50:43 i discovered termux 02:54:24 > (2 * 5.0e-324) * 1.5 02:54:26 1.5e-323 02:54:45 Whoops, that's the wrong one. 02:54:48 > (2 * 5.0e-324) * 0.75 02:54:50 1.0e-323 02:55:30 tswett: well calculations are IEEE, not sure about output. 02:55:42 Well, the rounding rule can't be "ties go towards zero", then. 02:56:30 towards even perhaps? 02:56:35 By the way, I propose that 2.220446049250313e-16 be called a "grain" and that 5.0e-324 be called a "speck". 02:56:46 That would probably make sense. 02:57:12 > [1.0 - 1, 1.0000000000000001 - 1, 1.0000000000000002 - 1, 1.0000000000000003 - 1, 1.0000000000000004 - 1] 02:57:13 [0.0,0.0,2.220446049250313e-16,2.220446049250313e-16,4.440892098500626e-16] 03:07:45 So, about those groups with unsolvable word problems (i.e. undecidable equality). 03:08:02 One of them is the group of (Collins 1986): https://en.wikipedia.org/wiki/Word_problem_for_groups#Examples 03:08:16 Which is somewhat complicated. 03:16:26 I guess we probably don't have a better one. 03:16:35 So, do I smell an esolang here? 03:16:45 Hi tswett 03:18:47 Hey there. 03:20:29 I think I do smell an esolang here. 03:20:52 Take Collins' group there. Just come up with some notation for concisely expressing elements of the group, and that's the language itself. 03:21:05 The semantics: you "run" a program by trying to reduce it to the identity word. 03:25:05 Wait a minute, that group of Collins is pretty much just a hacked-up version of a semigroup by Céjtin. 03:26:02 This variant of said semigroup is so simple that I can easily give an entire presentation for the whole thing in one IRC message: 03:26:51 ac = ca; ad = da; bc = cb; bd = db; ce = eca; de = edb; cca = ccae 03:30:57 -!- tromp_ has joined. 03:31:12 `fetch http://pastebin.com/raw/BGTV9E93 03:31:15 2016-05-05 02:31:12 URL:http://pastebin.com/raw/BGTV9E93 [3761] -> "BGTV9E93" [1] 03:31:23 someone compile that as bin/ciol 03:32:16 im not too good with gcc, i leave that to my ide :p 03:39:03 -!- Froox has joined. 03:40:02 Hi 03:40:09 `ciol rhello; 03:40:24 hello 03:40:57 -!- Froo has joined. 03:41:36 Hi 03:41:41 `ciol rhello; 03:41:47 hello 03:41:59 #slowego 03:42:09 -!- Frooxius has quit (Ping timeout: 244 seconds). 03:42:41 Anyone alive in here? 03:43:18 tswett? 03:43:22 huh. 03:43:47 Yup, I'm here. 03:44:33 -!- Froox has quit (Ping timeout: 240 seconds). 03:44:46 I'm here, but I'm not alive. 03:45:31 Lll 03:45:34 lol 03:45:45 i really adore termux 03:46:02 power =ful linux/unix like interface on your android' 03:46:03 -!- Froo has quit (Ping timeout: 240 seconds). 03:47:21 How do i tell a linux os to allow a file to be executed without chmod? 03:47:37 and in such a way i dont have to repeated that trick every time 03:48:05 tswett and pikhq ^ 03:48:08 You don't, really. If you want a file to be executable you have to use chmod on it first. 03:48:24 `` gcc -x c -o bin/ciol BGTV9E93 # oh well 03:48:37 No output. 03:49:04 Though, most things that generate executables will set things as executable for you. 03:49:23