01:43:28 <lambdabot> error: Data constructor not in scope: PostScript
01:43:40 <nakilon> there should be GetScript and PutScript
01:53:25 -!- oerjan has joined.
01:54:12 <oerjan> whee! finally solved that dread loopy puzzle!
01:56:19 <oerjan> <int-e> anyone remember this game: https://int-e.eu/~bf3/tmp/2k.png <-- yep!
01:56:31 <oerjan> the end of that loopy puzzle was a bit fascinating.
01:57:33 <oerjan> for a long time i'd only solved the bottom left completely, but there were many other separate regions where i gradually got information, but which seemed to stay in superposition of 2 or 3 options :)
01:57:41 <int-e> did it go clockwise starting in the top right?
01:57:56 -!- delta23 has quit (Quit: Leaving).
01:58:05 <int-e> hmm. that's a no, I guess
01:58:20 <oerjan> it was nakilon's puzzle in case that's not clear
01:58:51 <int-e> That's what I thought.
01:59:21 <oerjan> also, there was this spot i'd noticed in the middle right (and mentioned yesterday) which had a plausible uniqueness exploitation almost like the one you linked
01:59:34 <nakilon> oerjannow you can safely check the screenshots )
02:00:01 <oerjan> and finally, i got to the point where a counting argument connected that to something else in the upper right
02:00:36 <nakilon> do you mean the horizontal line section that could be joined to bottom and top only once?
02:00:58 <oerjan> so i chased it across the upper right until it reached the large near-island in the upper left
02:01:12 <nakilon> I thought about it and that's why I asked the question about do puzzles have to have a single solution, but then I've found something elsewhere in the bottom right and it solved the thing without the exploit
02:02:06 <oerjan> which showed how that island connected to the rest. and then i traced that information from both connection points - until i found a trivial contradiction near the middle left border
02:03:36 <oerjan> and once that trivial contradiction had been used, the _entire board unraveled with only simple rules_
02:03:54 <oerjan> so it was, as usual, something simple i had missed for a long time
02:05:20 <oerjan> (the lower left took a little longer to unravel than the rest, but only because the board was zoomed so large i couldn't see all the large loop parts instantly)
02:06:52 <nakilon> on the lower left I did several mistakes and had to backtrace a lot, but the solution was somewhere in the very middle, where when you enable one line it would make the cycle of 3 and once you disable it it starts solving down and then left
02:07:17 <oerjan> nakilon: it was sort of horizontal section with a diamond between, if that's what you mean
02:07:37 <nakilon> oerjan it was like 4 horizontal lines connected
02:07:45 <nakilon> to the right from center and maybe 1 up
02:08:32 <oerjan> argh let me see if i can find it without backtracking everything
02:09:13 <nakilon> this https://i.imgur.com/x9phtJ7.png
02:12:30 <oerjan> it was further down and to the right
02:13:01 <nakilon> once in the upper right I used to solve a portion of trying both enable and disable one line
02:13:20 <nakilon> then I screenshoted both results and applied what was common
02:14:34 <nakilon> if I was making an AI I would use this trick
02:14:52 <oerjan> i did try similar things (without screenshotting)
02:15:08 <nakilon> it kind of makes possible to solve further moves before solving the current one
02:17:27 <oerjan> it wasn't exactly horizontal, now i've looked at it
02:18:35 <nakilon> I now wonder if the trick could be applied to other things I was solving earlier
02:21:47 <nakilon> I feel like it should be somewhere on this wiki but the closest I've found right now is https://www.chessprogramming.org/Killer_Heuristic
02:21:57 <nakilon> though in chess the order matters
02:21:59 <oerjan> ok your screenshots were indeed not very spoilery :P
02:22:20 <oerjan> the last one looked a bit similar to my endgame
02:22:31 <oerjan> with bottom left remaining
02:30:07 <oerjan> <riv> i once wrote a postscript thing to make a certain fractal <-- hm i vaguely recall doing mandelbrot in postscript, on an actual printer. but it was only black and white and i think it was dog slow.
02:30:18 <esolangs> [[Minim]] https://esolangs.org/w/index.php?diff=86707&oldid=86704 * KakkoiiChris * (+15700) /* brainfuck */ Added syntax highlighting
02:30:45 <oerjan> although i might be misremembering whether i actually wrote the program myself
02:31:50 <oerjan> if i did, it was the most interesting postscript program i wrote directly
02:32:22 <esolangs> [[Minim]] https://esolangs.org/w/index.php?diff=86708&oldid=86707 * KakkoiiChris * (+60) /* Links */ Added GitHub repository link
02:32:56 <oerjan> i suppose mandelbrot is a bad fractal to choose if you want to use postscript's rendering strengths :P
02:36:14 <esolangs> [[Minim]] M https://esolangs.org/w/index.php?diff=86709&oldid=86708 * KakkoiiChris * (+126) /* brainfuck */ Added implementation note to top
02:53:02 -!- impomatic has quit (Quit: impomatic).
02:53:19 -!- impomatic has joined.
02:57:34 -!- impomatic has quit (Client Quit).
02:57:53 -!- impomatic has joined.
03:03:03 -!- impomatic has quit (Quit: impomatic).
03:03:19 -!- impomatic has joined.
03:04:48 <esolangs> [[Ark]] https://esolangs.org/w/index.php?diff=86710&oldid=86438 * Spargle * (+194) /* Ark: The esolang that is actually kind of useful. */
03:07:34 -!- impomatic has quit (Client Quit).
03:07:51 -!- impomatic has joined.
04:04:51 <pikhq> sounds like a fun challenge though
04:30:35 <esolangs> [[Minim]] https://esolangs.org/w/index.php?diff=86711&oldid=86709 * KakkoiiChris * (+8147) /* About */ Elaborated on the description of Minim, and added a brief history of the language's syntax
04:31:46 <esolangs> [[Minim]] M https://esolangs.org/w/index.php?diff=86712&oldid=86711 * KakkoiiChris * (-12) /* History */ Removed one layer of nesting from the History section
04:32:55 <esolangs> [[Minim]] M https://esolangs.org/w/index.php?diff=86713&oldid=86712 * KakkoiiChris * (+24) /* History */ Added separators between each version
04:39:00 <esolangs> [[Brainfuck implementations]] M https://esolangs.org/w/index.php?diff=86714&oldid=85173 * KakkoiiChris * (+139) /* Normal implementations */ Added link to Minim implementation example program
04:43:05 <esolangs> [[Truth-machine]] M https://esolangs.org/w/index.php?diff=86715&oldid=86569 * KakkoiiChris * (+3467) /* Minim */ Added syntax highlighting
04:47:56 <esolangs> [[Truth-machine]] M https://esolangs.org/w/index.php?diff=86716&oldid=86715 * KakkoiiChris * (+0) /* Implementations */ Fixed alphabetical ordering of Minim
04:50:30 <esolangs> [[Truth-machine]] M https://esolangs.org/w/index.php?diff=86717&oldid=86716 * KakkoiiChris * (-1288) /* Minim */ Removed line numbers and header comment
05:50:53 -!- oerjan has quit (Quit: Nite).
06:01:42 -!- SGautam has joined.
06:01:55 -!- SGautam has quit (Client Quit).
06:02:21 -!- SGautam has joined.
06:12:23 -!- hendursaga has quit (Ping timeout: 244 seconds).
06:14:21 -!- hendursaga has joined.
06:19:58 -!- Lord_of_Life_ has joined.
06:20:34 -!- Lord_of_Life has quit (Ping timeout: 258 seconds).
06:21:12 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
06:43:28 <esolangs> [[Backrooms]] N https://esolangs.org/w/index.php?oldid=86718 * Ch44d * (+19399) Add backrooms
06:48:50 -!- spruit11_ has joined.
06:49:43 -!- spruit11_ has quit (Client Quit).
06:50:43 -!- spruit11_ has joined.
06:51:11 -!- spruit11 has quit (Ping timeout: 252 seconds).
06:51:24 -!- spruit11_ has quit (Client Quit).
06:51:43 -!- spruit11 has joined.
07:18:35 -!- Sgeo has quit (Read error: Connection reset by peer).
07:19:38 <esolangs> [[Special:Log/upload]] upload * KakkoiiChris * uploaded "[[File:MinimLogo.png]]"
07:26:09 <esolangs> [[Backrooms]] https://esolangs.org/w/index.php?diff=86720&oldid=86718 * Ch44d * (+171) fix formatting problems and add more rule info.
07:32:11 <esolangs> [[Minim]] https://esolangs.org/w/index.php?diff=86721&oldid=86713 * KakkoiiChris * (+380) Added logo and info box to top
07:37:06 <esolangs> [[Minim]] https://esolangs.org/w/index.php?diff=86722&oldid=86721 * KakkoiiChris * (+7) Moved infobox and logo
07:43:33 <nakilon> is there anything like a CLI that would find the most common substrings? like some debug info for compression algorithms
07:44:10 <nakilon> I would apply it to the problem I want to solve -- finding the meaningful substrings in an array of strings
07:44:44 <nakilon> such as... #haskell #haskell_de #learnhaskell should tell me that "there is probably a word 'haskell'"
07:45:21 <nakilon> it's important to not say that "haske is probably a word"
07:45:48 <nakilon> no matter that this substring is even more common, it's not a full word
07:46:46 <nakilon> I've coded some thing, ran it for an hour or two and then saw the results aren't good; I'm missing something
07:47:48 <nakilon> for example, even if there are haskell_es and haskell_en it should not be confused to think that there is a word haskell_e
07:48:20 <nakilon> my approach was going down from the longest substrings and this became a flaw
07:50:08 <riv> that sounds like an interesting problem
07:50:41 -!- imode has quit (Ping timeout: 252 seconds).
07:50:55 <riv> the problem with running LZW on it and then checking frequency in the dictionary is it wouldn't respect word boundaries
07:51:12 <riv> but maybe you could do a modified version that does respect word boundaries
07:52:20 <nakilon> there should be some euristic
07:52:46 <nakilon> something that human intuitively applies to find the common words in such arrays
07:52:53 <riv> i dont know how humans do it
07:53:23 <riv> maybe it could be done with common pairs, 'ha' happens a lot, 'as' happens a lot
08:00:40 <myname> i'd say, strings that end at a word boundary probably are ends of words
08:02:56 -!- riv has quit (Quit: Leaving).
08:03:56 <nakilon> it should find the ends when substrings are concatenated, the '_' is rarely used in my data
08:05:24 -!- hendursa1 has joined.
08:08:38 -!- hendursaga has quit (Ping timeout: 244 seconds).
10:03:02 -!- impomatic has quit (Quit: impomatic).
10:03:19 -!- impomatic has joined.
10:07:34 -!- impomatic has quit (Client Quit).
10:07:50 -!- impomatic has joined.
10:32:32 -!- arseniiv has joined.
10:36:40 -!- jryans has quit (Quit: Bridge terminating on SIGTERM).
10:36:41 -!- Deewiant has quit (Quit: Bridge terminating on SIGTERM).
10:36:41 -!- craigoverend[m] has quit (Quit: Bridge terminating on SIGTERM).
10:36:43 -!- fizzie[m] has quit (Quit: Bridge terminating on SIGTERM).
10:39:39 -!- jryans has joined.
10:47:59 -!- Deewiant has joined.
10:47:59 -!- craigoverend[m] has joined.
10:48:11 -!- fizzie[m] has joined.
11:51:13 -!- SGautam has quit (Quit: Connection closed for inactivity).
12:23:32 -!- wib_jonas has joined.
12:51:01 -!- SGautam has joined.
13:05:29 -!- Sgeo has joined.
13:32:02 -!- simcop2387 has changed nick to mycroft2387.
13:32:37 -!- mycroft2387 has changed nick to simcop2387.
13:46:17 -!- hendursa1 has quit (Quit: hendursa1).
13:46:48 -!- hendursaga has joined.
13:47:40 -!- impomatic has quit (Remote host closed the connection).
13:49:03 -!- impomatic has joined.
13:58:02 -!- impomatic has quit (Quit: impomatic).
13:58:21 -!- impomatic has joined.
14:02:34 -!- impomatic has quit (Client Quit).
14:02:51 -!- impomatic has joined.
14:15:11 -!- impomatic has quit (Remote host closed the connection).
14:16:53 <int-e> "virtualenv" is such a presumptious command name
14:19:03 -!- Koen_ has joined.
14:22:07 <int-e> (python3 -m venv otoh is a bit too obscure)
14:22:40 <fizzie> Reminds me of some other tools where the Debian packages have had to rename the binaries because of excessive genericity. But I can't remember the examples.
14:22:48 <fizzie> "code" is a little bit in that direction.
14:23:15 <wib_jonas> fizzie: "nc". there are two or three versions of netcat, and one norton commander clone with "nc" as the name. and there are some more besides that.
14:23:40 <wib_jonas> I don't recall the others right now
14:23:47 <wib_jonas> I think there were two things called "rename" at some point
14:24:18 <wib_jonas> and it gets worse with clashes with windows commands: "find", "convert", "head" cause common conflicts
14:24:21 <int-e> fizzie: I remember the fight over the 'git' package name and that's not even a common word (though it is a word)
14:24:49 <int-e> Oh yeah, the Northern Commander
14:25:15 <wib_jonas> "find" and "type" are the same style of clash: they both have an ancient DOS meaning (grep and cat respectively) and an ancient unix meaning
14:25:23 <int-e> I'd complain about "type" but "cat" really isn't a better name
14:25:56 <wib_jonas> "convert" is both an obscure DOS utility such that the command name still exists on windows for some reason, and one of the public binaries of ImageMagick and its clones; I usually alias the latter to im-convert
14:26:25 <wib_jonas> "head" is the unix utility conflicting with some HTTP-related thingy that installs binaries named HEAD, GET, POST, I don't recall what
14:27:23 <int-e> some perl thingy... lwp-request
14:27:57 <wib_jonas> oh yeah, "halt". systemd installs a "halt" that does something confusingly different from the traditional "halt", and renames the traditional "halt" to "poweroff". I debugged that one for month, I just couldn't figure out why the new debian couldn't turn off the power on my computer.
14:28:06 <int-e> I use HEAD sometimes, never GET nor POST...
14:28:21 <perlbot> wib_jonas: head http://url/; returns the response code and server type from a HEAD request for a particular url.
14:28:44 <int-e> # cat ~/bin/halt \ #! /bin/bash \ echo try poweroff
14:28:58 <perlbot> wib_jonas: http://ip6.me/api/: 200: OK. Apache/2.4.46 (FreeBSD) OpenSSL/1.1.1d-freebsd
14:29:34 <int-e> Yeah, somehow `halt` leaves the computer switched on in a state that requires a BIOS-level hard or soft reboot (ctrl-alt-del works, IIRC)
14:30:17 <wib_jonas> oh, and "date" and "time" are also DOS vs unix conflicts
14:30:34 <int-e> it's just one of many nuisances that come with systemd
14:30:38 <wib_jonas> though now I just use my datei and dateu aliases on both unix and windows
14:30:46 <int-e> too much magic, really. "where are my core dumps?"
14:32:20 <wib_jonas> then there's "shutdown" and, indeed, "pushd", which are pairs of commands that serve the same main purpose on windows and unix, but are invoked differently
14:32:29 <int-e> (not cluttering the fs with year-old core dumpos is probably an improvement for most users though... but I disabled that thing because the whole journalctl business is too awkward)
14:33:08 <int-e> wib_jonas: This remeinds me of a previous discussion about git's and mercurial's UIs.
14:33:11 <fizzie> Of course Debian also has the "alternatives" system, which I think handles the cases like `nc` and `vi`, where there's several versions of a tool that does similar thing.
14:33:48 <wib_jonas> fizzie: and their new commands, such as sensible-browser
14:33:58 <wib_jonas> I think they have something like interactive-editor too
14:34:36 <wib_jonas> afaik debian isn't using alternatives for netcat though, the two or three different netcat versions just conflict with each other and can't be installed together
14:34:45 <int-e> . o O ( sensible-browser -> /bin/false )
14:35:49 <wib_jonas> it is used for vi and X, and has the result that's hilarious in hindsight only, that if you accidentally uninstall X.org then try to run the startx command, then it does nothing, exits with 0, and gives no error message, because somehow debian alternatives decided that if you have neither X.org or XFree86 installed then true is a suitable fallback
14:35:50 <int-e> eh, the alternatives system isn't bad
14:36:02 <wib_jonas> that's one I only debugged for hours
14:36:06 <fizzie> https://www.debian.org/doc/debian-policy/ch-files.html "Two different packages must not install programs with different functionality but with the same filenames. If this case happens, one of the programs must be renamed. The maintainers should report this to the debian-devel mailing list and try to find a consensus about which program will have to be renamed. If a consensus cannot be reached,
14:36:08 <fizzie> /both/ programs must be renamed."
14:36:34 <fizzie> (If they provide "the same functionality but different implementations", then it's handled via alternatives or "conflicts" dependency entries.)
14:36:50 <wib_jonas> fizzie: yes, and that's the policy currently broken with "netcat". not with "nc", mind you.
14:37:56 <wib_jonas> the two versions of netcat are both imperfect enough that you may want to have both installed to complement each other's mistakes, but you can't, so you just give up and try to learn socat or something
14:39:41 <wib_jonas> (one of them is so old it can't do ipv6; the other one is newer but lost the ability to listen and let the kernel choose the port number)
15:12:22 <int-e> wib_jonas: Oh, checking package names, they are: netcat-openbsd, and netcat-traditional ... so which one is the old one? ;-) (the former knows IPv6 so that answers that question)
15:40:23 <esolangs> [[List of ideas]] https://esolangs.org/w/index.php?diff=86723&oldid=84248 * Orisphera * (+1) /* Ideas related to esoteric operating systems, esoteric processors and esoteric computers */
15:43:21 -!- impomatic has joined.
15:46:33 -!- delta23 has joined.
15:47:34 -!- impomatic has quit (Client Quit).
15:47:51 -!- impomatic has joined.
15:53:02 -!- impomatic has quit (Quit: impomatic).
15:53:19 -!- impomatic has joined.
15:57:35 -!- impomatic has quit (Client Quit).
15:57:52 -!- impomatic has joined.
16:00:36 -!- wib_jonas has quit (Quit: Client closed).
16:06:11 <Corbin> Hm. Brainfuck has the concatenative property: Given any two programs, their juxtaposition is also a program. However, concatenation isn't composition; a wrapper would be needed in order to take outputs from the first program and feed them as inputs to the second program.
16:07:45 <Corbin> If only they composed! But they compose with a wrapper. This reminds me of commutativity; for many things, they don't commute, but they commute with an extra "commutator" term.
16:10:50 -!- riv has joined.
16:43:02 -!- impomatic has quit (Quit: impomatic).
16:43:19 -!- impomatic has joined.
16:47:35 -!- impomatic has quit (Client Quit).
16:47:53 -!- impomatic has joined.
17:33:27 <Corbin> Terminology question: What do we call it when a language turns unknown identifiers into searches on the filesystem? Like, unknown identifier "foo" becomes an attempt to read foo.sourcecode, or "foo/bar" tries to read foo/bar.sourcecode.
17:44:57 <Melvar> Hmm. Apparently on Fedora there are also two versions of netcat. Openbsd netcat, which is exclusively netcat, and nmap’s netcat, named ncat, which is both ncat and nc.
17:48:05 <Melvar> Also the halt manpage claims that poweroff existed before systemd, but that halt was usually a synonym of it. Not sure what actually halting is ever useful for, though.
17:49:28 <b_jonas> Corbin: that is called a shell
17:51:55 -!- impomatic has quit (Quit: impomatic).
17:52:12 <Corbin> b_jonas: Yes, shells have this behavior. But I feel like I've seen it, or variations on it, in other languages. I was wondering if it had a name.
18:00:44 <b_jonas> sure, matlab/octave does it too
18:01:53 <b_jonas> and the rather silly https://metacpan.org/pod/Shell module, which is not really intended to be used seriously.
18:11:53 -!- riv has quit (Quit: Leaving).
18:11:57 -!- impomatic has joined.
18:14:23 -!- riv has joined.
18:21:36 -!- imode has joined.
18:36:25 -!- imode has quit (Quit: WeeChat 3.1).
18:36:41 -!- imode has joined.
18:48:37 <esolangs> [[Minim]] M https://esolangs.org/w/index.php?diff=86724&oldid=86722 * KakkoiiChris * (+1) /* V2 */ Fixed typo
20:00:39 -!- dutch has quit (Quit: WeeChat 3.0.1).
20:02:43 -!- dutch has joined.
20:27:40 -!- impomatic has quit (Ping timeout: 265 seconds).
20:35:43 -!- delta23 has quit (Remote host closed the connection).
20:36:08 -!- delta23 has joined.
20:46:44 -!- impomatic has joined.
20:46:50 -!- impomatic has quit (Remote host closed the connection).
20:47:06 -!- impomatic has joined.
21:44:59 -!- Koen_ has quit (Quit: Leaving...).
21:52:17 -!- impomatic has quit (Quit: impomatic).
21:52:36 -!- impomatic has joined.
21:56:50 -!- impomatic has quit (Client Quit).
21:57:10 -!- impomatic has joined.
21:57:14 -!- arseniiv has quit (Ping timeout: 272 seconds).
23:13:34 -!- impomatic has quit (Quit: impomatic).