←2021-07-19 2021-07-20 2021-07-21→ ↑2021 ↑all
00:01:02 <oerjan> hm i think that's one of the trickier ones, but not unsolvable
00:01:09 * oerjan tries one
00:02:22 <oerjan> as usual, there's some easy catches around the outer rim
00:03:33 -!- vyv has quit (Quit: Konversation terminated!).
00:04:52 <oerjan> next i found some neighboring 4's, which tends to give a lot of information
00:13:29 <oerjan> argh i've made a mistake
00:13:39 <oerjan> (it was going too easily)
00:14:07 <oerjan> now time for the painful backtracking
00:14:25 <Melvar> I haven’t done these in quite a while …
00:15:13 <Melvar> Tried a Florets one, and apparently I still have muscle memory even though I don’t always remember why the patterns are valid.
00:16:54 <Melvar> There’s something with clusters of threes in the florets arrangement where in certain configurations a small inroad can solve most of the cluster.
00:18:26 <oerjan> oh found the error (maybe)
00:20:30 <oerjan> solved
00:22:36 <oerjan> i didn't have any clusters of three 3's
00:24:13 <oerjan> i don't think my algorithm for florets is anything other than the general ones
00:24:42 <oerjan> although there are many general tricks.
00:26:22 <oerjan> sometimes there is parity, which can pass information a long way before it actually gives a definite edge/absense. although for this one, i think it was very local.
00:27:46 <oerjan> *absence
00:29:33 <oerjan> (local being something like: if you have a meeting of three edges where you know from counting the parity of two of them, then you know the last one concretely
00:29:36 <oerjan> )
00:31:08 <oerjan> the thing with the 4s i mentioned is in general "neighboring numbers each one which is exactly one less than the number of surrounding edges"
00:31:23 <Melvar> Yeah, that’s pretty basic.
00:31:41 <oerjan> although in floret it happens for 4s, in some others you can have e.g. one 2 and one 3
00:33:16 <oerjan> snub-square is one that does that
00:38:23 <oerjan> one rule i found which is nice and took me very long to notice: if you have >-< pattern where two neighboring edges of degree 3, and you know that there are exactly 2 edges in the loop in the upper line then there are exactly 2 in the lower as well
00:38:45 <oerjan> *two neighboring vertices have degree 3
00:40:35 <oerjan> also, 0 <-> 0 and 1 <-> 3
00:41:13 <oerjan> but the 2 case is the least obvious, because the reasoning differs for each pattern
00:41:27 <oerjan> oh wait
00:41:46 <oerjan> 3 -> 1 but 1 can be either 1 or 3
00:43:01 <oerjan> i suppose that's also a form of parity, but with a little extra info sometimes
00:43:24 <oerjan> and in some geometries you can get long chains of deductions that way
00:43:50 <oerjan> honeycomb in particular
00:45:52 <oerjan> when you have a row of neighboring hexagons with numbers in them, and know the number in the loop among the three edges at one end
02:01:45 <zzo38> My idea to encode a canonical Huffman tree is: The first bit specifies whether the maximum code length is odd or even, and then the sequence of numbers, which can be abbreviated by omitting the bits that are already known by the time the previous bits have been read (sometimes all bits will be known, in which case that number can be encoded as zero bits).
02:02:59 <zzo38> The sequence of numbers is for each code length, the number of codes of that length, and then the values for the codes of that length, ordered starting with the lowest value of that code length, and then the highest, and then secondly lowest, and then secondly highest, etc.
02:04:35 <zzo38> What would you think of such a thing like this, please?
02:38:59 <int-e> Hmm, https://int-e.eu/~bf3/tmp/loopy.png is something I don't recall doing before: the inner white area mist connect to the outside somehow, but one of the routes (to the top) is blocked by known inner areas and a known line that bridges them... so the connection must go the other way.
02:39:10 <int-e> mist->must
02:40:04 <int-e> (I believe that this can occur in the standard orthogonal version as well, which I've played quite a bit)
02:41:46 <int-e> oerjan: I have used parity reasoning before, counting intersection of the path with some cross section of the whole board.
03:30:41 <oerjan> int-e: interesting, i'm not sure i have _ever_ directly used the fact that the loop must divide the board between an inner and outer part, the windows version i use doesn't show colors like that to make it obvious
03:30:53 <oerjan> jordan curve theorem ftw
03:32:23 <oerjan> is that new, maybe i should upgrade
03:34:45 <oerjan> although i quickly get there with my method by noticing that the line between the remaining 3's must be in the loop
03:36:23 <oerjan> *common border line
03:44:12 <oerjan> ok i'm confused, i seem to see two different solutions
03:46:47 <oerjan> wait duh one doesn't connect the inner white area
03:47:32 <oerjan> which of course i've just claimed i wouldn't use, so how do i solve it...
03:50:24 <oerjan> oh i see. i would detect that the line connecting the right 3 to the outside cannot be in the loop because it would close the outer part of it
03:53:35 <oerjan> which i suppose is a _form_ of "inner and outer areas must connect", but more limited (and perhaps a dual view?)
04:00:57 <oerjan> i suppose loopy's solubility check algorithm might not know about that logic you described, in which case it could not accept a puzzle that _required_ it to solve.
04:02:51 <oerjan> (istr tatham writing somewhere that some of the puzzles cannot generate arbitrary hard instances because they're limited by the solving algorithms included to check that they're solvable)
04:03:16 <oerjan> (which are not always fully general)
04:03:49 <oerjan> and loopy is probably np-complete
04:10:58 <oerjan> s/claims so/has an external link to such a claim/
05:19:31 <esolangs> [[Brainfuck++++++++++++++++++++]] N https://esolangs.org/w/index.php?oldid=86469 * CosmicMan08 * (+4048) Created page with "brainfuck++++++++++++++++++++ (or just bf19) is a joke derivative of [[Brainfuck]] by CosmicMan08#1975 ([[User:CosmicMan08]] meant to parody other brainfuck derivatives that j..."
05:22:13 <esolangs> [[Brainfuck++++++++++++++++++++]] M https://esolangs.org/w/index.php?diff=86470&oldid=86469 * CosmicMan08 * (+1) closed some parenthesis
05:32:10 <esolangs> [[Bf19]] N https://esolangs.org/w/index.php?oldid=86471 * CosmicMan08 * (+83) Redirected page to [[Brainfuck++++++++++++++++++++]]
05:36:01 -!- oerjan has quit (Quit: Nite).
05:38:05 <esolangs> [[Special:Log/newusers]] create * LarhoCherqi * New user account
05:46:45 <esolangs> [[Esolang:Introduce yourself]] M https://esolangs.org/w/index.php?diff=86472&oldid=86371 * LarhoCherqi * (+163)
05:52:24 <esolangs> [[User:LarhoCherqi]] N https://esolangs.org/w/index.php?oldid=86473 * LarhoCherqi * (+1) Created page with "v"
06:31:34 -!- sprock has quit (Quit: brb).
06:32:51 -!- sprock has joined.
06:38:58 -!- Trieste has quit (Ping timeout: 252 seconds).
06:40:51 -!- Trieste has joined.
07:00:55 -!- Sgeo has quit (Read error: Connection reset by peer).
07:55:02 -!- tech_exorcist has joined.
07:59:12 -!- riv has quit (Quit: Leaving).
08:06:42 -!- hendursa1 has joined.
08:10:07 -!- hendursaga has quit (Ping timeout: 244 seconds).
08:29:35 -!- Lord_of_Life_ has joined.
08:30:04 -!- Lord_of_Life has quit (Ping timeout: 256 seconds).
08:30:47 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
09:15:47 -!- arseniiv has joined.
09:58:06 <esolangs> [[Special:Log/newusers]] create * Matthildefuckingslut * New user account
09:58:47 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=86474&oldid=86472 * Matthildefuckingslut * (+62)
10:07:40 -!- Taneb has changed nick to nvd.
10:07:48 -!- nvd has changed nick to Taneb.
10:36:51 <esolangs> [[Esolang:Introduce yourself]] M https://esolangs.org/w/index.php?diff=86475&oldid=86474 * Matthilde * (-1) it's not "an sult" it's a "a sult"
10:53:28 -!- matthilde has joined.
10:53:32 <matthilde> hi
11:11:01 -!- Razetime has joined.
11:27:16 <Razetime> hi esolangers
11:58:38 <int-e> let's see if oerjan still logreads... I added the colors manually for illustration, it's not a feature of the puzzle implementation
12:32:04 -!- tech_exorcist has quit (Quit: tech_exorcist).
12:38:43 <Razetime> puzzle?
12:39:43 <fizzie> Razetime: Context here was https://int-e.eu/~bf3/tmp/loopy.png and the Loopy puzzle. https://logs.esolangs.org/libera-esolangs/2021-07-20.html#lL
13:01:47 -!- Sgeo has joined.
13:09:39 -!- hendursa1 has quit (Quit: hendursa1).
13:10:04 -!- hendursaga has joined.
13:43:35 -!- arseniiv has quit (Ping timeout: 258 seconds).
14:30:05 -!- arseniiv has joined.
14:37:26 -!- archenoth has joined.
14:46:35 <matthilde> .someone has been inpersonating me in the wiki, someone told me to contact you to talk about this
14:46:38 <matthilde> oops
14:46:41 <matthilde> hexchat won't work
14:46:42 <matthilde> aaa
14:47:47 <matthilde> .
14:53:11 <int-e> Hmm, fizzie maybe?
14:53:36 <int-e> matthilde: I think your complaint will be seen, though it may take a while.
14:54:34 <matthilde> ah ok
14:54:36 <matthilde> i was dming ais
14:55:00 <matthilde> in short uh, someone made an account under the name of Matthildef*ckingslut
14:55:16 <matthilde> and added "Hi my name is Matthilde, my site is matthil.de and I am an slut"
14:55:18 <int-e> Yeah, we have a wiki logging bot
14:55:31 <matthilde> when I first seen it, I took it easy and fixed their typo just because lol
14:55:38 <matthilde> but I came here to tell it before it gets out of hand
14:56:00 <matthilde> (if ever it even gets out of hand)
14:56:49 <Taneb> matthilde: do you know why someone would want to inpersonate you in this way?
14:57:34 <matthilde> absolutely not, someone guessed it was because of my sexual orientation but i am not really open about it
14:57:40 <matthilde> I have no conflict with anyone around
14:57:48 <Taneb> It's... not something I can remember seeing anything like on the wiki before
14:58:01 <matthilde> yup
14:58:22 <matthilde> i have no conflict with anyone that's the thing that makes me really confused
14:58:52 <matthilde> even less in the esolangs community
14:59:27 <int-e> Taneb: Well, we've had a few fights. The worst one so far was the ColorfulGalaxy one that spilled over from LifeWiki (as far as I understood)
14:59:53 <matthilde> i don't know what is the lifewiki
15:00:09 <matthilde> oh
15:00:10 <matthilde> ooh
15:00:13 <matthilde> that's the game of life wiki
15:00:15 <int-e> matthilde: It's an unrelated example.
15:00:19 <matthilde> ah ok
15:02:33 <int-e> There's usually the question with trolls who have not really done much yet (yes, they left an insult; we can remove that)... do you give them the satisfaction of getting banned.
15:02:59 <matthilde> i suppose not
15:03:31 <matthilde> idk i just took it easy and played their game by fixing the typo they left
15:05:24 <Corbin> Maybe somebody's irritated at your wiki actions? This sort of random lashing-out was something I saw a lot on English Wikipedia.
15:06:31 <matthilde> huh
15:06:43 <Taneb> Corbin: Matthilde has done not a huge amount on the wiki so I'd be surprised it that was it
15:06:45 <matthilde> lol that's weird cus as of today i only have done contribs in my own pages
15:06:52 <matthilde> i have around 5 contribs or so
15:07:13 <matthilde> i didn't even noticed it by myself, it's a friend of mine that notified me about it
15:07:17 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=86476&oldid=86475 * Int-e * (-61) Remove entry not following the instructions. Oh and it was needlessly offensive too.
15:08:04 <int-e> (entries should be at the end and signed)
15:08:05 <matthilde> that reminds me i should go more often in this wiki, there is very interesting stuff in there
15:08:16 <int-e> <3 having formal reasons for cleanup
15:12:53 <Taneb> I've been thinking for a while about writing an optimizing brainfuck compiler to a) learn how to write a compiler (I've never actually really done it before) and b) see if my idea for recognizing and optimizing loops that calculate a polynomial does any good on real programs
15:16:36 <int-e> b) probably not once it goes beyond multiplication
15:19:04 <int-e> The real challenge would be to identify control flow, and that's pretty hard. Basically you have to reverse engineer the tape layout. And you may have to deal with fancy unbalanced loops (my own terminology; a loop is balanced if its body doesn't change the tape pointer)
15:24:04 <fizzie> I think it's exceedingly unlikely anyone's planning to use a username like that (meaning the one with the insult in it) for contributing positively, so I'd be happy to just block it, though I've no idea whether doing that would have any sort of net positive effect on anything.
15:24:57 <int-e> matthilde: ^
15:26:36 <matthilde> i'll just hope nothing like that will happen in the future
15:26:38 <matthilde> we'll see
15:27:37 <int-e> yeah, if you can, don't overthink it. internet trolls don't need a reason to do something. having an opinion can be enough... a random tweet... an email... whatever.
15:33:02 -!- arseniiv has quit (Ping timeout: 245 seconds).
15:38:32 <Corbin> Huh, Brainfuck gives a Type 2 computer; it takes a stream of natural numbers (a computable real number) and emits another stream.
15:41:13 <matthilde> int-e: I like implementing brainfuck interpreters, more than writing brainfuck programs lol
15:41:29 <matthilde> brainfuck is very simple but can be a challenge to write an optimized interpreter
15:43:40 <Taneb> In university I won a brainfuck programming competition
15:44:21 <Taneb> It's my proudest achievement
15:46:14 <Taneb> (there were two people who entered)
15:48:08 <Taneb> The final challenge was to write an ASCII art banner generator. Someone here (I can't remember who) gave me a simple bitmap font they'd made and I used Haskell as basically a macro language for brainfuck
15:48:16 <int-e> Taneb: 2 people, so it was a competition.
15:48:39 <int-e> Right, I was wondering whether code generation would be involved.
15:49:01 <Taneb> Sadly I have lost both the generating and the generated code
15:49:11 <int-e> The trouble with writing Brainfuck is that it's fun... at small sizes... and gets very tedious very quickly.
15:49:18 -!- matthilde has quit (Read error: Connection reset by peer).
15:49:19 <Taneb> I could probably get the latter by asking the society that ran the competition nicely
15:50:15 <Taneb> int-e: yeah, definitely. I hand-wrote my entry for all the other challenges, though
16:03:18 -!- arseniiv has joined.
16:04:03 -!- matthilde has joined.
16:04:29 <matthilde> Taneb: ooh! nice!
16:05:02 <matthilde> i ended up 2nd on some esoteric programming language competition
16:05:16 <matthilde> basically we had to write an esolang that includes turtles
16:05:28 <matthilde> so i wrote turtleC
16:05:43 <matthilde> (cursed)
16:06:49 <matthilde> https://git.unix.lgbt/matthilde/turtlec/src/branch/master/programs/rpn.trt
16:06:57 <matthilde> an rpn calculator written in this language
16:21:48 <Razetime> the first thing that comes to mind when anyone says turtle is the graphics lib
16:28:34 -!- riv has joined.
16:34:48 -!- tech_exorcist has joined.
16:40:20 <riv> I highly recommend this
17:14:57 -!- Razetime has quit (Remote host closed the connection).
17:26:53 <esolangs> [[Woodchuck]] M https://esolangs.org/w/index.php?diff=86477&oldid=86455 * Rdococ * (-42)
17:27:17 <esolangs> [[Woodchuck]] M https://esolangs.org/w/index.php?diff=86478&oldid=86477 * Rdococ * (-7) /* Introduction */
18:09:30 -!- hanif has joined.
18:10:28 <hanif> 22:14:07 <riv> https://nickdrozd.github.io/2021/07/11/self-cleaning-turing-machine.html => cool stuff, thanks for sharing
18:26:53 -!- hanif has quit (Quit: quit).
18:40:20 -!- tech_exorcist has quit (Quit: tech_exorcist).
21:11:16 -!- user3456 has quit (Quit: ZNC - https://znc.in).
21:16:59 -!- user3456 has joined.
21:28:14 -!- archenoth has quit (Quit: Leaving).
21:33:01 -!- archenoth has joined.
21:35:23 <esolangs> [[Asm2bf]] M https://esolangs.org/w/index.php?diff=86479&oldid=82677 * Challenger5 * (-108) delete repeated sentence
22:11:30 -!- arseniiv has quit (Ping timeout: 258 seconds).
22:59:35 -!- matthilde has quit (Quit: bye bye owo).
23:21:20 -!- dutch has quit (Quit: WeeChat 3.2).
23:27:54 -!- dutch has joined.
23:29:17 -!- Lord_of_Life has quit (Ping timeout: 255 seconds).
←2021-07-19 2021-07-20 2021-07-21→ ↑2021 ↑all