00:17:03 -!- Lord_of_Life has quit (Read error: Connection reset by peer).
00:19:13 -!- Lord_of_Life has joined.
00:51:52 -!- Guest26nakilon has joined.
00:58:10 -!- Oshawott has joined.
01:01:36 -!- archenoth has quit (Ping timeout: 272 seconds).
02:05:42 -!- Guest26nakilon has quit (Quit: Client closed).
03:45:56 -!- dyeplexer has joined.
04:38:49 -!- arseniiv has joined.
06:12:34 -!- tromp has joined.
06:32:02 -!- sprock has quit (Quit: brb).
06:40:30 -!- sprock has joined.
06:55:26 -!- sprock has quit (Quit: brb).
06:59:30 -!- tromp has quit (Quit: My iMac has gone to sleep. ZZZzzz…).
07:00:01 -!- sprock has joined.
08:00:49 <shachaf> int-e: I had another application for distinguished points: Say you have a permutation f on 2^64 with a single cycle, and a starting point x, and f^n(p), and you want to find n. Naively it might seem like you have to do it serially, but you can just start at various points and build a map of the subgraph of distinguished points.
08:01:34 <shachaf> This is similar to a problem I was actually trying to figure out, though that one turned out to be much more easily parallelizable.
08:07:14 -!- hendursa1 has joined.
08:07:40 <esolangs> [[Ascii]] M https://esolangs.org/w/index.php?diff=85655&oldid=85654 * Dominicentek * (+1) Fixed a typo
08:08:54 -!- hendursaga has quit (Ping timeout: 244 seconds).
08:16:58 -!- Lord_of_Life_ has joined.
08:20:00 -!- Lord_of_Life has quit (Ping timeout: 258 seconds).
08:20:00 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
08:26:16 -!- Sgeo has quit (Read error: Connection reset by peer).
08:31:24 -!- nakilon has joined.
08:38:37 -!- nakilon has quit (Quit: rasengan did nothing wrong).
08:39:22 -!- nakilon has joined.
09:00:49 -!- delta23 has joined.
09:25:29 -!- nakilon has quit (Quit: rasengan did nothing wrong).
09:28:28 -!- nakilon has joined.
09:29:22 <nakilon> after two undocumented manipulations
09:30:45 <nakilon> 1. after https://docs.docker.com/engine/security/userns-remap/ you have to chown -R the mounted volume to that fake uid otherwise OS does not understand what user is trying to access it
09:31:52 <nakilon> 2. there are dozens of questions why dockerized nginx does not use the docker built-in 127.0.0.11 resolver so you have to hardcode it in nginx conf location directive
09:32:16 <nakilon> no one knows the answer why
09:35:51 -!- nakilon has quit (Quit: rasengan did nothing wrong).
09:36:50 -!- nakilon has joined.
09:41:15 <nakilon> you know how to understand that you should start closing tabs in chrome?
09:41:34 <nakilon> it's when you pin the tab and instead of becoming narrower it becomes wider
10:23:09 <int-e> "When a process accesses a file, its user and group IDs are mapped into the initial user namespace for the purpose of permission checking and assigning IDs when creating a file. When a process retrieves file user and group IDs via stat(2), the IDs are mapped in the opposite direction, to produce values relative to the process user and group ID mappings."
10:23:16 <int-e> (man user_namespaces)
10:24:52 <int-e> "This re-mapping is transparent to the container, but introduces some configuration complexity in situations where the container needs access to resources on the Docker host, such as bind mounts into areas of the filesystem that the system user cannot write to." is thoroughly uninformative though.
10:30:23 <nakilon> aha, "some configuration complexity" that we won't explain how to solve
10:30:27 <int-e> shachaf: Right. Rainbow tables work for permutations as well as for pseudo-random functions.
10:31:04 <int-e> Oh and near the end, there's this pseudo-sentence: "For example, if volumes are mounted from the host, file ownership must be pre-arranged need read or write access to the volume contents."
10:31:28 <int-e> (Something went off the rails with the grammar there.)
10:51:04 -!- velik has quit (Remote host closed the connection).
11:28:36 -!- oerjan has joined.
11:40:38 -!- arseniiv has quit (Ping timeout: 272 seconds).
12:01:43 -!- arseniiv has joined.
12:18:41 -!- velik has joined.
12:41:16 -!- queeries has changed nick to Soni.
12:52:53 -!- dyeplexer has quit (Ping timeout: 265 seconds).
13:01:03 <arseniiv> I read somewhere about a neat (? you tell me) format for approximate rationals which represents rational numbers with an error as pairs of rational numbers of bounded denominators: lesser the error, lesser the bound. It was claimed this way is better for some things but I forgot which. What do you think, and have you encountered something like that?
13:01:30 <arseniiv> pairs are lower and upper bound to the number thus represented
13:04:00 <arseniiv> obviously, one pro is that you can have as small an error as you want, paying with size of denominators and numerators. Hm I need to remember where did I read the specifics which problem was the author solving with this format
13:05:40 <arseniiv> OTOH the arithmetic might end up slow?.. Not just due to occasional bignums, just due to software implementation of this thing vs. for instance hardware floating-point
13:16:14 <Corbin> Can't help but think of continued fractions. There's an upper and a lower bound, and each coefficient alternates between improving the upper bound and lower bound.
13:20:21 <nakilon> this says there is something called Raku that uses "limited-precision fractional" but it's just fixed bit length limit https://en.wikipedia.org/wiki/Rational_data_type#Language_support
13:21:39 <nakilon> yeah something like https://en.wikipedia.org/wiki/Continued_fraction#Best_rational_approximations
13:22:59 -!- dyeplexer has joined.
13:24:46 <arseniiv> Corbin: ah yes, connecting those might be interesting
13:26:51 <arseniiv> nakilon: yeah, continued fractions are usually used for approximations, though there are some other uses I don’t remember right now. There is even a good arithmetic with them when implemented as iterators, which uses several related Möbius transformation neat things
13:28:03 <nakilon> IIRC projecteuler had a lot of problems with continued fractions
13:28:07 <nakilon> I was just skipping them lol
13:35:15 <int-e> Has anyone tried to use the github copilot for projecteuler? Also, has the copilot be sued for copyright infringement yet?
13:36:44 <nakilon> wow, automatized bug copypaster
13:37:14 <int-e> There are some egregious examples around... https://nitter.fdn.fr/mitsuhiko/status/1410886329924194309#m is an unambiguous copyright violation, and https://gist.github.com/0xabad1dea/7026665beb6de41c17c89beebfb1bb90 is ridiculous on another level (ask yourself: would you rather debug that code or start from scratch?)
13:46:44 <int-e> Heh, imagine the warning label "may contain code produced by AI"
13:55:14 <riv> today I learned that (x + y) - y can be 1 or 2 bits off in floating point
13:55:22 <riv> possibly more bits off depending on the size of the numbers
13:55:33 <riv> floating point is such a great hack
13:55:39 <riv> who would have thought of this
13:55:58 <riv> making a floating point number type that somehow works well for almost all applications
13:56:18 <riv> i think lua uses floating points for tagged unions
13:56:26 <riv> explointing the many NaNs
13:57:46 <riv> lol @ copilot stuff
14:09:05 <oerjan> > (0.1 + 1e100) -1e100
14:09:55 <oerjan> riv: it's pretty obvious that you can't keep anything with a big enough size difference
14:11:15 <oerjan> 1 or 2 bits sounds about right if they are almost the same size
14:11:43 <int-e> > (44.614 + 2^52) - 2^52
14:12:19 <riv> is that a trick to round a number to integer
14:12:32 <riv> > (1231231.122 + 2^52) - 2^52
14:12:36 <riv> > (1231231.722 + 2^52) - 2^52
14:13:12 <int-e> yes, with some severe limitations
14:14:30 <int-e> (basically it's restricted to the [0,2^52] range of numbers, and it depends on the current FPU rounding mode)
14:14:51 <int-e> and of course it's only for binary IEEE double precision floating point numbers
14:15:05 <int-e> which have a 52 bit mantissa
14:15:08 <oerjan> perhaps it technically works for > 2^52 as well?
14:15:17 <oerjan> (because those _are_ integers)
14:15:34 <Corbin> Note that Kahan has a classic technique for extending addition to work cleanly throughout the range of floats: https://en.wikipedia.org/wiki/Kahan_summation_algorithm
14:17:51 -!- Cale has quit (Remote host closed the connection).
14:18:59 -!- hanif has joined.
14:19:27 <oerjan> (2^52 + 1 + 2^52) - 2^52
14:19:33 <oerjan> > (2^52 + 1 + 2^52) - 2^52
14:19:42 -!- Cale has joined.
14:20:52 -!- hanif has quit (Remote host closed the connection).
14:21:13 -!- hanif has joined.
14:22:16 <int-e> > (2^52 + 1.0 + 2^52) - 2^52 == 2^52 + 1
14:33:40 -!- riv has quit (Quit: Leaving).
14:35:54 <nakilon> > (2^52 + 1 + 2^52) - 2^52
14:36:05 <nakilon> > (2^52 + 1.0 + 2^52) - 2^52
14:36:31 <nakilon> > (2^52 + 1 + 2^52) - 2^52 - (2^52 + 1)
14:39:06 <oerjan> too long since i've done haskell
14:39:10 <int-e> > (2^52 + 1 + 2^52) - 2^52 - (2^52 + 1) :: Double
14:40:19 <int-e> writing 1.0 is a cheap way to get a Double in many cases (by default type rules)
14:42:35 <nakilon> why does it says it's free? https://www.wolfram.com/engine/
14:44:54 <int-e> because you don't have to pay for it as long as you don't use it for anything useful?
14:46:11 <nakilon> Mathematica never was free
14:46:25 <nakilon> looks like now the GUI is what you pay for
14:47:39 <nakilon> oh damn, it just gives you two keys that have to be validated with website
14:53:10 -!- Trieste has quit (Ping timeout: 272 seconds).
15:15:24 -!- hanif has quit (Remote host closed the connection).
15:15:47 -!- hanif has joined.
15:15:50 -!- tromp has joined.
15:25:28 -!- tromp has quit (Read error: Connection reset by peer).
15:53:10 <b_jonas> will there be an ICFP contest 2021?
15:58:58 -!- hendursa1 has quit (Quit: hendursa1).
15:59:50 -!- hendursaga has joined.
16:03:57 <Taneb> b_jonas: I think I heard there would be
16:04:10 <Taneb> https://icfpcontest2021.github.io/ next weekend
16:08:36 <b_jonas> oh! so they didn't get the usual domain name http://icfpcontest.org/ for some reason?
16:11:06 -!- perlbot has quit (Ping timeout: 240 seconds).
16:12:18 -!- simcop2387 has quit (Ping timeout: 268 seconds).
16:17:09 -!- hanif has quit (Ping timeout: 244 seconds).
16:19:42 -!- perlbot has joined.
16:21:43 -!- simcop2387 has joined.
16:25:33 -!- hanif has joined.
16:25:51 -!- hendursaga has quit (Remote host closed the connection).
16:27:26 -!- hendursaga has joined.
16:39:53 <nakilon> it was github.io the last year too IIRC
16:41:47 <^[> nakilon: Raku is Larry Wall's latest esolang.
16:45:03 -!- hanif has quit (Ping timeout: 244 seconds).
16:53:55 -!- Cale has quit (Remote host closed the connection).
16:55:32 -!- Cale has joined.
16:56:07 -!- riv has joined.
17:22:25 <esolangs> [[Special:Log/newusers]] create * OrichalcumCosmonaut * New user account
17:28:51 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=85656&oldid=85586 * OrichalcumCosmonaut * (+313) /* Introductions */
18:14:48 -!- dyeplexer has quit (Remote host closed the connection).
18:19:08 -!- velik has quit (Remote host closed the connection).
18:20:44 -!- velik has joined.
18:44:36 <esolangs> [[Ascii]] M https://esolangs.org/w/index.php?diff=85657&oldid=85655 * Rphii * (+128) cleaned it up a little bit
18:48:11 <esolangs> [[Ascii]] M https://esolangs.org/w/index.php?diff=85658&oldid=85657 * Rphii * (+52) /* Initialization */ clean-up
18:49:13 -!- Sgeo has joined.
19:19:03 <velik> "[ echo \ruby 7+2"
19:19:07 <b_jonas> \ruby print %"[ echo \ruby 7+2"
19:19:10 <velik> [ echo [CR]uby 7+2nil
19:19:37 <b_jonas> \ruby print %q"[ echo \ruby 7+2"
19:19:40 <velik> [ echo \ruby 7+2nil
19:19:49 <b_jonas> \ruby print %q"[ echo \ruby 7+";2
19:46:19 <arseniiv> <riv> i think lua uses floating points for tagged unions => cool
19:47:08 <imode> floating points for tagged unions?
19:47:58 <riv> https://anniecherkaev.com/the-secret-life-of-nan
19:49:03 <imode> if a signal is available, it'll be used..
19:53:10 <keegan> I have some Arduino libraries that return NaN for "sensor not found" which I find rather distasteful, but such is life
19:53:51 <int-e> no available number
19:54:20 -!- chrysanthematic has joined.
19:56:59 <fizzie> In Chicken, (number? +nan.0) => #t, which makes sense but just feels a bit off anyway.
20:03:35 <fizzie> fungot: Do you use any floats?
20:04:03 <int-e> fungot: do you prefer rafts?
20:04:03 <fungot> int-e: the abbreviation is not really a simple macro to write if or cond expressions.
20:11:46 <int-e> fungot: what will you do once you reach the floating point?
20:11:46 <fungot> int-e: if you're just talking nonsense now to fuck with their brains", right?") and it only needs some way of specifying arguments.
20:15:43 <fungot> oerjan: hmm what's fgk?
20:16:45 <int-e> oerjan: I know right? it's uncanny.
20:16:52 <fizzie> fungot: Fachverband Gebäude-Klima e. V. hth
20:16:52 <fungot> fizzie: i've gone in a short function of 3 lines can be one line :)
20:17:11 <int-e> Wow, air conditioning.
20:18:10 <int-e> fungot: was that a zzo38 quote...
20:18:11 <fungot> int-e: got a link to the user, or something... how can you beat that?)? i'm using a design of my esolang
20:18:33 <HackEso> zzo38 is not actually the next version of fungot, much as it may seem.
20:22:35 <shachaf> int-e: People seem to be into rafts and they always talk about how Paxos is complicated, but I don't think it's so bad.
20:25:03 <int-e> What is... oh... well, I didn't consent to be drawn into a distributed computing discussion.
20:25:26 <int-e> All I wanted was to go with the flow.
20:27:54 <esolangs> [[Special:Log/newusers]] create * Sebosh * New user account
20:29:11 <int-e> fungot: do you think people should figure out two-party consent first?
20:29:11 <fungot> int-e: i was quite surprised, when i were to use any of those so i guess the gcc version fnord an implementation of syntactic closures ( mit scheme)
20:29:49 <int-e> fungot: you lost me there
20:29:49 <fungot> int-e: and thanks for the ( 1) ( b1. 2))
20:29:58 <int-e> fungot: fine, be that way
20:29:59 <fungot> int-e: like funge98 ' ( ' foobar)? i'd like to
20:30:03 -!- delta23 has quit (Quit: Leaving).
20:30:04 <shachaf> I don't think consent and consensus are really the same thing.
20:30:52 <int-e> oh but there's a significant overlap
20:35:56 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=85659&oldid=85656 * Sebosh * (+219)
20:37:04 <nakilon> \ruby eval's,m="",Hash.new(p=0);'+'+[-[<<[+[--->]-[<<<]]]>>>-]>-.---.>..>.<<<<-.<+.>>>>>.>.<<.<-.'.gsub(/./,?>=>'p+=1;',?<=>'p-=1;',?+=>'m[p]+=1;',?-=>'m[p]-=1;',?[=>'(',?]=>')until((m[p]&=255)==0);',?.=>'s<<m[p].chr;')+';fail s'
20:38:04 <nakilon> (based on https://gist.github.com/rdebath/3060affe89a739cf8624)
20:43:21 <esolangs> [[Language list]] https://esolangs.org/w/index.php?diff=85660&oldid=85647 * Sebosh * (+14)
20:45:02 <nakilon> "Please link only to existing article pages on this wiki.
20:47:57 <esolangs> [[Velik]] M https://esolangs.org/w/index.php?diff=85661&oldid=84306 * Nakilon * (+72) added \ruby invocation example
20:49:24 -!- oerjan has quit (Quit: Nite).
21:00:34 -!- chrysanthematic has quit (Quit: chrysanthematic).
21:38:46 <esolangs> [[FIFOrth]] N https://esolangs.org/w/index.php?oldid=85662 * Sebosh * (+453) Created page with "==FIFOrth== FIFOrth is my first esolang and it is basically Forth but with a queue (FIFO) instead of a stack (LIFO). I am currently working on an interpreter, which I'll link..."
21:38:50 <esolangs> [[Special:Log/upload]] upload * Zero player rodent * uploaded "[[File:Eeklogo.png]]": The logo for Eek!
21:39:26 <esolangs> [[FIFOrth]] https://esolangs.org/w/index.php?diff=85664&oldid=85662 * Sebosh * (+8)
21:40:39 <esolangs> [[Eek!]] https://esolangs.org/w/index.php?diff=85665&oldid=85525 * Zero player rodent * (+4)
21:48:20 <int-e> https://nitter.fdn.fr/MalwareJake/status/1411351168643706886#m is an excellent idea.
22:06:35 -!- arseniiv has quit (Ping timeout: 268 seconds).
23:06:34 -!- Noisytoot has changed nick to N7t.
23:10:12 -!- N7t has changed nick to Noisytoot.
23:13:15 -!- Lykaina has joined.
23:15:31 <Lykaina> just curious: is there still an #esoteric on freenode? i have no intention of connecting there to find out, so i ask.
23:32:48 <fizzie> Not registered, though.
23:33:01 <fizzie> (And with no ops or topic set.)
23:34:24 <fizzie> And there's been a total of one "conversation" after I stopped the logs, and that was about where everyone went (here).
23:34:38 <fizzie> So I guess "no" would be a perfectly arguable answer too.
23:40:49 -!- Lykaina has quit (Quit: Leaving).
23:58:22 <esolangs> [[FIFOrth]] M https://esolangs.org/w/index.php?diff=85666&oldid=85664 * PythonshellDebugwindow * (+31) Work in language