00:33:54 -!- mla has joined.
01:01:19 <esolangs> [[Stupidbf]] N https://esolangs.org/w/index.php?oldid=86783 * CosmicMan08 * (+1504) Created page with "Stupidbf is a joke derivative of [[Brainfuck]] by CosmicMan08#1975 ([[User:CosmicMan08]]). it sucks lmao == instructions i guess == {| class="wikitable" |- ! Instruction !!..."
01:01:41 <esolangs> [[Stupidbf]] https://esolangs.org/w/index.php?diff=86784&oldid=86783 * CosmicMan08 * (+27)
01:05:03 <esolangs> [[Stupidbf]] https://esolangs.org/w/index.php?diff=86785&oldid=86784 * CosmicMan08 * (-1497) Redirected page to [[Language of Laughing]]
01:05:06 <esolangs> [[Language of Lauging]] N https://esolangs.org/w/index.php?oldid=86786 * CosmicMan08 * (+1543) Created page with "Language of Laughing is a joke derivative of [[Brainfuck]] by CosmicMan08#1975 ([[User:CosmicMan08]]). it sucks lmao == instructions i guess == {| class="wikitable" |- ! Ins..."
01:06:24 <esolangs> [[Language of Laughing]] N https://esolangs.org/w/index.php?oldid=86787 * CosmicMan08 * (+33) Redirected page to [[Language of Lauging]]
01:14:24 -!- oerjan has joined.
01:27:27 -!- orichalcumcosmon has joined.
01:28:12 -!- mnrmnaugh has left (Leaving).
01:44:02 -!- mnrmnaughmnrgle has joined.
01:47:03 <oerjan> `learn The password of the month is too long for this irc message
01:47:07 <HackEso> Relearned 'password': The password of the month is too long for this irc message
01:48:58 <oerjan> ( arseniiv spoke about quasi-Fermatic passwords earlier )
01:50:18 <oerjan> hm this would have been a good use for `# , but of course i never remember that until afterwards
01:50:59 <HackEso> `# <comment>//`<command> is useful if you want to add a comment to HackEgo history for things like `sled or `le/rn.
02:02:38 <int-e> we should get fizzie to add links from the hackeso repo history pages to the IRC pages (not serious, no clue how feasible that is, but it'd require hacking hgweb for sure)
02:04:06 <fizzie> Hmm. If it's just based on commit timestamp to the date, that'd probably be reasonably easy. If you want it to link to the correct line's anchor, that's probably a little trickier.
02:05:34 <int-e> timestamp is probably fine, especially since the change might have happened off channel
02:05:58 <int-e> but eh... it's really not that hard to do that manually. a bit tedious, sure.
02:07:54 <fizzie> Yeah. I guess it'd actually be more than a "little" tricky to link to the exact line fully reliably, because HackEso and the logging are two separate clients, so they're not guaranteed to see messages in the same order.
02:15:40 <oerjan> you could compare the command lines, although there is some escaping i think
02:19:21 <oerjan> of course the logs can't be seen from inside HackEso
02:22:39 <oerjan> there is also a slight theoretical ambiguity because `run and plain ` are both stripped and not distinguished iirc
02:24:31 <HackEso> Sun Aug 1 02:24:30 UTC 2021
02:27:27 <HackEso> 12414:2021-08-01 <oerjän> learn The password of the month is too long for this irc message \ 12408:2021-07-01 <arseniïv> learn The password of the month is tempting in retrospect \ 12397:2021-06-01 <int-̈e> learn The password of the month is moving to Libera Chat. \ 12377:2021-05-08 <int-̈e> learn The password of the month is heavily guarded. \ 12374:2021-04-06 <int-̈e> learn The password of the month is hiding in plain sight. \ 12372:2021-03-04 <in
02:33:24 <oerjan> HackEso could probably collect all the info and build a url, and then there could be a web page that decoded it and searched to the right place. i presume. i don't know the web side of things.
02:46:49 <b_jonas> ` have you changed the error message for this?
02:47:13 <b_jonas> in what way is plain ` "stripped"?
02:48:32 <b_jonas> oh, you mean when HackEso records the commands in its hg log? or some internal log for fizzie when they don't modify anything?
02:53:59 <oerjan> the former. how would i know about fizzie's internal logs?
02:54:27 <fizzie> Yes, I think the hg commit message isn't entirely unambiguous.
02:55:08 <oerjan> although `run is pretty rare these days.
02:58:15 <b_jonas> oerjan: fizzie might tell about them, or you might have read the source code
02:59:02 <b_jonas> I almost never use `run but I *should*. it's an easy way to avoid rnoodl. ideally I should reimplement rnoodl to make it less mentally annoying, but I'm lazy.
03:00:10 <b_jonas> though part of the problem is that even after I reimplement it, I'll still hate it, and want it not to be present in commands that I run
03:01:04 <HackEso> IRC_NICK=oerjan \ PWD=/hackenv/tmp \ HACKENV=/hackenv \ HOME=/tmp \ LANG=en_NZ.UTF-8 \ IRC_NETWORK=Libera.Chat \ IRC_COMMAND=PRIVMSG \ IRC_IDENT=oerjan \ TERM=linux \ IRC_INSTANCE=HackEso-libera \ SHLVL=1 \ http_proxy=http://127.0.0.1:3128 \ IRC_HOST=sprocket.nvg.ntnu.no \ IRC_TARGET=#esolangs \ PATH=/hackenv/bin:/usr/bin:/bin \ IRC_MESSAGE=`` printenv \ _=/usr/bin/printenv
03:01:28 <HackEso> cat: /hackenv/bin/rnoodl: No such file or directory
03:01:35 <HackEso> perl -pe 's/([Nn])ooodl/"$1@{[o x(3+rand 7)]}dl"/ge'
03:02:42 <b_jonas> I'm fine with wisdom and \? running rnoodl to be clear, it's just \` and \`\` where I think it shouldn't run
03:04:21 <b_jonas> if you want to balance it out, make quote run rnoodl
03:05:06 <b_jonas> admittedly that would lead to a nested rnoodl syndrome, similar to the leaning toothpick syndrome, where some quotes contain lines that HackEso said and were already rnoodled
03:05:53 <b_jonas> rnoodl only affects exactly three os, not more than three
03:07:14 <b_jonas> the part of rnoodl that I hate is that it buffers too much. if a command ran with backtick prints a partial line then times out, you won't see that output, because rnoodl swallows it. there is never any reason for rnoodl to not print any input immediately, except sometimes the single last character when it just read something ending in "nood".
03:07:39 <b_jonas> something ending in "noood", sorry
03:08:08 <b_jonas> so a good implementation of rnoodl should write immediately after a read, except for that last "d", which it should ideally write after a short timeout
03:09:16 * oerjan goes to get some chocolate
03:12:13 <b_jonas> I did consider just modifying \` or \`\` to not run rnoodl, but I think that would be reverted quickly. I did consider modifying them to not run rnoodl if I (b_jonas) am running the command, but that would be reverted too, perhaps slightly less quickly. I even considered modifying \` or \`\` to not run rnoodl if I'm running it and hiding my tracks by modifying cbt and creating /bin/cat such that they
03:12:19 <b_jonas> both write a fake output if you cbt \` or cbt cbt or cbt cat, probably with some other extra behavior in cat to justify why I created it, and that would take much longer to discover, but after that you'd be much angrier with me.
03:13:40 <b_jonas> so my best option is probably to make a good implementation for rnoodl, and also modify *that* to exec cat when I'm running the command, or perhaps exec cat when I'm running the command and the command is \` rather than \q
03:13:56 <b_jonas> that would probably not get reverted, but it requires work
03:17:20 <esolangs> [[Special:Log/newusers]] create * Lexinathan * New user account
03:18:57 <b_jonas> of course I'm never really sure about what would get reverted, and the general problem of what gets reverted on a wiki and what gets to stay is probably politics-complete, i.e. harder than what humans can currently do
03:19:52 -!- delta23 has quit (Quit: Leaving).
03:20:13 <b_jonas> if you had repeated calls to an oracle that could predict if a change gets reverted on a wiki, you could probably use that to find optimal solutions to any political problem, like what the heck to do with the middle east
03:21:33 <b_jonas> well no, probably not find *optimal* solutions, just find solutions exponentially close in value to the optimal
03:23:18 <esolangs> [[Esolang:Introduce yourself]] M https://esolangs.org/w/index.php?diff=86788&oldid=86778 * Lexinathan * (+239) /* Introductions of Lexinathan */
03:23:21 <fizzie> What I've been thinking I should do is to make the log web pages produce proper cache validation headers (ETag and/or Last-Modified), because currently they contain none, but the content of past days never changes. Except (and that's probably the reason why I haven't done it yet) if I change something in the formatting.
03:24:22 <fizzie> Here's something I can never remember: which one out of the `-i` and `-I` flags of curl is the one that does just a HEAD request, vs. the one that does a normal GET request but also prints the headers as part of the output. There's probably some kind of a mnemonic for it.
03:26:38 <fizzie> Yes, I mean, I can look it up and/or just determine it empirically, the problem is remembering it the next time.
03:26:49 <fizzie> Maybe the long flags would be more memorable.
03:27:06 <fizzie> Though I'm quite capable of remembering that -i and -I do those two things, just not which one is which.
03:27:15 <esolangs> [[User:Lexinathan]] N https://esolangs.org/w/index.php?oldid=86789 * Lexinathan * (+107) me
03:29:21 <oerjan> "but that would be reverted too" <-- ye of little faith, we'd just expand it to read share/rnooodlhaters hth
03:34:12 <b_jonas> fizzie: -i for "include" header; -I is an uppercase letter to set the HTTP method, and http methods are usually written in uppercase
03:35:47 <b_jonas> -G means to set the method to GET; POST is the default if you give a body to post, and you can't post with no body (you can post with empty body but that's different); but -H was already used for something more important, so it's -I
03:36:24 <oerjan> <b_jonas> if you had repeated calls to an oracle that could predict [...] <-- this seems related to option 3 here https://astralcodexten.substack.com/p/updated-look-at-long-term-ai-risks
03:37:18 <b_jonas> fizzie: alternately, don't try to remember -i at all, instead try to remember -D to dump the header, but remember -D to print headers to a separate file, and use -D - instead, or -D with the same filename as where the normal output goes
03:37:31 <b_jonas> -D - instead, and redirect to a file if you wish
03:37:56 <b_jonas> you rarely want to include headers in the same stream when you write a script, and when you just examine the output then -D - ... | less will work
03:40:30 <b_jonas> but you do sometimes want headers in a separate file when you're writing a script, so remembering -D in general is more useful
03:41:50 <b_jonas> in practice, I remember almost none of those options, not -I, not -i, I look up everything in the manual except maybe sometimes -sS or whatever that pair of options is to mute the default performance info output on stderr when I'm writing to a file but still print errors
03:42:01 <b_jonas> I don't even know if it's -sS or -sq or -qQ or whatever
03:42:17 <b_jonas> yeah, looked it up, it's -sS
03:42:25 <b_jonas> curl is hard to use without looking up
03:43:05 <b_jonas> ideally I should implement a better downloading tool that does exactly what I need -- I have written one, wgetas, but that was like twelve years ago, the program sucks, my needs changed, etc
03:43:24 <b_jonas> and if I write the program, then I will set its syntax to something I remember
03:44:00 <esolangs> [[User:OrichalcumCosmonaut]] M https://esolangs.org/w/index.php?diff=86790&oldid=86762 * OrichalcumCosmonaut * (-18) pronoun.is has an ?or= query parameter
04:13:54 -!- dermato has quit (Ping timeout: 272 seconds).
06:13:36 <esolangs> [[Minim]] https://esolangs.org/w/index.php?diff=86791&oldid=86759 * KakkoiiChris * (+114) /* Operators */ Added increment and decrement operators to precedence tables
06:14:53 -!- Lord_of_Life has quit (Read error: Connection reset by peer).
06:17:30 -!- Lord_of_Life has joined.
06:22:31 -!- Lord_of_Life_ has joined.
06:23:25 -!- Lord_of_Life has quit (Ping timeout: 258 seconds).
06:23:45 -!- Lord_of_Life_ has changed nick to Lord_of_Life.
06:30:43 <zzo38> Now I wrote a PostScript code for parsing command-line switches, like getopt but always like POSIXLY_CORRECT mode. So, you can write: /Verbose false def << /v [/N {pop /Verbose true def}] >> (getopt.ps) run Verbose ==
06:35:36 <zzo38> http://sprunge.us/qT0uzl
07:17:28 -!- orichalcumcosmon has quit (Read error: Connection reset by peer).
07:18:37 -!- orichalcumcosmon has joined.
07:37:51 -!- orichalcumcosmon has quit (Read error: Connection reset by peer).
07:38:02 -!- orichalcumcosmon has joined.
08:03:51 -!- imode has quit (Ping timeout: 265 seconds).
08:08:10 -!- hendursa1 has joined.
08:11:04 -!- hendursaga has quit (Ping timeout: 244 seconds).
08:31:01 -!- Koen_ has joined.
09:21:17 -!- oerjan has quit (Quit: Later).
09:31:16 -!- Sgeo has quit (Read error: Connection reset by peer).
09:33:18 <nakilon> https://i.imgur.com/Gpi3LQn.png
09:33:47 <nakilon> found this edit in my library README dated to Oct 19 2020
09:35:45 <nakilon> and I have no idea what the hell is the bug there
13:29:44 -!- hendursa1 has quit (Quit: hendursa1).
13:32:56 -!- hendursaga has joined.
13:44:54 -!- Koen_ has quit (Remote host closed the connection).
14:04:48 -!- SGautam has joined.
14:09:20 -!- arseniiv has joined.
14:18:41 <esolangs> [[Special:Log/newusers]] create * R3D * New user account
14:28:27 <esolangs> [[Talk:OISC]] https://esolangs.org/w/index.php?diff=86792&oldid=32131 * Rdococ * (+367)
14:29:06 <esolangs> [[Talk:OISC]] M https://esolangs.org/w/index.php?diff=86793&oldid=86792 * Rdococ * (+3)
14:34:44 <riv> are there any AIs based on bayes law?
14:37:59 <Corbin> I'm imagining that, given a function which we are trying to optimize, we could estimate intervals around the minima. The edges of these intervals (their "boundary") would have a gradient based on the function's derivative.
14:38:47 <riv> the AI should make decisions based on logic and bayes law, to pick the best action
14:38:48 <Corbin> Given an evaluation of the function within any particular interval, we can improve our knowledge of that interval's minimum, and tighten the bounds.
14:39:21 <Corbin> This connects Bayes' law (information about minima) with differentiation-oriented optimization (information about gradients near minima).
14:39:55 <nakilon> looks like I finished the thing that finds the words in channel names
14:40:22 -!- imode has joined.
14:40:30 <riv> nakilon, can you show the code and explain it
14:41:05 <riv> Corbin, what about if everything was over finite sets?
14:41:16 <riv> then i wont have derivatives
14:41:30 <riv> e.g. an AI in a discrete simulated world
14:41:47 <nakilon> explaining it would probably take the same amount of time I wrote it lol
14:42:04 <nakilon> here is the result: https://dpaste.org/6LwV/slim
14:43:17 <Corbin> riv: Then the useful parts of (warning: memetic hazard!) https://intelligence.org/files/LogicalInduction.pdf apply, and we get 0-1 integer programming (Boolean circuits), which is NP-complete.
14:43:26 <nakilon> I see no errors except probably the line #32 that is a part of "cryptocurrency", should grep to see the input with this substring
14:44:20 <riv> the result makes no sense
14:44:53 <riv> how was this computed
14:45:14 <riv> there was many channels with pokemongo in the name?
14:46:13 <Corbin> riv: That said, somebody does have a recipe for creating generic AIs: https://people.idsia.ch/~juergen/goedelmachine.html As it evolves over time, it naturally exhibits Bayes' law by improving the explanatory power of its inner model.
14:46:15 <nakilon> example piece of input https://i.imgur.com/AqHIn9K.png
14:46:59 <riv> interesting
14:47:07 <riv> can you tell me how it does this nakilon ?
14:47:10 <nakilon> example piece with pokemongo https://i.imgur.com/7p1q88m.png
14:47:40 <riv> there is a tension between length of substring and number of matches
14:47:46 <riv> the longer a substring is the fewer matches it has
14:47:52 <riv> how do you balance it?
14:48:05 <riv> is there a name for this problem you solved?
14:48:31 <nakilon> I have no idea about the name
14:48:56 <nakilon> I asked local NLP chat and googled, found no existing solution
14:51:02 <riv> input is a set of I of N strings, output is a set O of M strings. You want M << N and you want entropy H(I|O) to be much smaller than H(O) I guess could be a way to set it up
14:51:08 <riv> i dont know for sure if that's right
14:51:36 <riv> correction H(I|O) to be much smaller than H(I)
14:51:51 <riv> maybe it's nonsense
14:52:05 <riv> the idea is that I can be compressed well using O
14:53:41 <riv> I think it's the wrong way to think about it, I am trying to make anything a nail for this hammer im reading about now
14:56:12 <riv> we want to find "common" substrings in a set of strings. Call a substring of length n common if it occurs more than log(n) times.
14:57:08 <nakilon> if you want to code and try to achieve the same success I can give you the input
14:57:26 <riv> I have no idea how to code this, that's part of why i find it interesting
14:57:44 <riv> I would maybe try to adapt LZW
14:58:11 <riv> I think LZW is not just a random dictionary compression alg. I think it is the canonical universal compression alg (for ergodic sources)
14:59:22 <riv> It will be nice if you tell how you coded it
14:59:39 <riv> but if you dont want to that's ok
15:01:44 <nakilon> omg almost deleted the source code
15:02:22 <nakilon> that the file was still open in it
15:03:30 -!- arseniiv has quit (Ping timeout: 240 seconds).
15:04:04 <nakilon> "but if you dont want" -- no, it's just complex
15:04:51 <nakilon> and involves my know-how algorithm twice
15:06:37 <nakilon> "I would maybe try to adapt LZW" -- I asked here like 3 days ago about it if there is "some text compression CLI with debug option to see common substrings"
15:07:20 <nakilon> (I'm still lazy to make any search functionality for my IRC logs)
15:07:47 <nakilon> anyway, there was no answer
15:08:33 <riv> you should use git
15:09:00 <riv> LZW is just an alg, not a specific command line program
15:09:12 <riv> so i would need to code it from scratch, especially to get common substrings out
15:09:30 <riv> know-how algorithm?
15:12:08 <nakilon> "you should use git" -- I didn't have any good results until the last hour so I didn't commit
15:13:28 <nakilon> I call it PCBR (Pairs Comparison Based Rating) -- very universal thing for sorting tables
15:14:27 <nakilon> works better than more classic heuristics functions that you apply "to sort tables by multiple axes at the same time"
15:15:47 <nakilon> it's like round-robin chess tourney but with some details
15:19:46 <riv> oh that is cool!
15:24:56 <nakilon> back in 2013 I wanted to figure out who's the best player on my fav Quake map so I've scraped thousands of player profile stats; the approach didn't work better than just dividing "score / time played" with some heuristics, idk why, but that was the start
15:26:39 <nakilon> then in 2015 I reimagined it and applied to filter the RSS feeds to leave only the most interesting posts based on their stats like score/comments/age and properties of the tags they belong to -- worked well so I made a gem
15:27:20 -!- Koen_ has joined.
15:29:03 <fizzie> We have an unsupervised-learning thing for morpheme extraction, that's pretty much the same problem I think.
15:29:20 <riv> morpheme extraction...
15:29:31 <fizzie> Although if I remember right, we really didn't care if the result was "correct", just that it works well.
15:30:02 <nakilon> in 2016 I decided to automate the moderation of the content in the company I worked at -- the 1-nn worked well after I've optimized the training set of 2000 items throwing away ~500 of them, and since you can't retrain the model 2^2000 time I used the PCBR to effectively traverse the tree of throwingouts
15:30:21 <fizzie> http://morpho.aalto.fi/events/morphochallenge/ -- it was an almost-annual challenge kind of a thing for some time.
15:31:37 <nakilon> and after these "milestones" I started to apply the PCBR so often for my scriptings it would be too big list to describe
15:32:04 <fizzie> And https://github.com/aalto-speech/morfessor for the piece of software. Not sure how obsolete it is these days.
15:32:18 -!- arseniiv has joined.
15:33:02 <nakilon> the first time I've realised the things works I tried to find it in any books to name the gem properly but couldn't so I had to make my own name for it; later I tried to google it again and only found a bunch of math topics on wikipedia, still no exact match
15:40:38 <nakilon> https://en.wikipedia.org/wiki/Multiple-criteria_decision_analysis
15:40:46 <nakilon> https://en.wikipedia.org/wiki/Multi-objective_optimization
15:46:10 <nakilon> fizzie what is the first column? frequency?
15:47:49 <nakilon> (here http://morpho.aalto.fi/events/morphochallenge2010/data/wordlist-2010.eng)
15:56:33 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=86794&oldid=86788 * R3D * (+100)
15:57:06 <esolangs> [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=86795&oldid=86794 * R3D * (+17)
15:59:33 <esolangs> [[User:R3D]] N https://esolangs.org/w/index.php?oldid=86796 * R3D * (+178) Created page with "hello, i'm R3D, and i have created some esolangs (Deadmind, LTCBCBYCII, RPL) check it out if you can contact me (if any interpreter bugs): discord: R3D#9999 github: redleader167"
15:59:52 <esolangs> [[User:R3D]] https://esolangs.org/w/index.php?diff=86797&oldid=86796 * R3D * (+4)
16:39:30 -!- Koen_ has quit (Remote host closed the connection).
16:51:25 -!- arseniiv has quit (Quit: gone too far).
16:54:48 -!- arseniiv has joined.
17:29:05 -!- spruit11_ has joined.
17:29:19 -!- spruit11_ has quit (Client Quit).
17:30:36 -!- spruit11 has quit (Ping timeout: 256 seconds).
17:31:22 -!- spruit11 has joined.
17:55:28 <esolangs> [[Pith]] https://esolangs.org/w/index.php?diff=86798&oldid=86407 * ZippyMagician * (+1318) Clean up page
17:57:55 <esolangs> [[Pith]] https://esolangs.org/w/index.php?diff=86799&oldid=86798 * ZippyMagician * (+14) Change wording of - operator notes
18:01:23 <riv> ok here's my plan
18:01:45 <riv> since we have a lot of short strings, it should be ok to just make a histogram of every contiguous substring of each string
18:01:59 <riv> I think that means you multiply the number of strings by T(n) essentially
18:02:08 <riv> O(m * n^2)
18:02:13 <riv> m number of strings, n length of strings
18:02:43 <riv> then i want to pick ones that have a good score
18:03:03 <riv> where score is some function that likes length and frequency
18:03:14 <riv> possibly log(length)
18:03:36 <riv> there's a lot of free parameters to mess with there, which is not great
18:04:12 <riv> what do you think of this approach?
18:17:54 <HackEso> The password of the month is too long for this irc message
18:30:52 -!- arseniiv has quit (Quit: gone too far).
18:31:07 -!- arseniiv has joined.
18:33:51 -!- arseniiv has quit (Client Quit).
18:38:44 -!- arseniiv has joined.
19:01:18 -!- Koen_ has joined.
19:13:13 <esolangs> [[LTMCBCBYCII]] N https://esolangs.org/w/index.php?oldid=86800 * R3D * (+4838) Created page with "Esolang, created by [[User:R3D|R3D]] in July 2021. Full name is Language That May Cause Brain Cancer But You Can Ignore It, the shortest one is BCL. Heavily inspired by IN..."
19:26:14 -!- vyv has joined.
19:40:23 <riv> hm so the problem I have with my idea is that I get substrings of substrings in my output
19:42:19 <riv> nakilon, https://bpa.st/5P6A
19:42:41 <riv> i dont think its as good as your approach
19:53:44 -!- Sgeo has joined.
19:59:35 <riv> > k -common substring problem: Given m strings of total length n, for all k with 2≤k≤m simultaneously find a longest substring common to at least k of the strings. It is known that the k-common substring problem can also be solved in O(n) time
19:59:37 <lambdabot> <hint>:1:46: error: parse error on input ‘of’
20:22:13 -!- dutch has quit (Quit: WeeChat 3.2).
20:30:18 -!- vyv has quit (Quit: Konversation terminated!).
20:33:56 -!- dutch has joined.
20:38:15 -!- Koen_ has quit (Remote host closed the connection).
20:54:35 <zzo38> My opinion is I think that the POSIXLY_CORRECT mode of getopt is better.
20:55:50 -!- arseniiv has quit (Ping timeout: 272 seconds).
20:56:31 <riv> POSIXLY_CORRECT? lol
20:59:00 <b_jonas> zzo38: for getopt, I agree that it's better than the GNU mode
20:59:07 <b_jonas> but I can mostly deal with both
20:59:28 -!- delta23 has joined.
20:59:41 <b_jonas> I try to write my scripts in a way that they work with both when I invoke a program, so put options first but still put ./ at the start of filenames that start with a hyphen etc
20:59:54 <b_jonas> and -e before a grep pattern that starts with a hyphen
21:01:56 <zzo38> So, the implementation I wrote in PostScript is always using the POSIXLY_CORRECT
21:02:35 <zzo38> But when I write script calling other programs also I will do things like you mention, and using -- in case there might need other file names/arguments with - at first, too
21:04:02 <esolangs> [[Truth-machine]] M https://esolangs.org/w/index.php?diff=86801&oldid=86717 * Oshaboy * (+6) Formatting
21:07:49 -!- Koen_ has joined.
22:10:47 -!- Thelie has joined.
22:29:36 <esolangs> [[Excellerated Short-Term Memory Loss]] M https://esolangs.org/w/index.php?diff=86802&oldid=86308 * PolySaken * (+70) /* Computational Class */
22:47:25 <esolangs> [[Excellerated Short-Term Memory Loss]] https://esolangs.org/w/index.php?diff=86803&oldid=86802 * PolySaken * (-70) /* Computational Class */
22:53:19 <esolangs> [[Unhaltingfuck]] N https://esolangs.org/w/index.php?oldid=86804 * Silver * (+1680) page created, along with haltingfuck
22:53:21 <esolangs> [[Haltingfuck]] N https://esolangs.org/w/index.php?oldid=86805 * Silver * (+1472) page created, along with unhaltingfuck
22:54:54 <esolangs> [[Language list]] https://esolangs.org/w/index.php?diff=86806&oldid=86771 * Silver * (+38)
22:58:07 <esolangs> [[User:Silver]] https://esolangs.org/w/index.php?diff=86807&oldid=86284 * Silver * (+74)
23:28:00 -!- Thelie has quit (Remote host closed the connection).