« first day (3749 days earlier)      last day (1077 days later) » 

1:49 AM
it seems that the monthly mini-golf is something that by consensus (+11/-0) we want to do; the esolangs showcase has some popularity but only +5/-0, so no opposition but not as much attention
though we don't quite have a consensus (+2/-3) on whether or not to host it elsewhere (I agree that a chat event for golfing is more on-topic to TNB than TNB usually is, though)
 
@OldSandboxPosts html tags ftw
 
@hyper-neutrino Just an FYI, my opinion on that is not a "hard" opinion. I personally think they'd be better outside of TNB, but I don't particularly mind either way :P
 
2:04 AM
Now that I've sort of stopped working on Ash, I'm not really sure what project to start on next
With school basically over, I've got 35 hours a week to work on anything I want, I just don't know what that would be :p
 
Vyxal but in JS?
 
Answer with Ash a bit, then realise how buggy it is (just extrapolating from my own lang dev experiences) then work on Ash v2 :P
 
2
Q: Apply at indices

caird coinheringaahingGiven: a blackbox function \$f : \mathbb N \to \mathbb N\$, a list of positive integers \$L\$, and a list of indices \$I\$, apply \$f(x)\$ to the elements of \$L\$ at the indices specified in \$I\$. For example, \$f(x) = x^2, L = [7, 6, 3, 9, 1, 5, 2, 8, 4]\$ and \$I = [3, 5, 8]\$ (using 1 inde...

 
@cairdcoinheringaahing No, I'm not finishing Ash.
 
@cairdcoinheringaahing Stopped working in this context is more "most likely abandoned before completion"
ninja'd
 
2:07 AM
I'm currently writing Add++ v6, because my code for v5.11 is so awful/spaghetti, and full of bugs, that I couldn't stand it :P
@RedwolfPrograms Then finish Ash, maybe? :P
 
It's my first attempt at a golfing language, and my initial vision is so different from my current views on how a good golfing would work that it'd end up as a horrible mess
 
Yes
That's how all first attempts at golfing langs end up, that's the point.
 
I'm planning something, but I know it'll be a super long term project and I'd rather work on some smaller things like Hunter-Fatherer for now
 
You create a horrible mess, then learn from it enough to create a less horrible mess, and so on until it converges to a decent language
4
 
2:09 AM
How do I misspell that every time...
 
Fatherer? erm
 
Hunter-Gatherer
I...that'd be a very different KotH
 
@RedwolfPrograms Fatherer... isn't that a bit nsfw?
 
I guess my general list of things to finish by summer break is the KotH, ranking system, and CGCC stuff like the blog, bots, and guild of reviewers
After that I'll work on my game for a little while, then maybe around July or August I can start on golfing language stuff again
I'm seriously considering making a non-golfing language first though
 
How do you plan that for ahead for such specific projects?
 
2:12 AM
Too much optimism
 
I have a vague idea of programming projects I want to try at some point in the near or far future
 
Same, but I feel better when I pretend I know what I'm doing :p
 
For example, a (non-trivial) programming language, interpreted in Jelly
 
I'll probably start work on some completely unrelated project a week from now and in October I'll remember I still needed to write my part of the blog :p
Oh, I need to work on Generic Review Tool a bit as well
 
Pretty sure we'll have the first blog post up in time for Winter Bash :P
 
2:16 AM
This might not be the best time to have tried to start the blog, since I wasn't able to do any work that week because of school, and nobody really knew what we were supposed to be doing :p
Maybe in a few days we can get as many of us in chat as possible to try to work out who's writing what, a summary of what each part will look like, and some deadlines for when they should be done
 
Fingers crossed
 
I might not be able to blog until late June/early july now
 
I'm glad Vyxal got LoTM when it did, because June's a busy month with exams and final assessments and all
 
TBF at some point, I might just say "screw it" and write the first one myself. I've made ~25 languages, so I can probably write a decent amount about language dev
But, I don't want to prevent others who might be just as interested, so I'll wait until a breaking point :P
 
2:23 AM
@cairdcoinheringaahing I started reading the spec for your recent question and thought, "That's got to be a builtin in Jelly and/or APL." Got to the last paragraph--yep, called it!
 
@DLosc That's where I got the inspiration :P
 
I think I've noticed that as a pattern in APL programming especially: "Here's a list of values, and another list with some indices. Do something with the values at those indices."
(Spoken as someone who never got very far with learning either APL or Jelly, but reads explanations of code-golf answers from time to time)
 
2:51 AM
I need a repetitive song lyric for the song composing thing - any ideas?
 
Aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaah! :P
 
@user Sure.
 
Wait, there's an actual song with that lyric???
 
i approve
 
3:04 AM
My score is 19 97/225 because I've posted just so many answers :P
 
3:28 AM
@Ausername you didn't know?
 
 
1 hour later…
@okie I'll probably do that next time.
 
:(
 
There aren't any more answers that I haven't continued off.
 
5:12 AM
sup
 
yes
 
How to find questions in PPCG which don't contain any python submission
Any nice search filter?
-"Python" is:question dosent work
 
you may need a SEDE for this
 
Oh no
 
@hyper-neutrino that sounds like something from the future
 
5:16 AM
data.stackexchange.com quite the opposite, it updates every sunday so it's usually from the past :P
 
6:02 AM
@programmer365 modify this
 
6:30 AM
7
Q: Machine Epsilon

leonardoMachine epsilon is an important floating point number to know when doing numerical calculations. One way to understand it is when this relation 1 + machine_epsilon > 1 does not hold. One (iterative) way to extract it is executing a small program like the one above: 10 EPS=1 20 EPS=EPS/2 30 EP...

 
> Thencequentlyforthupon
 
 
1 hour later…
7:47 AM
Yes
 
@Lyxal yes to whom?
 
Hn
 
I'm going to create a golfing language
@Lyxal
give me help
pls
 
@StackMeter yay!
a new golflang
what will be its name
 
containing no letters
@programmer365 Anti-Alpha
 
7:51 AM
Hello, just want to ask a very quick question: a user suggested [bit.ly/3nQ60Mu](this) -2 byte suggestion on "Help me count the Omer", is this acceptable? Can we have an anonymous function as the main answer doing the work and a few named helper functions at the bottom?
 
@ophact well i can't access the bit.ly link
 
it leads to a TIO
 
numbers are a thing in anti-alpha though
 
just copy paste the link
 
@ophact Yes, as long as all are counted.
Here:
https://tio.run/##jZNfT9swFMXf8ynuQ1ETYdKxP4CqpRGTtiGNdUg8AlLc5DZ1ce3IdshCW746u04KDPayFzdx7j3nd4/dJb/jNjeicgdKF/g4Tx7Dc3bKLtk0SibZYD3dQsHbwfosnEZbep0cpRkDxVcoWxisHx7C6eg42kKDeOurdhu0Q8V7x2nW/byI7FFxNh4O@yUIBuvQJe@izeY8XvEqDJEJMhZJ8iHF2GAleY7hKB6VLE8m4eg63sDmmo1ih9aFeZS68f6@i5JkmmbrwTrfkm4ejTGKl1qoDLLYVlI4//BafnpwmGZXg/WLyX3SqZM4Gw6j7c1gfd@7YEQoVgqqOTiMevYXg26G07/V/djksEvC975CaJikIkkIe8c7iOYtxOZfiiZKm7cUzTPFGMd@qNGTIm5vXiCvVU95eeXROtvx0c0WhIJu8/NJ@m48nXw8SY/GO@qbbZAFZ0mdTOrJYTq0QzJ8zLWyWmIsdRnOw@y7LmAm0VooDbEzmNMKdiEUQm1hrg3eoYmDH0o3gNy4BTS8tQwsv0MLXEq6R06QaBxcGC4sQoW6kgjn2hSsK9htVN3nO
 
7:54 AM
@Adám thanks, that is very much appreciated. I can probably golf my code a little further now.
 
@StackMeter so there's absolutely no letters. Hmm. You'll probably need extra Unicode then
Do you want it Prefix, Postfix or infix?
 
@StackMeter what will you use in their place?
 
@StackMeter also it is stack based or stackless?
 
8:34 AM
To answer all your questions, 1) Infix, because it's the only thing that makes sense, 2) There's surely enough symbols to uniquely map every one to a letter, 3) No, because I'm going to implement variables, and I'm not sure those two go together.
yeah there's 32 symbols we have more than enough
 
@StackMeter Sounds APL-like.
 
@Adám cool
but I literally don-t know how to process infix
 
well it depends on what kind of infix
 
@StackMeter transpiling would be nice
 
one option is to use a ton of tiers of precedence but it's not hard to tell why that isn't a very popular approach
 
8:46 AM
assuming you are writing interpreter in common languages like Python or Javascript
then you can easily trasnpile infix expressions
 
i don't think python lets you define new infix operators if that's what you mean
 
@programmer365 well I do know Python, so I might try and transpile infixes, because this is being written in Python (I'm not learning a language to make a language, when it is unnecessary)
 
if you want to discuss it more let's open a chat room on it
 
if you want to do a precedence based system you can also just
use operator precedence parsing
but again i would not recommend a precedence based system
at least not as the core of your syntax
if you feel like rolling with it and seeing what you can accomplish then go for it
but yeah
 
@programmer365 sure
 
8:58 AM
@StackMeter then hurry up open it
 
how do I open a room
 
go to site rooms and click open a new room and follow the instructions
 
open rooms list
csroll to bottom
 
ah I want to be changed back to Wasif
but i cannot change my username in next 30 days :-(
 
it done
 
9:03 AM
nice
now all discussion could be done there
 
9:33 AM

 Anti-Alpha: The Beginning

where stupid begins
 
@programmer365 Oh, that's who you are. I was wondering... Welcome back, by the way!
Hi!
 
10:09 AM
@ngn why do you have to emulate it?
 
 
2 hours later…
12:37 PM
in bash, is there a clever trick to copy all quiz-h.* files to their corresponding quiz-i.*? I feel I saw someone use brackets to do this somehow
 
In shell, I think you can "just do it" with copy quiz-h.* quiz-i.*
 
do you mean cp?
or is that a different shell?
cp quiz-{h,i}.tex does one file
 
Windows shell.
 
ah.. I know nothing of windows :)
 
You can get PowerShell for UX, but I don't know if you can get the good old shell.
 
12:42 PM
Okay, I shouldn't have let you guys do it in the first place, but please don't do it a second time. It's not really that funny and it's confusing and therefore disruptive especially when you talk about it in chat that much.
 
@hyper-neutrino That's why I spoke up.
 
94 messages moved to ­Trash
@Adám Yes, thank you for doing so.
I have "ruined the fun" in place of caird while he is absent. :P
 
@hyper-neutrino the fun is ruined
 
@Anush Oh, Wine should be able to run cmd.exe ― there you go! ;-)
 
@Adám I want a linux answer! :)
actually a bash answer
 
12:53 PM
@hyper-neutrino this is so sad
But could you please do a profile refresh for me?
 
@hyper-neutrino What exactly happened here?
 
What we did that one time but not last time
 
1:05 PM
@cairdcoinheringaahing my experience completely agrees
also good morning and why does all the conversation while I was away fit on two screens?
 
ngn
@user41805 good question
 
@Wezl because I trashed 94 messages
 
Can I at least keep the meme I posted?
As in can I post it here again. It seems somewhat relevant
 
ngn
@Anush for f in quiz-h.*; do mv $f ${f/h./i.}; done
 
@hyper-neutrino Mobile makes it difficult to trash chat messages :P
 
1:19 PM
mobile makes it difficult to do anything lol
 
Morning everyone!
 
ah yes, my birthplace
4
 
Bunch of spam flags in russian lol
 
what the hell is happening
 
1:29 PM
> But from checks in the queue, the participation of a person is still required, and not an autoclicker ass.
 
I don't trust google translate enough to decide based on it
 
Same
It looks like they're discussing fake reviews, copied off someone else with an autoclicker
Definitely invalid I think, but I won't vote since I don't speak russian
Looks like they've been cleared up anyway
 
@ngn cp not mv! :)
@ngn that is nice though. Better than for f in quiz-h.*; do cp $f quiz-i${f#quiz-h}; done
 
ngn
@Anush oops
 
but is it the most golfed is the question?
 
ngn
1:33 PM
probably not
 
I would pose it as a CMC but I am not sure what command line languages there are
 
@ngn How dare you write code that isn't the most golfed‽
 
@Adám :)
CMC (Linux command line.) copy all quiz-h.* files to their corresponding equivalent quiz-i.* counterparts.
 
ngn
@Adám coming from you, this sounds offensive
 
@Anush VTC Because this question is off-topic here. CGCC is not for general programming problems; you may want to try Stack Overflow :P
 
1:36 PM
@ngn Really? I was sure you'd get my tone right.
@Anush wine cmd copy quiz-h.* quiz-i.* or somesuch ;-)
 
@user :)
@Adám nice answer :)
@user the score is still bytes!
 
ngn
@Adám do i have to remind you what you&dialog did with my submission to the apl competition 2 years ago?
 
what's the meta consensus on allowing questions that have a practical use?
 
They're allowed but recommended to avoid
Unless they're actually interesting
 
@Wezl They're probably allowed, as long as you're obviously not just posting a homework problem
'Interesting question' trumps 'homework problem' imo
 
1:39 PM
Yeah, I'd think it'd be allowed. Plenty of challenges are potentially useful.
 
Plus "practical" and "golfed" are two things you will never get in the same answer lol
 
Like convert TIO post to CMC answer, and helping me count the Omer.
 
Unless you do atomic golf or fastest code
@Adám Lemme go VTC that :P
 
(why do people take my questions seriously??)
 
Because you asked it with a serious tone :P
 
1:40 PM
@Wezl Hmm, I'll have to think on that, as it's a rather complex matter.
 
@Wezl That's an off-topic question! VTC.
 
I say we go straight for a question ban :p
 
@Adám I like that you put a "." even after "VTC"
 
11 is the correct punctuation mark here though
 
@RedwolfPrograms You can go for a question ban if you want, I like the ability to ask questions
 
1:42 PM
I want to make a practical language. I have such a good idea for some features.
 
How about inputting and outputting erased implicit values?
 
@user I should collect all the times people have praised me for my consistent punctuation.
 
@RedwolfPrograms extensibility (ducks in case there are java people here (ducks in case they're offended by that parenthesis))
 
Java is literally the king of extensibility - we even have an "extends" keyword!
 
OOP is for people too weak for imperative
 
1:44 PM
?
 
OOP languages can also be imperative
 
If you need any structure at all you're doing it wrong.
 
I'm not sure you can get Turing-completeness without data structures
 
@RedwolfPrograms OOP is for people too scared of Unicode to learn golf langs :P
 
1:45 PM
Turing completeness is for people who can't just do the computing beforehand and tell the computer the answer
4
 
~~ OOP GOLF LANG TIME ~~
 
Who says you can't have an object-oriented golf lang?
@Wezl Gah, ninja'd
I'm gonna have to ask you ROs to leave now
 
I'm afraid I can't do that, Dave user
 
irc is the superior platform, and it doesn't have _built-in_ emphasis
 
ngn
1:55 PM
@user41805 i thought there might be some argument from irreducible computation (computation for which there are no shortcuts, you just have to perform it), but maybe i'm wrong
 
I see my CMC got the same number of VTCs as answers
 
@cairdcoinheringaahing Sorry, I don't get the reference
 
ngn
@RedwolfPrograms "попка" could mean "ass" or "supervisor" - i think it's the latter in this context
 
I figured the translation was incorrect, yeah
 
did you know Mozilla and Facebook use Mercurial instead of Git for most of their development?
 
2:00 PM
@ngn The same word could mean "ass" or "supervisor"? Interesting...
 
I would've marked them as not valid, but I do not trust Google Translate lol
 
@pxeger Why?
 
@user Facebook says it scales better
Although they're now working on their own partially inter-compatible SCM system called Eden
 
ngn
@user it's "supervisor" in criminal slang
 
Oh ok
@pxeger Pretty cool
 
2:05 PM
@user It's from 2001: A Space Odessey
 
Always knew you were just a bot
Apparently I can't tell the difference between you and Wezl
 
I _am_ Wezl :)
 
ngn
2:21 PM
cmc: list of totients for natural numbers below n
(or up to n, if you prefer)
 
@user Mercurial is easier to learn than git, although git is eventually renaming its commands to be more meaningful, so you don't have to use the same command to a) change current branch b) restore working tree files
 
ngn
@cairdcoinheringaahing builtins are valid but no fun
 
@user also git doesn't understand the concept of copies and renames very well, so it's hit and miss whether it will actually merge the right files together
 
2:26 PM
it's so bad that in one big merge I had to go through and manually rename all of the affected files first
while in another merge I just exported to Mercurial and merged there instead
 
ngn
2:38 PM
@cairdcoinheringaahing do you think jelly's R starting from 1 was a good choice?
 
@ngn Yeah. Personally, I prefer 1 indexing over 0 indexing, and R starting from 1 fits best with that
It can be a little cumbersome at times, but generally, it's better
 
I prefer 0-indexing
it makes returning a falsy value annoying for find in string/array stuff, but I have a secret plan for my next golfing language
 
NGL I have the same view of "1 indexing vs 0 index" as "tabs vs spaces", "emacs vs vim" and every other "programming debate", which is: I'll use whatever's easiest for me, and I don't give a crap what other people use
¯\_(ツ)_/¯
 
Real programmers use 0.5 indexing, newlines for indentation, and write all their code in excel
One character per cell
Guarantees good monospacing
 
I take back what I said. This is the one truly correct way
 
2:47 PM
@RedwolfPrograms new-lines for indentation?????
 
Real Programmers don't indent of course https://github.com/kparc/kcc#indentation
 
zero-width-spaces please
 
@cairdcoinheringaahing FTFY: NGL I have the same view of "1 indexing vs 0 index" as "tabs vs spaces", "emacs vs vim" and every other "programming debate", which is: I'll use whatever's easiest for me, and I don't give a crap what other people use anyone else is wrong and should die
 
Not 0x0a newlines, 0x85 NEL characters (from EBCDIC, included in Unicode)
 
ngn
1indexing is too painful for me. it's one of the reasons i never learnt jelly.
 
2:52 PM
@ngn Painful is a view I've never heard before, how so?
 
ngn
always -1 +1 to correct things
 
@cairdcoinheringaahing because it hurts :D
 
ngn
in most cases it doesn't matter, the pain is in the small subset where it does
 
@ngn If the entire language is consistently designed in the indexing (0 or 1 or whatever), you generally don't need to do that a lot
Add++ uses both 1 and 0 indexing, depending on the command/mode, which is fun ಠ_ಠ
 
disagree as a user of lua
 
ngn
2:54 PM
@cairdcoinheringaahing fun? my experience with apl tells me otherwise
 
@ngn "fun", not actual fun :P
It's just another thing I'm adding to my list of "awful behaviour that I need to fix" for Add++
 
add++++!
 
APL using ⎕IO←1 for the competition problems is annoying
 
ngn
not sure which is worse (for me) - always starting from 1 or leaving it as a choice
 
@Wezl Add++ v6 :P
 
2:56 PM
@Wezl well add++++ will obviously sometimes use 2-based indexing
 
@ngn definitely leaving it as a choice imo
 
0-indexing feels more right for practical purposes
1-indexing feels better when golfing
 
ngn
@hyper-neutrino and theoretical too
 
well i mean practical as in not golfing i.e. actually useful situations lol
 
The mathematician in me prefers 1 indexing no matter what :P
 
2:58 PM
ever since i learned why 0-indexing makes more sense in terms of pointers it makes more sense :P
 
ngn
@cairdcoinheringaahing in maths the residue classes mod n are 0..n-1, not 1..n
 
@cairdcoinheringaahing 0 indexing is nicer mathematically because of modular stuff
 
ngn
@cairdcoinheringaahing in set theory n is literally the set {0,..,n-1}
 
Unless you have a separate set of modulo-based builtins which do -1 and +1 before and after
 
2:58 PM
but since i usually golf in jelly 1-indexing makes sense when i'm thinking about how to do something with a total disregard for practicality, efficiency, or anything reasonable
 
Numbers were never meant to go below 1. "Nothing" as a number? "Negative" values? Complex?? Bah, return to normalcy :P
4
 
grrrrr
 
ngn
@cairdcoinheringaahing 1 is just succ(0) and 0's existence is an axiom
 
@ngn I disagree, these classes are sets that ciontain them all and it doesn't matter what representative you choose!
 

« first day (3749 days earlier)      last day (1077 days later) »