« first day (3918 days earlier)      last day (915 days later) » 
00:00 - 15:0015:00 - 00:00

3:03 PM
According to GitHub, I have 10 esolangs: Braingolf, LMBM, Clam, 2Col, Vacui, AlgiX, voovoos, sum-it, funcsational, BrainFlump,
at least half are functionally unusable, and at least half are unfinished and abandoned
technically Vacui is as usable as brainfuck, but only if you can accurately time your keypresses to 1/100th of a second
 
I've finished Grok and brainbox, and I have unimplemented ideas for Quest, Hello Hell, brainhex, Jumble, Cursed, Plum, Parity, and this new tacit lang idea.
 
Ah, I see you're a man of culture as well
 
lol
I hope to one day implement all of those ideas, but I have no idea when it will happen
 
BrainFlump is my only truly "completed" language, but it's also one of the functionally unusable ones
fun fact: Back in 2018 I got fired for working on sum-it on the job. That started a chain of events that led to me moving to Canada and getting married
5
 
esolang => true love, everyone write that down
i have exactly one finished + implemented esolang, and two that im implementing simultaneously
and one that i decided to abandon "until im better at programming"
wait, two of those as well
 
3:14 PM
i am currently making one (for the past 4 months)
 
I'm going to try to start a giant class-wide KotH in my comp sci class
 
it's not a real golfing esolang if you don't golf the interpreter code too
 
i couldnt wrap my head around implementing a lang that let you define, create, and remove functions with the same command
 
@RedwolfPrograms Might post it on CGCC too if it happens so that we can get some external competition
 
oh sick
 
3:15 PM
@Skidsdev uh
 
skidsdev is right
 
@RedwolfPrograms any idea what the thing will be about yet?
@Skidsdev alternatively, you could make the code be shaped like ascii art
thats always fun
 
@thejonymyster look, my 2 braincells are already at max capacity
 
@thejonymyster Not yet
Wow, I have now achieved the maximum possible lag
That message took like 40s to send
 
@thejonymyster I was expecting donut.c
 
Of my unimplemented esolangs, I'd say my favorite is Quest. I have a "spec"-ish thing written up for it on github
 
@AaroneousMiller the resemblance to an esolang idea I came up with earlier today is uncanny
 
i also tried to do fun code shapes once, this ones less impressive but im proud of it: my hq9+ clone and the esolang page
 
3:22 PM
@Skidsdev wow rly
 
literally on my commute this morning I had the idea for a 2d esolang where the source code is a fantasy rpg style "map", and the instruction pointer is the "hero"
which skimming the spec, seems to be exactly what Quest is
 
Great minds think alike or whatever the expression is :p
 
I think the world is just running out of unique esolang ideas :P
 
I read "Transmutation" as "Traumatisation" lol
 
3:24 PM
same thing, pretty much :p
 
do you have any example programs?
oh its in the github
my b
 
hmm, I think the cover up zeroes challenge might be possible in BrainFlump
 
@AaroneousMiller do you have something simpler like a truth machine ':D lol
lots of cool commands, but no syntax explanation \O/ {waa)
 
I've done another round of mods to my text editor: original, now
 
3:31 PM
I think this would be a truth machine in Quest, based on the info we have:
q
pnd
#ll
oh wait
q
pn+
#l+
ah shoot truth machines still need to output the 0 once
q
p+
nd
#+
last try
 
anyone know how to get the square root of a number as big as 2^2048 in python?
(that is 2**2048 in python syntax)
 
Any reason normal methods don't work?
 
@RedwolfPrograms math.sqrt gives an error message
 
Well I mean like newton's method and stuff
 
int too large to convert to float
 
3:38 PM
Presumably Python's math library works on floats
Like most languages
 
right
so what is the alternative?
 
Implementing it yourself?
 
could be but python normally has a library for anything you can imagine
 
Then google a python int sqrt library
:p
 
@Anush if it is exactly 2 ** 2048, then its square root is obviously 2 ** 1024
 
3:40 PM
I was ! via my favourite chat room :)
 
or do you just mean numbers in that ballpark
 
@pxeger true :) not exactly that number
@pxeger yes I do
 
sympy?
you can set custom precision on that, but I think there might still be limits
 
@pxeger sympy could work!
 
oh ok, cover up the zeroes is not possible in BrainFlump, because BrainFlump can't detect EOF from stdin
 
3:54 PM
would it be cheating to get length of input?
 
I think that would make it impossible
 
@Skidsdev yeah, according to my current idea of the language, that would be a functional truth-machine
 
you can't practically store more than 2 values at a time in BrainFlump. I couldn't store the length, current value, and last non-zero value at the same time
 
ah
thats tragic actually
 
3:56 PM
I mean you can, but you'll have a bad time
 
@AaroneousMiller i still dont fully understand the syntax or how movement is processed but having a simple example like that helps a bit
 
@AaroneousMiller I wrote that under the assumption that, from #, movement will happen in whatever way is valid, and it will detect that even though it can move right into the +, it can't go anywhere from there, and so must go up
 
how do you gather that assumption? did i miss a paragraph explaining movement :P
 
@Skidsdev I remember brainflump :P
don't remember anything about it, just that it exists :P
 
@thejonymyster BrainFlump has a single accumulator, to which input values can be read, and a 'dump', to which the accumulator value can be pushed. The problem is when you pop from the dump it pops a random value stored in the dump, so if there's more than one value in there, you can't reliably get either of them out
 
3:58 PM
incredible
 
Also, the interpreter is golfed :P
 
@cairdcoinheringaahing thanks to your hard work!
 
That's how you know it isn't well golfed, cause I golfed it :P
 
I'm pretty sure the concept of storing what each op does in an array, and indexing into that array based on the op's position in a string was your idea
aka this monstrosity:
`exec['v+=1','v-=1','d.append(v)','v=d.pop(randrange(len(d)))','v=ord(stdin.read(1))','stdout.write(chr(v))','0']['+-:;,.'.find(c)]`
 
such a monstrosity that SE chat refuses to parse it into a code block lol
 
4:01 PM
Yeah, probably. My Python golfing skills are rudimentary as best :P
 
@thejonymyster It was an arbitrary assumption based on the complete lack of info regarding initial state
If memory serves, BrainFlump was a sort of satirical parody of brainfuck derivatives that implemented various memory models, such as BrainFlack implementing a stack, and BrainFlueue implementing a queue
I invented the unusable memory model of a 'dump', and BrainFlump was born
 
@Skidsdev makes sense, i just was surprised by the dead end detection
 
oh btw @thejonymyster if you wan to grab another +2 rep by editing out the HTML you left in, I'll happily give you a few minutes to do so, before I edit it out :P
 
@thejonymyster it would definitely add significant complexity to the implementation. If you wanted more naive detection you'd have to add another column to the truthy machine like so
q
pr+
n d
#l+
that version also makes the loop more apparent
 
@Skidsdev Yeah, I forgot to put it in spec.md, but essentially it tries to continue straight after a command, and if it can't, then it tries each direction, going clockwise, until it finds a direction it can go in. It will not double back on itself, and if it finds no other way it can go, it is considered a dead end, which is equivalent to q. If it isn't moving, such as at the beginning, the first direction it checks is East.
 
4:08 PM
@AaroneousMiller ah okay, so in that case the one I just posted with the extra column would be correct, because otherwise it would go east into the + and then terminate
 
@cairdcoinheringaahing nah go for it lol, thanks though
@AaroneousMiller thank you it all makes sense now :D
 
@AaroneousMiller I won't rob you of the joy of implementing Quest, given its your language, but please let me know when you do I'd love to play around with it
 
Oh btw, if you're looking for a way to make an edit longer to be approved, if the post has an image in it, adding in a description of the image 1) pushes it over the 6 char min, and 2) improves accessibility for users on the site
5
 
oh! that is very useful to know yes:D
is there somewhere specific the description goes?
 
In the [...] part of the ![...](url)
 
4:14 PM
Where the placeholder says "enter image description here"
 
If it's left blank, it will say something like "Please add a description here"
Yeah that ^^
 
Basically, for anyone who uses a screen reader, the reader will read off the image description. Which... is useless if its the placeholder text
 
right
 
Been meaning to right up an SEDE query to find those posts actually :/
Im sure I'll get round to it at some point :P
 
Want me to? I'm bored :p
 
4:17 PM
You know I'm serious about one of my languages when I don't even bother documenting the operations, and just tell people to look at the source code instead :P
 
@RedwolfPrograms I know that if you do, I'm going to stop working on my homework, and start editing posts on the site :P
 
Too bad I already wrote one :p
 
How tf did I manage to correctly enter my password to a site when half-asleep and on mobile where the keys are too small. But can't do it correctly when awake on my laptop???
 
There are about 2800 such posts if my query is accurate
 
@RedwolfPrograms Got a link? :P
 
4:22 PM
I did, except SEDE sucks at permalinking and I reloaded by accident
 
this is a sign for caird to go back to doing their homework
 
But you have to promise to do your homework first
 
Just waiting for it to compile :P
 
4:36 PM
Ughhhhhh I hate Chrome console's new string formatting
I hate it so much
It's sooooooooo awful
And you can't disable it
 
whats it like?
 
It replaces characters with their proper formatting as if they were string literals
 
@hyper-neutrino Hey, I actually did this homework before even opening the SEDE link :P
 
So newlines are \n, quotes are \", etc.
 
@RedwolfPrograms See, I've been working on a query to identify posts with latex as images, which is much more complex. So for some reason, I thought it'd be more than just a single line query ಠ_ಠ
 
4:45 PM
No worries, it's quite clearly the SQL. It's been scientifically proven to cause your brain to shrivel up and invent slightly different variations of the universe in order to pretend SQL isn't as bad.
 
No, I know SQL is bad, but I just thought everything would be more complicated than expected in SQL :P
 
@RedwolfPrograms switch to Firefox!!!
 
Y'know, like a reasonable expectation :P
 
@pxeger I have, on my Windows laptop
 
oh... ChromeOS...
 
4:47 PM
On a chromebook it's a little to ironic (and slow)
 
use alert /hj
 
Cursed language idea: Interpret diacritics which look like single quotes or backticks as prepended single quotes or backticks
 
so àbcdè is the same as 'abcde'?
 
Probably with backticks and not quotes, but yes
Actually no
With my current idea it'd be `abcd`e
But it could be modified to work more sanely I guess
 
> cursed language idea
> sanely
I see a contradiction
 
5:03 PM
@Skidsdev It's kind of ambiguous, but I think n and p actually pop the top of the stack (if I remember my original intent correctly), so you'd have to duplicate the top of the stack, which requires gold, which requires fighting monsters, so I think a truth-machine (based on the current spec) would be this:
n0p+l!0S1#
+l1+
 
the endless war for truth u_u
 
@AaroneousMiller oh no
oh god I'm trying to print tabula recta in braingolf and loops are so horrifically broken
I think I need to rewrite this whole interpreter from scratch
 
@AaroneousMiller Actually, I wrote that message before I finished golfing the program, and it doesn't duplicate the top of the stack at all. It actually checks to see if a 1 or 0 was inputted by seeing if the Slime gives us enough gold to transmute the 0. If we have enough gold, we inputted 1, otherwise, we inputted 0.
 
I guess your brain is the interpreter now...
How do we upload somebody's brain as an interpreter to TIO?
 
Step 1: Wait for Dennis to return
Oh wow, SEDE is actually cursed
When you do ##Input## stuff, I think it inserts it (almost) literally into the query
 
5:10 PM
@cairdcoinheringaahing Alternatively we could have an SEChat channel where people post Quest programs and Aaroneous replies with the output. Then we can just permalink to the chat message :D
 
@cairdcoinheringaahing SQLi FTW
 
quick question: just how do i loop in jelly?
 
what kind of loop?
 
for example: repeat left link z times
are there others?
 
¡ is one way
 
5:12 PM
¡ does <link><repetitions>¡
 
There's tons of quicks that do looping related things, though
 
@Skidsdev I'll disable Aaroneous with an infinite loop while you grab the money :P
 
@Seggan while loop, loop until repeated value, etc.
 
tyvm
 
@cairdcoinheringaahing But it does have some unexptected behaviour when <link> is dyadic
 
5:13 PM
@Wezl I guess using a human interpreter solves the halting problem
 
@cairdcoinheringaahing like?
 
and by 'solves' I mean imposes a hard limit on program run-length/iteration count
 
Unfortunately Aaron will inevitably halt ._.
 
@Seggan It's essentially a generalised Fibonacci function
I'll whip up an example
Try it online! С is like ¡ but it returns each step
 
@Skidsdev All programs will halt in <my attention span> seconds, therefore Quest can't possibly be TC :P
 
5:17 PM
x f3¡ y gives f(f(f(x, y), x), f(x, y)) rather than f(f(f(x, y), y), y)
 
@cairdcoinheringaahing ty, but the program keeps going?
 
@AaroneousMiller if your attention span is anything like mine that makes Quest borderline unusable
 
lol
@RedwolfPrograms Why do you say that?
 
@Seggan What do you mean?
 
@RedwolfPrograms if every program always halts I guess that solves the halting problem :P
 
5:18 PM
@cairdcoinheringaahing return value -> next loop
 
Nooo I almost outgolfed a JS answer using some really cool compression stuff but adding the x=> at the front to make it a valid answer made it two bytes longer :(
 
:( sadge
 
@Seggan Oh right, yeah С just shows each step from the loop. If you notice, the fifth one e.g. is the pair of the fourth and the third, which is what ¡ does with dyads, rather than reuse the right argument each time
 
ty
 
5:24 PM
Alright screw it, time to re-write braingolf from scratch
 
@cairdcoinheringaahing so if i understand correctly, it moves the right to the left and uses the intermediate as right?
 
Basically, yeah
 
makes sense
 
wait, no. it moves the left to the right, and uses the intermediate as the left
 
nearly the same :)
 
5:50 PM
In case anyone's interested btw, there's currently not one but two controversies in the latest SO election :P
 
oh no
 
it seems odd to allow nominations right up until the end of the nomination phase, and not allow discussion after that phase. Surely you'd have a nomination phase then a discussion phase?
 
at least with the plagiarism the plagiarizer owned up to their plagiarism
 
5:59 PM
yeah, that one's slightly more interesting
cause. Shree is active in CHQ, and in the SOCVR, and looks to be exactly the kind of user who fits well with the "janitor" side of SO modding. But, damn, plagiarism is not a good look
 
somebody further down pointed out that some of Shree's answers were identical to other answers passed through a paraphrasing AI o.o
 
oh plagiarism, yikes. that could be what was being discussed in the so mod room yesterday
someone summoned all the SO mods in TL yesterday to look at that discussion so i figured it must be something really important
 
Personally I have no skin in the game. CGCC is the only SE site I'm at all active in, and so the only one I would feel comfortable voting in the elections of. That said I agree that plagiarism is not a good look for a mod candidate
 
i mean honesty is one of the most important things for a mod IMO, so... yeah, definitely not good
 
I've voted just because SO elections have the least "impact" from a single vote of all elections on the site
Plus, I do genuinely think one of the candidates in particular would be an excellent mod, so wanted to cast a vote for them
New profiles are live - mainly just some changes to make them responsive
 
6:10 PM
I have an impression that Shree has some difficulty with expressing their thoughts in English. All answers are very brief. In such cases it is tempting to use some ready phrases that are available. — Tadeusz Kopec 2 days ago
This is what it seemed like to me
 
Ollie's definitely captured my opinion on it well tbh
 
@hyper-neutrino to be fair, they were honest when the issue was raised. I don't think they did it out of malicious intent. As somebody in the comments pointed out it's possible their english written communication isn't very good. Although that's also not a great look for a mod on an english site...
 
Huh, so apparently mathjax + image descriptions don't mix (cc @DLosc) :/
 
Yep :/
 
I've swapped out the mathjax for the actual characters, should be enoug
Might open a report, because the edit preview doesn't show anything messed up with the image description
 
6:24 PM
Okay so...the company that provides little cartons of milk for my school puts jokes on the back of them, since the elementary school students get the same milk. And today's joke is uh...well it's a bit concerning.
"Why did the chicken do jumping jacks?" / "She wanted scrambled eggs."
Given the function of chicken eggs...that has some worrying implications
And...do the chickens want to eat their eggs?
Otherwise why would they want them scrambled
Or do the farmers want scrambled eggs and they've manipulated the chickens into believing they want them too
 
But... it's a milk carton... why not go with the much less troubling joke that's almost identical but with cows and milkshakes?
"Why did the cow do jumping jacks?" / "She wanted a milkshake"
 
@cairdcoinheringaahing Yeah, the difference between the preview and the actual results ought to be a bug. Whether MathJax should be supported in image descriptions could be or .
 
@Skidsdev well, sure, it's better to admit to a mistake than do nothing or deny it. but they still plagiarised in the first place and plagiarism is an issue of honesty inherently
and about the language barrier thing - I can see that being a motive for plagiarising but not a justification (which I don't think the original comment was trying to do anyway), and if by "malicious intent" you mean stealing zoe's nomination for the purposes of not having to do it, sure, there's an argument there, but there's still malicious intent in the form of misleading voters and intentionally taking others' work
 
@hyper-neutrino I wholeheartedly agree. Accepting responsibility doesn't nullify the act. Overall I agree with Ollie's comment caird linked to.
 
i would much rather read a nomination in broken english and being able to discern the candidate's competence. regardless of how important or irrelevant english proficiency is to moderatorship, what's the point in electing someone based on something they didn't even write? you just can't judge someone for a nomination that isn't theirs
as long as you can still get by communicating when necessary the only time i see language being an issue is if you frequently misunderstand posts and inaccurately assess if posts are off-topic, spam, rude, etc. but I don't see that really being an issue much
 
6:33 PM
The biggest mod communication is (correct me if I'm wrong hyper): suspension/deletion mod messages and meta discussions. Given the size of the mod team, wouldn't surprise me if mods with worse english skills can get another mod to handle those
 
I also forgot that there are non-english SO sites as well. I don't know whether mods are shared across the different SO sites, but if they are that would further lessen the relevance of english proficiency
 
It wouldn't surprise me if there are mods on SOes, SOpt, SOja and SOru who don't understand any english, so sharing mods would be a bit weird
 
In any case I come back to my original remark; I'm not in any way active on SE outside of CGCC so my opinion on the matter is irrelevant
 
@cairdcoinheringaahing yeah, pretty much. and things like talking to CMs about things, etc. but as long as you have someone on your team who can handle that it's not a problem for some mods to just do more behind-the-scenes stuff
 
Interesting that this and the last election on SO are the lowest in candidate numbers by far (6 each, every other one has had at least 10)
 
6:43 PM
are the number of slots lower too or is that not correlated?
(i definitely doubt there's causation but still curious)
 
Nope, election #11 also had 2 positions, same as #12 and #13
Tho most had 3 (except for #9, #4 and #2 which had 2, 4 and 4)
 
6:57 PM
hehe, I tried to make a truth machine with +³Ð¿. It works with 0, but with 1, my computer froze and I had to force shutdown
 
I'd suggest using tio.run/#jelly, rather than a local version :P
 
ye, ik i was thinking maybe TIO collected output at termination, instead of live. guess i was wrong
 
@Anush binary search for it, takes 1024 steps
 
@pxeger Same (on arqade), and Aaroneus too.
@pxeger Nope - I do that all the time (on my questions) :p
 
aha, the truth machine: Ṅ³¿
 
7:11 PM
gtg, o/
 
7:35 PM
@Neil good idea.
 
8:06 PM
Well I got hello world working in braingolf 2.0... now for everything else
 
8:56 PM
@AaroneousMiller This looks like an interesting challenge to answer in Acc!! if you want to try something without a source restriction.
 
*Cracks knuckles*
 
*Craccs
 
 
1 hour later…
10:03 PM
We should really find a good way to incentivize posting and voting on answers to old questions
The FGITW effect is so strong
 
More bounties maybe
 
i feel like we already have a bounty for answers to old questions - maybe that's just unanswered ones tho
 
It's more the voting on new answers to old questions
 
how about an anti-FGITW reverse-bounty system: if you FGITW a post (definition TBD) 50% of the rep you earn from it is taxed and put into a community pool to distribute bounties for incentivizing things like answering old questions
 
That would actually be kinda cool though :p
 
10:16 PM
50% seems harsh but that sounds like a good idea
Also, some FGITWs are legit good answers, so you might want time constraints on that
 
i mean my idea was 100% a joke so it's not meant to actually be a working idea :P
 
10:57 PM
@RedwolfPrograms giving a bounty to the people who vote the most in a year in the CGCC best-of could help
that way, people are more inclined to vote to get their stats up
 
No way to know if those are all just question votes or FGITW votes, which is likely
I feel that would enocurage both of those
 
Chances are that if you're at the top of the leaderboard, you're doing more than just fgitw votes
 
Specifically seeking out answers to old questions is more time consuming than either of those
 
honestly it would just encourage spam voting which isn't exactly in line with the goal
 
@RedwolfPrograms We could publicize the necromancers list more. That wouldn't really incentivize voting, though.
 
11:02 PM
User script could help...I was working on one a while back
 
@RedwolfPrograms if you sort by recently active, you can see old questions that have been bumped
idea: bot that posts new answers to old questions
 
@RedwolfPrograms Like the Random one but for old questions?
 
I was thinking more like collecting all of the new answers to old questions that day into one question-like UI where you can scroll through them and vote on them
 
I think instead of anti FGITW like hyper suggested, automatically awarding +10 rep to answers on old questions would be nice
 
Uh...that's an unfortunate misspelling
Fixed
 
11:07 PM
> ans vore
 
@RedwolfPrograms Good idea
 
@RedwolfPrograms Next Best-Of, I'm posting a "slowest gun in the east" category
 
Also thanks YT for cutting off "What is associativity" to this:
 
Oh man, I have so many inappropriate jokes I could make :P
 
@RedwolfPrograms First ans vore, now this. Redwolf you doing okay today?
 
11:14 PM
One of my teachers' names is exactly long enough that it causes gmail to repeatedly send notifications to me about my teacher posting a new...uh...yeah that's why I disabled gmail notifications
 
Idk, I think Redwolf is making it clear he likes... eating :P
4
 
@cairdcoinheringaahing That's when you challenge someone to a gunfight at high noon, fail to show up, and then 30 years later you track them down and shoot them after they've moved back to New Jersey, gotten married, and raised four kids
7
 
This feels suspiciously specific, are you planning on shooting an old rival?
 
Not anytime soon /s
 
Maybe in 20 years? :P
Fun fact, tio.run isn't explicitly whitelisted in Smokey
 
11:18 PM
Considering taking the UI for my userscript from the review queues
Where the question title and details are in a little box at the top
And you just have the answer formatted as usual
 
11:43 PM
It's a shame the largest number printable challenge forbids constants >= 10, otherwise braingolf would easily win
^ with an empty stack pushes 1,000,000 to the stack. ^ with only one number n on the stack pops it and pushes n^n to the stack (exponent not xor)
^^ times out on TIO
 
aw dang I just saw that the challenge bans infinite score
 
yikes
 
> You cannot use the operations * (multiply), / (divide), ^ (power) nor any other way to indicate them (e.g. 2 div 2 is not allowed);
 
oh right yeah, that too
 
@lyxal considering the max length of 100 bytes i think that rule is just to squash any cheeky submissions that print something without terminating (or floating point infinity)
 
11:54 PM
@UnrelatedString but it means that if you can get inf under 100 bytes, you can't submit that
 
hence why "aw dang I just saw that the challenge bans infinite score"
 
i feel like most languages that have floats make it pretty easy to get floating point infinity in under 100 bytes
 
yeah that'd be trivial in js for example
 
infinity isn't a real number anyway
 
11:56 PM
_=>1/0 floating point infinity in 6 bytes in js. And half of those bytes are just to make it a function
 
₈ĖĖs in Vyxal if constants bigger than 10 are allowed, and you can use 1/x
because sorting a float bigger than 139 produces inf for some reason
 
braingolf doesn't do floats :/
 
00:00 - 15:0015:00 - 00:00

« first day (3918 days earlier)      last day (915 days later) »