« first day (3750 days earlier)      last day (1096 days later) » 

7:43 AM
I am thinking about an anti-account-spam bot
 
sup
 
the sky
 
@okie you people are still continuing to spam?
 
yes
two more new account were created
 
what are they
I hope you aren't fake
Also this Lyxal is not fake too
 
7:51 AM
no I'm not
also, you may have noticed the new picture
 
@lyxal yes i've noticed your username change too
Lyxal -> lyxal
 
lowercase letters go brrr
 
search "www" on Users
 
@okie there is no such user named "www"
but there is a Www
 
user starting with www
 
7:55 AM
@okie there are 7 such users
you mean two of them?
 
4 of them now
2 of them are new
 
bruh project
 
*moment
Ftfy
 
if you have a query you think will give a high amount of spam accounts you can let me know in my office and i can go through it privately with the mod search
though www is only useful in normal search
but i usually only do profile spam cleanup either by going through the new users page after i haven't touched it in a while, or by using queries shared in the TL
 
8:23 AM
wow i have got 16 upvotes for this dumb one liner answer: stackoverflow.com/questions/59613201/…
I can't believe it
 
8:35 AM
Could I get some feedback on this? I know it's long, but I want to make sure it is crystal clear.
1
A: Sandbox for Proposed Challenges

Wheat WizardChallenge Statement The goal of this challenge is to build the 5 state Infinite Time Turing machine that takes the longest to halt. The rest of this challenge is some definitions to help you. Infinite Ordinals In this section I will introduce the concept of transfinite ordinals in a somewhat info...

Also suggestions on how to make it easier to approach are welcome.
 
8:50 AM
@WheatWizard One or two concrete examples of ITTMs (and how you'd calculate the halting time for those) would be great
And a joke question: Is halting problem solved for ITTMs (I guess it halts unless every state combination has a valid transition)? :P
 
Non-joke answer: The ITTM halting problem cannot be computed by TMs, even ITTMs cannot solve their own Halting problem.
 
CMC: prove in as few words as possible that the halting problem is undecidable
 
@JohnDvorak A Turing machine that solves it could make the liar's paradox.
 
The challenge should be called Busy Beaver++ (or Omega Busy Beaver)
 
@JohnDvorak I halt? No!
 
9:04 AM
@UnrelatedString A better version would be "I halt? Yes and no!"
 
9:26 AM
@JohnDvorak If I halt,dont halt
 
Unrelated: I've been working on an interpreter for lambda calculus with combinators, and as a test case I managed to write add m n where m and n are Scott-encoded
It feels so good when it works (the interpreter AND the hand-written lambda code)
I decided to not bother with mul, though I'll probably try after the interpreter is complete and I get to solve some challenges in it
 
9:59 AM
Made myself a userscript to make everyone look like Lyxal.
This is going to be fun.
 
10:28 AM
joke challenge: (never to be posted to main) create a function or program which takes in a value and outputs a truthy value if it is truthy and a falsy value if it is falsy. do not take it seriously.
this is [code-golf], so the shortest code wins!
 
@ophact Brachylog, 1 byte: m
no values are truthy or falsy, so it always errors instead
 
10:48 AM
@ophact Javascript, 4 bytes: x=>x
 
11:30 AM
-1
Q: Shorter way to write the math.hypot (Python)

Ciprian I have two points as a list in Python: a=[1,3], b=[5,6]. Using the math library, the hypot formula would be: math.hypot(b[0]-a[0],b[1]-a[1]) Is there any way to write this shorter? (I am using Python 3.6)

 
@ophact Stax, 0 bytes
 
@Razetime Isn't that "Hello, world!"
 
no that's Stuck, not Stax
 
@ophact What if the given value isn't truthy or falsy?
 
11:45 AM
@Ausername done
 
12:08 PM
0
A: Sandbox for Proposed Challenges

pajonkImplement OADD_MONTHS code-golfdate Introduction The concept of a function to add a month to a date, without overflowing if we reach the end of the month, is implemented in many languages/packages. In Teradata SQL it's ADD_MONTHS, here are some examples: ADD_MONTHS('2021-01-31', 1) => 2021-02-2...

 
12:44 PM
@ophact You can find a thousand different solutions here :P
 
Someone give this some upvotes, a new user managed to outgolf both Neil and Arnuald by more than 20%
Also I meant to type out a message last night to say I was going to go since I was tired and had a test today but I woke up at 4:30 in the morning with my face on my laptop keyboard so I guess that didn't work out lol
 
ngn
@RedwolfPrograms impressive
 
@Bubbler even more here
@RedwolfPrograms wow there some math going on there
ah it borrows from level river St's thing
 
1:35 PM
@okie ✅
 
1:50 PM
0
A: Sandbox for Proposed Challenges

Zizy ArcherRadio station hopping You are listening to a car radio. You are pressing seek up/down, moving you to the next frequency some radio station broadcasts on, to avoid all this pointless music and listen to all the ads, or vice versa. But there is a tendency of these broadcasts to have your radio jump...

 
On a school laptop for a state biology test lol
When I opened it, the battery thing said it had 4:20 remaining, so I physically cannot fail
 
means you have to try for 69%
 
(Oh this is a sock btw I don't want to log in on my main acct lol)
@rak1507 That'd be cool, but I'm planning on speedrunning this test
 
who's your main account or do you not want to reveal it
 
It's a puzzle ;p
 
1:53 PM
I would guess redwolf
 
@GoldFarmerSprow lowercase p!
 
Why's that?
 
i feel like redwolf would be on his chromebook lol
 
It's me lol
Anagram
 
Is "Gold Farmer Sprow" an anagram of Redwolf Frickin ninja'd
 
1:55 PM
lol it is
 
Gotta use a school laptop for this test, malware and all
 
That's a lot of rep for a sock
 
+32 from 1 Q go brrr
 
@GoldFarmerSprow ...which is why you don't want to reveal your account details?
 
1:56 PM
Login details mostly
This thing's almost certainly root cert'd
 
you love to see it
 
      'Gold Farmer Sprow' ≡⍥{w[⍋w←⎕C⍵~' ']} 'Redwolf Programs'
1
it checks out
 
I should probably go now, see y'all o/
 
bye and gl
 
@hyper-neutrino I've gotten +80 on MM today thanks to that bump :P
 
2:01 PM
lmao. nice :D
 
CMC: Given a unique list of pairs (natural number, non-empty string), output FizzBuzz from 1 to 100, but the divisible tests and the corresponding strings are defined by the input. For example, the standard FizzBuzz would be [(3, "Fizz"), (5, "Buzz")]. The output for [(2, "A"), (3, "B"), (5, "C"), (7, "D"), (11, "E")] would be:
Actually, forget that. Imma make that into a challenge Nevermind, it'll just be a rehash of the FizzBuzz challenge
 
2:17 PM
poorly golfed
wait hold on
@cairdcoinheringaahing it's not always a single character, is it?
 
@hyper-neutrino I have 11 bytes in Jelly, taking the numbers and strings as the left and right arguments
@hyper-neutrino No, any string
 
@cairdcoinheringaahing oh right that would probably make more sense for io format lol
 
It's a CMC, so I'm not particularly fussed about the IO format tbh :P
 
oh right i can use ȷ2 for 100
 
@hyper-neutrino I have 17 bytes to take it in the [(2, "A"), (3, "B"), (5, "C"), (7, "D"), (11, "E")] format, ew :/
 
2:23 PM
lol
 
Managed to get 16 through some register abuse :P
 
@hyper-neutrino Hint: try iterating over each integer, rather than working on the entire list
 
I've got 11 for taking it as 2 lists, and 14 for taking it as a list of pairs
 
2:31 PM
is there a convenient built-in i'm missing somewhere lol
 
is quite helpful
 
@cairdcoinheringaahing Charcoal, 20 bytes: E…¹℅e∨⭆Φθ¬﹪ι§λ⁰§λ¹Iι
 
but i think i know how i can shorten this now
 
would be 17 bytes if TIO was up-to-date: E…¹℅e∨⭆Φθ¬﹪ιIμλIι (takes input as a dictionary)
 
2:36 PM
Charcoal has dictionaries now? Cool
 
took me long enough lol
 
@hyper-neutrino 12, if we're counting the Y :P
 
If not, I've got 10 and 13 :P
 
sadge
 
2:39 PM
Python, 234 bytes :P
 
i don't see a way to simply remove a byte from mine
does yours just follow a totally different idea
 
13 mins ago, by caird coinheringaahing
@hyper-neutrino Hint: try iterating over each integer, rather than working on the entire list
 
@cairdcoinheringaahing 19 bytes in Arn I believe
 
surely you're not using registers and mutation
 
make that 16, I didn't realise that Charcoal's dict keys could be integers (too used to JS): E…¹℅e∨⭆Φθ¬﹪ιμλIι
 
2:45 PM
@hyper-neutrino No. I'm using ȷ2µ...)Y
 
i can never figure out how to make ) do what i want
 
It basically just means µ€ (i.e. do the previous chain over each element)
 
how does µ...) even work?
oh wait do you use ³/
 
Yes
Full programs ftw :P
 
that's very nice
is this what you had?
 
2:48 PM
Yes :P
 
nice :P
 
witchcraft
 
My 14 byte version is just disgusting: Try it online!
 
^^ that's just jelly for ya :D
> +/
 
Summing strings :P
 
2:49 PM
on strings? what is this heresy
doesn't that break jelly's type system
 
oh wait it just sums the lists together doesn't it
 
Jelly has tons of bugs with strings, cause it's implemented in Python
 
you can use things like unhalve to get multi-character strings which can sometimes result in... interesting behavior
 
2:50 PM
i think dennis or someone exploited that for a clever golf at some point
 
2 character long Jelly strings have a length of 1 :P
 
wtf :P
 
jelly strings are certainly cursed :P
05ab1e's always been overall better with string manipulation lol
i think vyxal easily beats jelly too
 
> Ṁ Equivalent to having both m and M flags
We heard you like flags, so we put flags in your flags
 
2:58 PM
what's even the point of that if flags aren't counted lol
i think i'm only going to add flags for I/O-related things. i dislike flag abuse but i hate inflexible/silly I/O requirements more so i'll allow flags to avoid wasting bytes on unnecessary and trivial format-related issues lol
 
Is this for a new language?
 
yes, if i decide to actually make it :P
 
Add++ has a large number of flags, but all of them are "separate" and don't really affect functionality in a golfy way. So to use the e, f, i and o flags, you have to do ./add++ -f -e -i -o file.add, rather than ./add++ feio file.add
 
./add++ -fe -fi -fo -fum file.add
 
CMM: Should also be applicable to questions about the Reopen Votes review queue (used alongside )?
 
4:02 PM
@cairdcoinheringaahing regarding this edit
 
@flawr What about it?
 
is it a new rule that the tag is not enough?
(to specify the challenge type)
I just would prefer the challenge without that comment
 
Sort of. I added it mainly because I noticed the wrong tags while writing up an answer and edited it all together, but if you don't want it, the tag should be clear enough
 
hm I updooted ngns answer
while martin's answer has a greater upvote-downvote difference, the ngn answer has a greater upvote/downvote ratio
 
I've updooted both tbh :P If the poster is a high rep user and is familiar with the site, then Martin's answer is less applicable. As I said, I mainly just added it in while removing the irrelevant tags
Feel free to edit it out if you want :)
 
4:09 PM
well the removed tags would be the second gripe I have with that edit:)
 
doesn't really apply here
 
I agree with removing those tags
 
Just because it was inspired by a common builtin in functional languages, doesn't make it a challenge. is just a bad tag as well, so I thought I'd take advantage of the "bump" from my answer and remove them
 
ok I see the point regarding
but I would say I'd still keep as functions like inits are very characteristic of FP
 
Finished the test in 20m lol
 
4:16 PM
how'd it go
easy I assume :P
 
Yep
Everyone was done within an hour, I just now realized I'm allowed to be on my laptop now :p
 
@flawr I disagree. Pretty much every challenge involves black box functions, it's basically a different tag for
 
I think tonight I'll get my NMetaP and NSP (or NP and SP) bots working, then write a meta post on moving them here
 
@RedwolfPrograms While you were gone, this weird "Gold Farmer Sprow" user showed up and started claiming they were a sock. Very weird :P
 
Hmm, I say we nuke their account :p
 
4:20 PM
alright, will do /s
 
CMP: Should NMP and NMetaP be combined into a single bot?
 
I don't think Meta is active enough for two bots to be an issue tbh
Maybe if we were getting multiple posts a day there, then it might be better to combine them. But, rn, we haven't had a post on Meta for 3 days, I don't think it'll have so much activity as to overwhelm the chat room with 2 bots
 
Next CMP: New Sandboxed Posts, Sandboxed Posts, or Sandbox Posts?
 
@cairdcoinheringaahing I would absolutely not equate these tags
has a very specific meaning, in that you're not allowed peek inside the function, you may only observe and use input/output
(disclaimer: I created that tag:)
It seems to me that you would like to reserve for challenges involving higher-level-functions
is that the case?
(And I don't agree with that either.)
 
It's less that "I would like to reserve it" and more "that;s how it's used on the site"
I think pretty much every challenge involves higher-level functions in some way, so, even if the meanings are different, that's what it's become to mean
 
4:29 PM
Then we should discuss renaming the tag or retagging the questions appropriately
 
@hyper-neutrino You also have a Github account called James Smith :P
@flawr Yeah, that seems better than a chat discussion tbh :)
 
@cairdcoinheringaahing how the hell did you find that :p
 
ah, lol
 
I mean higher-level functions are frequently a part of functional programming but they don't have to be, and at the same time they are not exclusive to functional programming, so it seems that these two concepts are orthogonal.
 
4:30 PM
Apparently James Smith is a very mysterious person, there's a file called about_me.html that's been completely empty since 2016 :p
 
@flawr I'm happy to hold off on any more retagging if you'd like to start a meta discussion. You're clearly more knowledgeable about the topic than I am; I'm just going by site usage, rather than what the terms actually mean
 
@cairdcoinheringaahing then you're ust exacerbating the problem :P
 
@flawr Not intentionally; I thought I was tagging appropriately until this discussion :P
And I'm always in favour of cleaning up the tagging system
 
well I first would need to survey the question in that tag a little bit
there aren't even that many questions
 
There is also the new tag, "For challenges where an exact output is not required but some property must still be fulfilled."
 
4:43 PM
@cairdcoinheringaahing So I just went through all the questions, and I think all but two of them actually fit to the topic of FP. There are two that I'd retag to instead.
 
@flawr Which 2?
 
wait we don't actually have a black-box-function tag
or do we?
 
No, we don't
But we can create it if no other tag applies
 
it might have more uses that the FP-tag
 
TBH my biggest hang up is: I'm sure most of the ones tagged with relate somewhat to the topic of FP. However, there's clearly a difference between how the tag is used here and the actual topic of FP, as shown by the edit that started this discussion. So it's more, are there questions on the site that relate to FP, but don't fall under the current usage of the tag? If so, what should be done?
For example, the current usage is definitely more "this challenge uses higher order functions", rather than "this challenge relates to FP". Those two overlap significantly, but they aren't the same
 
4:48 PM
I have to leave right now but I hope we can pick up the discussion again later!
 
Yeah, it'll be a good thing to clear up. o/
 
27 mins ago, by Redwolf Programs
Next CMP: New Sandboxed Posts, Sandboxed Posts, or Sandbox Posts?
 
New Sandboxed Posts
 
I like the other two more, for two reasons:
1. I can reuse it for the Guild of Reviewers
2. It makes Sandbox(ed) what shows up on smaller screens instead of New for both
 
Isn't the current feed called New Sandboxed Posts?
 
4:54 PM
Yes
 
If you're basically replacing the feeds, why not stick with their names?
 
Well if we're merging the NMPs we'd have change that one's name anyway
 
ngn
5:20 PM
@WheatWizard tbh, i think it would be lucky if that challenge gets any answers at all
 
It's definitely not hard to make a valid answer.
 
ngn
5:35 PM
people of tnb, behold the power of being stupid but persistent:
 
java
 
ngn
Congratulations, the answer you gave to problem 756 is correct.
You are the 155th person to have solved this problem.
 
nice
 
ngn
(something wrong with imgur or the "upload" button?)
 
solved in k?
 
ngn
5:40 PM
@rak1507 no, c
 
aw
 
ngn
language doesn't matter much here
@rak1507 i did the previous one in k
 
755?
 
ngn
yes
 
how do you prounouce "tnb" btw
 
5:41 PM
@Wezl tee en bee
 
ngn
@Wezl ^
 
Hi pajonk
 
6:00 PM
> TypeScript, as you may have guessed from the name, is a superset of JavaScript - SE Blog
JavaScript, as you may have guessed from the name, is a superset of Java
 
ಠ_ಠ carpets, as you may have guessed from the name, are pets owned by cars
 
Pencils, as you may have guessed from the name, are a superset of pens
Actually I guess that's sort of true, they can both write and erase
 
functions, as you may have guessed from the name, are fun
 
> Typescript, as you may have guessed from the name, is a superset of Purescript
> Purescript, as you may have guessed from the name, is a superset of Applescript
> Applescript, as you may have guessed from the name, is a superset of Actionscript
> Actionscript, as you may have guessed from the name, is a superset of Clojurescript
...
 
Just one > at the start of the message for a quote
 
6:06 PM
Yeah but I wanted to start multiple quotes not 1.
 
Don't think SE allows that in a single message :/
 
> Wezl, as you may have guessed from the name, is a superset of Lyxal
 
Yeah that's why I left it that way.
 
@cairdcoinheringaahing I feel the tag is used appropriately, in my opintion the challenges that use that tag seem to relate to FP. The fact that some of them use higher-order-functions is ... a side effect.
 
@Wezl Hi Wezl
 
6:23 PM
@cairdcoinheringaahing I would propose keeping the FP tag as it is, and maybe consider creating a new higher-order-funciton-tag
 
Yeah, I think FP may be a bit overused but it is still a useful tag.
 
6:51 PM
Ash would have gotten 9 bytes on this, beating 05AB1E by 3. No Jelly answers for me to compare against.
Wait no
11 bytes because it's sequence
 
FTFY
one of the few answers where an RTL jelly explanation could make sense
LTR would be equally valid. maybe more understandable, idk.
 
@RedwolfPrograms Jelly answer
It's longer than HN's tho :P
 
Oh I missed that lol
 
wait how does that work
 
Also for some reason my Ash program causes an infinite loop in the interpreter despite...not using loops (?)
 
6:58 PM
oh he used dyadic chain
am I using features not existent in 2016?
i find it unlikely i could so easily outgolf jonathan allan, though granted it was 4+ years ago and he's definitely improved a ton since then too
 
It looks like the chaining bit was the problem
 
oh wait, Ɗ is new (although $$ works)
 
The 1 is completely extraneous, and the only reason he used Ç is because the helper link is broken into chains
 
rip no chain nesting, lol
 
7:13 PM
0
Q: Output the Bee Movie script

Tux1I know, it's an old meme, but I wanna see y'all try anyway. Output all of the text at this Pastebin link. Use as few bytes as possible. https://pastebin.com/raw/Z9WHGMr5

 
7:28 PM
@hyper-neutrino $$ doesn't work. 1,2,1 monadic links (grouped by Ɗ) is different to a 1,(2,1) monadic link, where the (2,1) link is monadic
 
oh
wait why lol
does Ɗ do grouping/chaining differently from the regular monad chaining rules
 
Because S,PƊ is parsed as one link (S,P), but S,P$$ is parsed as (S(,P))
 
would 1,2,1 not chain as 1 - 2,1 under monadic rules normally?
 
No, it chains as S(v) , P(v) with Ɗ, but S(v) , P(S(v)) with $$
 
never mind
i'm stupid
 
7:31 PM
Before Ɗ and Ʋ, understanding chains was so much more important to good golfing :P
 
it chains as v = S(v); v = ,(v, P(a)) normally but v = S(v); v = (,P)(v) = ,(v, P(v)) with $$
 
fun fact:
 
forgot that 2,1 calls on the left argument in a normal link but grouping it with $ makes the "left argument" the active value
 
That is a fun fact :P
 
well.. . . ¯\_(ツ)_/¯
 
7:33 PM
Also, pxeger left before telling us the actual fun fact :P
 
I thought it would be funny to leave you all in horrible suspense
Please tell me whether it was actually funny
 
same energy :p but slightly shorter time frame
 
"featured on ifunny.com"
 
btw there is no fun fact
...or is there?
no
jk
unless...?
 
there is surely a fact, in which case whether or not it is fun is subjective
 
7:35 PM
Nope. There are no facts whatsoever. And that's a fact!
 
@hyper-neutrino The very first tweet should've been something like this, and then, just before the site dies, the final tweet should complete it :P
 
Gonna go take my math final now o/
 
\o
 
gl, have fun \o
 
gl don't fail
 
7:37 PM
Remember, 2+2=5
2
 
in the definition of "exponent" in google translate ಠ_ಠ
it works fine in google itself but not under translate
 
"23 = 2×2×2" - Google, a $1.6Tn company
@hyper-neutrino missing matmul contains getitem setitem delitem getattr setattr delattr get set delete iter next len await aiter anext enter exit aenter aexit...
Would probably be better to generate them dynamically using members of the operator module
 
what on earth are enter and exit
are the a-iter / a-whatever async version?
also lol there was definitely a smarter way to do it but i was too lazy
it was for the memes anyway
 
enter / exit are for context managers, i.e. with ... as ...:
 
enter and exit are for with statements Frickin' ninja'd
 
7:50 PM
a are async, yes
Python quiz: what do get set and delete do?
 
get sends a GET request, set creates an empty set and delete is fancy del
 
@pxeger those are words you use when you say "get ready, get set, delete python"
3
 
lmao
@cairdcoinheringaahing unfortunately no
 
@pxeger I'm shocked, shocked! Well, not that shocked
 
Side note: currently listing to the SMM2 OST and I must say this is an absolute banger
 
7:53 PM
1000! > 1000 ** 1000 right?
 
no
1000! = 1000 * 999 * ... 1000 times but 1000 ** 1000 = 1000 * 1000 * ... 1000 times
Nobody answered properly, so I'll tell you: they're descriptors, and they're called when the object is looked up/assigned/deleted as a property of another object, and are used to implement methods, classmethods, staticmethods, properties, and ORMs
 
python is too wack, i only understand like half or less of __ functions and how to actually use them lol
 
TBF most of the dunder methods are supposed to be for internal use
 
Python is like if someone looked at Ruby, said "no that's too logical" and then had an epileptic fit
 
I remember reading an SO answer that basically said "If you're writing dunder methods, you're doing it wrong"
 
7:59 PM
@cairdcoinheringaahing no, not really
If you're calling them directly you're doing it wrong
Writing them is kind of the whole point of Python
How would you write classes without __init__?
 
ngn
@pxeger python is older than ruby
 
Well, obviously a couple are supposed to be used :P
 
Tkinter is a pain.
 
__init__, __call__, __repr__ and __str__ are pretty much the only ones tho
 
8:02 PM
If you're using a dunder method to try to make CustomObject(*args) ^ CustomObject(*args) defined, your code is likely to be much more readable if you just use a .method call instead
In other news, I can't spell :/
 
Finished my math final
 
How'd it go?
 
I spent longer on that 11 question test than my entire 58 question state biology test lol
Hopefully well, I checked my work where possible
 
8:16 PM
what was the content
 
Some trigonometry, one question about derivatives, a bit about polar and parametric graphing
 
TIL wikipedia has keyboard shortcuts
 
I'm drafting an idea for an answer-chaining, but I don't think the style of answer-chaining has been done before, so I'd like some general feedback. Essentially, the answers each complete one isolated "step" in an overall process (e.g. outputting a text line-by-line). There are a finite number of steps, so the chain will always end after \$n\$ steps. Each answer has a score (e.g. code length) and the user at the end with the best combined score (sum, average, whatever) wins
The "task" of the \$x\$th answer will always be known beforehand, as it doesn't depend on the \$x-1\$th answer, but the answers individually are "meaningless" until you put them together to produce the final result
 
8:31 PM
sounds like some of the strategy would be choosing when to answer
 
Yes, it probably would
Is it a bad idea? Is it an challenge? The fact that they don't depend on each other makes me less sure
 
You'd probably need to be more specific about the task and scoring, at the moment it doesn't seem like it'd work well on its own
 
I've got a task and scoring in mind, and I'm drafting it now. It's more "broad picture, could this work?"
If not, I'll just discard my draft
 
Turns out building a (good) text editor from scratch in HTML is a lot harder than it seems
 

« first day (3750 days earlier)      last day (1096 days later) »