« first day (4279 days earlier)      last day (565 days later) » 
00:00 - 15:0015:00 - 00:00

12:04 AM
Update: GingerBot has now been running stably for 3 days
 
1
Q: Triangular polkadot numbers

BubblerGiven the infinite triangular array of positive integers, select every 2nd numbers on every 2nd row. $$ \underline{1} \\ \;2\; \quad \;3\; \\ \;\underline{4}\; \quad \;5\; \quad \;\underline{6}\; \\ \;7\; \quad \;8\; \quad \;9\; \quad 10 \\ \underline{11} \quad 12 \quad \underline{13} \quad 14 \q...

 
1:00 AM
Sandbox posts last active a week ago: Does the sequence shut the box?, Generate QR code from string
 
1:23 AM
0
Q: Construct a diverse array

dingledooperGiven an integer \$ n \ge 1 \$, your task is to output an array \$ a \$ consisting of \$ n^2 + 1 \$ integers, such that all possible pairs of integers \$ (x, y) \$ satisfying \$ 1 \le x, y \le n \$ exist as a contiguous subarray in \$ a \$, exactly once. For example, if \$ n = 2 \$, a valid outpu...

 
 
4 hours later…
5:52 AM
Me realising how incredibly annoying 99 bottles of beer is to golf
 
Yes
It really is
 
I'm at 371 bytes XD
 
 
1 hour later…
7:04 AM
Which language and/or library is golfiest for numerically finding the solution to an equation? Say you want to solve 4^x+6^x=9^x numerically?
 
@graffe Z3?
As long as the solution is integer, it deals badly with fractions
 
For general math, the best bet would be Mathematica
and for golfiness, Sledgehammer (which is basically a super well compressed Mathematica)
 
7:22 AM
@Bubbler I am guessing sledgehammer is not available to the poor?
 
think so
 
@mousetail that sounds a good idea. Could you show code for my example?
 
not that sledgehammer
 
Shame :(
 
totally different things
 
7:25 AM
My link looked fun :)
CMC solve 4^x + 6^x - 9^x = 0 to four decimal places at least
 
Something like this:

(declare-const x Int)
(declare-const y Int)
(declare-const z Int)
(assert (= (+ (^ 4 x) (^ 6 y)) (^ 9 z)))
(check-sat)
 
though sympy does have sympy.nsolve which attempts to numerically solve a function = 0
 
But it seems to be unable to solve it
 
This is a good site to play with z3: jfmc.github.io/z3-play
 
7:31 AM
(the 2 is where nsolve starts searching for a solution)
 
@Bubbler that’s cool
@mousetail thanks!
So sympy is winning…another other solutions?
 
interesting fact: if you enter a numerical value to wolframalpha, it tries to infer what it actually is (possible closed forms)
but it has a weird obsession to pi and e
so it fails to find that the value is actually log_1.5 phi
 
Btw anyone have thoughts about this argument - Users are arguing over some stuff, take a look at the comments
 
7:46 AM
Your comments? Or the comments in the other answer?
 
No not mine
The comments under the question, and under Mister's answer
I really tried to make my answer neutral, as I thought it's bets to avoid arguments. Did I do that right? :3
 
Yes, you express a opinion so people have the right to disagree.
Many people seem to agree, some disagree
 
If I'm correct (please correct me as I'm not super familiar with meta concerns), upvoting and downvoting in Meta is according to whether you agree or disagree right?
Not by usefulness
 
Yes
So even very high quality answers often have negative scores
 
Hmm I see
Thx for clarifying @mousetail!
 
7:51 AM
There are exactly 2 people who downvoted your answer and 2 people who upvoted the other one, probably the same people
 
Hmm might be a coincidence too, but a viable interpretation otherwise
 
There is no way to know for sure
 
One more question: Being neutral on meta is the best right?
 
What do you mean "Neutral"?
 
and with respect to what
 
7:56 AM
Like don't side with anyone
 
In opinion questions like this you need to include a actual opinion, ideally backed by other meta answers and documentation etc.
 
Opinions are different
 
If a question asks for your opinion provide your opinion, don't be neutral
 
I kinda meant don't forcibly place someone on one side of the argument and try to accuse
 
For sure don't call out specific names
that's different from neutrality though
 
7:57 AM
oh
@UnrelatedString ?
 
Unless of course you have reason to suspect only those specific people are causing trouble
 
@mousetail I see
 
Most meta posts are more about how to handle a broad class of situations rather than a specific one. So calling out the specific people behaving a certain way is unhelpful, since the goal is to create guidelines for next time.

However, if your conclusion is these guidelines already exists and just specific people are not following them for no good reason then you can call them out.
 
Hmmm agreed
Quick question can you flag a undeleted post? if your above 10k of course
 
Wdym? You can flag any post, including deleted ones
 
8:08 AM
You can flag nearly anything, other people, comments, flags, the sun
 
8:20 AM
@emanresuA but u have to be >10k to see it to flag it!!!
right?
or am I being dumb again
 
Yes of course
 
8:47 AM
@Bubbler that is interesting!
I guess I can't pose my CMC as a challenge a people will just write out 1.1868
which is a shame
CMC Find the first 30 digits of the solution for x^x - x^(-x) -x = 0
 
@graffe x≈1.5819790225351794653791459960674390425384594032203824871753101750854605877244182
 
@Fatalize code?
 
that's not code :)
 
It is, it’s just not mine
huehue
 
9:00 AM
:)
CMC output 3 different values of the triple a,b, c != 0 such that a^x+b^x - c^x = 0 has a solution for x.
@Fatalize harder to lookup in wolfram alpha :)
 
For any x?
or a given x
 
Impossible for x > 2
(fermat's last theorem)
 
That’s only for integers though
 
yes any x.. (4, 6, 9) is one. You have to find another three triples
not necessarily integers
(in fact I am not sure if there are any integer solutions)
@Fatalize is the question clear?
 
X must be the same?
for the 3 triples?
I can find as many as I want with X = 0 :)
 
9:09 AM
@Fatalize consider a =4, b = 6, c = 9
the solution x = 1.1868.. is a solution
@Fatalize but a =1, b =1, c =1 has no solution
so the task to find values for a,b,c for which there is a solution
and you have to find three such triples
@Fatalize is that clearer?
 
Not really
 
oh please tell me why
 
0,B,C is a solution for any B,C with X = 0
 
yes but I said none of them can equal 0
so none of a, b, c can equal zero
 
@Fatalize "the" solution
 
9:14 AM
@Fatalize Does it make more sense now?
 
Easy, take {i,j,k} = nth root of {1,2,3} and x = n
code: my brain
 
@Fatalize I like it. So let's take x = 2, a =1, b = sqrt(2) and c = sqrt(3)
@Fatalize Did I misunderstand?
 
it’s basically 1 + 2 = 3 everytime
 
very nice!
 
and you apply nth root to the triplet to counteract the nth power
infinite solutions, not just 3
 
9:19 AM
it's great
 
what do I win?
 
internet points!
 
._.
 
the solution 4, 6, 9 doesn't fit into that model it seems
 
I was hoping for loads of cash
 
9:20 AM
internet cash?
cash rules everything around me? (old rap track)
 
Works for any triplet a,b,c such that a+b = c
and any n for such a triplet
 
@Fatalize understand.. just wondering what model 4, 6, 9 is following
 
how did you find that it works for this one?
 
@Fatalize grammar... hehe
 
@DialFrost ?
 
9:22 AM
@Fatalize it's a math puzzle I was given. You solve it by dividing everything by 4^x and then noticing you have a quadratic equation
 
@Fatalize that sentence seems ... off
:3
 
Really? Feels completely natural to me
 
Btw guys I saw a notif earlier
It said something about a lang construction or smth?
@Fatalize it might be grammatically correct I think, but it sounds unnatural for me :P
I would usually say "How did you find out that it works?" or smth along the lines of that but oh well
U speak well enough (type) that I can't tell that your non-native :3
 
my eyes
 
@Fatalize ok new challenge, for which integer a, b, c < 10 does a^x + b^x - c^x = 0 have a solution?
 
9:25 AM
@Fatalize ? lol
 
your/you’re I don’t get how natives can mistake one for the other
 
maybe that should be my new CMC
 
@graffe I don’t get how that’s different? My approach still works
 
@Fatalize It's called slang :3 I type fast enough that I usually translate you're=>your
 
@Fatalize @Fatalize but it misses out 4, 6,9. The challenge is to find them all now
 
9:27 AM
oh integers
 
and integers to make the set finite :)
 
@DialFrost It’s called incorrect in my book
@graffe positive integers
 
@Fatalize why not
 
@Fatalize Ye it's incorrect, but I'm being informal :3
 
@graffe Otherwise the set is still possibly infinite
 
9:28 AM
@Fatalize well I would say the absolute values have to be at most 10
 
Just a tip: staying in sg for too long makes your english go haywire
 
Solution: try every triplet in wolfram and see if it gives a solution
Just from a few tries, I’m guessing there’s always a solution
 
@Fatalize I mean real solutions
 
And from the plots, I think you can show there is always one because a^x + b^x and c^x are both exponential shapes, one grows faster than the other to infinity but also converges faster to 0 than the other to -infinity, so they have to cross each other at one point
something like that
 
@Fatalize 7, 2, 3 doesn't have a real solution
 
9:36 AM
@graffe Didn’t expect that one could stay below the other all the time
 
@Fatalize also, if a, b or c can be negative, are there ever real solutions?
 
@graffe ? you mean are negative?
cause if they can, then just take positive numbers
 
@Fatalize for example, -2, 5, -3 (typo fix) has no solution at all
 
@graffe lol? x = 1 is one
 
what did I do wrong?
 
9:41 AM
you said 3 not -3
 
oops :)
 
But also Wolfram not finding a real solution doesn’t mean it doesn’t exist
 
@Fatalize that's a good point
 
But it definitely means that obvious approaches and tedious reformulations won’t work
 
@Fatalize this is good! Is there any way to make an interesting challenge from this?
I mean for the main site
 
9:53 AM
I doubt it
 
oh well :)
 
Most people don’t want to bother with floats
and you can’t find exact solutions anyway
 
@Fatalize ahem... sorry??
there are closed form solutions which is closed form enough
 
what does close enough mean
 
sorry about the typos
they could just list the integer solutions
 
9:55 AM
but how?
You think people will write code to find your 4,6,9 example?
 
two options a) list all integer triples with real valued solutions where the triples are below some size or b) output the infinite number of integer solutions
so b) would require them to find some pattern to follow I guess
 
Please think about writing a program that finds that 4,6,9 is an exact solution with no prior knowledge
 
is that for me?
as in should I be doing that thinking?
sorry I didn't quite understand what you meant
 
It’s a rhetorical question to make you understand that what you’re asking is not suitable for a challenge
5
 
@Fatalize I am not sure I agree. There are plenty of hard mathematical challenges
so it's more like "please write a program that will output integer solutions to this equation including 4, 6, 9"
 
10:00 AM
Finding exact solutions which involve non-integers, especially non-rational numbers, means you have to implement some kind of proof system, not just computations
you’re basically asking for people to implement a subset of Wolfram Alpha
 
I am not sure that is right
 
Which why I asked: can you write a program that will find your 4,6,9? Because I doubt you can, for good reasons
 
you don't have to find the real values for x. Just the integer triples. But you do need to be clever to be able to do that
@Fatalize I have more confidence in the genius users of this site than you ;)
 
If you need to find the integer triples that is an exact solution given any x, then you kinda have to find x anyway
 
@Fatalize not quite.. for example you define a set of solutions without explicitly working out all their solutions
 
10:04 AM
@graffe I was asking you to write it, not me. You can assume I’m a retard with 10 IQ, but you still haven’t shown that it’s a doable task
 
@Fatalize I definitely am assuming you are very clever!
@Fatalize I just meant that you already defined an infinite set of solutions
 
@graffe Yeah, and how would I write a program that would have come up with my argument automatically?
 
@Fatalize oh I don't think that is needed. The human writes the code
the human is the clever part
 
I simply don’t understand what you’re asking
 
the task is to write code to output all solutions, for example
or all solutions where the integer values are small
if I defined small to be absolute value at most 1 we can do it quite easily
but as this would be code-golf we don't want to just enumerate all the answers in the code if the range is larger
does that make sense?
@Fatalize I am sorry if I have been confusing/stupid
 
10:39 AM
ru kidding me
SE just logged me out of all my accounts
 
@DialFrost are you a criminal?
was it the result of a subpoena?
 
@graffe a what???
@graffe hehe maybe :3
 
@DialFrost As in "TRUMP REACTS TO JANUARY 6 COMMITTEE SUBPOENA LIKE A GUY WHO KNOWS THE WALLS ARE CLOSING IN" :)
 
@graffe I don't know about dialfrost, but as for me, depends if you consider stealing people's limbs a felony
And, yes, I realise that several governments say it is
 
@lyxal that is a great question and quite US centric
 
10:42 AM
but you're not several governments are you
 
we should have a cgse vote for that
a limb is normally 32 or 64 bits I think
 
@lyxal ? wuts going on I'm so confused :3
 
You don't know?
 
Not really
 
I'm the most notorious limb stealer to ever visit TNB
 
10:47 AM
hmm
interesting
@lyxal you dont sound tired anymore XD
 
do you nibble the limbs off?
 
@graffe nibble?
hell no
that'd ruin the resale value
 
nibble is 4 bits I think
 

Y'all just giving me more limbs to steal

Nov 4, 2021 at 5:57, 4 minutes total – 8 messages, 4 users, 0 stars

Bookmarked Nov 4, 2021 at 6:02 by lyxal

 
it would take between 8 and 16 nibbles to get a limb I think
@lyxal good point
 
10:51 AM
Oct 9, 2021 at 22:38, by user
@RedwolfPrograms Don't give him any ideas, we're tired of lyxal stealing our arms and legs, I don't we could cope without our tongues
see?
most notorious limb stealer to ever visit TNB
 
but you are still very welcome!
for anyone with a spare few minutes youtube.com/watch?v=qjAHw2DEBgw
 
11:04 AM
@lyxal heh
 
Since when has "Companies" been in the side bar?
 
a while
maybe since June, probably earlier
 
Huh, I'm blind I guess
3
 
Hell, it's been there as early as April 19
 
@cairdcoinheringaahing wow I just realised that lmao
 
11:17 AM
Never noticed either
 
12:00 PM
How do I make ATO links shorter :(
 
just use tinyurl
 
12:10 PM
Is it just me or is there a gap between rows of PFPs in the sidebar now?
 
horrible thought i had at 8 am: someone should make a language based around zalgo diacritics
 
already done :p
4
A: 1, 2, Fizz, 4, Buzz

user106728Fact, 11 bytes (SBCS) ef̀|3 5^ŹFh̃õ Explanation e 1e2 (i.e. 100) f̀ for i in 1:100 |3 5 Divisibility by [3, 5] (vectorizes) ^ String repeat by (vectorizes): ŹF "FizzBuzz" h̃ Halved (i.e. ["Fizz", "Buzz"]) ...

the repo is long deleted, but it was a language with zalgo
 
@lyxal Should make a language where a program is just one big stack of diacritics
so it’s 1 char wide
 
yessss
im assuming a bf derivative is possible too
with enough diacritics
 
I considered doing that once :b
never actually went through with it tho
 
12:17 PM
There was this recent challenge about code width that it won
 
found it thats hilarious
 
@Pacmanboss256 link?
 
21
A: Write a program with the smallest width

BubblerZalgo, score 5.221 0.222 ​̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̝̐̈̄̐̅̆̐̌̆̐̌̆̐̏̆̐̀̂̐̇̇̐̏̆̐̂̇̐̌̆̐̄̆̐́̂̐̀̂̐̄̅̐̈̆̐̅̆̐̀̂̐́̇̐̅̇̐̉̆̐̃̆̐̋̆̐̀̂̐̂̆̐̂̇̐̏̆̐̇̇̐̎̆̐̀̂̐̆̆̐̏̆̐̈̇̐̀̂̐̊̆̐̅̇̐̍̆̐̀̇̐̃̇̐̀̂̐̏̆̐̆̇̐̅̆̐̂̇̐̀̂̐́̆̐̀̂̐̌̆̐́̆̐̊̇̐̉̇̐̀̂̐̄̆̐̏̆̐̇̆̐̎̂​ Score this online! I think I found the perfect u...

 
12:35 PM
Love how it glitches the preview too
 
lol
 
Hard to believe that Ash development started (and finished) around 2.6 years ago
 
12:58 PM
@Fatalize Can we not use that word here please
 
1:15 PM
there are certain words that nobody should ever use, and that is one of them
 
Ah yes, the dreaded "haven't"
I hate contractions too
 
@graffe idk.. i think it is the one with most features
 
1:47 PM
@PyGamer0 it's a little sad as it has almost no devs
 
2:18 PM
0
A: Sandbox for Proposed Challenges

matteo_cGiven a word list, find all sets of five words of five letters each, such that the words of each set have 25 distinct letters in total. Rules Standard I/O rules apply. Words contain only lowercase (or alternatively, uppercase) alphabetic characters. You can assume there are no empty words (i.e. ...

 
Is CGCC being a bit slow for anyone?
 
Okay it seems like equality testing is very complicated even with symbolic math
WHAT DOES THE UNIVERSE HAVE AGAINST EQUALITY TESTING?
 
what
 
2:33 PM
Equality testing is more complicated than outputting the nth prime?
 
The way you do equality testing in symbolic math is by simplifying x - y and seeing if it turns into 0. But if you don't have every single possible way to simplify an expression implemented, which I doubt is even possible, your equality test might not work
 
what are you trying to do?
 
It's just representing math as a mathmetical expression
 
that sounds like a tautology: "represent math as math"
 
E.g., instead of a floating point, you'd store a nested data structure representing addition, multiplication, power, trig functions, etc.
 
2:36 PM
ಠ_ಠ
why would you do that tho
 
Because it gives you extremely accurate results
You lose a lot of data sometimes with floating point, and simple math can just fail for no reason (e.g., sqrt(2) * sqrt(3) != sqrt(6))
 
what are you doing that requires that level of precision?
 
It's useful in a golfing language
Since you never need to worry about precision requirements
 
or you could just never do challenges that require high precision
so that's your latest idea, a golfing language designed to be extremely precise?
 
No, that's just what modern golflangs do
 
2:39 PM
huh
 
E.g., Vyxal uses SymPy
 
I did not know that
 
@Ginger It's not just extreme precision, it's exact answers
 
so infinitely precise? :p
 
2:45 PM
cool
 
2:56 PM
Hmm...an existing Rust Symbolic math library is only a few thousand SLOC
It doesn't seem super comprehensive though
 
comprehend it then :P
 
Okay yeah its simplification capabilities are literally just like, "is the expression zero" lol
 
@RadvylfPrograms wow
is there no symbolic math library with as many features as sympy
 
00:00 - 15:0015:00 - 00:00

« first day (4279 days earlier)      last day (565 days later) »