« first day (5060 days earlier)      last day (50 days later) » 

01:24
I came up with another experimental golfing language help
Where tall narrow screenshot
This one would be more on the "golfy at all costs" side of the continuum
01:59
The idea is that the program is just a large number. The interpreter looks at the current stack, and produces a weighted list of all possible operations that can run according to their types and some hints which can be inferred statically (e.g., "this list is always finite" or "this number is always positive")
But there's a twist
It's able to look at past operations and run some predicates to determine if an operation would be "silly"
For example, if the last operation was negate, running negate again would be silly
(I assumed that was more or less where this was heading since I've also entertained a concept like that)
02:03
And these would be increasingly complex, taking into account things like commutativity and associativity of operators
and whether something would produce a constant that can be produced more concisely, whether it's a multi-op reimplementation of a single op, etc
 
1 hour later…
03:04
@rydwolf so, context-sensitive arithemetic coding?
neat :3
03:27
Copilot jumpscare
Pfft it gave a "write a readme for my profile" option
Trillion dollar company
Oh heck nah I got followed by a spam account
This is worse than that time someone posted a link to their nudes in the vyxal repository
03:49
0
A: Sandbox for Proposed Challenges

l4m2There are 10 prisoners in a prison. The prison has a central living room with one light bulb, which is initially switched off. Each day, the warden picks a prisoner at random and escorts him to the living room. The prisoner can either choose to turn the light bulb on or off, or do nothing, and th...

04:35
CMC: Given n, find the smallest positive integer who's prime factors sum to at least n
04:51
Assuming you mean all prime factors
Change the funny I to an uppercase one to go off distinct factors
Mfw it adds 3 bits
05:03
Is it smallest prime over n?
that definitely does seem plausible :P
is there any prime gap bigger than the prime at the start of it
The counter-example to this just being smallest prime >= n would be some n = p+2 where the next prime after p is greater than 2p
In number theory, Bertrand's postulate is the theorem that for any integer n > 3 {\displaystyle n>3} , there exists at least one prime number p {\displaystyle p} with n < p < 2 n − 2. {\displaystyle n<p<2n-2.} A less restrictive formulation is: for every n > 1 {\displaystyle n>1} , there is always at least one prime...
LMAO
So yeah, smallest prime over n
CMC: Given n, find the smallest positive integer who's prime factors sum to exactly n. (Always possible for n>=2 as 2*2*2*...*2(*3)
Any existing result about codegolf.meta.stackexchange.com/a/26615 ?
I can only find version where a single report is enough
@lyxal Why equal is shorter then ge?
?= probably happens more on the site
@l4m2 6 -> 8 otherwise use greedy algorithm (subtract largest prime != n-1 and recurse)
@Neil Input 148, Get 139x7x2, expect 137x11
hey, i couldn't count up that high in my head ;-)
06:39
61
A: Loopholes that are forbidden by default

DennisAbusing native number types to trivialize a problem It is common practice to restrict challenges to cases where input, output and/or intermediate values of the algorithm of choice fit into the language's native number type. At least for input and output, this is generally assumed even if not sta...

07:00
Frick ahh furries ruining capitalism
07:15
0
A: Sandbox for Proposed Challenges

l4m2Output all xxxs. No duplication allowed. A xxx is a finite set of pairs. Two xxxs A and B are equivalent if there are bijections F and G s.t. (x,y) in A iff (F(x),G(y)) in B. Shortest code win. Sandbox Notes Dup? What is xxx named?

Is it named?
 
3 hours later…
09:54
huh, F(n)^2 + F(n+1)^2 = F(2n+1)
@Neil that's... not true for arbitrary F
ah fuck hang on
i misse the n+1
yeah I copied this from OEIS so I'm pretty sure it's correct
@Neil golden ratio something something?
I mean, is there an intuitive explanation for this?
well, I have to admit, when I saw it, I was "what magic is this?"
10:05
that's also not true for F(n)=n^2+1
It can't be true for any function that can go negative
it works for linear functions only pretty much i think
no it's not even true for f(n)=n
It's not true for linear functions either
when in the fuck is it true then lol
@Neil Where on OEIS did you copy that from?
10:09
the page on F of course
Ah ok so we are looking for a very specific F
I assumed fibonacci?
That does start with F
Doesnt capital F refer to fibonacci in most contexts?
Or rather in this specific context, ig
It could be, IDK
10:20
It's fibonacci
7
Q: Induction proof of $F(n)^2+F(n+1)^2=F(2n+1)$, where $F(n)$ is the $n$th Fibonacci number.

abc... Let $F(n)$ denotes the $n$th number in Fibonacci sequence. Then for all $n\in\mathbb{N}$, $$F(n)^2+F(n+1)^2=F(2n+1).$$ I know how to prove it by using the formula $$F(n)=\frac{\left(\frac{1+\sqrt5}{2}\right)^n-\left(\frac{1-\sqrt5}{2}\right)^n}{\sqrt{5}},$$ but is there a way to prove it...

10:33
it is with great pleasure i inform you that despite The Daily Please Throw ARM Into The Sun Rant being canceled, EDA vendors are working overtime to ensure other, equally infuriating rants are created
Great, love to hear them
@lyxal wow there's a more general formula too
today's rant is directed at Synopsys, whose installers require you press a key to continue, because "You might be prompted for an input at a later stage.". if you haven't specified all the options for the installer, then you do indeed get prompted at a later stage, but if, like me, you are installing 34 IP products at once, you are of course incentivized to choose carefully your command line arguments and will not need to be prompted for input.
despite this, you will still need to press any key to continue even if all the options are satisfied, and once for each IP package you install
so you're saying it needs an option to disable that keypress?
or it could be intelligent enough to figure out that all IP packages only need --dir and --projectID
also the program i'm executing, on linux, to install the packages is a .exe which feel very wrong but i'm way past beyond that
 
2 hours later…
13:09
:/
i quite liked the mozz://a branding
The old logo was fine but the new one is nice too. Probably just needs some time to get used to
i do very much like the animation of the flag plopping on the ground it felt very nice
i guess i'm also not surprised but still disappointed at moz for spending idk how much on branding when it has better things to do with its money
The city I lived in spend millions of euros to add "central" to the name of the train station
For some reason no organization can resist pointlessly rebranding every now and then
true
my workplace did it recently, though they didn't have to spend any money just replacing "design" with "semiconductor" (yes, a final 's' is missing and it's sending me)
now we get to keep various "design" branded shit, i got a wireless charger for free
That's nice at least
Honestly, nothing I love more than getting random free trash
13:21
well none of my device have wireless charging but i'll be damned before i let a free gift go unclaimed
Exactly, I have like 4 terrible quality bluetooth speakers with different branding from conferences, but the terrible sound quality doesn't bother me as long as I don't use them
lololol
harvest the batteries from them at least
i just witness my manager take one of the free wireless chargers and start using it as a coaster for his coffee mug
fucking genius that's why he's a manager and not me
I mean if you don't have any devices that support wireless charging, you might as well
14:23
0
Q: The Graphing Calculator 3: Revenge Of The I/O

NatelolzzzPT 1 PT 2 PT 3 Last time, your challenge was to make a simple quadratic calculator, building on PT 1's simple line calculator... But continuing on the theme of increasing complexity (Hey, wait... is this a series now?) now... can you make it graph? (Use your PT 2 calculator as a base!) Write a pr...

@Themoonisacheese it's so simple and classy
I didn't even realize the new one was supposed to be a flag...
> Mozilla 👎
PFFFFF
It does SO feel like that
like what's that offset pixel even for???
a e s t h e t i c
not to mention it looks fucked up when the rest of everything isn't even on a pixel gid
grid
14:29
gid
@UnrelatedString for me it's the sans font except for the final 'a' which is serifed for some reason
i guess the font is generally ambiguously-serifed, that's even worse
yeahhhhhh
the a does also fuck with my eyes
and the serifs are so SQUARE too
wait I didn't even notice the bottom one has one curved corner but the top one doesn't
🤮🤮🤮
what the fuck were they thinking????
14:45
also the M on the flag isn't the same M as in the name
yeppp
and why even a flag
@UnrelatedString because it turns into the titular zilla
ahhhhh
so the weird pixel is the eye and it's just
that also explains the wierd offset pixel yeah
that would actually be SUCH good branding if it was the actual zilla and not a flag ????
think I still prefer the moz://a branding but that would still be super cool
14:50
i mean
at least now we can get dino game in firefox
because whatever the fuck is going on in the bar preference screen is not it
lmaooo
I was not aware of a bar preference screen game
(don't waste your time, it's really not that good)
...I could assume as much just from the URL 💀
i mean, if it was called "dino run game" would you think the same?
I'd think it had a chance
14:54
@Themoonisacheese LMAO
that's hilarious
i mean yes the concept of it is actually pretty fun
like you get to destroy ui elements that's fun but it's actually not very good
yeah actually reading the article it's so cute and fun that it exists but I do not care at all to actually play it :P
would be better if they, like, used the konami code to trigger it
 
2 hours later…
16:48
I can't decide if I should include the "bad ideas" section in my resume
On the one hand it shows I can reflect
On the other hand it's a terrible idea
Big piece of advice I got was to have one massive bloated "master resume" that you cut stuff off of differently for every job you apply to
but maybe that only makes sense in tech where resumes have to be kept down to exactly one page
One of my friends has an "evil resume" which is exactly what it sounds like and I think I'm going to make one
and submit it to lockheed martin
ahahahaha
>:3
Sections will include "agenda", "criminal history", "illicit skills", and "threats"
And maybe education, completely unchanged
17:08
ahahahahahahahhahaha
is this the gay agenda I keep hearing about
I think it's safe to say Rydwolf is never gettinga job
hasn't he already had a job
like in high school after moving out
If I understand the timeline right (It's quite complex) that was before the crimes
ah :P
that is a good point
@rydwolf also wait did something go wrong with the cybersec course or has it just been a lot of effort to keep up with
18:00
Turns out procrastinating a semester research project until a week before the end of the semester isn't a good idea
Especially when you're unaware until that date that 60% of the grade is mid semester checkins
18:51
TIL (derogatory) that Python type hints don't support higher-order generics
something like def single[I, C: Collection[I]](collection: C) -> I: is not valid
could you like hack somwthing together such that they does
so it can yes valid
not... really
I was able to rewrite my code to avoid needing that, but there are some things that it just makes impossible
can we ping python to fix
i forget are they receptive to customer complaints
18:56
just dont use python
CMC: given specifically a string representation of an integer, output the sign (specifically either + or -, and for zero, outputting anything else or nothing is fine)
@Ginger YIKES
@thejonymyster yeah they're aware of it
will be fixed soon™
oh thann goodness
i was worried itd be [tag:status: bydesign]
i offogt how to format that
the issue has been open for six years and was last commented on this Wednesday
3
19:01
forgot (cant edit on kobile??)
"such use cases are quite rare" my ass
 
2 hours later…
20:34
0
Q: Function that only errors

arutoneeDefine a function that, when given any input, will error. Defining the function itself should not cause an error. An ungolfed Python example is: def error(x): return [][x] # or lambda x: [][x] Shortest answer wins.

this is probably functionally a duplicate of something
since it's basically "print if the input is nonempty"
vtced as a duplicate of cat since I don't believe they're different enough to make the challenge interesting
 
1 hour later…
22:08
18
Q: Make your code error, but only sometimes!

rydwolfYour task is to create a program or function which randomly errors. Specifically, there must be a nonzero probability of erroring, but also a nonzero probability of running without error. An error is anything that causes a program to terminate abnormally, such as dividing by zero or using an uni...

That's more similar than cat
because cat is read input and write same output. This challenge doesn't necessarily need to act upon the input
fair enough
23:04
plenty of "throw a trivial error" challenges out there
y'know what i'll add a warning to the [error-message] tag
(43% close rate, that I know of)

« first day (5060 days earlier)      last day (50 days later) »