« first day (1822 days earlier)      last day (3022 days later) » 

10:00 PM
oh keep them coming, is currently the most promising candidate so I can finally get Taxonomist :P
 
-1
Q: Strategies for representing a given large integer using arithmetic expression

Christian SonneI have a specific number in mind, but it's part of a challenge I'm doing, and I don't want people to do (all) the work for me. Here is a number which has the same digits, but shuffled: 571316791592616713457839947344722355446006667431463981539128135232831531309148844832184388929174866010641466366...

 
whatever happened to non-challenge questions? or are we not doing that anymore?
 
> This site is for posing
^ @AlexA.
 
"posing" is a word that fits there
 
10:04 PM
AFAICT that's a perfectly valid question, according to meta policy. No idea why people are downvoting/closevoting it.
 
@AlexA. ^
 
@Doorknob shrug Seems off topic to me but you can reopen it if you want.
"What strategies should I use" is pretty clearly too broad IMO.
 
"... for compressing a very specific piece of data"
it's not off topic anyway
 
Then why haven't you reopened it? :P
 
10:11 PM
because I have
 
Oh okay
Caching
If general questions like that are acceptable, that makes +10 for question upvotes seem like less of a good idea to me.
 
they are like 0.5% of our question volume...
 
it's not like we couldn't use some variety
 
Or +10 for question upvotes
 
¯\_(ツ)_/¯
@Doorknob I guess, but I don't see why we would want questions like that. As is brought up often, we aren't a Q&A site.
 
10:19 PM
@Doorknob thanks
 
@AlexA. this is old and I disagree with myself on some of the things that I said in it, but: meta.codegolf.stackexchange.com/q/1470/3808
 
> Kolmogorov complexity, informally, is the amount of code it takes to produce a given string of text. When posting a challenge in this category, please make sure it adds something new to the challenges that already exist.
 
@AlexA. We aren't mainly a Q&A site.
 
@AlexA. Doesn't mean we can't use the Q&A part as well. It's a bit unfortunate that our (very small) Q&A part is mixed up with the challenges, but I like that we have both.
 
@AlexA. Isn't that true of all tags?
 
10:21 PM
Makes us even more unique in the recreational programming community.
Ideally we'd have separate software for the challenges, such that we could use the Q&A as an actual Q&A, about golfing tips, maybe esolang design, challenge writing, etc. That would be a real selling point for this community.
can someone write an answer to that question to suggest factoring the number? if it has small prime factors, Mathematica would be able to factor a number like quite quickly, and the factorisation would be useful to compress the number.
 
It doesn't seem like factorization would help much unless there are a lot of repeated factors.
 
It's astounding to me that teleporting all cars to space and letting them reenter would likely kill FEWER people than continuing to drive them.
 
@AlexA.
Heya, watcha doin?
(Good night)
 
Looks like Doug from the movie Up
 
I WAS HIDING UNDER YOUR PORCH ............... BECAUSE I LOVE YOU!
Dunno if that's an exact quote, it's been several years since I saw Up. :P
 
10:34 PM
@ballesta25 sure, but we don't know how artificial the actual number is
 
It's probably a very special number.
 
@El'endiaStarman It's exactly how I remember it, FWIW :)
 
> (I can't include the actual string here, because it contains some CJK characters that are banned by SE.)
 
In other news, our dumb comments are now famous on StackApps:
 
What characters would those be?
 
10:37 PM
(see left sidebar)
 
HOW JUIC AVOCAD PLZ
 
Question that just popped into my head: what's the expected time for trial division? As in, the vast majority of numbers are smooth numbers, so they will be factored quickly, but semi-primes and primes are not factored so quickly. Assume that the range is 1 to n and the distribution is uniform.
Further assume that division is a constant-time process, and that we're not stupid naive enough to divide by 4.
 
Are we stupid enough to divide by 15?
 
Not sure.
Skipping even numbers is really easy. Skipping composites in general requires additional bookkeeping that isn't really "trial division" anymore.
 
Well, there are n/log n primes below n, I think.
 
10:41 PM
It's not worth mentioning 4 then.
 
Yeah, that's the prime number theorem.
So we're not skipping composites?
 
Hmm. nah. Only even numbers after 2.
 
That's the same asymptotically
 
true, but still :P
 
10
Q: Size of largest prime factor

lhfIt is well known and easy to prove that the smallest prime factor of an integer $n$ is at most equal to $\sqrt n$. What can be said about the largest prime factor of $n$, denoted by $P_1(n)$? In particular: What is the probability that $P_1(n)>\sqrt n$ ? More generally, what is the expected val...

 
10:43 PM
I think if you use up extra bookkeeping space to remember that 2 is prime, it's not real trial division.
 
From there the expected size of the largest prime factor of n is n^.62433
No wait
 
@feersum But all you have to do is start at 3 and add 2.
 
But you want the smallest prime facotr.
 
The expected value of the exponent is .62433, which isn't the same.
 
Well, I was thinking of fully factorizing a number.
 
10:44 PM
 
And you can stop at sqrt(n).
 
I'm not sure if there's an answer anywhere in it though
Doesn't seem to be an answer anywhere. Just a bunch of squiggly stuff
 
The bottom of the last answer says it's O(x/log x) in p_n
p_n goes as n log n
so O(n*log n/log(n log n))
~= n*log n/log n = n
 
@ThomasKwa What are you looking at?
 
Edit 6 of the first answer
 
10:51 PM
How does this n relate to my n, though?
 
5
A: Average smallest prime factors

Daniel FischerAsymptotically, only the primes matter. If $p$ is the smallest prime factor of a composite $k$, then $k \geqslant p^2$, and so the contribution of composites to the average can be estimated from above by $\sqrt{n}$ via $$\sum_{p \leqslant \sqrt{x}} \bigg\lfloor\frac{n}{p}\biggr\rfloor\cdot p \le...

This looks better
 
So... average smallest prime factor of 2..n is quasilinear in n
 
interesting
 
So we want to take the sum of the smallest prime factors, subtract out the ones from primes, and add in the square roots of the primes.
 
I don't think ASPF is useful for this problem.
Well, maybe it is... we just need some kind of weighted average.
 
11:03 PM
But the answer should be ASPF(N) + the integral from 0 to N of ( sqrt(x) - x ) / N
 
can i haz upvotes on this:
no u cant
 
11:21 PM
Anyone know any resources about relatively unknown/uncommon encryption method that you can do in your head while writing? Preferably equally easy to decrypt while reading.
 
The "do in your head" and "easy to decrypt while reading" is kinda restrictive...
There's stuff like making your own script, which is basically a substitution cipher though...
There's making a code (making a new vocabulary)...
 
I know. I'm thinking of something that is secure in a psychological manner (takes people longer to figure out) rather than strictly mathematically secure.
 
There's "add the value of the last letter to the letter you are about to write down."
CAT becomes CCV
 
That's not really that uncommon, I'm pretty sure I've seen that. Maybe there's a forum somewhere this kind of stuff.
 
There's not a "key" to that, though, just the hope that people won't think of a rolling cipher.
There's steganography.
 
11:28 PM
Would be interesting to see if there's any research on how good the average Joe can recognize simple encryption methods.
 
It might depend on your use-case... how long are you anticipating other people having access to your writing?
 
It's just general interest, I don't really have a use case in mind
 
I would say... almost any kind of transposition cipher would be out of the reach of most people.
As long as it's more complicated than "unscramble the words"
Although... there are online tools to solve larger scrambled sentences.
 
Well double transposition is a good cipher, but immediately cracked by frequency analysis. And it's pretty time consuming for longer text.
 
A simple cipher: replace all instances of a with 4, e with 3, i with 1, o with 0, s with 5, t with 7, and b with 8; increment adjacent letters by the replaced number, and reverse the substitution, adding the replace number to the value.
Easy to do once you've figured it out.
 
11:40 PM
Why b?
 
B <=> 8
@CᴏɴᴏʀO'Bʀɪᴇɴ *by the replaced number + 1
 
@Doorknob i try for thirtee minut and no juic
 
@AlexA. i try for THREE OUR
 
wy you have nott tryed?
 
11:48 PM
cuz i dun wanna
an cuz ees dun wurk
 
i wan tha juic from avocad so badly but kanno get
 
eezee whith new avodad juicr 400
 
thwo eet on da grund
 
insart avodad and got juic
 
SE mods come in What the...!
 
11:49 PM
@AlexA. insart avocad and then GOT JUIC???
 
ES
HAPPEN
 
duznt compoot
 
@CᴏɴᴏʀO'Bʀɪᴇɴ That was kinda my reaction and I'm not an SE mod... :P
 
@AlexA. giv proof that your happen rilly happnd
 
That said, this does prove the insanity of two of PPCG's mods...
There goes all hope for graduation...
 
11:50 PM
may b i can b 1
 
0
Q: Coding Convention Conversion

Dariusz WoźniakIn that Coding Golf, you should convert one coding convention with TitleCase to lower_case_with_underscores, and vice versa! Specification Change the casing in a following way: If underscore character is a delimiter, change the casing to Title Case without any of delimiter. If there are multi...

 
@El'endiaStarman "I'm not an SE mod" wait, what? looks at username, is blue
 
I assumed "SE employee".
 
user image
2
DON SKWEZ HIM
 
11:52 PM
okay how do you find these things
 
not how juic avocad
 
@Doorknob put a carrot in the message
 
@Doorknob Bird juice
Side note: The image is called "avocado-scream.jpg."
 
Do we have a quadratic formula challenge?
 
Yes
I remember because I helped golf someone's Julia answer
 
11:55 PM
Ah. Where?
 
uh
¯\_(ツ)_/¯
get rekt avocad
 
I can't find one... are you sure there is one?
 
16
Q: Diamond Puzzles!

Ashwin GuptaExplanation: Last year in math class, on homework we would occasionally get these extremely simple, although equally annoying questions called diamond puzzles. These were basically questions where we would be given a sum, and a product then were asked to find the two numbers which when multiplie...

Yes ^
 
How do you check if a string has a character in Pyth?
 

« first day (1822 days earlier)      last day (3022 days later) »