« first day (1581 days earlier)      last day (3276 days later) » 
00:00 - 13:0013:00 - 00:00

1:00 PM
@MartinBüttner weird, but it seems like ?+)`,(1*);1\1 doesn't print at all while ?+:)`,(1*);1\1 does (dif is the : option)
 
yeah printing with loops is a still a bit odd
 
@PhiNotPi How much time do you think it will take? :)
 
@MartinBüttner at least I'm not losing my mind then
could the modifiers order matter?
 
I don't think so
it would only matter between : and ; I think
all other modifiers should be independent of each other
 
Leaderboard - 1000000 elections:
    39556470.50 (  260.13) - Priam
    39521401.50 (  258.66) - BestViableCandidate
    39477036.50 (  256.81) - CircumspectBot
    39374872.50 (  252.54) - BasicBot
    39367359.50 (  252.22) - PersonalFavoriteBot240
    39331909.50 (  250.74) - StrategicBot12
    39300368.00 (  249.42) - Kelly's Favorite
    39210888.50 (  245.68) - SometimesSecondBestBot
    39206211.00 (  245.49) - Optimist
    39018849.00 (  237.65) - NoThirdPartyBot
    38849866.00 (  230.59) - NoClueBot
 
1:04 PM
great, less thing to try out :)
@PhiNotPi where is ExpectantBot?
 
@PhiNotPi Sad, my bot still don't do much better :/
Would have liked to be able to test it to improve the algorythm
 
@Katenkyo You could run elections yourself by downloading the controller.
 
I love how most of these bots have strategies, and a lot just get beaten by PersonalFavouriteBot..
 
Is it just me, or does the "history" link do more harm than good on starred messages
 
@PhiNotPi Not really, I'm not on my computer, used my tab' to dev' it, and checked it's syntax using internet :3
 
1:09 PM
When I want to look at a chat message, I want to see it in context, but often times I think that that is the "history" (meaning chat history)
 
So couldn't even say if it were really executing before your ran it ^^'
 
I rarely, if ever, care about the actual history of the message, and that is visible from the permalink
I think they should simply get rid of the dropdown next to the starred message
as the two most common actions (going back in time, and starring) are still available
 
I'll be back home Monday, will do some hard testing/improvement then :)
 
@Katenkyo Actually, I think I might have been running the older version.
 
@PhiNotPi Ho, hope it will be better for the 10M run so !
 
1:14 PM
Rolling back the leaderboard for now.
Okay, I also forgot about ExpectantBot.
Well, everything should be updated now.
Expect a 10m run to be complete about 2 hours from now.
 
@PhiNotPi Yeah, saw that it take almost 9 mins for 1M, it starts to be really long !
 
Lot of cities in my area have already broke records for rainfall this month two days ago. And it has rained every day since then. And it's going to rain today again.
And the best part of this is that it's been sunny almost every day. It pours down heavy rain for ten minutes to an hour, then it stops and the sun comes out.
 
0
Q: Determine the Dimensions of a Rotated Rectange

Calvin's HobbiesThis Stack Snippet draws an aliased white rectangle on a black background given parameters for its dimensions, position, angle, and the grid dimensions: <style>html *{font-family:Consolas,monospace}input{width:24pt;text-align:right;padding:1px}canvas{border:1px solid gray}</style><p>grid w:<...

 
1:30 PM
@NewMainPosts A rotated rectange?
Not a rectangle? :3
 
I want to know how long we can make that typo last.
 
@PhiNotPi Some days, maybe?
 
wihch one ?
 
1:45 PM
The xx .. oddity is odd.
 
just 2% it
 
Sure, it's not hard to "fix", just weird to have to fix the input just so that it can look pretty.
 
I have a possibly dumb puzzling question
 
I have a possibly (probably) dumb answer?
 
So, let's say you have a safe that takes in an X digit number to unlock.
However, whoever designed the safe was lax on programming and makes it so any consecutive X numbers that match the combo is appropriate to unlock the safe.
So if the code is 1121 and you hit 111121, it'll unlock the safe.
 
1:53 PM
I am not able to get within the error ..
 
It would take a long string to have all X-digit numbers as a substring.
 
getting 25 26 for the second exapmle. .
 
@Geobits yes, but not as long as if you had to enter all X-digit numbers consecutively
 
Right. So what's the question? Looking for the shortest?
 
What's the fewest number of characters to get all of the possible combinations :D
With the worst performance being 40,000 because you do them consecutively.
10^X * X = worst performance
 
1:55 PM
That's probably not a dumb question. Might get some rep on Puzzling, and I'm sure it'll get a good answer.
 
I'm sure it will too, but no way for me to figure out if its right :P
the things i think of when I'm half-asleep in bed from my dog being awake
 
@Compass 10^X + X - 1 I think
 
Questions like that often get proofs (semi-formal mostly) in the answers if they ask for them.
 
@TheE I have no idea if you're right! :P
 
Cos there are 10^X unique codes and then the last code will have (X - 1) more digits at the end
 
1:57 PM
But if we use that for the root case of X = 1
ah
sneaky!
it'd probably be easy to check with 2
im not a math guy ._.
 
@Optimizer I figured 5% at 24 min leaves 1.2 pixels tolerance, which seems like it should perhaps be enough, but maybe not. How about 8%?
 
I am right not trying to get an avg from all 4 sides. Let me see
 
@Compass if X=2 the number is 90010203040506070809112131415161718192232425262728293343536373839445464748495565‌​758596676869778798899
which is 101 digits long
10^2 + 2 - 1 = 101 ;)
 
@Calvin'sHobbies in 40 24 example, I am getting 40 28/29 ..
 
Same if you shrink the 'alphabet' down. For x=2 and only digits 01, I get 10011, which is 2^2+2-1.
If you can prove you can form a string where every digit starts a unique substring for x>2, that looks like your answer.
 
2:05 PM
@Optimizer When I measure the euclidean dist from corner to corner in paint I get 24.7. Are you sure your alg is right?
I'll increase the tolerance anyway, just to be nice
 
@Calvin'sHobbies okay. I am doing some thing wrong.. need to figure out what..
 
@TheE sneaky
 
2:22 PM
@TheE I checked your answer and it works out. But you probably already knew that long before I did XD
char[] arr = "90010203040506070809112131415161718192232425262728293343536373839445464748495565‌​758596676869778798899".toCharArray();
for (int i = 0; i < arr.length - 1; i++) {
    String a = String.valueOf(arr[i]) + String.valueOf(arr[i+1]);
    for (int j = i + 1; j < arr.length - 1; j++) {
        String b = String.valueOf(arr[j]) + String.valueOf(arr[j+1]);
        if (a.equals(b))
            System.out.println("Found match: " + i + ", " + j);
    }
}
No output (meaning you win!)
 
Oh, and even if there was some output, I don't think that would mean you lose. But no output means you definitely win.
 
@Compass That's a problem, if I'm understanding correctly, which has been thoroughly solved before.
 
I think the only difference is that De Brujin is cyclic.
 
im not a math guy x-x
 
2:26 PM
Meaning 0011 works for 2,2, but we'd need 10011. It wouldn't surprise me if this had a separate name and was solved, though.
 
You basically just add the first few button presses onto the end of the sequence.
 
see, this is why you people are math people
 
I'm not a math people. Really.
 
I've actually solved / written a proof about this problem before. (non-cyclic)
 
I don't even have a math minor. If I did I'd be in jail! duh dun tss
 
2:28 PM
@Compass Wikipedia has your exact situation :P
> The sequence can be used to shorten a brute-force attack on a PIN-like code lock that does not have an "enter" key and accepts the last n digits entered. For example, a digital door lock with a 4-digit code would have B(10, 4) solutions, with length 10000. Therefore, only at most 10000 + 3 = 10003 (as the solutions are cyclic) presses are needed to open the lock. Trying all codes separately would require 4 × 10000 = 40000 presses.
 
2:40 PM
I answered the question by the link here, but I actually didn't read any of the conversation after your message here. (I happened to have the knowledge of De Bruijn sequences before, please don't blame me for just stealing the link to Wikipedia + the answer from @PhiNotPi :P)
 
I wonder if it would be any more interesting if we were looking for a subsequence instead of substring. Say the lock, once accepting a correct digit, doesn't reset and just ignores input until the next correct digit. (Terrible lock design, but whatever)
So then for (2,2), 0110 would work.
Nevermind, that ends up being really stupid. Forget I talked.
 
hmmm, surely that must have been studied... but due to the inconsistent use of "substring" and "subsequence" by some people my googling efforts are just turning up normal de bruijn sequences
 
You just need x copies of each digit, alternating.
 
oh yeah
do you even need to alternate?
oh
 
Well you can't just do 0011, then there's no 10
But 0101 works.
 
2:47 PM
yeah I was thinking you were talking about alternating the order, as you did in 0110
but yeah, just cycle through the digits n times
 
Like I said, it ends up stupid :D
 
I guess it would be more interesting for arbitrary sets of strings
I really don't get Area51 notifications
for my Nice Answer badges I only got a notification bar at the top when visiting Area 51. now I just got yearling which appeared in the normal achievements tab, but I didn't get the notification bar on area 51
 
3:07 PM
@Pietu1998 I didn't know they were called De Bruijn sequences so thanks :)
 
3:26 PM
Haha, I hit the reputation cap for the day before I even woke up. Now what do I do for the rest of the day?
 
Find a different SE site to terrorize for the day?
I hear Parenting is a good choice for that...
 
i think it would be interesting to see someone code golf a de brujin sequencer
 
The two-hour run of Strategic Voting just finished.
 
drumroll
 
Leaderboard - 10000000 elections:
   389541239.00 (  742.97) - ExpectantBot
   389008250.00 (  735.93) - Priam
   388528563.50 (  729.59) - BestViableCandidate
   388473282.00 (  728.86) - CircumspectBot
   387915294.00 (  721.48) - PersonalFavoriteBot214
   387190718.50 (  711.90) - BasicBot
   386894139.00 (  707.98) - StrategicBot949
   385708157.00 (  692.31) - Kelly's Favorite
   385152795.00 (  684.96) - SometimesSecondBestBot
   385024630.00 (  683.27) - Optimist
   382059378.50 (  644.08) - NoThirdPartyBot
 
3:40 PM
I think I might put a bounty on my array question
 
@Sparr which is that?
 
9
Q: Remove the mininum sum seam from an array

SparrThe seam carving algorithm, or a more complex version of it, is used for content-aware image resizing in various graphics programs and libraries. Let's golf it! Your input will be a rectangular two dimensional array of integers. Your output will be the same array, one column narrower, with one ...

I'm not exactly hard pressed for privileges; I can afford to lose some rep.
 
@Sparr I liked that algorithm very much. Surprisingly never heard of it before.
 
The graphical applications of it are pretty awesome. I was amazed when I first saw it in action.
 
yeah, I meant that part, seems like magic
 
4:00 PM
I actually expected to see some really short solutions quickly.
As it turns out, finding the minimum sum is super easy in a bunch of languages. Actually keeping track of the seam that generates that sum and removing it... not so much.
 
4:37 PM
puzzling.stackexchange.com/questions/15700/… a 0-key solution for the faulty safe
 
Despite my best efforts, I could not reproduce that issue.
Now I have to explain all these dead coyotes...
 
Add roadrunn.er
 
That wasn't in the bug report :(
See, this is why testers need to include all information. It saves lives!
 
has this idea ever been posted as a challenge? codegolf.stackexchange.com/questions/12042/…
 
ahhhh crap, I totally missed 44,444 this week :(
@Sparr it's definitely been sandboxed
 
4:45 PM
@MartinBüttner oh, I was also rooting for that
must. golf. further.
 
time to offer a bounty...
 
lol
 
A few days ago, I noticed that I had completely missed 8192.
 
don't really know what for though... mathematical analysis of Fission numbers?
(I feel like there was something a couple of days ago that I considered bountying but I forgot what it was...)
 
@MartinBüttner yeah, but did it run? It seems neat and simple.
 
4:49 PM
any tips for golfing that 55 byte monster ?
 
@Sparr I'm not sure... it's a bit hard to search for. it doesn't seem terribly interesting though, since it's just another generalised quine.
there's this related proposal: meta.codegolf.stackexchange.com/a/2001/8478
 
@Sparr actually outputting a self-descriptive string would be more interesting then outputting a string describing the source code.
(i.e. something like the first proposal I linked)
 
5:04 PM
I'm not sure that it's exactly analagous to a quine. The characters not having to come out in the source code order seems to make it different.
That is, there are ways you could further compress this challenge that you couldn't for a quine.
 
uhhh, I'm not sure I understand
 
in a quine, the position of the characters is information you have to store somewhere
that is extraneous information in this challenge
 
@MartinBüttner saw urs after posting mine ...
 
hmm, maybe in some languages. in CJam it's easiest to just sort the string representing the code and RLE it. (see my answer)
 
I spent about five minutes trying to come up with a way to word the challenge to exclude that approach in CJam and other languages with a "convert this block of code to its string represenation" function
 
5:11 PM
@Optimizer oh, I was just about to start golfing mine... yours is nicer
 
that function violates the spirit of the "don't read your source code" rule
I couldn't think of a good way to word it, though.
 
That is never considered as reading own source code
 
I think stringification is a legitimate way to write a quine
I could probably do this in the same amount of character with evaling a string instead of a block
one more character: "`'_'~++$e`{W%S*N}/"_~
 
'_'~]s
 
neat
wait, does that work?
 
5:16 PM
yeah
 
that doesn't create quotes, does it?
 
weird, right ?
oh. right
with quotes... it doesnt
i feel that if i dont write explanation of the two answers... i would loose potential up votes. but BGT...
 
0
Q: Character counts in source code

SparrWrite a program that outputs a list of the number of occurrences of each unique character in its source code. For example, this hypothetical program {Source_Print_1}; should produce this output: ; 1 P 1 S 1 _ 2 c 1 e 1 i 1 n 1 o 1 p 1 r 2 t 1 u 1 { 1 } 1 The formatting should match this examp...

 
you should know
 
5:19 PM
@Sparr I've added the quine tag. however it's solved, it is a generalised quine task.
 
the more I use wordpress, the less I like it
 
@NathanMerrill how about Ghost?
 
@MartinBüttner cool, np. you're right, maybe generating a self-describing sentence is more interesting.
 
@NathanMerrill when do you breakeven ?
 
@MartinBüttner never used it
 
5:21 PM
neither have I, but it looks sexy and uses markdown for formatting :)
 
@Optimizer already in the negatives, but my employer keeps it in the positive
the theme we are using uses something called Visual Composer
which is basically a bloated WSIWYG
 
@Optimizer I deleted mine btw
 
but it means I can't even edit html
 
@MartinBüttner sorry!
 
I have to figure out some weird language they created for VC
 
5:22 PM
@Optimizer nah, don't worry
 
5:49 PM
I just noticed that it looks like my Fission quine is now the shortest legit quine on Golf you quine for great good. o.O
 
6:21 PM
@MartinBüttner What's #0 in your quine?
 
Calvin is eager ....
 
6:39 PM
@Sp3000 the function definition. see this
 
You gotta simply love Dennis sometimes.
 
@Sp3000 is there no standard exec quine for Python that requires less code repetition?
 
So that's why my ears were ringing. :P
 
Dennis the Menace!
2
 
First time I come up with a non-standard quine and I don't get any points for it. Stupid rep cap!
 
6:45 PM
tell me about it!
 
Is the rep cap a per site thing?
Or overall?
 
per site
 
200 per day per site.
Only applies to upvotes.
 
but it can get screwed up sometimes
 
Possibly suggested edits.
 
6:47 PM
yeah especially with unupvotes
 
if you have potential 215 and someone retracts up vote, you get 190.
bam
 
@Dennis no downvotes also count, (so if you get a downvote, you can still reach 200... if you get an unupvote, you can't)
 
I meant rep from elsewhere isn't lost. Bounties, accepted answers and answer accepts are limitless.
My favorite: 20 upvotes you got no rep for and the next downvote costs you 2 points.
 
16
Q: New 10K tool: question close statistics

Shog9Two years ago, we rolled out some significant changes to closing, along with a massive number of changes behind the scenes to clean up the supporting logic that had grown crufty over the years. As with many ambitious projects, this one quickly pushed the limits of its 6–8 week implementation sche...

 
That's pretty damn sweet.
 
7:01 PM
I wish they were sortable.
 
It'd be nice to have links, too. Like a link to all closed for "too broad", etc. But still.
 
yeah
 
But especially links to the custom close reasons at the bottom. I can imagine opening the tools thing to see "I've voted to close this as off-topic because you're stupid", only to not be able to find what post it was on :D
 
@Geobits post an answer to the MSE announcement?
 
I would if I had an actual example. Apparently people flag them and they get cleaned up, making it a non-issue. I was sure I'd find a good example on SO, but they're pretty tame, tbh.
 
7:05 PM
I'm quite surprised that unclear is our most common close reason
 
I thought it would be broad or dupe.
 
would have bet on general programming questions or objective winning criterion.
 
we have high standards
 
no, we actually do. input format, output format, this, that, precision, brute force, blah blah, cats, nemo
 
7:06 PM
@Geobits it's quite amazing that we both named two other reasons and they're entirely disjoint, covering all (reasonably) possible reasons :D
@Optimizer that :D was about Geobits' and my guesses
 
@MartinBüttner Yea, but you won. Yours add to 23, while mine are only 17 :(
 
so.. I am not that funny ..
 
Nope
 
:(
 
:D
 
7:07 PM
@Optimizer no, not at all actually.
 
I blame the llama
 
@Geobits then again, 30 days isn't all that significant
 
Yea, I was wondering about the short range thing. It's hard to call it really useful without being able to compare with longer trends.
 
@Geobits there's an example for your other request: meta.stackexchange.com/questions/257449/…
 
the 'SN5 CJam solution is what I really hoped to see, in terms of cleverness
@MartinBüttner ^
 
7:15 PM
it is nice!
 
^ So cute. Yet so scary.
 
ppcg esolang stupid idea: define a language in which every unique string of characters defines exactly one specific program that solves a challenge on ppcg, with the language spec updated just a little too late to use it in any challenge unless an exact duplicate is posted later
 
I prefer reptiles with legs
 
@Sparr standard loophole
 
@Sparr problem being that no interpreter exists for any of these languages
 
7:17 PM
@Optimizer he's hiding or conserving heat, I think. unless my snake-fu is rusty, I think he can only strike from that position as far as the central S-shape reaches, not the coiled bits.
 
@Optimizer I think user23013 proposed that as a loophole but removed it because due to the interpreter thingy it's not even an issue
 
@MartinBüttner is a working interpreter a requirement?
note to self: do more work on marbelous interpreter at some point
 
yeah, otherwise you could solve challenges in English
 
@Sparr He's just hanging out, no strike-look to him right now (if my old snake's behavior is any guide).
His neck would be tensed up a bit more, drawing it back some instead of relaxing on his coil.
 
@Geobits oh, sure, if he's your pet then you know that. I was just referring to how far he could strike, if he was a stranger-snake :)
 
7:19 PM
@Sparr I've been question Marbelous's Turing completeness today
 
@MartinBüttner is there a simple challenge the completion of which would establish turing completeness?
 
Simulate a turing machine? :P
 
@Sparr interpret brainfuck
game of life, rule 110, simulate a 2-stack PDA
 
cyclic tag system
 
tag system
ninja'd
 
7:21 PM
yeah, rule 110 is what comes to mind when I think of crazy simple things that exhibit completeness
 
I've proven Retina's Turing-completeness with tag systems and rule 110.
 
New Challenge Posted: Prove your language is Turing complete! ;)
 
the problem with Marbelous is that the amount of storage you can have seems to be limited by the code (not the machine it's running on)
@Geobits hasn't that been done before?
or at least sandboxed
 
Maybe. Probably. Too lazy to look.
 
@MartinBüttner yeah, I've contemplated ways to change that.
 
7:23 PM
@Sparr you could take some inspiration from Fission's stack and queue I guess
 
I was slightly disappointed that the challenge wasn't actually posted right then.
 
I was thinking of ways that involve teleporting marbles around based on other marbles.
 
2
A: Sandbox for Proposed Challenges

QuincunxProve Your Language is Turing Complete Write the shortest code possible to prove that your language is Turing Complete. However, if the programming language is not Turing Complete because of something like limited access to memory, just state so. You can prove your language is Turing Complete i...

@Sparr but that still only allows you to have one marble per cell
 
sure, but if you can teleport marbles an arbitrary distance, then you have an unlimited board size to work with
instead of requiring a destination portal, make a two-input portal that sends one marble X cells upwards
 
ah yeah, fair enough... but you'd also need to be able to rewrite the board to retrieve them again
oh, upwards
yeah that might work
 
7:25 PM
it might. I don't think it will, but I'm no authority on the matter
and I'm perfectly ok with it not being turing complete
for now :)
 
I'll probably try proving Fission's Turing completeness some time soon, but I'll wait for the changes the author wants to make
XD
"I'm voting to close this question as off-topic because it is the obvious audit review task, probably."
@Geobits ^
(found on SO)
 
7:41 PM
obviously probable ._.
 
for my assymmetrical koth (catch the cat) should the catchers/cat submissions be in different threads?
id like to do just one
 
there's only 1 S in asymmetry
 
@MartinBüttner Nice. Someone posted the request as an answer now, though, so I'll have to content myself to one answer on that question.
@Compass Maybe it's a symmetrical ass?
 
that is how I read it, yes
 
cops and robbers usually have to separate threads,, do they?
 
@flawr: The first ones didn't, but they should now.
 
what about the 'assymmetrical koth'?
 
17
Q: Where should we put robbers?

DennisJust in case you didn't know already, we have a fairly new type of challenge: cops-and-robbers I enjoy these challenges very much, but I don't know if the current format (cops post answers, robbers post comments) is optimal. I see the following problems, in descending order of importance: Com...

 
I think the cat/catcher can be in the same post, as long as it's clear that they need to be labeled.
 
0
Q: Double-slit Quine

PlarsenObjective: Code two programs where each of the programs outputs both sourcecodes interlaced per character, a Double-slit Quine. The output from a Double-slit Quine is starting with the first character from the sourcecode of the first program. If the sourcecode of one of the programs is ...

 
7:49 PM
@Plarsen "Zipper" may be more idiomatic than "Double-slit"
 
We'll consider this once Oded spends some time optimizing the queries - I agree that 90 days at least would be useful here. — Shog9 ♦ 53 secs ago
 
@Geobits when some players make multiple submissions: all in one answer or each an own answer?
 
If each answer is unique normally it should be a separate answer.
Like, different languages and stuff
 
ok
 
Hmm, I've had players do both in the past. It didn't really bother me much either way, but I'd normally prefer one per answer.
 
7:52 PM
ok well then i'm gonna write it down that way
i try to make a graphical output this wkend and then i think i can start it
 
@Geobits nice
 
@Geobits the biggest issue comes when you have two languages, and one is horrible, the other is good
or one is wrong
at least then you can get credit for each individual language (or lose credit) for your rep.
 
I agree, as an answerer it's suboptimal. As a challenge runner, I don't care so much.
 
8:09 PM
@Calvin's Hobbies Yeah, that is true .. but perfect is boring ;)
@Calvin's Hobbies I edit it for clarity
 
I hate combined answers everywhere on StackExchange
lots of meta sites endorse them, which really annoys me
it breaks the voting system
it interferes with the comment system
oh, the top/accepted answer has 9 separate solutions in it? let me flip some coins to decide which of them to use
@MartinBüttner ^^ this is very very slightly aimed at your OEIS-aggregate answer to my meta question about math contributions.
if that answer grows, it's going to contain some very notable things, and some very unimportant things, and I am annoyed that I can't vote up the notable things.
 
8:35 PM
@Optimizer: Are you working on the double-slit thingie?
 
@Dennis Oh my... Are you being generous to not work on it if I am ? :P
You'd probably beat me big time anyways
 
Nah, I already have a solution.
 
and no. watching VGHS right now
 
I want to know if I should post it now or after midnight (UTC).
:P
 
so you are deciding between ruining my dreams now vs kicking my ass later ?
 
8:38 PM
You sound like you've never beaten me before.
 
Have I?
 
I remember Martin and us participating in a challenge and you beat me both.
 
must be Christmas!
 
11
Q: Calculate the troll toll to safely pass

tolosInspired by http://puzzling.stackexchange.com/q/626 In your adventures you arrive at a series of 7 bridges you have to cross. Underneath each bridge lives a troll. In order to cross the bridge, you must first give the troll a number of cakes as a percentage of the number of cakes you are carry...

 
@Dennis "me both"? Martin and Dennis are the same person...
 
8:44 PM
LOL
 
You both beat me?
 
nope
 
Not sure.
 
I meant, that first line is not equivalent of this one
 
I can see how it works, it's just more ambiguous that "you both beat me"
But it would be cool to know if someone had a high-rep sock puppet
For all I know you're all chatbots of Doorknob
6
 
8:54 PM
yes
 
Llamaknob revealed
 
9:27 PM
@PhiNotPi About how many elections does it take for the winners to be differentiated?
Do you think 1000 would be enough for testing a genetically evolved bot?
@Calvin'sHobbies DoorKnob isn't (maybe) ;)
 
9:59 PM
is there some way you could have less-random elections and just put the bots in a different order?
like card games where two teams play the same decks to see who is really better
 
10:09 PM
@TheNumberOne There's a reason I run 10m elections.
I think, if you look at the normalized scores, that scores that are about 3 points away from each other are differentiated.
 
10:35 PM
0
Q: Improving a Rubik's Cube scrambler's byte count

molarmanfulOkay, I recently wrote a Javascript Rubik's Cube scrambler that outputs a 25-move scramble. It's 158 bytes, but I feel like the code can be shortened further. The Gist is here, but for reference, I have the formatted code: function(){ for(var a=0,s=y=r=[],m='RLUDFB'.split(''),f=["'",2,''],x=Ma...

 
11:14 PM
hey any mac users in here?
has anyone ever made something in Pages, and then exported to pdf, to find that the pdf has moved text around?
 
11:56 PM
3 minutes and 1 upvote to rep cap ..
 
00:00 - 13:0013:00 - 00:00

« first day (1581 days earlier)      last day (3276 days later) »