« first day (1573 days earlier)      last day (3274 days later) » 
00:00 - 15:0015:00 - 00:00

12:01 AM
I found a funny, albeit slight horrifying, video by The Onion earlier today. "New App Lets You Work For Your Company Even While You Sleep" link.
 
sounds like a microsoft ad
 
@PhiNotPi yes i'm trying to work out another algo for the voronoi thing
@MartinBüttner would you like to help as well
 
uh, sure, but I'm only checking the chat occasionally at the moment (and won't have a lot of time tonight)
 
haha no worries
 
the "Coin Tossing Time Travelers" doesn't really need the time-travel mechanism to work :(
 
12:07 AM
so the deal is that IDL has the WATERSHED function exelisvis.com/docs/WATERSHED.html
 
reverting back your memory adds little to the game
 
sec
here's an example of the output from watershed (colorized)
so my thought is, if i can figure out an intelligent way to merge together the polygons there, i just do that until i have n
and then the polygon centroids will be my voronoi sites
i've been trying a ton of different ways to merge, and everything either fails horribly or doesn't get me far enough. (or runs my computer out of memory before it finishes.)
 
I'm looking at your current algorithm.
 
this would be another answer
 
Okay.
 
12:14 AM
sharing only the output means (i.e., triangulate the sites, draw the voronoi polygons in the z buffer, output)
my current algorithm starts large & refines
i want to start small & unrefine
for this method, i mean
my first idea was to create a heirarchical clustering tree for the watershed polygons in 3D (with the 3rd dimension being the watershed number), and then merge up the tree until you end up with N clusters.
but creating the cluster tree crashed my machine
my second idea was to turn the watershed vertices into a mesh, then decimate it
 
I think the key to creating a good image is to ensure that as you're merging, that the edges of the voronoi diagram continue to line up as much as possible with the lines of the images.
 
the problem with the mesh decimation is that i can't control how many mesh polygons i end up with
also the decimation doesn't merge them nicely enough
hah i just had an idea
we'll see how it goes
 
I've been thinking of some stuff....
 
go for it
 
@randomra I was thinking about your Time Travelers challenge. It would be interesting (as another challenge), if time travel was feasible, but that each round was deterministic
The outcome would have to additionally depend on input from the bots as well, obviously
 
12:30 AM
If we take the approach of trying to maintain the edges, then the merging of two cells on one side of the edge should correspond to the merging of two other cells on the opposite side of the edge.
 
hm
ok
 
MS Paint diagram time?
 
by edge do you mean image edge (like, from an edge detection filter), or a voronoi ridge?
 
@NathanMerrill Yeah, I'm planning one but how I can see it is much more complex. You would need a game with partial information (and maybe deterministic bots) and the one which teleports would keep it's memory while the other one loses its
 
Image edge.
Well, the goal is that the correspond to each other, a least for sufficiently high values of n.
 
12:32 AM
@randomra the opposite effect would be much more interesting
If you choose to teleport, you lose your memory
 
@PhiNotPi fair enough
 
in my case the randomness for the game would be the same after the teleport so the teleporter would know the future (part of it as he doesn't see the opponents "cards")
 
so basically what you're suggesting is: (1) do an edge detection & threshold to get a binary edges image; (2) find the nearest edge to two merging centroids; (3) find the nearest centroids to the merging ones, reflected across the edge; (4) merge the reflection pair and the original pair in the same step
 
In your version the gain for the teleporter would be the new, probably better randomness of the game, right?
 
I was thinking something along the lines of:
1. determine 100 coin flips
2. each turn the bots guess the flip
3. guessing the flip gives them a point
while the above example isn't more interesting than the original game
I think it can be extended to be more interesting
 
12:37 AM
I'm MS painting something right now.
 
ok cool
 
@NathanMerrill if the enemy keeps its memory why would you ever teleport in that scenario?
 
@randomra oh that wasn't part of my original idea
just another mechanic in some challenge
1. determing 100 dice rolls from 1-10
2. each turn the bots guess a number
3. if p1 guesses a number above the dice roll, he gets a point
4. if p2 guesses a number below the dice roll, he gets a point
The interesting thing about the above, is that time travel isn't always effective, as re-guessing a certain round may be ineffective
or even benefit your opponent
actually, I don't know why you would never guess 0 (or 10)
nevermind
 
Here's the way I am thinking of merging cells. The top is original. The bottom left is my idea, the bottom right is the "typical" way.
 
right
 
12:42 AM
By keeping things symmetrical across edges, we maintain them.
 
@NathanMerrill that can be fixed, and you would only learn if you&opponent win/lose and your guesses
there are some simple games which could work with some time-travel mechanics I tried to pick the simplest nontrivial one
 
When two cells merge, the area of the new cell is less than the sum of the two original areas because the surrounding cells grow some.
 
I definitely like your first idea
 
haha @PhiNotPi the idea i've been trying while you did mspaint has the weird effect of taking longer for smaller N
 
I just hope there are more
 
12:44 AM
this is a more complex game I started with:
May 16 at 12:19, by randomra
My current game is: both players have always have 5 card, they play out 1-1, and the bigger one wins the trick, most tricks wins. Does this have a name?
but coinflip is simpler
 
@PhiNotPi actually my idea doesn't work at all for small N. so never mind to that, blugh
it worked nice for N = 1000 though...
anyway
@PhiNotPi so that's the principle of it, but there's a problem that you run into
actually i'm not sure your method will have the desired result at all
 
@randomra if you did 100 rounds of that, and each round each player had 5 predefined cards from 1-20
that would be an interesting challenge
 
it has two problems
or rather, one problem, which manifests in two ways
 
@NathanMerrill with time-travel?
 
if you try to merge two points which are farther from an edge, their reflection may be on top of a second edge, which destroys detail
 
12:49 AM
Here's an idea....
Slightly different.
 
meanwhile, if you merge points near an edge, all you do is destroy detail along the edges
 
@randomra yes
 
hmm, that might work with the current mechanic too without losing memory
 
The idea is to set up some rules for when it is always safe to merge two cells.
 
hm, ok
 
12:50 AM
Like...
 
it's always safe to merge two cells when they're identically colored
and adjacent
(that's important, lol)
 
Well, not exactly. To make sure it's safe, all other cells adjacent to those two cells (which would be affected by the merging) should also be the same color.
 
so without memory loss that could be played/tested by humans too
 
@randomra I still like memory loss. It makes it true time-travel, IMO
 
But, in those cases it's almost certainly okay to merge them with no noticeable loss in detail.
 
12:53 AM
but to make it interesting, you could pass in a list of previous time-travels
 
So the first step should be to merge as many of those safe cells as we possible can.
 
@NathanMerrill but if both lose memory than the game just repeats
 
@randomra not if you tell them when they traveled in the past
 
alright, i'll see if i can come up with an algorithm for picking out those cells
 
and how many levers they have
 
12:54 AM
maybe, although that might be too little info
but that is my idea with the twist that the traveler keeps its memory
 
true. If you want it to go that way, then by all means, do it
it'd still be interesting
 
but that's pretty far away, idk if I will finish the coinflip :)
@NathanMerrill you had the NP-cracking cops&robbers idea, right? is that still aplan?
I like that
my only issue is that it might become a "who can run a program for longer time" challenge for robbers
 
@PhiNotPi picking out "safe" cells is not trivial, lol
 
@randomra Scoring is actually based on time. Its not a true cops-n-robbers, but the winning robber will be the one who solves all of the puzzles in the smallest amount of time. The winning cop will be the generator that takes the longest time to crack (on the robber who cracks him the fastest)
 
I see, so they are fixed sized.
 
1:01 AM
yes
 
Maybe you can pick out safe cells by checking if the single cell in question is the same color as all adjacent, without having to find a pair of them and testing adjacency to the pair.
 
semi-hardcoing might be a small issue, but not that serious imo
 
Any cell with the same color as its neighbors is safe, any two adjacent safe cells can be merged.
 
@randomra not really. If there is hard coding, then robbers will simply search for that hardcoding
 
I meant robber-hardcoding but seems like I misunderstood, so cops post programs generating puzzles
 
1:05 AM
correct.
the cops will also post a sample puzzle (but is free, at any time, to run their program again to generate another puzzle)
 
yeah, that seems good
 
or post several puzzles generated by their program
 
do you collect all puzzles into a test-set at some point for the robbers?
 
@PhiNotPi that's a good idea
ok then
 
@randomra actually, that's not a bad idea. The cop's challenge should be first
and have a due date
 
1:10 AM
I planned a bit similar battleship builder-guesser challenge some time ago
with number of guesses as score
(regarding your sandbox comment: a 0-sized teleport could only stop teleports cast at the exact same round.)
 
correct
 
@sirpercival In fact, I'm not sure if you even need to merge the safe cells. I think you could outright delete them.
 
The surrounding cells (being the same color) would fill in the space.
 
i guess it depends on how you're defining cells
 
1:15 AM
If the "cells" are the cells created in the watershed process, then you can't really delete them.
 
i was ignoring watershed for the moment
 
2
Q: Print strings without sharing characters

user23013You task is to write n programs in the same language. The kth program should output the (cardinal) number k in standard English in lowercase. So the first program should output one, the second should output two, etc. For a number having two parts, they should be separated by a hyphen like twenty-...

 
In case folks are interested, the controller for my next koth is almost entirely functional. You can test it out here. Just hit Run to see the testbots at work.
 
@sirpercival Here's my (not yet optimized) n-1000 wave. It's been running for about two days now, and isn't quite done. I estimate another 6-10 hours and it'll spit out a final product:
 
hahaha
very very nice
 
1:25 AM
Yea, it does really nice work, but on a time scale that makes me cringe.
I was aiming for a good quality low-n, damn the consequences.
^ n-100
 
excellent
 
@Geobits that's very nice!
now you only need a supercomputer :)
 
or a few... :(
 
@Geobits what's your algorithm that it takes so long?
 
It jiggles points around in a random fashion, scoring the entire image for each jiggle. Then it keep doing this until it cannot find a single better jiggle.
 
1:31 AM
yikes
lol
 
To "speed it up", I scale the image down by powers of two and work on each size until perfect, then scale up the points and rework it on the larger canvas.
 
I have an idea of using only triangles to give the algorithm greater freedom and than swap each triangle with 3 center points to generate the triangle-spahe by 3 voronoi-cells
 
The jury is out on whether that is very effective in terms of speeding it up, but it looks nice :D
@randomra Oh that's interesting. Should be much faster to compute which pixels go in which cell that way.
 
all hail @Geobits, arch-nemesis of heuristics
 
@Geobits do you jiggle all points at every step or just 1?
 
1:33 AM
		int distance = width/4;
		for(int i=0;distance>2;i++){
			int popCount=0;
			int walkCount=0;
			for(int j=0;j<n;j++){
				if(poke(j,distance)){
					walkCount++;
				} else {
					if(pop && poke(j,0))
						popCount++;
				}
			}
			draw(cPoints,true);
			if(debugOutput)
				System.out.println(i + ": walk=" + (walkCount/(float)n) + ", pop=" + (popCount/(float)n) +", dist=" + distance);
			if(saveFrames){
				String f = ""+frames++;
				while(f.length()<5)
					f = "0" + f;
				ImageIO.write(img, "png", new File(filename + "_" + f + ".png"));
 
0
Q: Caesar equivalence

DJ McMayhemTwo strings are "Caesar equivalent" if the distance (counting up) between the corresponding characters are the same. Yes, I made this term up. Here's an example: "Abc" and "Cde" are equivalent because distance from a-c == 2 distance from b-d == 2 distance from c-e == 2 The capitalization does...

 
All points. poke() jiggles point j by at most distance pixels.
If distance is zero, then it can pop it out and put it anywhere.
 
@Geobits it might also be more precise as you can move 1 vertex instead of rearranging the whole region (colorwise)
 
Yea, it sounds like a good idea. I look forward to your pretty pictures :)
 
ofcouse you only have N/3 triangles, and at the very end you can choose colors for all N voronoi-cell
@Geobits I'm definitely not doing it, just thinking about algorithms :)
 
1:37 AM
Well I'm not doing it. I'm gonna sit here and run these images until my computer throws a fit.
 
@Geobits your colors are messed up
are you using L*a*b*?
the shape is ver nice though
 
Using HSB to score, but just source-point colors for output. No averaging across the region. I was mostly concerned with shape.
 
@Geobits do you recalculate the image every time you move only 1 point?
 
I tried averaging, and it works really well for high-n, but for low-n it came out odd on a lot of images. Yoshi, for example, looks much better with flat colors.
@randomra Yes, which explains why my wave has been running for days :D
 
@MartinBüttner How did you manage to comment on a deleted post?
 
1:45 AM
for high n's that adds a lot of time
 
I tried scoring only for batches, but it became hard to tell which points in the batch were making it better or worse, and I got better results for single pixels.
 
@Ypnypn ehm... I'm a mod.
(also I wasn't aware it was deleted...)
 
@randomra Oh, I know. Trust me, I know :(
 
@Calvin'sHobbies I can't seem to figure out what the goal is. Do you have the spec written up somewhere?
 
@NathanMerrill Not yet. Basically the red and blue dots try to get to the yellow crosshair.
 
1:47 AM
@Geobits you can evaluate for each changed points but not the whole picture just the point's region
 
@Calvin'sHobbies exciting
 
@randomra I've thought about making that change several times, but haven't. I think I might have gotten a bit masochistic over the last couple days.
 
2:07 AM
0
A: Sandbox for Proposed Challenges

Calvin's HobbiesBlock Building Bot Flocks This answer is just to signify that this contest is currently being sandboxed. The full spec is in this meta post.

 
0
Q: Block Building Bot Flocks [sandbox]

Calvin's HobbiesThis is the sandbox post for my upcoming king-of-the-hill contest. Similar to RvB, it runs interactively in the browser using a Stack Snippet. Answers to this question will be loaded automatically by the snippet. Anyone may answer this question to test the game out but I may disqualify or even ed...

 
2:24 AM
@PhiNotPi first pass on the white yoshi killed almost 90% of our pixels, lol
 
That's good! (right?)
 
yes
it is
it won't do as well on other images, though
lemme see what it does to the bear
 
You can always increase the tolerance as to what counts as "the same color"
 
haha on the bear it gets rid of ~1.5%
yes
ok i need to do some actual work
@PhiNotPi we'll attack this more tomorrow, mabe
night all
 
3:07 AM
@MartinBüttner 26,f= is some weird way to get modulo :)
 
 
1 hour later…
4:35 AM
0
A: Sandbox for Proposed Challenges

da capo ariaClarence the Slow Typist code-golf Clarence is a data entry clerk who works at an internet service provider. His job is to manually enter information about the ISP's customers. He enters all of the data using a number pad which looks like this: 123 456 789 #0 The distance between the centre o...

 
5:25 AM
what's the shortest way in Python to create an n-by-n list of spaces: like [[' ', ' ', ' '], [' ', ' ', ' '], [' ', ' ', ' ']] for n=3?
the rows need to be distinct by reference, so that changing one entry doesn't affect others
it needs to be saved to a variable
this isn't a puzzle, I was just wondering looking at KSab's answer here
 
 
3 hours later…
8:15 AM
@xnor can you split into parts of 3 ?
 
 
3 hours later…
10:52 AM
1
Q: Count the number of vowels in each word of a string

Cool GuyThis is a rather easy challenge. Challenge Input will contain a string of maximum length 100. Output the number of vowels in each word of the string, seperated by spaces in the next line. Rules The string will not be more than 100 characters in length. The string will only contain alphabets ...

 
7
A: Tips for golfing in CJam

Martin BüttnerCorrect modulo for negative numbers It's often annoying that the result of the modulo operation gives the same sign as the first operand. E.g. -5 3% gives -2 instead of 1. More often than not you want the latter. The naive fix is to apply modulo, add the divisor once and apply modulo again: 3%3...

 
@MartinBüttner random walk ascii art snow flake
how does that sound ?
 
tedious :D
 
:D
like the non-ascii art one was not :P
 
no that one was fine
converting the data to ascii art sounds much trickier
especially the 60 degree rotations
 
11:01 AM
you can do a walk in ASCII.
then its 1 to 1 mapping to rotations
 
Is it just me, or does "Murder Positive Integers" leave out the vast majority of positive integers?
 
11:52 AM
That's why it's called selectively murder positive integers. But I'll agree it's not the most accurate title I could have given it.
 
0
Q: Scale up a Diamond Tiling

Martin BüttnerAny regular hexagon can be tiled with diamonds, for instance like so: ______ /_/_/\_\ /_/\_\/\_\ /\_\/_/\/_/\ \/_/\_\/_/\/ \_\/_/\_\/ \_\_\/_/ We'll consider the above a tiling of size 1 (since the diamonds' sides are made of one / or \ each). The same tiling of size 2 would look like...

 
12:08 PM
Im TIRED of people whom their challeneges exceed their answers.(seek to help as much as u seek to be helped)
 
12:30 PM
uummm... here on PPCG wouldn't that mean the opposite? that you should post at least as many challenges as you answer? writing good challenges is the hard (and selfless) part around here, not answering.
 
12:40 PM
can I get some votes on this?
0
A: Should there be defaults for tiebreakers in tag wikis?

Martin BüttnerIt just occurred to me that my suggested default tie breaker doesn't necessarily make sense for popularity contests. Therefore, I'd like to suggest an amendment to the accepted answer, and if this gets a reasonable amount of support I'll edit the accepted answer. In the case of popularity con...

 
I agree with it, very reasonable idea.
 
I just have one issue.
 
whew, for any fans of escape-the-room type games, go play elements by neutral RIGHT NOW
hi all
 
I think your ordering is broken
 
1:20 PM
@Optimizer what do you mean?
 
@MartinBüttner hi all should be before his first line
 
oh
I thought that was continuing from "I just have one issue"
 
1:36 PM
So on an overall basis, is JS better in golfing as compared to Python ?
 
I doubt it, but I don't have any real data
 
I don't think JS requires spaces, so that's a plus
 
you can do a lot to get around the spacing in python
 
@Optimizer Probably not: golf.shinh.org/lranking.rb
 
but how many times do you have to write a program that is long enough ?
 
1:43 PM
python has some nice builtins, though. I think it depends on the task, I've seen both come out ahead.
 
@randomra that seems to be outdated
even though it says updated today
doesn't have CJam, Pyth. What are the challenges it is computing on ?
 
@Optimizer what has nothing to do with PPCG
it's Anarchy Golf
 
Still, if it has GS, it should also have CJam, Pyth
also, I doubt that J is worse than GS
 
they are actually running all programs on their server. I'm not sure they're even aware of Pyth and CJam, so why would they support them?
 
Hmm. Challenge: Print the number of times JavaScript beat Python, a space, and then the number of times Python beat JavaScript. Regex for what constitutes a JavaScript answer and a Python answer to be provided by the challenge, as well as a Regex for parsing the score.
 
1:46 PM
@MartinBüttner and what is the credibility of it then ? :P How are we sure that the best attempt in JS was made..
 
how are you ever sure?
 
PPCG
 
if its here, and by trusted users :P
 
that's no proof that it's the shortest possible solution.
 
1:47 PM
That's Optimizer logic for you. He has a point buried somewhere in there, trust me
 
attempt*
 
@Optimizer J might be less general-purpose in terms of golfing and creating full stdI/O programs (which anarchygolf expects) adds like 8-10 chars
 
Can't we just assume that for every weak JavaScript attempt, there will likely be a matching weak Python attempt?
 
@Rainbolt but the scoring is not like that.
 
And throw out the results where there are like five Python golfs and only one JavaScript golf. Only consider results where the playing field was even.
There will be some false positives and some false negatives to balance it out, but over 500 code-golf questions, we should see a trend
Unless you think that people who write JavaScript are more or less likely to write weak answers than Python users? But why would they be?
 
1:51 PM
On that site, I have no idea about any of those numbers.
 
python is easier than javascript
to write in
i dunno about golfing
 
You can't see any code there, so you just sorta have to assume/guess that the golfers know what they're doing I guess.
 
Or (as I have said twice now) you can toss that assumption into the garbage (because it is weak) and just assume that for every weak JavaScript answer you also have a weak Python answer. I feel like I'm preaching to a wall now.
 
@Rainbolt that's assuming that one is equally likely to write a weak JS vs weak python answer
which may or may not be true
 
The problem with preachers is that they always assume they're righter than anyone else.
 
1:55 PM
i feel like i see crappy js more often than crappy python (though partly that may be that i don't know JS very well)
 
I smite you.
 
grc
chimes in
ES6 > python > JS for golfing imo
 
es6?
 
fancypants JS :D
 
@grc I was always considering JS as ES6 only.
while talking about golfing
 
grc
1:59 PM
then yes, I'd say it's better
 
I get almost aggravated when someone posts an ES6 answer to my challenges, because then I have to open FF. Google needs to get that shit working.
 
Is there no online interpreter?
 
there are many
 
grc
babeljs.io/repl seems neat
 
Biology has graduated: biology.stackexchange.com
3
 
2:02 PM
DNA badges
 
Ok, take my previous comment and apply them to snippets, which most recent JS/ES6 answers have been. I could copy paste each segment into an online interpreter, or run the snippet in FF.
@MartinBüttner We're next in line, right?
 
@Geobits yes, 27th May
 
of course
after Chemistry, I'm a bit disappointed with their vote buttons
 
@Optimizer I think 6 days is rushing it a bit. It's almost unbelievable.
The vote buttons were the first thing I checked :(
 
@Geobits as if it makes a difference
 
2:04 PM
@Geobits That's day golf for you
 
I hope they don't golf the layout. "Hey guys, I put the whole image into four pixels in the top left corner!"
 
Well, maybe an Easter Egg like that would be neat (in addition to the normal stuff)
 
grc
@Geobits you could get chrome to run an ES6 shim on PPCG pages
 
Oh that's an idea. Hadn't seen that before, thanks :)
 
I updated my ASCII Art of the Day series sandbox post with some new ones . All bit vague for now though.
I feel that keeping all in one sandbox post garners less attention ..
 
grc
2:07 PM
@Geobits not sure if it can handle new syntax though
 
Only one way to find out.
 
Ask on Stack Overflow?
 
Obviously.
 
Surely you aren't about to test it yourself.
 
Good god no. What would the point of that be? Then I would be deprived of the pleasure of getting others to figure it out for me.
 
2:09 PM
After all the hard work you have put into this corner of the Internet, it owes you an easy answer.
Just be sure to use good English and it will appear as though you wrote a good question.
 
I'm sure I can bullshit enough words in proper order to disguise my lack of effort.
(which should be appreciated as an effort unto itself, but...)
 
I just remembered one of the first things Peter Taylor said to me when I got here. "This is a probably a duplicate of that, but the zombie story you wrote around it is probably enough to not get it closed." (Not an exact quote.)
9
 
hahahahaha
 
Ha, I forgot about the zombies :)
Did we ever kill them all?
 
Why the heck did I make Wolf instead of Zombie?
I guess because the exercise we did in class used Wolves
 
2:14 PM
Reminds me of some story I read where wolves were "natural" predators of zombies and people used them for protection, but I can't think of the name at all now.
 
@Geobits to scifi.SE!
 
wolves vs endermen
 
@MartinBüttner That's sadly representative of the amount of detail many of their story-id questions get.
 
indeed
 
All I know is that it was at least 10-15 years ago that I read it.
 
2:16 PM
Good morning, folks! I'll probably be posting my challenge within an hour or so, so last call for loopholes.
 
@BrainSteel it's a combination of the code formatter and telescoping parentheses
 
That enderman reference reminds me. I made a portal in Minecraft last night, and it took me here:
Anyone recognize the reference?
 
Image not found?
 
Some guy expanded on it a little to make the reference more clear: frogtek.net/wp-content/uploads/2015/04/2015-04-27_03.02.38.png
 
@Optimizer are you doing the diamond tiling thing?
 
2:23 PM
Martin feel free to linkify my chat image. I let it go past the time limit on accident
I know a lot of people in here automatically right click inspect delete alt f4
 
@sirpercival Yeah, I kinda realized that... You think it's different enough?
 
is it offensively large or something?
as long as it doesn't blink, I don't see a problem
 
No, just don't like cluttering up the chat with images.
 
@BrainSteel i dunno. i haven't been around PPCG long enough to judge similarity acceptableness
 
@Calvin'sHobbies Is there a way to get the seed of the RNG after the game has run its course? I think I found a bit of a bug (either in the JSFiddle or the Baseline bot, I'm not so sure)
 
2:26 PM
On that note, when are we getting our copy of a pervy loli bot from the Anime Stack Exchange?
It adds just the right amount of uncomfortableness. I think it would cut down on our arguing a LOT
"I know we disagree but that's not the point! Oh... pervy picture..."
 
Better question: When are you going to post a male version in the anime chat room? :P
 
What is going on here?
 
Knowing you, I bet you would help me build it just to see the reaction.
 
are there any ladies who frequent PPCG/19th byte?
 
@BrainSteel The Anime chat has a bot that posts underage, sometimes scantily clad pictures of cartoon girls.
 
2:29 PM
I kinda like the anonymity
 
@Rainbolt Couldn't be hard. I think their bot is open, just need an appropriate list of pictures to point it to...
 
The bot doesn't have to be pervy, eh?
 
It has to be exactly as pervy as the female one. It's only fair.
Perhaps once we get past the shock value that I am after, we can switch it to hollister and abercrombie guys.
 
I'd much rather see random pictures of space or something.
 
Wow that first one was fail
 
2:32 PM
If there's one thing I've learned from anime in general, it's that bots designed by humans are likely to be pervy.
 
an APOD bot would be fun
 
Has anyone seen Ex Machina?
That white picture is reminding me of Daredevil
Oooooooh wow. I just had a major revelation.
For the entire Daredevil series, the bad guy stares at a white painting and he always sees hope for the city. The good guy sees the world on fire and always sees the bad in people.
Thank you high school English class.
 
That is... Clever...
 
2:48 PM
I decided after the season finale that I don't really like the show. The characters don't evolve at all for 13 episodes.
It's about time to start binge watching Game of Thrones. Three and a half weeks and the season will end.
 
@BrainSteel The random seed can't be set or retrieved in JS (see here). What happened that seemed wrong?
 
One of the Baseline bots just sorta stopped moving.
 
-1
Q: Code Optimization - Find the number

SwanandA Friend gave this puzzle: Find 3 numbers (say, A B C) such that ` ABC+ABC+ABC = CCC `ABC means ((A*100) + (B*10) + C). and also,these 3 digits must be distinct, I wrote a quick code to find out the answer: int main(void) { // your code goes here int hun,dec,unit; int final_nu...

 
Oh, that's fine. He was probably stuck in midair without a wall (dark gray block). Every time a bot moves they have to neighbor a wall.
Writing a rudimentary spec shortly so people can actually test stuff
 
Ah, that could have been it. I can't remember it so specifically. The Fiddle seems really cool! I await the spec :D
 
2:56 PM
I believe I've just completely finished the controller. But yeah, don't hesitate to point out potential bugs :)
 
It is rather sucky that you don't have a way of getting the seed. If you do run into a bug, it seems it would be a really useful debug tool.
 
00:00 - 15:0015:00 - 00:00

« first day (1573 days earlier)      last day (3274 days later) »