Grr, I'm being rate-limited by SE (in posting the two cops-and-robbers threads). Not even a diamond is enough to convince the site that I'm not a spammer :P
@Sp3000 Yeah, I was thinking about that. It was originally 80 chars, but I thought that would just be too unfair considering that the public static void et.al. boilerplate is already 45 chars...
I had a neat idea! I hope it's not a duplicate.. http://codegolf.stackexchange.com/questions/42916/print-the-twelve-days-of-christmas-with-twelve-different-coding-languages
So, I'm on a router that happens to block all traffic...except between 8 and 10. Unfortunately, it blocks my laptop at all times during the day. It's a Linksys Smart Wifi router.
I decided to poke around and found that I could load images, for one...but anything with a text content-type gets redirected to the Blocked page
but it doesn't do that to ones with Host: (anything).linksyssmartwifi.com or Host: (anything).linksys.com, those work just fine
...so I used my Windows hosts file to point tun.linksyssmartwifi.com to a VPS I got for this, and it goes right through the TCP tunnel to an instance of tinyproxy running on the other end. I'm using it right now. It's a bit slow but it's sped up a lot, ever since I moved it from the VPS to a dedicated server of mine.
Phoneword generator
This is codegolf and restricted-source
Goal
The goal is to write the shortest phoneword generator.
Phoneword
A phoneword, is a word, that is typed with letters in your phone which map to the numbers they display.
As example I use codegolf itself: 26334653 translates to C...
@Doorknob冰 I don't understand your mutation challenge. Can't I just replace each cop by a 120 byte Fibonacci program in the same language (and also get the highest possible score by doing so)?
Hi, I'm the author of the challenge Construct a graph. I'm considering extending it into a series of challenges, with the theme of constructing some object from a simple "blueprint". Is there an established format for challenge series on this site?
@MartinBüttner I tried snowflakes where the particles can drift in all 4 directions and it does give much sparser results - it seems to make a big difference. I think they are now too sparse so I'm experimenting with changing the probability of a particle sticking to the tree so some of them can drift deeper and fill the gaps.
It's specified in terms of a density rather than a number of particles, but I think these are a few hundred
They are sparse because as soon as a particles touches the tree it sticks, so it's unlikely many particles will get past once a branch has started to grow. I tried with only a 1 in 10 chance of sticking each time a particle moves onto the tree to make them denser:
@Optimizer Ha ha, very funny, but after looking at my code for the rotation-safe challenge (:n``+.{~.6=9*\.9=6{*3@+\}:f~.0=0f.8=8f;}/](\{+}*.,3@.,3@=3~@=!&.*n\+) that doesn't even work, I'm starting to think so... Really, the problem is that lots of functions just aren't implemented (like array index — how could I forget that one?!).
I then read about real snowflakes and discovered that the pattern is due to the conditions in the cloud changing as they grow - so they have periods of fast and slow growth leading to different shapes.
So I made the probability of a particle sticking vary gradually during the growth. They all have the same probability at any one time, but that probability drifts. I was hoping for layers of sparse and dense growth building up.
@PeterTaylor say, how would represent an empty 2x2 matrix, such that you don't need to handle it specially with regard to adding or filtering rows/columns? (I've got this problem in CJam, so I thought you might have a GS-specific idea, but now I'm sort of wondering how one would do this elegantly in any language)
@MartinBüttner All the particles are released from a single point at the edge of the image on the x axis, and a new particle is added at that point every time it is unoccupied. If the tree reaches that point no more particles are released, but any existing floating particles continue to drift around until they join the tree.
@PeterTaylor yeah, that's what I've currently got, but unfortunately I need to treat that case specially.
because all the array manipulation I do (including expanding it), is done by mapping onto rows.
@Optimizer do you have an unposted version of the rotation-safe numbers that's shorter than 46? can't believe that you didn't beat my code that I didn't try to golf yet :D
@Sp3000 I'll make sure to include an example of that. Would you mind adapting your reference solution to the new output format, so I can use it to generate test cases? :D
@MartinBüttner How quickly does an optimal solution become unrealistic to achieve? You could either insist on an optimal solution or score based on a test set that includes large enough cases to mean optimal solutions are unlikely, and the score is the sum of the toggles
The game BattleBlock Theater occasionally contains a puzzle which is a generalised version of Lights Out. You've got three adjacent blocks, each of which indicates a level between 1 and 4 inclusive with bars, e.g.:
|
||||
||
If you touch a block, then that block as well as any adjacent block w...
@MartinBüttner The purpose is that all classes have a shortcut to access; and that shortcut is the shortest, distinct piece of text to indicate a class.
here's another good one. i'm doing stuff with bunch of numbers (again, currently in strings). some are pandigital, some are not. i want to keep track of the greatest pandigital i've found so far
@Sp3000 i don't do much complex stuff, mostly basic problems. if you can't make use of any of that, python3 tends to win out just because len('input();print()') < len('raw_input();print ')
The game BattleBlock Theater occasionally contains a puzzle which is a generalised version of Lights Out. You've got three adjacent blocks, each of which indicates a level between 1 and 4 inclusive with bars, e.g.:
|
||||
||
If you touch a block, then that block as well as any adjacent block w...
