« first day (1587 days earlier)      last day (1647 days later) » 

3:02 PM
postage initiated
Haha, people like my Badger :)
For Hunt the Wumpus
@TheNumberOne Actually, it is pretty dangerous x) But it's not winning a lot ^^'
current scores :
ScaredyCat : 11075
Paranoid : 13005
Badger : 15540
Idomeneus : 16565
Pacer : 17300
MonsterHunter : 18260
rand : 22235
Trig : 31220

Total rounds: 7000
Humans killed by arrows: 4
Wumpus victories: 4580
Wumpus winning percent : 0,65
The problem is, when Badger survive alone, it will often die of starvation, and it is sad ^^
He prefers starvation over company :P
3:09 PM
Hahaha, why doesn't he eat some corpses so? He will stay alone AND will not starve to death !
I think a simple neural network would be good at this problem.
Q: Output every halting program (write a parallel interpreter)

PhiNotPiThe goal of this challenge is to (eventually) output every possible halting program in a language of your choice. At first this may sound impossible, but you can accomplish this with a very careful choice of execution order. Below is an ASCII diagram to illustrate this. Let the columns represe...

@PhiNotPi Can I still use threads if I make sure only one is running at a time?
why use threads? just start every program over from the beginning each time!
@PhiNotPi are exceptions in programs that catch them OK? Should programs only be excluded for an exception if it is not caught?
3:16 PM
@TheNumberOne I'm not sure how that would work.
What does everyone else think about the thread pausing method?
As long as the threads are still running code that he wrote, and not some other interpreter, sure?
@trichoplax If the tested program catches its own exception, then that is fine to be included.
3:20 PM
Actually, java Threads can't be paused like that :(
@feersum I'll try that if @PhiNotPi is okay with it :)
:( os.kill isn't available on Windows Python
@PhiNotPi What I had before you disqualified it :(
I'm kinda feeling like I just want to stay away from thread spawning. Some of it is rather creative, but I feel more interested in "true" interpreting.
ok :)
It would be kinda funny if one of the threads you tested were actually a fork bomb.
3:32 PM
^^ That's what I was thinking when I was running this, but good thing it never made it past 3-char programs
someone needs to do Bash so they can get to forkbombs
@PhiNotPi since we only output valid halting programs, we can recognize and refuse to attempt to interpret invalid programs, right?
fork() is not required to make any language turing complete, so you could just not run any program that creates it, and still be within the rules of the challenge
@Sparr Yes.
3:51 PM
@Sp3000 If it was I would have posted a KotH by now :(
@trichoplax ^ :D
@feersum It's what makes me like the idea of a language-specific KotH with players being functions - players cannot do anything between turns
same thing really with a Java one with classes, isn't it?
unless you don't want them to store data
^ Javascript is good for that.
is allergic to Javascript
If I get near it, I have trouble breathing
when I was in school the other children had to always use other languages
4:05 PM
@feersum I'd prefer them to not store data, so I can store data for them and ensure it stays within a size limit
@TheNumberOne That conveniently allows a KotH to be run in a Stack Snippet too :)
As much as I like language agnostic KotHs, I think they are a lot more work (to write and run). I'm tempted to run a series of language specific KotHs, each in a different language
isn't there some language that is directly compatible with Java?
@feersum you must have been to a weird school if that was an issue. in my school, the children were mostly using German (and occasionally English, French, Russian and Latin in class).
@PhiNotPi groovy, closure, jython, jruby, ...
4:12 PM
@PhiNotPi Scala
(Not just the JVM)
I think Scala was what I was thinking about.
I would be interested in a series of language specific KOTHs in different languages.
I've never used Scala/Clojure/Groovy & friends. Would it be possible for a Java controller to support classes written in those languages?
I'm just trying to think of ways to make class-based KOTHs be less language-specific.
@PhiNotPi They already do :)
Though, Clojure isn't even a class-based language, is it?
I would just have to install compilers for those languages, and it will work?
I think so. I've never used any other JVM languages though...
@PhiNotPi I suppose it wouldn't need to be a connected series, just a matter of noting which languages haven't had a KotH yet, and then anyone can pick one of those languages for their next one
Maybe I should make a meta post for people to suggest languages that they'd like to see a language specific KotH for, then we can see which language is most in demand for the next one
4:21 PM
Last year sometime (before I knew any Java) I wanted to make a Perl KOTH.
What I like about Stack Snippets KotHs is that visitors can watch them in action - they don't have to have/download the specific language
Are there any other possible languages that can be live-animated like that?
@trichoplax I'll probably try to hook up my Pi later this year such that it can act as a small KotH server that can just rerun a KotH all the time and always provide up-to-date results.
I have an idling Pi that I can somehow loan to the cause.
yeah trichoplax suggested last year that we could somehow set up a Pi farm to run KotHs in a distributed fashion
I think trichoplax has got one as well
4:31 PM
We should do that.
Am the only person around here who capitalizes all of KOTH?
okay I just generated some example data for Boggle Compression. string lengths 3 to 50, using hex digits. without any overlap, the minimum size is 1272. the minimum possible size you could compress it to (I think) is 92.
the number seems to vary between 90 and 100, with occasional spikes into the 80s or 100s
4:46 PM
how do you know the minimum?
or is it some lower bound?
@PhiNotPi I guess that question is equivalent to "is there any other language that everyone has an interpreter for?"
@MartinBüttner Yes I have the model B (the old B)
Started the time travel koth in Java, after coding in python. It's like writing a novel...
@randomra of course!
it is opensource=)
But I am not sure if the 'quality' of my code really is any good...
@PhiNotPi My earlier (months ago) suggestion was to have a number of pis in one location, and a KotH player per pi. I'm curious about whether you could make a KotH parallelisable though.
4:54 PM
@flawr I'm sure I can adapt :)
@flawr actually I already modified it, I think it's well written among the koths
@flawr You haven't put my scores on the leaderboard yet :(
the 1v1 result grid is useful
Q: How could we run KOTHs in a distributed fashion?

PhiNotPiThe idea was mentioned in chat earlier today that KOTHs could be run in a distributed fashion. I believe that there are several users (including me) who have Raspberry Pis that could contribute to a KOTH Pi Farm. This is an idea that I would like to see move forward. It would allow for up-to-d...

a star is allergic to Javascript ?
@Optimizer it's a regex, * is allergic to Javascript
5:01 PM
@randomra no its not
ok, wildchar
* has six sides
what do you call * then?
@randomra it's the lower bound
provided it's possible to arrange the strings for maximum overlap
5:02 PM
@MartinBüttner When you say "server", do you mean it would be online accepting new answers automatically too? Or just that you could start it running and leave it for a few days before posting the resulting leaderboard?
.. is the first time I have seen it written ..
@MartinBüttner if a fairly simple approach can get a couple hundreds below the 1270, it could probably be a reasonable test case
@flawr with the deterministic cats&catchers it could be a subchallenge to get the best score against them as a human
like your score is your total cat step - total catcher step
5:13 PM
@trichoplax Is the capitalization really that important?
No not at all - feel free to revert :)
There's a "rollback" option so you can choose which revision to revert to
If it isn't funny then I apologise - it wasn't really meant to annoy
There's an interesting psychological trick that, if I stop talking in a conversation, the other person ends up continuing to talk to fill the silence.
So it went from ";)" to "I apologize" with no prompting.
That's fiendish
I don't feel the need to fill general silences, but silences that may indicate offence get me every time
The edit's fine, I'll keep it.
lol if I'd known you were going to keep it I might have chosen the edit reason more carefully...
5:25 PM
I've been looking at this post for possible was to expand SS-KOTHs to other languages.
I was wondering about that too.
Calvin's Hobbies pointed out that the translation to JavaScript doesn't need to be done live - if you had code that would "compile" another language to JS then you could just submit the JS as your answer
So far SSKotHs have had functions as answers, so ideally there would be a program to take a function in one language as input and output a JS function
@PhiNotPi Wow that's a long list. Even has BASIC...
@trichoplax of course, a fully automated setup would be ideal
@MartinBüttner I guess it would be feasible for a language-specific KotH
5:40 PM
even if not it's just a matter of adding support for new languages as they come in
5:52 PM
@randomra just tried a larger alphabet, size 36. the minimum size only increases to about 130
also including lower case it goes up to 160
@MartinBüttner imo the question is how much can a baseline algorithm compress from the maximum size
if you can only improve 1270 into 1220 there might not be too much room for different effective algorithms
honestly, I'm not sure what the baseline algorithm would be
addin string 1 by 1, laying them around the blob's contour, using its letters on the edge
maybe it's not really a baseline, but an actual submission
Why is 2 allergic to Javascript ?
@TheNumberOne Sorry, it was way to hot today for me to start the computer, trying to keep my room as cool as possible=)
6:02 PM
@flawr btw I don't think you need to run 100.000 rounds
1.000 is probably enough, 10.000 definitely
Yes, I saw that there are many deterministic ones who just got a bunch of zeros at the end of their score=)
that too, but 100.000 is a lot for non-deterministics too
I take that as a compliment, so far almost everyone I worked with agreed that my code is a mess=)
Well lets see, I think 100 will already take long enough.
in code golf standards it's great :)
Thats a valid argument=P
6:16 PM
For my deterministic Nano Core War KOTH, I literally ran every possible scenario.
6:41 PM
okay, boggle compression is potentially ready now
although I guess a stack snippet for verification would be nice
whoa, the vote buttons are scrolling along with the posts now :O
it's not quite a dupe though
oh, they're doing A/B testing, so you might not have the sliding vote buttons
how do they tell which one is 'better'? do they prefer a layout that generates more votes?
No slidy buttons here. :/
Nor here, and I'm glad. I don't like to be stalked by page elements.
6:52 PM
Stalking ads are the worst.
sees "Is this word on the boggle board?" on the front page, promptly adds to the "2D questions to do" pile
@feersum essentially, yes.
they just activate it for 50% of all users and will monitor if it affects voting patterns
@Sp3000 it was the only boggle challenge we had that wasn't tagged boggle :D
I'll downvote all posts longer than one screen. That'll show them ;)
^ Martin's afraid
From memory I think all the 2D languages have yet to implement something that doesn't make them have to repeat the same thing before every letter
7:00 PM
@MartinBüttner u know? That's 1 line of CSS.
and not via putting everything on 1 line. Its simply 1 single rule on 1 selector
I believe their actual solution is this JS: $('.vote').sticky($('.vote').parent().parent().parent().parent().parent());
.vote { position: sticky; top: 0}
(okay, 2 rules)
wouldn't it need some more magic in CSS to make it stick to the right element?
no. Although, this would work only in browsers supporting position: sticky.
7:26 PM
@Sparr I'm looking into your idea for the stack-snippet & google sheets combo
please add Dropbox support as well. kthxbye
phew, just submitted a draft edit to the fission OEIS sequence
while doing so I've noticed that the sequence is actually really fresh
it was only added on 31 March, this year
@Optimizer "no"? how would it know then?
@MartinBüttner parent
well, but judging by the JS, it's not the parent it should stick to
which, in this case, is a left invisible sidebar, equal in length to the post
so it works.
just try it in firefox.
7:34 PM
hm k
@TheNumberOne I just started a new simulation with all the newest submissions, including yours (I had to limit it to 100 games=)
such a productive day in terms of upvotes...
this is a fun sequence with a really weird growth pattern: oeis.org/A121263 (see the table for the growth)
This makes the impression as if it would exceed factorial?
But seriously, who needs a sequence like this?????????????????
I've seen weirder stuff on OEIS :D
7:46 PM
Coming up with a sequence like that is one thing, but oeis still claims to have something to do with science....
@PhiNotPi It might be too late but codegolf.stackexchange.com/questions/51273 might be better as pop-con. It doesn't seem very golf friendly
As far as I am aware they only accept sequences that have a relevance at least a published paper?
the fission numbers don't have a published paper attached
Well, ok, some crazy number terrorists or relatives at work...
but that base dungeon thing seems to have been published by Sloane, so...
7:49 PM
I do not know whether this could be considered as a criterion of significance=)
1 4 2 2 2 8 2 1 9 2 4 5 3 2 2 2
@flawr OEIS allows pop culture seqs definitely. e.g. oeis.org/A247698 (video source: youtube.com/watch?v=D8ntDpBm6Ok )
But it still was submitted by a scientist I think
Nope: "Sebastian Zimmer: Undergraduate mathematics student in King's College, Cambridge. " did not expect that...
I though they were way more strict.
7:52 PM
@MartinBüttner oeis.org/A246651
nice :D
[Warning: there is an ad]
@Optimizer I thought that was funny too
It would have been even more funny, if oeis pages had ad themselves
7:54 PM
@MartinBüttner in oeis.org/A247698 on the video link
What the hell????? My esteem for oeis is rapidly deteriorating at the moment...........
oh right
@flawr no the OEIS page contains the warning
What kind of warning?
Or what are you refering to?
are you willing to click a link which contains a warning to an ad ?
@flawr The warning was next to a link to a video on youtube.
7:56 PM
if not, there is no point in explaining
Why 1 ??? Why ??
And here, I almost had Google getting more data about flawr..
@flawr Hey, pop culture mixing with math is not necessarily something evil. It's how to get the public and kids interested.
@Optimizer does dropbox have some way to submit data into an existing document, like google forms provides for a google spreadsheet?
that was a joke.
Seriously? Do people around here still take me seriously?
8:00 PM
@Optimizer When did they do that?
look up
can we expect a challenge about Calvin's ceiling now?
nevermind -.-
@Sp3000 I was thinking of making a language where you describe the pattern in two parts. The first part describes the shape that the characters are taken from, the second part represents a regex.
@Calvin'sHobbies I do not say it is a bad thing and I really like how oeis managed to become a central database for a branch in maths. But up to know I had the impression that they are very strict on what they actually accept. I could not imagine that they would include a sequence from starwars!!!
8:02 PM
@TheNumberOne are you talking about the 2d pattern matching challenge?
I love that that challenge is still alive
I'd really like to see something truly useful come out of it. Something that gets used in the real world.
cough something we can use to solve other ppcg problems cough
hi all
For example, with my idea, GOLF on a boggle board would be: ^**"GOLF
8:03 PM
I have a suggestion for a challenge... call a string bad if some prefix equals some suffix
a string is good otherwise
judging a string based on its fixes ?
count how many strings of length n contain a good substring
too easy/hard?
wow, that last line made it non-trivial enough
(beware my edits :) )
@TheNumberOne I want to match a 2x2 block of digits, where the bottom two are double the top two. so, "11\n22" and "43\n86" would match, but "12\n34" would not. How might you do that in your language?
that is, each of the bottom two digits is double the digit above it
8:06 PM
Err, I don't know.
@felipa no constraint on the strings? so you're counting among all possible strings?
@sparr I should fix the alphabet size
say that is fixed
@TheNumberOne I think a core part of the idea is that your language needs to support patterns that can't be easily described as a single boggle-like path of characters.
so, with the alphabet "abc" and length 3 the strings are aaa aab aac aba ... cca ccb ccc. among these there are substrings that are the whole string, plus aa ab ac .. cb cc?
@felipa code-golf ?
of the two-character substrings, the doubles are bad (aa bb cc) and the others are good, so any 3 character string that isn't aaa bbb or ccc has a good substring
8:09 PM
Is there any way for a Stack Snippet to the edit to post itself?
is that right?
@Optimizer hmm.. I prefer fastest-code :)
@PhiNotPi if you gave it your credentials, it could post edits. Don't do that :p
@PhiNotPi hence my idea to put the match results in a google spreadsheet. then another snippet could fetch from that to make the leaderboard.
@Sparr ^(l`>)^>"(00|12|24|36|48)*
@VisualMelon If anything I'd prefer to do my bathroom walls. There's totally an Einstein made of grout:
user image
I'm not even kidding. I have no idea how it got there
8:11 PM
you know, in the comics, you were trying to sleep, when you saw your ceiling .. now this comic would be so different..
@Sparr does it seem interesting enough?
@TheNumberOne Just updated the scores, it's seems you're rather number two and three.
@flawr :( I know of one modification that would significantly make it better.
@Sparr it has to be a substring.. so abc doesn't have one
a substring is contiguously indexed
8:13 PM
I think a substring is well defined here.
@TheNumberOne Feel free=)
@Optimizer the example Sparr gave didn't seem to be right
he seems to say something else there, not the definition of substring
and yes, by his logic also, abc does not have one
@Calvin'sHobbies I'd prefer to do those swimming pool tiles ;)
I have to do sadly
more another time
8:17 PM
@felipa are you talking to someone else somewhere else and replying here ?
@flawr As soon as I find the best single state automata for Hunt the Wumpus
@Optimizer I think "do" was just supposed to be "go"
can you edit that in ?
I could, but I don't think I will :P
by not ?
8:21 PM
@MartinBüttner I actually did want to do a mountain peaks look alike contest, but I don't have enough good images. (And those kind of challenges can be written by people that aren't me btw ;) )
yeah but if I write that kind of challenge I know I won't participate :P
(pool tiles still seem more fun than mountain peaks :P)
like you have stopped yourself in the past from participating in your challenges
"that kind of challenge"
"doesn't matter, answered"
8:23 PM
it would certainly be a popcon and I think to those I've only ever submitted baseline implementations
^ still getting a silver out of it
^ sour about badges
^ rich people judging
^ Lama
^ 2 and 3
8:27 PM
^ to the moon
^ arrow series broker
| arrow
@Optimizer I wonder how much I could sell my PPCG account for (it happens)
@Calvin'sHobbies how much ever it is, its cheaper than mine.
(or maybe not)
@Optimizer We could trade for a few weeks and see what it's like
8:31 PM
@Calvin'sHobbies we both would suck at it
Lama's don't like tigers very much.
Congratulations, @aditsu, for crossing 10k!
@Optimizer Yeah, I'd have to golf
@TheNumberOne Lama's pretty much don't like anyone else ..
thank you :)
8:32 PM
@Calvin'sHobbies and I'd have to ask at least 4 questions a week with 2 of them getting 50+ up votes
@Optimizer They like people who give them treats who poke them with a stick if they are bad.
@TheNumberOne no they don't
they spit on their faces
(Some of them.)
@Optimizer I rescind my offer
8:34 PM
@Calvin'sHobbies They hate such people even more.
@Optimizer Way to ruin one of my favorite artworks :/
not me. The Internetz
You proliferated
You rescinded the offer!
that looks so cool :)
8:41 PM
It even has chocolate chip cookies in it
Or raisin cookies :P
yip yip yip
@Calvin'sHobbies I approve of your taste in art.
@MartinBüttner Is that a joke given the taste of the above artwork?
8:44 PM
I visited a woodblock print exhibition in Tokyo a few years ago... let me see if I can recall the artist
@flawr Could you make the field constructor public ?
I think it was Kuniyoshi
I've been reading about some of Google's APIs, and I've determined one thing: I have no clue what I'm doing.
A: Sandbox for Proposed Challenges

NinjaBearMonkeyCount the number of words in a crossword code-golf crossword Consider the following standard 15×15 crossword puzzle grid. We can represent this in ASCII art by using # for blocks and (space) for white squares. # # # # # # # ### ## ## ...

9:20 PM
@felipa one of the substrings of "abc" is "ab" which does not have any matching suffix and prefix other than the degenerate cases of len=0 or len=2
9:31 PM
@TheNumberOne aren't constructors always public?
No, by default they are package private.
you mean protected?
ok yes of course!
@flawr no
well i never heard about 'package private'?
It means only the files in package main can access it.
9:33 PM
just looked it up
didn't know that
but i thoguth you can 'import main.Field'?
can't you access the constructor by thefault?
I can access everything but the constructors and the write method.
In other words, no.
9:47 PM
Q: Draw an ASCII Lightning Bolt

CzarMattIntroduction Partly inspired by this StackOverflow question, let's draw an ASCII Lightning Bolt. Write a program that takes a positive Integer n via STDIN or command line and outputs the ASCII Lightning Bolt below. Input Positive Integer n representing the number of lightning zig-zag tiers to...

A: Sandbox for Proposed Challenges

mbomb007code-golfquine Input Quine Write a program that takes an input string and outputs a truthy value if and only if the input is equal to its source code, and outputs a falsy value otherwise. Rules Your program may not read its own source. Standard loopholes apply. Scoring This is code-golf, ...

10:22 PM
I think I'll post the Samurai KOTH soon. There's been a drought of non-Java KOTH challenges recently.
10:33 PM
@Optimizer have you tried the crosswords thing in the sandbox?
Q: Parse Query Parameters

Benjamin WilliamsThe challenge is this: You are given a url that is as follows: http://sub.example.com/some/dir/file.php?action=delete&file=test.foo&strict=true There can be either http or https, there can be any number of subdomains or directories, and the end file can have any name. You are to create a ful...

10:59 PM
yay, time for Retina's Match mode to shine :)
@randomra ^
11:27 PM
Q: Coding hex board game

Renae LiderIntroduction Hex is a strategy board game played on a hexagonal grid, theoretically of any size and several possible shapes, but traditionally as an 11×11 rhombus. Other popular dimensions are 13×13 and 19×19 as a result of the game's relationship to the older game of Go. According to the book A...


« first day (1587 days earlier)      last day (1647 days later) »