« first day (1808 days earlier)      last day (3026 days later) » 

7:02 PM
@Sherlock9 So post your answer on the other challenge?
 
@TimmyD The other challenge was cluttered and had too many arbitrary rules
 
0
Q: What is the shortest LOOP program that outputs 2016?

Lew BaxterUse a minor restriction of the LOOP language described under Wikipedia's "LOOP (Programming Language)". The restriction is to eliminate constants. So, the language contains increment: xi++, decrement: xi−−, assignment: xi=xj, sequencing and loop. All variables are initialized to 0. The output o...

 
I didn't want to spend another hour trying to get my code to do all that
@PeterTaylor, if you don't mind can we reopen the recent arbitrary base conversion question. The first one is too convoluted and I can see this new one being, at the very least, an interesting easy question for the new users. Or a popularity contest, perhaps
 
Aside from the new challenge using the same base for both input and output, they look pretty much the same.
 
Well, I suppose you're right.
@PeterTaylor Never mind
 
7:13 PM
user image
13
 
Anonymous
Those base conversion questions are one of the cases where I would consider closing the old as a dupe of the new
 
Anonymous
The new one is a much better challenge IMO
 
@CᴏɴᴏʀO'Bʀɪᴇɴ jsconsole.com ?
 
Doesn't work on IE?
 
7:18 PM
I didn't test it on IE.
 
wait
It works.
I didn't see the input :P
 
Well, nice.
 
Anonymous
@CᴏɴᴏʀO'Bʀɪᴇɴ Spooder?
 
Anonymous
7:22 PM
 
Anonymous
spooder
 
⊙▃⊙
 
Anonymous
For when you're not sure whether to upvote or downvote:
2
 
Anonymous
 
7:23 PM
l'chaim!
 
What does @geokavel's avatar look like @Mego?
 
Anonymous
@CᴏɴᴏʀO'Bʀɪᴇɴ It's rotationally symmetric, so it will be identical
 
i have a question, should there be an archive of old KOTH's that haven't been fully tackled yet?
 
@Mego Just confirming.
 
7:39 PM
@geokavel That would be the Sandbox. Unless you mean something different.
 
i think it'd be more interesting if "metagolfing" referred to writing short programs that write short programs, and for the current usage "metaprogramming" or just plain old "writing a search program" could be used
 
@El'endiaStarman I mean ones that were posted, but have not gotten the best possible answers yet.
 
ah, okay
Well, what does "best possible answer" mean in the context of a KotH?
 
In the case of mine, I have some good mid-range intelligence bots. However, none that actually use strategy to predict what their opponent does. I think that's the next evolution.
 
Apparently you guys were just talking about that base conversion challenge hehe... exactly why I stopped in
 
Anonymous
7:46 PM
 
This chat is currently not working for me.
 
lol
 
hey guys!
 
My observation of the old one is that it's much more complicated, and only has 2 valid answers (neither of which would be competitive on the new challenge, which is meant to be much smaller)
 
@Mego That's...actually kinda neat.
 
7:47 PM
Any suggestions for changes that might have it not be seen as duplicate?
 
man
my throat hurts
I have snot everywhere
I have a headache
 
Anonymous
@Mwr247 Like I commented before, I'd be more inclined to close the old one as a dupe of the new one, because the new one is a much better challenge in my opinion
 
my ears hurt
 
that sucks
 
Thanks
 
7:49 PM
hello?
 
hello
 
hello‽
 
¿hola?
 
hallo?
 
7:54 PM
damn that was some weird problem.
 
hallo!
 
if i don't survive, tell my wife, "hello"
 
@MitchSchwartz what are you doing that might kill you?
 
Living.
 
7:58 PM
he's golfing
 
Death always follows life.
And very rarely, life follows afterwards. :P
 
all serviceable answers. :) (it's a futurama reference)
 
Very cool graphic.
It's worth noting that so many people die of heart disease and cancer...because we're so long-lived.
 
Should/Does anything get done about answers that don't actually meet the spec? This is assuming OP has vanished and has been notified by comments on their answer.
 
8:02 PM
@MorganThrapp Is the answer marked as non.competing?
 
Nope.
 
@El'endiaStarman So, if we lived shorter lives, we be the victim of much faster causes of death?
 
19
Q: Is there a place I can place my so-close-but-not-quite answer?

shelvacuI'm asking in general, but in my specific case I was working on an answer for Programming a Pristine World and got $><<"%c"*5%%w(w o r l d). The fatal flaw here is that #<< accepts an array, so if "%c"*5% is removed, $><<%w(w o r l d) is a valid program. I'd still like to post this for two reason...

 
@Rainbolt In a sense. You die of heart disease and cancer if you survive everything else. In the past, diseases and illnesses like the flu were much more deadly and widespread.
 
Yeah, I've seen that, but it doesn't address answers that DID get posted, but miss some part of the spec.
Though the linked one on that does, okay looks like comment and down vote.
Thanks. :)
 
8:05 PM
Not to mention wild animals, other humans, starvation, child birth, etc...
 
@El'endiaStarman I had the flu once. I slept on a couch for over 24 hours without waking up once. When I went to sleep, Mom was cooking one thing, and when I woke up 24 hours later, she was cooking something totally different. It was mind boggling.
 
Is someone willing to read over this CA challenge? meta.codegolf.stackexchange.com/a/8002/2867
 
> I intend to live forever. So far, so good. -Steven Wright
 
I've change it to be 3-state.
Also, I'm slightly debating whether the challenge should be to find the best rule or to write the program that finds the best rule.
 
8:07 PM
Best rule eliminates any debate about how long the program is allowed to run and on what hardware
Maybe score on best rule but require the actual program to be posted for validation?
 
Given a rule, anybody can validate the score based on the generation limit I've set.
 
I meant to validate that they came up with the rule, and didn't copy it from somewhere
 
15
A: Definitive policy about answers not meeting the challenge specification

DennisScope As I see it, there are five types of invalid answers: Answers that produce incorrect results. This is the most common type, and usually an accident. Answers that produce correct results, but break a rule of the challenge, ignore parts of the spec or violate a loophole. For example, ans...

 
Oh, so I should raise a mod flag?
 
8:13 PM
Someone seriously needs to consider breaking that meta post into pieces
 
@MorganThrapp Flag or, since we're already here, post a link to the answer.
 
If I have to read a short story every time I decide if I should delete vote, I'm probably just not going to vote
 
There's actually 3 different ones.
The second one is the one that really bothers me, because the only reason they're beating my answer is because they're not following the spec. :P
 
@Rainbolt Assuming you're talking about my answer, just stick to the TL;DR section.
 
You put the TL;DR all the way at the bottom...
 
8:19 PM
That's where it's supposed to be. Too long, didn't read.
 
Ah, I see
Else it would be "too long; won't read"
 
Exactly.
 
@Rainbolt I'm going to have "best rule wins" and ask that they write a description of how they found it (including programs they wrote).
 
I gotta say, there are some neat oscillators you can build with B3/S23 and B1/S1.
 
Hey, I found a still life. :D
 
8:27 PM
That's a loaf
 
...wat.
 
@El'endiaStarman Is there a way to make the grid bigger?
 
@FlagAsSpam That's on the to-do list.
I guess I'll do that next.
 
Coolio.
(this is cool. :D)
 
8:30 PM
:D
 
Have we had a Conway's Game of Life Cops and Robbers?
Cops make a pattern with a certain number of blocks filled after a certain number of frames, robbers try to find the pattern that made it?
 
that sounds fun
 
@FlagAsSpam That has the same problem as most CnR
There will be exponentially more Cops than Robbers. The amount of work is extremely unbalanced.
 
That's not even programming.
 
@feersum Isn't it?
It's boolean logic by taking steps to a solution.
 
8:36 PM
@mınxomaτ That's also what makes them so viral.
 
That's programming in my book.
 
@FlagAsSpam This problem is also often unsolvable.
 
@PhiNotPi I don't consider this a good thing. Made the last CnR pretty boring and cluttered.
 
@PhiNotPi Why would that be unsolvable?
 
There are many patterns that can be built which don't have any predecessors.
 
8:37 PM
I assume like in the programming language quiz, we'll have to find a starting config
 
Many GoL patterns vanish.
 
You could use a different solution - just has to be under a certain amount of blocks used.
 
In a cellular automaton, a Garden of Eden configuration is a configuration that cannot appear on the lattice after one time step, no matter what the initial configuration. In other words, these are the configurations with no predecessors. They resemble the concept of the Garden of Eden in Abrahamic religions, which was created out of nowhere, hence the name. According to Moore (1962), this name was coined by John Tukey in the 1950s. A Garden of Eden is a configuration of the whole lattice (usually a one- or two-dimensional infinite square lattice). Each Garden of Eden configuration contains at...
 
It's like block-golfing! (I know they're not called blocks. I don't know what they're called.)
 
Hello
 
8:38 PM
@FlagAsSpam In what way does "Fill in some squares on a grid and then run a GoL program" require programming?
 
We could require that cops must post a pattern with a predecessor.
You could also do something interesting with the scoring by considering how long it takes to evolve to that state.
 
@El'endiaStarman I think the idea is that to become safe, they must post their answer
 
@feersum "Fill in some squares on a grid in order to find a solution."
 
but the robber answer need not be the same
 
If robbers just have to match the final state, that takes care of the myriad ways to start.
 
8:39 PM
It's like "Type in some characters in order to find a solution."
@El'endiaStarman This is what I mean.
 
@FlagAsSpam I don't know what you mean. State the challenge idea again?
 
Should this be community wiki? If not, why?
 
Hmm. There's another problem: it'll probably be really easy to "crack" submissions unless you require that the number of steps from start to finish be matched as well.
 
@feersum Cops post a result after a certain number of steps. The cops have to find what the original code was (or a fewer number of alive toggled blocks) that produces the same result after the same number of steps.
@El'endiaStarman Yes - that's something I mentioned earlier.
 
Looking at some forums, GoL enthusiasts are really enthusiastic.
 
8:41 PM
That requires no programming by cops.
 
As of right now, I think this would be a pretty terrible CnR challenge. In order to improve it, we would have to place some extra restrictions on the cops.
 
@feersum Yes, it does - reducing the number of alive blocks.
 
@FlagAsSpam You're not making any sense.
 
@PhiNotPi maybe scoring by some composition # of alive and # of turns?
 
@FlagAsSpam I think what he's getting at is that all CnRs thus far have required the cops to write code.
If we don't have that, then perhaps it'd be a better fit for Puzzling.
 
8:42 PM
Not only CnRs, but all challenges on this site require writing code to be on-topic.
 
@El'endiaStarman But this is code. It's just a two-dimensional boolean code.
 
We would have to make it so that "some random pattern someone draws" does not automatically qualify as a cop. This also makes the problem for feasible for the robbers, since more restrictions = less raw brute force.
 
@FlagAsSpam It's a little more abstract than you think.
 
@mınxomaτ Well, yes, it has behavior given certain "shapes" (? I can't words this), but all programming languages do.
 
Creating a series of symbols that is capable of being executed in some Turing-completes system does not by itself constitute coding.
3
 
8:44 PM
^
 
To be programming, the creation of the code should be directed toward some goal known by the programmer.
 
FWIW I feel that it might not technically be fully on topic, but I think that it would be fun, and in the spirit of this site
 
That's what I was thinking too.
I'm not sure how Puzzling would react to having a CnR... :P
It's also a very different..."flavor" of puzzle.
 
@feersum Yes - the Cops were aiming to reduce the initial amount of alive blocks (I'm just gonna call them blocks) by logical process.
It is essentially code.
 
@FlagAsSpam: Post a question on Meta about it? I'd like to see more community discussion on this.
 
8:46 PM
Can someone please restate the exact goal of this CnR? I feel we might not be on the same page.
 
Cops post a pattern after n steps, robbers recreate the starting (or smaller) pattern.
As I understand it.
 
okay
 
@El'endiaStarman Smaller meaning fewer number of alive blocks.
 
@FlagAsSpam Why would they try to do that? You did not mention any goals relating to the number of alive cells.
 
8:47 PM
@feersum It would be easier for the robbers to solve if they had unnecessary amounts of alive blocks.
 
??????
 
Well, one thing we would need to figure out (empirically) is "given a random configuration of cells, how easy is it to find the predecessor?"
 
In the interest of avoiding exponentially-exponentially unbounded search space, and to give the robbers a hint, it may be beneficial to have the cops score based on alive blocks in their starting configuration and state what that is.
 
@FlagAsSpam: Why don't you come up with an example of what you have in mind? That'd probably help.
 
"13 alive blocks at start, 26 steps"
 
8:50 PM
@TimmyD This is my intention.
So, say I had this configuration after 59 steps:
 
how many starting squares?
 
These are both still lives at this point. A robber could either post my original solution:
Or, because he knows that they're still lives, he could just do this:
 
(It makes me really happy that my tool is being used for this. :D)
 
but @FlagAsSpam what about the # of initial starting blocks?
 
8:52 PM
@Maltysen 20.
 
then the trivial solution is incorrect
 
Although it could be golfed to 8.
 
oh I see
it can be <=
 
Yes.
 
So cops want to avoid trivial reductions and still lifes.
 
8:53 PM
It's still a ridiculously huge search scope for the robbers, though ...
 
It is too easy to create patterns that are computationally infeasible to crack, without even knowing anything about GoL or coding.
 
What if we also had cops post the bounding box?
 
@TimmyD Let's say we reduce it to a certain size grid, 20x20.
 
Click some squares, run the Life program, make sure that it didn't die out, and submit.
 
@FlagAsSpam 20x20 is huge, though, for this kind of stuff.
 
8:54 PM
@feersum but most starting conditions trivially reduce
 
> make sure that it didn't die out
 
2^400 possible states of a 20x20 grid
 
that's waht that step is for
 
@El'endiaStarman My input was arbitrary.
 
8:54 PM
@feersum but then like 90% of them become still lifes
 
Very small still lives.
 
or simple, easily recognizable oscillators
 
omg. you might have to try TEN times!
 
@feersum Would you mind showing us?
 
Showing what?
 
8:56 PM
As in, come up with a finishing pattern we could not hope to guess or best.
You're saying that this problem is way too easy for cops. (I think.) Should be easy to prove, yes?
 
Do you really need me to do it for you?
 
@El'endiaStarman I don't know how to phrase the "What constitutes 'programming' for CnRs?", what should I say?
 
21 initial alive blocks, 8 generations --
 
@feersum I don't precisely know what you have in mind.
 
8:59 PM
how many steps?
 
1000 in binary
 
oh didn't see that
 

« first day (1808 days earlier)      last day (3026 days later) »