« first day (1419 days earlier)      last day (3419 days later) » 

12:00 AM
Oh, nevermind, I thought it was just the unsolvable instance from GEB
 
nope
still, I don't get it... Peter says "it's just a graph reachability problem", but how do you terminate the search or limit it to a finite number of nodes?
 
that's the trick, I have a rule for bounding the search
 
I don't think you can definitively rule something out as unsolvable if it hits your bounds
It's kind of like the collatz problem
 
I think this is an 'easy' problem
not some unprovable thing
 
I think the example in GEB shows that it isn't
 
12:04 AM
isn't that a very simple unreachibility proof?
 
since you can apply rules forever, but you can never get to MU
yeah, the proof is simple, but it isn't a result of rule application
it's logic
that was the entire point of the example, IIRC
 
the bound isn't base on a number of rule applications
 
then what is it?
I think the point was that you can't prove that you can't get MU from within the MU-system
 
I don't like to share my algorithms anymore...then someone might code it in an abbreviation language ;)
the MU system can't prove anything, lol
 
oh, Dennis is called Dennis again! :)
 
12:06 AM
@MartinBüttner by "you have to edit them" I meant , you can leave everythign as is except for the tags while editing. (if that was not clear :) )
 
@feersum they still can if you post them :P
 
only if they manage to understand the program
 
@Optimizer that question was "in genuine need of editing" though ;)
 
too much effort for most people
 
@feersum is it in Python? :P
 
12:07 AM
I know, just making sure.
 
yeah, you are going to say I am in risk of Pyth
 
it will be obvious my answer was ripped off then
 
it will be obvious in any other case of "someone might code it in an abbreviation language" as well, because they'll usually say they use the same algorithm, or they won't explain the algorithm or something
 
@feersum Are you sure? General Post canonical systems are Turing-complete...
 
12:08 AM
that shows nothing about this particular problem
there aren't many rules in it
and they are limited in what they can do
 
I'm not asserting that the MU system is Turing-complete. Just that it is unwise to claim that it can't prove anything unless you have a proof of that.
 
I have a vague idea of how to prove it...but it would be too time-consuming
 
and by analogy, when I referenced the 3n+1 problem, I meant that you can't prove that a sequence of numbers is unbounded there just by applying rules
though if there is an unbounded sequence, perhaps you could prove it from outside the system
 
do you mean prove no sequence of numbers is unbounded?
I don't see how proving anything "inside the system" is a useful idea
all it allows you to do is write down lists of numbers
 
12:15 AM
"The conjecture is that no matter what number you start with, you will always eventually reach 1."
prove no sequence is unbounded, or cyclical then
cyclical without reaching 1
 
@MartinBüttner I just realised my Caesar answer affects newlines and tabs... (although feersum's use of translate would have been better)
 
And that problem can also be written using a substitution system, just like the MU-problem
"Hailstone sequences can be computed by the extremely simple 2-tag system with production rules a → bc, b → a, c → aaa."
 
@Sp3000 oh, are you going to fix it?
 
Maybe
I'm not really trying to golf anyway, I guess
 
you might as well keep the rep then :D
 
12:22 AM
Heh, not quite what I was concerned about but okay
Antigolfing: +42 bytes
 
@PeterTaylor I'm afraid your GS code has the same problem
it doesn't leave characters outside printable ASCII untouched
 
I'm assuming the JS one does too
:/
 
no that's fine
the regex only matches printable ASCII
 
[ -~] right
 
I just commented, the JS one is broken for empty string though...
 
12:40 AM
(for some reason I had the impression your code had to use printable ASCII too, which made things a fair bit harder :P)
(slipped my mind, I guess)
 
nope, you could probably do this in APL and write some code where only a single character switches
 
D: unicode, eh
 
1:08 AM
ASCII was good enough for my pa, and it's good enough for me
 
 
1 hour later…
2:11 AM
0
Q: Almost-correct answers

YpnypnSometimes answers are posted, which almost comply with the specs. They might be a clever, interesting take on the problem, but not follow one of the rules, or the program might crash in a few rare cases. What should be done with these answers? Upvote, downvote, or delete?

 
 
6 hours later…
8:27 AM
@MartinBüttner Will fix when I have time and then consider whether to edit and undelete or to post a new answer. If the change is too drastic then it wouldn't be right to start on +5.
 
 
1 hour later…
9:49 AM
@PeterTaylor fair enough
 
which post is in question .. ?
 
caesar cypher
wtf, I just got the bugdroid hat
 
why wtf ?
is there any issue in his submission ?
 
because I've never used or even downloaded the android app
 
someone hacked ur account ?
 
9:52 AM
@Optimizer it doesn't leave non-printable ASCII untouched
 
Ah ^^
 
@Optimizer there was no activity on my account either though
0
Q: Received Bugdroid hat without using app

Martin BüttnerI got the Bugdroid hat last night on CodeGolf.SE. The thing is, I never used (or even downloaded) the Android app. Did this happen to anyone else, or should I be worried that my account got hacked (in which case, why would anyone use the app then)? As far as I can tell, there wasn't any activity...

 
you can get it by droid, or get it by bug
4
 
LOLOL
 
this is good though, I wouldn't have been able to get the hat :D
 
10:06 AM
bluestacks.com
 
oh, neat
 
damn, you are catching up!
 
slowly... ;)
most likely, we'll end up tying... apart from Naruto, Red Baron and 30 Minutes or Less (and Hairboat and Eureka) they are all equally doable for both of us. so unless one of us is lucky getting one of those 3...
 
you have the gold one yet ?
 
I only need to post 13 more answers on the code golf tag
 
10:19 AM
ahh
 
@overactor You look different today, are you alright?
 
hat season inspires many to grow heads?
 
he's angry at Calvin'sHobbies
 
@MartinBüttner The hats made me get rid of my gravatar
 
I grew a llamma
 
10:21 AM
@Optimizer I that Carl btw?
In any case, I wanted to ask a question
 
I dont know. just took it from imgur
 
2
Q: Meta-atomic code golf

coredumpIn this meta-challenge, you will score your own atomic-code-golf submissions. More precisely, you will need to write a program P in language L such that the atomic-score of P is produced by P. Score The idea behind atomic-code-golf is to count language tokens instead of bytes. However, in prac...

Would
01
Be a valid Marbelous submission?
 
no
lets say a 100 bytes long marbelous code.
is given as input
what is the output ?
 
I might not get the question
 
if we didn't have multi-cell devices it would be really easy though
you'd just count every second non-newline character in the input
 
10:25 AM
@overactor the program should output the number of tokens in any other program of the same language provided as input
 
with multi-cell devices it's damn hard to do in Marbelous itself
 
@Optimizer aha
 
and the score is the number of tokens in your code
 
hmm
@MartinBüttner It really is
@MartinBüttner Are empty cells tokens though?
 
that challenge was so poorly specified, why bothe
 
10:31 AM
@overactor I'm pretty sure they are, as they affect the program
 
And what about comments
 
they are technically no-op devices
hmmm, comments...
@feersum this.
 
And empty devices that are implicit
 
this?
 
(At the end of a line for instance)
Marbelous is tricky when it comes to tokens
 
10:34 AM
@feersum I agreed with you.
@overactor oh good point
 
I feel like empty devices don't count
All they do is move around tokens
 
how about we define a board is a syntactical token. print "1" gg
 
They could be thought of as grouping symbol
 
@Doorknob冰 thanks for getting me Bugdroid ;)
 
haha, lol
 
10:58 AM
he did ?
 
@Optimizer Yeah. See balpha's answer to Martin's question.
3
A: Received Bugdroid hat without using app

balphaYou don't have to worry, your account is under your sole control. Unfortunately when creating the query to find out who deserves a hat, I thought we recorded which app was used to create a comment, but we do in fact record which app last touched the comment. And in your case, a comment that you h...

 
lol
 
11:13 AM
oh sweet, I've got my downvote privilege on MSE back
o.O
I posted this two years ago when MSE didn't even exist on MSO. apparently it got automigrated when they created MSE... and now it's closed for being off-topic there :D
also, now I know that you get no sort of notification if your question gets close voted or closed.
 
11:41 AM
man, I wanted your cypher question to go viral :/
I would personally give out a bounty to a code which can reuse more than 20% of the two programs in one another.
(instead of having two separate block , 1 for program, other for the cypher of the first program)
 
 
1 hour later…
1:01 PM
@MartinBüttner Lol, uhh, you're welcome I guess :P
 
1:20 PM
get me a naruto ?
 
1:59 PM
yoooo i have moderator tools now
 
Congrats!
 
i like this site: if you're good at making code small you're awarded with unimaginable power >:D
(i haven't looked yet but i assume it's unimaginable)
 
Haha management is throwing a huge fuss. Someone created a User Story under a User Story under a User Story. It broke their diagrams.
 
critical flowchart failure
 
The problem is that the top story is like "Make a product"
The mid level stories are all huge and encompass entire pages, like "Make a page."
Someone got smart and decided to make a third level that truly represented what a user story is: something the user said.
We'll get there. Management will catch up slowly.
 
2:08 PM
@Optimizer I was gonna do that as well, but feel free to go ahead :P ... (if you do, I might put another bounty on it, for the best submission using two different languages)
but I think this isn't the kind of question to go viral. it's way too complicated for that, "Caesar-Cypher-Mania" isn't a title that speaks to many people around the network I think.
 
I accidentally highlighted the top of the page and I thought we had graduated.
 
lol
 
haha, just found a typo in a challenge "coma delimited"
 
2:13 PM
I once wrote a puzzle on comma comas :P
 
@Optimizer warm welcome is back in the HNQ ;)
 
@MartinBüttner what do you mean by two different languages ?
 
if you often experience prolonged states of unconsciousness, and you spend your waking hours teaching people to be moral, is that an example of coma separated values?
 
@Optimizer the two parts of the submission don't have to be in the same language
 
is it written in the question ?
 
2:16 PM
@Optimizer yes
 
oooohhhh
dunno how I overlooked that
@undergroundmonorail how did u become a mod ?
and congratz!
 
@Optimizer i didn't, i just hit 2000 rep
which means i can access mod tools
 
:D oh!
 
thanks though :)
 
@MartinBüttner not giving me anything though :P
 
2:19 PM
@Optimizer also, you know how when you hover over a message it shows you what it was a reply to, or the person's last message if it wasn't specifically a reply to anything? for a whole second i thought you were pointing at my stupid pun and saying "how did this guy become a mod"
"come on"
 
which pun ?
 
> if you often experience prolonged states of unconsciousness, and you spend your waking hours teaching people to be moral, is that an example of coma separated values?
 
ah lol. I think it automatically picked your last message
 
my question was a fresh question.
 
2:22 PM
fresh. hip.
 
2:35 PM
hi
 
2:48 PM
@PeterTaylor I'm slightly disappointed that there's no combination of spells in WH, where you have to decide between using a two-hand gesture for 1 spell or using two one-hand gestures for 2 spells.
@PeterTaylor I've added test cases. Would you mind reviewing them, and having a look if you can think of any edge case I missed?
 
I just realized that PPCG is just like Minecraft, but way more judgmental. You built a castle? Someone else already built one. DUPLICATE! You built a dirt house in our beautiful city? OFF TOPIC!
I swear I had a good reason for that rambling. I was trying to think of a way to convince PPCG to come together on a physics engine
 
3:04 PM
@MartinBüttner why after so long ? (re: comment)
 
@Optimizer because I only now saw your comment
@Rainbolt That seems like... an odd way to do so... what kind of physics engine?
 
I didn't realize there were categories.
 
@Rainbolt let me rephrase that: for what purpose
Actually, I'm not even sure I understood what you meant. I read "to come together on a physics engine" as "to collaborate on writing a physics engine", but on second thought, I have no idea what you were trying to say.
 
Well, a physics engine has manageable parts right?
And each of those parts is all about optimizing some equation to make it very fast
 
It's a piece of software, so I suppose yes
 
3:10 PM
So it would make for a good challenge
I didn't mean "write it all in one go"
 
Oh you actually meant as challenges
 
Yes, like "Build component A of the physics engine."
 
I thought, just like the people in chat, for the fun of it.
 
After we have components A through ZZ, we could put them all together and watch an object fly around lol
 
here is a start ;) (although this isn't how you would handle collisions in a general-purpose physics engine)
@Rainbolt I'm pretty sure there are some parts of the engine that wouldn't make fun challenges (basically, the architecture that holds everything together)
 
3:13 PM
Yea, but I bet the architecture is the least performance intensive component anyway.
I'm sure all of the magic happens in the actual physics.
 
oh sure, but someone would actually have to pull everything together
 
2-3 people integrating and coming up with tasks
 
I suppose we'd need people who are familiar with existing physics engines (to my shame, I'm not)
 
Assume for a moment that I did know a lot about physics engines (I don't, but I could watch a series of lectures).
I would know how to pull the pieces together, and I could write the pieces, but they might not be very fast.
Wouldn't that look cool on a resume? "Helped write a piece of the fastest physics engine ever. See my accepted answer here."
 
haha, yeah. although I doubt that even the people on PPCG could come up with something better than people researching this stuff for a living.
 
3:18 PM
"fastest" -> "shortest"
 
@MartinBüttner That's a good point.
 
"We made a physics engine, it takes 30 million years to model 2 point particles, but its only 1KB"
5
 
JS4K or JS1K has shorter submissions
 
Updated, golfed 9K unnecessary bytes
;)
 
3:40 PM
Ok, I have the grand idea to end all grand ideas.
Let's build a physics engine in Minecraft.
"It only takes up 30 kilochunks."
 
the grand idea to end all grand ideas is to implement Minecraft in Minecraft.
 
4:44 PM
lets close that second answer quickly too :3
 
you need a red shirt? :P
 
^.^
how do you have it already :/
 
Does "Questions without an objective primary winning criterion" apply to the newest question? He has two objective winning criterion, but neither of them is "primary"
 
:)
@Rainbolt I was wondering about that. But in any case, it's definitely unclear what he's asking.
aaaand, deleted again
 
forget winning creteria. His question itself is unclear. See my latest comment
damn!
1 vote to close :( lets vote to undelete and then finally close it
 
4:47 PM
That's like saying "Forget your close vote reason, because my close vote reason." and that doesn't make much sense.
 
@Rainbolt No, I'm saying that it doesn't matter whether one close vote reason applies or not, because another one definitely does apply, so you could just choose the one there is no doubt about.
@Optimizer deleted counts for red shirt too
 
oh!
 
@Martin Sorry, I should have directed that towards Optimizer. "forget winning creteria"
 
but I cannot retract my undelete vote now :P
 
@Optimizer you are ridiculous...
(for undelete voting a bad, close to unsalvageable challenge, just so you can close vote it to get a hat... I think that's what Peter meant with "counterproductive to site quality")
 
5:00 PM
:D
for what its worth. I also think that I am ridiculous for doing that
 
 
1 hour later…
6:07 PM
So I was looking at some satellite photos
and apparently they're 11136x11136 pixels. That's 2^7 * 3 * 29
can anyone here explain how a choice like that is made
 
I think that is overthinking :)
 
@EricTressler that is odd
 
@EricTressler It might be more like 2^12 * 3
Because some images use 12 bit colours?
 
11136 X 11136 is even
 
Yes, but with some overhead, or whatever
 
6:16 PM
2^12 * 3 is still weird. I want to know why they picked that number
I'm going to just email NASA
Pretty sure they'll get back to me, it's a legitimate question
 
cc whitehouse
 
(in other unrelated news, it's Steam sale time)
 
Oh I lied the 12 bit thing is for radiometric resolution
 
@Sp3000 :)
 
Which describes how finely it can tell apart different energy readings
 
6:19 PM
This is not a crazy thing to do. I'm not trying to email the director of NASA or something
 
or something
 
cc NSA
 
^ Redundant :P
 
@MartinBüttner Funnily enough and as I thought, some games were cheaper to pick up from the auction
 
xD
 
6:20 PM
@Sp3000 I didn't participate in the auction at all, I used all my gems to craft boosters
sick, Savant - Ascent is reduced by 50%... now it's only 1 euro :D
 
Super Hexagon's on $0.98. Just for anyone who's been confused by Martin over the past week.
 
lol
its a good game ?
 
I'm surprised there isn't a super hexagon question in the sandbox :P
 
it's amazing (if you like these kinds of dexterity-based games)
it's also trippy as f*ck
 
6:22 PM
Try Hexagon and see: terrycavanaghgames.com/hexagon
 
what is these kinds of dexterity-based games
 
Oh my god, it's full of hexagons?
 
@Optimizer try it out
 
How would you do a Super Hexagon question lol
 
ok
 
6:23 PM
@Sp3000 haha, I've thought about that so many times while dying over and over again on that game :D
 
the question will start like
"As we know there is a super awesome game, super hexagon..."
 
@Sp3000 thankfully Savant is the only game on my wishlist that's sufficiently reduced right now.
 
(Weekly challenge theme? :P )
 
Given this hex and this position, output # of rotations?
 
Does the hexagon have to be regular ?
 
6:24 PM
Probably need to add more so it isn't boring :P
 
super hexagon playing bot
given the pixels of the screen ouptut which button to press
 
D: that'd be an... interesting image processing problem
 
and path finding !
 
@feersum not really
 
I liked the QWOP-playing bot challenge (no one else did though)
 
6:26 PM
you never need to plan ahead more than one segment
(as a human you do, because you can't see some of the patterns fast enough, so you have to learn them, but for an algorithm, you definitely don't need to plan ahead)
 
yeah I guess you could just draw lines radially, then find nearest one without something above it
 
then we need to design super duper hexagonerest mode in the challeng simulator
which requires deep pathfinding
 
except you can't even look ahead :P
@FryAmTheEggman the beauty of super hexagon is, there's some perspective distortion in some levels :D
 
lol the original practially gives me vertigo, no thanks ;)
 
@MartinBüttner You know, I always find booster packs weird because for some games it's cheaper to nab the individual cards from the market
 
6:32 PM
@Sp3000 well I do check if selling the booster or the cards is worth more before unpacking them
 
:P well now that gem prices have gone way down you can probably make a profit making boosters from that
 
I also got a foil holiday card from crafting a badge... that alone was one euro. ^^
 
:o
Wow
 
@MartinBüttner @Sp3000 so the whole game is like that only ?
dot line triangle square ...
 
@Optimizer yes, what more do you need? :D
 
6:41 PM
:D
I get bored of stuff pretty easily
 
Full game's more polished and has more pattern variety but basically that's the idea
It's the type of thing you keep hitting the restart button on
 
@Geobits Just checked the first page of the 2014 challenge. None that I can consider valid can be trivially adapted.
It reminded me of closing a loophole though.
wow my bugdroidbug from earlier today in surprisingly popular on MSE :D
 
K. I was going to go through and look for an example later, but if you're satisfied that any adapted answers would be non-trivial or non-competitive, that works for me.
 
well, there are 6 more pages... :D
 
I know. That's why I haven't checked them all :P
 
6:45 PM
@MartinBüttner the 2014 challenge seems unfairly biased for stack-based languages..they have no syntax rules so how is anything going to be invalid?
 
there was one (in R or JS) that might be adaptable, but I know that a much shorter answer exists
@feersum are we talking about the same thing?
 
yes
 
then I don't understand what you're saying/implying
 
I'm saying golfscript or something similiar has no syntax rules
 
so?
(in fact, they do, {} and "" have to be paired)
 
6:46 PM
so it's a lame rule
 
oh, I see
 
triangle in third try..
 
with "invalid" I meant cheaty answers, that use error output, output anything that contains 2014, or fetch the current time.
not syntactically invalid
sure you might be able to pick some of those answers, but I'm not sure you'll be able to actually make them output 2015
 
the answers have to be from the 2014 challenge? I thought they could be from anywehre
 
(at least not the short ones)
@feersum no
Geobits suggested disallowing the 2014 challenge as a source, because it could be too easy
but I don't think it's easy at all
 
6:49 PM
I don't think 2014 answers are any more useful than random code
 
yeah, that's what I was checking
 
I'm talking about this rule:
"You must be able to show that the original source block is syntactically valid in your language."
 
@feersum I know, but I'm not sure what it has to do with the original discussion between Geobits and myself. :D (or was your comment not related to that at all in the first place?)
 
my definition of "same" was that it was about the same challenge, lol
hmm I typoed 2014 for 2015
 
ahhh
that would explain it :D
 
6:51 PM
That makes things much clearer :)
 
@feersum So to answer your concern. Sure you can probably claim that any Java code is valid CJam, but why would you want to use a Java snippet for CJam if you might be able to solve it with a much shorter CJam or GolfScript source?
I don't think it's a problem.
the golfing languages will win again anyway, and I don't think it's a bad thing.
 
a Cjam code could use any answer on the site, but a Java answer could only use another Java answer
 
@feersum So could a Whitespace answer. I don't see how that's a problem. (The alternatives I can think of are worse, imo)
 
It does seem to say something about not hitting run time errors.. but i admit I don't really understand what Martin means there
 
@FryAmTheEggman that isn't a necessary requirement. hence the "in the best case"
 
6:55 PM
@MartinBüttner For codegolf.stackexchange.com/a/17068/21487 do the ords all collectively count as one code block or can you use,say, just ord('!')*ord(':')+ord('d')?
 
@Sp3000 they're all one block
 
Damn
 
@Sp3000 isn't that too long for you anyway? :P
 
... I should have counted before I tried :P
Although having a Python answer would be nice
 

« first day (1419 days earlier)      last day (3419 days later) »