« first day (1849 days earlier)      last day (1879 days later) » 

@mınxomaτ "Error"
@VoteToClose Works fine.
:D It's so floofy!
I just released Retina 0.7.3. See the CHANGELOG for details. It's mostly a couple of minor changes to improve golfitude, but there's also one new feature: $.n inserts the capture's length in substitutions. (cc @mbomb007, @NinjaBearMonkey, @randomra, @Sp3000, @TheNumberOne, @Dennis, @DigitalTrauma)
What, I don't get a ping?
Eh, fair enough.
I still got a ping. c:
@VoteToClose There there, we all love you :)
\o/ I am appreciated in an outwardly manner!
Thank you @MartinBüttner @Sp3000 for this priviledge.
@VoteToClose but you got one from mınxomaτ not that long ago
This is true.
Thanks @mınxomaτ @somebody
12:59 PM
Whoops, looked at TVTropes again
1:15 PM
I just posted a pop-con. >.>
It's been in the Sandbox for 4 months. >.>
I may've forgotten about it until a few minutes ago.
Q: What time is it again?

VoteToCloseWhat time is it again? The Inspiration The Challenge I'm tired of seeing the same old-same old clock in my top-right-hand corner of my screen. It's time for a new clock. While the clock clock sated my lack of interesting clock for a little while, it simply will do no longer. Your job: make a b...

ohai there
@MartinBüttner facepalm Yup. Thanks.
seems a bit broad
I had it in the Sandbox for 4 months. >.> Nobody mentioned that. Any ideas on how to fix that?
not off the top of my head.
1:20 PM
the clock clock is nice though :P
Yeah. I saw that and was like I'm making a challenge out of this.
Is there a way I can put it directly in the challenge?
doesn't look like it
:\ That's a little saddening.
unless you screencap it and put a gif into the challenge
1:22 PM
o-o Not the worst idea, but it wouldn't show current time. :\ Oh well.
what's with the barrage of garbage challenges?
@VoteToClose what OS are you using?
@orlp >.> I don't know. I hope you aren't talking about mine.
@somebody Mac OSX Mavericks.
we've done a challenge about that one
1:25 PM
@somebody I don't understand the output of this.
update with any changes to the system clock
@VoteToClose RTFM
@somebody That helps very little.
Oh, hey, I found the docs on it.
reading the clock section
end of page
@VoteToClose how will it update with any changes to the system clock?
@somebody On my challenge?
1:28 PM
That's really up to you. I mean, it's part of the process, reading the time. >.>
It basically restricts the use of using sleep(1000) or the like.
but basically you need access to system time?
what kind of libraries are allowed?
is it allowed to drain battery a lot
1:42 PM
Unless explicitly forbidden or in the standard loopholes, you are allowed to do as you please.
i wonder if the userscript should add link to loopholes
@Downgoat ^ Good suggestion.
Maybe next to the "Post Answer" button?
i'm thinking of using three.js + electron + node.js for clock challenge
just i have no idea what to do
2:08 PM
code challenge: make a website that crashes chrome even if I prepend "view-source:" to the url
1 hour later…
3:10 PM
@MartinBüttner Just pulled it on TIO.
3:31 PM
thanks :)
Did you see the radiation softened quine?
like I said, it would have to be a language where our (makeshift) quine definition doesn't apply
3:50 PM
I wonder how hard it would be to find a fixed point that doesn't require a decoding function.
Probably very hard.
yeah I was wondering earlier whether you could find a ???128b:c quine in CJam
4:02 PM
hi somebody
@Dennis Pull Vitsy? (I found out that I left debug code in there and forgot to take it out. ಠ_ಠ)
Thanks. :)
You ninja'd my pulled. :P
;) I saw the edit happen.
I forgot to remove "1;" output for debugging from the "if" method. ಠ_ಠ
I've done that many times...
@VoteToClose I'm a little conflicted because I think the clock challenge is a tad bit broad (esp. since Tweetable Mathematical Art got locked 10 days ago) but a mod posted in the challenge. So, to compromise, have a close vote...
and a clock
4:12 PM
@Sp3000 :o Noice.
I would vote to close as well if I could do such a thing.
@Dennis Why can't you?
My "vote" is binding. If I press the close button, it's closed.
because he's a mod
we can't "vote"
Sad. :c
Then again, not sad?
4:20 PM
Too bad I never saw it in the sandbox. Clearly, four months is not enough time...
4:30 PM
I normally leave my stuff in the sandbox for 5 days. xD
Yeah, there's not really a point after that. Once it's off the first page, it's highly unlikely somebody would find it by accident.
^ That's my thought behind it.
ohai Conor.
The hate downvote was removed ._.
4:43 PM
o-o do wat
Guys are you proud of me I spelled Liechtenstein correctly first try
It sounds like that distance function but it isn't
BTW isn't there a "most controversial posts" SE query?
@CᴏɴᴏʀO'Bʀɪᴇɴ SELECT [Id] from Posts WHERE OwnerDisplayName is 'Seadrus'
^ exactly what I was trying to do XD
4:48 PM
160 days? GG.
Hay @jrich
anybody here know javascript?
Let's break that 160 day streak! high-fives
That's some strong willpower, resisting the guiles of TNB for 160 days
i've only taked here once before lol
but i'm back now!
Nope, nobody knows JavaScript. We all use good languages, like COBOL.
4:49 PM
@jrich I DO
oh ofc. I'm too young for this.
JavaScript? Can I eat this?
Haskell ftw!
lol. so I'm trying to test a possible radiation resistant quine
@flawr Don't you'll get JavaSick
Q: Radiation Hardened Quine

minerguy31As you should (hopefully) know, a radiation hardened quine is a quine that you can remove any one character from and still print its original, pre-modified source. The thing is, with most of these you can only remove one character; otherwise everything breaks down. This is where this comes in; yo...

but it outputs to console.log so I dont know how to brute force it...
4:50 PM
You want to brute-force check every combination of a removed letter?
hesitant to post if im not sure its valid
Code? I won't steal it ;)
its ~900-1000 chars i think
That's a quick Python script
4:51 PM
I'll write a checker real fast
I have node.js if that helps...
idk if python has a js evaluator
var string = "your quine thing";
for(var i=0;i<string.length;i++){
  var toTest = string.slice(0,i)+string.slice(i+1);
... yeah, then eval it
but it calls console.log so...
4:52 PM
console.log = function(x){return x}
Try that.
huh. but the code uses setTimeout to call the console.log
It should still work.
oh wait
import subprocess, itertools
s='code here'
for i in itertools.permutations(s,len(s)-1):
 with open('foo.js','w') as f:
 print subprocess.check_output(['node','foo.js']) == s
which returns a number to identify i think
Mego to the rescue.
4:54 PM
@Mego thanks! will try
(not a python expert)
How about console.log = function(x){window.RESULT_HOLDER = x}, then check the value of RESULT_HOLDER.
Now it checks for proper quinage
but @Conor the for loop would reset RESULT_HOLDER multiple times
Could you put your code in a pastebin or something? I can't help you much if I don't know what your code is ^_^
4:55 PM
how about a jsfiddle?
That'll work.
this will be embarrasing if I'm totally wrong here...
ah crap hang on
I think there's "trivial" Lenguage solution for any n to that radiation hardened quine challenge (similar to jimmy's solution for radiation softened quine)
I'm not in it to win i guess
i only got n=1
4:57 PM
A: Sandbox for Proposed Challenges

muddyfishReading a timeline code-golf Inspired by Draw a timeline. Given a timeline of events, output the date for each event. A timeline will be given in the form 2000 --AF-C-------E--D---B-- Where the first line shows a known point in time. The known year is always the character below the fir...

which was surprising difficult for js
I'd be surprised if we didn't have a plain radiation hardened quine golf yet (only n=1, shortest code wins)
that'd be fun
I'd love to try golfing this
(if not, dibs on posting it)
+1 from me!
4:59 PM
An interesting problem would be "adding N characters to the source code" from some point in the source code.
should it remain a quine or still print the original source code?
no more ads on the right on google searches
I think if it should still be a quine you get the same problems as with the radiation-softened quine: only things like unary code or quine built-ins will be possible.
@MartinBüttner I don't know. Either would be interesting, I think that former would be easier.
@Conor how's it looking?
5:17 PM
A: Sandbox for Proposed Challenges

Martin BüttnerRadiation-hardened quine code-golf quine source-layout We have a challenge for radiation-hardened quines, but it's a code challenge that asks for quines that work for larger numbers of removed characters. This challenge is about the standard definition of removing a single character and is a...

@MartinBüttner I'll presume REPL's are disallowed
@Martin why no functions?
@jrich I am rather confused by it, but I think I have a solution.
(By the way, my name is in small caps, so @Conor does not ping me (; )
@jrich it just seems less clean for quines.
can anyone actually find a radiation hardened quine other than the popular ruby one?
5:21 PM
@MartinBüttner sounds cool. I think i might have just written one in js, still validating
@Cᴏɴᴏʀ O'Bʀɪᴇɴ what did you find?
I though I had something, but apparently not.
@Cᴏɴᴏʀ O'Bʀɪᴇɴ what is the problem with my code?
oh yeah, and for some reason it doesn't seem to work in node
It's rather hard to wait for a response in javascript; your code looks legit.
@Cᴏɴᴏʀ O'Bʀɪᴇɴ really!? :D
Yeah. I've tested a few random places, and they seem to work out okay.
Still working on the entire brute force though.
5:24 PM
@Cᴏɴᴏʀ O'Bʀɪᴇɴ sounds awesome
Why setTimeout, out of curiosity?
I may be missing something in your code.
because you can call it with template strings
eval doesnt work i dont think
You are right, it doesn't. But since eval is a shorter keyword, wouldn't that be shorter? As in, eval(`...`)?
If you remove a backtick the comments trick still makes it valid
calling with parens would make it fail if u removed a paren
That's smart...
5:27 PM
removing the first backtick makes it valid code and the setTimeout just evaluates to an unassigned function and does nothing
using settimeout also allows us to keep going if there are erros
A: Sandbox for Proposed Challenges

Luis MendoBlock-diagonal matrix from columns code-golf Inspired by Copied from this question of Stack Overflow. Given a matrix A, create a matrix B such that the columns of A are arranged in a block-diagonal fashion. For example, given 1 2 3 4 5 6 the output would be 1 0 0 4 0 0 0 2 0 0 5 0 0 0 3 0 ...

Well, I'd post it, it seems very legit.
@Cᴏɴᴏʀ O'Bʀɪᴇɴ will do! Thanks!
My pleasure :)
5:43 PM
@MartinBüttner I've posted what I believe is a valid radiation hardened quine in JavaScript. feel free to check it out. @CᴏɴᴏʀO'Bʀɪᴇɴ Thanks again for the help! codegolf.stackexchange.com/questions/57257/…
My pleasure!
5:54 PM
Yikes I need to fix something in pl before xnor posts Kleene star
Link me to the sandbox thingy?
@quartata If it's in the sandbox, you're already SOL
@quartata best development is panic development
5:55 PM
Shit outta luck
Oh hahaha
I don't think I am
Adding a language/feature after a challenge is sandboxed is no different than doing the same after a challenge is posted to main
Right that is if you add say a built in. This is just bug fixes and eval..
5:57 PM
It's still a new version
Still invalid
Can't use language features implemented after the question was posted
(in the sandbox, on the main site, etc.)
Isn't it only a problem if the answer has a chance at winning?
Er.. meta link? I have never heard this.
5:58 PM
There is no difference between a bug fix and adding a new feature
@CᴏɴᴏʀO'Bʀɪᴇɴ why would it be?
What if you never visit the sandbox, and fix a bug
I knew about the main policy obviously
@Adnan Can't prove it
Would that make you submission invalid?
Ahh, true
5:59 PM
There is no way to prove that you never saw the sandboxed question
That's why there's no exceptions to the policy
That makes all those TPLQ ??? submissions invalid.
@Doorknob Say I have an esoteric language X that measures anywhere between 100 and 200 characters longer than anything else. If I fix a bug in which input was handled differently whilst a challenge was in the sandbox, and it does not affect the chance of said language winning, then why would anyone have a problem? IMO, it should be okay.
@CᴏɴᴏʀO'Bʀɪᴇɴ then mark it as noncompetitive
oh right
I'm an idiot.
Sorry ^^"
6:03 PM
A: Loopholes that are forbidden by default

DennisUsing prior knowledge to circumvent other loopholes In particular, gaining an unfair advantage by inventing new languages for future challenges that are being sandboxed, discussed in chat, are your own, etc. For example, creating a language that solves a task specified in a sandboxed post with ...

@quartata The linked loophole specifically mentions ??? for TPLQ
I would argue that they are invalid on that basis, but ¯\_(ツ)_/¯
ah, good old missing-arm-shrug :P
I think we should approach it communistic: spread the bytes across all the languages, because they belong to everyone.
I wasn't around for TPLQ, and frankly I don't care enough about it to argue that submissions in a months-old, closed challenge aren't valid/competitive
6:08 PM
Man what is going on with my net
Loading posts is taking like 2 minutes
hm I thought that was just me
maybe it's something on SE's side
Maybe the site is slow because they're flipping the birds graduation switches for us
Does anyone know how I can modify the shunting yard algorithm to include the number of arguments passed to a function on the stack?
As in, shunt("max(6,7,8) => 6 7 8 (number of arguments) max or something
@Downgoat any reason you abandoned both of those sandbox posts instead of trying to address the questions in the comments? when people say they don't understand something about the spec, that doesn't imply they think the challenge is crap. most of the time it just means they don't understand something about the spec. Both of those looked like interesting challenges, provided they were actually presented a bit more clearly.
6:14 PM
@Mego That loophole specifically says "unfair" advantage. Bug fixes are not unfair. Built ins are.
@quartata If it makes your code shorter to the point that it's a competitor for winning the challenge where it previously wasn't, it's unfair
@quartata Fixing a bug in a feature you want to use for the challenge is unfair.
well anyways brb gotta go disqualify a bunch of answers
@MartinBüttner I couldn't think of a clear way to describe the spec for both challenges so I've just deleted them because I don't feel like working on them now, I might clean up the spec later
@quartata such as...?
6:17 PM
Two Jelly answers on my primitive roots challenge, an answer on Catalan Numbers. Plus I gotta check the change logs of all languages used in answers on my challenges
@jrich the scoring for radiation hardened quine got finally changed (which is probably good for you because it means its no longer possible to achieve an arbitrarily good score).
also I have to add a note to grocery store disqualifying all languages made since October
@MartinBüttner Is Retina on GitHub?
That's why I like to add the line "You may use a language newer than this challenge or a version made after this challenge on this challenge so long as it doesn't solve the challenge itself."
That is a bad idea
You can still make a language tailored towards a specific challenge without solving the entire thing
Using a newer language/feature doesn't make the answer invalid, it just makes it non-competitive
6:23 PM
@CᴏɴᴏʀO'Bʀɪᴇɴ That RX answer on atbash is invalid
Hmm Jelly's commit history is a little tangled
@CᴏɴᴏʀO'Bʀɪᴇɴ The accepted answer on Atbash Palindromes.
Oh wait that's my question
6:25 PM
@quartata it would help if you actually posted links to the things that you're talking about
@MartinBüttner Ah thanks, couldn't find it.
@Doorknob On mobile, copy pasting isn't working with chatSEy for whatever reason
I'll take a closer look when I get back
@Doorknob Are you using the graduation script, by any chance?
@Mego the what?
Hmm nvm even with disabling Tampermonkey, it's still slow
6:42 PM
@MartinBüttner I noticed. Does that make your sandboxed post too much of a dupe?
I don't think so... I'll give it some thought though
Because now technically I'll try to golf my answer to get a better score... and then when you post copy-paste
(tbh i think your question is better/more reasonable, but just putting it out there)
@Doorknob I still think this is an extreme interpretation of that loophole. It specifically cites as an example making a zero or on character built in that solves the challenge, and does not specify "grey area" things like fixing a bug with arrays. Also, users who cannot see deleted posts will have no idea when something was sandboxed and so have no idea whether they can use a language feature or not.
i think we need a meta post on this; the loophole is somewhat vague.
@quartata It's interpreted strictly because there is too much room for abuse otherwise. While your intentions may be good (you just want your language working properly), intentions are impossible to prove. The cost of closing the loophole (losing some otherwise-valid answers because a language feature had a bug that was fixed after the challenge was posted/sandboxed) is less than the cost of leaving it open (tons of crappy answers that exploit features added after the question was sandboxed).
Because a language is defined by the implementation and not the specification, there is fundamentally no such thing as a bug.
6:51 PM
MSVC's comments are really informative.
Q: The middle-square method

AdnanIntroduction The middle-square method is used for the generation of pseudorandom numbers. However, this is not a good method in practice, since its period is usually very short and has some severe weaknesses. How this works? Let's take an example: For the seed, we pick 123456: Seed 123456 ...


Related conversation

Jan 6 at 5:32, 8 minutes total – 16 messages, 2 users, 0 stars

Bookmarked 7 secs ago by Mego

@Adnan Dood you're a challenge writing machine
@Doorknob That's not what I had in mind when I wrote the loophole post, but I guess I didn't express my thoughts as clearly as I thought I did. I think we should have a separate meta question to sort what what can and what can't be done once a challenge idea has been sandboxed. Other forms of prior knowledge will still be problematic though.
@Mego Hahaha, thanks :)
7:01 PM
@Dennis Can mods see deleted comments?
My stance is: language versions past the earliest description of an upcoming challenge (main post, sandbox post, comment, or chat message) should be ruled as non-competitive due to prior knowledge. For deleted commands and chat messages, the mods would have to be the arbiters, since us plebians non-mods can't see them. Of course, this requires an unambiguous definition of a "description of a challenge", which will be problematic.
@Mego Should that be comments rather than commands?
"I didn't see it" isn't a valid excuse. It's impossible to prove, and given how quickly information spreads on the internet, once something is typed and posted anywhere, it's not unreasonable to assume that a large fraction of the planet's population has seen it within a short amount of time.
7:08 PM
@trichoplax Yeah it should, I no type so good
That may be clear cut, but it would involve a lot of work to verify it. It also doesn't cover the most difficult-to-verify form of prior knowledge, namely self-answers of off-SE communication.
Also, I think it's entirely unreasonable to make a language feature (or entire language) non-competitive just because an idea was brought up in chat or the sandbox a year ago.
@Dennis It also seems somewhat unreasonable to me, but I'd rather have a clear-cut policy. If a better alternative is suggested, I'd support it.
@Dennis I'm sure the community can apply common sense to those cases
@Dennis Off-SE communication is impossible to verify/detect/inspect, so that wouldn't be a valid reason for non-competitiveness. If a person is willing to go to those lengths to cheat in a PPCG challenge, they probably won't last long enough here to make much of an impact.
@Maltysen Have you been to meta? Common sense is tossed out in favor of pedantry.
7:13 PM
I'd give you a wonderful example but SE seems to be down
Q: How-to [HTML Rain Matrix]

Roblo SoulHello and if you're reading this you either want to know How-to make a Binary Code Raining Program or want to know how to convert it into an HTML/Binary Code Raining Program. So let's get started just [Copy And Paste The code in the #'s] # <html xmlns="http://www.w3.org/1999/xhtml"> <head> <...

See the comments on this answer (and the extended discussion in chat) for a great example of how common sense isn't a viable method of determining validity here
SE why are you dying :(
Is anyone else having problems? I am not.
7:20 PM
I got a 522 from cloudflare just a minute ago
Seems to be somewhat better now
It must be the hamsters' lunch break
@El'endiaStarman The electron thing finished computing, but the result was destroyed :(
@CᴏɴᴏʀO'Bʀɪᴇɴ awww :/
Q: The knight's next tour

wizzwizz4We've all heard of the Knight's Tour puzzle: find a route for a knight that passes through all the squares on a chess board. But let's be honest, it's a little bit boring. So, let's give the knight a bit of a challenge. Task Write a program that takes the knight through all the squares on an ar...

7:57 PM
@Mego and @Doorknob, honestly, I think that being insistent on disqualifying languages that had bug-fixes after a challenge was posted in the Sandbox (especially) or on Main is/will be counter-productive and possibly threaten the fun of this site. I've never really thought of this site as a competitive one. Anarchy golf would be better for that.
If you try to close every single little loophole, you run the risk of being rules-lawyered. I don't remember where exactly I read it, but there was a popular forum where the only rule of conduct was "Be nice." (I do remember that this inspired Stack Exchange's own code of conduct.) The administrator of that forum simply banned everyone who violated that rule, and the community was well off for it.
People who try to find ways around such rules are deliberately misunderstanding them, which proves that they do understand said rules. In particular, regarding bug-fixes to languages after challenges are posted, I see the reasoning in that we want to disallow new features that are an unfair advantage.
I think we should just say it that way, and downvote people who quite apparently misunderstand that rule (and leave a comment expressing such an opinion). That person will eventually learn to stop due to all the negative feedback. Harder to enforce? Absolutely. But I think it'd be better this way.

« first day (1849 days earlier)      last day (1879 days later) »