« first day (1370 days earlier)      last day (3456 days later) » 

12:19 AM
0
A: Sandbox for Proposed Challenges

millinonCompile source code into English This is an idea I had that would make for an interesting personal project, and I figured that it couldn't hurt to see what could be done with it as a contest. Any feedback would be great! A compiler is typically a program that converts a program written in o...

 
 
2 hours later…
2:41 AM
@COTO Regarding programming being required, I draw a higher line that "programming is the challenge" rather than it just being a medium. I might as well link my existing meta answer: meta.codegolf.stackexchange.com/a/1962/20260
I guess my question is: what do you envision an interesting answer to your challenge being?
 
@xnor There used to be a "POVRay" contest online where programmers were invited to create the most stunning 3D scenes they could (in POVRay) in 512 characters or less. The results were asounding and the code was beautiful. I couldn't have fathomed some of the things that came out of that contest.
 
so i agree that if people made interesting comics with algorithms, that would be a good use of coding
people seemed to do that with the mathematical art question (though i'm not sure i approve of it still)
 
In this case, I'm hoping to see the creative ways people come up with rendering people, textures, borders, text blurbs, etc., programmatically.
 
i'm not getting that from your scoring
the crucial part being "programmatically"
i understand the char restriction forces efficiency
 
Can you draw just a stick figure, or can you draw Calvin hugging Hobbes in front of an oak tree, with texture, and having both recognizable--in a tiny fraction of the code that would reasonably be necessary to render such a thing unencumbered.
 
2:48 AM
but could easily imagine a stick figure with a speech-circle telling a funny joke (forced in with lots of text compression) getting loads of upvotes and winning
 
There might be some of that. I suppose I can take out "funny" as one of the recommended reasons for upvoting.
 
@Calvin'sHobbies Did you have a psychic sense that calvin and hobbes was mentioned :-P
@COTO That would help, I think
 
Frankly, even coming up with a decent stick figure comic in 512 characters is bound to be a challenge.
 
actually, the whole comic thing
look at XKCD or dinosaur comics
lots of people find them hilarious though the art is simple
i think what you want is clever rendering
 
Apparently...
 
2:50 AM
and asking for clever rendering directly is more likely to produce what you want
 
Well I'll reword with an emphasis on artistry then.
Give me a few minutes.
 
i know you have a subjective judgment or what's good, but that won't stop the legions of HNQ from upvoting simply on the work "comic" in the title
 
There's zero upvotes so far. And I proposed the challenge precisely because the last few challenges I've posted have had tougher-than-usual criteria and tanked. People here clearly, clearly want simple, open challenges where they can program creatively and/or non-competitively if they want to. Where the winning criterion is open.
 
@COTO, don't know about clearly, it's a perennial source of discussion and close votes
at least there are definitely answerers that enjoy answering them
 
I was aware of the art meta discussion before, but I did a good faith search for existing challenges, determined that mine was within reason and suitably novel. I also thought to myself "If not PPCG, where on Earth could a competition like this take place?", which is a relevant question.
If you posted this on an art board somewhere, you'd get laughed out the door in a matter of seconds.
 
2:59 AM
unfortunately, i think that's true for pretty much every closed question
except the ones that are clearly programming questions for stack overflow
the community doesn't think they belong on PPCG, but there's no similar place for any similar kind of question
 
So why all the angst against programming questions that involve art? I get the "posting pictures of a cat" (i.e. completely unrelated to programming) angle in the meta discussion, but this attitude that "programming may not be used to produce art" bewilders me.
 
githubphagocyte has a good response to the meta question pointing out that the amount of art should not be a reason to close, but rather the amount of programming
and i agree
but i still have concerns that answers to such questions are basically judged on the art and creativity of the output
and not on the code that produced them
 
Well, as I say, the fact that nobody who isn't an expert programmer can participate in this challenge is sufficient for me to conclude that it involves programming.
And since the quality of the art is entirely predicated on the entrants' ability to squeeze out graphics from a tiny bit of code, it seems like a reasonable challenge.
At any rate, we shall see what the community thinks.
If it's closed, I'll accept that. I definitely don't agree with it, but I'll accept it.
 
i'm generally not a fan of questions where coding is simply the medium to produce results, like KOTHs where the code just gives your move in a prisoner's dilemma type game
but such KOTHs are popular and seem to be accepted by the community
so perhaps i'm in a minority
i'm not voting to close the question, but if i had to guess, i'd put the higher odds on it being closed
please do post more on the meta threads on these topics
 
My personal criterion would be: could somebody who isn't a competent programmer take part in this competition?
And thanks for the heads-up.
I won't start any new meta topics on the discussion.
 
3:06 AM
there's existing ones that i think are in want of more discussion
 
BTW, I've reworded with an emphasis on artistry, including some specifics.
 
i think those changes help
why comic though?
why not just art?
i think "comic" has a strong pull towards humor and minimality that is liable to make upvotes not for art
 
Because "art" to me implies something far more broad, and I'm genuinely more interested in comic-book style art. Maybe I could change it to "graphic novel", if you think that would improve the quality of the answers.
 
there are of course, artistically rich comics, but that's not what i think of when i see the word
i like graphic novel but i think it implies something too long
 
How about including "Artistically Rich" in the title?
 
3:09 AM
++
half of HNQ will only read the tile :-)
 
What is HNQ?
 
hot network questions
 
Ah. So "the hoi polloi".
 
it's a blessing and a curse
when questions from PPCG appear there, they get lots more viewers
and answers and upvotes
but often ones not in the spirit of our site
exactly
for better or worse, popularity contests tend to attract them
 
Well, if we get a spate of neat 512-byte artwork, I'd consider that a boon.
 
3:11 AM
and their subjectivity is often quite different from site regulars
 
I'll add "artistically rich" to the title.
brb
 
unrelatedly, sorry your Java satire got deleted
it looks like that took a crazy amount of work to make
and i thought it was witty and hilarious
 
Thanks. It did take a while. And it took 270 rep with it when it went. But Doorknob explained why satire is taboo and I got over it.
I've added "artistically rich" to the problem title, and "This is a competition about artistically rich comic book (or graphic novel) style artwork. Humour is a nice touch, but should not be used as a principle criterion for judging entries." to the scoring description.
 
great, i think that gets to the heart of it
 
Thanks for the input.
 
3:17 AM
good luck!
 
Late here. I'm off for the night.
 
and good night
 
Thanks. And cheers. :)
 
 
2 hours later…
4:54 AM
0
A: Sandbox for Proposed Challenges

Super ChafouinMondrian painting description langage This challenge consists in coding an interpreter for a Mondrian painting description language (MPDL). Language definition The language operates on a stack of rectangles. A rectangle is defined by its upper left coordinate, lower right coordinate and color....

 
 
1 hour later…
6:12 AM
Just found this, I can't stop laughing, thought I might share it:
 
 
2 hours later…
8:26 AM
@PeterTaylor The accepted default for functions actually seems to be that they can be unnamed, so I'd apply the same to verbs and blocks.
 
I do actually say named function too, because I think it's slightly fairer on the languages which don't have anonymous functions.
 
yeah sure, that's fine, I'm just saying for the default, I'd probably go with unnamed.
I've edited the post to include "function-like constructs"... hope that's not too vague
@Geobits Regarding the scoring of the regex proposal... I figured I could just scale the denominator with the number of the answer. Then I just have to figure out a reasonable scaling... but I suppose I'm not too bad at regex so I just could play the game myself for a while and how fast answers grow.
 
8:55 AM
(unless people add in extraneous characters to make subsequent matches harder)
 
9:14 AM
maybe I should do "match all odd-numbered submissions, fail on all even-numbered submissions"?
 
9:38 AM
by "odd" and "even", do you mean ones of the same parity sort of thing? otherwise that will have to end after 2 iterations, as the second answer will have to match the first, and the third will have to match the first but not the second?
 
 
1 hour later…
10:54 AM
0
A: Sandbox for Proposed Challenges

WillDraw the Tree of Life A codegolf contest to render the tree of life data (which is in the simple Newick format) sideways in ASCII, using the same formatting as this example: __/a __/ \b \ _/c \_/ \d \e Note that the distance in the source data is ignored; the tree should be dr...

 
11:50 AM
@VisualMelon no that's exactly what I mean... every answer has to match 1,3,5,7,... but not 2,4,6,8,...
I gave it a try earlier and it's a much harder problem
 
12:07 PM
@MartinBüttner oh, sorry, I assumed it meant matching the same language as the given answer, rather than matching the expression
 
What's the 'Shells' room that COTO made doing?
 
Ah maybe. What does Core Code mean then?
 
there's a definition in the challenge I think, but I didn't read it
 
12:24 PM
Oh yeah.
 
12:50 PM
Is it just me or does Python's grammar specs look a little off? docs.python.org/3/reference/grammar.html
Seems to me like doing

while 1:
pass

requires 2 newlines after the pass :/
 
grc
1:22 PM
@Sp3000 As I understand it, you only need 2 newlines in the interpreter. The interpreter doesn't know if you're going to continue the indentation after the first newline, so a second is necessary.
 
I see...
Might have to shuffle the newlines a bit to get the "usual" behaviour then...
 
@grc do you mean the REPL?
 
Is it possible to have a flow chart that is n choose 3?
actually, I know its not, nevermind
 
I don't even know how a flow chart can be a number.
2
or what n is
 
1:38 PM
Well, I'm trying to develop a system where people can design board or card games
So, one of the things I need to do is a turn-flow
the actions they must (or can) do
A common option in games is to pick 2 cards out of my hand (or 3 cards)
and I'm struggling to get that into my flowchart of actions
 
2:11 PM
Some feedback in the art meta discussion is requested, as the site really, really needs it.
Sorry to post and run but work calls. Somebody gotta make dat moh-nay.
 
 
1 hour later…
3:39 PM
@Geobits haha, I love the top close voters list :D
 
Top close voters list? Does that exist?
 
data.SE
Stats: 19 reopened in the last 60 days, Top reopeners, Top closers. Your claim that "nobody reopens" doesn't hold up, IMO. — Geobits 23 mins ago
 
ah
Lol, I'm second on the Top closers list.
 
yeah, but Peter's margin is absolutely mental :D
(not very surprising though ^^)
 
CJY's margin is more surprising :D
 
3:50 PM
especially since he hasn't voted since 2012
 
It's good to keep in mind that that only counts votes that ended up in an actual close/reopen, though. If the votes expired, they're not counted there. So most people's numbers are probably a bit higher in both lists.
 
19 reopened in 60 days is higher than I imagined, though. We only get, what, 4-5 questions a day? that's 300 max for the time period, so 19 is a decent chunk of that.
 
@MartinBüttner I don't know whether that number is actually correct.
For me, it says that I last voted on "07/23/2014", which is incorrect.
 
Disclaimer: I didn't write those queries, just searched them. Look elsewhere for blame ^^
 
3:54 PM
@Geobits the 4-5 don't include all the ones that get deleted really quickly though. I'm not even sure if closed ones are counted
@ProgramFOX As Geobits said, it could be the last of your votes that went through ^^
 
Yea, I'm not sure exactly what is counted in the "questions per day" stat on Area 51.
 
@MartinBüttner I have some from this month who went through, and those are also old enough to be included in the Data Explorer.
 
Either way, a question is reopened roughly once every two days, which isn't bad for our volume IMO.
err.... every three days. WTF.
 
It's always fun when you're trying to do some golfing and you end up learning how to use git to submit pull requests.
Also hi, I haven't been around in a while.
 
I was thinking the same. Hi!
 
4:00 PM
What's up?
(also the pull request i sent included one commit that added exactly one character lmao)
 
Same old stuff here. You?
 
not a whole lot
 
@undergroundmonorail Hi!
 
@ProgramFOX yooo
 
@Geobits any opinions on the alternative version of the regex challenge? (match on odd, fail on even)
my main concern is that it's a bit harder to keep track of a test sets
but that is probably similar to the 95 quotes challenge
the other thing is that a decent scaling for the scores is even harder to predict
 
4:22 PM
After submitting a pull request, is it normal to get butterflies in your stomach and feel the urge to refresh github a million times while softly mumbling "notice me, senpai"? Or is that just me
 
no I only get that when posting questions on PPCG and waiting for notifications :P
 
4:39 PM
is this a room where I can ask general question about code golf?
 
you can try ;)
 
I was looking for solution for #108 on this page codegolf.stackexchange.com/q/40376/32451 and when I found it it, and wanted to post, it say's I need 10 reputation cuz it's protected. BUt I have 100 rep, so I dont get it :/
 
@Dexa association bonus doesn't count
sorry about that
btw, you'll find more people busy with that challenge in this room
 
meeh, never mind two answers came in the meanwhile :/
oh, ty
 
@Martin Not sure. It definitely seems harder to keep track of, like you said. But your example outputs do seem to go a long time without growing much, so it might be better if you want it to get really complicated within 200 posts :P
 
4:53 PM
I'm fairly sure the even/odd version gets interesting within 20 posts
 
It would make it easier (for answerers) to track if you kept an odd/even listing in the body, but that would mean keeping on top of it really well.
 
I wanted to do that anyway, but I'll probably go with Calvin's approach of having people include the test set in each answer
(writing "Calvin's" is so odd, as that's not his name :D)
hmmmm, I'm not sure what to think about this
1
Q: Sabotaging the coding standards

vszThere are various coding standards enforced at software companies which have the goal of increasing code reliability, portability, and, most importantly, readability in code written jointly by different developers. Two notable examples are the MISRA C, and the C++ standard developed for the JSF ...

the actual code that's written only goes in spoiler tags, but it undoubtedly requires some intricate programming knowledge
 
I'm not sure that you even need code there, since it just asks for an explanation.
 
well you probably need to provide an example
 
I was just going to ask about this. Both the help center and the pop-contest tag wiki don't seem to consider challenges that don't require to write any code.
 
5:06 PM
good answers will probably use syntax quirks
 
Hmm. I think this is one of those questions that depends mostly on the quality of the answers.
 
6:07 PM
@undergroundmonorail I've submitted a pull request that changed one character. OTOH by correcting it I caused the XML which the program generated to actually match the schema, so it was accepted quite quickly.
@MartinBüttner I think it's a bureaucratese challenge. It's about hiding the true meaning of an English sentence. Better suited for lawyers than programmers.
 
@PeterTaylor Hm yeah, that's one approach. But I think interesting answers will abuse the coding standard by lawyering their sentences, but by introducing some innocent looking formatting standard which actually breaks the syntax in some cases.
actually no, that's not possible in the given challenge
the rules should be used to break working code which seems to follow them but doesn't
that would require some technical or linguistic ambiguity in all cases, I suppose
two reopen votes on this... what on earth...
 
@MartinBüttner I was flabbergasted when I saw COTO complaining about reopen votes not being cast, because I regularly see reopen votes cast on questions where there are unanswered clarification requests in the comments.
5
 
yes, that's a correct edit
That's how Stack Exchange has to be spelled.
 
I like the thought of SE mods running a script that searches for "StackExchange" so they can edit it :P
 
yeah that's fine... but how does a mod who doesn't seem to be very active here notice that within a few minutes :D
 
"StackOverflow" in the image description is not changed...
Fixed.
 
btw, the SE logo totally reads "StackExchange"
 
6:31 PM
Yea, but that's a logo/trademark issue. When spelled out in prose, it should have a space. It's been brought up on meta before.
 
yeah, I get that... just saying... the logo's a bit misleading if it's that important for people to spell it with a space :D
 
How is Program FOX able to submit 1 character edits? And yes, I intend to spell it Program FOX (with a space) forever.
 
You failed twice at typing my name :P
(no space)
But anyway, you can submit these edits if you have the edit privilege.
 
Next thing you know, they'll set up a bot to take the space out of "Program Fox" :D
 
Lol :D
 
6:38 PM
@MartinBüttner Found it.
 
thanks
 
It's been a long time since I did a pacman -Syu and I can't tell if it's hanging on (re)building database or if it's taking that long because there are so many changes to make...
Oh, nevermind, it started going while I typed that.
I'm a bad arch user. The only reason I'm even updating the system is because Desmume kept crashing and I thought "maybe if I update everything the problem will fix itself so I don't have to diagnose anything".
 
Wait, that's a bad reason to run an update? :P
 
I think, in theory, I'm supposed to update because "i haven't done that in like 2 days", not because I'm too lazy to track down a bug :P
Especially if it's been months since the last one, haha
 
But updates fix bugs, right? crosses fingers
 
6:48 PM
by the way, update: it did not solve the problem. currently looking to see if there are other ds emulators for arch
Doesn't look like there is. Last resort: I'm about to pull a Windows.
brb rebooting
 
I've heard good things about No$GBA under Wine, but I haven't tried it.
 
We've had more bugs created in the last week than the four weeks before it put together. I blame ReSharper.
People are literally checking in changesets with the comment "ReSharpened blah.cs"
 
But think of how sharp the code will be once they're all done :P
 
I cut myself on a 1 once. The 0s are pretty harmless though.
 
My real problem with it is that I am making localized changes that (hopefully) affect only the area I touched, while they are breaking everything.
 
6:54 PM
Oh, and rebooting didn't fix it, but running desmume through sudo did. Uh, that's... probably not good.
 
Also, shameless plug for my own SO question:
0
Q: How do I get the error code of netsh?

RainboltI capture the output and the exit code of a process like this: Process proc = new Process { StartInfo = { RedirectStandardOutput = true, UseShellExecute = false, FileName = Environment.SystemDirectory + "\\netsh.exe", Arguments = "http delete urlacl url=ht...

Help me earn a living please.
 
7:13 PM
@ProgramFOX If I edited titles like that, I would remove the trailing "?". "Getting error code of netsh?" is grammatically incorrect.
 
@Rainbolt Is it? Well, I thought I saw an example like that on the good-title page, let me take a look.
 
There seem to be a few, like "Turning Linux USB power on and off?"
 
@Rainbolt Yes, I saw these so I thought it was good to made it end with a question mark.
 
I think you could only really use a question mark there if it was an "echo" sentence. Like "How do I X?" "X? Oh, that's easy."
 
But if it is not grammatically correct, I'll keep that in mind for the future.
 
7:16 PM
As far as grammars concerned anyway. But this is the internet so all bet's are of ^^
 
@ProgramFOX Well I take back what I said. Most of the good examples are also grammatically incorrect. So carry on the way you were lol
 
Okay
 
I just got the announcer badge for sharing a link that was visited by 25 unique IP addresses for "Survival Game - Create Your Wolf". But it doesn't tell me where the link is...
 
@Rainbolt The link is just the link you get when you click "share" on your question.
 
@ProgramFOX I don't know where I shared it though.
 
7:26 PM
@Rainbolt Do you still not have 100 favourites on that? Wow.
 
97 haha. I am soooooo close.
 
@Rainbolt oh, no idea then
 
@Rainbolt It doesn't distinguish between referrers, afaik... so if you've shared it in multiple places, the 25 is the sum for all of those links
 
@MartinBüttner Oh, I see. The link is the same every time I generate it. I didn't notice that before.
I thought it spat out a fresh one each time
 
@Rainbolt it's just your user ID that's appended
holy cow, a Nice Answer badge for a sandbox proposal in 2 days...
 
7:36 PM
@MartinBüttner Linky?
 
the regex thing
one sec
 
@MartinBüttner I'm a little confused by your challenge. Is every submission both a string and a regex?
 
for its own purpose it's a regex. for the all subsequent submissions it's a string.
 
So abc is a regex for its own purpose.
Then the second submission is a, and abc is a string?
because "abc is a string for all subsequent submissions"
So the regex "a" matches the string "abc"?
 
7:42 PM
I'm more confused than I was before.
Oh wow, I guess I always had the verbs backwards
 
A little Googling told me that Regexes match a String. I always thought a String matched a Regex.
 
oh
thinking about it, that is odd
 
I thought a regex was a machine that you fed strings to
And the regex came back and said "yes" or "no"
 
that's what it does
but I can understand your perspective insofar that "an instance matches a pattern" and not the other way round
 
7:45 PM
@PeterTaylor Some specific challenges: Chess Compression, Chess Tournament. I think your response covers the first well, and data compression is definitely useful in programming (but note that there is no code).
The second is basically "find a simple chess strategy and write code for it", which I'd say doesn't take much programming skill at all.
 
but somehow regexes match strings
 
@MartinBüttner In that case, my only addition is the part in bold: "you are to answer with a regular expression which matches the submission right before yours but none of the others before it."
Because it can match the other submissions, as long as they came after you.
 
right, that makes sense
but I'll probably change that part of the challenge anyway
I tried it this morning and it's way too easy
I'm currently considering "match all odd-numbered answers before you, fail on all even-numbered answers before you"
 
How about this: "You have to match all of the submissions after yours."
Title it "Psychic Regex-Fu"
 
Match any five previous answers (and no others). You'd have to give five+ starters, but...
 
7:49 PM
@Rainbolt [\s\S]
that's badly golfed
\s|\S
or [^] in ECMAScript
 
@MartinBüttner I was working on it for a bit but then school got in the way...
So, basically, no, not ready at all.
 
@Geobits That one is interesting.
 
I don't regex enough to know exactly how hard it would be, but it seemed like a good idea to me. You may have to play the game to see if it holds up.
 
I think it depends a lot on the initial 5 strings.
I need to continue the even/odd game first though... I only got to pattern 9 this morning :D
 
That may be an idea for another challenge, though. You're entering a tournament (game to be decided), but the mob has paid you to take 5th place. Entries must play the game with current answers and take 5th exactly to be valid.
 
7:58 PM
heh, we could just make that a KotH
"The winner will be the 5th-best bot."
sounds cool
 
@Geobits I love the simplicity of that idea. I don't even think the game needs to be all that complicated.
 
okay, the even/odd is stupid if I don't also alternate the matching and failing sets
 
You could literally just output "Win" or "Lose" and have a rule for resolving ties.
 
Yea, any old game would probably work.
Just gotta find something that hasn't been done.
 
You would end up with "Win a bunch, then try to lose just enough." bots, and "Try to stay around fifth the entire time" bots. I wonder which strategy is better.
Also, exactly fifth place becomes exceedingly impossible as the number of entries grows. How about "X% away from fifth place"?
 
8:04 PM
Mobsters don't know math, they just know what bet they made. If it ain't fifth, they kill you :P
2
Seriously, though, to have a single objective winner, X place makes the most sense to me.
If it's KOTH.
 
So buck the system and let a percentage of the submissions win.
 
I have a feeling it would get closed if I did ;)
 
so just to follow that up with the regex challenge... the problem with even/odd is this: the regex will usually be either inclusive or exclusive... so either it will match everything except a small superset of the even answers or it will match nothing except a small superset of the odd answers. in either case, every second submission will be added to the set in which it will correctly match/fail on itself. therefore, the follow-up answer can always just change or add a character
(to ensure uniqueness)
 
If someone closed my challenge that intentionally allowed multiple winners I would seriously blow a gasket.
 
8:10 PM
Geobits tricked me. He already knew the answer :-/
 
Who, me?
I still need to work on that thing... stupid life.
 
@Geobits I suppose starting with either 6 or 10 would make sense?
or, I could start with 5 strings, and provide the first submission, which brings the count up to 6
 
Any number 5+ should work I guess, but 10 would force at least some effort from the first answerer.
 
how about these 5: lower-case alphabet, upper-case alphabet, digits, common regex meta-characters, other symbols
and then my first submission, which could be ., I guess
 
Seems reasonable.
 
8:18 PM
Not to be outdone by Calvin Hobbies, I think it's time for a Domino KotH challenge!
 
You mean a KotH to play dominoes? :P
 
Omg this is brilliant. If I make each submission both a submission and a compiled version of every submission before it, then I never have to compile anything again!
Also, each submission must mail me a nickel for each submission before it.
 
@Rainbolt that only scales with the square of the numbers :(
 
Someone needs to create a graphical output domino challenge
 
You mean like mail-mail? That means I'd have to buy a stamp. I haven't done that in at least a year or two...
 
8:23 PM
someone needs to figure out how we can reasonably conduct domino challenges with the SE format
 
Where each submission outputs a gif that follows from the one before it, and the end result is some sort of PPCG made movie.
@Geobits I went to the post office the last week and paid $4.50 to mail a letter because I didn't have a f***ing clue what I was doing and the office folks were obviously desensitized to morons like me.
I thought "Surely the post office has white envelops and stamps right?"
 
Sounds about right. I actually have no idea how much a stamp costs now. The last time I needed one, I looked it up online, found out that it had changed (twice) since I'd bought my book of stamps, so ended up just using two.
 
@Geobits Lol, it doesn't
 
@MartinBüttner Introduce a minimum Levenshtein distance requirement?
 
8:41 PM
@Rainbolt hm that seems like a very arbitrary restriction which just complicates things even more. this can probably be fixed with 5 shorter initial strings
trying these 5 now: a, AB, 012, .[\], !"#%&
that is a lot more interesting... at least I'm going back to previous answer, realising I could have shortened them, because the optimal answer isn't all the obvious... and I've got hopes that the patterns will start to grow soon
it's amazing how this challenge idea, which sounds really decent at first (at least if the votes are any indication), is so completely infeasible in the first 3 versions I wanted to post it :D
 
@MartinBüttner This is code golf right?
 
well, kind of... shorter is better, yes, but the score is more complicated than just regex size
 
9:06 PM
@Geobits you keep saying you're not too familiar with regex. would you mind playing this variant for a bit and tell me how it goes, so I can gauge the spectrum a bit? I'm currently trying to play optimal, but it's unlikely that the actual answers will remain near that optimum if it's easy to slip (which I can't really judge)
 
9:20 PM
Can't right now, but I'll take a look at in a while. Familial duty and all that :D
 
sure, no rush
Actually, don't worry... the new starting set doesn't solve the problem, and no set can
At some point there will be a situation where there are 5 strings of some length n>4 (say 9)... when you get there, there are a dozen ways to match 9 arbitrary characters while using a pattern of less than 9 characters.
last chance: matching all but 5 strings
 
@COTO Continuing the discussion from your meta thread
You say that "In other words, the observable consensus is that an art programming challenge should be allowed so long as programming is a principle and integral component of the challenge." but I think that's exactly what is under dispute
(i mean that programming is principal and integral to your question, not that that is the consensus, I don't know if it's the consensus)
 
9:48 PM
overall, I think you're claiming consensus supports you much more than it actually does
your question had 3 upvotes and 4 downvotes when it was closed, which doesn't support the idea of a cabal of superusers going against the will of the masses
i do encourage you to extract a more tightly scoped challenge and post it to sandbox
maybe something like imitating the style of a particular comic that has a very distinctive art style
or simply drawing a hard-to-make texture like wood or snow or sand
i also think that banning graphics libraries that give you more than basic "draw a shape" functionality is necessary to stop it from being a challenge of "who can import the best library to draw the cool art for you" and make people write the needed code themselves
 
10:36 PM
0
A: Sandbox for Proposed Challenges

chilemagicThis is an idea for an incremental answer-dependent question. I'm interested in comments. I will keep it here for a while and probably reword it at some point. Print the name of the language of the previous answer restricted-sourcecode-golf The first answer to this question must print the na...

 
11:13 PM
0
A: Sandbox for Proposed Challenges

Beta DecayKeep me cool! code-golf Challenge This may be another implementation of the Travelling Salesman problem, so please notify me if it's a dupe. Walking around Marks and Spencers, I noticed that they had air conditioning units placed randomly around the shop. Wanting to keep cool, I wondered what...

 
Hey, for this, do you think would work or is bog-standard best?
 
For those of you who didn't already know that Parenting was a joke:
7
Q: Should a 5 year old excuse themselves after farting?

trejderMy five years old daughter sometime farts silently, and she doesn't excuse herself. Every time I detect such a situation, I remind her that she should try to avoid it and leave the room, going to a toilet, but if she fails to do that, she should excuse herself. My wife says this is wrong. She cl...

 
@Rainbolt I saw that earlier :) Here's another gem.
 
Why is Sparr the first commenter on both of them?
 
@BetaDecay Maybe I don't really understand the monkey challenge, but wouldn't the optimal probability for each key basically be the inverse of its frequency?
I think Sparr's the first commenter on them because he just joined today and felt like getting some activity in ^^
 
11:26 PM
@Geobits Oh yeah... It's just frequency / total number of letters... I should have thought about that...
 
I'm not saying that makes it bad, exactly, but if it turns out that's optimal you might want to make a careful check for duplicates in the character frequency area.
 
@COTO I think the "too little programming" vs "too little art" point is well taken and well-agreed on as a point in githubphagocyte's much-upvoted answer
 
@Geobits Yeah... So would it work as a Rosetta code?
 
@COTO but i don't feel like many art challenges also have too little coding, not just too much art
where coding doesn't just mean that writing code is required to participate, but that it is the central challenge and aspect on which submissions are judged
 
@Geobits "all but 5" doesn't work either... it just delays the problem until you can do ^(?!.{11})$ and similar... gahhh...
 
11:35 PM
@Beta If it works at all, I don't see why not.
@Martin That sucks, lol.
 
of course that would work with a constant denominator of 10, but then your issue with the scoring applies
 
@COTO oops, too much negation: i feel like many art challenges also have too little coding, not just too much art
 
What if instead of bare regex, each string was a (short) program to generate a regex matching (some subset of) previous entries?
 
what's the input for the generator program?
I'm thinking a mix of my first idea and yours could work: match the previous one + 4 others.
doesn't exactly make things simpler though (in terms of the spec)
 
What's harder (spec-wise) about that?
 
11:45 PM
well, I meant... it seems a bit more annoying to keep track than either "the last one" or "5 arbitrary ones"... probably not a big deal though
 
What if each entry chose which previous ones the next entry has to match? (no two chosen sets of 5 the same)
 
lol
I was just racing back to the laptop to post a similar idea :D
 
Nice :D
 
what if each entry chose whether it went into the match or fail set?
 
That sounds nice, you could sabotage easy pickings I guess.
 

« first day (1370 days earlier)      last day (3456 days later) »