« first day (1734 days earlier)      last day (3102 days later) » 

Anonymous
5:01 AM
Yeah the bots are kinda slow
 
@El'endiaStarman Voting to reopen. I didn't read all the arguments but I don't like when challenges (that are clearly not terrible/spam) are closed by a single user.
 
Anonymous
And by kinda I mean I've seen faster-moving geriatric patients
 
@Calvin'sHobbies [nods] I have a tendency to try and make the strongest case I can from the outset. Hence the long post...
 
Anonymous
@El'endiaStarman I agree with your arguments, and cast the 5th reopen vote.
 
Anonymous
If you compare the number of bytes with other languages, you will see why no one is using C++. — CroCo 4 mins ago
 
5:03 AM
But anyway, it would probably be a good idea to actually settle the question of whether to close good/better/newer duplicates once and for all (at least until consensus changes).
 
Anonymous
People like this annoy the hell out of me
 
Anonymous
@El'endiaStarman We had a decent consensus that we shouldn't last time it was brought up, iirc
 
I appreciate your reopen votes, @Calvin and @Mego, by the way. Thanks. :)
@Mego I think I remember seeing that in chat, yes.
 
Anonymous
@El'endiaStarman No problem man. I like the challenge, and will get cracking on it
 
@Mego Everyone is entitled to an opinion. ¯\_(ツ)_/¯
 
5:04 AM
On Meta, however, it's not particularly settled.
 
Anonymous
Probably in the morning though
 
Anonymous
11
Q: Closing old question as duplicate of a new one

Martin BüttnerWhat is our policy regarding closing an old challenge as a duplicate of a new one? I know this happens on other sites (at least SO), if the newer question is generally better or usually if it gathers better answers (such that the old one acts as a "redirect" to the canonical answer). Now this do...

 
I think if we generally allow posting new challenges that are duplicates of old challenge but are "better specified" or whatever, we're just going to end up repeating ourselves every year or so. We will no longer have interesting content.
 
Anonymous
Seems pretty settled to me
 
When I miss that a challenge can be solved by simply calculating a GCD, it's time to go to bed...
 
5:07 AM
haha
 
Anonymous
To me, there's a simple set of criteria:
 
@Mego Not really. 2 vote difference.
 
Anonymous
@AlexA. 2 vote difference, with no discussion in over a year
 
@Mego Before I voted on the answers, it was +3/-1 (I think?) versus +1/-1. Even now, +5/-2 versus +1/-2 is not a particularly strong consensus.
 
@Mego That doesn't make it a consensus.
 
5:09 AM
0
Q: Meta privileges help page inconsistency

Alex A.There appears to be an inconsistency on our Meta privileges page in the Help Center! The unflappable dev Chris Jester-Young and our benevolent overlords in the SE Community Team have lowered PPCG's meta participation privilege reputation threshold to 1. Excellent news! That means that new users ...

 
@AlexA. Yeah, I don't think that's acceptable. (Part of why I don't think my question runs afoul of that is that the older one is four years old.)
 
One issue is that our current challenge format is much stricter than it was when the site started.
So posting "repeats" can add value in a way.
 
At what point is it a problem, though?
 
Anonymous
1. Are old answers *valid* in the new question, or can they be trivially transformed into answers for new questions?
2. Are old answers *competitive* in the new question? (mainly applies to code-challenge, koth, and fastest-code, but also somewhat to code-golf, since CJam/Pyth answers may not be valid due to being created after the question)
3. Is the old challenge free of flaws that can be fixed by edits that won't invalidate answers?
 
Anonymous
If the answer to all 3 is "yes", then the new challenge is a dupe in my eyes. If not, then it isn't.
 
5:17 AM
@El'endiaStarman Well, that's hard to define. :P This is partly why I try to avoid using existing math/cs ideas directly in my challenges.
 
5:29 AM
@Calvin'sHobbies Yeah, methinks I'll go that route now that you mention it... :P
 
Let's see how this goes:
0
A: Create a programming language that only appears to be unusable

DennisChangeling ShapeScript ShapeScript is a naturally occurring programming language. Shape shifters (or Changelings, as they prefer to be called) can transform into a set of instructions that allows them to process data. ShapeScript is a stack-based language with a relatively simple syntax. It is...

 
@Dennis Hahaha my APL is 20 bytes, yours is 3.
 
<{(     bird
>[]     tv
( {)    mustache
[><]    hourglass
There need to be more bracket characters :(
 
@AlexA. Yours is portable though. ngn doesn't have GCD for floats.
 
orly?
I had forgotten about for GCD. Does that not work in ngn?
DOMAIN ERROR: ∨ is implemented only for Gaussian integers
u_u
 
5:41 AM
Best you could do with ngn is the trick I used in Pyth.
 
@AlexA. Put that emoticon in bracket form!
 
First multiply by 2.
 
@Calvin'sHobbies Uh how
 
)
 <
)
Or just continue talking about on topic stuff.
 
@Calvin'sHobbies I like this but it doesn't look like an emoticon on first glance.
@Dennis Where in J/R.5Q/RiFJJ are you multiplying by 2? o.O
 
Anonymous
5:47 AM
@Calvin'sHobbies What's a topic?
 
@AlexA. Just wait for the two-byte J answer. :P
 
Are you working on one of those now?
 
@AlexA. I divide by 0.5, since that casts to int.
 
Oh okay
I forgot that dividing by 0.5 is the same as multiplying by 2. I cannot math.
 
Nah, I have to go to sleep. I just wanted to post the cop to feersum's challenge.
 
5:49 AM
Pffft sleep.
Overrated.
 
6:01 AM
 
:P
Now I'm the one who's going to sleep.
 
%+./ would work as a monadic function, btw.
 
So I just woke up
 
Good night, y'all!
 
Anonymous
I should go to bed, too
 
6:05 AM
G'night!
 
Anonymous
Peace out girl scouts
 
I'm heading to bed too once I sandbox another challenge... :P
 
6:18 AM
0
A: Sandbox for Proposed Challenges

El'endia StarmanTwo mountain hikers want to meet, but must stay at the same altitude This question is totally inspired by this question on Puzzling SE. For ease, I have reproduced the original question here: Two hikers are separated by a two-dimensional mountain range, like the one shown below. The mountain...

 
o_o
That was less than a minute!
 
6:56 AM
Woah, this place is eeeeeempty!
 
 
1 hour later…
8:05 AM
The last message was posted 1 emptiness ago.
 
8:17 AM
:c I can't edit my posts for my latest update of Vitsy.
 
8:47 AM
hi all
 
Hello. c:
 
 
1 hour later…
9:52 AM
...I just more than halved an AppleScript answer...
0
A: Shortest code to produce infinite output

mcgrailmAppleScript: repeat until 1 < 0 say "hi" end repeat

For those that can see edit suggestions.
 
@VoteToClose I've voted to reject, on the basis that golfing edits are generally frowned upon (leaving a comment is fine)
 
10:31 AM
nobody likes my latest java answer T_T
a wild upvote appears ^_^
 
11:12 AM
@Sp3000 Mkay. I've been edited before, thought it was accepted. Whoops.
@Sp3000 I left a comment - is it fair?
 
Yeah looks okay. I dunno, might have to check with @MartinBüttner about golf edits, but personally I wouldn't like it too much if I woke up one day and saw my answer golfed without me testing it
 
editing the actual code in an answer has always been disallowed
 
Hmm, I have an unusable language but it is a lot of work and will be probably cracked in a couple of hours time...
 
11:33 AM
@VoteToClose What the others said...
@randomra so? :)
 
I have finite time
 
I couldn't possibly think of a better way to spend it ;)
 
If I post only the specs, you would have probably solved it and written an interpreter before I could do mine
@MartinBüttner people say earning money is useful in long term, meeting people outside of the internet is also considered beneficial by some
 
11:52 AM
I'm writing a neural network. One of my inputs to this neural network is an ordered list of objects (where each object can be represented as a set of values)
because the order may be important, how do you construct a neural network to handle the list?
 
@NathanMerrill how many elements (value sets)? how many values per element (set)?
 
variable number of elements
fixed number of values
 
@NathanMerrill do you have an idea if the value sets' order is not important?
 
no, I don't have that knowledge
you mean the order of the elements?
it may be important, that needs to be something the neural network is able to eventually assign a high weight to
 
fixed number of values as in 2 per list element, or fixed alphabet?
by how many I also meant size, 5 or 500 list elements are not the same
can you give an input example?
 
12:01 PM
sure
in python:
[("apple",5,"green",90), ("banana",3,"blue", 1), ...]
so each element has the same elements
in this case ("string", int, "string", int)
typically that's easy with neural network
simply enumerate the possible strings
(which I can do)
and assign each of them a network
the problem is the list
it may have anywhere from 1 to infinite elements
the order may be important
 
is [("apple",5,"green",90), ("banana",3,"blue", 1), ...] one example input or the whole training set?
 
one example input
 
you use IDs for the strings?
 
what do you mean?
the strings can be enumerated, they typically only have 4-5 possible values
 
so the network only gets the ID (1..5) of the strings
 
12:07 PM
sure, yeah
I know how I would write a neural network for a single element in the list
 
do you have a method for feeding a simple arbitrary long list to the NN?
 
ok, now I see the main problem :)
 
oooh, I think I got an id3a
idea*
what if I add the position of the element in the list to the element's neural network
then run each of the elements through the same neural network
then average the results?
max the results?
perform a different list operation on the results, and feed that to a neural network?
 
12:33 PM
This could be fed with arbitrary length input:
A recurrent neural network (RNN) is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. This makes them applicable to tasks such as unsegmented connected handwriting recognition or speech recognition == Architectures == === Fully recurrent network === This is the basic architecture developed in the 1980s: a network of neuron-like units, each with...
16
Q: How are neural networks used when the number of inputs could be variable?

Jeremy EAll the examples I have seen of neural networks are for a fixed set of inputs which works good for images and fixed length data. How do you deal with variable length data such sentances, queries or source code. Is there a way to encode variable length data into fixed length inputs and still get...

max, avg,... could be good if you know they can provide the logic you need, eg. you won't be able to compute the parity of a list (which RNNs can as linked in the SO page)
 
I don't care about parity
RNN are effective if one element is dependent on the next, which is not my case
 
so the order doesn't matter,just the position?
 
Hard to say anything more without knowing the problem.
you have to relax the problem but idk what info can and can't be disregarded
(it's often hard to say even if you know the problem)
 
I'm doing general game playing
I'm trying to represent a collection of cards (like a person's hand)
sometimes the order of the cards in your hand is important
but a collection does not have a max size
 
12:51 PM
wow, you are more optimistic about the NN's capability than I am
 
well, I'm not using solely neural networks
I'm using TD learning to learn the game
but I need some way to represent the state of the game at a given point in time
and neural networks appear to be the best option
 
probably the wrong place to ask, but does anyone know is this uses PPM or SMbus? Or even some other protocol?
 
Hey guys
Still working on Microscript II
Any feature requests?
 
Needs moar jQuery.
 
What.
 
1:05 PM
Oh, sorry. When I looked at the original, it didn't have enough jQuery. Did you add more?
 
1:17 PM
18
Q: Golfing strings in Fourier

Beta DecayChallenge Given a string as input, golf down the Fourier program which outputs that string. In Fourier there is no easy way to output a string: you have to go through each character code and output that as a character. Fourier The language is based upon an accumulator, a global variable which...

The bounty is running out :P
 
1:32 PM
(removed)
 
2
A: Shortest Code that creates a Deadlock

VoteToCloseosascript, 0 bytes Oh, yes, this is a thing. Essentially, I used to believe that this was an infinite loop until told otherwise in a comment to this answer.

Applescript wins. :D
 
I don't think blocking input read is a deadlock
 
1:48 PM
Yeah, neither do I.
 
2:03 PM
hi
I want to pose a code-golf challenge to write a very simple interactive graphical game. The idea is to collect the shortest code in each language as in codegolf.stackexchange.com/questions/57617/…
 
Hello
 
If my question is marked as a duplicate, would it be best to delete it or leave it (as it already has a bunch of answers)?
 
anyone think of a suitable game? I thought of breakout, pong but they might be a bit too big
 
@GamrCorps You can't self-delete if it has any upvoted answers IIRC.
 
@Geobits Oh, I never knew that, so just leave it be then?
 
2:06 PM
Yea, it's no big deal for it to be there closed.
 
@Lembik maze game where user types left/right/forward/backward to move
@Geobits ok
 
Yea, a simple maze seems easiest/smallest/simplest for a catalog type.
 
that seems like a good idea. I just don't want it in ascii
should I specify the maze myself or let people generate a random one?
 
If it's not ASCII (or at least text-based), it's going to rule out a lot of the weird languages that catalogs usually attract.
 
@Lembik it would be hard to do it in non-ASCII since most languages cannot display pictures, etc.
 
2:07 PM
@Geobits right.. I don't want it to be text-based
@GamrCorps I understand
 
And you can always specify a test maze, but have users create random ones.
 
Maze generation has been done already, so it might be best to focus on the "display a maze and move around in it" part.
 
but I feel that we shuold have non-text based simple challenges too. Maybe someone can invent a compact language which is good at solving them :)
@Geobits Thanks. That makes sense
of course this is close to pacman :)
 
Or "take a maze as input and etc".
So you can feed in different test mazes.
 
@Geobits I am worried about how to pin down the spec to be honest
 
2:09 PM
When in doubt, sandbox it
2
 
i.e.what exactly should be displayed on the screen
@GamrCorps true :)
 
Good. Being worried about the spec is better than not being worried about it :P
 
:)
I was tempted to have a challenge to place one single yellow pixel in the center of the screen :)
I have no idea what the smallest C code, for example, is to do that
or Java, or Haskell etc.
maybe I should start with that? what do people think?
 
Is it OK to draw other pixels?
 
"Given a set of lines in (x1,y1,x2,y2) format, plot them on the screen"
 
2:12 PM
@feersum no!
@Geobits That seems nice..maybe as a precursor to a more interactive challenge
 
This will be good for machines that only have one possible screen size
 
there is a political point here :) Which is that it annoys me how hard it is to do what I could do on a ZX Spectrum in the 1980s
 
Variable resolution? You're living in some sort of fantasy world! :P
 
which language do you think will have the longest solutions?
and still finite :)
 
Maybe better would be x and y given as percentages of the screen/window/etc. So 0.5,0.5 is the middle.
 
2:17 PM
hmm... maybe it should move the pixel in a circle clockwise
that seems a better compromise
not too trivial but not interactive either
ok I think I can write this version easily :)
but anyone think it's a terrible idea?
or a great idea :)
 
Maybe it would be beneficial if you specified exactly what kind of purple @Lembik
 
@BetaDecay I thought of that but.. how?
 
magic?
 
Hex code or RGB value
 
@Optimizer The usual solution :)
@BetaDecay is it easy to convert from one to the other?
my first downvote! :(
with no comment
 
@BetaDecay thanks... rgb(128, 0, 128) surprises me a lot!
I had no idea that was purple
it's annoying when people downvote without commenting....
 
Yeah. The challenge to me slightly underspecified
 
@Lembik "I will maintain a list"... why not add the leaderboard snippet to do that automatically?
 
maybe it's someone who loves their text only language :)
@MartinBüttner that's a great idea.. could you give me a hand with that please? I have never tried it before
 
@Lembik Check meta: there's a question especially for that
29
Q: Leaderboard Snippet

Martin BüttnerI've been using a stack snippet for a while to generate leaderboards for my simpler or more popular code golf challenges. I want to share the code for this snippet here so others can use it more easily as well. Feel free to put feature requests and bug reports in answers, or let me know in chat....

 
2:36 PM
alternatively, just copy it out of one of the recent challenges that had it. all you need to do is find QUESTION_ID in the code and change it to your question's ID.
your spec seems a bit contradictory though
first you say that the animation has to fill the entire screen but then you say that the circle can occupy 75% of some window?
 
let me try to add the leaderboard code first :)
 
also if you're trying to post a challenge that is a decent reference for "shortest code to perform some simple graphical output task in every language", maybe you should take the time to sandbox it for a few days and actually let people give you some feedback for the spec.
2
 
even (especially?) simple challenges need solid specs.
 
ok.. let me address one thing at a time :)
 
2:40 PM
For example, the catalogue challenges
 
is my question_id 62095 ?
 
Yep
Ninja'd
 
0
Q: A single pixel moving round a circle

LembikThis is a graphical output challenge where the task is to give the shortest code per language. I will maintain a list of the best answer for each language at the bottom of this question. Task Your code should plot a single purple pixel (hex value #800080 or rgb(128, 0, 128)), moving clockwise...

 
I literally copied and pasted some code and changed QUESTION_ID only
could someone look please to see if I did it right?
 
2:45 PM
It is typically suggested that you change the OVERRIDE_USER field as well.
 
to my user id?
 
@SuperJedi224 done I hope
assuming that is ok.. I will move on to the next problem :)
 
you could just... you know... delete and and sandbox it, and fix all those things before anyone starts working on this... because I'm pretty sure the 1x1 window size won't be the last thing someone points out.
 
@MartinBüttner which part says that it has to be full screen?
 
2:49 PM
0
A: Sandbox for Proposed Challenges

Christian IrwanReverse the Meeesesessess upp teeexexextext. code-golf This is inspired by Monday Mini-Golf #6: Meeesesessess upp teeexexextext Background ETHproductions have difficulty from entering text to his usual webpage. Everytime, the text will be meeesesessessesseesseessedessed. Your task is to help ...

 
"Nothing else should be shown on the screen except for the pixel."
 
@MartinBüttner I feel deleting it will annoy people even more than not having sandboxed it at this point :(
 
@Lembik I seriously doubt that.
2
 
@MartinBüttner ah.. "screen or window" is that better?
there is already an answer!
err.. two
 
sure, the longer you wait, the more answers there will be...
 
2:52 PM
the snippet sort of works!
for example reason it has only one entry now
ok.. any other problems that need addressing with the question urgently?
 
yeah, because the other answer's header is wrong
 
@MartinBüttner ah ok
 
@Lembik my point is, addressing them urgently cannot address them as well as if we just had some time to think about this.
 
Currently working on defining the semantics of addition for Microscript II.
 
what about the framerate?
 
2:53 PM
@MartinBüttner I 100% agree I did the wrong thing
 
what about languages that can only output vectorised graphics?
 
and I am sorry
I just really don't want to delete a question with answers
 
@Lembik You don't seem to realise that you can fix it at any time though.
 
@Lembik I don't think you can...
 
@Geobits I think that's only after they get upvotes
 
2:54 PM
@MartinBüttner I am not sure I understand the concern here
 
@MartinBüttner Both are at +1 now.
 
@Lembik the notion of pixels doesn't make sense for vectorised graphics
 
I mean re: vectorised graphics
@Geobits don't! he will just point out I could have avoided that :(
 
@Geobits well I can still delete it... :D
 
@MartinBüttner :)
 
2:55 PM
Sure, you can nuke it from orbit if you want ;)
 
@MartinBüttner I feel the question of whether a 1 by 1 vector is a pixel or not is quite niche :)
I would say it is
 
What do you guys think of the Microscript II draft specs so far?
 
@Lembik I don't understand what you mean by a 1 by 1 vector in this context.
 
It's a 1 unit by 1 unit rectangle in some unit.
 
@MartinBüttner I think you were saying that some languages only show vectors, not pixels
 
2:57 PM
next thing: an inclusive challenge that aims to address as many languages as possible would probably also allow writing the animation in a common file format to disc or stdout.
 
a vector that covers exactly 1 pixel is a 1 by 1 rectangle
@MartinBüttner I explicitly don't want that. My interest, which I understand is not 100% ppcg mainstream, is in useful tasks that I could do trivially in BASIC 30 years ago :)
 
@MartinBüttner That seems to miss the point, which is to demonstrate a way to draw on the screen.
 
@Lembik the point of vector graphics is that they are arbitrarily scalable, so they have no relation to the number of pixels they cover
 
it annoys me that animation and interactivity are somehow regarded as non-mainstream these days
 
@feersum that's fair enough I guess
 
2:59 PM
@MartinBüttner I see your point although it is quite subtle :)
if someone asks, I will answer that they can fix the screen or window width so that the vector takes up exactly one pixel
 

« first day (1734 days earlier)      last day (3102 days later) »