« first day (1736 days earlier)      last day (3100 days later) » 

6:00 PM
Happy Halloween people
 
That's not funny. :/
 
Hm :/
 
Wow, that's a good kick XD
So what am I going to do after Halloween? I really like my new avatar... :P
 
Keep it?
 
6:08 PM
I'm changing mine to a ghost with a smiling face.
 
I will take off my costume after halloween.
@Zgarb Regarding HPR, do you have any hints on how to print the number?
nvmd, I think I found a way.
 
@TheNumberOne you mean the snake skin?
 
No, the witch hat :)
Yay, I solved HPR!!!!!!
 
Congrats!
And thanks, I guess. :P
 
is FORTRAN still being used?
(in production systems)
 
6:27 PM
0
A: Create a programming language that only appears to be unusable (Robbers' thread)

TheNumberOneHPR, by Zgarb The code: #(*#(!(-)(#(-)()))()!(-)(-)#(!(-)(#(-)()))())(!(-)(#(-)()))#(!(#(!(#(!(-)(#(-)())*#(!(-)(#(-)()))())(!(-)(#(-)()))#(!(-)(#(-)()))())(!(!(-)(#(-)())#(!(-)(#(-)()))())(!(!(-)(#(-)())#(!(-)(#(-)()))())(!(!(-)(#(-)())#(!(-)(#(-)()))())(!(-)(!(!(-)(#(-)())#(!(-)(#(-)()))())(!...

 
@Optimizer yes, yes it is
 
Mortarboarded :D
The sad part about mortarboarding though is that I don't get any rep from all these upvotes I'm getting now...
 
@quartata still read that as "motorboated"... every time
 
6:43 PM
@quartata Well, you managed to beat me to it. (Although I've come within 10 points of it at least twice.)
 
6:55 PM
Ziim is a great language... it's simply impossible to explain the code with anything resembling imperative pseudocode... so instead all I could do is tidy up the graph represented by the code and use that as an explanation o.O
2
A: Simple cat program

Martin BüttnerZiim, 222 201 196 185 bytes ↓ ↓ ↓ ↓ ↓ ↗ ↗↙↔↘↖ ↖ ↓↓⤡⤢ ⤢ ↘ ↖⤡ ↖ ↙ ↙ ↕↘ ↑ ↙ →↘↖↑ ↙ ↑ →↖ ↑ →↖↘ ↙ ↑↓↑ ⤡ This will probably not display correctly in your browser, so here is a diagram of the code: I can't think of a simpler structure to solve the problem in Ziim, but I'...

 
@SuperJedi224 Why doesn't this work in MicroScript: i"em"2s"ykoops"d2sa
Outputs a bunch of spaghetti :/
 
@MartinBüttner is there a program you used to make that graphic? or was it manually done?
 
Timwi provided me with a tool for doing that a while ago. I don't think it's published anywhere. It's not just a tool for generating diagrams... it's actually how I wrote the code (you can copy a text representation out of the program).
 
hmm
 
@timothymh hi
@timothymh are you about?
can anyone translate "Is it acceptable if the movement speed is defined by the key repeat speed? i.e. that the alien moves a certain amount every time a certain keystroke is pressed? " for me?
 
:P
 
@Lembik I think it means, each time the key is pressed the character moves a certain distance so the speed is determined by how often the key is pressed, or the key repeat speed
 
@Vɪʜᴀɴ oh ok.. so the answer is no :)
 
@quartata the i instruction is numeric input, does that help you any?
 
@SuperJedi224 Right
I put in 3
So it should put 5 to the stack
 
7:10 PM
a outputs as characters
 
After me
 
Ohhhhh
Huh
Hmm how should I do this then
 
@TheDoctor did you really use the word unfair :)
@TheDoctor it's hardly fair on C either :)
 
:c
 
7:11 PM
I mean processing basically wins all these graphical-output challenges except for BASIC as far as I can tell
and maybe Mathematica
 
Unfortunately, the p command in the original Microscript includes a newline
 
I don't know if you can do proto space invaders in mathematica
how on earth is the size of a scratch program measured?
 
Usually by the byte count of the corresponding scratchblocks code
@quartata In Microscript II, you could do it as Nps"spooky"p2+p"me"
 
interesting
@TheDoctor why can't processing handle single byte filenames? What is the restriction?
 
Did they shorten the chat edit period on us?
 
7:17 PM
I couldn't tell...
two minutes?
 
I think it used to be 5.
 
that must have been a long time ago
 
Last minute costume: 418 error
For anyone who is actually doing something today.
 
or 419 - carrot
 
419 is marked as UNUSED.
 
7:23 PM
Also great
 
oh, 418 is USED ? :P
 
Well, it has an official error type associated with it.
But it's still marked as UNUSED anyway.
419 doesn't even have an official error associated with it.
 
@Lembik processing needs the extension to figure out how to load it
 
@TheDoctor ah I see
 
I have just invented a most excellent language that will outgolf Dennis.
 
7:25 PM
@TheDoctor on another note.. is it easy to make a crude drawing tool in processing so you can just draw with your mouse?
 
@TheDoctor processing rocks :)
I should learn it
 
yes you should
 
you may carry on winning as my 24fps rule cuts out the old BASIC variants it seems
 
anyway i gtg
 
7:27 PM
bye @TheDoctor
 
bye
 
Bye D:
 
@Lembik Thanks! That was my first ever PPCG answer :)
 
Hello @El'endiaStarman
 
7:44 PM
Hello!
 
How goes it?
 
Bye gys
 
Bye @SuperJedi224
 
@SuperJedi224 Sees you
 
7:52 PM
20
Q: Halloween Golf: The 2spooky4me Challenge!

MegoA current internet meme is to type 2spooky4me, with a second person typing 3spooky5me, following the (n)spooky(n+2)me pattern. Your mission is to implement this pattern in your chosen language. You should write a program or function that takes a value n (from standard input, as a function argume...

Chaine, 15 bytes: {i~}:-,}:{2+}me
 
@trichoplax my pleasure :)
 
Hmm. @Lembik I think I may be able to use a TI-84 CSE emulator that runs the calculator at 4x speed. Would it be acceptable to get 24 FPS that way?
(By the way, the solution isn't trivial in TI-BASIC; there's no native sprite support)
 
8:07 PM
^ and what a shame at that
 
@CᴏɴᴏʀO'Bʀɪᴇɴ wat
 
@quartata :D you noticed
 
@CᴏɴᴏʀO'Bʀɪᴇɴ how
I don't even wat
Rotor is 16 I don't see how 15 is possible even in a hypothetical language
How does this code work?
 
@quartata I'll enlighten you
Chaine is a string-based language; it's sole purpose is to be good at strings.
 
Alright
 
8:10 PM
The :...: accesses a base 93-encoded index in a dictionary
-,} is the b93 index of spooky
 
lol
 
Wow
That is impressive
 
Why do you need the "me" then?
Is it shorter than looking it up?
 
Yes.
:(WY: = me
The dictionary is sorted by alphabetical order
Unfortunately, it is an noncompetitive answer, since I just invented Chaine
 
8:12 PM
Do you have an interpreter for it yet?
 
Yeah!
Lemme git it up
(Warning: a lot of commands are still unsupported, and there is no user-friendlyness yet)
 
Sounds like it could kick some ass in too
 
Anonymous
@CᴏɴᴏʀO'Bʀɪᴇɴ You'll have to buy me dinner first
 
Anonymous
 
Anonymous
Successful rep farming
2
 
Anonymous
8:15 PM
@CᴏɴᴏʀO'Bʀɪᴇɴ what is this I don't even
 
...crap, I can't just output everything as characters. 2 and 4 don't work as ASCII codes...
 
Now I wish people would wait a couple hours before upvoting my stuff
Mortarboarding isn't all that fun
 
....double crap, outputting as integer adds a space.
And I don't have a num-to-int conversion command yet. :(
 
Anonymous
@CᴏɴᴏʀO'Bʀɪᴇɴ Go ahead and post it and mark it as non-competitive
 
@Mego :D Fantastic!
 
Anonymous
8:18 PM
And now that I successfully farmed a bunch of rep, I'm going to post a bounty
 
@quartata Welcome to the club!
 
Anonymous
I will give a 150 rep bounty to the first solution that is eligible for the gold bonus and validates inputs based on pattern matching, rather than my simple solution of running the programs through the interpreters. — Mego 8 secs ago
 
Anonymous
30
Q: Seriously, GolfScript, CJam, or Pyth?

coredumpSome time ago, the following question was asked: GolfScript, CJam, or Pyth? Based on the title only, I thought that it would be a very nice challenge, but unfortunately, it turned out to be a question asking for tips. Here is the challenge I wanted to read: Who said golfing languages were not...

 
Anonymous
150 rep for anyone who can solve ^ using a not-dumb solution like mine
 
8:22 PM
@Mego I believe that question is underspecified.
 
Anonymous
@ThomasKwa How so?
 
It says 99% accuracy on a Bayesian solution is sufficient.
What does 99% mean?
It will correctly recognize 99% of programs? How can that be scored when there are infinitely many?
 
Anonymous
It means alpha=0.01
 
Anonymous
Confidence value is a better way of putting it
 
Anonymous
So the solution has to be 99% confident (using Bayesian analysis) that it is the language(s) it spits out
 
8:25 PM
what do you mean by "using Bayesian analysis"? Where do the data for such analysis come from?
The question asks to identify which languages the program is valid in, not which language the program was probably written in
 
yay
 
translate: Voici l'interprétation de Chaîne.
(from French) Here is the interpretation of string.
 
Anonymous
@ThomasKwa From the test cases and a statistical model. It's impossible to be 100% certain that a finite string of symbols is a valid program in a given language (without actually trying to run the interpreter). So, the threshold is 99% confidence.
 
@Dennis Actually, Chaîne is the name of the language, so it remains the same.
 
8:28 PM
Tell that to Bing.
 
I mean its what I called the language. I made a programming language called Chaîne, and there was its interpretation.
 
Anonymous
I would expect a deep learning approach to be fairly successful at that challenge
 
@CᴏɴᴏʀO'Bʀɪᴇɴ I know, but it's an automatically generated translation. I didn't copy-paste.
 
8:29 PM
Here is Chaine's interpretation/rendering
 
@Dennis I know as well, it just goes to say.
 
@quartata Therefore, .
3
 
@Mego Passing the test cases is a separate restriction from being "99% confident", right?
 
Anonymous
@ThomasKwa "If you use a fuzzy/bayesian approach, "without a doubt" means with a high-level of confidence (you score 99% with your classification, for example)."
 
@El'endiaStarman ಠ_ಠ
 
8:30 PM
[whistles nonchalantly]
 
Anonymous
@ThomasKwa It's saying that, if you give it any finite string as input, it should output the correct language(s) with 99% confidence. It must correctly pass the test cases, but the 99% confidence thing is in regards to additional testing.
 
So 99% of finite strings should be correctly recognized?
 
Anonymous
It must give the correct output on the test cases, and correctly identify any program with 99% confidence.
 
Anonymous
@ThomasKwa Statistically, yes.
 
My Bayesian confidence level is always 102%.
 
8:32 PM
But there are an infinite number of finite strings.
 
Anonymous
@ThomasKwa Not a problem for statistics.
 
and a very large number of finite strings of bounded size.
Yes it is
What if my program only correctly recognizes strings of length >1 googolplex?
That's 100% of the finite strings, because there are infinitely many passes but finitely many fails
but it's useless.
 
@Mego Unless you specify a distribution, it kinda is.
 
Anonymous
You are confusing the meaning
 
I wonder what you could do with a Pyth program that's 1 googolplex characters long...
 
8:34 PM
@El'endiaStarman Anything/crash the internet
 
Anonymous
If you take a random sample of finite strings and pass it as input, it would correctly identify at least 99% of them.
 
@CᴏɴᴏʀO'Bʀɪᴇɴ Turing would say "no". :P
 
@Mego a random sample on what distribution?
 
Anonymous
@El'endiaStarman import universe
 
@ThomasKwa Weibull
 
8:35 PM
@El'endiaStarman Fine. theoretically anything
 
@Mego That reminds me of Isaac Asimov's "The Last Question".
Let there be light!
 
Anonymous
@ThomasKwa A finite random sample of the infinite set of all finite strings
 
@Mego Random how? It may pass 99% using one distribution but merely 1% using another.
 
@CᴏɴᴏʀO'Bʀɪᴇɴ ......uh, still "no". I'm referring to the Halting Problem by way of computable/non-computable numbers.
 
8:36 PM
@El'endiaStarman OHH. Right. Well... Only your sanity bounds your output.
 
@AlexA. Does Weibull have some properties that make it elegant in this case?
 
Anonymous
Ok, maybe I'm explaining this incorrectly
 
@CᴏɴᴏʀO'Bʀɪᴇɴ import cthulhu
 
@ThomasKwa Nope
 
Anonymous
8:37 PM
The program, if it uses statistical inferencing, should be 99% confident about its answer
 
Anonymous
I'm sure it's possible, for any pattern-matching program, to construct a string which yields a false negative
 
Is ΣWeibullpdf(x)*2^x, x=0 to infinity even finite?
 
Anonymous
Because it's getting uncomfortably close to the territory of computability theory, the Chomsky hierarchy, and Gödel's incompleteness theorem
 
Anonymous
It's like how regex can't parse HTML
 
@ThomasKwa I don't think so
 
8:41 PM
I'm deciding whether or not t do all the project euler problems... Any suggestions?
 
Anonymous
@AlienG python
 
@Mego I want to do them in python, C#/Java and maybe in one more...
 
@AlienG JavaScript.
 
Anonymous
@AlienG Start with python. Builtin bignum support makes it a lot easier to solve some of the problems that involve massive numbers (though the best solutions often involve exploiting number theory to not have to deal with massive numbers). Then port them to other languages that don't do bignums as well.
 
I've been trying to do all the Project Euler problems in python, but for one I had to do JavaScript.
 
8:45 PM
Well I like Python and all but I usually use C# (90% of the time) so I'm sort of torn...
 
@Mego You just took one step up the abstraction ladder; that's further from defining the concept.
 
Anonymous
@ThomasKwa Let's put it this way. The solution should almost surely identify every finite input correctly.
 
Anonymous
I'd be happy if the solution could correctly categorize every solution currently on the site as CJam, Pyth, Golfscript, or none of the above (probably perl).
 
Anonymous
But for the sake of my bounty, it is only required to correctly identify all of the test cases in the OP.
 
I like to do the Project Euler problems in APL.
Which reminds me, I should do more of those.
I think I did the first 10 and then forgot about it.
 
8:54 PM
@CᴏɴᴏʀO'Bʀɪᴇɴ How many have you solved so far?
 
@El'endiaStarman Lemme check...
 
@AlienG I highly credit Project Euler with helping me learn Python.
 
A pitiful 12 problems
 
1-10, 14, and 89
 
8:56 PM
gotcha
And that's an interesting spread. :P
 
How about you? (And yes, it is XD)
 
um...
135 problems...
 
ಠ_ಠ
nice :D
 
:D
I've solved the majority of those, basically all but the first couple dozen or so, with Python.
 
Anonymous
Uh-oh
 
Anonymous
8:59 PM
PE is saying the password Chrome has saved for me is incorrect
 
Nice! What was the other language(s)?
 

« first day (1736 days earlier)      last day (3100 days later) »