« first day (1852 days earlier)      last day (2982 days later) » 

9:07 AM
would an ibus keyboard for cjam/the esolangs that use really weird chars be useful?
 
Finally done with the post
 
wait
 
0
A: The Holier Numbers

KatenkyoLua, 317 Bytes I had some troubles doing this, some things in Lua don't work as I think it does. I will have to try and play with them if I want to golf this down. Explanations are coming, but for the moment, here's the golfed code! function f(y)h=0(y..''):reverse():gsub(".",function(c)h=c:fi...

 
does the userscript have auto-TIO execution for answers without a TIO link?
 
If someone find why the table.sort doesn't work, I would be happy to know ^^
 
Anonymous
9:12 AM
@orlp I tried using a linear-time algorithm and ended up with a worse score somehow
 
@El'endiaStarman I've never seen it before, and it I was not able to find a reference that using ans is acceptable. This is clearly hardcoding and the specs say that if not specified, you have to provide a function or a full program.
Or do you see that differently? I just searched thou meta and did not find any clue why this should be acceptable either.
 
Anonymous
@flawr You have a point, though Ans is the closest thing TI-BASIC has to command line args
 
@Mego No, you can specify input variables in TI-BASIC programs!
 
D: why does everyone use TI-BASIC instead of CASIO-BASIC?
 
  :addThree(n)
 :Prgm
 :Disp n+3
 :EndPrgm
 
Anonymous
9:23 AM
@flawr Really? I never knew that. Oh well, down with the Ans answers.
 
Also there is Input and Prompt
 
and X if the program is a graph
 
How can a program be a graph?
 
if the output must be a graph in some way?
 
What would the code of such a program look like?
 
9:36 AM
5X+1
input in the graph field of the TI
 
Well that is just and expressions and does not ask for any input
 
Anonymous
X is also a normal variable
 
Anonymous
(outside of the graphing feature)
 
@orlp I tried writing my own version and got the same score as @Mego to 15 decimal places...
 
@feersum can I see the code?
 
9:39 AM
But when you want a program you'd have to write 5X+1->y1(X) which is a snippet that defines a graphing function.
 
OK
If you look at the code you will regret it though
Probably has bugs too.
 
@feersum UnboundLocalError: local variable 'nz' referenced before assignment
print f([1, 1, 1, 1, 100000])
 
@orlp Oh, in the big else block it should say addall = xs / not01
 
Anonymous
I haven't refreshed chat in a while:
 
Anonymous
 
9:46 AM
Funny how the test cases didn't trigger that.
 
Anonymous
And now I don't want to
 
@feersum ye, making test cases is hard
 
Anonymous
@feersum None of the test cases will end up with an initial list with sum > 32767 with the method you and I use
 
Yes it will
E.g. [1, 1]
In mine, that is
 
Anonymous
What score do you end up with if you change the generation to 1000 histograms (with seed 0 still)?
 
Anonymous
9:49 AM
I get 8.006724986235526e-06
 
Anonymous
Wait that can't be right
 
8.0067249862355206e-06
for feersum's code
 
Anonymous
Oh wait yes it can
 
Anonymous
So his is slightly better for the 1000 set
 
no
not significant
probably floating point error
 
Anonymous
9:50 AM
Oh true, at that point it could be rounding errors
 
Anonymous
I think this score may be optimal (or as close to optimal as we're gonna see)
 
Anonymous
So it really just depends on the complexity
 
Mine has a lot of extra code to eke out some differences of 1
But after squaring it might be almost too small to show up
 
I have a more complex benchmark
 
Anonymous
Which really means it's more of a challenge (and thus a chameleon challenge)
 
9:53 AM
t = []
for _ in range(1000):
    l = 1 + min(int(random.expovariate(1/100.)), 1000)
    l2 = 1 + min(int(random.expovariate(1/100.)), 1000)
    l3 = 1 + min(int(random.expovariate(1/100.)), 1000)
    li = [int(random.expovariate(1/1000.)) for _ in range(l)] + [0]*l2 + [1]*l3
    random.shuffle(li)
    t.append(li)
yet this still doesn't manage to trigger the overshoot case like [1, 1, 1, 100000] does
oh wait
edited the code
now it triggers it
 
Do you have an answer validator?
i.e. that you don't have illegal zero/nonzeros
 
for your code
assert sum(o) == S
assert all(bool(a) == bool(b) for a, b in zip(o, h))
 
@Dennis I updates my quine answer with what I think is a true quine
 
@Mego I figured two beta function challenges were a little much, but what do you think I should do now?
 
I demand fully graduated function challenges.
 
Anonymous
9:56 AM
@Sherlock9 Pick one or the other
 
Anonymous
@orlp 8.006772588874205e-05 for seed 0
 
Any preference for which one?
 
@Mego that's different than feersum
he gets 8.1980761942105327e-05
 
@ThijsterHaar OK, I'll check it put when I'm awake. (It's 7:00, I should probably go to sleep.)
 
There's multiple ways to write this up. I'm not even sure whether to do any restricting besides disallowing beta function built-ins
Maybe gamma as well
 
Anonymous
9:59 AM
So I get the better score on that data
 
@Mego could you edit your answer such that it defines a function like feersum, and then you call that function for every line?
with open(argv[1]) as f:
    for line in f:
        hist = literal_eval(line)
        print(model(hist))
and you define model
 
Anonymous
@Sherlock9 What I would do is either just do the complex one, or do the complex one with the caveat that languages without complex support are still welcome and may solve the real version, but aren't competing
 
@Mego even better is
import fileinput
for line in fileinput.input():
 
@Mego Brilliant. I'll do that.
 
Congratulations on graduation!
 
Anonymous
10:05 AM
@orlp Updated
 
@Mego don't make 2 versions man..
just delete the old one
 
@orlp but there are still Seriously 1 programs, and they aren't back-compatible
like python 2 and python 3
 
Anonymous
@orlp Done
 
@Mego Done. Check it out again when you have the time, thanks
 
Is it possible to pull updates from Sourceforge svn using git?
 
10:11 AM
@feersum ok, mego's solution for now is better
tested with this
 
OK, I don't plan to mess with it any more
 
@ThijsterHaar Nice one :) Lots of room for golfing, but nice one :) A quick one to start - I don't know why people are so fascinated with ", but you can do 3d* for '
 
10:29 AM
Looks like Grace Note has a distinct avatar for each of the 152 Stack Exchange sites.
 
@Dennis Did you every find your significant figures answer?
I have found use in "{:.[sig figs here]g}".format(a)
 
Yeah, I'll switch all of Jelly to SymPy.
 
Alright. That works too
 
TIL Grace Note is actually a cactus
 
format is equivalent to what I'm doing. It doesn't work if a is too big for a float.
 
10:36 AM
@Dennis Oh hell. All right. And good luck with Jelly
 
Where's my swag?
 
Also, how would you recommend I go about learning Jelly, bearing in mind that my main experience in programming is Python?
 
@Optimizer These haven't appeared.
 
@AlexA. why is calvin gone?
 
@Optimizer Helka doesn't want to get a trademark strike.
 
10:38 AM
wait, which calvin?
 
@Optimizer Helka Homba (formerly Calvin's Hobbies).
 
@flawr I never got your email through meltmail, unfortunately
 
@Sherlock9 Jelly is completely different from an object-oriented language like Python. Start here.
 
Ah good, so I'll have to build my brain back from the ground up. Excellent
 
"an object-oriented language like Python" <-- Not the first thing that'd come to mind when I think of Python, but OK :P
 
10:44 AM
or of object oriented languages :-D
 
I would have said that Python is an imperative language. What do you guys think?
 
*C-based?
@RikerW userscript looks still broken
 
@Sherlock9 yup, imperative I think
 
@Sherlock9 no, python (like many current languages) is multi-paradigm
 
Multi-paradigm has become a trend :3
 
10:54 AM
I hope you agree Java is pure OOP ;-)
 
nope
look for 'paradigm' here
 
@somebody it still stays OOP, even with Java 8
> that is concurrent, class-based, object-oriented
 
Java doesn't even have first-class functions.
 
If you start saying a language isn't OOP but Multi-paradigm because it could be used in an imperative way, there's a problem in the definition.
 
Plenty of languages have been adopting functional style primitives.
 
11:00 AM
@Katenkyo ctrl+f, paradigm
 
@somebody I did, but I won't change my mind here.
 
OTOH, Haskell has acquired a bit OOPisation, too. Not much in terms of runtime polymorphism, though.
 
On the same point, it isn't because Java 8 now support some tools to allow functionnal programming, that it adds the functionnal programming paradigm to its CV
 
Java has never been functional. In either meaning.
 
Else, you couldn't say of 99% of languages that they aren't multi-paradigm, which then lose it's meaning
 
11:03 AM
> Java works on every device... until Word of God says otherwise.
 
@JanDvorak Java does what you ask it to do, slowly, but it does
 
@Katenkyo until a new version comes out. If you're lucky the original author is still active and signs the applet again.
 
@JanDvorak because you upgrade your compiler without needing to do it on production code maybe? :)
 
Well, no. When a new version of Java comes out, the old one flat out refuses to work. Literally.
 
So how do you explain that some companies still use java 1.5?
A compiler doesn't magically refuse to work
 
11:07 AM
e.g. written in Java 7, when run with Java 8, chat gradually becomes smaller
 
At least for web-facing stuff it does. Desktop apps are inherently insecure, so there's no point to require the latest runtime for them. New version can still break hard-coded paths, though.
 
@JanDvorak some companies uses a 1.5 compiler for their web-facing application (while not the best thing to do..)
I've worked in one, and automatised the nightly build/test/deploy for their java apps
 
But, 99% of the applets written in 1.5 compilers aren't signed enough to let newer runtimes run them. And older runtimes just resign.
by web-facing I mean applets, not servlets.
 
@JanDvorak Oh, so you were speaking about the code in itself..? Anyway, if you upgrade compiler, you're ready to do some changes in your code, so it doesn't change that
 
I mean that 99% of java applets on the Internet flat out refuse to run, and their authors aren't around anymore to do the necessary dance.
 
11:14 AM
never encountered that, so can't say anything.
 
Shall I toss you some examples?
Wait, no, right now I'm unable to tell if applets refuse to run because they aren't signed or if it's just because a new Java version has come out since I've restarted my browser.
 
I'd be glad, I'll look at them after my break
oh
 
Seems I did restart the browser in the meantime.
 
Looks like my version of Java is obsolete too, I'll do that later :)
 
I've tried the site exception list, too. Either it takes a really long time to update or I don't know the right syntax (and it's not specified anywhere either). Perhaps I need to know what CDN the applet gets pulled out from (problematic because of actual CDNs and also to find out) or to enable it for each URL separately (plain dysfunctional)?
 
11:33 AM
Meanwhile, I've been doing an activity to nest programs in multiple languages. Any suggestions for this? New tasks? New solutions? Improving scores?
 
also, what is the really long number for?
 
@somebody Looks like, but I don't want to make an ourobos program. It's very hard.
@somebody This is a ruler so that I can count bytes easily, in the IDE/editor/interpreter/whatever.
*ouroboros
 
@zyabin101 it's a lot easier with -5----10---5----20---5----30 etc
what does get question of the day do in the userscript? (when are they added to the cookies?)
 
@somebody Also, this program is way too long for the defaults of my activity (57 sloc that is way bigger than 64 bytes).
 
12:16 PM
Hello?
 
hello
 
12:39 PM
@SP3000 That's a good tip! I'll see if I can golf the code (especially try to aim for as much 2byte pointers instead of 4byte pointers as possible) later today :)
 
@ThijsterHaar He's @Sp3000.
 
1:18 PM
@xsot i think you meant 55 instead of 57? i wouldn't have tied you if you didn't give hints about there being a magic number, and that it could be brute forced :)
 
@LuisMendo How do you suppose I can best frame my beta function question? Let complex inputs be? Forget complex inputs and have only deal with real inputs? I'm not sure how best to approach this
 
i found 15 with this program, and didn't find any alternate formulations ideone.com/PiIKJf
were you able to pass with a different magic number somehow?
 
@Sherlock9 allowing complex inputs makes programs more difficult (which is not a bad thing).
er, requiring*
 
I figured. And it removes quite a few built-ins since most of those require real inputs, as far as I recall
So, I suppose I'll post the challenge in a few hours. Any other tips, additions, ideas, edits, etc. are welcome
 
@Sherlock9 you forgot the http://
 
1:29 PM
Welp, that's for the moderators to fix, if they decide to
 
1:40 PM
0
Q: Having both an "easy" challenge and a "hard" challenge

Nathan MerrillI've been working on some upcoming challenges of mine, one of them a Traffic Light Simulator code-challenge, the other a Mafia king-of-the-hill. There are certain elements I can include that makes the challenge significantly harder. In my Traffic Light Simulator, I can either make the roads all...

 
2:30 PM
@HelkaHomba you might want to change your name in PPCG MC server chat's message
 
2:44 PM
Oh my gosh, Ton Hospel has found PPCG.
 
@xnor i was actually one of the committers in area 51, then i got sidetracked for a while (and had to write admins / tech support to access my account since myOpenID got shut down in the interim)
so i'm partially a traitor :p
 
3:05 PM
So many DDoS attacks suddenly...
A bunch of 80 to 180 MBit/s attacks from Israel to our company's servers. Also, someone is attacking/scanning the codegolf.xyz TLD, so I shut that down for the moment.
 
@Lynn wow
 
Who is Ton Hospel?
 
The Chuck Norris of Perl golf.
 
How did he start with 101 rep when he isn't a member of any other site?
 
@Rainbolt he has exactly 10 upvotes
 
3:14 PM
I think he just has 101 rep from answers?
 
Oh, whoops. I missed that lol
 
3:34 PM
My day out ends soon. :/
 
3:56 PM
@somebody Did I break it?
@zyabin101 Sorry to hear that.
STAY WITH US
@Calvin is now @HelkaHomba. Aw.
Still calvin there.
 
@AquaTart Thanks for the edit! I hadn't noticed it was a catalog
 
np
Catalogs have kinda died out.
 
^
 
@Sherlock9 Perhaps separate the real and complex case in two challenges. I'd start with the real case and, depending on the answers you get, post the complex one later. I think the complex one may be much more difficult than the real one. In Matlab (which is what I knowe best) thee gamma function only accepts real inputs
@Sherlock9 But then, I may not be being objective. I know a 10-byte answer for the real case :-P
 
@orlp I agree but El'endia really likes the name :P
 
4:06 PM
I know that, I just realized it was the weekly gain not total.
 
You can sort it in other ways too.
 
I was happy that Dennis had 4,099 but was confused about the rest.
@AquaTart I just realized that.
As to the person who merged @All with me, ಠ_ಠ.
 
in The 2nd Monitor, 41 secs ago, by Pimgd
then I can have a new, unmolested duck
The things I've seen in the 2nd monitor....
 
^
 
4:14 PM
in The 2nd Monitor, 1 hour ago, by Pimgd
@JNat why does the logo come off when you rub your duck
 
lol
 
in The 2nd Monitor, 27 secs ago, by Pimgd
We take ducks seriously here at CR
 
Apparently.
 
@HelkaHomba So why the name change?? Just curious
 
Copyright things.
 
4:15 PM
in The 2nd Monitor, 1 min ago, by Aqua Tart
I kinda was hoping you meant a clean build of Rubberduck or something
 
He didn't want to get in trouble if he gets swag.
 
Hey, I'm in hot network questions!
23
Q: Picture of an old friend in ASCII art

Luis MendoI hope this picture looks familiar to you. It's one of Pacman's ghosts in his "vulnerable" state, after Pacman has eaten a power pill. The challenge Picture our ghost in a little frame, using ASCII art. At normal scale (more about this later), each square in the above image should correspond...

 
@LuisMendo people are suckers for Pac Man
 
^
 
@AquaTart Hm. What do you mean? (non-native here)
 
4:17 PM
They see pacman and go OOH PACMAN *UPVOTEUPVOTEUPVOTE*.
 
Aah... ok :-)
 
Don't worry.
 
@LuisMendo "Sucker" usually means someone easily fooled/fleeced/cheated.
 
In this case suckers means easily supsceptible to PPCG charm
ninja'ed
And mine was better. PPCG charm FTW.
 
Anyways, definitely an idiom.
 
4:22 PM
@AquaTart I had a pacman KoTH that didn't get very many answers, and wasn't really that interesting of a challenge
 
Thanks! I only knew the negative meaning of "suckers". As in "hey, you sucker".
 
but it still got lots of upvotes
 
Yep, case in point.
@LuisMendo English is weird
 
^
@NathanMerrill you seem to really like KOTHs.
 
@AquaTart Indeed! Which makes it a lot of fun. You never stop learning
 
4:23 PM
Yep :P
 
I love them
 
KOTHs are cool but hard to organize :/
 
its why I joined the site, and its 90% of my participation here
 
For example, the first time I saw "methinks" I thought it was a joke :-D
 
@AquaTart wait for my upcoming controller. It takes so much work out of them
 
4:24 PM
! ooh
 
@RikerW What would it be about?
closed as too broad
 
@VoteToClose I finished LetterRobot, and I'll upload the source to Github tonight.
 
Either KOTH style with everybody attacking each other, or code-challenge to the highest score.
I would have to write a controller/interface to let the bots do stuff easily.
 
A code-challenge to design a high-scoring nethack bot would be cool but I know for a fact there's an optimal strategy
 
4:27 PM
high-scoring would be a code-challenge
 
@AquaTart Aw.
 
a KoTH would be with them attacking each other
 
@NathanMerrill sorry brainfart
> It is possible to achieve extremely high score (MAXINT on 64-bit platform) by exploiting bug SC343-8 - "Re-animating a statue containing gold produces too much gold.", as it allows duplicating for an effectively infinite amount of gold.[2] Berry ascended with 9223372036854775807 points during Junethack tournament.
:P
 
Anyways, since everything is simply a reskinned nethack, a nethack KoTH is already done
 
why do people not like mice
they're so cuuute
 
4:39 PM
Apparently, Marky is dead.
 
Sneeze
 
@Hosch250 Nope, Marky is absolutely alive! \o/
 
He said he was dead!
 
Well, ...
 
Also, ETHbot is taking a revanche in the Battle of the Bots! Live on Boop Magg Beepot, Magg Beep Boopot and Beep Boop Maggot!
 
5:08 PM
> Not going anywhere for a while? Grab a Snickers.
 
I wish I could have a Snickers :(
 
@NathanMerrill What features will this controller have?
 
I'm aiming for complete invisibility of STDIO communication. You define a turn structure, and it'll automatically serialize and unserialize to and from other languages
also, various other packages, such as scoring
I'm building my own serializer, so that submissions don't have to do difficult parsing. (I'll be building parsers on other various languages, but those parsers would be optional as well)
the input format is var1;var2;var3;
and lists are seperated using commas
unless no commas are necessary
 
5:25 PM
@AlexA. Come up with a name for a regular, non-golfing language to be made by me.
 
@PhiNotPi Phi's Lango Supreme
Phi's Non-Golfo Supreme
 
Phi's Lango Supreme -> PLS -> primary lateral sclerosis
 
RegularScript
 
TotallyNormalScript
 
OneTrueScriptingLanguage
 
5:27 PM
BowlingScript?
 
phiscript
 
OneScriptingLanguageToRuleThemAllScript
 
PhiScriptNotPiGolf
 
CAPS (CAPS ain't PiScript)
Wow, So impressive and huge, not reading it...+1 though — Mr. Derpinthoughton 3 hours ago
LOL
 
!GolfScript
 
5:32 PM
PhiNotGolfScript
 
RugbyScript (it's about as opposite of Golf as you can get)
 
hahaha
HockeyScript. Hockey is like violent golf.
 
PHI Has Interfaces
 
Protected Health Information?
 
check var peeks at the top of the stack ... var1 slashing var2 does div ...
 
5:36 PM
hahaha
 
Got a call at 1:30 from a friend's son who was at the airport, waiting for his father. My friend's cell was unreachable, his computer was off. The son doesn't know Spanish nor how to get to his father's house, so getting there himself wasn't an option.
After trying to reach my friend for a few hours, I had no other choice but to go to the airport myself and pick his son up. He was naturally worried, but apparently I know his father better than he does. It turns out that the 24th at 00:30 doesn't mean the same to him as to the rest of the world. The day only changes after he has woken up...
 
So it's not "tomorrow" until after he wakes up? Makes a certain amount of sense ...
 
What was the UTC time?
 
5:54 PM
0
A: Sandbox for Proposed Challenges

Hurricane996Crash the webpage [tag:code_golf] Inspired by a question(please step forward if you know which one) on Worldbuilding.SE, Cthulhu has been released over Facebook. Luckily, your twitter is linked to FB, so you can post a script that will crash the page so the can reset it. Character count is based...

 
ETHbot is invisible, apparently:
 
@ETHproductions wow, this really is bot writing season
 
in Beep Boop Maggot, 40 secs ago, by ETHproductions
@MarkyMarkov <ETHbot> I'm not seen
 
@ETHproductions I'm not sure if I'm _UNK but I'm not sure about that.
 
@ETHproductions did you just let ETHbot chaat with marky?
 
5:57 PM
@AlexA. haha
@RikerW ya
 
rofl
 
in Beep Boop Maggot, 1 min ago, by ETHproductions
@MarkyMarkov <ETHbot> Are you mean.
 
@Dennis rofl
 
@PhiNotPi Teh language of GOLFINGNESS
@AquaTart Okay.
 
I don't remember if it got fixed in 3.6
 
5:59 PM
@PhiNotPi NormalNotGolf
 

« first day (1852 days earlier)      last day (2982 days later) »