« first day (1806 days earlier)      last day (3031 days later) » 

1:00 AM
@NathanMerrill Hmmm
 
@PhiNotPi An ad that size might lead to downvotes, but you could look into subtle product placement ;)
 
How about [intersection coordinate, side the lights are on, [new lights]?
And only work with the ones that have changed
 
I'm sure there's been a lot of challenges which have introduced people to games they'd never heard of
 
I'll bet Spacewar! was one.
 
that would work
 
1:03 AM
@Sp3000 I used to play Nethack a long time ago. Picked it up again because of the Minimalist Nethack challenge
 
0
A: Sandbox for Proposed Challenges

CallodacityCalculate the Langelier Saturation Index if the local pool! Anyone who has operated a commercial swimming pool will know that maintaining a good Saturation Index (SI) is important. Water will dissolve anything it touches until it becomes saturated - when this happens the excess solutes will prec...

 
@NathanMerrill Is this going to be like a ? Most optimal traffic management?
 
Thank you @NewSandboxedPosts.
 
@PhiNotPi For the Bloom filter challenge, I like the idea of explicitly specifying two hash functions, but allowing people to use their own if they prefer.
 
^^not
 
1:04 AM
Are you @TanMath's sock?
 
@trichoplax Any idea what two hash functions could be? One of the two could literally be ord()%32.
 
@RikerW I have no socks!
 
@TanMath That must be rough on your feet.
 
^
 
@RikerW No.
 
1:06 AM
@TanMath Are you sure?
 
Let's all put in 50 rep and have a sweepstake on how many more times RikerW can thank the bots before being chat banned
 
@quartata the latter. Trying to come up with a good scoring that punishes making cars wait too long.
 
@trichoplax Do I get all the rep at the end?
 
@RikerW I have sock socks... I have no sockpuppets but!!!
 
@RikerW Um, no
 
1:07 AM
@trichoplax yeah!! let's do this!
 
So you will bet on my survival time?
 
I guess 25 more times
 
Chat kicked, yes.
Chat banned, no.
But other than that, sure!
 
kicks rikerw
 
Maybe use the sandbox room though.
@BlockCoder1392 I think you are @TanMath's sock.
You both abuse bots.
 
1:08 AM
@NathanMerrill Make it so all the cars have a preset place they're trying to get to. When all the cars make it to where they want to be, the number of ticks that have passed is the score
You'd have to find a way to deal with strategies optimized for the specific cars though
 
@quartata good idea.
 
@RikerW huh?
 
@PhiNotPi Is it golf? If so I like the idea of specifying hashes that are not suited to golfing, so the best answers will be ones that choose a more golfable hash
 
@RikerW
 
You don't thank them, and try to stop people from thanking them.
 
1:09 AM
@RikerW I dont have a bot that works
 
@trichoplax Yeah, it's golf. Then the next thing would be placing requirements on them making their own hash functions.
 
Sup @Calvin'sHobbies
 
@Calvin'sHobbies keeps disappearing and coming again...
 
@BlockCoder1392 I mean @NewMainPosts, @NewMetaPosts, and @NewSandboxedPosts.
 
@RikerW Maybe you could use the sandbox room for thanking bots :P
 
1:10 AM
For some reason when I try posting my bot in JS console i get "docunet.getElementsByClassName is not defined WTF???
 
@quartata I don't want the sum/average. It leads to ideal strategies of leaving the busy routes open, with the other roads at red
 
@trichoplax Only if you link them into there too.
 
@RikerW I do? Wat?
 
@BlockCoder1392 you kicked @RikerW
 
6 mins ago, by BlockCoder1392
^^not
 
1:11 AM
@NathanMerrill Ideally you don't tell people what the preset positions are.
 
@BlockCoder1392 docunet?
 
7 mins ago, by RikerW
Thank you @NewSandboxedPosts.
Then two down was the earlier quote.
 
@PhiNotPi Would it work to just place a requirement on the probability of a false positive? Is that straightforward to calculate?
 
I think you could make an executable for a controller + honor rule not to decompile
 
@quartata each of the feeder roads have a busyness rating
 
1:12 AM
@NathanMerrill Ah.
OK.
 
which is known at the beginning
 
So the plot thickens.
 
the user doesn't know where cars are heading exactly
 
@RikerW So?
 
maybe we should change "Thank you" to "I will use you to my advantage" in every one of @RikerW's posts
 
1:12 AM
When does the simulation end?
 
@BlockCoder1392 You disagreed with thanking the bot.
 
@RikerW so? its 3confuzing5me
 
@trichoplax If the hash functions are (practically) independent & uniform, then the false-positive rate is easy to calculate. I would just have to place some requirement on "independent and uniform."
 
Maybe something like the largest amount of red lights a car got
 
@BlockCoder1392 its 4confuzing6me
 
1:13 AM
If the hashes are not independent, then false positive rate could be all over the place.
 
@BlockCoder1392 You think the bot shouldn't be thanked, so you are supporting abuse and by extension abusing them.
 
@quartata a fixed number of ticks
 
OK
 
@quartata Life is a simulation and it ends when we die
 
anybody like the idea?:
2 mins ago, by TanMath
maybe we should change "Thank you" to "I will use you to my advantage" in every one of @RikerW's posts
 
1:15 AM
@AlexA. I see you're one of them weird digital physicists
 
^
 
@TanMath No. That would require moderator intervention to change each of his posts, probably against his will.
 
@PhiNotPi Do you want to allow someone to choose 2 hash functions that are neither independent nor uniform, but happen to give a probability of false positive less than your cap because they fit together well?
 
tfw lost a vmf for a map and you have to decompile it
 
@quartata ¯\_( ͡° ͜ʖ ͡°)_/¯
 
1:16 AM
@trichoplax Sure.
 
Too many TLAs
 
@RikerW Well what I meant is that it's so confusing and I wouldnt thank someone to give me a really confusing lecture on a code golf right?
 
brb having fun with my now func_detail-less map
 
@quartata I'll miss u
 
@BlockCoder1392 Oh, you weren't talking about the thanking the bot for its hard work part?
YOU AREN"T @TanMath's SOCK!!! YAY!!!
 
1:17 AM
@PhiNotPi I like the opportunity this leaves open for golfers, but it depends on being able to find a way to measure the probability across all possibilities
 
You'll miss me even more if I don't optimize it and let it take 2 years to compile
 
@RikerW ah, come on!
I wish I had a sock!
 
vvis amirite fellas
@TanMath A sock! A sock! My rep for a sock!
 
@TanMath You can have @FaxMachine.
 
@RikerW who is he??
 
1:18 AM
@TanMath Alex's "official" sock
 
@quartata I want to see him...
 
@TanMath Here.
 
@BlockCoder1392 thanks!
 
@trichoplax Since there's two hash functions, and 32 bits, there's 528 possible hash results (compared to 94 possible characters). I could put a limit on the number of hash collisions.
 
@AlexA. let loose the hounds
 
1:22 AM
@PhiNotPi Any given bit string could be the result of between 0 and 94 characters, right?
 
@trichoplax Yes. The characters are added individually, so there 2^94 possible sets of characters.
 
dammit, This function has another bug
 
@Mego what is your best Seriously answer?
 
If someone rewrites it for me I'll give them 20 rep.
 
Since there's only 32 bits though, the bloom filter is not very effective once it gets to a certain number of included characters.
 
1:25 AM
@PhiNotPi Hmm. Much to large a space to calculate the probability accurately for unusual hash functions
 
Kind of disappointed there are so few categories in the best of PPCG 2015. Personally I wish there were a "Rookie of the Year" award. There've been some amazing newcomers this year.
 
That's why I'm thinking on putting a limit on the number of hash collisions (the number of characters that hash to the same pair of bits.)
 
How would you feel about setting a large batch of test cases and measuring validity solely based on those?
@PhiNotPi I think it's reasonable to insist on zero hash collisions after 2 inputs
I think I used the wrong terminology there...
I mean after 2 inputs, the 2 inputs are always recoverable
 
@Zgarb That was really cool.
 
Any comments before I post this?
3
A: Sandbox for Proposed Challenges

NinjaBearMonkeyExpand the numbers You may remember in first or second grade using expanded form to learn about place value of numbers. It's easier to explain with an example, so consider the number 123. In expanded form it is represented as 100 + 20 + 3, which helps a young mind visualize place value. It is re...

 
1:29 AM
Can anything be recovered from this?:
-5
A: Sandbox for Proposed Challenges

TanMathDo-nothing Polyglot Challenge This challenge requires you to write a polyglot which contains as many types of comments possible. The comment must say This is a polyglot comment. The program must do nothing at all. Besides comments, your code may contain no-ops from the language(s) you are usi...

@NinjaBearMonkey good question! looking forward to answering it!
 
@TanMath Wisdom about what types of challenge not to post in the future.
 
@NinjaBearMonkey ooh
Looks good.
 
@El'endiaStarman They've also made a 3D version, which looks really spooky.
 
@trichoplax This is actually impossible, I think.
 
Aaand posted
 
1:35 AM
dammit didn't get a chance to prepare my perl solution
 
Hey this is pretty nifty:
1
A: Hotel room numbering

feersumSnails, 32 bytes A \*{l\#,|r,9d.,{|=~r,9{|l.,\ \#

 
@Doorknob ooh
 
if sum(s^2 for s in S) < sum(t^2 for t in T), then is sum(s^1.5 for s in S) < sum(t^1.5 for t in T)?
assuming s and t are always positive
 
Which gives me an idea as to how to validate solutions: There's 94 choose 2 = 4371 possible pairs to two letters.
 
1
Q: Expand the number

NinjaBearMonkeyYou may remember in first or second grade using expanded form to learn about place value of numbers. It's easier to explain with an example, so consider the number 123. In expanded form it is represented as 100 + 20 + 3, which helps a young mind visualize place value. It is reminiscent of how you...

 
1:39 AM
@NathanMerrill Not if the values are all between 0 and 1, for example
 
if they are all greater than 1?
 
We calculate the number of false positives generated by those 4371 tests.
 
(which they will be)
 
@trichoplax ^
 
@PhiNotPi Refreshing my brain on Bloom filters and it's still fuzzy, but doesn't that allow for fitting 4371 pairs into 32 bits without collision?
I'm trying to get my head around it
It just requires that any given pair have a maximum of one bit in common, right?
 
1:42 AM
@Mauris @Phrancis @orlp @DigitalTrauma @BetaDecay @Fatalize @Quill, got your time capsule entries.
 
cool
 
That makes 64 valid entries. Not even close to covering all of ASCII, though. /cc @Calvin'sHobbies
 
Here's my logic: Let's say that after adding some number of character in, a false positive is generated when testing for some character X. This means that the two bits X is assigned were previously flipped by two other characters, which could be Y and Z. Then, a bloom filter with just the characters Y and Z in it would still give a false positive for X.
 
@Doorknob I hope by next year everyone wont have lost interest in this :/
We could do a mini week-long time capsule :P
 
@Calvin'sHobbies Says a long-time user of a four-year-old beta site.
 
1:45 AM
@PhiNotPi Oh yes I get that. I was thinking that you had added X and there existed a Y which had the same hash as X
 
Haha. Month-long isn't a bad idea, actually. Or we could still wait the whole year; if people forget about it then the "ohhh yeah that thing!" factor may play a part. :P
 
@El'endiaStarman 4.95
 
@El'endiaStarman I've only been here 1.5 of those
 
Well, I'm not sure what you were meaning when you said "after 2 inputs, the 2 inputs are always recoverable"
 
@Maltysen Are we really that close to five years? Wow.
 
1:45 AM
Chat minichallenge: Write a program which reads a line of STDIN and sends Calvin an email with the input as the body... in one week's time
 
@Calvin'sHobbies Okay, not quite that long, but still.
 
@PhiNotPi yes - that wording of mine was inconsistent with what I'm thinking. Sorry about that. I probably need sleep...
 
okay
 
@Sp3000 But note that Calvin will only respond to emails that contain actual spam (the meat)
 
@PhiNotPi So we can require that 2 inputs will always set at least 3 bits, but those 3 bits won't necessarily identify the 2 inputs uniquely?
 
1:47 AM
@El'endiaStarman WOOHOO 5 years without graduatin- :| :/ -__-
 
@trichoplax Correct
 
@Doorknob what did animuson say to FlagAsSpam's swag request?
 
Are we the site that's been in beta the longest?
 
Someone onebox the comment. Trying to on mobile will be painful :P
 
Does that mean we win beta?
 
1:49 AM
@Calvin'sHobbies nope
Check A51
Board and Card Games (@Rainbolt hangs out there) is at 1907
 
PPCG is #5, following Audio-Video Production, Writers, Homebrewing (Beer, Wine, etc.), and Board and Card Games, which has been in beta for 1907 days compared to our 1807 days.
 
Wow, exactly 100 days difference.
 
(area51.stackexchange.com/?tab=beta&page=2 for people who don't want to go find it)
 
@El'endiaStarman Only 28 from a nice round number
 
1:51 AM
Wait I though BCG graduated already
huh
 
Apparently not. Also, their questions per day and answers per question ratios are worse then ours.
 
100 days difference? And it only takes 90 days to go from public beta to graduation...
 
@trichoplax That hasn't been true since....ever?
 
@Quill He said he could not give us that :/
 
1:52 AM
@El'endiaStarman We have the most green requirement thingies though of those sites
 
so basically rip our chances of graduating
 
@quartata oh that sucks
 
CR is lucky.
 
Maybe you ought to do like a campaign to get site stats up
 
We tried that
 
1:53 AM
@Calvin'sHobbies Wow, yeah. #2-#4 all have questions/day in the red.
 
@Doorknob is trying for 1 question every 2 days i think...
 
Twice
I do not think it is a stats thing at this point
 
Our questions per day are (is?) steadily rising. Our eventual graduation is inevitable
 
@RikerW right
 
Good so far?
 
1:54 AM
Yep
I have one scheduled for today
I try to post as close to UTC midnight as possible
 
Cool
lol
1
Marijuana

Proposed Q&A site for medical and recreational users interested in the various aspects of Marijuana.

Currently in definition.

 
So when I get home I'll immediately choose one and post it
 
> Recreational Users
 
I have like 4 different ideas for challenges. I should probably write some of them
 
@Doorknob `Choose one`?
 
1:55 AM
@El'endiaStarman Looks like of the betas that have 2k+ days, only area51.stackexchange.com/proposals/1321/… and area51.stackexchange.com/proposals/2578/philosophy are on par with us in terms of most green requirements
 
Maybe we're just not a good fit for the network or something. We're excellent in all categories except for questions which we are Ok in. Sites have graduated with worse (far worse) stats
 
@RikerW I have a file called stuff.md with like 15 queued challenges
 
@Maltysen Or add any you don't want to write to the donations meta post...
 
@Doorknob lol
I was thinking about one called I can't believe it's not Büttner!. :P
6
 
@Calvin'sHobbies Interesting.
 
1:56 AM
@RikerW Please post that
 
@RikerW 11/10 so bad it's good
 
It will be nearest neighbor for a butter ad vs Martin's profile picture. Outputs that if closer to the buttner image. :P
 
I could believe it. It's a very distinctive trench coat
 
@RikerW wut?
 
@Maltysen Compares input to two images, and outputs "I can't believe it's not Büttner!" if it is closer to Martin's image.
 
1:59 AM
oh lol
 
How do you pronounce Büttner anyway @Martin? Is it like butter, or like butte (byoot)?
 
Bohtner is closer.
Wachtler is even harder.
 
somewhere between buht and burt?
 
Yes.
Think a throaty growl. :P
 
I guess I'll try writing my nine deaths of the ninja one
 
2:00 AM
Do so pls.
I will sandbox the buttner one in a bit.
 
Like the {x} sound in Lojban? (Pronounced like a throaty h, sort of like Loch or Bach in some accents)
 
gtg finish math for now.
@Doorknob Yes.
Exactly.
Now, screw hyperbolas.
 
@RikerW don't be mean to hyperbolas ;-;
 
I assumed screw was an adjective and this was some kind of weird spiral
(I don't mean weird in a mean way...)
 
Nine days from fanatic.
 
2:04 AM
@Doorknob Nice!
I hope it's a good character. I had a tough time picking.
 
@trichoplax Now I wonder what a hyperbolic screw looks like...
 
I have enough difficulty with metric/imperial
 
@RikerW What the heck? There's no /x/ in any pronunciation of "Martin Büttner" I can think of...
 
@Mauris Lojban.
I'm guessing they have different pronunciations for many characters.
 
@RikerW This one @Mauris
 
2:08 AM
Welp. I made the same erroneous assumption as Mauris. Never mind...
 
Martini Butter
is close enough
 
I had the advantage of growing up with that sound in Welsh words like chwech
(begins and ends with a throaty rasping phlegm-clearing sound)
 
Interesting product
 
I would say [ˈmɑɹtɪn 'bjʊtnɚ] in English and [ˈmaʁtin ˈbʏtn̩ɐ] in German and I just realized how I misunderstood the conversation but I spent too much time cobbling together this IPA not to hit enter on this line, hi.
 
2:12 AM
It is actually Walked-ler in the american way, and closer to Valk/x/er in the german way.
 
@RikerW 4/10 too much differential geometry, not enough pictures of hyperbolic screws
 
@Maltysen Agreed.
 
martin martin can almost become minty train ram
 
Hyperbolic screw 2:
YAY!!!
Technically a hyperbolic worm gear, but whatever.
 
Close enough!
-1 not a gif
 
2:16 AM
lol
user image
2
 
@RikerW I was actually thinking of how to combine a hyperbola and a helix, but this is cool too.
 
Hyperbolic gear gif: ^^
 
@RikerW ooh
 
@El'endiaStarman Same.
Technically invalid for non-screw hyperbola...
 
Wow. There's a challenge there somewhere
 
2:17 AM
@trichoplax Isn't there always?
 
@Ampora Fair enough... But this one seems particularly interesting both visually and code-wise
 
Cool thing from google images (same search):
 
I was actually thinking of one recently: partition the plane based on an adjacency graph.
 
@Calvin'sHobbies Give me a few hours - I need to stare at this more
 
2:19 AM
White space indentation in python is terrible
even python is sad about it
 
Like if I give you an adjacency graph for the faces of the cube, you could partition the plane any number of ways so long as the parts have four sides, there are six of them, and they're arranged appropriately.
 
I still like my idea better. :P
 
@Cyoce It's wonderful. I don't have to manually close loops and conditionals.
 
No
Even python is sad
def f ( x ):
          ^ sad face ):
see?
 
^^^^^ Very much agreed.
 
2:21 AM
Whitespace indentation even makes the source code unhappy
 
:26708147 I'm glad you removed that or I'd still be there searching for the join
 
@RikerW Gifvs don't one-box.
 
@RikerW WHHYYY!!!?!
 
^^^ So people don't have to load the gifs we keep pasting if they don't want to
 
2:23 AM
@Calvin'sHobbies I've been there before. I ain't going there again. Wanna conserve my precious time...
 
Is the character ‾ okay to use in a challenge?
 
@PhiNotPi What challenge?
 
One I'm planning.
 
About?
 
2:25 AM
@PhiNotPi If you mean required in the output, I often do see people complain about any non-printable ascii that may be required
 
@ThomasKwa SECRETS!!!
 
Squiggly lines /‾\_/\/\/‾\/\/‾\/\/\_/\
Not required in output, will be in input though.
 
I'll stop now. :P
 
K I'm done with 9 deaths of the ninja except for test cases
 
Okay.
 
2:29 AM
Ehh, I'll just get rid of the non-ascii
 
For my sandbox challenge: ^
 
@RikerW Cool pics.
 
Thanks. Including the last one? :P
 
That one is a bit odd...
 
@RikerW Please don't post huge, off topic, animated GIFs. The two I deleted alone weighed 10 MB. Some of us have slow internet connections and pay per byte.
 
2:31 AM
lol
Okay. Sorry.
 
I can't believe it's not Butter!
 
For my `I can't believe it's not Büttner!` challenge.
 
@Dennis And some of us are on mobile atm, and for that I thank you
 
Sorry @Sp3000.
 
0
Q: 9 Deaths of the Ninja

MaltysenInspired by this conversation in chat. Your goal in this challenge is to emulate a ninja and count how many deaths he has left. Specs You ninja starts out with 9 deaths left. He also gets an integral starting health as an input. Then, he takes as input a list of events in his life that alter ...

 
2:35 AM
@RikerW No worries - Australian law now says companies need to text if I'm approaching my monthly limit :)
 
Okay.
Lucky you...
 
There's usually a day or two delay though, so as long as I don't binge browse in a day!
 
Okay.
Netflix on your phone? :P
 
Heh, now that is a bad idea
 
My squiggly challenge is in da' box.
 
Try it, you'll like it. If you don't we will absolutely give your data back...
 
Does anyone see something obviously wrong or something that can be improved?
1
A: Sandbox for Proposed Challenges

J AtkinTuring's truth tables code-golf Alan Turing - Wikipedia Truth tables are very important in logic. They are a very useful tool to help you visualize how a function works. In this challenge, you are going to make a program/function that generates a truth table. Ops that you must support: &: ...

 
0
A: Sandbox for Proposed Challenges

PhiNotPiSmooth vs. Chunky vs. Broken Squiggles Based on Chunky vs. Smooth Strings. Squiggles /\_/\/\__/\/\/\/\_/\_/\ are fun to make on a keyboard when you are really bored. But not all squiggles are created equal. Some squiggles are smooth, like \___/, and some are chunky, like /\/\/\/\. Others are...

 
Thank you @NewSandboxedPosts.
 
1
Q: 9 Deaths of the Ninja

MaltysenInspired by this conversation in chat. Your goal in this challenge is to emulate a ninja and count how many deaths he has left. Specs You ninja starts out with 9 deaths left. He also gets an integral starting health as an input. Then, he takes as input a list of events in his life that alter ...

 
2:43 AM
Thank you @NewMainPosts.
 
@RikerW You are a very thankful person :P
 
lol
 
I've heard it said that gratitude killed the cat
 
@trichoplax they are feed bots, not cats.
 
@RikerW I was more thinking that you were like the cat ;)
 
2:55 AM
@RikerW Uh, looks like you should be more careful.
 
@Dennis So....RikerW will commit a murder-suicide?
With kindness.
 
Any feedback on this before I post it? meta.codegolf.stackexchange.com/a/8019/2867
I almost decided to skip the sandbox on this one.
 
> answered 19 mins ago
o_O
 
PhiNotPi is trying to overtake Downgoat.
 
^ he gets it
 

« first day (1806 days earlier)      last day (3031 days later) »