« first day (2119 days earlier)      last day (2726 days later) » 

12:05 AM
Hello @Oliver!
 
heya @oliver
 
Hiya!
 
@Maltysen what's it like
@Oliver hello
 
It's a super hard competition where you have to come out and survive with the power of math
You have to count the math first to win :P
 
>_< i knew that much /s
@Zizouz212 come out?
 
12:19 AM
:D
 
come out of america into canada?
 
No.
Me Canada.
 
canada best
 
@EᴀsᴛᴇʀʟʏIʀᴋ its pretty fun
 
Nah...
 
12:21 AM
you have to be fast
 
Faster than a turtle should suffice
 
12:56 AM
@Zizouz212 ;_; trichoplax is fast
 
What?
 
he is turtle
> When a constructor or a method is called, what form of data can the caller pass as parameters (when the parameters are of primitive type)?
Question: what does the above even mean
is it asking what type it is?
how it is stored in memory?
idk, question is ambiguous
 
1:38 AM
You rang?
@Downgoat That's my CG flair, by the way - my PPCG flair looks better...
 
Just got the job offer. :D
14
 
@El'endiaStarman where?
 
Weird, at ~4.400668GHz I'm picking up an FM station from 105.7MHz
 
@PhiNotPi Agora Games. They're a subsidiary of Warner Bros.
 
@El'endiaStarman Nice! \o/
 
1:50 AM
Agora Games is a video game development company that works with game developers to build online features and web-based communities for video games. It has worked with publishers such as Activision, Konami, RedOctane, and Nintendo, and with game studios such as Vicarious Visions, Neversoft, Splash Damage, id Software, Vicious Cycle Software, 1st Playable Productions, and Gas Powered Games. Major League Gaming acquired Agora Games on August 18, 2009. == Projects == Call of Duty: World at War [1] Guitar Hero World Tour [2] Guitar Hero: Aerosmith [3] Dance Dance Revolution [4] Guitar Hero III: Legends...
huh
> The topic of this article may not meet Wikipedia's general notability guideline.
._.
 
@El'endiaStarman Noice!
 
@El'endiaStarman ? Wikipedia says it was acquired in 2009 by MLG, which was acquired by Activision Blizzard this January
oic, Wikipedia isn't updated
 
@El'endiaStarman Woooo!
Congratulations!
@ΛεγίωνΜάμμαλϠΟΗʹ That sounds really interesting
 
2:09 AM
@Dennis btw on my alternative TIO client thing I fixed the things you mentioned such as now allowing multiline input for args, and a lot better responsiveness on mobile
 
@El'endiaStarman Congratulations! :)
 
@Downgoat The CLA doesn't work at all for me. As soon as I enter more than one character, everything gets truncated. Also, how do I add additional CLAs?
 
crap :| did not test all browsers so just a POC
I was thinking CLAs should be split as \S+ or strings
 
In what browser did you test it? It looks quite interesting in Firefox.
 
Safari, and chrome 54
at least partially in chrome
 
2:15 AM
Chrome 54 is giving me the CLA issue. The CLA works fine in Firefox, but I have to scroll down to see the I/O fields.
 
hm :/
._____. wat is firefox doing
 
(You made me google qhm.)
 
@TimmyD You didn't think I was gonna let Powershell beat Java, did you? :P
 
;_; firefox y u gotta be so weird
(firefox thing should be fixed tho)
apparently firefox does not properly support box-sizing: border-box
 
Are you sure? My client uses it as well.
@Downgoat For CLA-heavy languages like Jelly, I don't think that's a good idea.
 
2:22 AM
@Dennis I mean either that or firefox cannot properly do subtraciton
it seems like the border is calculated incorrectly but rendered correctly. (or the other way around)
 
@Downgoat see tf2 room
 
18K! 20, here I come!
 
Thanks to you all for the congratulations. :D
 
<3
 
@Downgoat I'm stumped. The calculated width (1920x1080, maximized window) is 1523.2px, but if I set it manually to 1530px, it fits...
However, if I un-maximize then re-maximize, it doesn't fit anymore...
 
2:44 AM
@El'endiaStarman For the lulz, a graph of the number of Agora Games employees in the last 5 years:
 
@El'endiaStarman congrats! Any chance you could refer me?
(Just kidding) I don't think I want to move to NY
 
0
Q: "I am Lord [name]" anagram generator

Ghoti and ChipsIn Harry Potter and the Chamber of Secrets we see that Tom Marvolo Riddle is an anagram for I am Lord Voldemort. Task Write a function that the user can pass 2 string arguments to as inputs: the first name (e.g. Tom) and the pseudonym (e.g. Voldemort), such that the output is an actual English w...

 
@NewMainPosts Haha, I love that users name
 
I might be obsessing, but "best possible last card" in a poker hand doesn't really mean the same thing as "card giving you the best possible hand"
 
@ΛεγίωνΜάμμαλϠΟΗʹ Huh, that's actually pretty interesting. They're expanding again and will soon be moving to a new location a block away.
 
2:57 AM
@El'endiaStarman Yeah, was tricky navigating through web.archive.org 's copy, Agora's site has gone through quite a few redesigns
 
Eg, supposedly with 7d 8h 10c Jd, 9d, 9h, 9c, and 9d all tie... but 9d appears slightly worse to me
 
Why would it be worse? Suit only matters to a straight if it's also a flush.
 
@Geobits Because there are only 52 cards and there are other players
 
Ah, so you're saying 9d is least likely to prevent a flush?
If that's the case, then yes, you're obsessing :P
 
9h, 9c, and 9s deny straight flushes that 9d doesn't
Yeah, I think I am :)
 
3:01 AM
Hmm. It doesn't say whether the river card is communal or not.
That would make a difference.
 
@El'endiaStarman What was the job responsibility? Congratulations! You The economy deserves it, it's not like the job market for developers isn't lacking people.
 
Oh that's a good point
 
That would be such a pain to golf though :P
 
@Downgoat browserstack; they set up a real VM with the real OS to test on browsers; all in browser. Keep signing up and using the 30 minute trial. (only counts time when you are actually testing)
 
It's always nice when a new user responds positively to close votes & their respective comments
 
3:06 AM
^
 
@NathanMerrill It earns rapport with mods too.
 
@NathanMerrill :O this happens?
 
@noɥʇʎPʎzɐɹC The trial comes with rather severe limitations though, including access to only a tiny portion of the browsers they emulate.
 
@Downgoat look at the above challenge that got posted
 
@Downgoat Very rarely, but it happened here.
Ninja würde
 
3:07 AM
@Dennis Most of the major ones, though. No mobile or older versions, but they have a wide range.
 
It helps when the poster has a bit of experience with SE as a whole. Not too much, mind you, but a bit.
 
Just wait till new user sees golfing lang ;)
 
1
Q: Finding The nth Prime such that the prime - 1 is divisible by n

Ando BandoProblem The goal is as the title says to find the nth prime such that the prime - 1 is divisible by n. Explanation Here is an example so you understand the question, this is not necessarily the way it ought to be solved. It merely as a way to explain the question given 3 as an input we would...

 
@noɥʇʎPʎzɐɹC Don't know for sure yet. Probably a lot of working with and designing databases.
 
@ArtOfCode We need you
 
3:19 AM
@El'endiaStarman Congrats! :D
 
Thanks! :D
 
Say @Zizouz212, are you familiar with ChatExchange? (I dimly recall you being familiar with it)
 
Relatively speaking, yes
 
Nevermind we might have gotten it working
 
It's been quite a while since I've worked with it... I think they migrated it to Python 3 so I kind of gave up
 
3:28 AM
Actually maybe not

 PPCG TF2 Server - Shut down now

This room is kept for posterity, but the server has been shut ...
Join here and help us with our authentication woes
This worked fine locally
 
@NewMainPosts Interesting. So this is just the T(n)th prime, where T() is triangle number?
Does that make this a dupe of "find the nth triangle number" or "find the nth prime" I wonder?
 
What does that have to do with triangle nummbers?
 
Oooh wait, I'm looking at this wrong. I didn't look at the triangle closely at all >_>
 
I even checked to see if OEIS mentioned anything like that. :P
 
I got there from OEIS, but like I said, read it wrong
I'm gonna just stop mathing now.
 
3:34 AM
I stopped mathing a long time ago :P
 
Almost done building my Planck! Just gotta solder stuff and then put on keycaps.
 
I always thought Plancks were shorter for some reason >_>
 
They are, this picture is just really zoomed in.
 
new challenge idea: write code that outputs the date you joined stackexchange. the catch: it must correctly output the date of the user running the code
 
What is a Planck?
Google is giving apparently irrelevant results on Planck's constant
 
3:37 AM
try "planck keyboard"
 
oh
 
I don't really see the appeal, but more power to ya :D
 
^ smaller shift key = more work on pinky?
 
@Mego 1) Hello, long time no see; 2) I beat your answer here
 
@Downgoat Well, I'm putting my shift key on the right thumb. Much less work for the pinky :P
 
3:40 AM
3) Shouldn't the D6 entry in commands.txt say "D6 (╓): pop f,n: push first n values where f(x) is truthy, starting with f(0) (the function is called each time with a stack only containing x)"
 
@Doorknob oh i (think) I see what's going on
 
Where does the ever important caps lock key go?
 
Yeah, or scroll lock?
 
;_; my favorite key, Prt Sc, is gone in Planck layout
 
I wasn't originally going to have a capslock key, but I decided to stick it on Raise+Esc just for the heck of it. :P
 
3:44 AM
:( I cannot swap caps with escape
 
That's my planned layout so far
 
@Downgoat a) What does Print Screen do these days besides take a screenshot, b) why is that your favorite key?
 
@Sherlock9 to answer your second question:
it is only key you can eat without any repercussions
 
... eat? O_o
 
He is a goat
 
3:46 AM
@Mego Thanks to your new tag, I've been receiving a significant number of votes and answers on my old challenges, haha
 
I am goat :P
 
I see
 
@DrMcMoylex There's a new tag?
 
Question: do people often refer to you as hacker even though you can't even hack /usr/bins permissions back to 755 from 777 :|
 
17
Q: New tag proposal: decision-problem

MegoInspired by a question in chat from a new user, I thought it would be a good idea to have a tag for challenges where the output is a truthy value if the input meets certain criteria, and falsey otherwise. decision-problem would make a good tag name - it's precisely what these types of problems ar...

 
3:50 AM
Thank you very much
 
@Doorknob That's one of the nerdiest things I've ever seen. I love it!
 
Haha, thanks :P
 
wat
4:12 AM
@Doorknob How do I get started building a keyboard?
 
Don't ask me. This is the first time I've ever even soldered anything. :P
 
First (and last) time I soldered something, I had a lot of coffee but no breakfast, so I was trying to hold these micro processors in tweezers while my hand was shaking.
 
@DrMcMoylex O_o
 
It didn't turn out that great
 
1
Q: Abstract Syntax Tree Golfing

isaacgI'd like some feedback on an idea for a new type of challenge that I have, namely Abstract Syntax Tree golfing. Code Golfing is a lot of fun. But many of the tricks that get used in a given language are the same for every challenge. I'd like to ask a challenge which, instead of counting bytes, c...

 
4:16 AM
As long as it didnt blow up you can consider it success :D
 
In that case, it was definitely a success, haha
I did have to use a lot of Flux though
 
@El'endiaStarman Pytek Time™
(in regards to your comment)
If Half Life 3 beats us I suppose that will definitely be a wake up call
 
I'm trying to write comments and tests for my code, and its so boring
 
4:36 AM
@quartata I know, right?
Half Life 3 was actually used as an example by one of my interviewers during the whiteboarding part. :P
 
I've seen my share of code that has redundant, useless comments that are basically a slightly reworded version of what the code says. I'd bet you could write an "auto-commenter" that does that
Don't know how you can cheese your way out of tests though
 
It's easy if your program is a programming language.
I simply put my answer and the test cases as tests for each code golf challenge I answer in it.
 
> 1. A clear specification of what constitutes a correct submission, so that it is possible to indisputably decide whether an entry is valid or not. Test cases are highly encouraged.
2. An objective primary winning criterion, so that it is possible to indisputably decide which entry should win.
aren't these two like the same thing in essence?
 
nope
one determines what an acceptable submission is
one determines what the winning submission is
 
Wait, you mean having the shortest code and giving the correct outputs aren't the same ??
 
4:45 AM
@NathanMerrill ah
 
5:10 AM
Egads... I must be tired... I wasn't obsessing enough earlier. Given 7d 8h 10c Jd, and a non-communal river... 9s is unambiguously the best card.
And I think even 9c is second best, 9h third best, 9d last... they all have a preference order, none are tied
 
Why 9c over 9h? High club straights are blocked by your 10c, but not high hearts.
Not that it really matters what rank if your opponent pulls a straight flush
 
@LuisMendo I created a repo for a bunch of my Bernoulli Factory code: github.com/PhiNotPi/BernoulliFactory I've slowly been accumulating a bunch of classes for a bunch of different operations/formulas.
for example, I define x/(x+y) by means of a recursive algorithm: new Or(new Mult(new Neg(y), new OverSumMinusProd(x, y)), new Mult(x, new Neg(this)))
 
5:34 AM
Geobits: That's exactly why. 9h prevents 10 high heart straight.... and therefore, very slightly increases your chance of winning; say, instead of 9h, they happened to get 9c and thus a 10 high straight non-flush. Then you win.
These are really tiny things but possible
9c doesn't rule out anything along those lines; I think it's just better because of odds, since you have 2 diamonds and only 1 club, of simple flush
(better than 9d that is)
Okay, leaving again... got to watch this cat and then get sleep
Egads... the logic's right but I mixed up the cards... 9s prevents 10 high spades; 9c prevents 9 high clubs, 9h doesn't really prevent anything but flushes. There, now the chat record's fixed despite my brain being fuzzy
(Still, "prevent 10 high spades" includes having 10 high straight non-flushes, etc)
 
Since you have the 8h, the only thing 9h blocks is 9-Kh. On the same note, 9c only blocks 5-9c, since you have the 10c.
Both would beat you though, so since they block one hand each, I'm not sure it matters.
But yes, spades is clearly best and diamonds worst ;)
I mean, yes, both c and h lower flush odds too in addition to straight flushes, but the same amount for each, and they all beat a straight.
 
6:07 AM
Hello
 
Hi
 
Right, but 5c 6c 7c 8c 9x does not beat you, when x is s, h, or d. More specifically, 5c 6c 7c 8c 9x cannot beat you, unless x is c. 3 of the 9's lose, one wins. You don't want that one win...
There is no symmetric scenario with hearts., where you get this specific preference.
 
I got the Socratic badge, woo!
4
 
@Zgarb Congrats! Hopefully I'm not that far behind
 
6:24 AM
@Zizouz212 what?
 
Made a new lang called Doxical!
6
Q: Manoeuvre the grid!

Qwerp-DerpBriefing: You are a bot, in a 2D grid that extends infinitely in all four directions, north, south, east and west. When given a number, you must move the bot so that you get to the target number. Here's how the grid works: You can move in 4 directions: north, south, east or west. Once you move...

More promotion of this (also, Doxical is based off of this)
 
6:47 AM
SpaceX asked permission for over 4000 satellites that can provide 1 Gbps low latency internet to the whole world. businessinsider.com/…
 
6:59 AM
I don't think that takes 4000 satellites. Sat internet can already provide up to 50 Mbps (for consumers). Seeing how the technology improved since the satellites launched that power the current one, we might need fewer, but beefier ones for Gigabit. Still, there is absolutely nothing you can do about the horrible latency.
 
They won't be in the geostationary orbit, that might help with the latency.
 
Finally finished the docs for Doxical :D
I kinda wish there was a display thing for Github repo links
Hello?
 
7:15 AM
@Geobits Final analysis... you're right! 9h blocks King high heart straight... that was my missing symmetry. Exactly as many straight flushes are possible to beat that hand decorated with 9h versus 9c.
 
7:42 AM
Almost as good as the tong one.
 
8:15 AM
Hello?
 
No one knows OCaml, right?
 
8:29 AM
@quartata hm?
 
8:51 AM
@Doorknob Why don't you get a Typematrix?
 
Well the French test wasn't that bad TBH
All of my tests are done! I am free!!!
 
for now
 
:(
@Fatalize What do you think of Doxical?
 
9:07 AM
I don't even know what that is
@Qwerp-Derp ^ ?
 
"Python 100%" off to a bad start :p
"Results in the interpreter crashing, because you're moving back to a square you're already on."
You mean "you were already on" right?
 
@Fatalize Why???
 
As in a Mario game where a platform you walk on disappear?
 
Yeah
 
9:17 AM
@Qwerp-Derp jk, I just don't like Python :p
Does the interpreter crash because of lazyness or does it return a proper error like "you can't revisit a square you already passed on"?
 
@Fatalize It returns a proper error
 
<<
Value: 0
Shouldn't that crash? Or is the counter at 1 initially?
Or does it divide by the counter after incrementing it?
 
The counter is at 1 initially
 
Move west: Divide the value by the counter (value /= counter)
 
Don't worry I got confused
 
9:21 AM
so I assume < will compute 0/0
 
Wait what
Whoa also what is prolog it seems kule
@Fatalize Are you still there?
 
9:42 AM
Yes
What does < do if it doesn't do 0/0?
because from what I read that's what it does
@Qwerp-Derp It is cool indeed. Try it!
 
9:59 AM
@Fatalize It errors
 
but in your example you say that << results in 0 instead of an error
 
10:17 AM
Wait whoops
Uhhh
It does 0/1
Which equals 0
And then 0/2
I just saw 0/0 and < so I assumed you were talking about the sidecase
I got confuse :(
 
wait
when you start the counter is 0 and the value is 0 right?
 
0
A: Sandbox for Proposed Challenges

FatalizeIs it a stochastic matrix? A stochastic matrix is a matrix of probabilities used in the context of Markov chains. A right stochastic matrix is a matrix where each row sums to 1. A left stochastic matrix is a matrix where each column sums to 1. A doubly stochastic matrix is a matrix where each...

 
When you start counter is 1
value is 0
 
That doesn't seem indicated anywhere
So that means that moving will compute +-*/ before incrementing the counter?
 
 
2 hours later…
12:19 PM
Hiya
 
@Fatalize That thing is gigantic :P And not mechanical, and lacks thumb keys, and isn't programmable.
 
@Doorknob It's really not gigantic, I have one. Alts and space are on the thumb
(and enter kind of)
It's a bit expensive though I'll give you that
 
12:38 PM
0
Q: Head, Shoulders, Knees and Toes, Knees and Toes

Kevin CruijssenIntroduction: I think we all know it, and it has probably been translated in loads of different languages: the "Head, Shoulders, Knees and Toes" children song: Head, shoulders, knees and toes, knees and toes Head, shoulders, knees and toes, knees and toes And eyes and ears and mouth and ...

 
1:10 PM
ರ_ರ < eyeless disaproval face
 
the proper way to decide whether something is on-topic is to pray to jeff atwood and sacrifice a goat
 
@Downgoat ^
 
@MitchSchwartz NOT DOWNGOAT. Upgoat is okay.
 
1:27 PM
Meh, goat is goat. It all stews.
2
 
Anonymous
A local radio station was announcing that it was moving to a different channel yesterday. It did that by playing pre-recorded messages between songs (with no regular advertising or DJ chatter). Well, between song - the only song they played all day was Closing Time by Semisonic. Over and over.
 
Anonymous
@El'endiaStarman Congrats!
 
Anonymous
@Sherlock9 Good job! I don't know why I didn't think of that.
 
@Mego Still better than the never ending loop of "Last Christmas" that's about to hit radio stations around here soon
 
I love christmas music, but I can't stand any of the classics
 
1:39 PM
I don't mind listening to Christmas music...when it's not Christmas season.
@Mego Thanks! :)
 
careful or you'll anger santa
 
All I want for Christmas iiiiiiiiiiiiiiiis…
 
Silence
 
Well tbh I don't really want anything
I get a new Star Wars every december that will do
 
Anonymous
@mınxomaτ Said radio station exclusively plays TSO and Mannheim Steamroller for Christmas music
 
1:48 PM
This severely messes with my head in a way I don't understand yet.
2
 
@Fatalize The Rogue One
 
That's a really distracting gif. o.O
 
It also doesn't add anything
 
@trichoplax It actually came from a comment on the one I linked to.
 
0
Q: Repetition-free regexp for "inclusive or" pattern

kjoNB: This question was motivated by a "real-world" programming problem, and in fact, I originally intended to post it on StackOverflow. What made me post it here instead is the requirement to solve the problem with a single "repetition free" regular expression, which is somewhat gratuitous from a...

 
2:13 PM
have we had a code-golf challenge to give a random non-decreasing sequence of integers of some length?
 
@Geobits Down to 67 ...
 
I was just looking at that question, satisfied that neither yours or the C# answer had changed overnight :(
You win this round though. Ain't no way I'm dropping another 40 lol
 
@ArtOfCode The goat was having really weird problems with ChatExchange authenticating. If you could take a peek at the errors in the transcript that would be great
(tf2 room)
 
sure, gimme a few
 
@Lembik Would that be equivalent to the partial sums of a series of non-negative integers? Would you want the jumps to be arbitrarily large or chosen from a defined range?
 
2:17 PM
@Geobits I was really stressing out, because I couldn't think of a way to shorten the code any, until I came into work this morning and said to myself "Why don't I just generate 'em all and check membership?" Lo and behold, way shorter.
 
@trichoplax the idea is to specify a random 1... n and say that the list of k integers must be from 1..n and the list must be uniformly chosen from all non-decreasing lists of that length
 
@TimmyD Oh, I'm surprised you hadn't tried that yet, since a couple answers use it.
 
@Lembik Does non-decreasing allow for k>n?
 
@trichoplax no
 
Is 1, 1, 1, 1 non-decreasing?
 
2:20 PM
yes
 
Or are the numbers chosen from the list without replacement?
 
1,1,1,1 is non-decreasing
the only tricky part is the unformity requirement
 
Is k specified as an input, or must you choose over all possible k?
 
@Geobits Yeah. It wasn't until this morning that I actually saw those other answers. On harder problems like this one, I generally try to avoid "spoilers" for myself to see what sort of solutions I can come up with independently.
 
@Geobits you had to go and find a solution with numbers...
we would have been tied
ugh
 
2:23 PM
@TimmyD I do at first. I don't usually change mine to a direct port of anything I've seen, though, since it normally ends up longer in Java anyway :P
 
It's time for Processing to enter the stage
 
@Poke I knew there had to be something :)
 
> Prohramming
 
The art of raising prize-winning pigs.
 
So that should be prize-winning rams :P
 
@Dennis Eleven abuse to the extreme.
 
I might be able to use numbers too... honestly anything to get that arrayCopy out of there
 
@trichoplax it's specified in the input. The input is n,k
 
Yeah, I tried the arrayCopy thing. Too bad Java doesn't have a shorter slice.
 
2:26 PM
@trichoplax some sort of time limit will need to be imposed otherwise there is a wildly inefficient solution
 
Tried the recursive method with Strings too, but that was a disaster.
150 or so
So many charAt() :(
 
yeah charAt and even substring
 
@Lembik Generate all valid sequences and pick one at random? Or generate all sequences of length k (valid or not) and keep picking one at random until it's valid?
 
or if you tried to go with a character array with string input .toCharArray()
/me shudder
 
@trichoplax right.. both should not be allowed :)
they are easy to ban by saying the code must run in a few seconds for n = 100 and k = 10
 
2:29 PM
I still don't follow why k<=n. Is that just a rule of the challenge?
 
@trichoplax oh I see.. maybe we don't need that restriction
good point
 
I don't know if the restriction allows more approaches or not. I don't have an intuition for what approaches are possible yet
 
let's cancel that restriction :)
 
So for n=2, k=2 we have
1, 1
1, 2
2, 2
each with probability 1/3
 
Free Amazon Prime month ended one day before The Grand Tour starts. >:U
 
2:33 PM
When does Grand Tour start?
 
...
In one day.
But only in the US, UK, Germany and Japan.
 
@trichoplax yes
@mınxomaτ and also on the Internet.. cough
 
That sucks. They should release to more parts of the world. For example, the new Top Gear (with the new hosts) were available online on a BBC site
 
@KritixiLithos it doesn't really stop anyone watching it does it?
 
@Lembik You got a point there
 
2:37 PM
They could enact the new Top Gear right in front of me for free and I still wouldn't watch it.
 
@Lembik I'm not sure if it's a duplicate but the uniformity requirement makes it sound new to me
 
@trichoplax cool
 
The sandbox should clear that up though...
 
1
Q: On this site we obey the laws of thermodynamics!

Luis MendoAnd in particular the second law: the entropy of an isolated system increases over time. For this challenge, An "isolated system" will be taken to be a program or function (abbreviated as "program" from now on); The passing of "time" will correspond to iterated executions of the program's outp...

 
2:45 PM
@trichoplax as soon as I have time :)
 
^^ @xnor @flawr Posted. Note the score function, following @Angs comments
 
@LuisMendo nice, you figured out a fix
 

« first day (2119 days earlier)      last day (2726 days later) »