« first day (1693 days earlier)      last day (3133 days later) » 

7:00 PM
@Mauris How do you make J paint boxes with ASCII characters? My console is showing Unicode boxes.
 
Paint?
 
Show, draw, print. Whatever.
 
:P
 
@Geobits Ditto until my doctor saw my blood sugar level and warned me to change my diet if I didn't want to get diabetes. I've now lost all that weight gain and am back in the "underweight" bracket.
 
I haven't had that one come up yet. So far so good, not in any of the danger areas. We'll see what happens in a few years.
 
7:10 PM
Danger Zone
 
I keep typing Geobits in my emails
Luckily I haven't hit the send button on any of them yet
 
My phone autocorrects things to Geobits.
 
We have this thing called a Geoserver, and it isn't one of our products, so when I talk about it I get it confused with Geobits
 
Dr. G.O. Bits
 
Hmm. Well this is awkward.
 
7:13 PM
Oh hey, a paper I worked on for a few years got published! appliedclinicaltrialsonline.com/…
8
 
Congrats!
 
Thanks!
 
I've definitely learned too much today, Alex's last name has put me over the edge.
 
>.>
 
He's semi-related to a godly lion.
 
7:15 PM
I thought "Ararslan" was just one of those cool video game usernames that you came up with as a kid.
 
Haha no
 
Unfortunately my last name is always taken before I get the opportunity to make my first character in every single game I play
 
That's surprising
 
I swear this guy in Diablo II would log in exactly every 89 days just to make sure I couldn't have my name
 
My name got changed slightly to Geobits when others started showing up with my old one.
 
7:16 PM
:/
 
So what was your old one?
 
What did it used to be? George Bitson?
 
Geobyte?
Lunabits?
 
@ಠ_ಠ I don't know you :P
 
Fedorabits?
 
7:18 PM
Sep 13 at 15:42, by Doorknob
@ಠ_ಠ WHY DO YOU EXIST
Geoboats?
 
@Geobits You know me better than you can imagine.
 
Nope.
 
As it turns out, @ಠ_ಠ is Geobits' son.
 
@AlexA. Shut up, uncle!
 
He wouldn't ask me that question... he knows the answer :D
 
7:19 PM
If Geobits knows you in ways better than he can imagine, then are you saying that your imagination is more capable than his?
 
@Rainbolt Sure, I even have the reference imagination.
 
You should imagine a search term that will show you the answer, then. There's at least one I can think of.
Come to think of it, it's been mentioned in here, too.
 
Even a reference implementation can be bugged. I did not find it out up to now...
@Geobits Another proof, in case needed: As you know all odd numbers are primes, so 2 must be odd.
 
All odd numbers are primes? When did that happen?
 
9 is certainly prime.
3, 5, and 7 are prime, so by induction, all odd numbers are prime.
 
Induction works best on normal, ferrous material. You're using odd numbers, so it doesn't work out as well.
 
@Geobits Don't give @BetaDecay any reason to show us his hobs again.
 
> A liberal professor at a famous university lectured his class on what numbers were and were not prime. He started out by saying that the odd numbers 3, 5, and 7 were prime, but went on to say that 9 was not. A certain student, disapproving of simply being told by an "expert" what was or was not prime, raised his hand and asked a question.
"You say 9 is not prime, correct?"
"Correct," replied the liberal professor, who did not like being questioned by his students who obviously were nowhere near as smart as he was.
 
[snort] Albert Einstein. Of course.
 
I don't think you're supposed to snort people.
 
7:33 PM
It's a pain drying and powdering them first, but it's doable.
 
"[forcefully expels air through nostrils]" is a bit verbose.
 
Unfortunately, just verbose enough 'round these parts.
 
So why isn't it [featn] then?
 
@ಠ_ಠ I hadn't come across all of those before. One or two of them are quite good.
 
@ಠ_ಠ Then it would be mistaken for featuring Albert Einstein like they do on songs.
 
7:36 PM
@ಠ_ಠ Would you have known what featn was? Plus, it's ambiguous.
 
@El'endiaStarman No, but I had to look up 'snort' too.
@PeterTaylor Well I just found them via google, because I recalled a similar joke about a engineering student at an oral math exam.
 
Some Noses Often Require Trumpeting
 
haha
 
doesn't this encourage upvoting of bad answers? codegolf.stackexchange.com/help/badges/58/sportsmanship
 
Only if you think all answers except for yours are "bad" I guess.
Unless you just mean indiscriminate voting just to get the badge.
But that's a thing for many badges.
 
7:45 PM
Having the 'most upvoted' sorting as default does not help either.
I suspect most people are only looking at the top 1-3 answers anyway.
 
The other voting ones are generally "use" not specifically "up-vote", but I realize that most times there aren't actually answers that deserve to be down-voted.
 
So for getting a good score the quality is IMHO not the primary ingredient, but rather the order of the submissions.
 
I don't mean to start this conversation. I was looking through the badges and that one was the only peculiar find.
 
(that was bad, and I should feel bad)
3
 
7:47 PM
So votes usually have not much to do with quality.
 
Yes, views is correlated to upvotes. Answering quickly gets more views. Sort order keeps it on top. FGITW is born.
 
@Dennis The default differs from system to system. On Windows, j804 gives me ASCII box drawings.
You can type 9!:7 '+++++++++|-' to set it to ASCII (or, by changing the stuff in that string, to anything else)
 
@Dennis mine draws the required characters by default
 
It tries to read your term settings, probably
 
@MartinBüttner Is there no esolang yet that accepts number inputs only in their prime decomposition?
 
7:52 PM
I think most people don't come up with esolangs purely to loophole golfs :P
 
Dennis is probably already on it
 
I think all golfing languages are designed to shine in golfing situations, whether abusing loopholes in an obvious way or less obvious way.
 
That's a pretty blatant one :D
 
I mean look at cjam, with built in isprime/factorization/permutation functions.
 
some non-golfing languages have those as well
 
7:56 PM
Anyone here ever used Ada?
 
But most non golfing languages that have those usually have a vast library, while cjam does not.
 
I've just watched a documentary about Ada Lovelace and it sounds like a nice language to learn for commemorative purposes
 
These commands appear to be quite specific when you look at the number of cjam commands.
 
@ಠ_ಠ You mean like Python?
 
@aditsu I think we may have discussed this before... were you intending to reserve one each of e* and m* (for some *) to extend to 3-character built-ins at some point?
 
7:57 PM
Sure it has a bunch of useful stuff, but "I can only take input as prime factorization" means to me "I'm only ever going to use this language in golfs where I need to factor things."
 
@BetaDecay I had Matlab/Mathematica in mind.
 
@ಠ_ಠ Mathematica is probably the weirdest language on PPCG
 
I wholeheartedly disagree with that.
 
Damn. The one time I can't use the [alex-is-wrong] tag.
 
I was thinking more because of the huge number of built ins
 
8:02 PM
Mathematica is a little freaky :<
 
what's freaky about Mathematica?
 
Sunset[].
 
I don't think there's another language with so many abstract functions
 
@Mauris eh, that's nothing
 
Which don't require to be included
 
8:03 PM
And it seems you can solve 80% of the challenges here with less than three Mathematica commands.
 
Mathematica builtin: WatchesYouWhileYouSleep[]
 
@Mauris okay this is maybe a bit weird:
 
They're not exactly "built-in"... Mathematica scans your brain to determine what you need. If it's not already present, it creates a new function and adds it to both itself and online documentation. It's simple, really.
7
 
Yeah, I dunno. The syntax just looks scary, so many &s and @s and whatnot? I barely ever understand what's going on.
Hahaha, yes
 
8:04 PM
@Mauris not as bad as Perl :P
 
Or lisp
 
(also that's only golfed Mathematica)
 
@MartinBüttner Well that was unexpected.
 
Very true. Or most actually freaky golfing languages
@MartinBüttner Are there actually Map[] and Reduce[] and stuff?
 
@MartinBüttner Do you mind editing of your answers with trivial grammatical errors I see?
 
8:05 PM
@SvenTheSurfer no, thanks for the proof-reading
 
@Mauris Map[] draws a map and Reduce[] makes a balsamic reduction.
 
@Mauris yes
 
Though I guess Mathematica's no different to Python
 
I know @ is composition and // applies (maps?) functions or something... I'd like to see an overview of what all the punctuation means, really
 
although Reduce is Fold (and there isn't even syntactic sugar for it)
@Mauris not quite
@ is prefix function invocation. // is postfix function invocation
@@ is apply and /@ is map
 
and composition is @* (and there's /* for right-composition)
 
@BetaDecay There are a lot of books in her face.
 
@ಠ_ಠ Bookne. It's a recognised condition ;)
 
Does anyone of you know Matthias Wandel (Engineer/Youtuber/Woodworker)?
 
I saw his lego domino row building machine on Reddit.
 
8:12 PM
He's got a kid now=)
(That poor thing=)
 
@MartinBüttner I don't intend to do that, but I will probably use x* for more 2-char operators
 
is there any reason not to leave that option open?
 
@aditsu Why not use a negative space unicode character?
 
@MartinBüttner I will probably leave it open, i.e. I won't use up every single combination of characters
 
reserving one would be useful if it's supposed to have a reasonable mnemonic though
 
8:17 PM
@ಠ_ಠ Those sadly don't exist (or do they? Unicode is so full of hacks, gosh.)
 
(I was thinking ex and mm or mx)
 
@ಠ_ಠ first of all, I have no idea what that is; second, I want to make it ASCII-only
I will definitely use ex for something (there are at least 3 operators fighting for that name)
 
@aditsu I do not know whether there is, but at least in latex there is the very useful command \! which produces a negative space, so I thought if there is a latex command, there must be an unicode symbol=)
 
if you use x for 2-char operators, which single chars are left? only nkv?
 
@aditsu It seems there are at least zero width spaces
 
8:19 PM
oh, and y I think
 
also u
 
that's quite a lot actually
 
well, n and k will probably show up in the next version
maybe v too
 
@aditsu k as inverse f?
 
no, for more stack stuff
 
8:23 PM
oh?
2
 
well, could be overloaded too
 
what "stack stuff" :P
 
there are a few things that are still hard to do: wrapping top n items in array, copying top n items, moving n-th item to the top
 
oh, the last one would probably be my favourite of those
 
Nope, no one else is starring Martin's oh?. Whoever you are, it only works once per day ;)
 
8:39 PM
@Mauris Ah, yes, it's right there is base.cfg: case. 'Linux' do. ... BoxForm=: 0.
 
9:14 PM
Today's reminder of why I'm not keen on perl - these comments:# "What's in the box?" "Pain."
#
###########################################################################
#
# This is where all the scary things happen: parsing lines into
# paragraphs; and then into directives, verbatims, and then also
# turning formatting sequences into treelets.
#
# Are you really sure you want to read this code?
 
Dune reference :p
 
Is it typical to count unavoidable spaces in Befunge programs as part of the byte count?
As in, spaces needed for aligning flow controls.
 
That's what I thought.
 
9:45 PM
Am I just bad at math or is this due to Java rounding errors?
float num = 0.12438f;
int fifthDecimal = (int)(num * Math.pow(10,5)) % 10;
System.out.println(fifthDecimal); // prints 7 instead of 8...
Someone dug up an extremely old answer of mine on StackOverflow and pointed out an issue with it. Then I got to tinkering with it and now I can't see how my examples were ever correct
 
Just did that with Python and I get 8.
 
Interesting
Well anyway, the commenter's concern was that the solution fails for negative numbers
This was just something I found while messing with it
 
The nearest representable 32-bit float to 0.12438 is less than 0.12438, but the nearest representable 64-bit float is greater than 0.12438.
 
round instead of cast to int?
 
Can't round. I need the digits to remain the same
The challenge is to pick the nth decimal digit of some float
Crap, I've given myself away by referring to an SO question as a "challenge". Now you guys know that I play kiddie golf while you guys are doing the hard stuff.
How is this for a model Stack Overflow answer? stackoverflow.com/a/26048414/3224483
 
9:58 PM
The question itself is flawed, since if we say num = 0.3 it depends on whether the nearest representable float is e.g. 0.29999999996 or 0.300000000001
 
Well whatever the case, I exacerbated the issue. I tested with this:
float num = 0.12438f;
System.out.println(num); // prints 0.12438
It's not until after I do some math on the number that the rounding error becomes noticeable
 
If you do System.out.printf("%.20f", num), you will see that the 5th decimal digit is actually 7.
 
I wonder if printing does some auto-rounding on the number.
 
what @feersum said. "0.3" isnt' a valid float, so askng for the first digit of 0.3 is not asking for the nth decimal digit of some float.
 
10:13 PM
sigh after going back and forth with this guy he tells me he is using Python to test my Java algorithm
Python handles negative number modulus differently
 
I understand the reasons for both types of negative modulus... I just wish someone in math/compsci in the 1800s or 1940s had said "this is how we're all going to do it, and fuck the other guys"
 
10:36 PM
 
I want to have Python's modulus but C's negative division.
 
@Dennis not relevant. this is about picking one of two existing standards, not adding a new one.
 
@Sparr alternatively, languages could just include both modulus and remainder separately (like Julia does)
 
Or you could do like OCaml, and only have the remainder operator, but call it "mod".
 
Just give modulus a higher precedence than unary minus. It's already higher than subtraction.
:)
 
10:48 PM
I just found out about a site feature called a "post notice" meta.stackexchange.com/questions/191981/…
 
They're great.
 
I think it's also the name for the "duplicate of" boxes that appear at the top of duplicate questions.
 
I don't get it. What is "post notice"?
 
I think this is an example of an automatically-generated post notice: codegolf.stackexchange.com/questions/58039/…
The box at the top of the question.
 
meta.stackoverflow.com/questions/250148/… <-- still my most wished for SE/SO feature
 
10:56 PM
@PhiNotPi Actually, post notices appear at the bottom and are added and removed by mods only.
 
@El'endiaStarman Where the "marked as duplicate by" box is?
 
@PhiNotPi Not exactly. Lemme find an example on C.SE.
 
Is it like the gray boxes that say 'this answer sucks' or something like that?
 
Yes. There are three general kinds of post notices that I can add to an answer (see image below). When a question is locked, that counts as a post notice too, but of a different sort, so to speak.
@feersum @PhiNotPi, all of the post notices on PPCG can be seen here: codegolf.stackexchange.com/search?q=hasnotice%3A1
 
11:24 PM
I haven't seen any grey boxes in those questions, just things like open bounty notices.
and "locked by" notices
So I guess PPCG mods haven't ever used custom post notices.
 
I guess your mods haven't added post notices to any answers.
 
nope: Chuck Testa
 
Bahaha. I remember when that was going around.
 

« first day (1693 days earlier)      last day (3133 days later) »