« first day (2095 days earlier)      last day (2751 days later) » 

1:00 PM
@miles yes. I just added my results which are similar
It might have been better also to have used floating point numbers between 0 and 1
to prevent the permanent being huge
then it would have been a challenge to keep the precision I suppose
@miles did you follow the link I pasted about doing 256 bit arithmetic quicky?
 
Hey, does anyone here have an Amazon Echo?
 
@BetaDecay What happened to you D:
 
I'm slowly decaying
 
Hey, does anyone here have an Amazon Echo?
3
 
1:14 PM
@Lembik yeah I just read it, it seems familiar to something I read on StackOverflow about sse and bigint also. I'm planning on submitting a version in Java, right now it only does n = 27 in about 48 seconds but thats using BigInteger which is a big decrease over just using long.
 
@miles Cool!
I should have added Java to the list of wanted languages
 
Java for a fastest code???
 
in fact.. any suggestions for "system language"? That is ones that run fast
@TuxCopter Java is about half the speed of C, right?
 
@Lembik did you compile their programs using their recommended settings, it seems odd why the gap between Dennis and Hospel is so large for you
 
The JVM is really slow
 
1:17 PM
There's Python entries but those were sped up with PyPy
 
@miles 38 versus 53 seconds? It's not so huge
@miles yes.. pypy can be great
 
@BetaDecay Why does it say colour.box
WHat challenge is this
 
@TuxCopter it's not really slow. All the benchmarks I have seen say it is about half the speed of C
@ASCII-only my challenge?
 
I'm seeing 27 vs 35 at n = 33 but I'm not sure
 
Java is like Alzheimers; it starts slow and eventually takes away your memory.
 
1:18 PM
@Lembik Link?
 
@TuxCopter I am not sure where you are getting this info from
 
@Lembik It's basically a joke, but it's also half-true
The JVM is generally really slow
 
@TuxCopter which part is true?
@TuxCopter but look at the benchmarks!
 
@Lembik It start slow and it take all your memory :P
 
Java today is much faster than it used to be and can even approach the speeds of c/c++ in some tests, but it still has that stigma of being slow
 
1:20 PM
you would have to show me some benchmarks for this
 
Someone downvoted antsy permutations. :/ I guess they didn't appreciate my lame pun.
 
@miles well "used to be" must be referring to the 90s?
you can write slow Java code if you are not careful
 
I wonder how slow GMP is compared to native support (or LLVM IR's arbitrary precision integer is)
 
If Java is slow it's mainly because of the GC
 
I mean it's slower than C, but I wouldn't call it slow in general.
 
1:22 PM
I'm not that old, I began with Java 6, but I guess installing the JRE for runescape might give me an earlier date
 
-1
Q: Add an option to disable upvoting for your post

Mitch SchwartzThis feature request applies only to codegolf.SE and only to main, not meta. I think it would be cool to be able to make a post (question or answer) that doesn't allow anyone to upvote it. The reasoning is specific to this site. There is, conveniently, already an objective scoring method associa...

 
> rem
 
@NewMetaPosts That would be hard to implement
(I think)
 
1:24 PM
maybe someone actually understands SSE and can take advantage of it
that would be cool
for my challenge
 
There are a lot of languages that are slower than Java
 
I'm trying this weird idea of holding the n-bit gray code as a boolean array to see if it would increase speed
 
C#.net has some slower scores on that site @Lembik linked.
 
e.g. python
 
@miles sounds cool
@miles do you think I should do something about the big num problem?
it wasn't really meant to be a big int challenge
 
1:27 PM
@NewMainPosts This is at 4 close votes as unclear. With someone talking about posting an answer I think this could do with being put on hold until there has been some discussion of how to make it clearer.
Oh already done...
 
^ fitting avatar
 
Me?
 
@ASCII-only Colour box? What do you mean?
 
@BetaDecay The watermark :P
 
61
Q: Why do people still say Java is slow?

TheLQFor a long time in SO and in other places Java has the reputation of being slow. From jokes to many comments in questions and answers, people still believe Java is slow based solely on experience with it in the 90s. This is my issue: we have disproved (most) of the reasons that people believe Ja...

 
1:29 PM
@Lembik maybe modulo a prime like most of those online programming contests?
 
Yay my Celeron has SSE4
 
\o/ I have SHENZHEN I/O
 
@Lembik the challenge is fine as is, I don't think there's a real need to change the output format to double precision float/etc
 
@TuxCopter Good, now waste 100 hours on it over the next week :)
 
@ASCII-only I see no watermark... :/
 
1:36 PM
brb when an external event make me go out of SHENZHEN I/O
 
@miles ok thanks. Mod 3 would have been interesting :)
 
I think the Wiki page said something about mod 2 or powers of 2 also
 
@miles mod 2 is "easy"
mod anything else is hard I believe
 
mod 1 :P
 
:)
 
1:43 PM
Miles: does your processor have AVX2 ? That helps a lot for 64-bit multiply. Lembik's processor cannot do fast 64-bit multiply
 
oh that's a good point
gcc is probably optimizing more since I do have AVX2
 
hi @TonHospel
I thought my CPU was OK :(
 
It is, it's just before AVX2 which added a lot of very useful vector ops
 
just wait till next fall, Intel will hopefully release consumer CPUs with AVX512 (was dated for 2016 but delayed)
 
1:44 PM
I just realized that I in fact use the wrong kind of multiply for your CPU too since not even 32-bit multiplies get properly vectorized
 
@Lembik To quote Weird Al, "Throw that junk away man, it's an antique!"
 
.o/
 
Dang, so close to rep cap
 
The author of this challenge has added another bounty, this time for a somewhat related but clearly different task. What should be done about this?
 
1:59 PM
@ETHproductions I don't see the bounty. Is it an open ended one?
 
^ ditto
 
@trichoplax Ugh, CTRL+C didn't work. Click the link again now please
 
^ same. Perhaps you're confusing the protected banner?
 
@Downgoat I just pasted the wrong link
 
Definitely an incorrect use of bounties
Though I don't understand why he didn't sandbox/post a challenge instead
 
2:01 PM
@ETHproductions I hate it when that happens. Lucky you didn't have anything confidential on the clipboard...
 
@trichoplax I would've noticed if it was anything but a PPCG link. Unfortunately, it just happened to be that...
 
SHENZHEN crash every time I launch it ;_;
 
The new bounty appears to change the requirement of the challenge, inviting answers that are not valid under the original rules. You could post this new requirement as a new challenge, rather than including it in this one. Otherwise the whole challenge is likely to be closed as it no longer has a single objective winning criterion. — trichoplax 43 secs ago
 
While we're on the topic, I'd like to drop a casual reminder that this bounty-related meta post exists and is lacking a good answer :P meta.codegolf.stackexchange.com/questions/7260/…
3
 
I think my comment stands, but it isn't actually possible to close the challenge for the next 7 days due to it having the bounty...
 
2:08 PM
;_; only 25 people have clicked on sandbox ad too date
 
@Downgoat 25 more people who don't post straight to main sounds like a job well done to me
 
@trichoplax Unless most of those 25 people are people who already use the Sandbox...
 
^
 
Fair point. I feel my optimism evaporating...
 
@Sp3000 the problem with this is actually defining "winning criterion". Aka, we don't have problems with people that bounty for "clever solution". What if I wanted to bounty a submission because it was fast and golfed (but not necessarily the fastest one posted)?
if we disallow only objective winning criterion, then people will simply make it unobjective :)
 
2:13 PM
I think the problem is when there is a direct conflict with the winning criterion. If I say "cleverest solution" or "best explanation", then the answer still has to be valid, which means making reasonable effort to meet the winning criterion. But if I say "fastest solution" on a golf challenge, it gives an incentive to not meet the winning criterion, which is polluting the set of answers rather than improving it
 
I totally agree, but it's hard to define where the line is
 
Maybe we need a "bounty sandbox" ;)
 
I should say there's a reason why I've posed a question but haven't provided an answer :P
 
I think at least some of the questions in your meta post can be answered, so maybe an answer than does so and leaves the rest as subjective would help?
I want to give this some thought
 
" if I say "fastest solution" on a golf challenge"
Why not just post a fastest code challenge instead…?
 
2:15 PM
Because you're lazy
 
@Fatalize I definitely agree with that. It wouldn't count as a duplicate and would keep like solutions together for meaningful comparison
 
what if I post a bounty for a "fast solution that's still somewhat golfed"
 
fastest code and code-golf almost always result in completely different answers, so cluttering one with answers from the other would be strange
 
The same goes for the example that started this - posting a different output requirement. It should just be a separate challenge, but now that the bounty is posted it can't be undone
 
is that allowed on a code-gofl?
 
2:17 PM
@NathanMerrill How would you score that?
 
you wouldn't score it
its a bounty
no need for objective scoring
 
O_O 7/8 of newest questions have been closed
@SE we need posting rep limit pls
 
@NathanMerrill If it was the main winning criterion you could find a way to make it work well. As a bounty, it still conflicts with the true winning criterion, encouraging answers that distract from the desired answers
 
Though I wonder if it's possible to come up with a good metric that takes both the program size and the runtime to score an answer
 
@Downgoat There's already a rate limit for people who consistently get downvoted
 
2:19 PM
@trichoplax so, we have to "judge" whether or not something conflicts with the true winning criterion?
like, least whitespace could easily conflict with the true winning criterion
 
@NathanMerrill Yes, there are always going to be grey areas. I still think it's worth having some guidelines and updating them over time
 
I don't understand bounties anyway. Why would I ever give away my precious green points?!
 
@Fatalize You must never downvote, either, then?
 
I still think that the bounty of "fast solution that's still golfed" is a valid bounty, but I also think its heavily into the grey area, and not everybody will agree
 
2:21 PM
23
A: We're not a Q&A site. But what should be done about it?

Alex A.Increase the "Ask Question" Privilege Threshold to +2 I should note that I'm not convinced that this is a good idea but I figure I'll mention it to see how others feel. On Q&A sites, the point of posting a question is to get help, to ask a question and get an answer. There shouldn't be a barrie...

@Fatalize I've posted and seen others post golfs where there is a time limit for a particular test case. So the winning criterion is "shortest code that is at least this fast" and that seems to work well. So it think it's fine if it's the core of the challenge, rather than tacked on as a bounty in conflict
 
@TimmyD I do, but rarely. I get bounties tbh I'm just kidding, I just never felt the need to submit one :)
@trichoplax And what @NathanMerrill was suggesting is kind of the opposite: "fastest code that is at least this short"
 
If you're keen on your precious green points, you can sometime profit from a bounty if it leads to rep on a question/answer of yours...
 
or some mix of the two. Bounties don't need to be objective. I could just say "fast code that's golfed", and pick the one I think did the best
 
@Fatalize I think that's a step more defined than Nathan's example, as it sets a specific maximum length for the code. That makes it objective, but there are problems with max length code as it's hard to choose a length that suits all languages. Similarly it's hard to set a time limit for the other approach since a meaningful limit for one language will make the task impossible for others
The problem we have here is that the challenge author is the one posting the bounty, and has edited the winning criterion of a challenge with a huge number of answers already
 
> This bounty will be arbitrarily awarded to a randomly selected answer on this question. I will do this by attempting to give my cat a belly rub, then counting the resulting scratch marks and taking the answer numbers modulo the scratches, then sorting in ascending order and taking the smallest.
 
2:28 PM
A separate but contributing problem is that the bounty prevents voting to close the challenge
@TimmyD Interesting. Objective but not reproducible...
 
How do they select users that will receive swag upon graduation btw?
 
first 2 pages of all rep iirc
 
@Fatalize Whoever was in the first two pages when they graduated (not at design time)
 
so that would be the first two pages in february then?
 
Yes. Which pobably doesn't include you sadly
I think you were at 6k or so
 
2:36 PM
7165 precisely
 
I think bottom of page 2 was about 8k
 
@quartata :'(
I thought I had a chance...
 
I wasn't even here...
 
I was on page 2 on september 22 2015 apparently
 
@quartata I feel like it might actually be based off of design time.
17 of the 50 most recent questions have negative vote totals
which feels like an abnormally high number to me
 
2:47 PM
0
A: Alternative winning criterion bounties

Nathan MerrillYes and No There are two kinds of bounties we can disallow: Bounties that reward answers that don't fit the submission requirements: "Draw a cat" on a Fibonacci challenge. Bounties that are unrelated to the scoring criterion, and have a tendency to push answers away from the scoring criterion...

 
Also, the swag thing reminds me to be careful about falling off the 2nd page of users.
 
I'm impressed at how the Pyth, Jelly, and Osable answers on this question are all currently at 7 bytes.
 
@El'endiaStarman although the Jelly one will go down to 6 :)
 
Due to English language causing severe misunderstandings of the intention of this challenge, I strongly feel it has been reopened too soon, and would benefit from being on hold as unclear again until the discussion has played out.
0
Q: 100 Prisoners and a light bulb

jacksonecac100 Prisoners and a light bulb You may have seen this puzzle over on the puzzles exchange or perhaps heard it somewhere else. If you are unfamiliar with this puzzle check out: 100 Prisoners and a Light bulb The gist of this puzzle is that there are 100 prisoners, each one is separated from one...

 
3:21 PM
Mmm ... 15 rep ...
 
Ah yes, those delicious green points.
 
@PhiNotPi Wait, really? Holy shit I need to get on it then I still have a good chance.
I'm pretty sure it's based on graduation though
 
I think so too. I hope I'm wrong though, cause I would love swag.
 
3:38 PM
32
Q: Puzzling Stack Exchange - Top User Swag!

animusonWe want to congratulate and thank everyone that helped this site and community grow healthily into what they are today! If you can find your name among the top 72 users in the first two pages here, we have a surprise for you! We want to send you a token of our appreciation for being among the mo...

They point to the "current list" of top users.
no mention of "at graduation time"
 
22
Q: How will graduation/election swag work with the changes to the "graduation" process?

CatijaHooray, Design-Independent Graduation is happening! Sites that have been deserving of graduation all over SE are getting their elections and (hopefully eventually) getting the "Beta" label removed from their site name. Part of the graduation and election process is the Swag sent to users who are...

Judging by the Chemistry and Movies post the swag was promised before the design came
Also Quill said that's how it works and he knows everything about how this dumb thing works
 
Any clue our position in the waiting list for site design?
 
Like I said, with the amount of features we asked for, I'd be surprised if we got our design in a year
We're definitely a low priority
(understandably)
 
hmm... the "print a triangle/hourglass" guy is trying to call off his bounty, to create a separate question for it.
problem is that he's already created a bounty on the original post
 
Mods can refund bounties
 
3:56 PM
0
Q: Travelling Pumpkin Problem

YodleBackground: Jack is a pumpkin that enjoys spooking the citizens of the villages near his pumpkin patch every Halloween. However, every year after someone lights the candle inside of him, he has a limited amount of time to spook everyone before the candle burns out, thus being unable to spook an...

 
Oh, goody! I just received an email from Aron Smith at ar337a@nyc.rr.com telling me that the IRS is "supenaing" my company, and I need to open this Word .doc in order to respond.
 
:O
 
Seems legit.
 
It's them haxx0rs again.
Though I would trust HMRC to pull off something like that.
 
The word document contains the text "rmdir system32"
 
4:06 PM
Somehow I installed Office 2007 on my Arch Linux system
and I now have a box popping up saying "Upgrade to Windows 10"
ON LINUX
wtf
 
hmm... wat
 
I go to my Wine menu and Windows 10 upgrade assistant shows up
3
 
I understand Microsoft wanting everyone to use the latest version of their operating system, but lol.
 
Change the Windows API version in winecfg.
 
It's Windows XP for Office 2007 compatibility reasons
 
4:28 PM
@TonHospel I've setup my i7 920 again which does not have avx support, and for n =33 (all 1's) I get 2m46.738s for yours and 1m1.893s for Dennis. Unfortunately I'm not adept enough to see what the compiler is doing differently to make such a difference
 
4:51 PM
This is my favorite comment of all time:
I disagree. STACK OVERFLOW IS HEREBY CANCELLED. — Jeff Atwood ♦ Dec 10 '10 at 10:41
3
 
@miles all 1s is quite a mean case :)
 
5:55 PM
I would normally add a bounty now
but the answers are so awesome already I am wondering if it is worth it
a quiet chat room!
 
6:18 PM
Has there ever been any more information on when we properly graduate? New design, and whatnot?
 
November 1st.
10
 
Mm-hmm
 
6:41 PM
0
A: Sandbox for Proposed Challenges

OliverPrint an alphabet party hat Your task is to print this exact text: z yz xyz wxyz vwxyz uvwxyz tuvwxyz stuvwxyz rstuvwxyz qrstuvwxyz pqrstuvwxyz opqrstuvwxyz nopqrstuvwxyz ...

 
0
Q: Totally Invertible Submatrices

TimmyD(inspired by this question over on Math) The Definitions Given an n x n square matrix A, we can call it invertible if there exists some n x n square matrix B such that AB = BA = In, with In being the identity matrix of size n x n (the matrix with the main diagonal 1s and anything else 0), and A...

 
Anonymous
 
@NewMainPosts I first read this as "Totally Invertible Submarines". Now I'm disappointed.
 
@Mego What is the point of this edit?
Ah
 
@mbomb007 Those are called bathtubs.
 
@mbomb007 "Totally immersible submarines" sounds even better!
 
@TimmyD ♫ We all live in a yellow rubber ducky, a yellow rubber ducky, ... ♫
 
Anonymous
7:15 PM
@mbomb007 It's not all that it's quacked up to be
 
Anonymous
...I can feel the incoming chat ban
 
Everyone gets the song reference, right?
 
@mbomb007 Me too!
 
Anonymous
@mbomb007 If they don't, we have a chair, rope, and the entire discography available
 
@Mego Meh. One song is enough.
 
7:17 PM
@mbomb007 Yeah. It's that one song from that one really unpopular, talentless, unheard-of British music group.
 
Anonymous
@TimmyD The Rolling Stones didn't do Yellow Submarine
 
^^ What are these dashed lines?
 
@Mego Oh, I was going for U2.
But Rolling Stones, that works, too.
 
Anonymous
@TimmyD They're Irish. If you call them British, Bono will whine for a few years to no effect.
 
Oh, right.
My bad.
Besides, I'm from the US. Britain, Ireland, Scotland ... they're all pretty much the same to the majority of USians.
 
7:22 PM
:/ SHENZHEN I/O don't work
 
@TimmyD At first I thought that US. was US, and that you were saying you were from all four. lol
 
Anonymous
Ireland is part of the British Isles, but is not part of Britain or the United Kingdom, with the exception of Northern Ireland, which is part of the UK.
 
Anonymous
And the Irish are very proud to not be part of Britain
 
And Britain is regretting its Brexit
 
@Mego Oh yeah. I watched the CGP Grey video.
 
Anonymous
7:24 PM
It's all very confusing
 
Same with the Holland video.
 
I think I've seen the vids you're talking about. Pretty confusing stuff.
Althought the main point of the Holland video was that it's really "The Netherlands", not "Holland"
Holland is a region within The Netherlands
 
Right.
 
@mbomb007 I don't know anyone who (claims to have/wanted to) leave
in the first place
 
Anyone here have advice on writing/keeping change management documents for software projects?
 
7:30 PM
use git commit messages properly?
 
The thing is - I do....
Our information security officer is pushing for us to have documented changes per release for each project
 
@muddyfish There are plenty of articles/interviews
 
(ie: word document stating a bunch of things who/what/when/)
 
@mbomb007 I'm just saying I live in a part that was/is pro-European in the first place
 
@muddyfish Mmk. There were lots of people that voting "Brexit" to make a "statement", but didn't think it'd pass. They feel incredibly stupid after doing so.
 
7:35 PM
I know. I saw that a day or so after the vote
 
8:06 PM
-2
A: 100 Prisoners and a light bulb

BlacksilverC, 468 bytes #define LEADER 0 bool bir[100];bool light = false;int main(){int i,p;srandom(time(NULL));for(i=0;i<100;i++){start:p=(int)random()%100;if(p==LEADER){if(light==true){light=false;continue;}else{goto start;}}if(bir[p]==true){goto start;}else{if(light==true){goto start;}else{light=true;g...

Does my comment on this sound fair?
I've tried to be neutral, while still getting across that the answer doesn't match the challenge.
 
Also, the fact that there are more than 30 comments on that challenge speaks volumes about its clarity and readiness for main.
 
Yes I definitely think the challenge needs serious rewording, but the answer seems incorrect and ungolfed, even taking into account how unclear the challenge is.
 
Yes.
 
The challenge has been closed twice today, and each time it has had reopen votes before it has had a chance to be improved. I don't understand that.
 
Much like brexit, protest votes abound :P
 
8:18 PM
Ugh, I hate it when I'm working on a tough problem at work and I accidentally close my browser window, walk over to the cookie vendor, buy a cookie, eat it, and then reopen my browser window.
 
When competing in any form of programming competition, math test, whatever it is, the #1 rule is always to read the instructions carefully. They didn't.
 
@flawr That guy in the GIF is like "Agh, so that's why you paid me to do this, gah!!"
 
@flawr That's both genius and horrible.
 
@TimmyD Is there a way to move those comments to chat?
 
8:26 PM
I wrote a 130-byte Element answer today... TBH I'm not sure why I even bothered.
 
@DmitryKudriavtsev Once the comments are dealt with they can be deleted by the comment posters, or they can all be deleted by a mod. A mod can also move them to chat if they have potential to be interesting to future readers, or if further discussion is needed. If they just suggest an improvement that has now been made, there doesn't seem a need to preserve them though.
 
ok. so comments are inherently ephemeral?
 
Yes - they are intended to be. They often end up staying around indefinitely but they should be written taking into account that they could be deleted at any moment.
If a comment says something very relevant to the question or answer, it's worth editing the info into the post so it will still be available if ever comments get deleted
For that particular challenge, there is still a lot of improvement required before the wording will be clear, so the comments may not be moved or deleted until the challenge is ready.
 
So I had dead pixels. I run JScreenFix for 8 hours, and nothing happens.
Today, I force restart the computer and it's 100% gone.
 
8:44 PM
Is there no end to what can be fixed by turning it off and on again??
2
 
Most forms of life, I guess.
 
Yeah, but if you're having a bad day, people say to just go to bed and you'll feel better in the morning. That's the equivalent of turning it off and back on again.
 
True, but that probably actually works less often than it does for tech ;)
 
-1
Q: What's different?

OliverGiven two strings, print a change log of what's changed! Please print the added words in square brackets [] and the removed words in parentheses (). Examples: Input | Output ---------------------------------------------+------...

 
@trichoplax A lightbulb.
 
8:55 PM
True... :)
 
0
Q: Import this into your own language

boboquackThe Zen of Python must surely be one of the most beautiful Easter eggs in any language. (It is created by writing import this) So why not create a version for your own language? For those who don't have access to Python 2 or 3, your code should output the following: The Zen of [YOUR LANGUAGE],...

 
I just realized that an entire portion of my API is unnecessary, so I was able to get rid of it :)
 
@trichoplax Hand grenade?
 
I'm beginning to regret my comment :P
 
Oh, I was sorta ninja'd.
@trichoplax A relationship!
 
9:06 PM
[citation needed]
 
Anonymous
Hmm... That last challenge and a challenge I saw in the Sandbox got me thinking. Maybe we should have an off topic close reason of "this challenge gives an unfair advantage to certain users".
 
I'm staying well away from any comments on that...
 
@DmitryKudriavtsev ephemeral is such a cool word!
 
@Mego too specific, I think.
 
9:11 PM
I still don't know what it means, but it's just nice how it roll off your lips=)
 
@NewMainPosts This is far more a dupe of this than it is rickroll
 
Text-only KC challenges are really easy to dup without being a duplicate
but even then, "Happy Birthday" and rickrolling have repeated lyrics. Zen of Python has none
 
Wait, Python 3 doesn't have cmp?
 
Anonymous
@quartata Nope, it was removed in favor of key-based comparisons
 
Why
 
9:17 PM
Without repeated lyrics/words/phrases, it's just "who can compress this fastest to get FGITW votes". Which I'm sure we have a relevant dupe target for somewhere around here...
 
Anonymous
Because cmp sucked and key-based comparisons don't
 
@TonHospel @Dennis It turns out the running times are quite inconsistent even though I close all applications. I have run both of your code a few times and taken the fastest run. Which turns out to exactly equal once I use Ton Hospel's improved gcc flags
 
0
Q: Proposal for a new off-topic close reason

MegoThis sandboxed challenge and this revision of another challenge got me thinking. Clearly, we don't want challenges where certain users have an unfair advantage over others (such as the length of their username (especially since changing your username for a challenge is not allowed), or having bee...

 
+1 for nested parentheses in prose.
Too bad we don't have stats on how often this is an issue. I'm all for new close reasons if they'll get used often enough to justify it, but I'm not sure this one is so common that we can't just use the "custom" option.
 

« first day (2095 days earlier)      last day (2751 days later) »