and then I got this one because I also want to calculate the LCP: arxiv.org/pdf/1101.3448.pdf

I got source code too - sites.google.com/site/yuta256/sais

@aditsu the dc3 algorithm simply produces the lcp array too. See the end of cs.helsinki.fi/u/tpkarkka/publications/jacm05-revised.pdf

@Lembik I think SA-IS is faster though

yes! But do you care about speed?

if you can't actually implement it :)

dc3 is comprehensible :)

but it isn't the fastest method in practice you are right

@Lembik well I'm trying to fix that

:)

good luck!

so I guess you're not familiar with it..

what's this for

suffix arrays

Solving any particular problem?

well, finding the longest common substring

@Lembik Well otherwise you could have an unlimited alphabet, which are quite different cases. So I'd explicitly state which one you're aiming for in the challenge.

"Intuitively, …" [extremely non-intuitive stuff following]

@Lembik btw I was reading that too, it stops making sense about 1/4 of the way through

Hᴇʏ ᴀʟʟ

@flawr I think for any finite length string, the optimal number runs occur when the string is binary

@aditsu let me know where you get stuck and I can help

hi @feersum I am working on the problem of getting answers for my challenge

@Sherlock9 Hey part of all

I get stuck at completely different things in each document; anyway, I gotta refuel now, bbl

@aditsu ok but if you choose a document I have give you and get stuck please let me know :)

@feersum this one codegolf.stackexchange.com/questions/84592/…

I didn't understand that one

@Lembik I'm not looking into DC3, but only SA-IS

Yᴏᴜ ᴋɴᴏᴡ ᴡʜᴀᴛ I ᴊᴜsᴛ ʀᴇᴍᴇᴍʙᴇʀᴇᴅ? Dᴇᴀᴛʜ ɪɴ ᴛʜᴇ Dɪsᴄᴡᴏʀʟᴅ sᴇʀɪᴇs ᴛᴀʟᴋs ɪɴ sᴍᴀʟʟ ᴄᴀᴘs :D

@Lembik "I think" is usually not a satisfactory result in math

depends @flawr , aren't all conjectures without proofs "I think" results?

@Sherlock9 The book? Is it worth reading?

Yᴇᴀʜ, Dɪsᴄᴡᴏʀʟᴅ ɪs ᴀ ᴘʀᴇᴛᴛʏ ғᴜɴ sᴇʀɪᴇs

@Lause yes, and that's why they aren't satisfactory

zyabin101 did you get your answer about compressing the flag?

@Sherlock9 writing that in the notepad will have to go see my librarian !

@flawr well some of them are still considered "as good as true" if they have good evidence

@Lembik Well how can you be sure that your testcases actually are optimal?

It reminds me that I have to buy some bookshelves

I'm at 500+ Comic/Manga/Books, and it's pilling up in boxes atm..

@Lause well "as good as true" really is not satisfactory from a math poitn of view

@flawr you still can base other results off it, like in RH case

Sure, assuming it is true but all those efforts are in vain if it proves to be false.

@Katenkyo I ʀᴇᴄᴏᴍᴍᴇɴᴅ ᴘɪᴄᴋɪɴɢ ᴀ ʙᴏᴏᴋ ᴀғᴛᴇʀ ᴛʜᴇ ғɪʀsᴛ ᴛᴡᴏ ᴏʀ ᴛʜʀᴇᴇ ᴀᴛ ғɪʀsᴛ. Tʜᴇ ғɪʀsᴛ ᴛᴡᴏ ᴀʀᴇ ɴᴏᴛ ᴀ ɢᴏᴏᴅ ɪɴᴛʀᴏᴅᴜᴄᴛɪᴏɴ ᴛᴏ ᴛʜᴇ ʀᴇsᴛ. Iᴛ's ᴀ sᴇʀᴇɪs ᴏғ sᴇʟғ-ᴄᴏɴᴛᴀɪɴᴇᴅ sᴛᴏʀɪᴇs ᴀɴᴅ ᴛʜᴇʏ ɢᴇᴛ ʙᴇᴛᴛᴇʀ ʟᴀᴛᴇʀ.

Please no all caps.

@Lause So should we say P =/= NP is true until we find an evidence? ^^

@Katenkyo Tʜᴇʀᴇ ᴀʀᴇ ᴄᴏɴᴛɪɴᴜᴀᴛɪᴏɴs ʙᴜᴛ ᴛʜᴏsᴇ ᴀʀᴇ sᴘʀᴇᴀᴅ ᴏᴜᴛ

@flawr Tʜᴇsᴇ ᴀʀᴇ sᴍᴀʟʟ ᴄᴀᴘs, sɪʀ

@Katenkyo not true just more likely to be true than the other option

@Katenkyo would you dismiss a result only on the basis it asumed P=/= NP?

@Sherlock9 I will document myself before buying them in this case. Thanks for the advice of skipping the first few ones

Skip them at first

@Sherlock9 He's saying "Stop using small caps." (and I agree). They're hard to read.

I personally think the watch or witches storylines are better introduction than the rincewind ones

You can go back to them later. They're alright, but different from everything from about the third book on.

@El'endiaStarman Fair enough :D

@Sherlock9 So what

@Lause the thing is you can't disprove P =/= NP by saying I found a case where P = NP

Just being pedantic. Stopped

So no, I wouldn't

@Lause Absolutely. Guards, Guards!, Mort, and Wyrd Sisters are all interesting starting points IMO

@El'endiaStarman I find small caps far less annoying than caps, and as easy to read as lowercases. Only my 2 cents though

@Katenkyo Actually yes, you can. It's been proved that a P solution to any NP problem means that P solutions for all NP problems exists and can be constructed from the first P solution.

@Sherlock9 If I like this serie, I would eventually read them even if you told me to burn them ^^

@Katenkyo I never meant that. Only though if you had a result saying - Goldbach Conjecture is true if P=/=NP , it would be an interesting result, even though it depends on P vs NP

@Katenkyo Haha, fair enough

@Sherlock9 I also enjoyed Most von Lipwig's storylines very much. And the film version of going postal is very good.

@El'endiaStarman Forgot that it was a problem were we had to define P and NP as strictly different or stricrly equal, my bad

@Lause I've been reading the Discworld series in sequence anyway (against the advice I posted above haha) but I'm at Small Gods, so I am nowhere near Moist's introduction

@Lause The funny thing here, is that I wouldn't publish something like that if I found it because I would feel like it isn't complete, and for me, not complete= garbage

@Sherlock9 I do that as well now, but I had some moments in my life when I read them when they came out

I don't have a great mind, if I can't explain every little pieces, or at least understand them, I can't say what I discover is something worth attention

@Katenkyo Well I think that dependence of two problems can be an useful information as well

@Lause And I would say you're right, but that's how my ill brain works ^^'

@Lause One example of this is that Taniyama-Shimura conjecture proves Fermat's Last Theorem, which I find very interesting

Oh yeah that is way cool

@Sherlock9 Subtle nitpick: proving Taniyama-Shimura* proves Fermat's Last Theorem. The proof that they were equivalent is separate and came earlier.

We all get turns to be pedants! :D

I would be thoroughly suprised if a proof of NP =/= P => Goldbach came out though

@El'endiaStarman Subtle nitpick: they're not equivalent, Fermat is a corrolary of Taniyama-Shimura afaik

The implication 'true -> true' works both ways.

@Fatalize Oh, I got that wrong, then.

yeah Modularity => Fermat

"Andrew Wiles proved the modularity theorem for semistable elliptic curves, which was enough to imply Fermat's last theorem. Later, Christophe Breuil, Brian Conrad, Fred Diamond, and Richard Taylor extended Wiles' techniques to prove the full modularity theorem in 2001"

@feersum I more meant NP=/=P=>Goldbach without proving or disproving either of them

@feersum If that were the case, there wouldn't be so many statements like "this theorem is true, but not the converse".

@Lause What? I'm talking about the Fermat stuff.

Ah sorry

@El'endiaStarman Are you saying true->true is logically incorrect?

No, I'm saying that "If A, then B" being true does not mean "If B, then A" is also true.

True implies true, to be clear.

I'm saying that if A = true and B = true, then A implies B and B implies A.

If you proved FLT but not modularity , with our current knowledge you wouldn't be able to get modularity from FLT

@feersum So, if A is "a square is a rectangle" and B is "3 is a prime"...

@El'endiaStarman Yes.

Well the issue at hand seems that feersum is talking about values of logical statements (for which implications are actually correct) and El'endia is talking about the logical statements themselves, disjoined from their values for which the implications are nonsense

I'm not so sure you can treat true and false in logic like functions in higher-order functions

^^

> Man pages without a name section are as useful as refrigerators at the north pole.

@Sherlock9 0/10 doesn't cover the cases Chatgoat and Upgoat

Would not smile again

Katenkyo is sad. Draw him many smiles to cheer him up! :)

@zʏᴀʙiɴ101 I'm actually more than sad, I'm as salty as the dead sea because I have to do some Responsive WebDesign

:(

ohai @RenderSettings?

For extra fun:

Noncontradiction: "If all animals are goats, and no animals are goats, then Dennis has been outgolfed."

@LegionMammal978 Has Dennis ever been outgolfed?

@Sherlock9 Just using Dennis having been outgolfed as an example proposition; after all...

Hi.

@DestructibleWatermelon Hi.

@BusinessCat Hi.

Want to take more work on MtG creatures?

Remember how i was making that language? Well, I'm still making the spec, but I'm not sure how to implement repeatable control flow

well, "where" to implement it

Oh wait, idea in my mind suddenly

I LOVE my new Fantasque Sans Mono. \o/

really

Fonts are fun

I made a bad font once

I still use it though

Who's into Magic: the Gathering and has enough creativity?

(for making creature cards)

I have creativity, but I don't play M:tG

but i think its interesting

@zʏᴀʙiɴ101 I'm into MtG Turing machines but not much else

@LegionMammal978 The machine is long broken.

The Gatherer has no entries for "Artifical Evolution". :(

I think I've done some fair work on this spec

@DestructibleWatermelon Where are the FireMelon, AirMelon and EarthMelon?

@Katenkyo There don't exist such fruits. o_o But good joke.

Well, you see

those are not liquids

@zʏᴀʙiɴ101 They exist, they are the divine Elemelons

there do exist mercurymelons

and also gallium melons

but only in very tropical areas

(Sadly, it isn't mine, I think I've read it long time ago somewhere on the internet)

@DestructibleWatermelon How would you go about destroying a WaterMelon?

You can get rockmelons, from really hot places

@DestructibleWatermelon So, you wish to do creature cards for MtG, no?

@DestructibleWatermelon EarthMelon and AirMelon can be liquid in certain temperature and pressure. And FireMelon being composed of plasma, it is technically a liquid too :D

Air ceases to be air when liquid

@DestructibleWatermelon Would it be possible to create a HeliumMelon?

and earth, refers to soil

Helium is a superfluid or something

(Wow, I have 3 posts on the starboard!)

it acts too weird

@DestructibleWatermelon When cooled down to a liquid, that is

yep

So... any DryIceMelons?

oh wait, not all liquid helium is superfluid

dryicemelons would not exist (because it wouldn't be ice)

but perhaps CO2 melons

in reeeaaallly high pressure

there would be no one to enjoy it

Even more insane:

Watermelonmelons?

wait

you would need a blender

also, how high is the pressure for liquid co2

A dryicemelon would be a really cold airmelon, wouldn't it?

Speaking of elements: who wants to play Alchemy together in the Sandbox?

@zʏᴀʙiɴ101 ?

@LegionMammal978 littlealchemy.com

Oh, sure

@BusinessCat Ping me if you want to work on creature cards for MtG.

whoa

15 seconds in and i've already made an atomic bomb

wtf

also that is a great out of context post right there

There, just reset it

@flawr Hi.. you asked how can I be sure the test cases are optimal

@flawr I exhaustively tested all strings with 3 symbols. You are right I don't have proof that 4 wouldn't be better. This just means that an answerer may possibly find a clever solution

@feersum is there anything I can do to make it clearer?

little alchemy hint: don't combine anything with something that's underlined

@Lembik I dunno. I didn't try super hard to understand it.

@feersum OK. I am a bit worried that smart people aren't understanding my challenge so if there is anything I can do to clarify it please let me know

@feersum or just ask anything that is taking more than a minute to understand please

Which question is this, Lembik?

@Sherlock9 codegolf.stackexchange.com/questions/84592/…

please ask if anything is unclear!

Ohh. Yeah, I get the general idea, but I'm having trouble figuring out how to implement it

Doesn't help that I should be finishing my report for the end of term

Good luck, Lembik

@Sherlock9 thanks!

@Sherlock9 you may find I double the bounty in a few days :)

Duly noted

hmm, I wonder if I should contact one of the guys who invented SA-IS, he's from HK :p

@aditsu why not?!

    SortedSet<string>[] ssSsE_0 = {
        ssSsE__0[0][1], ssSsE__0[0][2], ssSsE__0[0][3], ssSsE__0[0][4], ssSsE__0[0][5], ssSsE__0[0][6], ssSsE__0[0][7],
        ssSsE__0[1][1], ssSsE__0[1][2], ssSsE__0[1][3], ssSsE__0[1][4], ssSsE__0[1][5], ssSsE__0[1][6], ssSsE__0[1][7], ssSsE__0[1][8], ssSsE__0[1][9],
        ssSsE__0[2][1], ssSsE__0[2][2], ssSsE__0[2][3], ssSsE__0[2][4], ssSsE__0[2][5], ssSsE__0[2][6],
        ssSsE__0[3][1], ssSsE__0[3][2], ssSsE__0[3][3], ssSsE__0[3][4] };

@LegionMammal978 In production, it would be the same thing, but with SortedSetOfStringWithAnNameHorriblyLongAndWithoutMeaning__0 instead of ssSse__0

@Katenkyo But yeah, I kind of use Hungarian notation with no actual variable names

So ssSsE_i would be a SortedSet<string>[] and ssSsE__i would be a SortedSet<string>[][]

@LegionMammal978 Yes, but why name it ssSsE in the first place? If you don't want to use a variable with a meaning, then just go for even shorter -> ss

@Katenkyo The entire thing is systematic

    class C1
    {
        public int i0, i1;
        public string s0, s1, s2;
    }

    class C2
    {
        public int i0;
        public string s0, s1;
    }

    static void Main()
    {
        bool b0 = true;
        DataContractJsonSerializer dcjs0 = new DataContractJsonSerializer(typeof(S0));
        int i0 = 0;
        List<C1> lSc1E0 = new List<C1>();

@Lembik The scoring is really confusing

35°C WTFFFFFFFF

hi

@LegionMammal978 When even abreviated name start being too long, I usually go for one or two letters -> d0 :D

@TùxCräftîñg And the effing humidity.

@TùxCräftîñg It's hella hot here too (Nice, France)

@Katenkyo (Nice, Portugal)

too ealry? =P

@Katenkyo It's not so hot here at home (Engels, Russia)

normal, you are in Russia ;_;

@Poke mon

@zʏᴀʙiɴ101 mon. you saw nothing

@Lembik I now have a running program, but only for the alphabet {0,1}

find ./ -type f -name *.js only return js files in the current directory and dont search recursively

@TùxCräftîñg -R?

@TùxCräftîñg Ew, gross. It was 36 c here yesterday.

@zʏᴀʙiɴ101 unknow predicate: -R

(I did have to Google Fahrenheit to Celsius)

._.

@flawr Ah.. what is confusing about it? It just a way to allow non-optimal answers to get points

@flawr awesome!

@flawr what can you get up to?

@DrGreenEggsandIronMan There are a few nice points you could memorize for the everyday usage: -18°C=0°F, then 0°C = Freezing point of Water = 32°F, 20°C = Room temperature = 68°F, 38°C = body temperature = 100°F, 100°C = boiling water = 212°F

@Lembik Pretty much as high as you want, as long as you wait long enough

It is just an exhausive search

the body temperature is 37°C AFAIK

@flawr Obligatory xkcd link

@TùxCräftîñg you're probably right, I just had in mind that 100°F was originaly chosen to match about the body temperature

ok

@TimmyD "223kg: your mom (also incl makeup)"

files=$(find ./ -type f)
for file in $files; do
    echo $(echo $file | grep *.js)
done

@Lembik It's about O(n^2 * 2^(n-1))

dont work .____________________.

@flawr ah so probably to about 25?

just needed to change grep *.js to egrep \.js$

depending one what language you have coded it in of course

@flawr I knew a couple of those. Like 0 == 32 and 100 == 212. Body temperature is 98.6 though.

@Lembik Matlab, n=10 already takes about 10 seconds

@flawr ah ok. This may be a situation where python is actually faster

n=12 about 20sec, n=13 about 40 sec

so 25 seems a little optimisitc

@Lembik matlab is just an interpreted scripting language, so it is quite slow

right.. like python I suppose

@TùxCräftîñg Do find . -name '*.js' -print

@flawr doesn't it have a JIT these days?

I mean a JIT compiler

@quartata thx :D

gci .|?{$_.Name-like"*.js*"}

very suboptimal

code-golf :p

In that case

ls -R *|grep \.js$

@Lembik it does, but it still is quite slow

PS C:\Tools\Scripts\golfing> grep
grep : The term 'grep' is not recognized as the name of a cmdlet, function, script file, or operable program. Check the spelling of the name, or if a path was
included, verify that the path is correct and try again.
At line:1 char:1
+ grep
+ ~~~~
    + CategoryInfo          : ObjectNotFound: (grep:String) [], CommandNotFoundException
    + FullyQualifiedErrorId : CommandNotFoundException

:p

weak

especially if you cannot vectorize your code

@Lembik So what happens if someone posts e.g. code that returns some strings for n=2...10. Then someone else posts acode that returns strings for n=2..15, but the ones for n=2...10 are slightly worse.

Who wins?

no-console is on by default in eslint ;_;

Dang, this is a great answer to that Iceland-flag challenge.

Wow, and I just watched the numbers tick up to a Mortarboard for that user.

Is the room free enough so I can post some request safely?

@flawr the n=2..10 person

So it isn't. Foils my plan.

@Lembik Oh ok

@Lembik patterns are cool

@Lembik still no answers I see

I attempted to write something today, failed miserably. On the bright side I now know RegEx backrefs and (?:non-)?capturing groups much better now

i love DuckDuckGo

nerd

;_;

i love Google

i just have to type ascii table to get the ascii table, directly in the search engine

my connection is slow. really

@TùxCräftîñg Bing does that, too.

casual, not memorizing the ascii table /s

;_;

anyway, the best search engine is DuckDuckGo.

and the best editor emacs.

@TùxCräftîñg you are my antichrist

google and vim

google ;_; seriously

I like google, I'm a google fanboy

also my job is kind of to use Google AdWords to advertise so

yknow

fs.exists is deprecated WTF ;_;

Just open it and catch the exception?

@TùxCräftîñg use this

function exists(path) {
  try {
    fs.accessSync(path, fs.F_OK);
  } catch(err) {
    return false;
  }
  return true;
}

thx

OMG 2 spaces at indentation ;_;_;_;

fiiine

                                        function exists(path) {
  try          {
                                  fs.accessSync(path,                 fs.F_OK);
     } catch                                       (err) {
         return false;
    }
                                  return true;
          }

@TùxCräftîñg What's wrong with that?

better?

._.

but WTF

@TùxCräftîñg how so? I personally don't have a preference, my text editor just chose 2 spaces

lol ... support ticket gets escalated "Why is it still so slow uploading these dozen-MB files? We upgraded from a single to a pair of T1s last week."

@MatthewRoh I'll give it a shot with node.js

I mean

can I actually just unzip the file and look at the new file size?

literally the worst bomb detector since it detonates the bomb lol

@charredgrass try with 42.zip

hey, you said to golf it, not to make it work well

where do you say I can't detonate the bomb?

I'll add this to comments so others can discuss

also what mbomb said

rip, probably will be close-voted

-4 votes and 5 close votes in less than 5mn ;_;

7 more rep until I can see close votes :P

^^

> I don't think this is a frequent enough issue to warrant using one of the custom close reason slots. But I do think it's a big enough issue to warrant not only a downvote and a close vote but also a delete vote

> Subscriber represents, warrants and agrees that it will not contribute any Subscriber Content that... (e) contains a virus, trojan horse, worm, time bomb or other computer programming routine or engine that is intended to damage, detrimentally interfere with, surreptitiously intercept or expropriate any system, data or information, or (f) remains posted after Subscriber has been notified that such Subscriber Content violates any of sections (a) to (e) of this sentence.

no answers ;_;

@TùxCräftîñg I will totally attempt that tomorrow morning when less drunk

for now, cya

Is this code-golf'-ish?

We do have a fastest-code which kinda fits what you're describing

Ah.... almost

fastest-relative-code (javascript to c) is pretty much the idea.

Still .... would you like a challenge like that?

I'm just scared that it might get down-voted instantly by people who might think it doesn't belong

Well sandbox it and see what people think. I can't really tell what it would be like without a more fleshed out challenge spec

Ok cool

@RuHasha its a bit weird though.

so people have to kind of levarage on bad things in C which are good in JS ?

I guess... The idea behind it comes from people saying dynamic languages are slow. So naturally the question is "what sort of things are less slow"

yup

so they write all answers in 2 languages? C and JS ?

Yes that would be the plan.

But what would make it tricky is that someone could optimize another persons C implementation, changing their score.

hmm.. try it out in the sandbox to see the initial response/suggestions

@MartinEnder Is there a way in Retina to do a Replace with overlapping matches for each one? Or append a result for each overlapping match?

retina.tryitonline.net/…

@TùxCräftîñg Want answers, get 125 rep for a bounty.

vimeo.com/7133810

(An effective amount of 125 rep, which is two components: 75 rep is the "put bounties" privilege, plus 50 for the minimal bounty.)

^^ this men is a fcking genious

.______.

My project is dead ._.

@zʏᴀʙiɴ101 All your projects seem to be of the form "do it for me please"

What? I don't see any project being in any such form.

There must be downstars, right here, right now!

downstars ._.

@Fatalize I really don't see any of my projects being in such form.

You just think that every my project is annoying, just like me, and should exit chat forever, sooner or later, just like me.

Everyone thinks I'm a spammer in this room.

@zʏᴀʙiɴ101 I just think people would be way more interested in your projects if you present something that is more than just "here's an idea"

which in itself is not bad at all, giving ideas is always good

but you always repeat "Who's interested to do this" for quite some time, wtihout working on it by yourself first

@Fatalize I don't have enough creativity for this project.

@Fatalize I did prepare a base card list for my MtG set, just that I didn't release it.

@zʏᴀʙiɴ101 That's what you believe but if you try you can do those projects by yourself, even if incomplete

When I decided to rewrite all of Brachylog in Prolog, I didn't ask "who's interested in doing that?", even though I wasn't confident at all I would be able to do it

But I started and it worked

And now a few people got interested

I'm not hating on you (nor do I think anyone here is), but you need to believe in yourself a bit more

If you come up with all those project ideas, you probably have enough creativity to do them

@zʏᴀʙiɴ101 and ultimately getting things done by yourself feels great, even if no one cares about what you've done

and you will not feel bad because people are not interested in what you want to do

@TùxCräftîñg Yep

@zʏᴀʙiɴ101 TL;DR believe in yourself