« first day (1892 days earlier)      last day (2954 days later) » 

2:00 PM
@Lembik It's a full blown Ubuntu container with sudo enabled. Travis gives you Ubuntu or OS X and AppVeyor gives you Windows Server.
 
@Lembik Travis is a CI though - it tests if things succeed, so that means no interactivity
 
CI?
 
@Lembik Continuous integration.
 
@MarsUltor Well you can output whatever you want to the log, which is just a text file. Input is provided by the test commands (which can be any bash command).
 
I am learning a lot :)
 
2:02 PM
Basically, runs every build to tell you if anything is wrong. It can be integrated with GitHub as well to allow/disallow pull requests
 
have any ppcg challenges used it yet?
 
afaik not
 
@Lembik No
 
as this is all new to me I am not sure I should be the first
but I would love to copy what someone else does :)
 
@mınxomaτ I know, but I didn't say you couldn't. I just meant no c9/Docker-style interactivity
 
2:03 PM
unless someone wants to copy my challenge it
I would happily help them
 
Yeah, I'm not sure how people would commit
unless someone ran a bot
 
of course someone needs to actually submit an answer too :)
1 hour and no answers....
 
@MarsUltor They fork the tester repo add their code and send a PR.
You don't even have to merge it to have it run.
 
is there also free gpu usage online?
 
@mınxomaτ Multiple languages may be a problem though
 
2:06 PM
@MarsUltor No. Just set the language to blank and everone adds their installation command to the install script. This way no one can complain about missing languages / compilers / packages etc.
 
@Lembik Fastest code, big code challenges, and pop-con may actually take several hours before one wants to post his answer
No one want to post something he's not proud of
 
@Katenkyo you make a good point
 
Finding an algorythm and a data-structure as performant as possible can be really hard
moreover, fastest code will often be win by C submissions (because C is fast), everybody knows that, but not everybody is fluent in C
 
@Katenkyo True.
@Katenkyo that's not so true about C
 
So be patient, you made a good challenge, there will be an answer :)
 
2:09 PM
@Katenkyo thanks :)
one of my trickier challenges was won by a nim answer
I mean fastest-code challenges
and another by Java
 
"often" ^^
 
the point being that it's just painful to write complicated code in C
although it would be faster if you had the energy.. of course fortran is faster still :)
 
But it will be faster, because compiled in less ASM instructions than higher level languages
 
I am disappointed by how few fortran answers we have
 
Not everybody knows it ^^
 
2:11 PM
I know! :)
but also not everyone knows jelly
and yet we have plenty of answers in jelly :)
 
ven
nobody knows Jelly :P
 
:)
 
@Katenkyo That is an impossible statement. Is always depends on the program. Loop unrolling creates massive instruction lists than stress your L-Cache (gcc's -O3 and funroll-loops e.g.).
 
ven
in Jelly, 2 days ago, by Dennis
I don't understand my language. ._.
 
Jelly was designed here for use here, though. Of course it will be overrepresented here ;)
 
2:12 PM
@mınxomaτ what statement was that in response to?
@Geobits :)
 
@mınxomaτ That's why I'm speaking of chosing a language, and not about an algorythm or a data structure
 
@Lembik Just hover over the message.
 
@mınxomaτ oh I see.. the problem with optimized compilation of C is pointers afaik. This is why fortran can often be better optimized
 
@Lembik but if someone doesn't want to painfully develop in C, he won't chose fortran/ASM either ^^
 
I believe that fortran is pretty much the only language that was specifically designed to be easy for a compiler to optmize
 
2:15 PM
@Lembik Not really, pointers always exist Wether the compiler adds it or not. The Intel Fortran and C++ compilers don't have that much of a performance difference. Both blow other compilers out of the water.
@Katenkyo ASM is only painful if you use it wrong.
@Katenkyo It's algorithm. But "because compiled in less ASM instructions than higher level languages" is so impossible, it's not even wrong ;-)
 
@mınxomaτ I don't get your point about pointers. If the compiler can't work out what the code is doing it can't optimize it well. That's essentially the problem
 
@Lembik Fortran is basically the perfect mashup of "language so old that classes are starting to move away from it" and "language so verbose that if you tried to golf it it wouldn't look like you're trying very hard"
 
@mınxomaτ so I may be using it really wrong :)
 
@Sp3000 :)
 
@mınxomaτ How much instructions are wasted for a simple i=i+1 in Java compared to C? or worst i++!
 
2:18 PM
in any case.. nim rocks :)
 
looks at FORTRAN code
 
@Katenkyo Here are five books that explain how and when to use assembly and when it doesn't make any sense: agner.org/optimize (I wrote one, too, but it's in German :P)
 
I didn't say it is true in all situations, but most instructions in C will compiled in less ASM instructions than Higher level languages
 
someone just needs to create a FORT language (like MATL vs MATLAB)
 
263
Q: Is Fortran faster than C?

quinmarsFrom time to time I read that Fortran is or can be faster then C for heavy calculations. Is that really true? I must admit that I hardly know Fortran, but the Fortran code I have seen so far did not show that the language has features that C doesn't have. If it is true, please tell me why. Pleas...

that's a more accurate answer than I gave
 
2:19 PM
@mınxomaτ Even though it is oriented about C++ optimisation (I hate this language...), I'll bookmark it, it may be interesting, thanks a lot.
 
@Katenkyo FEWER not LESS :)
 
@Katenkyo ASM is always used (or it should) for optimization, never as a standalone language. Those books assume that you know that.
 
Anyone know of a way to run bash on windows? (Without VM)
 
@Lembik Can't edit, thanks for pointing it out
 
@GamrCorps It's coming soon as an update.
 
2:20 PM
@Katenkyo my pleasure :)
 
@mınxomaτ really?
 
@GamrCorps Yes. Really. winbeta.org/news/…
 
@GamrCorps Sourcetree comes with git bash.
 
@GamrCorps Or use cygwin bash or this
 
But until then you can use win-bash.sourceforge.net
 
2:22 PM
@mınxomaτ I never had to go into so much runtime that I have to exploit some ASM for optimization. As I said, I may be wrong about ASM as it is painful to me :)
 
Thank to all of you guys, Ill check those out
 
@mınxomaτ Turtled
 
@mınxomaτ That's the first time Microsoft Hyped me...
 
cygwin bash is slow. I prefer git bash
 
I prefer Linux zsh :)
 
2:31 PM
there is git bash?
 
btw, salut
 
ven
coucou
 
so, I recently upgraded my old phone to a new old phone :)
I'm pretty happy about it
unlike most new phones, it runs android 6
 
@aditsu the important question is what version will it run in 12 months time
i.e. will they actually upgrade the kernel
and... can you code for it? :)
 
in 12 months it will probably run android 7
 
2:38 PM
@ven A fellow frog?
 
ven
@Katenkyo hon hon baguette
 
@aditsu you mean they will give you OTA upgrades?
most networks/phones don't do that
 
I don't have experience with android, but I could probably work something out
 
which is why android is a disaster
 
@ven We're invading ppcg.lol slowly! :D
 
2:39 PM
@Lembik it's a Nexus 5
 
ah then you will be good :)
great choice
how much was it?
 
2nd hand, about US$100
 
that's a great price!
I want one :)
 
come to HK and you can probably get one
 
I thought the US was now the home of cheap electronics?
 
2:41 PM
@aditsu Hell Kitchen? :3
 
having overtaken HK
 
Hong Kong
 
:(
 
@Lembik "not" or "now"?
 
oops
so.. what is the best phone for under 300 dollars these days?
 
2:42 PM
maybe you can find something similar on ebay, I don't know
 
I am not sure I trust ebay any more
 
but my phone comes with warranty from the shop
somebody recommended me a Zuk Z1
 
@Lembik Android is hardly a disaster ;)
 
@Lembik Nokia 3310 :)
 
@Geobits I think for security it is
@Geobits I mean it's what I use.. but I am aware that I am taking a big risk
@Katenkyo good call!
the security problem is simple.. people, like me, run versions of the kernel that are 12months+ old
without any of the patches.. because we are not offered a kernel upgrade
in fact almost everyone does
 
2:45 PM
I weigh the risks based on experience. Neither I nor anyone I know has ever had their phone exploited due to a security flaw, and I know a lot of people with androids.
 
That's almost entirely the fault of the network operators ... who shouldn't even be in the middle. I don't need Comcast weighing in on a Windows update, so why does Verizon get to weigh in on a cell phone update?
 
@Geobits this is a good point so really it depends on your appetite for risk
@TimmyD yes
 
@Lembik My kernel is 30 months+ old :D
 
@Katenkyo :)
 
let me see what kernel I had on my old phone :p
 
2:46 PM
the latest version of android for the galaxy s4 is 5.0.1 for example
there just is no more recent version
 
Actually, if you don't do risky things with your phone, you don't have to care a lot about the potential breaches
 
this ^
 
2.6.32
 
@Katenkyo I am not sure that is 100% true
 
there were some media-related exploits
 
2:48 PM
Even on an outdated kernel, it's more secure than most desktops :P
 
That is, unless you're scared about a stranger see the message you send to your wife, but again, it's a pretty small things and won't harm you even if someone who don't know you and who you don't know have them
 
kernel version 3.4.0-4760392 whatever that means
@Katenkyo I am worried about people reading and writing my emails
as my phone has full access to my gmail account
which includes lots of personal details in the X GB of emails
 
@Lembik That's a thing I call "risky"
 
but we all use our smartphones to read our emails
and no one types in their password each time they want to do that
 
But even here, it's still safer than having your account info on a USB-storage, right?
 
2:49 PM
not really. I mean how are you going to remotely hack into usb storage?
 
I don't read my email from my phone, so there's not risk about that
 
@Katenkyo ah :) hence the nokia 3310
 
I don't pay or anything with it
 
I am not so worried by the paying part
for some reason
 
I read emails from my phone, but only I do. Nobody's hacked into it.
 
2:51 PM
Security, like most things in life, is a compromise. Here, you're compromising risk vs convenience.
 
@TimmyD well.. I wouldn't mind if my phone upgraded itself every now and then!
that wouldn't be inconvenient
 
@Lembik If your phone is shutted down, how are you going to remotely hack it? :)
 
@Katenkyo I am not sure I understand what you mean
 
A phone contains encrypted datas, a USB does not
 
@Lembik If you care about security, be proactive and update it yourself. It's not very hard.
 
2:52 PM
@Geobits it's not possible!
there is no later android version
for the galaxy s4
 
@Lembik Please tell me this phone model that is impossible to root/upgrade.
 
see above
 
I guess Geobits is telling you to install a custom ROM
 
ah...
 
@Lembik If you're USB-storage is connected to one of your computers, it becomes possible to remotely hack it using your pC
 
2:53 PM
I can't say I am very confident about what you get if you do thta
@Katenkyo true
 
@Lembik You get the exact same thing as if you don't
 
@Geobits thanks.. I know about that but I need to know a person who has done it first to tell what you really get
 
Just be care, it's like installing a linux: you could destroy your phone ;)
 
@Katenkyo I will buy a google phone instead next time. It's the easiest solution
 
2:54 PM
Meh, I've never bricked anything. Most have very simple step-by-step instructions
@Lembik You get the latest updates (assuming you install the latest), exactly what you're saying you want.
 
Rooting your phone to install a custom ROM is defined as "very hard" for the average person.
 
@TimmyD But we here are a bunch of guys who hacks around with dark features of obscure languages, it should be good enough ^^
 
@TimmyD For the average person, I'd say that's true. For the average person here, I wouldn't.
 
@Geobits I used to read the forums for these things and they always listed weird bugs.
@Geobits is it all very safe and simple these days?
bugs like bluetooth not working
 
@Lembik Android Always had weird bugs
 
2:55 PM
It's a lot simpler than it was with my old HTC magic for damn sure :P
 
:)
 
I rooted my old phone and my tablet, not the Nexus 5 yet; also, never installed a custom ROM
 
2 hours and no answer!!
 
I won't say it's 100% safe, but I will say I've rooted 20+ phones/tablets with no bricks.
 
Mine will not ring/vibrate on incoming message if the message app' is open
 
2:56 PM
@Geobits how many were yours? :)
there are lots of simple things I don't get about android :)
 
If you open the app, it opens the last conv' you had, if you want to chose an other, you have to close it again, then re-open it :)
(it's one of many things :D)
 
ah!
 
@Lembik Umm... HTC Magic, Galaxy S, Galaxy S2, Xoom, Galaxy Nexus, Nexus 5... I think those are the only ones of mine I did.
 
@Lembik no answer to what?
 
2:57 PM
@Geobits Ah, I didn't realize you were specifically talking here. I thought we were still speaking in the abstract population-at-large.
 
oh, a challenge; here is my chance to break the threshold
ah, fastest-code, hmm
 
The only android I've had a real issue rooting recently was a Kindle Fire HD 7", and that's only because it's barely an android at all.
It made me even more sure that I won't ever buy an Amazon device.
 
Kindle Fire is to Android as TimmyD is to a developer.
 
@TimmyD Something strange and without a defined shape?
 
Basically :P
 
3:00 PM
@Lembik looks like too much work :p
I'll do another one instead
 
That entire challenge looks like it is intended as code for "Go away, Geobits".
;)
 
damn, that Jelly is quite something
 
I felt the same, math are a field I hate to see placed in theory...
Despite I'm fine applied math fields (Electrotechnics, Mechanics, Electronics...)
 
3:25 PM
@Doorknob How would you filter those whose length <2 ?
 
in Pyth?
 
yep
 
f<lT2
 
no # for me? :(
 
well, you have to take the length first, which means you can't use #
 
3:28 PM
ok
 
oh, f!PT is golfier
 
wow
does 05AB1E have a:rep(5)?
 
@orlp I frequently see these edits (which change a gif into a link) (and I appreciate). Is this mods doing it, or did you make the edit?
 
@orlp "oh my god I'm gonna drown you" :p
 
3:34 PM
@NathanMerrill I did
I generally display a gif for like 10-30 sec
then edit it
 
ah cool.
well, thank you :)
 
that way people get to see it
 
I don't like watching it on repeat
yeah
 
but are not forced to have flashy stuff on their screen for minutes
 
I think I just saw the loading gif
 
3:38 PM
I posted to CR for my second time.
They have my soul.
 
I'm not sure what the point is yet
 
Your soul is going to end up eminently readable
@NathanMerrill Takes a while to grow...
 
oooh, you can change the color of the lights
 
@himarmjr what did you do to your username
 
bah, this is worse than a clicker game
 
3:40 PM
@Downgoat chhanged it
Is explain xkcd down for anyone else?
 
@himarmjr yes
 
@Downgoat after a conversation with @Himarm on scifi chat.
 
@himarmjr Only the gallery for me
 
@himarmjr why no capitalization
 
funness
himarm has a record of sucky spelling
 
3:42 PM
@NathanMerrill You can makes the lights wide or narrow beam too...
 
yeah, but that was more obvious
the little circle wasn't at first
 
OK - I found the colours before the beam width...
 
I didn't realise the triangle controlled both direction and width
 
@himarmjr I never thought I would be so bothered by capitalization ._.
 
3:45 PM
Image from origin 'http://xkcd.com' has been blocked from loading by Cross-Origin Resource Sharing policy: No 'Access-Control-Allow-Origin' header is present on the requested resource. Origin 'http://www.xkcd.com' is therefore not allowed access.
 
there's a person in my garden
I have no idea what they are doing there
 
@NathanMerrill My lights grew a pillar (perhaps a bird bath?)
 
@himarmjr Works fine for me, however the styling is broken.
 
does your pillar have a circle around it when you click on it?
it feels like I should be able to control him, but I can't
 
@NathanMerrill I lost that garden due to being away from the tab for too long, but I think everything is selectable (and deletable...) even smaller parts of large objects like trees
@NathanMerrill As far as I can see all you can do is choose the lighting and delete things (like pruning a garden, I guess...)
 
3:48 PM
@trichoplax which reminds me: I can't stand how chrome dumps memory
 
@NathanMerrill Is that what kills my garden? I'm using Firefox and I've tried opening it in a separate window now rather than just a tab, to see if that stops it being lost
 
@NathanMerrill -1 ... focusing three beams of light on a tree doesn't cause tree to combust
 
@TimmyD Have you tried putting them all to maximum colour temperature...?
(I haven't - I can't promise anything...)
 
Doesn't seem to make a difference.
 
related: xkcd.com/1603
 
3:50 PM
@trichoplax yes. If chrome detects you are low on memory, it'll dump the memory of the last-used tab
lol I got an octopus
 
The most recently used tab rather than the least recently used tab??
 
least recently used
 
Ah OK. That's better. But still annoying that it doesn't save it to disk
 
hurray, I have over 100000000000000 reputation now! \o/
 
4:04 PM
-3
Q: Minimum Permutations for a polindrome

DaniBetCan someone help me with an algorithm which displays the minimum permutations of letters to become a polindrome?

 
oh well, there are 10 types of people in the world...
 
@aditsu Reminds me of a joke
> There are 10 types of people in the world. Those who understand binary, those who don't, and those who think this joke only makes sense in base-2.
 
4:09 PM
@TimmyD That's 11 types. As I remember correctly, the most popular interpretation doesn't have the "this joke only makes sense in base-2" part.
 
it's obviously using base 3
and looks like you're in the last category :p
 
How can I find a list of chat stars by a specific user?
 
user stars isn't public data
 
@NathanMerrill I mean, messages by a specific user that were starred.
 
Does anyone know how to fix left-recursion?
 
@GraceNote o7
(It's supposed to be a salute)
 
Welcome back @GraceNote. You are always welcome here.
 
Thanks. And yeah I can recognize a salute.
When I was in high school, my modern european history teacher was a First Lieutenant in Finland. So the goon squad (which was myself and two other students) got into the habit of saluting him whenever we passed him in the hall.
 
Was looking through the code of that JS newsreader and I noticed the default feed is Digg. wow this is old lmao
 
4:45 PM
hrm, I got 16GB of RAM sitting on my desk, wish I could plug them into the motherboard without rebooting (and ideally without opening the case too)
 
oooh, adding another CPU would be nice too
 
@zyabin101 the data explorer might help? I'm not really sure if it can query chat messages
 
@aditsu I don't think so.
 
Heya, Grace Note. What brings you by?
 
5:00 PM
it was one of my favorite challenges, but I don't remember getting passive rep from it
 
imagine taking the site and leaving all the other text as it is, but removing all reputation and votes
 
♫nothing to kill, or die for♫
 
it's easy if you try
 
5:16 PM
0.857 rep per day from one answer. Up until the question was closed as off topic...
 
5:32 PM
are you gonna edit that statement daily or count the rate period end with the closing date?
 
music is good but video kinda sucks
 
That's not the point
 
that's not the byte
 
@quartata I'm not sure if you are rick rolling
 
5:35 PM
It's the Unikong tune
They didn't actually make it which makes me sad
12
A: Is the Rick Astley chiptune available to download anywhere?

Adam LearAs I mentioned in the comments, we didn't make this particular rendition of the timeless classic. The credit (also listed in the game itself) belongs to Toast024 on Youtube. We've edited it to create a version better suited to looping indefinitely, but here's the tune itself: https://www.youtube...

 
-7
Q: If you can find three positive integers k1,k2and k3 as below I shall disprove FLT

Unnikrishnan Thekkeppat If any one could find three positive integers k1,k2 and k3 as attached equation, then I shall disprove FLT. Other wise if there is no such equation exist, FLT is true by 3 page proof.

 
pfffft
 
very downvote, wow
 
I just re-read it and I wish I hadn't recommended Mathematics SE
mm whatever
 
why not? that seems to be the right place for it
the other 2 suggestions, on the other hand...
 
5:45 PM
So... are we saying we don't want proof of FTL (either way)? Harsh, man. Harsh.
 
FTL is known to be impossible :p
 
FLT != FTL
 
Fermat's Last Theorem? Fermat's Little Theorem? First Law of Thermodynamics? ... Disproving any one of those would be incredible.
 
It is an actual YouTube channel. It posts YouTube test videos every 20 minutes or so.
 
5:47 PM
Fluffy Little Things!
 
@NathanMerrill I anagram things. So maybe not ==, but ~=.
 
Your question "Andrew Wiles is an idiot, hahaha" has been put on hold as off-topic
5
 
Basically
 
@MitchSchwartz Which question?
 
Well he can have fun searching that haystack
@zyabin101 The one above
 
5:49 PM
@MitchSchwartz I didn't know that this was asked.
 
@zyabin101 andrew wiles was the guy who proved fermat's last theorem
 
Sir Andrew John Wiles KBE FRS (born 11 April 1953) is a British mathematician and a Royal Society Research Professor at the University of Oxford, specialising in number theory. He is most notable for proving Fermat's Last Theorem, for which he received the 2016 Abel Prize. Wiles has received numerous other honours. == Education and early life == Wiles was born in 1953 in Cambridge, England, the son of Maurice Frank Wiles (1923–2005), the Regius Professor of Divinity at the University of Oxford, and Patricia Wiles (née Mowll). His father worked as the Chaplain at Ridley Hall, Cambridge, for the...
ninja'd
 
The proof is not light reading to say the least
Well, maybe to some people
 
After several hundred years and thousands of failed proofs, I would be flabbergasted if it were light reading.
 
Well, it would be light reading if the abc conjecture is true.
 
5:52 PM
it's not supposed to take much more than the page margin
8
 
hahahaha
 
> the Annals of Mathematics
I read that wrong ._.
 
I was more impressed by that Perelman guy
 
> Then, seemingly as an afterthought, he noted that that meant that Fermat's last theorem was true. Q.E.D.
Wow.
"Here's a recipe for making a really delicious cake. Oh, and, no big deal, it also shows P=NP."
 
Never mind that the reason Wiles went into mathematics in the first was FLT. It only took him proving several other theorems about elliptic curves and modular forms to come up with "Owen Bytheway, this proves Fermat's Last Theorem. Q.E.D."
 
5:59 PM
> Owen Bytheway
4
I like that.
 
hey I know that guy
 

« first day (1892 days earlier)      last day (2954 days later) »