>>28801
lolkwut

>>28802
He used to have a girlfriend who was a C# developer and she taught him everything. He's now employed doing PHP, Node.js, and JS development, but the girlfriend has left to work in Finland. It's okay though, because he uses the fat ducats from his Node.js work to pay for the pussy he crushes every week.

Well done, Mr. Backlelure.

>>28801

> go KKK every weeks and have sex with premium girls.

I would never have considered the Ku Klux Klan to be the place to go to pick up chicks.

>>28808
I think this is an american/european culture misunderstanding. KKK must have a different meaning where Beatrice is from.

>>28857
orly?

The reasons depend greatly on the language in use, but in general the reason is this: The coding style denotes that capitalized variables are constants.

In some languages this is only a guideline (i.e. it's not enforced by the runtime), but in other languages capitalized variables are how constants are created.

In other cases, it can just be an influence some coders have because of their work with other languages.

http://en.wikipedia.org/wiki/CamelCase
It's popular because it allows you to identify words more easily. I'm not really sure why people don't just use the underscore character instead, just tradition probably.

http://en.wikipedia.org/wiki/CamelCase

>that's just how it's done

There are no secrets answer here. Because, well, that's how it's done.
You can do whatever you want. But remember, someone is going to read your code. It's better if both of you have the same standards.

Java style:
FirstSecond for classes
firstSecond for variables and methods
FIRST_SECOND for constants and enum

I've seen:
first_second
firstsecond
first-second

Comment too long. Click here to view the full text.

I just leave them lowercase in my C program. I have functions like
list * newlist();
void appendtoken(list * l, char * t)

etc.

>>28836
*encoding

Pretty stoned when I posted that, but yeah that's basically it.

Server side could be something like:

money = 4052
wins = 3

Then there'd be some thing that'd take those variables, combine them into one generated string like '5Hzuo6pl" and you could enter that on the site to continue a session. The key would be forever re-usable by anyone. So the session isn't being stored, it's being generated by the key.

>>28838
So long as you don't care about making it unique and the information you want to encode is well defined, this is straight forward. If you have three 16-bit values you want to save, just stick them together, convert to base64, calculate some kind of checksum, and stick that on the end to prevent cheating.

I don't know ruby, but I'm sure all of the heavy lifting has been done for you.

https://en.wikipedia.org/wiki/HTTP_cookie

You may not even need a gem. If you want the session to be usable by anyone, just encode it into the url. Instead of trying to store the data the variables are holding, store the data for the session on your server, give it a unique id, and then put the id into the url.

>>28845
When you use Eclipse it'll make a debug keystore for you, wich it automatically authenticates. Only when you export the production apk you'll have to select your own key, passwords.

If you build in debug mode, it uses the debug keystore automatically. If you want to automate release builds, you'll have to use eclipse or ant, as the previous poster said.

>>28787

>it takes less time

Depends on the time constant associated with each algorithm.
Depends on the overhead associated with each algorithm.
Depends on the data size.

Big O is highly theoretical anyway, OP question is dumb. I said the same thing on a similar exercice in /math/.

>>28803
Theoretical yes, but also very applicable.
I'm sure you do know, but in case someone does not: If some algorithm has lower big O than another, then no matter what time constant or what overhead there is for either of the algorithms, the one with lower big O will be faster always when the data size is above some limit. This limit is rarely too large to completely disregard big O in the common case, but in special cases data sizes may be small enough for a 'worse' algorithm to perform better.

To op's questions:

>100n^2 vs. 2^n

I would just bracket the value. I'm not sure if there is a more efficient way to do that but I can't come up with one atm.

>where does big O notation come in?

I kinda said it earlier, but to make it more succinct (more or less straight mathematical definition with squiggles translated into English):
Function f(n) belongs to O(g(n)) iff there exists positive values m and k so that f(n) <= k*g(n) whenever n > m.

So for example
if f(n) = 100*n^2 and g(n) = 2^n
we could test k = 1 and m = 100:
100*100^2 <= 1*2^100 <=> 1000000 <= 1267650600228229401496703205376

Comment too long. Click here to view the full text.

>2logn - nlog2 < -2

so that's when I said "fuck you"

>>28803

>Big O is highly theoretical anyway

Only in a world where you don't give a shit about how long anything takes or how much memory it takes.

>Big O is highly theoretical anyway

That statement is such horseshit.

While yes, an algorithm with a worse big O can outperform a better big O algorithm on a small data set, if your data set is small who gives a fuck about how fast the algorithm is anyway. Also, almost any non-trivial data set is going to be big enough to shit all over your shitty x^n garbage.

>>28830
I last used it in the late '90s. I'm pretty sure that it was stand-alone at the time, though I could be wrong.

>>28832
That said, my opinion on the subject is probably not worth as much as it once was. nb

Start with Python or C.

It's best to pick up a book on programming Python or C. Then pickup a Linux machine or install one with VirtualBox or VMWare. Compile all your shits on Linux, and you will have a good time.

I know it's going to be a pain in the ass to setup, but it's well worth it in the long run. If you just want to learn scripting, use this book with Linux:

http://www.amazon.com/Practical-Commands-Editors-Programming-Edition/dp/013308504X/ref=sr_1_2?ie=UTF8&qid=1371537907&sr=8-2

Since you're just starting out, all this syntax is going to be foreign as fuck. Give it time to sink in, and most of this will be straight-forward.

>>28815
Agreed with this guy. Batch is pretty much useless and can't even be considered real programming. Python has beautiful, neat syntax and it's very easy to pick up. Also, consider starting your journey by watching some tutorial videos instead of reading massive manuals which tend to be more confusing than helpful (at least when you're starting out). It will also help you if have a specific thing you'd like to code, it gives you a chance to code something you're actually interested in, so it's more fun and you'll be more motivated to work on it so you'll improve faster.

In any case, I wish you the best of luck and I hope you'll stick to it. Coding can sometimes be a bitch (I just started learning C and C++, didn't sleep in the last 24 hours, spent the last 10 hours coding something trivial and troubleshooting compiler and external library issues, and I'm still confused as fuck), but it can also be one of the most rewarding things you can do with your time. Most code starts as something simple, just to make a specific task easier. But then you keep improving that code, adding and changing tiny bits, until it evolves so much that it almost has a life of its own. It's no longer just a bunch of code, it becomes a world unto itself.

I'll hope I'll experience some of the worlds you create in the future :)

>>28842
dude thank you so much. I googled that book and found a link to a torrent file with that and about 30 other books about linux programming.

Alright, got a question.

I've set some session variables in the main program, but there is seemingly no way to access them from another object.

Example:

[p]
class Tester
def test_method
return sessions[:testvar]
end
end

configure do
enable :sessions

Comment too long. Click here to view the full text.

hmmm

class Tester
  def test_method
    return sessions[:testvar]
  end
end

configure do
  enable :sessions
end

get '/' do
  sessions[:testvar] = "hey"

Comment too long. Click here to view the full text.

Might as well not make a new thread for this.

Is there any way to save things on Heroku? I guess it's read-only, so while I can make a number/stat save on it, it later goes away as it's never been written. Do I use a database for this, or is it just not possible with Heroku?

>>28829
Push it into a NoSQL datastore. Heroku includes a few options for you to choose from.

btw, tl;dr is after the break
nb

This won't probably work but think about how easy you actually have it, for at least half the people that graduate from CS it means debts hard studying and maybe other personal problems but once they're out, working and making money feels like heaven. Now think about countries like spain where engineers are delivering the newspaper because it's so hard to get a job.

Or if you're just bored find a personal hobby to distract yourself from work, so that when you come back you feel more relaxed and willing to work.

>>28800
Is it okay to know that in some way I'm taking advantage of this job's flexibility? I'll probably stay here for another 2 years before moving onward and upward. And by no means will I leave this job for some menial programming position.

Also, I haven't graduated. I've only gotten my basics out of the way. I plan to finish a degree online at WGU with their BS in IT-Software to make sure I can transition when I'm ready.

What do people do to hold interest in their hobbies? I tend to get obsessive and flair out quickly (manic/depressive stages).

>>28807
Normal people have problems, problems make people ambitious, ambition makes people work.

No offense but most people I've seen who have these "depressive" stages have their lives handed to them, grow some balls and finish your education. Program a game, hack a console. You have skills most people don't.

Or if you just want to get distracted play a sport or learn to play an instrument.

>it's not working

stellar bug report, simon

>>28822
it wont draw the shapes

>>28820

Change 'draw' to 'paint' for your method

>>28824
thanks bra

>>28824
>>28825

Dont you need to override onPaint or something.. Or was that .net?

>>28735 You know, .NET isn't a net working lib per s?.


>>28723
I started a week ago on learning C#. I also knew java, obj c, python and web development languages. And I can tell you, it's a fucking breeze. It's almost the same as java. Never had a problem(with the language) that i couldn't solve by the first google hit. I'm pretty confisent i can write the same software in c# that i can write in java, and that after mot even a fulltime week. I had a little experience in .net in visual basic though. But .NET is only for the ui stuff and interacting with windows(partially the same thing). Plus that lib is sublimely documented aswell, if you ask me.

>>28723
Whoops, one thing though, if you make a new project in vis studio. always press the save all button at least once. else the project files arent saved on disk, and are kept in memory. Lost two days of work by forgetting to press that button(my computer crashed). i'm now repeating it and its the worst thing i have ever done involving software development.

>>28736

Huh? I did not mention .NET anywhere in my post; Qt is an application framework, Boost is a set of libriaries, both are written in C++. Besides, .NET is much more than UI stuff/Windows interaction.

>>28723
Why don't you learn .NET or JVM languages?

> C, C++

Could get you a job in embedded sytems. But you'd need to know quite a bit about low-level programming.

> HTML

If you add javascript, you could get a work as web designer.

Considering the languages you've already used, you would probably also get an entry-level job in some language that you haven't used yet. Just apply everywhere or consult a recruiter.

>>28771
How does one actually contact a recruiter? I have had several contact me but I can never find them.

>>28689
Good libraries implement each library function in separate object (.o) files, this enables the linker (ld) to only extract and link those object files from an archive (.a) that export the symbols that are actually used by a program.
Whatever stdlib you're using is terrible and shouldn't be used.

>>28793
I would assume it's just libc.a that gets linked. It's from glibc 2.17-2 in arch repos.

>>28700

>Windows also (afaik) does not support multiple processes using the same shared library instance in memory

It does, actually. It's how Chrome manages to not eat up all of your memory when you have a thousand tabs open. It's commonly overlooked in windows though, because the task manager doesn't show shared memory statistics by default.

>>28796

>>Windows also (afaik) does not support multiple processes using the same shared library instance in memory
>It does, actually.

in the old days, it was possible to overwrite things in memory, like the IAT for whatever dll you had loaded. has this changed? as i remember, the changes only affected one process, so does windows do copy on write now when things like this happen?

>>28797

>does windows do copy on write now when things like this happen?

Yes. In fact, that is why the IAT exists in the first place.

For those who don't know, IAT stands for import address table.

All functions that reference external libraries have to have their function pointers filled in at runtime. (If the linker is loading the libraries dynamically, the compiler can't know where in memory an external library is going to be loaded.) Each program is going to have it's own libraries that will have to be linked to various shared libraries, so they'll need different function pointers in those spots. For efficiency reasons, copy on write works page by page, so if any memory on a given page is modified, the entire page has to be copied to all of it's clones before the write can happen. This means that every page that has at least one reference to an external non-shared library will have to be copied to every clone. (Which will in practice be nearly every page.)

All functions that would normally point to an external library instead point to an entry in that executable's IAT, and each entry in the IAT holds the function pointer to the actual external library functions (which are filled in by the linker at runtime.) This way all of the modified memory that would need to be copied is kept inside as few pages as possible, reducing the memory overhead and allowing copy on write to be possible. (At the cost of adding in a extra jump for every external function called.)

Learn about heuristics, complex system, graph theory, statistics, alpha-beta pruning and other similar algorithms. Artificial Intelligence is just applied graph theory with some math and psychology.

>I imagine a game where there's a bot that plays players and learns from them

It sounds great in theory but in pratice it'll never be as efficient than a real AI.
You can definitely get emergent behavior and I'd love to work on something like that someday.

>http://aigamedev.com/

Lot of fancy screens. Lot of words to explain either nothing or something trivial.
Premium articles. PREMIUM ARTICLES
No thanks. Well that was my impression after 1/2 hour of reading. There are some decent articles I guess.

Why not LISP?

>>28770
Protip: when you say shit like that, not only will nothing come of it, but people of any experience will tune you out.

>>28782
jeez dude...
I agree you aren't gonna want to use LISP for an actually decent and useful program, but LISP can be good as an intermediate step in explaining some AI topics.

But either way, Charlotte's gonna use python...

>>28788
You know, Lisp isn't only about learning. It may seems weird, but it's an actual, useful, and very expressive language. You can do in 50 lines what may take 1000 lines in another more complex language, like C++ or Java.

http://c2.com/cgi/wiki?LispIsTooPowerful
http://www.paulgraham.com/icad.html