21 replies to this topic

[karligula]

... you cross your fingers every time you run your code (I seem to be doing that a lot lately(!))

[Reedbeta]

LOL

"Run it, change something, run it again to see if it works" is a valid, if slow and error prone method of software development...

[Mihail121]

Don't forget also that the "printf" debugging is a completely legal and widely used way for correcting bugs and memory faults!

[eddie]

Hah. I think every programmer is like that, really. Granted, if you're doing that on HelloWorld, you've got issues. :)

A good way to improve the outcome of your running is unit testing, however. :)

[davepermen]

i love it in c#, where i can actually predict the errors while compiling. i _bet_ that'll get a deadlock here..


it's locking!! it's locking!!!! who payes the beer now, dudes?!!? :D

[Alex]

I find it even more scarry if you compile something (to have an excuse to check ur mail or so) that can impossibly work cause it violates at least 2 fundamental assumptions made all over the code..and it runs without any obvious glitches...

Alex

[.oisyn]

@ Alex: heh yeah, I had something similar last week. I am working on a scene management system that does all the culling and passes all the appropriate data to the renderer. You can mark a model with flags telling you want it to be dynamically lit and whether it should cast shadows. It all worked fine for some time now, until I saw the statement setting these flags (not my code) using the logical or operator (||) instead of the bitwise one (|). Turns out that it was working correctly because I, in turn, wasn't reading the bitmask correctly. d'Oh!

[Wernaeh]

You shouldn't become a programmer if you spend an entire week of your holidays playing with your shiny new stage piano, instead of finding that
nasty multithread deadlock bugger :)

Well, perhaps I should reconsider that... After all, as long as you drink your
daily caffeine and don't go to bed before 5 a.m., I guess it doesn't matter
what kind of keys you actually hit :o)

Cheers,
- Wernaeh

[Ed Mack]

... when you estimate the length of daily chores perfectly.

[kariem2k]

LOL,but that happen when you enter a new field of programming but after some time you will not cross your fingers any more :), you will be sure that it will work.
also as a programmer you are like an inventor who is making a mathmatical proof,he of course he has the tools (Mathmatics knowrldge) and you have the knowldge of programming.

[ProgramWizard]

...when you get bored during programming. Seriously. I've heard people say how much they hate it, and then turn around and say that it's their calling in life.

[eddie]

Meh, I get bored with it sometimes too. :)

But that's what video games are for. :D

[Superfly Johnson]

Lolgorithm Says Hi

[monjardin]

Ugh. Not the stupid Lolgorithm. It's really not funny to most people that didn't go to your school.

[eddie]

You could've left it at "It's really not funny".

[moe]

...your first program ever said something other than "HelloWorld".

...you don't know the difference between an array and a linked list.

...your test functions have a name other than foo, bar or foobar.

...you look into the mirror and don't think about how to compute the reflections.

...you think "2 + 2 == 5 is true for extremely large values of 2" is not funny.

...you don't have a coffein level greater than 5%.

...you're in danger and your first thought is not F1.

and you should not be into game programming when...
...you see a house beeing build and you don't think: "why is it rendered in wireframe?"

[Jynks]

the longer u leave it b4 u run it the more you love to test it

[ikk]

when thinking about upgrading hardware instead of improving code performance

[ProgramWizard]

What's a linked list? :sad:

Do not strike me down, mighty Moe, I've only been programming for 10 months!

[eddie]

It's a data structure that holds both data, as well as pointers to other elements that together form a list.

Looks something like this:

template<typename T>
class List
{
public:
    T m_Item;
    List<T> *m_pPrevious;
    List<T> *m_pNext;
};

I've ignored functions for simplicities sake.

The idea is that as you add elements, you chain them onto the pointers. Usually NULL on m_pPrevious or m_pNext signifies the ends of the bounds.

Clear as mud?