I have been learning C and C++ over the last few years...

My problem is i don't know where to go next. I know how to use pointers, classes etc... in C++ but i haven't written an actual program, only assigments from the university(calculating arithmetic means, mean deviation etc)...

I have thought of joining some opensource project but i do not have prior experience...

Perhaps you are lacking in concepts? You understand the gritty little implementation details but not the grand schemes? Then why not learn Scheme? It will give you a new perspective on the meaning of calling functions, and change the way you think about programming so that you can advance conceptually.

Almost nobody has experience with programming as part of a project until they start. Perhaps take a look at the source code of your favorite applications and see how they work. Make some improvements that you've always wanted, or hack on a plugin for some program or another. Get involved with mailing lists or chatrooms.

Try to implement Perl 5 data structures in C++. You'll learn a lot about copy-constructors, reference counting, autovivification of references.

Make a small game like pong; Using GUI (e.g. SDL)

I say just think of a cool idea... could be a game, could be some program you've wanted to have, and then start writing it. You'll learn a lot by actually applying your coding knowledge to a real project. If you can't think of any good ideas, then joining an open source project is probably one of the best ways to become a better programmer.

Sometimes I would just implement some complex data structures or algorithms in c or c++, or some utilities for my desktop, just for the experience. It doesn't really matter what you write, as long as it has more complexity of a full program (classes, functions, abstraction, etc...).

Good luck!

Find a question asked on these forums that requires you to produce some code. Solve the problem, and submit your solution. Nothing beats a real-world problem for learning to create working code. Once you have cut your teeth on a few of these, you will probably have encountered enough of your own real-world problems that you can create a solution for. The more solutions you have rolling around in your head, the more solvable problems you will notice.
--- rod.