Sorry for the late contribution to the thread. There are a few things that I would like to mention.
First of all, we must understand that sniffing around other peoples' networks without their permission is not right ethically or legally (I agree with Fawad on this one). So, the first thing you people need if you have so much time for games
is that you have your own network for experimentation.
Once that is done, you can start with taking one operating system at a time, studying it along with its weaknesses and then trying to exploit those weaknesses. Detailed documentation of all this will be invaluable.
If we can get this done, we can give some detailed reports to the concerned authorities regarding the results of our research along with directions on how to secure their systems. This will hopefully benefit us in the long run.
At least, what we will get from all this "game" is that we will learn a lot. We will also be serving our country if we provide some valuable information to people who need it to secure their networks.
Though, I still feel that this task should be done by people who are getting paid for it but I guess that since they are not doing it in a way that satisfies most of us, we should help them a little.
This way, we will not be poking around into unwanted areas and will also be fulfilling the quest for knowledge (a llittle too sentimental )
If I may suggest, a good starting point will be to take a look at IPSec and the FreeSWAN project. FreeSWAN implements IPSec (Encryption at the Internet layer) so that applications don't have to bother with data encryption and other such issues. Two computers running with a Kernel supporting FreeSWAN (patched?) automatically take care of which algo to use for encryption/decryption and other such issues. The whole procedure is transpraent to the user.
The first step we people can take is to start using FreeSWAN, testing and validating its functionality and then creating a document that makes it easy enough for technicall staff of most organizations to implement this. This can help us big time.
The second step would be to code support for stronger algorithms into FreeSWAN. AES (Rijndael) is the new encryption standard (deprecating DES) but it is well known that it is not the most secure one. We can poke around with cryptographic issues and try to make an even more secure version of FreeSWAN (Military grade?)
These are just ideas, I would love to hear about any ideas that you people have.