If you want to make a splash in any community (programming or otherwise) one of the great ways to do it is to take something that is considered impossible or too hard and make it possible or trivially easy.
In my Open Source life, I've managed to do it twice. I was repeatedly told that PPI was impossible (and they were right), and I was repeatedly told that it would be impossible or stupid or way too hard to install something as big and complex as an entire IDE from the CPAN.
I've been told by some of Perl's old hands that ExtUtils::MakeMaker implementation was also considered impossible, that generating makefiles on the fly like that was just too hard.
Finding and executing a headline high profile project like this isn't just valuable in it's own right, they can act as game changers for an entire area of development. Especially if the headline project doesn't ONLY prove something is possible, but does so with style and elan.
Being told that something is impossible or stupid that I know SHOULD be possible has become like a rag to a bull for me.
Being told that nobody wants to consume Perl games via the CPAN, that everyone wants to build standalone installers was such a moment. The CPAN provides SO much value, why should every single game author have to invent their own installers and embed Perl? How can that possibly be a good thing if a better option is available?
And so after some cajoling to the new (awesome) Perl SDL maintainers, we've started the process of bringing Frozen Bubble (1.0) to the CPAN, just to prove it's possible and how easy it is to distribute games in this way.
If you'd like to help, or just watch the process, you can checkout the game from http://svn.ali.as/cpan/trunk/Games-FrozenBubble
I once looked into getting FB installed on Windows...I didn't have the time, and never got back to it, and the FB maintainers seemed to have no interest in making it easy to install on Win32.
As for doing the impossible (a little OT), I really like the way perl tends to let you do that. Several times people think I'm kidding when I suggest spending a few minutes or so writing something in perl rather than several hours/days doing something manually (even to supposed developers/programmers who should know better).
