I agree wholeheartedly, although I personally dislike xmonad. We need more experimentation; we need lots of people trying out random ideas. Most new ideas wil

I agree wholeheartedly, although I personally dislike xmonad. We need more experimentation; we need lots of people trying out random ideas. Most new ideas will fail, and even more implementations will fail, but the good ideas will slowly rise to the top. If we don't see more failures than successes, we're not stepping outside the comfort zone enough. So I present a challenge: take a common piece of software/functionality that you use and come up with a new approach. It doesn't have to be a good approach, but it has to be different. Implementations aren't necessary here, but they often help to turn a bad idea into a good one. For example, I don't believe much has been done with tabbed browsing in the past few years. It'd be nice to see a tiling window manager inside the browser, with tabs acting as a sort of taskbar. Would this idea work well? Doubtful, for most people at least. But who knows what idea might come from it if it was implemented?