Minimum Lovable Product

I liie esr's definition (for community building, but works for MVP too):

> plausible promise. Your program doesn't have to work particularly well. It can be crude, buggy, incomplete, and poorly documented. What it must not fail to do is (a) run, and (b) convince potential co-developers that it can be evolved into something really neat in the foreseeable future. http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral...

In some cases, the problem you're solving might adequately define the "promise" you're trying to fulfil.