Usually I avoid using SPs so I can be database independent. If you need to switch database and don't have SPs you don't have to rewrite all that logic. Another reason not to use them is that the DB is difficult to scale. You can avoid using valuable CPU time for SPs on the DB server if those functions can be executed on the application or middle tier layer (that one scales well) as well. But when I don't have scaling problems, or don't plan to change DB I use them as well.

It's cool in the sense that most people don't do it. :) At least, most people who release open source stuff.