> find a problem/task that interests you or is worth solving - and work on it until it's solved, regardless of how difficult

I guess part of the problem is that those problems tend to be rare. The way I feel, is that if I have a problem, it's likely already been solved (in which case I'll just use that solution), or people much more intelligent than me are already working on it.

Another issue seems be be is that as time goes on, the problems I take interest in seem to become much more difficult to solve. This may be because they are technically complex, but can be for other reasons too. Often I don't even know what I need to be learning / researching to solve a problem, and that basically halts me.