Personally, I believe the most important problems are mostly relationship problems in other words politics. The solutions there are simple but not easy. For example, world peace is simple: Everybody just stop killing each other. However, getting everybody to agree on that is ridiculously hard.
My general thoughts are:
1. I have a certain set of skills and strengths that I can contribute. I should use them instead of doing stuff I'm not good at because that should optimize the global productivity.
2. I should work on things where return of investment is best. So do not work on things where a return is unlikely, e.g. intergalactic travel or fixing email, or where the return is not worth it, e.g. cheaper ice cream or a flappy bird clone.
3. Keep track of big problems where the time/opportunity has not come yet. This is inspired by Hammings talk on research. It also applies to lots of free software though. For example, if Twitter would shut down tomorrow, then the Fediverse would be there to catch the fallout. So it is of some importance to work on open alternatives to popular services. The iPhones success was mostly waiting for the right time, when mobile Internet, mobile computing, and app-store market structure all were ready enough.
Unfortunately, in terms of my hobbies I'm still mostly suffering from the Paradox of Choice. Too many options to make a decision, so I don't do anything. Let check Hacker News again...