- Easy stuff, the kind the some proper digging within the docs or the logs would give the answer. This always seemed like the majority, even more within the last few years. I don't think I should spend my time answering those. I used to ask questions on forums before and got schooled a couple of times from people telling me to look it up in the docs or from other questions, it was more valuable than a direct answer itself.
- Stuff that require digging, like 1-hour or more. Those are interesting, but again, I have my own issues that require 1+ hour research that I could spend my time on. I know not everyone is doing it for the points, but I didn't feel like the transaction is worth it. The knowledge I'd gain from researching a topic I don't currently care about to answer a random question might seem valuable (and it may be), but I think it's somewhat similar to the value of watching those short "educational" youtube videos, you feel like you learned something, but 2 days later you basically forget the whole thing. Maybe it's just me.
- Stuff that you're the/a authority on, like if it's your language/library/framework/toolkit. I'm not either of those, but if I was, I don't think I'd be able to spend time answering random questions on a site I don't control. I'd probably feel much more compelled to give lengthy answers on the tracker/forum/qa site of the project itself.