I used to try to work a bit on physics articles. That actually wasn't the problem there -- the problem is that there are so many cranks, and they (being cranks

I used to try to work a bit on physics articles. That actually wasn't the problem there -- the problem is that there are so many cranks, and they (being cranks) are so insistent on their incorrect views that it becomes a constant, tiring battle preventing any real work from being done. The only rational course is, once it has been established that you're dealing with a crank, to completely disengage and smack them down with the official policy. (They never have any reliable sources/citations to support their edits.) It's tempting (since they are usually provably, mathematically wrong) to try to convince them of their errors, but experience shows that never works. I saw a lot of people just burn themselves out doing that -- in this case, expertise was being driven away by reluctance to rely on policy.