[bkcooper]

The Feynman lectures certainly will teach you a lot, but I think it would be a fairly trying textbook for learning a lot of the material for the first time. Among other things, a very serious deficiency for self-education is a complete lack of problems to work (which just reflects that these really were lectures).

[DennisP]

What would you recommend for self-education?

[bkcooper]

That's a tough question. One of the most appealing things about the Feynman lectures is the breadth and how self-contained it is. I don't think there are really good analogues for that. The closest thing I can think of is one of the monster first-year physics tomes like Halliday and Resnick. That will teach you much of the basics (large chunks of Vol. 1 and 2 of the Feynman lectures), has huge numbers of problems, but it is of course less fun. It is also probably comparably or more expensive.

Going much further than that gets tricky, because you will usually need more development in math in concert with the physics. The omnibus "engineering mathematics" type books will cover a lot of it but I don't really like them. Boas's Mathematical Methods in the Physical Sciences would be something I would look at. For more detailed looks at electromagnetism and quantum mechanics, I (and many others I know) really liked David Griffiths's textbooks. The Feynman lectures make an excellent supplement to these for the different perspective and interesting physical insights.