The interview reads like Russell offering up science and math to expound upon Hume's answer to Descartes. I appreciate that it's more detailed, and some of the examples are excellent. It does boil down to basically that, though.

The lesson to me is that Descartes is always going to be valuable because Hume's approach does make people lazy. Russell appears to "school" the interviewer over and over again in exactly that way: actually, no, you're assuming too much, and by assuming less, you may actually get to a better contingent truth even though you still have to accept some shit on faith (or "instinct.")

That's kind of beautiful. If you think about it, it's a great apology for the idea of the devil's advocate. The guy whose position is "actually, no, you can't really know much of anything" keeps you honest, even though you're never going to accept his position, because, well, it sucks and you don't want to starve to death or treat your dad like he's an illusion with no moral significance. But if you let yourself get pushed by it to a point, your own work will benefit.