To be honest, I'm thinking of buying it myself, from what I've read in the responses. I don't intend to ever build any game using DX, or using any platform specific code/lib, but the fact that it involves creating an RTS and terrain generation, and is being well spoken about, has got my mouth watering. I'm all for interesting reads on concepts I feel attracted to.
For me, specifically, I think it might be a good source of further knowledge that I might end up translating/incorporating into my actual projects (I don't think I'll be attempting to develop an RTS anytime soon). For you, well, I don't know. You could follow along using the old tools, and it should still work on any windows machine (that's my guess).
One thing to keep in mind is that once you're done you'll want to step up to more up-to-date technology, and that will involve further reading and learning, kind of like starting over, in a way. Might still be a lot easier than to follow along with the book while trying to translate it to its modern counterparts, which, as far as I know, involves not just using different keywords, but writing different code in a different mindset for different specific reasons.
Or you could still buy the book and read it on the side, as you focus on something else with some other book. To be honest, I can't really say what would be the best option, since the most important thing is that you're approaching the subject in a way that you really enjoy and keeps moving you forward. What I think I can say is that, from what it seems, you'll probably not be sorry to have bought it, regardless.