I've been having a lot of fun recently picking out books to get with the faculty development money I need to use up. I feel a little like a kid in a candy store; there's all these cool books that I've either used personally or had recommended at some point, or are just classics that "every mathematician should have." I've realized three great things so far.
First, Half.com totally rocks. (When it works. I've also run into weird glitches, like one book that always appears for sale, but which is never available when you actually try to buy it... it's a phantom book I think. Or the time I received confirmation that a book had been shipped, followed by a note from the sender that "I can't send you this book" and a refund three minutes later.) When it works, you can get some older texts for cheap, and there's tons of old math texts that are still great. I don't think this would work as well in a lot of other fields, but things don't usually stop being true in mathematics, so we still read and reference a lot of old books. A lot of new ones too, but still plenty of old ones. (I remember one of my professors in college joking about a false "theorem" that was "proved" in a journal article one year, and found out ten years later to be incorrect. He said the theorem was "true between the years of...". My partner was also surprised to see I referenced papers from the 1800s in my dissertation; he considered anything older than perhaps 1980 to be tremendously dated and highly suspect.)
Second, Dover totally rocks too. I discovered Dover has been going crazy putting out ten-dollar paperback reprints of tons of great math books. (Again, this works great if old texts are still useful to you.) They always had some good stuff, but it's sort of amazing how much is available now. I'm waiting on a lot of these because they're so cheap and plentiful now, and mostly using grant money for more expensive (and harder to find) texts. But eventually I think my bookshelf will overflow with them.
Third, I must still really like math, because I'm really enjoying this process and it's exciting to think about all the things I could be reading. (I know you would assume I still like math, but every once in a while, I find myself wondering.)
I'm just slightly disappointed that my splurge is coming to an end. But I've promised myself that every few months I can buy myself a new math book, even if it's kind of expensive. I just have to remember that it's not just spending money; it's professional development.