Recently, I've been seeing a number of companies and projects springing up around the idea of publishing blogs as books. The examples I'm aware of are Blogbinders, qoop, LJBook, and most recently, book this blog, but I'd bet there are more. What I'm wondering about is how useful the blog-directly-to-book pathway is. Wouldn't an application that aggregates your blog posts into an editing environment (like Word or OpenOffice) be more useful? Can we really smooth over the formatting differences between web and print well enough to produce (automatically) a nice-looking book 100% of the time? I'm a little skeptical.
From my (admittedly cursory) browsing, it looks like blogbinders have a human in the middle of the process, and qoop certainly did for their only title so far, John Battelle's SearchBlog. Requiring a human being in the loop raises costs and introduces scaling issues.
There's a fine tradition of publishing diaries, going back at least to Caesar, but unless you happen to be famous, or have a blog that's truly interesting a high percentage of the time, the way to monetize blogs is more likely to be on the Hardball Times model, where you build up an audience, and then sell them work they're interested in.
But if blogs really are the ultimate vanity presses, then there may indeed be money in printing them, if you charge the blogger enough up front. It will be interesting to see how all this shakes out.