A few months ago, a friend of mine made an intriguing suggestion. "You know, you should publish a collection of your columns," he said. "Really," he said. "It could be a nationwide best seller on this campus." Gosh, I thought. Me, publish a book? I liked the idea, but dissertation research being what it was (and still is), I filed it in the "Think About Later" category for the time being.

It remained there for four long months, through the whirlwind of doing research, giving a talk, teaching a new class, doing more research, writing up four fellowship applications in just under three days, sleeping (occasionally), fixing the mistakes in th e research, flying to D.C. to interview for a fellowship I didn't get anyway (nice hotel, though), and the small matter of looking for a way to eat and have a roof over my head next year. (I'd just go to grad school, but I've already been to grad school!)

About two weeks ago, though, the idea resurfaced. I decided if Chad Strawderman could do it, then I could do it. I realized next that if I was going to do it, I'd better do it soon, or else everyone would graduate and people would just say, "John Keisling ? Who?" Since there was no time to try to get a publisher to underwrite the costs (and I'm not exactly Thomas Sowell anyway, at least not yet), this meant my entry into the wonderful world of self-publishing.

Along with Post-ItTM Notes and the Aegis missile defense system, self-publishing is among the very pleasant offshoots of modern technology. It used to be that a press run under about 1500 books wasn't worth doing because the printing setup costs were so h igh. Those days are gone now, and people self-publish every imaginable kind of book, even just a few copies for the family. With a little knowledge and experience, you can do just about everything yourself.

Of course, the drawback to doing something yourself is that you have to figure everything out yourself. In my case, this quickly became problematic. Which columns should go in? Just the Wildcat ones, or some from the Stanford Review? Would UA readers need footnotes to "get" the Stanford ones, or would they "get" them at all? Should I organize them by topics? What topics? What size should the book be? What font? What font size? What leading (the space between the lines, a problem at first I didn't even kno w I had)? How many pages would it have? Should I put book titles in italics or underline them?

That was just the editing. I first had to decide if it was worth it to even print the silly things. What would it cost? How much to do layout? To copy? To cut and collate? To bind? To print the cover with one color? Two colors? No colors? A photo? One col or and a photo? One color without a photo? What size photo? (Actually, what photo?) How many people would actually buy the things? I couldn't afford a market survey. How many would I order? How long would it take? Why was I in a fetal position, sucking my thumb all of a sudden?

Once I calmed down, things began to clear up. All it took to figure out the loss/profit projections was a simple application of everyone's best friend, mathematics. I simply defined the independent variables, derived the formulas for the dependent relatio nships, discretized the underlying Hausdorff space, and expressed the absolute profit as a discrete Booleanized five-dimensional hypersurface. What could be simpler? It turned out I could do it, even with a two-color cover and a photo halftone, and still keep the cover price quite reasonable and the risk in a decent range. I weighed the decision carefully, consulted my Magic 8-BallTM, and decided to go for it.

My opus would include all 55 Wildcat columns, 8 from the Stanford Review, and one new one. It would be called Essays of a Right-Wing Maverick. (Catchy, eh?) In the next few days, I wrote the intro, designed the front cover, pulled a bunch of quotes for th e back cover, had the layout done at a local publishing outfit, proofed the copy, carted it down to a printer, and sent it on its way.

The rest depends on my loyal fans. (Hint, hint.) Still, self-publishing is rather rewarding in itself. If you have the time, the writings, and a little capital, I'd highly recommend it. Just remember to Booleanize those variables.

John Keisling's new book will be available soon. To reserve a copy, please e-mail keisl@math.arizona.edu or call 325-0351. The author is a math Ph.D. candidate whose column appears Wednesdays.