Self-publishing

Long ago, when the world was new (well, to me it was), a fellow named Don Lancaster was writing about strange things called "TV Typewriters" (hey, it was like a TV—it was some time before many computers had any graphics, much less graphics that needed more bandwidth than TVs could provide—and what you typed was displayed on it). He also wrote about starting your own business...

...and about self-publishing with print-on-demand. Computing and advances in printers made it possible to avoid the "vanity presses" and their fees (and scams, as the SWFA points out). Instead, you printed the books yourself at the rate you needed; no need to keep a huge inventory.

Things have gone even further, now; there are companies that will do the printing and shipping for you, in the same way that cafepress.com prints and ships T-shirts, mugs, and the like. For example, the Cannery exhibit in SL can be seen in printed form via a book available on lulu.com.

So... just as computers have made it far less expensive for people to do many other things once limited to a select few, they've made it less expensive to publish a book, and I hope this will have the same effects...

...but on the other hand, Sturgeon's Law does apply. Self-publishing isn't new, just a lot cheaper, and I found what turned out to be one example of it long ago as an undergraduate.

While rummaging through the stacks in an obscure corner of the math section of the university library, I found a small book, published some time ago--I forget whether it was in the early 20th century or the late 19th. It grandiloquently proclaimed that its author, a 33rd degree Mason (ooooh....), had solved the problem that had defeated mathematicians for so long; he had squared the circle! Except that he hadn't, really...

Squaring the circle, for the non-math majors, is, or rather was, a problem, and it dates back to the ancient Greeks: how, using standard geometrical constructions in Euclidean geometry, can you draw a square with the same area as an arbitrary given circle? Standard geometrical constructions means all you get to use is a compass and an unmarked straightedge. Mathematicians beat their heads against it for ages until, in the 19th century, algebraists described what geometrical operations let you do in terms of finding zeros of polynomials... and proved that pi, which is a factor in figuring the area of a circle, is not just irrational, but "transcendental," meaning it's not the zero of any polynomial with integer coefficients, so all those earlier mathematicians were beating their heads against the wall for nothing... well, except for all the math that resulted from trying to do it.

The mere fact that you can't do it, though, didn't stop some, um, determined folks... including the wealthy man (and 33rd degree Mason!) who talked himself into thinking he'd solved it, wrote and printed a bunch of copies of a little book announcing his "triumph," and then sent them out to land-grant colleges and universities across the United States. There they were obligingly put into the stacks, a curiosity left to molder between occasional looks from curious folk like my younger self.

I have to wonder what, and how much, writing of this era will be read by people who will look at each other with a wild surmise... and then burst into howls of derisive laughter.

P.S. Why the constraint to integer coefficients? Well... if you allow arbitrary coefficients, trivially every real number r is a zero of the polynomial x - r... and WLOG (without loss of generality) you can always use integers instead of rationals--just multiply out by the product of the denominators, or their least common multiple if you're feeling miserly.