With a fair amount of fanfare and an official launch by a major British bookseller and publisher On Demand Books released the Espresso Book Machine 2.0 last month. The launch included British bookseller and publisher Blackwell's installing an Espresso Book Machine in their Charing Cross shop. The EBM, as it's affectionately known, produces a paperback, perfect-bound 300 page book from a digital file in about five minutes. The basic process is that once the printer/purchaser has selected a book from the catalog, or supplied their own digital ready to print file, the book is printed on 8.5 x 11 paper, duplex or "two up," the cover is simultaneously printed on a four color laser printer, the pages are joggled mechanically to align them, clamped, milled, and the cover is stuck to the pages with hot glue; then the book is trimmed, with trim size ranging from 4.5 x 4.5 to 8.5 x 11, or "A4" size. The EBM at Blackwell's has 500,000 books in the database, ready to print, with more expected. The pages print at roughly 100 pages a minutes, so an average paperback would take about 5 minutes, all told.

The good news is that there's a great deal of potential for books that are out of print to be brought back in print, and that books that have limited appeal or lifetimes of use—say software manuals, or, my personal favorite, scholarly monographs—can be fairly easily printed and produced by would-be readers. Note I said easily; the costs of paper and binding and service do not make a book printed by the EBM a bargain, by any means. But it does mean that books with limited shelf life need not be "out of print," and unobtainable. That's very good news indeed.

I'm less sanguine about some of the prognostications I've seen around the 'net about the EBM changing the nature of publishing, making books affordable, or removing "the middleman." First of all, if you actually look at one of the books thus produced, the cover stands out immediately as a less than standard professionally produced cover; that's in part because of the stock, which in part determines the ink used, but mostly, it's because it's a 4 color laser printer; that's pretty much the same laser printer you'll find in a business office, the one that looks a lot like a photocopier because it does that too. It's nothing like the quality of printer a trade publisher uses for conventionally made mass market paperbacks. Nor is the binding the same quality as a mass market paperback, in part because of the hot glue; it's not the same kind of hot glue used on trade paperbacks, because that hot glue requires ventilation.

But most of all, I worry about the typesetting issues, frankly. If the user chooses trim size, for instance, that means the typesetting and layout are blown, unless the publisher provides multiple versions—and that's expensive. The typesetting and layout are important not only because of aesthetics and readability, but because if the book is paginated with a table of contents, or index, or, horrific possibilities, footnotes, those are enormous potential problems if there's a layout change. I also want there to be some way of paying royalties for books that merit them; no one is mentioning how that is, or will be, done. I note that the official description from On Demand regarding the quality of the books printed by the EBM is "library quality." That's pretty much meaningless; the accurate description is "print on demand" quality; good enough to read, but not something to read over and over again, not as durable as a conventional mass market paperback, even.

Please don't misinterpret me; I think the EBM is fabulous, and I think every campus should snap up as many as they can. But I do not think EBMs remove the need for editors, designers, typesetters, and artists. I do not think they represent much of an opportunity for change to publishing work flows. And, having read slush, and lots of vanity published books, I shudder at the pie-in-the-sky "everyone can be a publisher" scenarios. I'm reminded of the first few years after the Mac Plus was born, with the Apple Laser Printer as a sidekick. People started creating newsletters that used eight fonts on a single page, six of them display faces, two of which were used for body text, and yes, that really did happen, and yes, I was one of those people. I got better. But there's an assumption that the cost of producing books is in the printing and in the warehousing. That's not really the case; there's the same amount of overhead in creating a book that sells five copies and one that sells 50,000 copies up until the point where the book becomes a physical object. But those pre-print costs aren't going to change; and no, there's really not a lot to improve in that process, really and truly—if you want a quality product to land in the reader's hands.