Printing books on demand missing the point

News reports indicate that Espresso Book Machines (EBM) are being installed in some Australian bookstores and US libraries. The EBM is a Print-On-Demand (PoD) book printing service using a laser printer and binding machine which prints books from PDF files on demand. But it will still take several minutes to print a book and so I am not sure this is a viable service. Also I don't think the future of university librarians is in the book printing business, it is in running the teaching facilities in the learning commons.

For PoD you need a printer which is quick enough to do it in the time people are prepared to wait, but cheap enough to buy and easy enough to run. That is very difficult. Print on demand, deliver next day is much easier.

High speed laser printers cost hundreds of thousands of dollars. To make them pay their way they have to print a lot of material each day and they need trained staff on hand to keep them working.

There are still economies of scale in printing, even for print on demand. If the book is not available within a few seconds, then for many customers they will be happy enough to have the book delivered overnight. With the delievry time changed from minutes to hours, the codered can be consolidated at a larger facility servicing a city (or a country). A large high speed printer can then be used with staff who know how to work it.

But PoD is largely irrelevant to publishing. It is a bit like seeing email to fax interfaces as important to the future of written communications. We spent a lot of time worrying about email to fax interfaces about fifteen years ago. It turned out that worrying about faxes was a waste of time, as email quickly supplanted it. There are still fax machines and email to fax is useful, but not very important.

Being able to print documents, such as "books", is useful, but not very important. The real document is the electronic version and print just an option, useful for some limited purposes.

The ACS has made this transition with its Journal of Research and Practice in Information Technology. For most of its 40 years, this was a traditional printed publication sent out by mail to subscribers. We still produce printed copies for those who want them. But I expect we will be down to about 200 print subscribers out of 15,000 by the end of this year.