Perhaps too eagerly, last year I agreed to write an article on print on demand for the Library Journal.  I'm soon to submit it to them in a horrid mess, which time pressures force upon me.  It's been a really interesting opportunity though, and it has very much alerted to me the nexus of happenings in digitization programs, localized services, print on demand vendors, distributors, and digitized book repositories.

One of the things that I will write about is On Demand Books, which is a company founded by Jason Epstein and Dane Neller.  On Demand is offering a distributed print on demand solution called Espresso; the New York Public will be installing one of these machines soon.  It's large yet - think early Xerox Corp. (Haloid Corp.) equipment - and the business model is maturing.  Nonetheless it offers a myriad of interesting opportunities, some of which I will discuss in my article.

One of reasons for the interest in print on demand among libraries is the possibility that they may soon have access to significant digital repositories.  The most prominent example, and one near and dear to my present heart at the University of California, is the potential digital largesse of works being made available via the Google Book Search Library partners program.  In the program, libraries get back a digital copy of their works scanned by Google.  Obviously, for works that are in copyright, there is a very limited number of things that libraries can do with these copies.  For public domain works, however, almost anything is fair game, including printing off your own copies.

One of the challenges of the Google Book Search program for libraries is that the quality of the images delivered to libraries is uneven, and certainly not archival.  Without belaboring the details, it is fair to say that Google's effort is focussed on the indexing of the texts to power discovery, and a marred display image is an acceptable compromise to make in order to reach the magnitudes of digitization necessary to make the operation - an industrial one in scale - sustainable.   But  not-pretty images pose a problem for print on demand, which would potentially produce not-usable, not merely not-pretty, books. 

As my colleague John Mark Ockerbloom at the University of Pennsylvania notes, it is not acceptable to print a book for a scholar or a student of uncertain fidelity, where the reader discovers late one evening, weeks after obtaining the copy, that eleven pages in the middle of a critical chapter are entirely illegible.  

For these repositories to be acceptable, John points that what we should do is to establish a clearing house or registry of these digitzed works - and it is conceivable that OCLC might be in a position to create one - and then make it available for commentary on issues of usability by qualified librarians. 

In other words, if a faculty member requested a print of a book, a librarian could verify whether it met minimal standards and could give it a rough grade, certifying it to a certain level.  They wouldn't try to correct or itemize the errors, but rather merely note this was a readable work, or readable but for the preface.  In such a fashion, particularly if universities could ever figure out how to work together to make a centralized repository of public domain works, one could know simply by looking up the work whether it was printable or not.  (There are other uses too, of course, and if the registry included in copyright works, you could also imagine it being married to rights management cataloging.  But that's another story altogether).  

Given the challenges of these new, mass digitized repositories, one is forced to look for collections of material that are clean and clear for print on demand.  One source for these materials are existing print on demand vendors like Amazon's BookSurge and Ingram's Lightning Source.  These  POD vendors actually provide two distinct services (at least): they maintain a digital repository of works, and they provide the actual book production process.  An increasing number, albeit small fraction, of their works are produced by university libraries who digitize collections to high quality; the University of Michigan partners with both BookSurge and Lightning Source to deliver print on demand solutions for some of their rare or fragile material.  However, a larger number of works in the digital repositories come from publishers, who essentially use the POD vendors as a digital repository for their catalogs.  

This raises all sorts of interesting possibilities and questions.  In the Google Books Program, one of the things that has most irked the publishers is the fact that a copy of in-copyright works goes back to the contributing library.  I think it is clear that we are moving to a world where many publishers are seeking to control their digital product, either directly in their own repository, such as Harper-Collins, or via the auspices of vendors like Ingram.   It is not a stretch to imagine that publishers will increasingly seek to license digital holdings to libraries for uses ranging from print on demand to digital lending, in the same way that journal publishers and aggregators have moved to such licensing models.  One can see the opportunities that a vendor like Lightning Source has in such a model.  They could even establish a partnership with On Demand Books, nominally a competitor of print on demand solutions, whereby they provide the source repository with adjutant rights control and royalty payments to an On Demand machine at a library.  The publisher, in such a system, would be able to much more tightly control the digital copy. 

(In fact, if there was a registry of On Demand locations, a user could choose the location that he or she desired to print and package the product, perhaps e.g. using their library preferences from OCLC's WorldCat; there are interesting potential co-service offerings between library information vendors and publishing distributors).

Today, I was fortunate to speak with J. Kirby Best, the CEO of Lightning Source.  Kirby indicated that they were interested in exploring a wide range of new business models, and would certainly consider a partnership with On Demand Books in which On Demand served as a local printing point of presence for their digital works repository.  Lightning Source understands they occupy a valuable territory as - at least, what they hope to think is - the Switzerland of the publishing industry, presenting a wide variety of services to both producing and consuming clients.  

Lightning Source is pursuing the addition of high quality and high value works to their digital repository; they clearly appreciate the partnerships they have been able to achieve with universities for rare and scarce materials.  This kind of long tail material, if aggregated by Lightning Source, could be one critical source for a high quality repository of public domain works of great scholarly or even leisure value. 

Given this level of interest, I asked Kirby if Lightning Source would be willing to fund digitization projects at university libraries which were focussed on obtaining high enough quality scans and images to power print on demand (as well as other digital delivery) solutions.  He indicated that they were willing to consider such proposals, and indeed were working with one small pilot already.  I inquired as to whether, in such arrangements, Lightning Source was willing to contractually commit that partner libraries were able to retain the source high quality images produced, and Kirby responded in the affirmative.  (I also confirmed that Lightning was willing for this information to be public.  Evidently, they had simply not been asked if they were willing to entertain such collaborations). 

I think this is an extremely interesting opportunity for libraries and distributors; generally print on demand is an extremely interesting space for publishers, distributors, and libraries.  We need to spend more time thinking through these spaces.