If you’ve reached the “Preservation” chapter in Planned Obsolescence, you might recall Kathleen Fitzpatrick’s observation that, unlike in print, where simply using a book can interfere with its ability to be preserved, in digital “the very point of… preservation is ensuring future usability” (144). You also might have noticed that she returns on a few occasions to work The Andrew W. Mellon Foundation has done to advance thinking and to fund projects and tools at the intersection of digital preservation and access.
As has come out in some of my introductions of myself to the class, I happen to work at Mellon in the very program (Scholarly Communications) housing the projects Fitzpatrick hones in on. Though our grants to LOCKSS and Portico predate my time here, there’s a range of other projects on both our preservation and conservation and access and library services fronts that could be worth a Google or two.
I thought I’d use this space now, though, to offer a brief peek behind the curtain at another project we’re funding–jointly, with the NEH–that has digital preservation and access components to it (despite its existing in the publishing area of our portfolio). It’s called Humanities Open Book, and it’s designed to help university presses and libraries make the best out-of-print humanities titles in their back lists open-access and freely available to scholars and to the public. We convened a small group of HOB project directors just last Thursday to discuss the opportunities and challenges native to these sorts of digitization and publishing efforts, and I thought I’d share here just a bit of what I heard, and what I’m still thinking about, in a kind of generalized pseudo-workshop debrief:
It’s not as easy as you’d think. There seems to be something unarguably good about sharing created knowledge, perhaps especially when it stops circulating and becomes stagnant or invisible. Common obstacles to even getting going with this sort of work, though, include obtaining authors’ permissions (some are dead, some are hard to reach, some distrust or philosophically object to OA publishing) and securing appropriate copyright clearance (even with fair-use policies, each image, big or small, photograph or artwork, illustration or map, will have its own side path to journey down). For this reason, text-heavy titles that are only recently out of print are exponentially easier to publish than are architecture, art history, or design titles on the outside fringe of “public domain” territory; a year of work on each might result in 600 publications of the former, while only 18 of the latter. Commission some new forwards/introduction essays or fresh cover designs and your timeline can extend well past what you had originally projected.
If it’s not accessible, it’s not actually OA. Digitizing out-of-print titles as EPUBs might make them “open,” but it’s often not enough on it’s own to confirm true “access.” Extra work needs to go into remediating these titles and making them ADA-compliant, work which might require annotating and converting texts and, as importantly, images into machine-readable formats. Factor in the fact that some titles might be written in non-English languages featuring diacritics that aren’t easily picked up by OCR-like technologies, and you really have to go beyond the simple scanned PDF or EPUB to demonstrate a true commitment to OA.
Even when you’ve done it, it’s hard to know how you did. In one project director’s words, HOB allows titles to be “reborn,” with books disseminating into the hands of readers in up to 150 different countries, in one case. Routing these titles through a range of aggregators and distributors like JSTOR, Project Muse, HathiTrust, Google Books, (etc.) might aid this kind of increased visibility and exposure, but may also result in duplication or redundancy of content across platforms. Does this ultimately help or hinder discovery? Moreover, without the ability to consolidate OA usage metrics across these platforms, there seems to be no efficient or consistent (or standard, in Fitzpatrick’s terminology) way of reporting on if or how these recovered texts are being used. (For what it’s worth: there’s a recent Mellon grant out to the University of Michigan to support cracking this nut.) Since getting organizational buy-in beyond the “soft money” of grant-funded support might very well rely on such analytics, this seems to be a critical area of focus in the larger conversation about preservation and use.
What are we doing this for? Is the goal of projects like HOB to churn out the largest number of out-of-print texts as possible? If so, perhaps presses and libraries start to lean to simpler, text-exclusive projects in literary criticism, history, and philosophy. Or is the goal to figure out how to overcome the obstacles of more difficult projects involving significant out-of-print titles that might otherwise be lost to time? If so, perhaps organizations begin to prioritize image- or design-heavy titles, or ones that invite new contextualizations in our political climate (e.g., in Indigenous or Black studies), with a focus on establishing model, replicable, and sustainable workflows. Related to this: one project director noted how the lack of an online source for buying and selling ebooks (i.e., no Amazon) in Latin America had resulted in an increased market demand/ potential for preservation/publishing projects like HOB, while another project director showed off his organization’s use of a Python script tracking WorldCat holdings across the globe to see whether popular titles in one area of the world are noticeably absent from others. Could a geographically-focused strategy for selecting out-of-print texts parallel or even complement the mission-driven approach of the area-focused strategies above?