I have a chapter scheduled for a forthcoming book entitled Rethinking Collection Development and Management, edited by Christine Copp Avery. I’m writing on the ways that print and digital preservation intersect and while there is probably a whole monograph to be written on that topic alone, here’s the current version of the chapter for your review and comment until I get around to that. I’ll replace this with the pre-print and the final version as the work moves towards publication: Print and Digital Preservation.
Jacob – I was going to send this as an email – but didn’t quickly find an address for you, so I’ll do it as a long comment. Most of my comments are copy related.
Interesting use of the FRBR model to look at preservation. I liked it, and found it helpful, although I find FRBR’s concept of “Work” too abstract and theoretical to be useful. (And now I’m at the end of the article and I see you also say Work-level preservation is too abstract.)
“There are numerous methods available to care for a repair items…” p. 4 probably should be “and repair items”
“Damage can occur to an Item in one node of the network without affecting the existence of the rest of the WEM chain.” P. 7 “WEM chain”? New term – I’m guessing it is short for Work Expression Manifestation.
“There is a potential problem here if the publishers DIP is substantially different from the publishers own SIP or DIP, especially if there were to contain metadata or higher quality data than the publisher released through their AIP.” It would seem to me that it should be “own SIP or AIP” and “released through their “DIP”.
And shouldn’t “publishers DIP” have a possessive apostrophe? – publishers’ or publisher’s DIP.
“The roles played by copyright, intellectual property law, and license terms are key policy issue for library preservation.28 However, these restrictions are solutions to fiscal problems, not intrinsic characteristics of digital information.” I’m not yet convinced some of these are not intrinsic characteristics of digital info – not that I have any evidence to support my case, but I remain suspicious.
“OhioLINK was established in 1987,31 the same ear when the University of California system also created a shared collections…” p. 13 I assume that should be “year”.
“Access to adequate digital versions of the works kept in shared print networks…” p. 13 Just curious – do you mean “works” in the FRBR sense? Because you brought up FRBR and its terminology, I am now more alert to the terms. I don’t know that your use of “works” in this sentence agrees withthe FRBR meaning of the terms, but it’s something to pay attention to.
“…and accessible in a format that is intrinsically able to be ubiquitously available, even if law, economics, and policy proscribe use.” P. 14 My gut has a hard time with that line. The fact that law, economics, and policy proscribe use, make the idea that it is/might be intrinsically able to be ubiquitously available almost moot. To me, digital objects/collections are so inextricably linked to the restrictions of law, economics and policy that I think their ability to be ubiquitously controlled/unavailable is almost as intrinsic. (Please pardon what may be my lingering, latent uncertainty – or even modest distrust – of digital material.)
Thank you, Kevin! Your content- and copy-editing are both helpful and taken to heart. Your notes about intrinsic qualities of digital objects are especially helpful. I need to find better language for this, to emphasize the degree to which I’m talking about digital raw materials. Perhaps I should say write “bitstreams” when I am refering to the intrinsic qualities and save language like “digital objects” for more general cases.
An example: it’s very hard to reproduce perfect copies of a marble statue – different blocks of stone respond differently to all that chiseling, plus a sneeze at the wrong time can be decisive. Even if you did a CAD/CAM production line process to eliminate human error, say the way you might built a car, it still takes substantial time and raw materials to make the thing. Even as cheap as printing has become, it’s still non-trivial in terms of time and cost to produce extra copies of a printed book – you need paper, ink, a press – so some degree of scarcity is built into the format, too. One can additionally establish law and policy to protect those things, but the raw work of creating them is a barrier to copying in itself. Far less so for digital objects, and so much less so as to approach being no barrier at all. Indeed, just to read this comment, it has to exist in three copies – on the server storage, in the server memory, and in your computer’s memory – and along the way it’s reproduced in packets across the network, probably cached by your browser, and written to redundant storage as a backup. Also telling is that your reading of this post doesn’t preclude its reading by others, something that is not the case with printed books.
Now, it does take a huge investment in infrastructure to manufacture the computers, generate the power, write the code, and such like, but those are sunk costs. They don’t matter much on a per bit basis. Bitstreams are sort of like the second pill from a pharmaceutical corporation. The first one costs a billion dollars in R&D, but after that it’s pennies each. (Not surprisingly, pharma is another area where IP protections are incredibly important.)