Charlotte Roueché pointed me to this transcript of a piece from ABC Radio’s Perspective slot: ‘Our Biblioclastic Century‘. The author, Robin Derricourt, an academic publisher with a background in archaeology and history, makes some well-observed points about online publication and the need for sustainability of publication and citation if we are to retain the intellectual and academic output of our culture. With none of this can I disagree. However, he then ends this short, pithy piece with the somewhat knee-jerk conclusion:
I know that my grandchildren will be able to go into a library and read an article by Einstein, a book by Newton, or a manuscript by Captain James Cook, and those by their minor contemporaries. I do not know that they will be able to access the reports, documents and articles that I can read today only on some present day institution’s website. In fact I can be pretty sure most of this will not survive.
And when our own civilisation finally ends, as each civilisation does, where will be the repository that maintains what we now have as knowledge, perhaps even through some future dark ages, for later societies to inherit? They will still have Aristotle, or Darwin, but they may not have the 21st century equivalents to read.
It is important to recognise that this is the well-thought out fear of an informed and intelligent person, and that those of us working for digital sustainability therefore need to communicate our aims and achievements more widely. I cannot help, however, but point out a logical fallacy in this argument: Derricourt assumes the existence of the physical library full of books (as well he might, the library is an institution that will not go away any time soon). But the library has not always existed, and it was by no means automatic or self-evident that the library would come into existence.
If these cultural and academic institutions had not come into being at several points in history (often associated with the courts of kings or religious communities), then books would be in no better shape that websites are now (or rather websites in the world that still exists in Derricourt’s imagination, which was the world of the early Web of the 1990s). Individual copies would have circulated in private collection, some would occasionally have been copied, but not on the scale and with the rigour that we saw in Mediaeval monasteries, for example. The idea of the repository that holds a copy of everything published in a certain domain, whatever its perceived worth, would not exist. A private collection or library could easily be burned or thrown into the trash at the end of its owner’s life, or when moving residence (and not all trash-heaps are as future-friendly as the sand at Oxyrhynchus). The library changed all this, and thanks to the libraries and scriptoria, and later printing houses and repositories, copies were made and works were preserved in multiple places, on durable materials, and with rigorous standards.
On the Web, some might say, we do not have libraries to do this job for us, and so when one private collection (a privately registered web domain, say) disappears due to its owner moving residence or losing interest or failing to keep up payments on the domain registration or service provision, all will be lost. Irrevocably and permanently. (No great loss, others would argue.) However this is not true. There are libraries in the online world. There are digital archives and repositories; the Internet Archive and various search engine caches (among other entities) may be able to recover the lost website from 1998 that Derricourt mourns. Digital libraries set out to make multiple, well-archived, backed-up copies, in open standards and formats and registered with Digital Object Identifiers, of all works in their purview. In short, there are libraries on the web. And it is not therefore true that, as Derricourt argues:
Let’s be realistic – all [sc. online content] will disappear, because no web site is permanent. Only a physical library can maintain and transmit to future generations our heritage of ideas, knowledge, discovery, speculation, literature. I can more easily find an 1898 print article than a 1998 document published on the Web.
In fact, as the world becomes more connected and the Internet becomes the source and the repository for more and more of our information, libraries are going to come under increasing pressure to cut back their accessions, to digitize and archive (or even destroy) their paper collections, and to become custodians of digital rather than physical artefacts. (Don’t get me wrong: I will be in the front line of the fight to defend libraries against this offensive, but the pressure will be there.) It is by no means automatic that physical libraries will always be the best source of cultural and literary preservation in our grandchildren’s time. If no one has bothered to digitize even a 2008 print article, then the 1998 website will be easier to find in one hundred years time. I don’t fear for websites. I fear for paper archives that no one is digitizing.
Thanks for the post, Gabby. I agree with you that there is pressure for libraries to cut back on accessions and digitize their holdings, but I don’t agree that institutional digitization necessarily equals long-term preservation. Unlike books (which can sit on a shelf for years, and assuming the environment is okay they’ll be just fine), digital content requires constant curation. And I’m afraid that a lot of libraries are jumping on the digitization bandwagon without thinking through the long-term consequences of that decision. (I hope I’m wrong) I still believe that digitization is, in fact, the way of the future (heck, it’s the way of the present) but we must ensure that digitized library content (and the born-digital research projects that many of us who read the Stoa are engaged in building) remains available for the long-term. We can do this by publishing our work under open access licenses and encouraging others to reuse our data (and to make their own work available under open access licenses) [something that, unfortunately, many libraries, at least those digitizing medieval manuscripts, are not wont to do], by taking advantage of long-term repository storage offered by our own institutions (if we’re lucky enough to have that option), and by storing our work in other publicly available repositories (I’m thinking here of the Oxford Text Archive (http://ota.ox.ac.uk/) and the Scaife Digital Library, currently under development). Multiple copies of data, located in several different repositories (very similar to the model of brink & mortar libraries, really) are more likely to survive over the long term. I think it’s likely that content being digitized today can be available in 10, 50, 100 years – but it is going to take a lot of work. Open standards and Digital Object Identifiers are only the start.