The question as to how we are doing as a DP community was expressly posed by Paul Wheatley of Leeds University (and avowed digital preservation geek), but it was echoed in other presentations, panels and workshops throughout the conference. How are we doing as a community? Are we making good progress or getting stuck in old issues? Are we closing the gap between theory and practice? Are we pulling in knowledge and tools from other disciplines, especially IT? Are we getting things done? Obviously, these are no simple yes or no questions. But let me share some opinions with you that were voiced this week at iPRES2012 – by Inge Angevaare
Paul Wheatley spoke at the first plenary session on Wednesday. Our host, Seamus Ross, had welcomed the participants by saying that he was impressed by the papers at this iPRES. “They show how far we have come as a field. This is rigorous, thoughtful work. It is all grounded.”
Knight: “10 Years on we are still pretty much talking about the same things”
In the following plenary keynote, Steve Knight of the National Library of New Zealand took a different view. He pulled out his notes from previous conferences, especially iPRES2008, and concluded that “We are still pretty much talking about the same things. Tools like DROID and PRONOM etc. didn’t work properly then, and they still don’t work properly now. The wish list from this year’s Future Perfect Conference (New Zealand) did not differ that much from the wish list four or ten years ago. Knight noted a “mismatch” between standard documents such as OAIS and “what we have to do now”. Knight also quoted conclusions from the Aligning National Approaches to Digital Preservation (or ANADP) conference that our present preservation systems are pretty much untested (see also my blog post then).
Wheatley: “We are duplicating efforts”
In his subsequent cRIsp presentation with Maureen Pennock, Paul Wheatley pointed at lots of duplication in digital preservation research. Now, a certain amount of duplication – or rather: trying out different models and pathways – is useful and necessary. But when you look at this list of initiatives on cost modeling, shown by Wheatley …
… you do begin to wonder if all of that is really effective and necessary.
From hobbyists to artisans to industrialists
On the previous day, during the “research challenges workshop” Knight had presented a paper by Peter McKinney describing the various stages of development in digital preservation, from hobbyist to artisan to industrialist. McKinney argues that the transitions are not clear-cut. Some organizations may still be in the hobbyist phase where others are in the artisan phase. But – and this is me speaking – it seems that very few of us are yet in the industrialist stage. But given the amount of data we are dealing with, it would seem that that is where we need to go. McKinney proposes a massive digital preservation “war games” based on real data to establish exactly where we stand and what research challenges we have. Now that would be a true international testing effort!
At the workshop not everybody agreed with terms like hobbyism and artisanship. Sheila Morrissey of Ithaka (the organization behind the Portico digital archive) argued that there has been a lot of “industrialist” work – it just hasn’t been done by “us”, the digital preservation community at this conference. It’s the likes of Google and Amazon and major cloud service providers that are doing the “industrialist” work. Are “we” reinventing the wheel?
Are commercial services maturing?
Most memory institutions are not buying into cloud solutions by large commercial partners. I need only remind you of Australian National Archives’ Michael Carden’s speech recently at ICA2012 (see blog post), who said “this is core business, and we have to do it in-house.” Library and Archives Canada Director General and CIO Ron Surette disagrees. “It is all about control. And you can still have control over information stored by commercial partners.” What about “trust”, then? Surette: “When those commercial companies fail, they have a much bigger problem than non-profit institutions. In that way, I tend to trust them more. But of course you should never rely on one supplier; you should always have a redundancy.”
Perhaps wanting to do all the work ourselves is about hating to lose our jobs, Surette speculated. Interestingly, there was a poster at the conference about the US Chronopolis distributed preservation network based at the San Diego Supercomputer Center (SDSC) entering into a behind-the-scenes collaboration with not-for-profit cloud services provider DuraCloud. And during the APARSEN session on trust (see post), Raivo Ruusalepp suggested that, as much as we may distrust commercial services providers’ solutions, for may small organizations they may be the only way to achieve some type of preservation without massive investments. Ruusalepp is a consultant for the preservation of business archives in Estonia.
Lots of modelling and frameworking in European projects
On a personal note, I may add that the likes of DuraCloud are at least providing very practical solutions for redundant storage and data health checks – even though they many not go all the way in terms of trust and preservation planning. Whereas a lot of work coming out of the major European projects (which accounted for something like 70-80% of the presentations!) seemed to focus more on theoretical work: capacity models, evaluation frameworks, etc. Which look impressive on a slide but sometimes – I admit it – go over my head (and not only mine).
Let 1000 flowers bloom, or more design?
At the end of his presentation Steve Knight asked whether we should let the proverbial 1000 flowers bloom or have more design in digital preservation. Some of the answers from the Future Perfect Conference were (click to enlarge image):
Knight also quoted a suggestion by Laura Campbell of the Library of Congress at Aligning National Approaches to Digital Preservation (2011, p. 29) to establish: “an international preservation body or association that would focus on policy aspects of digital preservation. Such a coordinating body might be aided by an advisory groups of experts to help identify what is most at risk and most important to preserve. This group could focus on content and changing forms of communication or trends in certain disciplines. Establishing a common index of already preserved content in a virtual international collection, regardless of where it is housed, could be a valuable service of such a coordinating body. It’s not about the preservation body itself, it’s about the results. Second, we might expand the notion of a national digital collection to an international digital collection. I think it’s worth talking about how such a collection might be made accessible broadly.”
To my mind, that is an interesting idea to develop. As Knight suggested, perhaps we can take our lead from the requirements for organizing digital preservation which were developed at ANADP: