Preservation, DP 2.0 and the happy shopper

Twitter: Rory McNicholl (from #dptp slides): Is there a tool that covers the entire OAIS model? (next slide shows full diagram) “What do you think?” https://twitter.com/jisckeepit (2:33 PM Oct 22nd)

So there goes another weekend, of relaxation, reflection and little event, interrupted as usual by the weekly supermarket shopping trip. We don’t have a car at the moment, so the list has to be planned to avoid pedestrian overload. I managed to restrict my selection to just four full bags this week (not including the case of wine, but we won’t dwell on that). On returning home the goods had to be distributed to various storage destinations: cupboards, vegetable rack, fridge and freezer. This takes a little time and effort, but not a lot of intellectual exertion. Most stuff has its place. I did wonder whether the minced beef would be needed for the week, and if it should go to the fridge or the freezer (I chose fridge, and later changed my mind). I noticed that the fridge was unusually full, and found myself removing some green cheese, out-of-date yoghurt (I’m more sensitive about ‘use by’ dates than I was), and something that I decided was a less of good idea than when I had bought it (clearly my selection and appraisal skills could be better).

Those reading this blog in search of enlightenment on digital preservation rather than shopping will have seen where this is heading. This may not be digital preservation, or even information preservation, but it is preservation. There are, however, a number of lessons from this simple process:

1. None of us are specially trained in this form of supply and storage (much to the concern of food safety technologists)
2. It’s mostly straightforward but occasionally we have to think a little harder about certain decisions
3. It’s not an infallible process, as stuff gets wasted for different reasons
4. It’s not a highlight of this or any other weekend

It’s striking how most of these lessons are reversed when it comes to digital preservation. Last week I attended part of the Digital Preservation Training Programme. People were being specially trained. The content was extensive, comprehensive and, often, complex, particularly in the inter-relations between the different modules (which I had the chance to scan after attending). We are aiming to be infallible, because we can’t really admit otherwise (not yet, anyway). And it is very much a highlight, in this case aimed at actual or prospective information and digital preservation professionals. We are regularly reminded elsewhere that digital preservation must be an urgent priority.

But should we expect this to apply to everyone involved with information or repository management? After all, the food storage example is founded on experience, both real and received, indicated in food labelling and packaging and date information, and on the implementation of storage devices with simple controls and explicitly developed and refined for purpose and simplicity. If this were a digital environment, we might call it food storage 2.0.

Where is DP 2.0? According to my reading of DPTP, it’s not here yet. Let’s be clear, DPTP is mostly reflecting state-of-the-art. Simply, the art is not where it should be for wider uptake. So who should be held to account for this? Well, for one I’ll hold my hand up for the Preserv projects. We had hoped to get further, and the groundwork has been laid, so perhaps we will make it in this project. When I hear about the software tools available to help with DP, and invariably PRONOM is mentioned again, I realise we were saying this five years ago. However good PRONOM is and whatever refinements TNA has made to PRONOM in that time, notably adding DROID, it is still short of what is needed. But this is a community responsibility, not that of one agency or tool.

What’s needed for DP 2.0, especially at the more complex, technical end such as file format management and metadata, are embedded knowledge and decision bases, accessed by interfaces tied to information management and storage environments.

Widely used open source software for creating repositories has been variously criticised for creating a limited framework, e.g. in terms of ingesting content, not being “of the Web”, etc., but in fact the strength and flexibility of these services has always been based on providing a series of interfaces (deposit, manage, access) to a series of data management processes. This is all integrated and is only limited by the use to which it is put. In DP we have a single interface to a single process, or a series of tools and interfaces, but not integrated.

The good news is that we are beginning to see open knowledge registries that are accessible through interfaces that build in preservation planning support, are tied to widely used content management services such as IR software, with built in storage selection. The complexity is hidden. The content manager should not have to be an expert on file formats to manage preservation any more than the happy shopper needs to be a food technologist.