In our first episode of 2016 I sat down with Steph Taylor and Ed Pinsent to talk about the new, and free, OAIS online course they launched last year.
As part of the plan to move away from teaching OAIS in the classroom, the team released this module as the first step towards developing an online learning offering. It was originally intended to run for four weeks, to gauge interest, and to see if maintenance of this new resource would be possible.
The Course surpassed our expectations by gathering a good deal of interest from an international audience of students, and we’ve learned that creating virtual learning is both achievable and rewarding. So it’s been decided to keep the course available online for the foreseeable future.
If you’re an information professional wishing to renew your understanding of the Open Archival Information System, or are just starting out in digital preservation – why not give it a go? The course consists of 3 modules and we estimate it will take between 4-6 hours to complete.
In this weeks episode: ‘A Beginner’s Guide to the OAIS reference model’ of the DART podcast series we discussed:
Why the team decided to make the OAIS reference model their first short online course
Why OAIS is a useful starting point for people who wish to understand digital preservation
How the reference model could change for the better – see the DPC’s review initiative and OAIS Community Wiki page for further information
We also discussed the way the online course maps OAIS to a generic preservation workflow, helping to teach the nature of the preservation continuum
Steph also gave a quick preview of her talk at the upcoming: What I Wish I Knew Before I Started: DPC Student Conference, which takes place on Friday, 22nd January in Senate House.
Episode 5: Beginner’s guide to the OAIS reference model
You can follow us on SoundCloud or subscribe to us on iTunesU.
Transcript: Beginner’s guide to the OAIS reference model
Frank Steiner: Well I’m once again sitting here with Ed & Steph, our resident digital preservation expert. And we’re starting off the New Year by talking about the OAIS reference model, and a free online course the guys launched in November last year. So tell me a bit about the new free online course. Obviously in the bigger scheme of things you run the digital preservation training programme, it’s currently face-to-face. It’s undergoing a bit of a review. We had the digital preservation and training skills survey and are currently looking at some of the results and findings. Why did you choose the OAIS reference model as a first module to put online?
Stephanie Taylor: Well, I think we’ve been teaching the OAIS reference model in the DPTP course pretty much from the beginning I think. And although we feel that it is something that’s useful for people to know it can be quite time consuming to teach face-to-face. So it was our first choice for online training so that we could point people at something before the course. And also I think because it’s a standalone thing. It’s useful for many people to just be aware of the reference model, what it does and know that it’s there. It’s referenced a lot within digital preservation, both the good and bad. So it’s useful to know what it is.
FS: I can see Ed is itching to jump in there!
Edward Pinsent: It does underpin the course, it has done for a long time. I think ever since we first set it up we wanted people to try and understand that model. And to begin with it was William Kilbride who years ago made these valiant, heroic efforts and it took him about three hours I think, three hour face-to-face teaching to try and explain what the model is. But on the other hand he was not just explaining the model, he was using it as a kind of peg to hang on some very interesting concepts about the bigger picture of digital preservation and how it would work, and developing it from there. So we tried to build on that.
And I think one of the things we like to do on a course is demystify complicated things, and this one of the complicated things we like to explain. Because people are still to this day somewhat mystified by what the model means and its terminology. It’s particularly hard to understand it if you just try and pick up the standard and read it cover to cover. It doesn’t really explain itself very clearly.
So one way or another I think we’ve tried to evolve a narrative that explains it. And one of the things we’ve done alongside that is try to map the OAIS reference model which is rather abstract and diagrammatic and vague, and try and map it to something which is actually real. And if you do this online course and you do the second module, you’ll see what I’m talking about, because it’s called a walkthrough. And the idea is that it will take you on a start to finish journey of a digital object as it passes through the stages necessary to achieve digital preservation.
So that’s our value add, in a sense that is like the second layer we’ve tacked on to the OAIS reference model, supplying a lot of things which aren’t in the model at all. It comes from our heads.
FS: Well, I think the practical example was always quite useful because if I remember correctly I think that’s also something William did in the original course.
EP: Yes, he did, definitely, practicalities. You have to keep on going down these rabbit holes as Steph calls them. Or as William calls them, the deep dive. You start trying to explain a simple thing like ingest and suddenly there’s ten or fifteen other things dependent on this apparently simple action. So that’s what we like to do. And the other thing is the jargon quite frankly, because the OAIS reference model is full of strange, weird words, which need unpacking and need to be made clear. Which again I would say if you try and read the standard cover to cover you quickly become bogged down in this rather impenetrable thicket of jargon.
FS: Sounds a bit like my PRINCE2 course I did a few years ago.
EP: Exactly, you have these terms driven into you through repetition and then after a time you just think what does it really mean? So I hope that’s the reason that it’s going to add to the value. And as Steph has said, the OAIS reference model works very nicely as a standalone course. And the initial concept that we had last year of this will just be one part of a much bigger learning environment.
FS: So the idea then is that you would use the OAIS reference module for people to get a basic understanding and then come on one of your face-to-face courses actually knowing what you’re talking about. So you shift the focus. You don’t have to stand up for three hours and explain it to them, because hopefully people will have a common level of understanding.
SP: That’s definitely one aspect of it. And I think another is that we find, particularly on the beginner course that we run, that I think the single biggest stressful thing that people have encountered before the course is usually the OAIS reference model. If you mention digital preservation somebody somewhere will soon pop up and point you at the OAIS reference model. And I think it does cause a lot of anxiety both for people coming on the courses and just generally for the community, for people that are new to digital preservation. We thought it would be a good thing as Ed said, to try and demystify that for them, make it a bit more accessible.
EP: Like it or not, it is a standard and like it or not, it’s become rather lingua franca as we might say. The common parlance, of those who talk about digital preservation, even if they support the model or not, they keep referring to it. And even if they don’t like the whole thing, they sometimes tend to like individual bits of it. Because we still like information packages for instance. An information package is a good thing. And even people who do research data management are starting to buy into the idea of an information package, which is kind of interesting. They don’t even know where it came from in some cases. Then they get the idea of a package.
ST: And I think also as digital preservation moves on that there’s certainly trends at the moment to maybe move away from OAIS in some circles. It doesn’t necessarily fit everything. We don’t want to go into that in too much detail. But again, for people that are starting out, at some point if they want to explore that and get involved in those conversations it’s useful to know what the starting point is, even if that isn’t where you end up.
FS: I always say ‘you need to know the corners before you know which one to cut’. So in a nutshell what is the OAIS reference model?
EP: It’s a way of trying to convey the idea of how digital preservation ought to work. And the way it does so is by explaining the roles people have to play and the things they have to do. It’s trying to get across the idea that planning is every bit as important as playing around with computers. And that people have as much a part to play as software or a piece of respiratory system or anything like that.
But I think the thing to stress is that it is only a model and it is a reference model. It is not meant for instance to be the blueprint for how a system would work. I can enlarge upon that. And because it is only a model it tries to be extremely generic and it never ever gets into specifics of file formats or tools or even conversation processes. It just broadly describes the ways that you would build an information package and why you’re doing that. But it keeps on stressing the planning element, the administration element are all very important.
And it tells a story as well I suppose you could say. There’s a thread of changing the object as it moves through time, because this is meant to stand for a long period of time. It’s meant to be…how can I put it, have longevity. So you really have to plan for the future. It points out a role for archival storage which is very interesting because Arkivum has pretty much built its business model on an understanding of what archival storage means, as determined in that model.
The model outlines I think six or seven things that archival storage ought to be doing in terms of how it would back up and how it validates files and how often it does so, which is by the by completely different to network storage. It’s defining a completely different way of storing the digital files specifically to get success in preservation. And Arkivum have learnt a lot of lessons from that.
FS: A working practical example of parts of the OAIS reference model then?
EP: Yes, exactly. And, the important thing here is parts of it. Because the OAIS reference model has been put down to one piece of paper where you can see everything happening on one sheet of A4, it makes people think, well, surely it’s possible to build a single system that does that. And one of the things we teach is that that’s not the case.
ST: I would say it has some very useful parts. And certainly when I first started looking digital preservation as something that I still like to teach people who’ve experienced digital objects but maybe not in a preservation sense, is the idea of the information packages that Ed has already mentioned. And I think that a very important thing to understand, the transition that an object will go through to create different information packages working through. It’s very fundamental to an understanding of digital preservation. For me one of the big negatives as Ed has already mentioned is, it is not a specification for a system. And unfortunately in one way or another it has been treated like that.
EP: Totally, you get people over the last ten years have been saying we’re going to build a system that is compliant with this model. And you’ll even get vendors saying they’ve succeeded in creating a system that’s compliant with this model and therefore it’s worth buying for these reasons. But that wasn’t the idea behind it. As I’ve said, there’s a lot of human planning elements encapsulated within this model which a machine simply can’t do. They may feed into the process but I think they stand apart from it.
Whatever you do, storage cannot be part of a system. Storage is storage; it’s servers, it’s cloud based storage, whatever. I think people have become a bit more sceptical about it now. I mean the very, very high powered digital preservation experts are so jaded and blasé now that every time they go to a conference they play preservation bingo. As soon as they see the OAIS model on a screen somewhere they score ten.
But don’t forget it is still an international standard, and by the by there’s a whole audit regime which has grown out of it which is another interesting development to do with certification for trusted digital repository status, of which OAIS plays a big part.
FS: You alluded to the fact that people are reviewing the OAIS reference model. Do you want to say a bit more about that?
SP: It’s interesting that in 2016 there is an official review of the model, and that the Digital Preservation Coalition, the DPC, will be coordinating the UK input and possibly wider, European input perhaps, into the conversations around that. So they have a Wiki up at the moment about that which we’ll put the links in the notes that’ll go with this podcast. But again, another good reason for us to put out this short course for free at the moment was to get people to have a look at the model, learn a bit about it, and then maybe carry on reading a little around it, and be able to take part in those discussions and contribute to the suggestions that we’ll all be making about where the model should go after 2016; what’s next for it.
EP: And the DPC are not only running that wiki, they’ve published two technology watch reports now by Brian Lavoie explaining the workings of the OAIS model. The second one is a quite recent update and review of this thing. But the interesting thing is, and I was hoping to see more activity on this DPC wiki, but one of the major drawbacks I think of the model, which I was hoping more people would say, at least one person has said it, is that there isn’t enough of a pre-ingest stage, put it that way. And ingest is really the point at which the system starts to take hold of a digital object and begins preservation actions. The model doesn’t really describe what happens before it gets there.
In the real world what happens in that space of time is of great interest to anybody who’s an archivist, a records manager, librarian, an information manager, a research data manager or anybody who’s dealing with digital objects that have been created by other human beings. Because we have a tremendous interest in learning what software they use to create it, what they did during the lifetime of that resource, how they’ve referred to it, how they’ve catalogued, what metadata they created it using it and so on. The long tail as some people call it, which to my mind and to other people’s mind as well, the OAIS reference model doesn’t take enough cognisance of that. And I think there needs to be a way to capture it and reflect it, and not only acknowledge it but build it into the process.
FS: That was going to be my question, because you said earlier that the OAIS reference model is quite wide and tries to be everything to everyone, and encapsulates everything. So adding this long tail, wouldn’t it make it more complicated? Should it be part of it? Should it be a separate model?
EP: Well, join the debate! Join the debate and add your voice. Perhaps it would make it more comprehensive and better integrated I think.
FS: If it was part of it was all part of the OAIS reference model?
EP: Yes, if it was part of it.
ST: And I think as well if you’ve ever considered explaining digital preservation to people that are coming new to the subject, the areas that Ed’s mentioned about pre-ingest are very, very important to get people to understand and start thinking about how they would manage that for their own systems and places that they work. And it totally ignores what is a very big part of the whole process I think.
EP: I’ll put it the other way around to you, Frank. I think anybody who’s worked as an archivist or librarian or record manager recognises the holistic view. And that the life cycle of the record and all of these other models that have been around for a long time all recognise that early intervention is essential. And in a way it’s OAIS that’s wrong and new, trying to depart from that. It fails to build on what is already a good, solid foundation of best practice which is known to work. And it makes this assumption, erroneous I would say, that the package that comes into ingest is perfect.
FS: The perfect record, if you like?
EP: Yes, but it’s hardly ever the case and records need improvement.
ST: I mean I think the review this year is very timely with it as well because I do think that’s probably a legacy thing of having been around for quite a while, digital preservation is now much more embedded into people’s organisations. Many more people are engaging with it. And I think again, all this pre-ingest and holistic view that Ed’s just outlined, all those kind of things are now pretty much everyday work for many people. So it makes sense for the model to be updated and take that into account I think.
FS: It does, but also thinking of…just like my small piece of experience and working with the research data management team upstairs, I think part of the research data management plan is you don’t think about once the record comes, you think about [inaudible 18:18] research, what systems you use. So [inaudible won me over 18:19].
ST: Well, I think research data management, the things that we’ve touched on are true of objects created in many different ways. But specifically in research data management I think it’s very important to get that right, because it really is important to get in as soon as possible with that, and to also understand things that are probably for a different podcast. But things like when an object is actually complete. So research data management is potentially making it all even more complicated. So a robust reference model that could also deal with that would be very welcome at the moment I think.
FS: Steph, I’m conscious of time, so just remind listeners, the conference you’re going to be at?
SP: Yes, we’re going to be at the Digital Preservation Coalition, DPC Conference for a student conference, it’s going to be here at Senate House in London. I’m going to be speaking about the basics of getting started with digital preservation from a non-technical point of view. So I possibly will mention OAIS as something people should be looking at.
Because it is aimed at students, people that are just starting out hopefully in a career in digital preservation we’ll also be making them aware of the online course and hope that that will be useful for them as a free course that they can take and hopefully increase their knowledge about preservation as well.
So it should be quite a good day. The morning is about digital preservation itself and then in the afternoon there’s a series of speakers talking about how they got started with their own careers that now involve digital preservation. So I think it’s going to be a very good day.
FS: Great. Thank you very much for your time today and good luck next week.
The post Dart Podcast – Beginner’s guide to the OAIS reference model appeared first on DART Blog.