 See, someone's notes, left behind. OK, don't know who those notes are. When most people actually first hear my job title, they usually assume that I work with the digitised collections at the Turnbull Library when I actually couldn't be further from the truth. I'm many a surprise to learn actually that I work in arrangement and description of unpublished collections and that I specialise in one digital materials. That the Turnbull Library is actively collecting, describing, preserving and providing access to born digital materials makes us pioneers. Very few collecting institutions worldwide can actually make this claim. Furthermore, they're usually even more surprised when they learn how long we've been involved in this activity. Like any frontier pioneer, we've had to learn by trial and error. Consequently, we've had to make some changes to our procedures and improve our practices. However, there's still some major challenges we have yet to resolve, but I'm getting ahead of myself. Like any good archivist, you need some context. So, the Alexander Turnbull Library is a library within the library. According to National Library Act, and I'm just quoting here, the purpose of the Turnbull Library is to preserve, protect, develop and make accessible for all the people of New Zealand the collection of that library in perpetuity and in a manner consistent with their status as documentary heritage and taonga. And I'd just like to emphasise that word in perpetuity. It makes it quite hard when you're talking about born digital collections. Within the library, there's several well-large and well-respected unpublished collections which are listed up there, which are managed by curators. Processing unpublished collections involves arrangement description on what many of us call A&D. UNESCO defines this as the process of putting archives and records into order in accordance with the accepted archival principles of governance and original order. Respect the fons. This is the guiding principle of archival theory, and I hope many of you know what I'm talking about here. Providence. This is where collections are described by the organisation or the individual that has created them. And original order is the order in which the collection was fashioned by their creator, and this is retained in the description. All our unpublished finding aids are made accessible via Ta Pui, our unpublished collection management system. And this is something I kind of have to emphasise quite a bit. So, all unpublished collections within the Turnba Library actually do contain component of born digital materials. A lot of people are actually quite surprised by that fact. But what is born digital? It is digital objects that originate in digital form. This differs quite a lot from, of course, analog materials which have been digitised or reformatted for preservation, like my colleague before me talked about. And I just like to emphasise actually this digitised image on the side here was actually taken the year I was born. And it's a man at ANZ Bank working on a computer. And of course, banks were one of the first adopters, early adopters of computers. It's 1978, by the way, giving away my age. Born digital materials actually commonly come in to us in a number of carriers, as ones actually seen here actually all within our collections. Some people probably forget those old rails. We do have those in the collections as well. And many might remember when the first zip disk first came out on the market. Pretty amazing stuff. 150 megabytes, wow. Oh, wonderful. We've even received entire computers to us, been donated to us. Sorry, Morris G's old Macintosh SE computer that's stored in our Curio's collection alongside Catherine Mansfield's typewriter. So it gives a great complement to the collections. But we also receive born digital materials via email and on cloud-based storage servers, like Dropbox as one. Born digital by its very natures fragile. They all depended on hardware and software, which is under constant threat of obsolescence. According to the conditions of the library, which I mentioned, this makes it really difficult for us to consider. Especially when you think the very first donation of any born digital objects into the collection was actually made in 1983. We commonly, from that point of time, we actually used to print out the files onto paper, put them in a folder and describe them as physical entities. This actually changed in 2000, sorry, and we started describing them as physical entities in themselves. This work was done by library staff throughout all the curatorial sections, but it quickly became evident to them that they couldn't sustain it. There was just so much coming in that they couldn't keep up with the work. So in 2007, my point when was made. There was a number of things that were evident at that time. Born digital in particular was not given priority, mainly for a number of reasons, but primarily staff resources, but also because born digital really doesn't have a physical, obvious, tangible backlog. Just goes on a server and stays there. It's invisible. They don't need to worry about it because it's not piling up around their desk. It's true. This meant the first part of my job was really to work on what was, by then, a quite considerable backlog in some of the collections. And in 2008, our wonderful digital preservation system was launched, Rosetta. Now I just want to move on to three case studies which really illuminate the changes and challenges in improving practices that we've done regarding collecting and managing our born digital materials. Many might recognise this wonderful man. Michael King is probably, well, in still many ways, one of New Zealand's most favourite and beloved, beloved, should I say, historians. We received his papers throughout his life, but we didn't receive his born digital materials until his, from his estate, which was in 2008. So quite a bit after he passed away. So much of the digital material was received nearly 20 years after its creation. And priorities meant that actually the files were not arranged and described until 2010, several years after the initial accession. And I've always actually argued, in many cases, arranging and describing a born digital collection takes a lot longer than an analogue traditional collection. You can't spread out a messy collection of digital files on a workbench and see just by glancing at them that they belong to each other and put them in piles. One thing that was really evident with this collection too was the fact that the analogue description had gone through the whole process and they'd created 32 separate series. And if you're not too short series, it means it really goes back to the whole idea of original order. It was putting them in groupings that was created by Michael King himself. And I want to emphasise, I would never create a series called Digital Files, or ones based on their file formats like word processing documents or Excel spreadsheets. Please try to avoid that. I've tried to fix mistakes like that in the past. It's not pleasant. And of course, I couldn't go back to the donor. I couldn't ask about the hardware or software in which the files were created or the nature of his work practice. For instance, it was really quite obvious. Well, I was pretty certain, about 90% certain, he composed his emails and word processing files and then pasted them into emails and sent them. But vice versa, I couldn't really tell, but that was one thing that was really evident. I'll just put it anyway. The collection actually contained 83 disks, both a Mac and PC. And you can see there's a zip disk and a couple of zip disks in there and also double density disks. This meant actually reading the disks was quite difficult. Required, I think, three different computers. Joe knows a lot about this. And the copying the files was difficult. And of course, another thing that was really difficult with this collection was the disk to actually discover doing the process of ranging and describing that 2008 accession. And they usually were found on the bottom of the dusty boxes or into file within folders. But unfortunately, the person doing the description of the time didn't document where the disks were located. So that meant it really hard for me trying to reorganise them back into that series structure that was created. I knew that they belonged there somewhere, but it was quite hard. Also, there was no clear folder structure except for the disks in which the files were saved to. Of course, if you think of the size of those disks, there was no way that one of his books couldn't... They couldn't be fit under... About three or four disks probably for one draft of his books, for instance. Like, you just couldn't tell whether or not those disks even belonged together as well because you could tell there were chapters one to nine and then 10 to 13 on one disk. But because they hadn't said that they were found together, I couldn't verify that they actually were the same draft of the... They created the same... Anyway. Yeah. And the file names were not always helpful on identifying the content. That was one that's, I think, a really big lesson with this. Just because it says it's something doesn't really mean the contents are exactly what they might say they... Well, appear to be. Yeah. And also, the duplicate file names don't actually guarantee that they're duplicates. You know, file naming conventions of a particular creator, they might just reuse that same kind of file name constantly. And with this collection, there was, from the original 939 files, there ended up being 144 files, which isn't actually quite a... There's some bigger culls that we've had in our collections, which you'll see later. And also, the fragility of these files were compounded by the delays. I guess the discovery, the disks, their technical appraisal, their arrangement and description has meant that only in this last year, the very last of his files have been preserved and made accessible. Thanks to Jay. Over there. I'll just give a little snapshot of what the collection looks like on our... That's the online channel, if you're not too familiar with it, but you can sort of see this collection description, series down the side. But, yeah, I'll just sort of get... And then moving on. Damon Farrell, his freelance photographer working in... He was based in Greymouth, but he provided an incredibly unique perspective on the West Coast way of life. He took photographs of sports teams, events of opening of the McDonald's, for instance, the Tatoo and Wearable Arts Festivals, and also privately commissioned photograph work. We purchased this collection in 2008, and it is still to this day the largest born-digital photograph collection we have. His files were downloaded from his computer, and they were handed over to Turnbull Library staff. This was the first collection, believe it or not, in which the born-digital component of the collection was given priority over the analogue collection. It was a really well-organised collection as a result, though. There were very well-organised into folders as any commercial photographer will tell you. They're pretty well-organised anyway. If the customer comes back to them and wants reprints. And the more... I guess one thing also, particularly because he was a photographer, is that the vast majority of these file names were nonsensical. They were generated by his digital camera, of course. But the one thing that's really interesting about born-digital photographs as well, in particular, is that they have a huge amount of significant properties in their metadata, such as captions, copyright information, keywords, the model maker, the camera in which the photograph was captured on, and even geotagging. And this is really, really important and actually greatly assessed with arranging and describing. However, I'd like to point out, of course, as you can see, there's still a very significant-sized chunk of analogue collection that still has yet to be processed. So, it just shows you the impact. We've put the emphasis on the born-digital, and, unfortunately, it has meant that the analogue has yet to be arranged and described. So, we haven't got it perfect. And that's just a little snapshot. And you can see some of his photographs. There's a lot there. There's a large collection of some fantastic shots in there, if you're getting any inkling from that wonderful belly dancing shot. Then moving on to Keith Locke. Keith was a long-standing Green Party politician and a son of a well-known activist, which many of you might know, Elsie Locke. We made several visits to him in his offices prior to his retirement in Parliament in 2011. And when we discussed the arrangement of his born-digital files, he was really keenly aware that there were really fragile and needed a lot of work on them. But this actually really helped us in the long run. And on the last day in office, his very last day, we went into his Parliamentary office and downloaded directly from his parliamentary computer, noting the hardware, software and so forth. And of course, one thing that was really evident with this collection was that the arrangement and description of both the analogue and the born-digital, they were described concurrently in a series structure was created using both of those. And I should actually mention that, of course, still, though, the digital specialist was working on the born-digital collection. And I'll talk about it a bit later. And the digital files were actually very well organised into clear folder structures, and there were very few technical issues with the files. However, there is a common practice within Turnbull Library, and it's maybe for other collections, where we tend to transfer collections based on format. And one thing, I'll just scroll through too. It's probably too blurry out there, but there is one, there was a large, there's several transfers, one to the ephemera collection and one to the photograph archive. The photograph archive was actually a quite extensive collection, and it actually contains 6.42 gigs of photographs, and that is yet to be arranged and described as well. So even though we described the manuscript-born digital side, there's still, yeah, still got a lot of work to do. And that's sort of got some parting thoughts really around these and these mainly questions and not many answers, but the vast majority of the born-digital materials that we receive usually comes in part of a hybrid collection, which includes traditional analogue materials, and I don't see this changing for a very long time. And it's important to note that the digital and analogue collections are created and used in conjunction with each other, and they shouldn't be in isolation from each other. For instance, just thinking back to Michael King, and with many literary collections, they will usually create a draft on their computer, print it out, do annotations, then create a new draft on their computer, and then so forth and so forth and so forth. And I don't really see, I do my daily job. I can see it. I'm not going to change my practices for a wee while, I don't think. Anyway, and furthermore, of course, as I mentioned before, the arrangement and description of born-digital is still being done by digital specialists when it's becoming more prevalent throughout the collections. Staff are being trained to gain more confidence, but this is taking a wee while. And in many cases, it's a very steep learning curve for many. And the size of collections. The size of the collections we're receiving are getting bigger and bigger and bigger. For instance, a recent donation we received contains 94,000 files. That's 194 gigs. This has a huge impact on our resources. And the tools that we use to assist with these are just not sustainable in the long term. Do we need more specialist and powerful tools? Yeah, that would be great. But is this the only solution? Do we need to collect less or more intelligently? Personal digital archiving is becoming very popular, particularly with CloudBlaze storage systems. And also the digital afterlife is becoming a big concern overseas. I mean, should our role really be about advocacy and education? It's just a question. And curatorial management. Curators really need to consider and weigh up the cost of digital preservation when appraising collections in the first instance. In the past, they used to say, yes, we want it, but they're really having to look in a file-by-file basis at some of these born-digital collections and having to reassess some of their collection policies. And I mean, the King Collection in particular is a very good case in point. It was likely we would have never gone to the links that we did to ensure that files were accessible if it was a less prominent or well-known or beloved writer. And I think it's a myth to think that born digital, just because it's digital, that everything's automated, we still handle many of these files on a file-by-file basis. And also, the curatorial management areas are getting blurred with born digital. You can no longer distinguish a collection by format anymore. It's clearly as you could in the past with analogue. Physical storage items impacted their intellectual control as issuable items. But with digital, this is no longer relevant or necessary. In fact, you want to retain a collection together for contextual purposes, dividing up and transferring it to other curatorial areas based on format just makes no sense anymore. And in 2011, actually, the Turnbull Library went under a really large restructure looking at this and we moved away from an organisation based on format to really function on what's called the matrix model. And this actually set up the arrangement description team. I should emphasise, before this time, Turnbull Library had very much silos based on curatorial sections. And what was unique about my position, I was allowed to go into each of those areas before they were all jailed into one big happy family, so to speak. But during that time period, a lot of arrangement description practices was happening in isolation from each other. A lot of people don't realise it, but there was a lot of disparity in some of the arrangement description regarding some of this stuff, and that's coming to a head. In particular, with our... I guess with the digital materials are really pushing, in one respect, the collection management systems that we have. They're really traditionally based on analogue physical archival collections. And in particular our beloved Tapui system is an age bespoke product which was developed before the adoption of international archival descriptors standards, and it's actually coming to an end of its life, really. And we're hoping to have a replacement system that is compliant with these standards so that our metadata is more easily shared with the world. Another thing, of course, is the digital era has really revolutionised the way that information is created and used. Bond digital files are easily created, copied, exchanged, conserved and deleted. This means the application of the archival principle of respect to fonts is not as easy as it was to apply. I mean, you look at, like, folders were based on how many physical files you could fit, wedge into it. Now you get digital folders on a computer that contain thousands of files. And who do you know who actually created the files in some respects when, especially in Microsoft, the beauty of Microsoft, as soon as you open a file, you might not even change it and close it. It changes the metadata and behind it and says that you are the last person and so forth. And another thing, of course, with the discoverability and the access to our collections, we acknowledge that the access and delivery is not ideal for our Bond digital. But the researchers, in honesty, are not fully engaging with these parts of our collections yet, so we really don't know what future expectations will be. I can think of a few, but there could be more. I mean, would they want a keyword search or the Bond digital content and surface only the relevant files? That would be great. Whereas the entire output of a creator, I mean, we got Judith Binney's computer, for instance, and seeing her entire files and seeing how she interacted with them, that would be wonderful. But also, digital archaeology is becoming a really big growing area of research, excavating the chips, as they call it, and understanding the obsolete technologies, which Jay knows all about. But I just want to say there's really no silver billet to any of this, and that's actually what makes it so exciting and rewarding and why I've been working with Rolf for so long. And... Yeah, any questions? Yeah. With the material that had been deleted, I'm thinking about Stafford, who had the... Sorry, Stafford had the Robert Creely archive and his hard drive, and they were coming from that email that he deleted and also putography that was in his browser cache. So, they had decided that we had to curate that material? Yeah, I mean, we get that quite a lot as well, particularly because, as you probably might not be aware, but a huge amount of our actually writers and creatives actually prefer to use Mac. And we work in a PC environment at work, and you actually get to see a lot of stuff that you get the shadow files and all sorts of things. But also, we see the bins, of course, and when you're looking at those disks, but that really comes down to a curatorial decision. And also, I guess, if they put it in the bin, they had a... They don't want it to come to us in some respects. It comes down to a question. We do bring that up every now and then. And then Michael King actually did retrieve quite a lot from his bin, particularly when there were partial drafts and things like that. But yeah, for some of the things that comes down, you just have a discussion with the curator, because I mean, I've just been given the collection to work on. Discover some things that perhaps we're never intended to be seen by people. Yeah, there's... There is some... Some... There is a lot of questions around. I think it's a case-by-case basis, really, when it comes down to it, though. So there's no set policy on... No, not really. ...safe it and then make a decision on it? We do save it, and then we make decisions, yes. But I should even say so. We do retrieve everything we can at the time, and then we make a decision what we have in front of us. Usually it comes down to that collection policy and the curator's decision around retaining particular files. Yeah, but anyway. Yeah. Oh, yeah. So are you going to do anything about the interface with the NVHA? We would love to... Oh, we've got a wish list. You can talk to that, Jay. So... We've got a wish list. It's been growing. When you say that, what do you mean? What part of the interface do you want my role? The public interface, for Browns, for getting access to the... Plenty of image viewer and... Sure. ...in the interface and... Absolutely. And I think everybody who works in the premises would say that it's not perfect, it's not beautiful, and we deserve, we want better things. And there are a few kinds of ideas of the work on behind the scenes that we want to push out. Well, obviously, these things take time. And I think fundamentally, I think what we need to hear, we need to be told the library, so maybe just email Bill and say, hey, I want some better tools, because I don't think you're doing a time... I don't even explain it in the way that I think is as best for that collection. So please say those things and say them out, please. So we can get the mandate to do something about it. So you need to leverage your relationship with its leaders because, after all, you didn't build the product with them? Yeah. So I think what we need to do in conversation spaces is not appropriate for a classic story, but my response would be, Rosensu is one product, it's a presentation mechanism, it's not an access mechanism. And that's the thing, yeah. That's the thing that preserva... Yeah, we've been concerned with the preservation, but now we are getting really into the... providing access and really mining that some of that stuff in behind some of the data that we've got, yeah, like some of those digital photographs that have a lot of background metadata to them. But, yeah. What? Ah, yeah. Yeah. It's so easy now for everyone to say everything. Oh, absolutely. I'm just thinking. Yeah. How do you place some kind of files or photos you eventually are going to come in space? Yeah. So you certainly don't run out of space on how you're going to space? No, that's true. This is the problem. I mean, that's what we're discovering. Preserving, yeah. What if you talk about advice role to the public and share it for it comes to you? Yeah, I mean, there's some... over in the States, I know the Library of Congress, they run some really interesting workshops with people, with concerns about how to manage their own collections. And I think that's a way that we can... not saying that we definitely want all your stuff to come to us, but just educating people how they can better manage their own collections, because in some cases it is just out of hand just thinking that, like, the last collection I mentioned. But there's some really interesting collections, though, sorry, and I'm probably talking too loudly into that. Another collection I worked on a wee while ago, which I didn't use as a case point, was Maura Shabbot, and if most people don't know, he suffered from Alzheimer's. And so he relied heavily on his computer in the later part of his life and expressed quite passionately his... His despair, I guess, in losing his words. And I think as a writer, isn't that the most thing you prize most? And he talked about that quite a bit. But also the actual shared disorganisation of his files actually just sort of reflected his disorganised mind. And I think some of that kind of stuff is really interesting as well. I mean, you want to retain, like, he had all these things all over the place, but I kind of questioned, just because there were drip kits everywhere, that actually sort of blishtated how scattered his memory was, and, you know, but anyway, that's... And that's a case-by-case basis, but, yeah, yeah, anyway. So, tidying him up for that big case list? Yeah, absolutely. Absolutely. And that's where context comes back again. Yeah.