The Alexander Turnbull Library’s chief librarian, Chris Szekely, delivered the annual Friends of the Turnbull Library Founder Lecture to the Friends on 18 June 2009, discussing the challenges ahead for the Library and the giant steps already being taken to meet them. We have pleasure in presenting the full text of his lecture which poses and answers vital questions about how the Library will look in ten years time and beyond.
Research libraries throughout the world face similar challenges in re-thinking traditional approaches to collection management and service delivery in the context of the digital age. As New Zealand’s pre-eminent research library, the Alexander Turnbull Library is no exception. With extensive collections built up over the last 120 years that now extend into the digital realm, the Turnbull must think about ways in which it delivers to its purpose of preserving, protecting, developing and making those collections accessible to all the people of New Zealand; indeed, accessible to anyone, wherever they happen to be.
What are the challenges?
An immediate challenge is the issue of collection growth and space to store the collections. The Turnbull’s collections are large, and they continue to grow.
How do we house them in conditions that will ensure their continued preservation? We need space to store them. And ideally, appropriate space in terms of temperature, humidity and air quality. While we can certainly look at new ways to reconfigure storage arrangements to maximise the facilities we currently have, there is an inevitability of the need for more.
In this regard, the Library currently makes use of offsite storage, and it is likely that this will continue. We use this type of storage now for materials we seldom need to access, for example microfilm masters and newspapers that have been copied. However, it is not hard to imagine a future where we call upon offsite storage for materials we may need to retrieve for researchers on a more frequent basis. In other research libraries this is already a standard practice.
Will books and print die away in the digital age, and therefore lessen the need for physical space?
As things stand we are not observing a decrease in the number of publications that are produced annually in New Zealand, at least not in monograph form. And of course, unpublished material produced over the last ten, twenty, fifty years or more is likely to find its way into the Turnbull over the next ten, twenty, fifty years and further into the future. So clearly, there is no getting away from the reality that we need to find suitable solutions to the issue of space constraints.
Many of the challenges we face relate to the question of access. Traditionally, access to the Turnbull collections, for research purposes, has meant coming to Wellington and spending time in the Library. The role of the librarian in cataloguing and describing collections has been essential. If we don’t know what we have, how can researchers find what they are looking for?
The challenge of keeping on top of cataloguing and describing collections is not new, and continues to be an ongoing requirement and challenge. I am pleased to say that we expect to have electronic catalogue records to most if not all of the Turnbull’s published holdings including monographs, serials, maps and special printed collections available online within the next three years.
Describing our unpublished holdings of manuscripts and archives, photographs, and oral histories is a somewhat more complex endeavour. I will return to that a little later.
We have entered an age where there is greater opportunity to access the collections online, and indeed, an increasing expectation from researchers and broader audiences that the collections should be accessible online. Electronic catalogue records are an important starting point. Increasingly though, the expectation is for online access to the actual content of collection items. This means digitising, and creating suitable platforms for online delivery.
The Turnbull has of course, for many years made material directly available online, primarily through its pictorial website Timeframes. There are currently around seventy thousand images available in this way. To some, seventy images may sound like a lot, until one considers the pool of over five million images in the collections.
A proliferation of other Turnbull collection websites has emerged over the last few years. These range from the recently launched Manuscripts and Pictorial website, which delivers the papers of Sir Donald MacLean, through to the highly-used Papers Past website, which delivers over a million pages of 19th and early 20th century New Zealand newspapers.
What are some of the challenges that come with digitising the collections?
Choosing what to digitise is a challenge. How do you choose from nine kilometres of manuscripts and archives, two kilometres of newspapers and serials, one hundred thousand drawings, paintings and prints, five million photographic images? Until recently, the Library’s digitisation activities have been largely driven by client demand through the placement of photographic orders. Around five thousand images a year are uploaded in this way.
The issue of how to choose is best addressed by having a sensible strategy; one that considers and plans for the optimal use of available resources, and maximises the amount that can be digitised and made available online. The Library is currently drawing upon its experience and learnings derived over the last few years to craft a strategy that will be completed this year. I expect this strategy to take into account a number of drivers, including preservation imperatives, known areas of high-interest research, and general appeal.
Digitising alone does not necessarily mean that access is created. A good example to illustrate this point is Papers Past. Nearly a million pages of newspapers had been available to read on the website for some time. However, the researcher still needed to trawl through page after page with eagle eyes to find what they were looking for. It was essentially an online way of reading microfilm. It was not until the website was made text searchable that the true value of newspapers online could be effectively realised.
We need to recognise that digitising more of the collections means that those collection items become available to anyone who wants them for whatever purpose. And in that regard there are potentially millions of users around the world who may have an interest in our online collections, but who have no need or desire to access the specialist expertise we have to offer.
And that is okay. In fact, it is desirable, as we would never have the resources to meet such a demand. In some respects this is just a new version of the tactics deployed by the Turnbull in the seventies and eighties to lessen the demand on staff time and stress on original items. At that time, the National Library Reference Service and Pictorial Reference Service were created as buffers of filters. It can be argued that digitised collections online serve a similar purpose.
We also have to come to terms with the fact that our ability to protect the integrity of digitised images, and how images will be used will increasingly diminish as the volume of material increases. For some material we do have guardianship responsibilities and a facilitation role around rights management. However, while we can make explicit our expectations around what we consider to be appropriate uses, and request that users acknowledge the creators and stewardship of collection items, ultimately our ability to enforce these requirements is limited in an online environment. We must therefore think carefully indeed about what material we choose to digitise and how we make it available online.
So, what will the future Turnbull Library be like?
In ten years time, will the Turnbull continue to serve researchers who require hands-on access to physical items? Yes. Of course the collections are available now to any researcher who has the good fortune to be in or get to Wellington and come through the door of the building and find the bit of the Turnbull Library they want. (For a first time user that in itself can be a challenge). Once they have found us, they will need the requisite knowledge to use our systems and tools, or at the very least have the confidence to ask for our assistance. Could access be made easier? Undoubtedly.
It is useful here to share the definition of a research library as developed by the Guardians / Kaitiaki of the Turnbull Library.
“A research library collects and preserves research materials for future generations in order to generate knowledge. Onsite users of a research library will normally be those whose research needs require access to the physical collections or pre-digital substitute formats such as microfilm.”
In ten years time, will the Turnbull continue to serve researchers who require one-on-one access to curatorial expertise? Yes. Curators and specialists are very much a continuing and integral part of the Turnbull’s future as a research library.
The British Library recently noted in its strategic plan that the two most important assets it has are its collections and staff expertise. It is no different for the Turnbull Library. I fully expect that the next generation of researchers will continue have access to curatorial expertise, both onsite and at a distance. Access to this expertise will enhance and enrich the endeavours of researchers.
However, we cannot take the ongoing availability of this expertise for granted. If a curator leaves, there is inevitably a gap to fill that will never be filled in quite the same way, with quite the same knowledge set. We have an urgent need for more comprehensive succession planning in this regard.
The British Library’s response to this issue was the establishment of a curatorial development programme. We need something similar at the Turnbull.
In ten years time, will the Turnbull continue to collect extensively in the areas of New Zealand and the Pacific, and rare books? Yes. But again our ability to collect cannot be taken for granted. While we may seek to collect New Zealand’s contemporary published output comprehensively through legal deposit, our ability to collect effectively in other areas is again largely and appropriately dependent on our curatorial and specialist expertise. Our curators are entrusted with the responsibility to develop and build the collections. They have the skills and the knowledge. Our ability to maintain, develop and support this expertise is not assured though, and we can do better in terms of how we position our experts to do their jobs. This is an area of risk. We need that plan.
Another element critical to developing the collections is the area of donations. As we know, this Library is founded on the extraordinary gift of Alexander Turnbull. And since 1920 much of the Turnbull’s unique and expanding collection holdings have been built on the generosity and trust of thousands of donors. Many of you will recognise this is a photograph of Sir Joseph Kinsey. Upon his death in 1936 the Turnbull received the Kinsey collection, perhaps the most sizeable donation after Turnbull’s founding bequest: some 15,000 volumes as well as furniture, artefacts, manuscripts and photograph albums.
Donors continue to be fundamental to the Turnbull. Last year alone the Library accepted gifted items from over five hundred individuals and groups. Donors cannot be and are not taken for granted. Accurate information, good will and good communication are essential to maintain the confidence of donors, past, present and future. Recent media focus on this issue is a timely reminder of that.
In ten years time, will the Turnbull have ample and appropriate storage for its expanding collections? I don’t doubt we will find space. We need to consider options that include making more effective use of the spaces we currently have available to us, as well options such as offsite storage. A challenge will be to ensure that storage solutions do not unduly compromise the ability to deliver access.
Another challenge will be optimising the quality of the spaces in terms of appropriate environmental controls. The better conditions we create in terms temperatures, humidity and air quality, the more likely it is that our collections will last in perpetuity.
In ten years time, will the Turnbull have a more substantial portion of its collections digitised and accessible online? Absolutely. As mentioned earlier, we have some seventy thousand digitised pictorial items online now. It is conceivable that we will have over one million items out there over the next handful of years. This will principally comprise items from the photographic archive, as well as drawings, paintings and prints, maps and ephemera.
We have over a million pages of historical newspapers online now. We fully expect that this will grow to several million pages over the coming decade, with an increased number of titles that extend further into the early 20th century.
Apart from newspapers, we have a modest amount of other published material online as well. Te Ao Hou, and the Transactions of the Royal Society are two notable examples. Again, we can expect growth in this area, as well as opportunity to work cooperatively with other agencies; for example, the New Zealand Electronic Text Centre and university libraries.
We have a small amount of manuscript material online now, and some early learnings around appropriate online delivery systems. This amount will inevitably grow. However, perhaps not at the rate we would like. Again this is where collaborative approaches and partnerships come in play.
Some more issues to note: increasing the amount of material that is available online is a gradual process. And there are concerns, and indeed some anecdotal evidence to indicate that some researchers will focus their research efforts solely on those resources that are online; either because they mistakenly believe that that’s all there is, or because they choose not to look any further or dig deeper. Thus there is a fear that digitisation can skew the research process, or contribute to sloppy scholarship.
In this regard Turnbull is but one player in the scholarly equation. We certainly have a responsibility in terms of reader education, and should have the full breadth of our collections appropriately catalogued and described online. Of course we should do our best to make our procedures, service processes and finding aides self-explanatory and user-friendly. In the academic arena, there is much more we can do in terms of developing partnerships with tertiary providers in order to promote and provide pathways into the collections as primary research resources. Ultimately though, should a PhD student choose to base their research solely on online surrogates in preference to viewing and engaging with the real item, that’s their call.
I have heard it said many times that the Turnbull is perceived as an exclusive domain for academic researchers and the scholarly elite. I do not believe this to be true. Researchers who use the Turnbull come in all shapes and sizes: from academics to genealogists; hobby enthusiasts to senior high school students. The one thing that generally unites them is that they are motivated by a sense of purpose. They are acting on an identified research need.
While serendipitous browsing is very much a part of the public library experience it is hard to see how this could operate in a physical sense in a library which has almost all of its primary source collections necessarily protected in secure non-public areas.
At least it was hard to see until the Internet came along and with it increased expectation of ubiquitous online availability.
Consider this image (Bovril image slide). This photograph was digitised recently and uploaded to Timeframes. It forms part of a corpus of material that aligned to a nationwide programme to commemorate the 90th anniversary of the [Turnbull Library]. We also included it in a batch of photographs that was uploaded onto the Armistice Flickr commons as part of a trial (Bovril image on Flickr slide).
Almost immediately the photograph was “discovered” by someone who was part of an online group dedicated to the enjoyment of Bovril, and in accordance with the Flickr commons licence, it was reproduced on their website. The image has since been viewed many times over.
The point is, the image was of interest to a wider group, or an audience that was different to one we might have expected. An audience most likely not dedicated to the creation of new knowledge; a group that simply wanted to appreciate the image through their own lens, ie. the enjoyment of Beefy Bovril.
It begs the question: does the Turnbull Library, as a research library, have any interest in audiences who are not researchers? I suggest that we do. And we do for a number of reasons. Firstly, at a very basic level while we may have diminished ability to ‘protect’ images from the collection when they are in an online environment, we nevertheless still care about where they are and how they are being used.
But we also have an interest because we have a mandate in our legislation to make the collections accessible to all the people of New Zealand. (Ironically, most of the people who have looked at the Bovril image are overseas viewers).
The Flickr Commons experiment aimed to surface material from the world’s public photographic archives, including those of the Library of Congress, the Smithsonian and the Getty. A second aim of the Commons is to demonstrate how input and knowledge from expanded and diverse audiences can potentially help to make these collections even richer.
Here (shipwreck slide) is an example of a photograph with an inscription: Wreck of the “Derry Castle” Auckland Islands, 22 March 1887. “Rocketpad” believes that this photograph must have been taken at a later date, and appears to have reputable evidence to support this. “Rocketpad” has also posted a very helpful link to take the interested viewer to another aspect associated with the incident.
The value of this type of online sharing of information will be self-evident to experienced researchers.
Here (Ship Garthsnaid slide) is another image from the same collection, which generated many online comments ranging from a general exclamation of what an amazing shot it is, through to where the photographer must have been stationed to capture the shot, and what kind of camera he must have used. But here’s a comment from “Roshi8” (Flickr comments slide 2).
“Roshi8” believes they know who the original photographer is and has given a personal provenance. We in turn have posted a response to tease out the provenance. “Roshi8” has yet to respond, so until they do, and until they can provide a more documented source for their information, a researcher would possibly treat this provenance with a little more caution. But who knows? Maybe “Roshi8” will get back to us and validate the source.
This type of online engagement outside of the Library’s own website environment has got tremendous potential for the research community. And yes, while there may be lots of other non-research oriented comments in the online discussions, I think we would be unwise to ignore the possibilities that are open to us with these kind of opportunities.
The other thing to note here with the Flickr example is that we have deliberately pushed collection items into domains and places that attract audiences who in all likelihood would never seek to visit our website, or search our online catalogue. We are, in a sense, taking the Library out to where there are diverse audiences, rather than expecting them to come to our door, online or otherwise, and use our systems, procedures and tools.
Another obvious benefit is that wider audiences, including researchers, are exposed to the Alexander Turnbull Library, and the Library’s profile is enhanced.
There are other opportunities and benefits for the Turnbull in consciously choosing to engage with audiences beyond our usual research constituents. Let’s return to Papers Past. There is general acclaim for this website, now that we have made it full text searchable. What used to be an arduous research process of scanning through kilometres of microfilm, a process that could takes days, weeks and years, is now a process that can take minutes and hours. The benefits to researchers are enormous. This has been achieved through a process called Optical Character Recognition, or OCR. But the process isn’t perfect. And often times you will end up with situations like this (slide) where Mr “Polson” becomes Mr “Poison”. Here’s another example (Papers Past gobbledygook slide). The extent of accuracy depends on the clarity of the original item, and even with the best copies available, that quality can still be variable. What we want is good clean text, not gobbledy-gook. But realistically we just don’t have the people resource on staff to go in and re-key these errors. There are however, people out there who would love to help us with this, if we’d just let them.
Here’s a letter (letter slide) from someone in Australia. Probably what is motivating this letter is the knowledge that at the National Library of Australia where they have the Aussie version of Papers Past, you can register to do help with this. The NLA has an army of self-propelled volunteers throughout the world whose passion is to correct these types of mistakes. Incalculable people hours are spent cleaning up inaccurate OCR text.
We could be in that space too. And I fully expect in due course we will be. The point I am making, is that within that expanded audiences there are truly people out there that genuinely help us out and help the research community out.
Before I move on from Papers Past, I would like to let you know about a significant milestone on the near horizon. Next week we go live with the latest release of Papers Past. Every newspaper on the site will be fully text searchable. More newspaper titles from across the country will be available, including the Ashburton Guardian, the Kaitiaki NZ Nurses Journal and the NZ Truth. And the date range will increase. Currently the date range is up to 1920. From next week, it will inch up to 1932.
And finally, issues with the slowness of the search capability have been addressed. Papers Past will be Papers Fast.
In the coming decade, will the Turnbull Library have all of its collections appropriately described and catalogued? Yes and no. It depends on what one means by all of the collections, and what one means by ‘appropriate’. As mentioned earlier, we are fairly confident that we will get to a good place in short time with our published collections in terms of comprehensive online catalogue records.
Our unpublished collections, such as manuscripts and photographic archives are a more complex proposition. The idea that we would have an online catalogue record for every individual item is unrealistic.
Here is a hypothetical example to illustrate the point. One collection, let’s call it the Chris Szekely collection, might consist of 50 boxes of material, each with 50 folders, and in each folder there are fifty handwritten letters. That amounts to 125,000 individual items, for one collection alone.
To put this in context we have over 9 kilometres of manuscripts containing thousands of discrete collections. We therefore concentrate our initial efforts in providing descriptive records more broadly at the level of each collection, rather than the millions of individual items contained therein.
A similar situation exists in the photographic area, where we have nearly 5 million photographic images, predominantly in negative form. Again, while we have a reasonable level of intellectual control at the collection level, the idea that we would aspire to a full individual catalogue record for each individual image is daunting and ultimately unrealistic.
Yet, if we aspire to digitise portions of these collections and make them available online, it is important to attach some identification, or basic metadata to each item. So there’s the conundrum. With digitised material we are going to have to agree on a base level of description.
Again, I expect our curators and collection specialists will play a leading role in this process in terms of agreeing standards and a consistent approach to the way our collections are described online. Part of this leadership role necessarily involves working productively with others across the Library to find sensible solutions.
On the matter of leadership, I would like to take this opportunity to publicly announce the appointment of the Associate Chief Librarian with responsibility for Turnbull Research Collections. Ronald Milne has accepted this role. Ronald is currently the Director of Scholarship and Collections at the British Library. Prior to this he was the Acting Director of University Library Services at Oxford University in charge of the Bodleian Library. Ronald will be joining us in mid-September, and along with my other recently appointed Associate Librarian David Reeves, we will seek to find and implement solutions to some of these issues.
In ten years time, will the Turnbull have more substantial digitally-born collections? Undoubtedly. As we know, “digitally-born” refers to things that were created on a computer as a digital file and include such things as websites. Changes to the National Library Act in 2003 required the Library to collect electronic documents – such as websites – with an expectation that they would also be preserved in perpetuity to be available for future generations. The reality of these times is that a vast and ever increasing amount of New Zealand information is now created and published digitally and may only be available on the web for a fleeting moment. If we don’t collect it, then we face the very real possibility of a future blackout in the documented history of this country, for which future generations of researchers would not thank us. My key message here is that not collecting digitally-born material is not an option. It doesn’t mean we stop collecting works on paper, and it doesn’t necessarily mean that resources are drawn away from existing collecting endeavours.
The National Library received substantial funding to build a digital heritage archive into which Turnbull’s digitally-born holdings are deposited, preserved and retrieved from. While most of the rest of the world are still pondering on the complexities of digital preservation, New Zealand is one of the first countries in the world to have got on and done it.
It is also relevant to note that substantial baseline funding was achieved; new money, to employ additional staff to manage the increasing volume of digitally-born material. For Turnbull, that meant more staff to select and describe new digital acquisitions, and the ability to build and imbed new expertise.
As an interesting aside: digitally-born material now includes an increasing number of cartoons. Many of you will know that the Turnbull is home to the New Zealand Cartoon Archive. It is perhaps less well-known that the bulk of the cartoons we accession into the collection are received as digital files. Many cartoonists now create their works directly on computer, and deliver their cartoons via email. We have arrangements in place whereby several cartoonists, such as those featuring in major dailies, will send us the digital file of their cartoons as they create and submit them for publication. Last year alone, nearly two-and-a-half thousand cartoons were entered into the Turnbull collections as digitally-born items.
In ten years time, will the Turnbull have a more visible and coherent physical presence within the National Library building? Ideally yes. Currently the Turnbull Library research spaces inhabit bits-and-pieces of public and collection areas across four floors, linked by a fire escape and one lift – possibly the slowest lift in Wellington. I would like to see those component parts brought together into a more cohesive and consolidated space. A space that had a front door, so you knew when you were entering the Turnbull Library. A space that you point to and say unequivocally, “there’s the Turnbull Library”.
We certainly have an aspiration to do this in the context of the National Library’s building project that is currently being re-scoped. However, we are not yet at a point where we can say it is do-able. My immediate focus is on the collection storage issue, in terms of how we might prioritise project dollars.
Finally, in ten years time will the Turnbull Library still be New Zealand’s pre-eminent research library? Yes, certainly in terms of the size and nature of the collections. Our particular challenges will be around how we most effectively marshal and apply the requisite specialist expertise that is critical to a research library in terms of collection development and access. And critically, how we best leverage from the tremendous opportunities that the digital age offers us.
I began this lecture with a slide of a well-known image of Alexander Turnbull. In closing, it seems appropriate to share an image from a more-recently acquired picture of the Library’s founding benefactor. This marvellous painting of Turnbull by Gavin Hurley was purchased just last year. I like to think it presents a fresh face to Turnbull’s ongoing future role as New Zealand’s pre-eminent research library dedicated to the generation of knowledge. A library respectful of its ninety-year history as a public institution with strong traditions grounded in sound curatorial practice; a library confident to embrace a digital future for the benefit of researchers, and the broader audiences beyond.