Library Users of the Future... Or, projecting outward from that fringe of researchers we struggle to support today
Notes for a talk I gave at the Anybook Oxford Libraries Conference, Oxford University, 24 June 2015
The following text represents my notes rather than precisely what was said on the day and should be taken in that spirit.
<iframe src="//www.slideshare.net/slideshow/embed_code/key/uMHC2xBlPtn33Q" width="425" height="355" frameborder="0" marginwidth="0" marginheight="0" scrolling="no" style="border:1px solid #CCC; border-width:1px; margin-bottom:5px; max-width: 100%;" allowfullscreen> </iframe>
I'm not sure what I was thinking when I proposed the title of this talk. I'm not a futurologist. I'm a historian. And I was trained to spot, critique, and - ultimately - laugh at teleological leaps of faith. 'The work of X anticipated the work of Y' is anathema to me.
But for all that a grumpy view of Whiggish history precludes looking ahead within the auspices of the past, it shouldn't preclude looking ahead per se, of looking beyond the here and now into the future. For as Jo Guldi and David Armitage's The History Manifesto (2014) reminds us, speaking to the future is a gateway to power. And power is useful.
My feelings about the future of all things tend to lurch disconcertingly between super excited and grim. Every sunlit vision seems to cast a dystopian shadow. A case in point. SLIDE I recently heard about CRISPR, a defence mechanism bacteria use to defeat their bitter enemy: the virus. Most of the time viruses beat bacteria. But sometime bacteria win and when they do they snip out of sequence of DNA from the virus they've defeated and insert it into their own DNA. The next time the bacteria meets that virus, boom the bacteria wins. Why this is awesome is that scientists have figured out how to use this natural process to efficiently and cheaply trick bacteria into snipping out sequences of DNA of the scientists choosing, meaning - potentially - crops with better resistance to disease, cattle with decreased risk of developing degenerative conditions, species where physical weaknesses are suppressed, the gene pool of the human race irretrievably altered, woolly mammoths made out of elephants, SLIDE pigs with wings... You may have noticed that I slipped from awesome to grim there (that is unless you want pigs with ineffectual and unwanted wings). The future usually feels like that, a place full of polarities. The future of research libraries has the same feeling sometimes: SLIDE make everything open! worry that we make ourselves redundant in the process... SLIDE automate tagging of all the things! get moaned at when the automation fails and simultaneously make some people question the purpose of humans in the process... SLIDE digitise everything! come under pressure to burn the books... SLIDE collect lots and lots of complex born-digital data! struggle to figure out how to make it accessible...
SLIDE Nevertheless we need futurology. As David Edgerton argues in his Shock of the Old, futurology and the accompanying social obsession with novelty have been with us some time. And so by now we should be quite good at critiquing these apparently grand lurches. Frederick Morton Eden, writing in the context of poverty and poor relief in the 1790s, wrote:
With regards to mechanical knowledge, it is probable that we are still in our infancy, and when it is considered that, fifty years ago, many inventions for abridging the operations of industry, which are now in common use, were utterly unknown, it is not absurd to conjecture that fifty years hence, some new contrivance may be thought of in comparison with which the steam engine and spinning jennies, however wonderful they appear to us at present, will be considered as slight and insignificant discoveries.
The State of the Poor, 3 vols., 1797
This is futurology of the most fence sitting variety. But it has its virtues. A little self-reflection always help here. Pick up a vision statement or report on an emerging trend produced by your institution in the near past (the last two decades), map that to what has happened since in both the world and your institution, and you'll find - I suspect - a combination of optimism that was never quite met and fears that were never quite as bad as they seemed at the time. And the reality of that future you traversed? Well that was likely dominated by business as usual, by managing expectations, by getting there, by reassuring the fearful, by calming the over-excited.
SLIDE So when we talk about research library users of the future we must remember that we do two things:
- first, inevitably perhaps, we project outwards from the fringe of research library users we have today
- and second, we evoke complexities and opportunities that might vanish, persist, or harden, or all three, as that future starts to happen, that might make us and all our futurology look daft, mundane, or brilliant, or all three.
And so at the risk of looking stupid, the hope of looking brilliant, and the likelihood of looking meh when someone finds this talk on my blog in a few years time, here are a few predictions projected outward from that fringe of researchers we struggle to support because, for now at least, they lack numerical significance, if only for now.
Research Library Users of the Future will use research patterns not yoked to print and playback
SLIDE Libraries are good to supporting research that uses digital objects in forms yoked to print (for image and text), to playback (for audio-visual). They are not good to supporting research that uses digital objects in forms yoked to digital. Steven E. Jones's meditations on digitisation are useful here (Jones, 2013). In The Emergence of the Digital Humanities Jones argues persuasively that although digitisation was originally conceived as a means of replicating a thing, of converting a physical object into ones and zeros with direct preservation and access benefits, recent research in the computational humanities has revealed digitisation to be a gateway for the transformation of physical things into new research objects with their own associated affordances and challenges; a transformation that can enrich, connect, and reconfigure the original data point, the thing itself. Our presentation of digital objects forms yoked to print keep this transformative potential from the bulk of our users, they aim to ensure - to quote Siegfried Zielinski - that users are immersed 'in images and sounds without noticing the transition [that is, between the users and the machine they use] and, what is more, without knowing that one was dealing with a precisely prestructured, calculated construction of visual surfaces and temporal sequences' [Zielinski, 2008; 259]. In short, by trying to recede from view we intrude into what our users are doing. What is the alternative? SLIDE Zielinski again: 'Interfaces to these worlds must be designed to maintain an inherent tension with the worlds outside the machine'. [Zielinski, 2008; 259] Now "must" is a bit strong for our purposes, but I like this because introducing tensions into our interfaces can help researchers to develop novel use cases that help us help them exploit the affordance of digital collections that are not yoked to print. So, in the same way that putting wonky/honest OCR alongside digitised pages on Welsh Newspapers Online helps get researcher mindsets from searching and reading newspapers to mining newspapers for patterns, it helps get us libraries from deliver mechanisms characterised by posting disks full of data to infrastructures characterised by computational access that facilitate efficient, flexible mining. Here we need to follow industry, bring compute to the data (probably in some sort of cloud) and let researchers use the tools they want to use and let research that uses digital objects grow without us getting in the way. And there are models out there, from Hathi Trust Research Centre and Australia's NEctAR infrastructure in particular, that we'd do well to follow.
Research Library Users of the Future will be all about the born digital
SLIDE Okay, so perhaps not 'all about'. But consider this. The BL has, give or take, 150 million physical items. We have 57 million items in our catalogue. Since non-print legal deposit came into force in Spring 2013 we have collected over two billion webpages. In sum the collection profile of a 250 odd year memory institution has fundamentally changed in just two years. And this is not just about numbers. Hitherto the ratio of 'proper documents' to ephemera has been very much in favour of the former. This has now flipped. The same goes for personal archives. For even the beloved manuscript isn't 'safe' from the bitstream: collection bequeaths are increasingly hybrid, including 'manuscript' material held across internal memory, external storage from floppies to flash drives, data dumps, email archives, and clouds connected to smart devices: big bucket of stuff that contain a range of 'proper' documents and super-useful secondary information objects that such as unpurged download folders and browser caches that humanists are likely to find just as useful for understanding a life or a historical phenomena. These archives are packed with ephemera. And because - unlike dusty uncatalogued boxes of stuff - data begets metadata these ephemera are discoverable, and (although too big to know and check against all the compliance criteria we'd ideally like) they can be made available to our users, through discovery systems and accompanying research workflows tuned to 'proper documents'. These born-digital archives are our archival collections from now on: they are the history of the post-1996 world, of the world after the public deployment of the web. Research orientated users today don't quite know they exist, but they will soon and they'll be demanding: not least because as an undergraduate starting her degree this year was born in the 1997 (hence the mid-90s are now very legimately the past), and as digital objects born in the digital ecosystems they come with heaps of born digital metadata that, supported right, could enable research that fully exploits the promise of the digital. SLIDE That could places all this data into radical contexts the historian Tim Hitchcock has urged us to explore, and into the important contexts that Tim Sherratt - he of Trove - argued in his Digital Humanities 2015 keynote can make a difference to the lives of everyday people.
Research Library Users of the Future will need someone to help them get from A to B
SLIDE Research Libraries have for some time played a crucial role in getting researchers from A to B, whatever A and B might be. I see that role as accelerating as our users, particularly our humanities researchers, develop questions where data, metadata, and computational skills are the right tools for the job. For, if nothing else, we are good at data and metadata and probably better than we realise at computational skills. By computational skills I don't just mean the ability to write code, though - as Andromeda Yelton argued in a recent ALA Library Technology Report - it helps, I mean the wherewithal to participate in and critique computational research projects in the arts and humanities and beyond. SLIDE This November I'm experimenting with offering training of this kind. The programme of unix, Git, and Open Refine skills build on Software Carpentry (an established software skills platform aimed at scientists), The Programming Historian, and the British Library's internal Digital Scholarship Training Programme: a rolling programme of whole day course that has been offered by the Digital Research team to colleagues, and that has been warmly received, for over two years. The response to Library Carpentry - nothing I might add to do with saws, nails, sanding, or maker spaces - has been massive, we are thoroughly sold out, and this is indicative of the positive, forward-thinking state of the library profession. Building software skills across resarch libraries is important, for once we hone and build confidence in these skills we can collaborate with researchers to get them from A to B, from hand-crafted to data-crafted research. And by collaborate I mean co-produce, not support. We are at a crucial moment in the humanities where the benefits and acceptance of co-production and co-authorship have a chance to be realised. Librarians should get in on the act. SLIDE Look at what Saskia Scheltjens and University of Ghent are doing with their Library Lab, a co-production space where faculty can work with librarians to develop skills, projects, knowledge. SLIDE Or a Jisc-funded project I've been involved with where a team of librarians, academics, and research software engineers working with PG and ECR humanists to turn their research questions into computational queries run against high performance computing facilities and the data derived from those queries into visualisations using reproducible and easy to adapt scripts; work that means a lot to the researchers and that we all - scholar, engineer, librarian - are credited for.
SLIDE If these are all by sunlight visions then you might ask: SLIDE where are my pigs with wings? They are many (we at British Library Digital Research are, after all, far from techno-evangelists and are as likely to go for social solutions to problems as technological solutions, and indeed - along the way - reject the whole discourse of technological solutionism (my thoughts on Morosov and his anti-solutionism position are online elsewhere). But rather than turn the end of this talk into a rant, I'll pick just three by way of conclusion:
SLIDE That we don't unyoke ourselves from print and playback - through stasis - erect higher and higher barriers to the novel data-rich research more of our future users will want to undertake. Without going too far an forgetting out roots.
SLIDE That we fail to reassess our attitudes to risk and deny future users the full promise of born-digital collections. Without going too far and failing to uphold our responsibilities.
SLIDE And that in our desire to attend to the 99% we fail to be active participants in thee growth of the 1% of computationally enabled humanities scholars we encounter today into the fair chunk of the research library users of the future they will become. The historian Adan Crymble tweeted last week...
I wish you could borrow computer scientists from the Library. Like you can books.
— Adam_Crymble (@Adam_Crymble) July 16, 2015...perhaps we should try and become that place. Without going too far and alienating our base.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Exceptions: quotations and embeds to and from external sources