Innovative use of crowdsourcing technology presents novel prospects for research to interact with much larger audiences, and much more effectively than ever beforePosted: August 25, 2011
(Originally published in LSE Impact Blog, 25 August 2011)
In the push to make clear and unquestionable links between research and its effects on society, academics with seemingly esoteric projects might struggle to make their work accessible and interesting to the public. But projects centring on Scots language dictionaries, tattered Greek papyri and Bentham’s philosophy of utilitarianism have all made the jump through innovative use of crowdsourcing. A growing number of projects, such as Ancient Lives, Transcribe Bentham, Old Weather and Scots Words and Places, are making sophisticated use of the web to actively engage the general public as contributors to their research.
Old Weather, for example, invites the general public to transcribe naval logs, thus providing crucial meteorological data for climate scientists, as well as opening up sources for the history of the British navy. Transcribe Bentham works with a range of groups, in particular schools, to decipher the numerous papers of Jeremy Bentham. For such projects, securing user contributions is about much more than impact. They provide a venue for communities outside academia to play a meaningful role within university research, providing insight and knowledge, saving time, and facilitating the route towards high-quality outputs.
It is worth remembering that crowdsourcing predates the digital era; the Oxford English Dictionary was initially built on contributions from volunteers and there is a long tradition of active contributions from the public within many fields of the social sciences.
But the development of crowdsourcing on the internet has rapidly accelerated the sophistication of its methodologies. Recent projects have been particularly adept at using social media, developing refined mechanisms for ensuring that contributions are quality assured, working with large data sets, and creating interfaces that interact in a way that reduces complexity and confusion.
These developments mean that there are suddenly novel prospects for future projects to interact with much larger audiences than previously, and to do so in a much more effective manner.
Of course, there are plenty of research projects that do not lend themselves to this kind of public engagement whatsoever. That’s fine. But for other projects, even those that could seem recondite in nature, there are opportunities to explore.
So as crowdsourcing advances, a vital factor will be the sensitivity with which the needs and motivations of those taking part are understood. If the research community engages the public in a utilitarian sense, as just cogs in a larger research wheel, then the whole methodology will become imperilled. Understanding what moves an inhabitant of a specific community, a child in the schoolroom, or the ‘silver surfer’ with a new internet connection, and making sure their input is suitably recognised is crucial.
Engagement, as Chris Batt pointed out in his report on the topic, must be a two-way conversation “knowledge co-creation and exchange rather than simply knowledge transfer: a dialogue which enriches knowledge for mutual benefit.”
The task of the University of Oxford’s RunCoCo team was to develop guidelines for projects wishing to develop digitised collections by asking the public to upload their own content or adding information to existing resources, as happened with the highly successful Great War Archive. Equally, the Citizen Science Alliance is working according to firm principles on how to interact with their users, as articulated in Arfon Smith’s podcast on the success of the Galaxy Zoo project. Indeed, the Alliance is now looking for other researchers with whom to work with and is requesting proposals for ideas.
If crowdsourcing is to continue to be embedded in research, then it is the principles and thinking drawn from RunCoCo or the Citizen Science Alliance that need to be adopted, adapted and implemented. There is a wealth of UK research that can be enhanced by the involvement of a engaged, knowledgeable and passionate UK public.
Given the dramatic events concerning the recent riots in England, I was interested to find news of an AHRC-funded project Around 1968: Activists, Networks and Trajectories, based at the Department of Modern History at the University of Oxford.
According to the news item on the AHRC website, the project undertook “hundreds of interviews with former activists from the 1968 revolutions which shook Europe have been analysed and put online”. Useful context, I thought, for the very different types of riots that were happening in England.
However, on arriving at the website there were a number of issues that impeded me from getting access to the interviews.
First of all, I had to prove myself as a bona fide researcher. I’m not actually a researcher, but given that I work in a university context, this was not too difficult. Nonetheless it was irritating not to get immediate access.
But once into the database, it remained tricky to use the online collections.
The primitive search interface presumed one already knew exactly what type of content was in the database. There was no context to explain the precise nature of the content, nor how to browse through it a meaningful way.
Then when one did find an actual result, one was presented with a catalogue record. The interview was itself hidden away at the very bottom of the page, with little indication of what the speaker may be discussing.
The audio files were massive, some over 100MB, which tested the normally rapid Joint Academic Network (JANET) connection when the downloading process began. And rather than the files being cut up into smaller, more digestible snippets, one had to listen to the entire recording to glean any sense of it (although transcriptions for some of the material did help).
For anybody unfamiliar but interested in the archive, this all added up to a disappointing website. The site was (and remains) a great opportunity for the research team to engage with a multitude of interested parties. But they way the content was presented restricted its usage to a very narrow set of scholars.
To an extent, this is the result of the slow moving wheels of scholarly tradition.
Academics have a long and fruitful history of undertaking oral research with individuals and communities. For much of this time, there has been no feasible way to share the data collected with parties outside interested research groups.
And most importantly, those undertaking oral history are familiar with the need for anonymising or protecting the voices of those who have ‘dedicated’ their selves and their identities to a project, particularly in contested political or social areas. In legal terms, it’s called data protection, but ‘identity protection’ seems a more apt phrase.
This tradition means that researchers have not really considered disseminating their research collections. But the changing digital environment and its potential to attract and engage audiences outside the academy put a radically different spin on this. It offers researchers an ideal channel to disseminate aspects of their research to a much wider audience; an audience, consisting of taxpayers that are funding the researchers, that is showing much greater interest in why academic projects are funded. Neglecting this audience is no longer feasible.
So we need to have a different approach to compiling and disseminating oral history. Researchers must be more proactive in explaining why oral histories should be made openly available. Often, interviewees have a fear that appearing on the Internet will make their position more vulnerable. But on a world wide web that that contains billions of billions and pieces of content, where individuals expose their identities is myriad and often self-defeating ways, one oral history is often little more than a merest drop in the ocean.
And if there really are identity protection issues which need to be adhered to make the research project work, teams need to devise strategies to disseminate their research data. For example by exposing metadata, anonymising recordings, creating summaries, or editing versions of the original interviews. Although challenges exist, it is possible to make research outputs available to other researchers and a broader public, while respecting the concerns of interviewees and safeguarding privacy.
There can be no real excuse for getting half a million pounds of government funding and then allowing the fruits of that research, of a topic with relevance to contemporary concerns, to be available only to a narrow band of scholars.