Crowdsourcing and social engagement in libraries: the state of play
Posted by Rose Holley, 29 June 2011
As Manager of the Australian Newspapers Digitisation Program and now Trove at the National Library of Australia I have had the wonderful opportunity to utilise and promote crowdsourcing and social engagement technologies over the last three years. This would not have been possible without the ideas and assistance of everyone on our small Newspaper and Trove team. I wrote an article called „Many Hands Make Light Work‟ two years ago which outlined the activities and motivations of Australian newspaper text correctors in the first six months of service. This generated wide international interest in the library and archive world. Australia was under the spotlight for breaking new ground and demonstrating library innovation. The two main things that people asked are: “will the user activity continue?”, and “how can these technologies be applied more widely in the library/archive world?” I shall attempt to answer these two questions in this blog post. 1. Will the user activity continue? The newspaper text correction activity has expanded in leaps and bounds and continues to grow. This is despite the fact that since start up little has been done to change the process or motivate correctors and no extra staffing has been available to support users/volunteers. I remember getting very excited when a million lines of text
had been corrected after the first four months. Now we have two million lines corrected every month and this figure continues to increase. There are nearly forty million lines of text corrected to date, by over 30,000 volunteers. The Trove annual report gives highlight statistics. We think we are doing things right and other people seem to agree since we were very honoured to recently win the Excellence in E-Government Award for Service Delivery for Trove/Australian Newspapers. The keys to success of newspaper text correction seem to be:
The simplicity of the task The addictiveness of the task Volunteers wanting to help a worthy Australian cause.
I have compared and analysed the growth of user activity with other similar services such as Wikipedia and Distributed Proofreaders. This has helped me to identify and write about the top ten tips for crowdsourcing, so that librarians and archivists can make their own crowdsourcing projects a success. I also gave a presentation to the NAA/CAARA Archives 2.0 workshop on this topic. Using certain features in the design and functionality of your interface/website can make it a success. Crowdsourcing functions go hand in hand with utilising social media features. So, still being a reference librarian at heart I am going to recommend you to read a book on this topic: Malone, E, and Crumlish, C. 2009. Designing Social Interfaces: Principles, Patterns, and Practices for Improving the User Experience. O'Reilly Media. Our users are motivated by being part of an active social online community and Trove has a forum, twitter, blog and YouTube presence so that they can engage with each other and us. This is critical to maintaining the crowdsourcing activity. Social engagement has been a learning curve for us, and we have been operating in a „pilot‟ mode for the last six months. One of our team members recently gave a presentation on the social engagement pilot. 2. How can these technologies be applied more widely in the library/archive world? I think there is greater if not more potential to utilise crowdsourcing in archive collections, rather than library collections. It can be used as a means to improve our basic functions (collect, describe, organise, deliver, preserve). For example most handwritten documents in archives still cannot be effectively converted into text by a computer program (OCR) and are therefore not full text searchable. The current
method is to transcribe these documents manually into text files. Very few libraries or archives would even consider starting such a mammoth task that has no end in sight. However users of these documents particularly genealogists have a strong inherent desire to help other users. Large transcribing projects have been started by volunteers, rather than libraries or archives and have been going on for years. Most notably these include transcription of birth, marriage, death, and shipping records (i.e. people‟s names). However until recently the technology to do this was fairly clunky, it did not involve the utilisation of social engagement tools, or harness a willing crowd of library users, so involvement and activity, although good was not maximised. Now we know so much more we can put it all together and I really believe librarians and archivists have a significant role in this area. Five good examples of this are:
Transcribe Bentham from the University College London. Transcribing the Jeremy Bentham archives. Civil War Diaries Transcription Project from the University of Iowa. Digitalkoot from National Library of Finland. Uses gaming technology to correct historic newspaper text. Waisda from Netherlands Institute for Sound and Vision. Uses gaming technology to add search terms to archival films. World Memory Project from the United States Holocaust Museum. Transcribing 17 million names from Nazi records of Holocaust victims.
I have been a member of the RLG Partners Social Metadata Working Group for the last two years. We are now in the final stages of publishing three reports with our research findings. We have released a summary discussion document. This outlines our key findings and makes recommendations for how libraries and archives can effectively and easily apply some of these new technologies into their services. But technology alone is not the answer. We need to look firstly at what people want to do, then the basics of what libraries want to do, and then ask “how can we harness new technologies to achieve all our needs?” We need to learn the art of working „with‟ our users not doing things „to‟ or „for‟ them. Charles Leadbeater calls this „The Art of With‟ (this is a great read). Our users want to have meaningful opportunities to participate and contribute, and be offered the opportunity to do so; to share easily and think and work laterally; to have viable ways to collaborate and get things done. Librarians and archivists (information management experts) have the tools and knowledge to work effectively with our users and come up together with good practical ways to do these things.
The Manager of the Australian Newspapers Digitisation Program and now Trove at the National Library of Australia had the wonderful opportunity to utilise and promote crowdsourcing and social engagement technologies over the last three years.
The keys to success of newspaper text correction seem to be:
• The simplicity of the task
• The addictiveness of the task
• Volunteers wanting to help a worthy Australian cause.