A big long list of cool social tools to check out. Thanks Tony Hirst!
Charlie Rapple introduced the fourth plenary session - this one on Researchers' Social Behaviour - by announcing an experiment to take questions via Twitter.
Audience members (and those following remotely) sent questions to @UKSG and it appeared to me to be the most efficient Q&A session so far at the conference. Perhaps the size and formality of the Pentland Auditorium has discouraged some in-person questions via roving mic, and the virtual posing of questions within a 140 character limit was less daunting and more considered?
Network ecology and the knowledge economy: why researchers need to get online and social.
Tony Hirst, Open University.
Tony gave a super-speedy rundown of what social tools are out there and being used by scientists. He showed examples of the many methods by which researchers are engaging online, how they are creating objects that are themselves social, and how publishers could embrace this.
Hirst started by setting the scene of the traditional environment where if you don’t publish, you are no one. Metrics such as the h-Index give an indication of an individual author’s reputation based on how embedded their formally-published work is within other articles, known as the citation network.
These days, citations are also being generated informally (particularly in the physics field):
• An example of this is Nature Precedings, a forum for scientists to discuss preliminary findings pre-publication.
• Researchblogging.org has had good exposure with it’s (at least) fourth plenary mention at UKSG 2010; it allows blogs to be linked back to cited articles in a well-defined way so that cited articles can be informed by the discussion, and vice versa.
• PostRank is a look at social activity around blog posts e.g. # views, # comments, # backlinks, # friendfeed shares, # votes.
• F1000 allows academics to filter content, generating post ranks even before those articles become cited. Quite a formal social network essentially providing ‘pre-citation ratings’.
So whole networks of discussion are being built around an article.
Increasingly, these social networks drive how work can be discovered. Google search engine originally was different because of pagerank to determine how important a page was. But this is just one factor for them now. If you have a google acc, all your google searches are logged – now they customise results based on your previous use of results.
Social content is very connected – you can associate your google profile with e.g. your twitter account and so info such as ‘who you follow’ is used to inform what results you get from google. So your results are really yours, including for example results from people you follow on twitter. Social info is being mined heavily and it’s therefore very important to nurture and curate one’s social circle (true too in real life ;-)) and choose your friends wisely as you’re making a public declaration of who you are and what you like.
A brief warning about the risk of ‘deanonymisation’ where thought-to-be anonymous data could be tracked to individuals. For examples Netflix (for dvd rentals) released information from a competition and someone managed to identify some of the borrowers based on their borrowing behaviour, and patterns of their searches. This will be an issue with the release of govt data. BUT these fears should not prevent people from releasing data.
There are some services that do filtering for you such as ResearchGate which is kind of a facebook for researchers where you can recommend a friend in similar area.
How can publishers learn from these networks?
Participate in the conversation. Some people think twitter is a toy for the kids – not necessarily a good opinion. Twitter can be used to identify projects, to relate projects, crystallise work, to discover people at and during an event, for example Hirst created a ‘Hashtag Community’ for this #uksg conference. Also...
• Tweets can be used to automate captions on YouTube.
• Jiscri (JISC Rapid Innovation programme) - generating references for only a small element, paragraph or dataset of the full article
• Gapminder – set up different datasets
• Embedded video is increasingly popular on third party sites
• Many eyes – generate visualisations and support commenting around the data and embed in your own publications.
• Wolfram has introduced the idea of ‘living documents’.
• Tagging means that users are generating metadata for us
• Contribute back to the community – we saw yesterday how the BBC is using Wikipedia to populate their site, using its information but also contributing and enhancing Wikipedia in the reciprocally. The Chemistry community is also doing that – a CAS-number on a Wikipedia articles mean that the community has vetted it.
• Look at trails like Tesco does with its clubcard. Google gives us trails to follow for its advertising. And follows you – it remembers where you’ve been based on where you’ve mean. Leave trails on del.icio.us, Slideshare etc.
• Use Yahoo’s pipes to aggregate and mashup the web’s content
Publishers can get as much value as they want from this.
Questions from Twitter & Audience:
Q: You’ve shown a lot of social networking sites and tools, how do we keep all our networks up to date?
Hirst: Through proactivity. Every so often prune your twitter network – remove people who don’t tweet or only Re-tweet what you already know. Cull them.
Q: You talked about a lot of services but didn’t include your own opinion on some quite contentious subjects; on the subject of personalised searches, do you think it is a good thing to have google filter out results based on your past behaviour? Not just the privacy issue but what if you want stuff want the wider world, influences you haven’t come across or wouldn’t normally consult. Does it need an ‘unrecommend’ button?
Hirst: There is a danger yes but your networks are your own to cultivate so you can shape what you see. People in networks are also in other networks and these scale very quickly. But sure, I subscribe to things that are tangential to my work and want to see alternative opinions to mine, I strongly believe in serendipitous discovery. When you build networks you take responsibility.
Q: With all these sites can you recommend a good password keeper!?
Hirst: I don’t trust them and have actually developed my own algorithm which wouldn’t be wise to share with you all!