Social filtering of scientific information - a view beyond Twitter
Thu, 07/01/2010 - 4:13pm | by daniel
This comment has received input from a number of FriendFeed users, as detailed in this thread, and was jointly blogged today by Björn Brembs (FriendFeed; blog post), Allyson Lister (FriendFeed; blog post) and Daniel Mietchen (FriendFeed; this blog post).
"It's not information overload, it's filter failure." (Clay Shirky)
Bonetta (2009) gave an excellent introduction to the micro-blogging service Twitter and its uses and limitations for scientific communication. We believe that other social networking tools merit a similar introduction, especially those that provide more effective filtering of scientifically relevant information than Twitter. We find that FriendFeed (already mentioned in the first online comment on the article, by Jo Badge) shares all of the features of Twitter but few of its limitations and provides many additional features valuable for scientists. Bonetta quotes Jonathan Weissman, a Howard Hughes Medical Institute investigator at the University of California, San Francisco: “I could see something similar to Twitter might be useful as a way for a group of scientists to share information. To ask questions like ‘Does anyone have a good antibody?’ ‘How much does everyone pay for oligos?’ ‘Does anyone have experience with this technique?’” It is precisely for such and many more purposes that scientists use FriendFeed, which allows the collection of many kinds of contributions, not just short text messages.
Also in contrast to Twitter, comments to each contribution are archived in that context (and without a time limit), providing a solid base for fruitful, threaded discussions. In your user profile, you can choose to aggregate any number of individual RSS or Atom 'feeds', including scientific publications you bookmark in your online reference manager (e.g. CiteULike or Connotea), your blog entries, social bookmarks (Google Reader, del.icio.us, etc.), and Tweets; and any other items you wish to post directly to your feed. You then look for other users whose profile is relevant to your work and subscribe to them. Every individual item posted in your subscriptions will then appear on your personalized FriendFeed homepage, plus optionally a configurable subset of the feeds you subscribed to. You can choose to bookmark ('like') any of these items (Facebook copied this 'like' functionality just before it bought FriendFeed), comment on them, and share discussion threads in various ways.
At first, this aggregation of information and threaded discussions might seem daunting. However, the stream of information can be channeled by organizing it into separate sub-channels ('lists'; similar to but more versatile than 'folders' in email), according to your personal preferences (e.g. one for search alerts). In addition to individual users, you can also subscribe to 'rooms' that revolve around particular topics. For example, the "The Life Scientists" room currently has 1,267 members and imports one feed.
The feature that makes FriendFeed truly useful is its social filtering system. Active discussions move to the top of your FriendFeed homepage with each new addition, which automatically brings them to the attention of you and everyone else who reads those feeds. In a sense, the most current and the most popular entries compete for attention at the top, making notifications unnecessary. This means that your choice of both rooms and subscriptions affects and filters the content you see. In that way, for instance, you could set your preferences such that you would only see papers with a certain minimum number of 'likes' among your colleagues. Alternatively, you can opt to hide items with zero likes or comments, ensuring that only those that someone found interesting will reach you. Thanks to a very fine-grained search functionality, threads also remain easily retrievable.
Some of the synergistic effects of the many scientists interacting on FriendFeed are already apparent at this early stage of adoption. FriendFeed provides a convenient way to microblog from conferences by means of dedicated threads or discussion rooms created for the event, thus allowing to share comments within and across sessions, or even with people not physically present at the meeting. Such conference coverage has even received direct (e.g. ISMB09, BioSysBio09) or indirect (e.g. ISMB08) support from the conference organizers.
Above and beyond conference coverage, scientists use FriendFeed to share papers, experiences on laboratory equipment, resources for teaching, or anything else commonly asked at mailing lists. A number of real-world scientific collaborations have already been sparked from such interactions. Collaborative grant proposals have been initiated, submitted and some of them approved after the idea was passed around and discussed on FriendFeed. Several bioinformatics problems have been solved by code-sharing and advice. Articles in scientific journals have been published by FriendFeed users after meeting and discussing on the platform [1-5].
Of course, since FriendFeed was not designed for scientists, there is room for improvement in terms of usability for scientific purposes. For instance, files can only be uploaded upon starting a thread, not while commenting on it, and there is currently no functionality which infers a measure of reputation to a user from his/her contributions (though the wide-spread use of real names somewhat allows that to be imported). As with all online contributions, citability and long-term archiving are unresolved issues, as is the permanence of services whose source code is not public. Fortunately, the development of social networks tailored to the needs of scientists is actively being pursued from various angles. The Polymath projects, in which researchers collaborate online to solve mathematical problems, provide a number of examples. The recent award of two NIH grants of over $US10M each for exactly such purposes is another. Ultimately, the continued enthusiastic adoption of the sophisticated variants of social filtering tools by a broad community of researchers interested in sharing their science will only increase the usefulness for and thus the capabilities of the online scientific community.
1 Lister, A., Charoensawan, V., De, S., James, K., Janga, S. C. C., Huppert, J., 2009. Interfacing systems biology and synthetic biology. Genome biology. 10 (6), 309+. http://genomebiology.com/2009/10/6/309
2 Saunders N, Beltrão P, Jensen L, Jurczak D, Krause R, et al. (2009) Microblogging the ISMB: A New Approach to Conference Reporting. PLoS Comput Biol 5(1): e1000263. http://dx.doi.org/10.1371/journal.pcbi.1000263
3 Neylon C, Wu S (2009) Article-Level Metrics and the Evolution of Scientific Impact. PLoS Biol 7(11): e1000242. http://dx.doi.org/10.1371/journal.pbio.1000242
4 Daub J, Gardner PP, Tate J, Ramsköld D, Manske M, Scott WG, Weinberg Z, Griffiths-Jones S, Bateman A. (2008): The RNA WikiProject: community annotation of RNA families. RNA. 14(12):2462-4 http://dx.doi.org/10.1261/rna.1200508
5. Huss & al. The Gene Wiki: community intelligence applied to human gene annotation. http://dx.doi.org/10.1093/nar/gkp760
Bonetta, L. (2009). Should You Be Tweeting? Cell, 139 (3), 452-453 DOI: 10.1016/j.cell.2009.10.017