Cameron and Shirley have just published a paper on article level metrics, in our old favourite PLoS Biology. (Aside: why PLoS Biology? I guess no one would have found it in PLoS One…)

They make the point that because of the sheer volume of scientific literature (PubMed alone, which covers just a subset of the biomedical literature, is indexing over 800,000 articles annually), we need good

filters for quality, importance, and relevance

The filter most scientists use is journal based—we have our favourite journals, general and specific, and we rarely look outside them. The journal Impact Factor (IF) is the ‘objective’ measure of this. Cameron and Shirley summarize some of the problems with the IF, and then go on to look at article-level metrics and the problem with comments.

Comments is something I’m very interested in. One of the failings of the current f1000 sites is that you, the user, can’t leave feedback or engage in any meaningful way. But commenting on papers is something that really hasn’t taken off, despite several (on-going) experiments. There are a number of reasons for this, and I think that a major one is that people just can’t be bothered logging on to different sites. That’s where Google Sidewiki comes in, except I think it’s actually quite a faff and badly implemented too.

Back in summer, a group of us from various places (in London) got together to start discussing a federated approach to communicating the social interaction around scientific objects (i.e. sharing comments between sites).

I’m starting to re-visit this, and if you would like to explore this with us drop me a line.

As Cameron and Shirley conclude,

… in the spirit of science, let’s keep learning and experimenting, and keep the practice and dissemination of science evolving for the times.

We’ll be there.