Impact Factors - the story goes on, and on...

"I first mentioned the idea of an Impact Factor (IF) in Science in 1955. With support from the National Institutes of Health, the experimental Genetics Citation Index was published, and that led to the 1961 publication of the Science Citation Index." Little did Eugene Garfield realise that journal-based IF's would still be a sourse of heated debate in the second decade of the 21st Century.

In a recent article in PLoS One, Cameron Neylon and Shirley Wu examine the case for article-based metrics and find them wanting. In their concluding remarks, they state that:

"Indeed, the fundamental problem of which paper to read can also have different contexts. Which new papers are relevant to you? Which papers should you read if you are going to pursue research question X? Which papers do you need to read before submitting your paper?"

Quite so, but I suspect that the main influencers in each of these three circumstances are: the title of the article; the name of the author(s) and the aims and scope of the journal, respectively.

IF's matter for assessment purposes, but relevance is a personal thing, not something that can be measured by the wisdom of crowds.

Are informatics tools too difficult to use?

A new RIN report, entitled "Patterns of information use and exchange: case studies of researchers in the life sciences" highlights that fact that, although different, all of the areas studied show very compex patterns of information use. Despite this, many of the groups studied used only a fraction of the information sources and tools available to them - one of the reasons given being that they just didn't have the time to learn how to use them effectively...!

Putting data on the page

I was interested to see the article "Predicting new molecular targets for known drugs" in this weeks Nature. It is a good advertisement not only for "open data", but also on how additional functionality could be added to journal articles in the future. The idea that drugs can be "magic bullets", with incredible specificity in search out a unique target turns out not to be a biologically achievable goal. Chemicals inevitably interact with a variety of targets, producing desired, undesired and unexpected effects. Some of this information is buried in the literature, much has been mined and assembled into curated public and proprietary databases. There is much to be gained by linking this information and visualizing it as a network of interactions. There are some examples in the article, and there are companies such as Ingenuity and GeneGo who do this on a commercial basis, and also develop sophisticated viewing software. Others such as Symyx and Thomson Reuters Prous simply market the data. But wouldn't it be better if the abstraction and visualization of this type of information was an integral part of the publishing process?