Sunday, November 12, 2006

Is blogging like journal peer review?

In a post A Comparison of Blogging and Journal Peer Review, Dennis D. McDonald discusses an article in February 2006's The Scientist by Alison McCook, which article had previously been reviewed in IPBiz.

McDonald highlights some of the similarities between blogging and journal publishing in peer-reviewed journals:

we in the blogosphere know that things aren't as laissez-faire and wild-west as some would have you believe. Blogging stars ("A-Listers") have evolved, authors sweat over their Technorati statistics, and ranking and rating systems such as Digg are watched very very closely by some. A mention on page one of Techmeme or Techcrunch will send some bloggers into paroxysms of euphoria.

So despite the seeming differences between the world or journal peer reviewing and the world of the blogosphere, there actually are some similarities:

Prestige and reputation, even though measured differently in the two communities, are important motivators.
Bloggers crave incoming links; journal article authors crave published citations to their journal articles.
Secondary publishing is important to both communities. For journals, abstracting and indexing services provided indexing vie structured and unstructured indexing vocabularies. For bloggers, listing in services such as Technorati or is critical.
Both "communities" rely on the actions of individuals behaving as members of a variety of different but intersecting social networks.

IPBiz notes that PageRank was loosely patterned after the "science citation index" of Eugene Garfield et al. Thus, the similarity between "craving links" and "craving citations" is hardly accidental. It was intended.

McDonald also notes: I am saying, though, that there is an underlying similarity to the two systems that has less to do with what is being published than with the role that social networks play in the process of creation, publishing, and distribution.

Google doesn't evaluate what is being published. People type in keywords and get results. The results are not ranked on "what is being published." The results are ranked based on linking. The exact SAME text will have different ranking based on measurement of links.

In the science publishing business, publication in a flagship journal such as Science and Nature guarantees a higher level of citation than if the SAME material were published in a journal of lower impact.

McDonald also stated: The intended end result [of peer review]: published articles represent high quality work and are actually improved through the review and revision process. McDonald didn't mention that the journal Science passed through two separate articles of Hwang Woo-Suk which did not represent high quality work, and which were hardly improved through the review and revision process. McDonald also didn't mention the work of Jan-Hendrik Schon, wherein third parties were communicating with the flagship journals about Schon's publications, informing the flagship journals that the work was wrong, and the flagship journals ignored the warnings.

Returning to the central theme, bloggers are not evaluated on the accuracy of what they write, but who links to them. Linking may relate to approval at some level, but it is hardly analogous to peer review. Peer review, to first order, is about the plausibility of the logic, not about whether the underlying experiments are accurate. This deficiency was abundantly manifested in the evaluation of the work of Schon and Hwang (see for example 88 JPTOS 239).

Peer review can include a discussion of uncited (or perhaps unappreciated) references. The topic of "prior art" does not come up frequently in blogging, except perhaps in discussions of plagiarism (e.g., Southern Illinois University, Ohio University, the Vermont Congressional candidate Martha Rainville). The topic of prior art is of less emphasis in scientific journals than it is in the patent business, as the current re-examination of James Thomson's embryonic stem cell patents is going to demonstrate vividly.

Look also here.


Post a Comment

<< Home