Re: OA Now

[Matthew Cockerill <matt@biomedcentral.com>]

On 29 Mar 2006, at 20:34, Peter Banks wrote:

> The perfectly reasonable drive for more access to scientific 
> information does not require a movement to OA journals.
>
> In my field (Endocrinology) at least, there are far more highly 
> cited OA papers from non-OA journals than from OA journals. 
> According to ISI, there were 421709 citations in Endocrinology 
> and Metabolism in 2004. 35.3% of the total citations came from 
> only 4 journals--Journal of Clinical Endocrinology and 
> Metabolism, Endocrinolgy, Diabetes, and Diabetes Care.
The fact that, in most fields, including Endocrinology, 
established journals have, so far, continued to play a dominant 
role can't logically be used to indicate that a move towards open 
access is impossible, or unnecessary. It simply indicates that 
such a move has not yet happened on a wide scale in that 
particular field. Given the huge momentum that well established 
journals have, few expect traditional journals to disappear 
overnight.

c.f.  the fact that most music, by value, is still (I think!) 
currently bought on CD, doesn't change the fact that paid 
download of tunes looks set to become ever more important.

If you look outside Endocrinology, other subject areas provide a 
different picture. They can positive evidence of what is 
possible, which is surely more significant that negative evidence 
of what 'hasn't happened yet' in a given field.  For example, in 
Bioinformatics and Genomics, the journals BMC Bioinformatics, BMC 
Genomics, Genome Biology, Nucleic Acids Research and PLoS 
Computational Biology are leaders in the field and all make 100% 
of their research open access, with the end result that a large 
fraction of the articles and citations in that field are oa (it 
would be interesting to gather the actual stats on how high the 
proportion is)

> The first two journals, from the Endocrine Society, publish 
> reviewed manuscripts upon acceptance. The Society's Rapid 
> Electronic Publication makes research papers accessible to 
> subscribers up to 12 weeks before the print and online journals 
> are published. Diabetes and Diabetes Care are freely accessible 
> 3 months after publication, and accepted manuscripts may be 
> placed on acceptance in any institutional reposity. A Rapid 
> Electronic Publishing option is likely soon. An institutional 
> subscription to any of these journals is less than $1000--a bit 
> less than the "price of a Toyota Camry" so often mentioned as 
> the standard journal price in the general media.
>
> By comparison, BMC Endocrine Disorders is not even ranked by 
> ISI. It may well be a good journal, but the statistics aren't 
> there to cause authors to choose it.
If the scientific community relied entirely on ISI impact 
factors, no new journals would ever be started, since no new 
journal has an impact factor for the first few years. Even more 
problematically, ISI's process for deciding which journals it can 
afford to spend the money to track is highly subjective, with the 
result that many of the most highly cited new journals are not 
yet tracked.

For example, ISI's own cited reference data shows that BMC 
Biology would be in the top 5% of all tracked ISI journals, by 
impact factor, if it were tracked - and yet ISI does not yet 
track it. The problem here, if there is one, isn't with open 
access, but with the tracking systems of one particular indexing 
service. ISI is not (or certainly should not be) the ultimate and 
only arbiter of what works in science publishing.

Fortunately. ISI is not the only source of metrics as to the 
success and quality of journals - there are plenty of others 
(Scopus and Google Scholar, for a start).

Matt Cockerill
Publisher,
BioMed Central