RE: Berkeley faculty statement on scholarly publishing

["Matthew Cockerill" <matt@biomedcentral.com>]

> "Seen in that context, the cost of scientific publishing, even with the
> existing inefficiencies, is relatively affordable, in that it amounts
> only to a few percent of the overall cost of what the funders spend
> doing the scientific research in the first place."

Peter Banks wrote:

> To the naive--that is, most journalists and members of Congress--such
> statements seem highly plausable. A few percent? Chump change in the
> research enterprise!
> 
> Except it isn't chump change, it's a diversion of research funding and a
> net loss for science.

It is a myth that Open Access is, somehow, a costly additional option that
would have to be paid for on top of the existing publication process,
resulting in a net additional cost for science.

What are the additional costs in making research openly available,
compared to making it available online under toll-access restrictions?  
They are negligible.

As Harold Varmus and Mark Walport, amongst others, have repeatedly stated,
there is no reason to suppose that Open Access publishing will involve any
more costs than traditional publishing. The activities and costs involved
in managing peer review etc are one and the same, whether those costs are
paid upfront by the funders of the research, or after-the-fact, in the
form of toll-based access.

So from a macro-economic point of view, no additional funds should need to
be diverted from science in order to deliver open access to scientific
research publications.

If the amount of science that gets published grows, then the total costs
will go up, but that would be true whatever the model of payment/access.

> Conducting less research to support open access might make sense were
> there strong evidence to support the contention that OA will
> "dramatically increase the effectiveness of scientific communication,
> and therefore will help the progress of science." So far, however, that
> proposition rests on faith, not evidence. Effective communication does
> not consist in shoveling out reams and reams of manuscripts; it consists
> in devilvering information in a way and at a time that empowers crtical
> decision making, whether in patient care or research. Just as we now
> insist upon evidence-based medicine, we need to insist on evidence based
> informatics. A major public policy initative like OA needs more evidence
> behind it that has so far been presented.

While to some, the benefits of open access vs restricted access seem
self-evident, I do agree that it is important that these benefits should
not simply be taken for granted, but should be analysed and quantified.

Evidence for the benefits of Open Access *is* now accruing rapidly. e.g.
PNAS just reported that their open access articles, no average, receive
50% more accesses than their subscription-only articles in the first month
after publication. This strongly suggests that, even for a widely read
journal such as PNAS, there is plenty of demand for access to scientific
research articles that is failing to be satisfied by the current
subscription model. For less well-known journals, which are less widely
subscribed to, one might expect the increase in rate of access to be even
more pronounced - this can and should be tested.

Also, recent developments in bibliometrics are showing that Open Access
leads to research having more impact in terms of citations. e.g.  
http://eprints.rclis.org/archive/00002309/

Most of the citation impact studies done so far have relied on data from
ISI, which tracks only a fraction of open access journals. The arrival of
Google Scholar means that that the citation rates of *all* open access
journals can now be tracked easily, and compared to traditional journals.
This should result in some even more informative analyses in future.

Regards,
 
Matthew Cockerill, Ph.D. 
Director of Operations
BioMed Central