RE: journal and publication costs, corrected figures

[Jan Velterop <jan@biomedcentral.com>]

A better way to look at this perhaps is to take a representative sample of
publishers (broken down in categories, perhaps), take their annual journal
turnover and then divide by their annual article output. That should give
an average cost per article to academia as a whole, in the aggregate.
Input-paid open access charges, which occur only once per article, so
represent the sum-total cost to academia for that article, can then be
compared.

But the argument for open access is evolving beyond pure
'bibliotheconomics' now. Open access is so beneficial for the development
of science that to limit discussion of its impact to an economic one is
doing the concept injustice. If there is an economic argument to be made,
it should at least include the economic benefits of a greatly accelerated
research environment, in which access and re-use of published data and
literature is fully open and unimpeded.

The idea of free and unimpeded re-use of scientific material is key to
open access, indeed should be part of the definition of what open access
is. Publishing always was about re-use of scientific results. Ideas were
taken in from reading the literature, 'processed' and enriched with new
data and insights, and published again in a new form. 'Standing on the
shoulders of giants' to the nth degree. The deeply collective nature of
science is represented in the tight web of references and citations that
characterises the scientific literature. With the increasing
data-intensity of modern science, and the large amounts of data involved,
simply 'taking it in' has become impractical, and the restrictive
copyright rules on re-use have become a real impediment to progress. This
is understood and beyond controversy when it comes to scientific data.
Indeed, many journals insist that the raw data underlying research
articles should be deposited in publicly and freely accessible databases.
This concept now needs to be taken further and include research results
published in scientific articles.

Jan Velterop

> -----Original Message-----
> From: David Goodman [mailto:dgoodman@phoenix.Princeton.EDU]
> Sent: 21 January 2003 02:37
> To: liblicense-l@lists.yale.edu
> Subject: Re: journal and publication costs, corrected figures
> 
> The argument is intended to be, that if this analysis applied to not
> merely a few individual universities, but to all universities, then it
> might add up, because the hundreds of thousands of articles produced
> elsewhere would be paid for where they were produced.
> 
> As I say in another message I just posted, I do not think this data proves
> this. I think it is suggestive only. I posted it because I 
> was suprised at the approximate equivalence to the $1500 proposed by 
> Harnad et al., and I wanted to see the reactions of others.
> 
> As you say, it does not take account of the nonscientific 
> literature and the nonjournal serials. It is not easy to disentangle the 
> relative proportions of money spent on them, but let's guess that 60 % 
> of the serials expenditure at Yale is scientific serials: then 
> you're spending $6 million vs 3857 science articles or $1557/article. 
> That's not the same as $2169, but it's fairly close.
> 
> David Goodman