Re: Self-Archiving and Journal Subscriptions: Critique of PRC

["atanu garai" <atanugarai.lists@gmail.com>]

On 23/05/07, Phil Davis <pmd8@cornell.edu> wrote:

> The economic analysis of subscription versus author-pays model 
> was first calculated for the Cornell University Library, and 
> then generalized for 113 Association of Research Libraries. 
> The second link provides a spreadsheet where basic assumptions 
> for the calculations can be modified.  There was considerable 
> debate on liblicense when these reports came out.  To avoid 
> redundancy on this list, readers are directed to view those 
> posts.
>
> http://hdl.handle.net/1813/193 [Cornell Library Report, p.26]
>
> http://hdl.handle.net/1813/236 [ARL calculation spreadsheet]
Clearly there are two aspects here: one is funding and another is 
the author's self-interest in promoting his/her research i.e. 
through citation impact. I support open access because of its 
broad goals of providing toll-free access to a large number of 
audience very righteously. But my points of contention are not 
the goal, but the ways of achieving it. The way open access model 
is designed and being practiced is not very convincing for me.

For instance, I have gone through report Phil Davis cited on open 
access implementation in Cornell and various other research 
findings. The author's motivation for submitting preprint in IR 
is very little - one can not at least at the initial stage 
predict and hope for the possible citation impact. Or, for that 
matter, authors in most cases, would give citation impact of the 
journal article a low priority, the greater priority is to 
disseminate research findings to fellow colleagues who are doing 
similar research in home or abroad. But in many cases, this is 
being done through informal communities of practice and 
close-group research scientists often share internally their 
research articles - preprint, post-print, all versions. IR and 
open archive may not be very helpful or unhelpful in that 
practice.

The question is - are the open archive and IRs best ways to 
promoting access to the research findings. Researchers and 
librarians find particular research articles using a wide variety 
of tools, most important among those are indexes and abstracts, 
not the google. Now, a typical open access or closed-access 
journal will have multiple authors and those authors will publish 
their preprints (and in some cases post-prints) in a wide variety 
of IRs and open archives, located and indexed in their own 
institutional repositories.

The problem obviously does not arise in the case purely open 
access journals, wherein all the articles are kept in the OA 
journals site itself and there is no need for the users to locate 
the pre-print version of the journal articles.

But in the case of journal with open access mandate that allow 
authors to submit their preprints or post-prints in their IRs, we 
can not guarantee that the article will be submitted in IRs. This 
proposition is not hypothetical either - consider these optional 
open access journals being published by Oxford Journals - 
http://www.oxfordjournals.org/oxfordopen/. I am wondering if 
anybody of us could locate the pre/post-print versions of the 
articles published here in the IRs of the authors. I could not do 
this.

That means, from a purely methodological point of view, it is not 
the appropriate way for the institutions to maintain their IRs 
and mandating the authors to deposit their pre/post-print copies. 
This may or may not happen. But we are here talking about a 
notion of guaranteed access to the most important articles by a 
researcher on the web. If there is no guarantee that a researcher 
would not be able to access at least a pre-print copy of the 
research article, then the very purpose of the OA is defeated. 
The researchers then obviously has to rely on the subscription 
based databases, which in the midst of urgent need can supply the 
required article.

This does not do justice to

1. business model (be it author-payee or subscription based)

2. access and citation impact.

For the purpose of taking advantages of both these objectives, 
there are two options - either a journal would be fully open 
access, meaning all costs and processes would be managed by the 
journal publisher or it has to be closed access. The apparent 
solution to find the relevant article might be to give the link 
to the IRs from the open access mandated journal articles, but if 
publishers agree to do that, there is no point in making it a 
subscription-based model. And if journal publishers are agreed to 
give the links to IR deposits, there is no need to maintain a 
separate IRs by the institutions themselves, as publishers will 
not see any business in making the final version available for 
sale.

-- Atanu Garai Globethics.net