Re: In the news (Georgia State)

[Thomas Krichel <krichel@openlib.org>]

Ian.Russell writes

> 1) Strictly speaking, arXiv is an electronic preprint server so 
> the papers there may not be the published version.  Are 
> researchers in physics happy to use that version?

I can't comment on this since I am not a physist. In my time as 
an economics researcher I have seen so many mistakes in journal 
papers I don't believe in peer review at all.

There is one particular (quite well known) economist I have read 
several papers off, all contained severe problems, he just made 
up the Maths just to please him. The world would be much better 
place if there would be more public review of papers.

> Would researchers in human medicine be happy to use a version 
> of unknown providence?

Papers are still signed by authors, so the provenance is clear.

> Would librarians be happy with that situation?

The happiness of librarians is not particularly important.

> 2) The content that I am familiar with on arXiv almost always 
> associates a posted article with a journal.  Authors very 
> quickly add 'submitted to Physical Review E' or 'Published in 
> Classical and Quantum Gravity' (or whatever journal) to their 
> preprint. Why?  Well to get the authority / credibility / 
> imprimatur / brand identity of the journal.  This is tied to - 
> but not exclusively gained by - the peer review process of the 
> journal.

That's not a reason to buy a copy of the journal.

> It is very important to note that for many, many years (going 
> back to pre-web) journals have NOT been the method of primary 
> dissemination in some subjects. arXiv may provide access to 
> content, but trustworthiness and authority - for the time being 
> at least - still comes from journals (whatever business model 
> is used).

Journal papers are read by subject experts. They can judge the 
paper on its own merits, and don't need a journal to tell it is 
valuable. All they need is some channel that tells them about the 
latest papers in their field. In RePEc we do this with NEP: New 
Economics Papers, at http://nep.repec.org, a creation of mine.

> What would happen to academia if the primary mechanism of 
> identifying trustworthy content and assessing the order in 
> which to read papers was taken away?

But I never suggested to remove peer review. Simple economic 
sense dictates that those who get the benefit of peer review, 
i.e. authors, rather than readers, should pay for it.

> 3) As someone who represents society publishers I find Thomas' 
> final point very interesting.

Let me add some more elements to this idea.

Research is published to advertize skills of the academic staff 
of an institution. Institutions are in the business to maximise 
attention to the research results that are produced locally. When 
the library of institution buys access to a journal, over 90% of 
the material in that journal will contain material coming from 
other institutions, then it subsidizes attention to research 
results from other instutions. You don't need a PhD in economics 
to see that this makes no economic sense. A rational institution 
will pay nothing for research produced elsewhere and will spend 
all its efforts to make its results widely available.

> I would be even more interested to hear any ideas for 
> mechanisms to facilitate the flow of money away from library 
> acquisition budgets to scholarly societies.

Attention as per the previous paragraph comes not only from prime 
publication, but can also come from hosting and supporting 
aggregates of contents, enabling new peer review tools etc, and 
that would be ideally done with scholarly societies.

However, this will not happen because library staff don't have 
the required IT skills. That in turn is the failure of library 
schools. In my own school I have been trying insist on more 
technology courses, but my preachings are falling on death ears.

With cheers from Novosibirsk (cloudy, 20C)

Thomas Krichel                    http://openlib.org/home/krichel
                               RePEc:per:1965-06-05:thomas_krichel
phone: +7 383 330 6813                       skype: thomaskrichel