RE: Versions

["David Goodman" <David.Goodman@liu.edu>]

Heather explains very well why "self-archiving" is deliberately designed
as a temporary measure. It serves the practical purpose of getting some
useful version of a paper available to every potential reader.

I do not think its advocates have ever suggested it as a method for
permanent publication. As I understand them, they always say that not just
peer-review and version control, but permanent access is guaranteed for
"green" journals by the conventional publisher, not by the archive or the
author, and that provisions for these matters is not part of the proposed
system--only current access (now also being called supplementary access).

Certainly archives of very high quality in all these respects could be
designed, but they would be more elaborate, costly, and harder to
implement than the needed supplementary access--perhaps very close to Open
access journals.

As an aside, some of the examples in Heather's argument have resulted in
warfare. Others may just be wasted effort: it is appropriate to study the
growth of Darwin's ideas in the successive editions of his books, but this
cannot be said for every scientific paper.

I also note one type of appropriate change in archived material: when
non-persistent links change, most careful authors will (or at least
should) want to update them , at least in a note.) Will authors be tempted
to correct material that turns out from later work to be erroneous? I
would think it appropriate to place such comments in a appended note, but
I suspect some people will change the text. I even think some authors who
disagree with a referee's enforced change, or prefer their style to the
editor's, will change it right back again when they can.

Some adocates of "self-archiving" disagree with me on the importance of
these considerations; I think we all agree about the need to make use of
what techniques and systems are currently available.

Dr. David Goodman
dgoodman@liu.edu

-----Original Message-----
From:	owner-liblicense-l@lists.yale.edu on behalf of Heather Morrison
Sent:	Mon 6/21/2004 10:02 PM
To:	liblicense-l@lists.yale.edu
Subject:	Re: Versions

On 19-Jun-04, at 6:09 AM, Anthony Watkinson wrote:

> It may seem that I am trying to score points off OA protagonists but I 
> do think there is a question here which is not actually an OA/anti OA
> question. If anyone is interested in problems of authenticity see
> www.bic.org.uk/securing%20authenticity.pdf
>
> Anthony

I would agree with Anthony that the issue of different versions, while
relevant to OA, is not particularly an OA question.

There have been different versions of works since ancient history, which
has led to some interesting discussions and controversies.  There are
differences in translation of the Bible and other religious works, for
example.  Revised editions of published works, and slightly different
editions of the same work, have been around for some time.

What is different now is not really OA, but rather the flexibility of the
electronic medium.  The different versions problem is one that applies to
individual and organizations in our daily work.  When we're working on a
business document, for example, it may go through many different
iterations, and it can be a challenge to ensure that we correctly identify
the final one, particularly if this is one that is approved by our
organization or committee.

This might be an issue for the self-archiving author; when choosing a
preprint for posting, some authors will have multiple drafts to select
from, and it would be a bit surprising if the wrong draft was never, ever
chosen.

There are occasions where multiple versions are clearly beneficial, and
the electronic medium facilitiates this.  A pdf version for those who like
pretty printed pages, a text version for those who are visually impaired
or have difficulty downloading large files, an html version to allow for
easy following of links - or, like PLOS, the scientific version and the
layman version, nicely tied together.

Another potential future possibility - I haven't seen this happen yet,
although perhaps others might have - is that authors could update their
works after publishing.  In some cases, there may have been advances since
the time the article was submitted for publication.  While authors might
not wish to continually update their articles, in the situation where they
need to do their own editing before posting to an institutional archive,
perhaps it might be tempting to update at the same time?  If the author is
knowledgable and their work does not need much editing, such a version
might well be superior to the peer-reviewed published version.

Multiple versions is only a serious issue where peer review has indicated
either substantive changes (as in, an error in mathematics has been
detected leading to some change in reporting of results), or where
linguistic editing is essential to avoid misinterpretation (as in, the
author meant to say NEVER amputate but unfortunately left off the word
NEVER by mistake).

If the peer review and editing process is considered to be important, then
it is important for authors and publishers to ensure that the final,
edited work is the one that is published (with perhaps a link to more up
to date information, if warranted).  The safest way to ensure this is if
publishers provide the final, printable copy to the author for
self-archiving in institutional or disciplinary repositories.  Otherwise,
even the author who is willing to self-edit and has the best of
intentions, could easily make a mistake in copying the edits.

cheers,

Heather G. Morrison