RE: Results of the NIH Plan

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

No one expects error-free journals, let alone author's manuscripts. Given
that "ALL clinical information is potentially dangerous!" it is reasonable
to expect authors and publishers to do what they can to minimize the risk.

The primary place for eliminating such errors is in peer review, and a
responsible author chooses a journal with high standards of peer review in
order to ensure that any mistakes he might have made will be detected, and
his paper thereby improved.

After peer review, generally the final check of the paper finds a number
of insubstantial errors, and the publisher fixes them. This does not
affect the scientific quality of the paper, and it is not critical whether
the author fixes them or not.

But suppose a substantial error is found. In this context, perhaps iot is
a misplaced or missing decimal point.  The publisher will correct it--one
hopes before publication.  Any person would naturally notify the author.
Does the publisher have any special responsibility beyond that?  I think
he does, since he generally is in the special position of being the first
to detect the error.  Fortunately, even if the publisher is thus
irresponsible, some alert reader will see the error and tell the author.

Should the publisher now allow the author to correct it at the author's
own trouble and expense? The ADA considers itself justified in preventing
a person who make a dangerous mistake from correcting it-- or at least
that people who make three dangerous mistakes: leave out a decimal point,
follow the NIH regulations, and let the ADA publish their work

To be fair, that should be the ADA or another of the publishers who have
the same practice. Such publishers are not merely standing by when authors
and peer-reviewers err. They are actively requiring that potentially or
actual mistake-prone material be perpetuated.

Even some of the advocates of "Green" OA accept that "In the real world
... the ADA policy is perfectly reasonable and responsible."  That may be
the real world as seen by those who are mainly concerned about publishing
journals.

For most, the real world is where errors are made and journals cannot
always be afforded--where OA is important for readers and the public, not
only authors and their publishers. If the ADA truly cared for the
dissemination of high quality work in its important area, it would
encourage -- and even insist-- that authors adjust their versions to
correct mistakes found during or after production.  Even if the NIH were
to make this more difficult, the publisher would not hinder it further.

They would certainly not force that known errors remain. Only those  
who primarily want to destroy OA would insist on that. 

Dr. David Goodman
Associate Professor
Palmer School of Library and Information Science
Long Island University
dgoodman@liu.edu

-----Original Message-----
From: owner-liblicense-l@lists.yale.edu on behalf of T Scott Plutchak
Sent: Sun 9/18/2005 12:41 PM
To: liblicense-l@lists.yale.edu
Subject: RE: Results of the NIH Plan
 
Authors having access to the "final copy reflecting all substantive
peer-reviewing" is not the point here.  It's the fact that despite
everyone's best efforts, no one can guarantee error-free publication.

There was a case some months ago in which we (Lister Hill Library) were
alerted by letter to the fact that the abstract of a recently published
article contained a typographical error, that, if followed, would have
resulted in deadly dosages being given to pediatric patients.  I'm afraid
I don't have the citation right now, and I don't have the time to look it
up, but it was a reputable journal with good quality controls. But even
with the best of efforts, it is inevitable that sometimes these things
slip through. (That's why journals publish errata, after all.) What gives
editors of clinical journals nightmares is that on rare occasions, a typo
in one of their journals could actually kill somebody. In the case I'm
referring to, the journal was able to quickly alert its subscribers
(individuals and institutions) as well as making a correction in the
electronic version.  If, however, this had been a journal which actively
supported, to the letter and the intent, the NIH policy, they would have
had no control over the author's manuscript.

Whose responsibility is it in that case to insure that the change is made
in the author version?  And who is liable if the change is NOT made, and
because of that some infant dies?  If the journal had, as you suggest,
provided the author with the final, approved, fully peer-reviewed AND
copyedited preprint, the error would still have appeared.

You ask, "What makes you think these people would carelessly post
potentially dangerous information on the Internet?"  Nobody thinks authors
are being careless, for heaven's sake.  But ALL clinical information is
potentially dangerous!  In the real world that we have to deal with, the
ADA policy is perfectly reasonable and responsible.

T. Scott Plutchak
Editor, Journal of the Medical Library Association
Director, Lister Hill Library of the Health Sciences
University of Alabama at Birmingham
tscott@uab.edu