To bolster Adrian's argument about people not reading papers:
I periodically identify PDB entries that have not been released
becuase they are on hold until publication. But the paper was
publshed months earlier. This means that nobody has read the
paper, then tried to look at the coordinates, and then asked the
PDB to release them.
My suspicion is that many structural papers are not read beyond
the author list and title, if at all.
[I am not faulting the PDB in not identifying that the structure
should be released. Typically the author list has changed and the
title is completely different from what was submitted to the PDB.
And there are fewer of these so I suspect that the journals are
getting more reliable about communicating with the PDB. And
it looks like the PDB is better at finding these.]
Frances
=====================================================
**** Bernstein + Sons
* * Information Systems Consultants
**** 5 Brewster Lane, Bellport, NY 11713-2803
* * ***
**** * Frances C. Bernstein
* *** [email protected]
*** *
* *** 1-631-286-1339 FAX: 1-631-286-1999
=====================================================
On Thu, 10 Oct 2013, Adrian Goldman wrote:
?then the issue is to reduce the number of papers people publish: this is the
central problem in the system: nobody reads them, nobody cites them, etc etc.
There are papers out there - quite a number - that have no cites, meaning that
even the authors weren't interested in them. A long time ago, when I was at
Yale, Fred Richards said that people should be judged on their 10 best papers,
and that was all you should be asked to put into a grant or whatever.
If we (the funding agencies, governments etc etc) did this, the number of
papers would go down, there would be less rubbish to review, less money to be
made by Elsevier and the open-access journals, less money wasted on the whole
process - and even the current peer review system would work better because we
would have more time to spend on properly reviewing that little that remained.
My personal contention is that anyone who is publishing more than 10 papers a year isn't
reading and understanding their "own" work - and yet there are many senior
authors that have published 300+ papers in 10-15 years.
Adrian
On 10 Oct 2013, at 09:11, Miguel Ortiz Lombard?a
<[email protected]> wrote:
Ciao Roberto,
I'm sure the current research system works better in some fields than in
others. It depends on a number of factors, perhaps the more important of
them the amount of publications produced. Or it may be as we say in
Spain: everybody talks about the party according to how much fun is
having :-)
Agreed that peer-reviewing is a continuous, endless process. But can we
afford relying on the cleverness of the next generation to carry out our
present work and mend our present problems? That's why I tried to make
the distinction between peer-reviewing and really existing
peer-reviewing. In some fields the latter may get closer to the former,
sure. You assume that papers are read beyond their title, abstract and
conclusions, that they are read critically and understood, that when
flaws or reproducibility problems are found these are reported, that
those reports are ever widely registered by the community. All that
happens, fortunately, and more likely when the paper is a "big one". But
how often does it happen, especially in "hot" fields that produce
hundreds or thousands of papers a year? Because science is not only
about "big papers", or is it? So, is really existing peer-reviewing
actually helping separate grain from straw? How often papers acceptance
or rejection depend on factors that have hardly anything to do with science?
Again, I don't think that these problems, if they exist and are not a
product of the imagination of some of us, can be solved by simply
improving the peer-reviewing procedures.
Cheers,
Miguel Ortiz Lombard?a
Architecture et Fonction des Macromol?cules Biologiques (UMR7257)
CNRS, Aix-Marseille Universit?
Case 932, 163 Avenue de Luminy, 13288 Marseille cedex 9, France
Tel: +33(0) 491 82 86 44
Fax: +33(0) 491 26 67 20
mailto:[email protected]
http://www.afmb.univ-mrs.fr/Miguel-Ortiz-Lombardia
Le 10/10/13 08:16, Steiner, Roberto a ?crit :
Many (more) reviewers ???? - [panic on Roberto's face]
Isn't real peer-review just a question of standing the test of time?
A piece of work blatantly wrong will sooner or later be picked up by someone
(although I acknowledge that wrong papers can have serious consequences on
one's ability to
get funding). Limitations on a piece of research due to whatever reason will
be hopefully lessened by other authors or the next generation(s) of scientists.
Overall, I don't think the current system is really that bad.
Cheers
Roberto
On 10 Oct 2013, at 06:57, miguel <[email protected]>
wrote:
(Sorry if you get this twice. The first time as marked as junk by our email
server. Well, it may be junk after all...)
Hi Marco,
Impact factor is the last refuge of the publishing system as it is.
Precisely because in this ocean of untrusted publications we tend to
believe that high impact factor journals deserve our respect. This is
more or less all right: among those who have investigated the issue some
are more pessimistic than others about the quality of papers published
in those journals. Yet, it is hard to believe that their papers are
generally worse than those of not-so-high impact factor journals. But
from a scientific point of view, taking into account the evolution of
research and publishing, the trust that we give to high impact journals
is, in my opinion, wishful thinking.
Concerning peer-reviewing, I don't think that adding more opacity will
help. On the contrary. What I believe, but I don't have any proof of it,
is that peer-reviewing is useful only if it is more transparent, engages
in a real scientific discussion (understood as a conversation, not as an
exchange of messages separated by weeks) and is open to (many) more
reviewers. But that alone will not help if the way research is done does
not evolve at the same time.
On Wed, 9 Oct 2013 18:56:32 -0700, Marco Lolicato wrote:
Hi scientists,
this interesting topic brought back to my mind a similar discussion I
had with a colleague of mine and now I want to share it with you guys.
As Vale already pointed out, the peer-review process seems to be far
from an ideal system: there are many papers in which one of the author
is himself the editor of the journal in which the paper is published;
the impact factor of a journal is becoming the "only" way to judge the
quality of a paper (and of the authors) [example: one of the European
Commission grants has as mandatory eligibility criterium that the
applicant should have at least one paper published in a "high IF
journal"...I'm asking...Why?].
I have also the suspect (from my insignificant experience) that some
papers are accepted in really high IF journals without a clear
peer-review process, but basing the decision mostly on the authors
listed in that paper.
Anyway, for those reasons and more, I was wondering if maybe is
nowadays needed to revisit the peer-review process. One thing that
immediately came out was: the authors of a papers should be hidden to
both the reviewers and the editors, so that paper will be judged only
on the intrinsic quality and not from the names on it or from the
country.
I'm looking forward to see your opinion.
Marco
Il giorno 09/ott/2013, alle ore 15.00, Miguel Ortiz Lombardia ha scritto:
Hi denizens,
Now that Biology has gone missing, at least in the programs of the
funding agencies in this part of the world, the reflections that I'm
going to expose concern at best that even smaller field of natural
philosophy that we euphemistically call, not without a twist of candour,
"biomedicine". At worst, they only concern the world whose limits are
the limits of my language.
As I understand it, the main purpose of really existing peer-reviewing
is to act as a filter. By selecting those papers deemed publishable it
spares us the herculean task of reading every possible piece emanating
from our overheated brains. This actually reveals a big problem of
really existing research (with the caveat expressed in the first
paragraph). But I'm not going to venture into that problem: more clever
minds have drowned in its muddy waters. Back to the point, if the need
of publishing were not such a strong source of inspiration and we
researchers would feel the compelling necessity of publishing only when
we could write well-structured and thoughtful papers, full of useful
data and rich in new ideas and hypotheses, we could then read a
reasonable percentage of the papers concerning our fields of interest.
In that utopia, peer-reviewing could be a continuous, transparent and
open process that would involve a relevant part of the community. Not
likely to happen and probably for good: knowledge seems to progress by a
combination of slow accretion of small steps and sudden
(re)interpretations of those steps.
But what is interesting to see in that utopian/dystopian possibility is
that really existing peer-reviewing suffers from a fundamental problem:
statistical significance. Because, what significance is to be deposited
in the opinions, whether reasonably argued or not (another thorny
Pandora box I won't dare to open), of two, three or at best four people
acting as editors or reviewers? Anonymous people in the latter case, to
complete the scene.
In the tension between these requirements trust is suppose to build up
and give us a reasonable path to pursue our noble endeavours. In my
insignificant opinion, in the current state of matters, trust is
seriously broken. Too much pressure to publish, too many journals, too
much money to make from publishing, too restricted and opaque a
peer-reviewing system... As a corollary, my impression is that while
many of us suspect we live in a bubble, we all seem to tacitly expect
that we will not see it explode. A good friend of mine once offered me a
book about the Spanish Armada; no joke. Its title was "The confident
hope of a miracle".
To rebuild trust we need, among other things, to rebuild our tools. And
we better do it before the next big bang. Research is not the only human
activity involving knowledge and its transmission, we could use some
curiosity beyond our noses.
Vale.
Miguel Ortiz Lombard?a
Architecture et Fonction des Macromol?cules Biologiques (UMR7257)
CNRS, Aix-Marseille Universit?
Case 932, 163 Avenue de Luminy, 13288 Marseille cedex 9, France
Tel: +33(0) 491 82 86 44
Fax: +33(0) 491 26 67 20
mailto:[email protected]
http://www.afmb.univ-mrs.fr/Miguel-Ortiz-Lombardia
El 09/10/13 20:04, Navdeep Sidhu escribi?:
John Bohannon wrote about his experience writing "a computer program to generate
hundreds of unique papers." Thought some of you might find it of interest:
John Bohannon. Who's Afraid of Peer Review? Science 342 (Oct. 4, 2013) 60-65.
DOI: 10.1126/science.342.6154.60
http://www.sciencemag.org/content/342/6154/60.full
Best regards,
Navdeep
---
Navdeep Sidhu
University of Goettingen
---
--
Miguel
Architecture et Fonction des Macromol?cules Biologiques (UMR7257)
CNRS, Aix-Marseille Universit?
Case 932, 163 Avenue de Luminy, 13288 Marseille cedex 9, France
Tel: +33(0) 491 82 55 93
Fax: +33(0) 491 26 67 20
e-mail: [email protected]
Web: http://w2.afmb.univ-mrs.fr/Miguel-Ortiz-Lombardia
Roberto A. Steiner
Group Leader
Randall Division of Cell and Molecular Biophysics
King's College London
[email protected]
Room 3.10A
New Hunt's House
Guy's Campus
SE1 1UL
London
Phone 0044 20 78488216
Fax 0044 20 78486435
