peer reviewed journals & quality of work

by Peter Apian-Bennewitz, Physics PhD, April 2013

Around end of April 2013, I've looked into ways of publishing work of mine. Since it wasn't funded by public money nor paid work for clients, I was free to choose.
Some links that I had found on the current situation with publishers of scientific papers:

• http://www.guardian.co.uk/science/blog/2013/jan/23/open-access-publish-paywalls-victims-perpetrators
• http://www.nature.com/news/mathematicians-aim-to-take-publishers-out-of-publishing-1.12243
• http://oa.mpg.de/berlin-prozess/berliner-erklarung/
• http://p2pfoundation.net/Peer_Review
• http://futureofscipub.wordpress.com/
• http://code.google.com/p/gpeerreview/
• http://en.wikipedia.org/wiki/ArXiv

IMHO, there are strong points not to use commercial publishers, if one can avoid them. Some of the arguments:

• by standards of a commercial company, some big publishers provide a rather unstructured, badly written, dysfunctional web interface for authors
• selection of reviewers seems questionable in some cases. By my experience, and extrapolating from their feedback, quite often at least one reviewer is obviously not familiar with the subject at all. (sorry folks, your work is appreciated, not your fault)
• selection of reviewers is mostly steered by "experience" or status in the field. Well, good idea. However, under the publisher system, a young bright mind will never get a chance to review a paper of an "established" writer, independent of its quality. This system is too feudal. And it is open to prejudice/jealousy by fellow co-workers. If you believe the scientific world is 100% "pure", wake up.
• charging for download limits access for readers to those with money or access to a library at a large institution.
• just the fact that a paper appears in a "reputable journal" does not guaranty its quality per-se. This is obvious, but tends to be forgotten frequently.
  There are many examples of uncorrected errors in "official" papers, some substantial, which had escaped the review process (see above). In contract, there is a wealth of high quality information available otherwise, e.g. in the form of white papers by manufacturers or consultants.
  The quality of contents varies always and in any case, the final selection of his/her sources is left to the researcher.
• conclusion: Unless your funding or career depends critically on the number of "official" papers, why bother with money making big publishers who offer badly designed support to their authors and who limit the access of potential readers to your work ?
• alternatives: There seem to be two procedural things required for a paper (well... beside contents !):
  a) a "frozen state" of a paper and a way to reference it: filing it with an archive and getting a handle that states what was published and when. E.g. arXiv.org, see above
  b) a way for others to comment on it, - anonymously, if you must. In general, the principle to "publish first, then do a review process" seems a nice thing, since it avoids time-lags and prejudice/jealousy by reviewers. And, hey, as a reviewer, you're not paid in any case.