As a student in a traditional condensed matter physics group, I was taught for many years that for every conference you visit, you write an article for the proceedings. In my experience it was mainly seen as a gesture to the organizers and to the community. Several times I have responded to the request of organizations like SPIE to contribute a 10-page article to a conference. In later years I was surprised to find out that this attitude toward proceedings is not shared among all researchers. So what is the role of conference proceedings in the present scientific system, should we write them, are they a waste of time, or are they perhaps worse than that?

Why proceedings may be useful
Conference proceedings have fulfilled several useful functions in the past. A conference volume provided a sense of community; by contributing one is acknowledged as an active member of the field. Conference volumes were distributed among the community in hardcopy, in which case they provided reference material for workers in the field. Thus it was an effective way of addressing the relevant people. Proceedings were a way of getting one’s work known in the community before major results were published in a peer reviewed paper. Finally, proceedings can be used to provide background information or to present data which would otherwise not be published elsewhere.

Why proceedings are a waste of time
Most importantly, writing proceedings takes time which could have been used to write a real peer-reviewed article. There is intrinsically no new information in them. Many times I have painstakingly recovered a proceedings relevant for my work only to find out that the authors had literally copied text and figures from one of their published papers. Many societies have proceedings which are inaccessible for non-members including institutions. Examples in my field are SPIE and MRS. Conference proceedings do not count in most of the important evaluations and grant applications. There is simply no box in the online forms where they should be filled in.

Why proceedings can be bad for your career
Next to their waste of time, proceedings may be potentially damaging to your CV. Proceedings, at least in physics, are not well-cited. Of course there are always the rare exceptions where proceedings have yielded a Nobel Prize. When included in bibliometric evaluations, proceedings count negatively on the scientist’s average citations per paper and add to the apparent ‘crap’ factor which is the long list of uncited work trailing the citation histogram of most of us. Some evaluation panels use the average citation per paper to assess the impact of individual scientists or of entire research institutes. As many conference papers appear in special volumes of regular journals like, in my field, Journal of Luminescence, Physica B, and Phys. Stat. Sol., it is not easy to filter them out in a computerized analysis. I can even point out rare occasions where an original conference paper was upgraded to a regular journal paper by the conference organisers. Still with no citations as a result however.

Reporting unpublished results in the form of a proceedings is dangerous and makes no sense unless this is the only way to beat the competition. Preliminary results published in conference proceedings can prevent publication of similar results in regular papers. A critical reviewer can use the argument that the authors have already published similar claims elsewhere, even though it is in an unread, obscure proceedings paper. This fact automatically means that in principle proceedings cannot contain any new results worth publishing elsewhere.

In short, the use of conference proceedings, at least in the physical sciences, seems very limited nowadays. With the exponential growth of peer-reviewed publications, becoming available online as soon as accepted, and open-access journals like the Arxiv, the average scientist has no time even to keep up with literature in its field. Therefore the arguments in favor of proceedings appear outdated.

I suggest we should stop writing and publishing proceedings altogether and use the time to do useful work or go to the beach. Most of the above arguments by the way hold for compiled volumes, which cost even more time to prepare.

Follow