How a journal should be

Recently, thanks to my friend Peppe Liberti, I read this interesting paper by Jason Priem and Bradley M. Hemminger, two researchers at UNC, about the “decoupled journal”. The paper is interesting because it is about a topic I have been thinking on lately by myself. Now I want to share my thoughts with a large audience.

There is something in the scientific publication system that is just not going in the right direction. Most researchers are focusing their attention on the peer-review model for selecting papers. True, peer-review is less than perfect, and maybe there is a better method to select good papers from bad papers. However, I agree with Priem and Hemminger that this is not the real problem. The real problem is with the system of journals for disseminating scientific knowledge.

The problems is that journals cost too much; many researchers, especially from poor countries, cannot afford to pay a huge amount of money for subscription to journals. For my university, we are talking of more or less 1 million euros for subscribing to the most relevant journals in all disciplines, and for example they only subscribed to IEEE and not to ACM. That is a lot of money. Moreover, closed access to publications reduces the visibility of a paper; therefore, most researchers now just put the pdf of their papers on their web site for free download, bypassing or just ignoring the copyright (that has been duly transferred to the editor of the journal).

However, if revenues from subscriptions are completely cancelled by allowing a “free for all” self publishing rule, who is going to pay the editors?  How to support the whole process? The whole system is in danger!

The Open Access model

The solution that has been proposed by some editors is “the authors of the paper should pay”. Therefore, with Open Access, the publication process is modified as follows:

  • authors of prospective papers can submit to the Open Access journal freely
  • the paper follows a regular, traditional,  review process
  • if the paper is accepted, the authors must pay a publication fee, usually proportional to the number of pages
  • once the paper is published, it is made available for free forever on the editor web-site

I don’t like Open Access journals, because in my opinion there are several problems with this approach. The first one is that the fee is quite high: typically, I have seen  approximately 100$ per page, therefore a paper with 10 or 15 pages costs about 1500$. The editor I have been in contact uses a small “page” and a large character size, in order to maximize their income (of course). This cost must be charged on research funds, increasing the overall cost of doing research. So, now researchers from poor countries will be able to read freely, but will find problems  in publishing: it does not look like a great advancement.

The second problem is on the goal. These editors are naturally interested in publishing as much as they can, because their revenues are proportional to the amount of papers they publish, not to the amount of readers. This goes in the direction of encouraging publishing at all cost. Will we have a lot of papers that nobody reads?

As you may know, government agencies that fund universities are now asking for “quality” measures, and most of these are based on counting the number of papers. As an example, the “Legge Gelmini” that reforms Italian universities will require a wanna-be researcher to have published a certain number of journal papers in order to access the selection process. Therefore, young researchers will be encouraged to publish more and more, just for the sake of passing the limits and gain access to the profession. And guess what is happening? Editors of Open Access journal are actively pushing young researchers to submit papers, and senior researcher to be guest editors of special issues. I personally receive in my e-mail Inbox an average of one invitation per day to submit to an Open Access journal.

I don’t think all of this goes in the right direction. So, what to do?

My ideal

In my opinion, the old model (the reader pays) is not that bad, we can still use it. So I came with the following considerations:

  1. Let’s get rid of the hard paper entirely. Every researcher has a printer in his office or in his lab.
  2. Let’s get rid (eventually) of the pdf format. A paper is much more that a sequence of characters; it has links to other papers, data to be analysed and compared and re-used by other researchers, code snippets or simulation code, etc. Let’s make every paper an hypertext, for example using ePub as a common format (or any other standard, it would just be fine).
  3. The main service provided by a journal is to store and catalog papers, make sure they do not change, assigning them a unique number (a DOI), and assess the quality with peer review. All of this can be done at a very low cost, if everything is managed electronically. In fact, the hardest work (reviewing) is done for free by the research community!
  4. Therefore, we can reduce the cost a lot. For example, a subscription to a set of journals should cost as little as a few dollars per months. If a service like lastfm costs 2$ per month , an editor can probably offer the basic services roughly at the same order of money.

But a modern journal can offer much more!

  1. Publishing of (amended) reviews, to understand how and why the paper was accepted;
  2. The possibility to download additional material (data sets, graphs, code, etc.);
  3. A continuous interaction between the public and the authors, using for example a public comment forum for each paper, where only registered (and paying) users can comment;
  4. A ping-back service, with which authors are notified when someone is citing their work;
  5. RSS feeds of recent papers, Editor’s pick, commentaries, etc.

Basically, the idea is to apply some of the techniques used in social networks (with obvious care). Then, fantasy is the only limit.

The low cost of access should make the issue of authors publishing their own work on the web a non-issue: searching on Internet is not the same as searching on a dedicated web site, which can provide many more services. Eventually, I think many researcher will just subscribe for such a low amount.

A dream?

Of course, there is still a long way to go. The technicalities are not completely solved yet (for example, it is not possible to properly format equations on ePub: the only way to do it is to produce a small image to be embedded in the file). Also, I think many big editors will not easily renounce to the big money their are making on us.

I am convinced that, eventually, there will be something like this out there. How long we have to wait?


5 thoughts on “How a journal should be

  1. Pingback: Come dovrebbe essere una rivista scientifica « BATS blog

  2. I agree that the current system needs fixing. Don’t quite know what the fix should be, though. I like some of your ideas very much; in particular, I think it would be great to have a summary of the reviews made available (although not anonymously — I’d like the editors to take ownership of the review comments, and to accept the responsibility of defending these comments if necessary).

    I don’t know whether a forum for continuous interaction would scale with the number of publications an author has, though — I’m not sure I have the cycles to monitor/ respond to comments about papers I wrote in the 1980’s…

    • The “comment” features seems to be unpopular, mostly because people are scared of receiving bad or insulting comments. You’re concerned with a different problem, and it is something to think about. However, if an author does not remember anything about his past paper, it probably means it was the paper was not so relevant, so why a reader should care? 🙂

      • On a related note, the ResearchGate network

        has a “microarticle” blog-like feature in which you’re supposed to provide a summary of an article results that you published, and every other researcher can provide technical comments. This is the kind of interaction that should be targeted across researchers.

        The current conference-based system simply doesn’t work: almost always you don’t have the paper available before the conference, papers are technically complex and so hardly you can come up with some insightful technical comment in the 20mins you hear to the author’s talk (where rarely technical details and equations can be provided), except things like “this is not clear…”, “can you please explain that more detail…”, etc. Ok, you can always study the proceedings between the welcome reception evening and the first talk the morning after — skipping some sleep.

        On a further related note, I came across this new HiPEAC/ACM publishing model that tries to address the wasteful repetitions that we usually have across journal and conference papers:

        Any comment is welcome.

  3. an additional aspect is the financial one: we, the authors, the funding tax payer, have to *pay* go give away our copyrights for free, so that an organisation, e.g., IEEE, ACM, can then make money from it – of which we dont get anything – and then we again and other researchers have to *pay* to read – this is just plain wrong


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s