`We have to be able to recover our costs'

Interview with Prof. Martin Blume, Editor-in-Chief, American Physical Society.

In 1996, Professor Martin Blume, a condensed matter physicist at Brookhaven National Laboratory (BNL) in the United States took over as the Editor-in-Chief of the American Physical Society (APS). It was growth time for electronic publishing and the Internet, which were establishing themselves as the preferred modes of scholarly communication. The APS, which publishes the prestigious physics journals Physical Review and Physical Review Letters, had to adapt and evolve to the digital age.

In the past seven years, Blume has brought about impressive changes in APS' publishing enterprise by successfully leading its efforts in electronic publishing. As of July 1, 1997, all of Physical Review and Physical Review Letters became available electronically, and many new features, such as searching and linking, became possible online. He was instrumental in building up the APS online archive consisting of all published journals of the Society going back to 1893 when Physical Review began publication in Cornell University. The task was completed about three years ago. In the context of the ongoing debate on open electronic access to scientific information, in particular professional journals, Blume's experience with the APS has enabled him to have a definite perspective on the issue.

Blume earned an A.B. from Princeton University in 1954 and a Ph.D. in physics from Harvard Unversity in 1959. He was a Fulbright Fellow at Tokyo University from 1959 to 1960, and a research associate at the Atomic Energy Research Establishment (AERE), Harwell, England, from 1960 to 1962. He joined Brookhaven as an associate physicist in 1962.In 1984, Blume was appointed Deputy Director of Brookhaven Lab, a position he held until 1996, when he became Editor-in-Chief of the American Physical Society. He has remained a senior physicist (on leave) at Brookhaven.

Blume's research interests include theoretical solid state physics, magnetism, phase transitions, slow neutron scattering and synchrotron radiation. Blume predicted magnetic resonance scattering in 1983, and it was first observed experimentally by Namikawa at the Photon Factory in Tsukuba, Japan. Today every major synchrotron in the world, including the National Synchrotron Light Source (NSLS) at Brookhaven, uses resonant magnetic X-ray scattering. In recognition of this, Blume was awarded the 2003 Advanced Photon Source Arthur H. Compton Award, instituted by the U.S. Department of Energy's Argonne National Laboratory.

Blume was in India in November 2003 as a visitor to the Tata Institute of Fundamental Research (TIFR), Mumbai, and the Indian Academy of Sciences (IAS), Bangalore. R. Ramachandran had a wide-ranging telephonic interview with him on the issue of open e-access. Excerpts:

What is your reaction to the recently launched open- access journal PLoS Biology of the Public Library of Science and the basic philosophy of PLoS? Does the APS subscribe to the idea of free access to its archives and current issues of journals?

We would very much like to be able to provide open access to our journals. Again, we are not for making loss. We have to be able to recover our costs. The Public Library of Science - let me characterise it this way - is open-access to readers but is toll-access to authors. We are the other way around. We are open-access to authors but toll-access to readers. Only one of our journals has page charges or article charges and that is Physical Review Letters and there it is essentially voluntary.

In discussions in India I was told that people have never paid page charges because they cannot really afford it. But if we went to open access for readers, we have to recover the costs in some other way. We have to charge somebody for it and there would still be a problem.

But the population of readers is much larger than the population of authors. In that sense, when we say open access it addresses a much larger population of the scientific community than what the policy of charging the readers does.

This may be true in the life sciences. In physics, it is not a broad general public that is looking to read our journals. People who need access, who are going to read these technical papers, have access because they subscribe. This is what it really amounts to. I don't see that we are going to gain a large number of readers. We would still like to be able to do it. The question is how to find a way to do this and still recover our costs.

Let me mention one experience that we had 10 years ago. Then page charges were a significant part of the money that we brought in than they are now. At that time there was an experiment where Physical Review D (High Energy Physics) had no page charges. And at that time there was still this anger in the physics community of the U.S. over the super-collider, and high energy physics and condensed matter physics were at odds with each other. I am a condensed matter physicist and was a member of the super-collider Board of Overseers. I was something of a skunk, you see. But the fact is, there was still this hostility. The condensed matter physicists said, `here are these guys who are trying to get all of the money in the scientific budget and you are not charging them the page charges'. At that time we could not afford to do this and so we reinstated page charges. There was a revolution. People did not want to pay. The life sciences has other sources of income besides the page charges - advertising, for example. In physics we have no such sources of revenue. Not many people are interested in advertising.

When you have an online edition, are there possibilities of raising revenue through advertising on the Web?

The amount of money we can make this way is trivial compared to subscriptions and our costs. So we have to look at this [open access issue] carefully. We are making a number of attempts at this. A few years ago there was a Nature debate on this. I wrote a section in this where I said that we would like to be able to provide access and we could do it right now provided everyone who subscribed now would agree to continue this contribution as sponsorship instead of a subscription. Then we would be able to open it to everybody. But that would mean that we would have to count on the people who now subscribe not to say, `Oh boy! It's free. We don't have to contribute any more.' We could not afford to do that. We are breaking even with the money that comes in now. If they don't have to pay they are not going to.

I don't know if you are familiar with public television in the U.S. Public television is funded by contribution, and a couple of times a year they have what is called `pledge week' where they get pledges from people that they will contribute. It is rather annoying because these replace commercials for the entire week. We would have to do the same thing. Every time you try to download an article during `pledge week', you would have to watch for five minutes or so somebody pledging before you got the article. This was facetious but it highlighted the problem that we face.

I myself have been in this position with page charges in the past. What happened with Physical Review D was that people started boycotting our journal and started publishing in Nuclear Physics, which did not have page charges but which cost about 10 times as much on a per page basis to the institution. If page charges and article charges have to be paid out of the authors' grants, as happens in the U.S., then the authors are faced with a dilemma. Either they pay the page charges or they send a post-doc or a graduate student to a meeting. The cost would be about the same. And they generally are faced with choices like this. It is not going to be easy to convert to that mode of operation [which is the basis of open access journals like PLoS Biology]. By the way, APS does have one journal which is open access. This is Physical Review Special Topics: Accelerators and Beams. We are doing this with money from the large accelerator laboratories.

How is it doing? The costs of an online edition are probably much lower...

It is the peer review and the composition that cost the most in these things. And we are looking to reduce those costs. But the reduction is likely to be minimal. One of the things you have to recognise is that our costs scale with the number of articles that are submitted to us. That number of articles has tripled in the last 20 years. We receive about 25,000 articles in a year. This is more than a hundred every working day. And this year's submissions seem to be about 7 per cent above last year's. We find that if we make our internal processes more electronic, we can save money.

The Public Library of Science (PLoS) is charging $1,500 an article, which is not far off from what we would be able to do, provided we got this from every paper. Now the fact is, only one-third of the papers come from the U.S.; a third from Western Europe and a third from the rest of the world. The papers from the rest of the world are those where people will not be able to afford the charges. So, for every one that we lose we will have to collect it from somebody else. This is part of the problem and the question is whether this is going to be sustainable.

How much does the cost come down by when you have only the online edition?

Has the Ginsparg e-print archive impacted the APS journals in any way?

We make use of it. We mirror it as you know. We were the first mirror in the U.S. We use it for submissions, for example. We link to it so that if people refer to an article [in the e-print archive], we link to the article. Authors can post their articles in the e-print archive in advance of submission. In fact, we make use of that. We send the referee to the e-print archive [for downloading and refereeing]. It saves us money. In addition, we allow the final (published version) to be posted on it so that whatever is published is there. It is a form of open access. Most of the articles of high energy physics get into the e-print archive. Still this does not stop people from submitting because they want the peer review and the peer-reviewed version to be read. So, to answer your question, it has not impacted and it provides us with the benefit.

But has open access to the e-print archive impacted your subscription base?

Not that we can see. In point, this is a consequence of our collection of articles and a consequence of the fact that the e-print archive has things going back 10 years perhaps, growing during that time, but before that it is only the APS archive containing everything that we published since 1893.

Unlike physics journals, why do others, like Nature and Science not accept papers once they are posted on any e-print archive?

I don't see it does them any good. I believe Nature no longer has such a policy. Science still has. We just had a meeting at the National Academy of Sciences on electronic journals. I asked the Science editors whether they still had that policy. They said that they still did. And Paul Ginsparg, who was sitting with me, remarked that they still had that policy but they had the tendency to ignore it when it was a paper they really wanted to publish. The chemists persist in this and I can't explain it. I think it doesn't make any sense. It doesn't do any harm and, I think, having it posted there does good. It just seems to bring more people back and forth to our side as well.

What is your reaction to the recent pronouncements at various fora - the Budapest Open Access Initiative, the Berlin Declaration, the Bethesda Statement, the Principles and Plan of Action at the World Summit on Information Society (WSIS), the ICSU documents, the CERN-RSIS Conference submission to the WSIS and so on - all of which urge a movement towards open-access publishing. Do you see any impact of these on the APS' journal access policies, in the medium term or the long term?

JHEPJournal of High Energy PhysicsJHEP

How do you see the future of electronic distribution of scientific information, in particular progress towards open access?

There is one way in which we can bring our costs way down. I don't know if this is going to happen, but we have to be prepared to think about it. We can become what we call a `virtual journal' on top of the e-print archive. That is, authors post their articles there and, when we do peer review, essentially all we need to have is a `Table of Contents' which goes back to the e-print archive. You just link to it. But there is a problem with that, very likely we would still have to download the articles ourselves because we have to have, I believe, in our possession, a copy of the article which we have peer-reviewed so that it can't be changed in arbitrary ways. We have to have that which we have peer-reviewed, so that there is a public record and we maintain the archive ourselves. So that puts you back with more costs and the question is, do you do such things as correcting English, which would also bring the costs up? Also - here I admit to a romantic view of these things - I still remember when I first had the article published in Physical Review. I had the manuscript written by hand and equations put in by hand and then seeing it in this format. I still would like to see the look and feel of an article. Well, this is a romantic view. If you can't sustain it in the electronic era, you can't sustain it. But for the moment, people seem to appreciate it.