Journals and scientific knowledge...easy access?

The article below is about seven years old and most still holds true today. Most scholarly journals are not interested in the dissemination of knowledge but the art of cupidity. And the issues covers other areas of knowledge.

"Science for Sale"

by

Danny Kingsley

May 29th, 2003

Australian Broadcasting Corporation's Gateway to Science

There is a quiet revolution going on in laboratories and behind closed editorial doors. A few brave souls have recognised the potential of the Internet and are trying to fundamentally change the way scientists communicate their results. In doing so they are taking on the biggest players in the scientific world - the publishers of scientific journals. The Lab's Danny Kingsley investigates.

In the beginning

To understand how publishing houses became the real powerbrokers in the world of science, we have to go right back to the beginnings of science as a discipline.

Science used to be a solitary pursuit. If you were rich enough (or lucky enough to have a patron) you pursued your area of science alone, building a telescope in your attic or collecting things and trying to classify them. You may have been in contact with a few people who had similar interests, but basically the only way to let anyone know about your work (and there was no imperative to do this) was to write a book.

That all changed in 1665, when Henry Oldenburg created the Philosophical Transactions of the Royal Society of London, which is still published today. This was the first public record of original contributions to knowledge, and acted almost like a patent office - once your discovery was registered in the Philosophical Transactions it belonged to you.

This was the beginning of the modern journal system, where scientific results don't 'exist' until they are made public - it's no good making the biggest scientific discovery ever if no-one else ever finds out about it.

Fast forward to now - the role of journals in science (in theory)

Despite there now being an abundance of journals, in almost all languages, the system has pretty much remained the same. Scientists undertake their research then write it up in a 'scientific paper', which they submit to be published in a journal.

Scientific papers are the main result of scientist's work. Once a paper is accepted by a journal's editor it is subjected to 'peer review', where it is looked at by other people in their field to determine the research is solid and makes sense. After any necessary modification the work is published.

Having results in the public domain, like a journal that anyone can read, opens the results up to be challenged or added to by other scientists who can try to replicate the work. If they make an improvement on it or find a mistake, they then write their own paper and have it published with the new information. It all adds together to make one large body of public knowledge, each step brings us closer to the 'truth'.

Scientists who do particularly outstanding work will often find their papers referred to (also known as a citation) in many other papers. Generally, these are the scientists who are promoted and funded.

Sounds fair doesn't it? Sadly that utopian version of science no longer exists (if it ever did).

The role of journals in science (in reality)

Publishing their work has become the main focus of most working scientists - it has become a numbers game.

The problems can be clearly traced back to 1958 and a man called Eugene Garfield, who started up a company called the Institute of Scientific Information (ISI). To be fair, Garfield could never have anticipated what would eventuate from his system.

The citation system of referring to previous work can be seen as a huge web of information. It shows which papers are linked to each other, which papers are seminal papers, which ones are destined for obscurity.

Stretch your imagination back to well before the Internet, before faxes, even before computers were popular. In 1958, the only way to find out if a particular paper was seminal was to look at the bibliography of lots of different papers (which meant going to the library and searching through volumes and volumes of bound journals) until you created a picture - the seminal papers being those that were continually be cited.

Garfield saw that by linking up this information, you could save everyone the bother. So by the early 1960's the Science Citation Index (SCI) was in place. This was (and is) a computerised listing of the papers that had cited a particular paper. To date the SCI contains a few thousand core journals, a small fraction of the copious number of journals out there.

After a while the ISI began to publish something called the 'impact factor' (sometimes referred to as the IF) of the journals it used in the SCI, in effect rating these journals against one another. This system ignored that the journals came from different disciplines and often were completely unrelated to one another.

With this move, SCI was ready to drift into a whole new business area. That of career management tool. Suddenly scientists were being rated - not on whether their work was of high quality (remember it is the journal, not the article that is being ranked) but on where they can get their work published.

Suddenly all the power is sitting in the hands of the journal editors. Scientists need to get their work published so they can get citations and get research funding and tenure. Today Australian (and other) universities and the government use the impact factor as a measure for justifying research funding and academic appointments.

Feel the power - journals as the steering wheels of science

The editors of scientific journals are in a very powerful position. Those that head up very high impact journals - like Science, Nature or Cell, are bombarded with thousands of articles for each space they have available to publish. They must also vie with each other to publish the sexiest research first, and negotiate with scientists who play journals off against each other for who gets to publish their research.

So editors become quasi-reviewers as they sift through and decide which of the papers will be sent off for peer review. Meanwhile the hapless scientist is waiting - they can't send their paper to another journal until it has been officially rejected by the first one.

Even once a paper has been accepted it can take up to 18 months to be published. The journal has moved from being a method of letting other scientists find out the latest research, to being an archival record of past work to be assessed by bureaucrats.

This time lag also challenges another tenet of the traditional 'system' of science - that of replication of work to iron out any inaccuracies. Let's face it, with the pressure of constant publication to keep the numbers up, who has time to replicate someone else's work? For that matter, who has time to check back original papers cited in other work?

There is other evidence that the system is crumbling. Over the last few years several high profile discoveries that were published in high profile journals were later retracted, causing many to ask about the robustness of the peer-review system. Recent studies have also shown deep flaws in the citation system, but that's another story.

Money makes the (scientific publishing) world go round

As with most things in life, it all comes down to money, and how to make a buck - and there are plenty of them to be made out of scientific publishing. The prices for scientific journals have risen around 200 per cent over the past 10 years. For example the annual cost to an institution of subscribing to the British Medical Journal is A$625 and Nature is about A$1500.

In the days of the Philosophical Transactions, the printing press had only been around for a couple of hundred years (coming from Germany to England in about 1474). Individual printing houses printed the articles and they wanted to establish legal claims over the texts they were printing. They developed a system where they 'owned' the text - creating the concept of intellectual property (which when you think about it is a very odd term).

This still happens. Scientists hand over their intellectual property when they submit a paper to be published. That's right. It's not the author who owns the intellectual property (including copyright) on the work, it's the journal. To add insult to injury, the scientists' institution's not only have to pay for a subscription to a journal to see their work in print, but sometimes scientists are also asked to pay a fee when they submit a paper to a journal.

This situation has been described as the 'Faustian bargain' by some scientists - a deal with the devil. They are tied to the publishers because without them they cannot get promotion or sometimes even funding. But they are giving away their work, their copyright.

Shhh! The library's role in the story

Between the scientists and the publishers lies a group of professionals that are often forgotten in the world of academic publishing, but who are fundamental to this story - librarians.

For many, many years now there has been a crisis going on in the hushed aisles of library stacks the world over. It is referred to as the "serial price crisis." Libraries simply can't afford the ever-increasing costs of journal subscriptions.

To give you an example, in 1991 the UNSW library had a yearly subscription budget of $4.7 million for science technology and medicine journals alone, just over half its TOTAL book allocation budget of $9 million.

The underlying problem is that journals are monopolies, argues Mr Jan Velterop, publisher of BioMed Central. This manifests itself in 'big deals' where libraries are in contracts that include a package of journals and online access to journals from a particular publisher for several years, with restrictions on early cancellation.

So what can anyone do? All the power is in the hands of the journal publishers and the scientists and librarians will just have to lump it. That was the situation until the 1990's, when the biggest thing to hit the scientific world first started to make an impact.

What is this newfangled thing called the Internet?

The Internet began as a Cold War concept for controlling the tattered remains of a post-nuclear society - where information was kept in a series of hosts (servers) so if one was destroyed BY a nuclear bomb, the information would not be lost. In the very early 1970's a small group of universities and government research institutions became linked by a network, called ARPAnet. In 1982 TCP/IP, a common Internet language was invented.

In 1990 ARPAnet bowed out, leaving the 'Internet' in its current form, and in 1991 the code for the World Wide Web was written.

Suddenly, scientists had a way to communicate with each other quickly, easily and cheaply, and a few of them started to realise that here was a way out of the mess they were in.

The year 1991 was notable in the push towards electronic journals. There was a considerable amount of material published that year arguing for electronic journals, and a few people started them up. In general these early attempts were unsuccessful - at this stage most academics did not have a computer at their desk, and navigation around the Internet was difficult because the World Wide Web really didn't take off as the main browsing code until about 1995.

One electronic 'journal' that did begin in 1991 and is still going today was started by Paul Ginsparg. He began his physics preprint server at the Los Alamos National Laboratory. He then invited his colleagues in high-energy physics to submit their papers to it. With the new server, physicists could check many new developments in their field with just one stop.

The success of the pre-print service showed that print publication was related to career management, while the digitised version placed on the public server dealt with the management of intellectual quests.

Over the past 12 years there have been several attempts to set up free access electronic libraries (see viva la revolution), but none of these really challenged the printed journals at their own game. But in December 2002 the announcement of the first completely free, peer-reviewed journal was made by a group called the Public Library of Science (PLoS).

Towards a free future

The PLoS is a non-profit organisation of scientists committed to making the world's scientific and medical literature a public resource. With a AUD$9 million grant from the Gordon and Betty Moore Foundation, the PLoS is creating two peer-reviewed online journals in biology and medicine with the goal of cornering the best scientific papers and immediately depositing them in the public domain.

According to the PLoS, the journals will be governed and operated by scientists, retaining all of the important features of scientific journals, including rigorous peer review and high editorial and production standards. All published works will be immediately available online, with no charges for access and no restrictions on subsequent redistribution or use.

This model is nearly identical to that of BioMed Central. The difference is that BioMed Central is a commercial enterprise, despite having the philosophy of authors retaining copyright.

So who's going to pay for it? Apart from the set-up grant, the journals will initially ask most authors to pay about US$1500 per article. This would probably be covered by their institution - and be incorporated into the grant funds at the beginning of the tender process. The advantage to the institution is they do not have to pay the literally hundreds of thousands of dollars to subscribe to a journal to see their work in print.

Conclusion

So will it work? Will a bunch of scientists committed to the public access of science be able to take on the publishers and win?

The new system relies on scientists being prepared to forego publishing in prestigious journals to support the principle of science as a public resource. And there lies the rub. Many scientists will argue that there is nothing wrong with the system as it currently stands. Others will not be prepared to 'risk' publishing in a start-up journal.

The publishers found a way around the initial threat of electronic journals and there is now even more money at stake. So this is the time for scientists as a collective to decide what they want. It's in their hands.