The Naivete of Scientists

Although in some disciplines, it is relatively uncontentious, the rise of open access publishing has produced a lot of comment in others. In one of my two disciplines, computing science, this form of publication is still the minority, and still raises comment. For instance, Michel Beaudouin-Lafon has commented suggesting this scientists are highly naive about the costs of publishing. He argues that scientific publishing is intrinsically expensive, and that open access will have negative implication for science as a whole.

Over the years, commercial STM publishing has become a cutthroat business with cutthroat practices and we, the scientific and academic community, are the naive lambs, blinded by the ideals of science for the public good-or simply in need of more publications to advance our careers.

— Michel Beaudouin-Lafon

Personally, I think that “naive” is the wrong word; scientists are often not good at operating in a co-ordinated way. Although, we work together in small groups, and sometimes in large groups, in general, we are still very much a cottage industry; at any one time the number of scientists working in a distinct discipline is not that large, even on a world-wide basis. Of course, this works pretty well for scientific advance; we are not a production industry, but researcher. No one knows the best way forward, and we need to experiment to find out. But it does mean that we often play second fiddle to those capable of more co-ordinated action; compare for example, scientists to the medical community with its tightly controlled professional bodies. Or, of course, the STM publishing industry, particularly as it has become focused in fewer and fewer competing publishers.

For example, ACM spends several million dollars every year to support the reliable data center serving the Digital Library

— Michel Beaudouin-Lafon

Clearly, it is true that the cost of data centres and storage are not trivial. But the cost of servicing data has plummeted over recent years. Scientific papers largely consist of storing words and figures; these do not take up much space. The laptop I am working on has a copy of my email directory; it’s not complete but it carries most of my outgoing email since 1994 and a lot of the incoming; this is a lot of words! But the total size is now less than 5G, which will fit on a 3 pound pen drive, or my phone. Now if ACM were storing research data, then it would be a totally different issue; the costs here are significant, problematic and rising. But they do not.

The ACM might spend several million dollars a year, but the bottom line here is that this does not account for the cost of publishing. The Wikimedia foundation which supports Wikipedia spends around 10 million dollars a year, in total, on one of the top ten websites in the World. This is about the daily cost of the whole scientific publishing industry.

The quality of a journal is typically measured by its impact factor

— Michel Beaudouin-Lafon

And a very bad measurement of journal quality it is too. As someone who works in two disciplines at once, I constantly get hit by this: my best computing publications have laughable impact factors when compared to my bio publications; when judged against computer scientists, however, my bio publications have such high impact factors, that they have to be ignored as outliers.

At $5,000 per publication, my lab is broke.

— Michel Beaudouin-Lafon

It is not clear where the $5,000 figure comes from, as most open access is less than this. But, anyway, this argument makes no sense. Our labs are already paying a vast amount of money for publications; usually this is squirrelled away in overheads, taken from our budgets before we see the money. And, although it doesn’t happen so much in computing, many journals levy significant page charges.

They are the big pharmaceutical labs and the tech firms who publish very little but rely on the publication of scientific results for their businesses. With author-pay, research will pay so that industry can get their results for free. Is this moral?

— Michel Beaudouin-Lafon

Open access on its own is not enough. we also need public disclosure about the process. Perhaps the examples of the pharmaceutical funding journals directly are unusual. It is not so easy to tell at the moment. In this context, it could be argued that the last thing we need is the pharmaceutical industry paying for the results of science. Of course, conversely, the pharmaceutical industry could argue that they already do pay for the (publically funded) research by way of taxation.

While they are interesting, all of these arguments really miss the point: the pharmaceutical industry already get their results for free, as their subscription fees do NOT pay for the research just its publication. The publishing industry also get the results that they depend on for free or with page-charges by charging the authors. And for every paper that researchers publish for free, they pay more to read someone elses.

So, we are already in the situation that we are told is not moral.

It is important to understand that the scientific community is largely at fault

— Michel Beaudouin-Lafon

There is some truth in the idea that scientific community has let itself walk into the situation, but ultimately I feel, that this is like blaming the financial crises on those recieving subprime mortgages. It is true that it is scientists who submit their best work to expensive closed publishers; but, especially in early and mid “career”, we do this to safe-guard our futures.

The problem with the subscription model is not the model but the fees.

— Michel Beaudouin-Lafon

Quite the opposite. Ultimately, I don’t pay the fees, so how much do I really care? But the subscription model prevents re-purposing, it limits access, it prevents competition. I work at a university as a scientist because I value the ability to be able to swap and discuss my work. I want the general public to be able to access my research. Dissemination of knowledge should be part of my job; I think it is reasonable that I, or my employers, should pay for it.

Which is not to say that the level of fees are fine; they are not. They are far to expensive under any model.

The added value provided by publishers is twofold: reputation (the value of the imprimatur), and archiving (the guarantee that the work will be available forever).

— Michel Beaudouin-Lafon

And this is it? Is this all that we are getting, given the costs? Especially the the reputation comes from the work, not the journal, and the archiving should be a rapidly decreasing cost.

Actually, in practice, I think the current publishing industry brings more value; selection of reviewers, sometimes copy-editing and, critically, advertising of the content. But, again, times have changed, and publishing practice in these areas has not.

The only other area in publishing where authors pay to get published is called the vanity press. Do we really want to enter that model?

— Michel Beaudouin-Lafon

This is a low blow, nor is it true. Many people pay for their own publishing costs. The government pays to publish election results; health service pay to publish public health information; companies pay to publish product safety recalls. All circumstances where the value to the author of public awareness of their content far exceeds the income they would recieve from charging. And the biggest example of this is the advertising industry.

Nor is the implication that this will necessarily result in low quality true. Consider the blogosphere; of course, there is much junk, the standard of science journalism is very high; frankly, when ever respecting sources like the BBC start talking about pixie dust, it’s probably at least as high-standard the as mainstream media.

All this aside, what do I, as a scientist, actually care about? Some of these leap to mind:

Open access was built on the basis of replicating the existing publication. PLoS for example did this precisely so that it did not challenge both the business model and the publication procedure at the same time. How much of the costs stem from this? I think that we, as authors and readers, should know. How much of the millions the ACM spends on it’s data centre is involved in managing access controls, for example? How much on advertising? How much at booths at meetings?

Open access has opened the door, but now we need to challenge and change the process. Hosting data is not free nor is archiving. And, yet, I can find own my website from 2002 and enjoy it’s gaudy colour scheme all again. If this blog post is so exciting to the world, that the load brings the server down, you will be able to read it on coral cache. The peer review is expensive and time-consuming; I know because I’ve organised enough of it for BioOntologies. But then I did not get paid for this and how many of the real costs of peer-review do publishers bear? And discovery and selection? Well, we have google, and I follow my peers on twitter.

Author fees are not a solution. […] Finally, nonprofit publishers should take advantage of their unique position to experiment with sustainable evolutions of their publishing models.

— Michel Beaudouin-Lafon

And on this, I could not agree more. Our experiment with Knowledgeblog suggests that we can get 90% (or 80% or 70% depending on who you ask) with commodity software. It’s only a small start, but then I was on the mailing list that saw the first email about the creation of wikipedia, and that wasn’t long ago.