Thursday, October 29, 2009

Buying PDFs: Truth and Consequences

On-line journals are great. You get immediate access to the latest research and can download a fully searchable PDF for later use. Journal digitization has revolutionized the pace of science communication and increased the reach of formerly obscure journals. Through the wonders of the Internet, anyone can get access! Right?

Not so fast. On-line journals offer full benefits only to those whose institution has a subscription. If you don't have a subscription, you're out of luck. . .mostly. Thankfully, it is possible to buy PDFs of individual articles. Right?

Not so fast. It is indeed possible to buy PDFs of articles from most journals. . .if you can cough up the money to do so. Let's face it. PDFs are expensive. Ridiculously so. A three page note from Journal of Vertebrate Paleontology might cost nearly $30! A six page paper from Geology runs $25. [end of this post] Is it really worth it?

Big Problems
I won't pretend that journal sales don't support professional societies I care about. I won't pretend that publishers shouldn't get compensated for their services. I won't pretend that everything is going to turn open access tomorrow.

But, I'm not shy about saying that the current system stinks. No matter how you slice it, $30 for a PDF article is unaffordable for a typical consumer of paleontological publications. The new membership rates for the Society of Vertebrate Paleontology, which include electronic access to Journal of Vertebrate Paleontology, cost $115 for e-journal only. Assuming around 120 articles per year (a rough count for JVP in 2008), this works out to a little less than $1 per article (and remember that this assumes that 100 percent of membership dues go to the journal subscription - certainly not the case). Why is there a 3,000 percent markup for non-members? Wouldn't journals be able to get by with, perhaps, a 500 percent markup over the member rate? [IMPORTANT NOTE: I am not intending to single out JVP and SVP exclusively; nearly all journal publishers are guilty of this problem. . .it just so happens that JVP is the example most relevant to most readers here]

Let's put it another way. At the current costs of PDF articles, buying every single article in the journal would cost an individual around $3,600 per year.

And let's look at it from yet another angle. Institutional paper subscriptions to JVP are $270/year for US institutions (pre-Taylor & Francis switch; we don't yet know what the cost will be post-switch). This works out to an average of$2.25/article. . .for a format that is much more expensive to deliver than an electronic document!

And one last angle. . .the Mesozoic Birds: Above the Heads of Dinosaurs edited volume, which is a beautifully produced and scientifically important work, includes 20 articles, ships in hard cover and retails for $100 new. This works out to $5/article within the book. Contrast this with $30 or more for a single digital file.

Commenting in a recent thread here at the OSP, a librarian noted that libraries are paying between $2-$4 per PDF article when bought in bundles from for-profit publishers like Elsevier (which, incidentally, continues to post profits in the midst of the recession). Although it is certainly fair to have volume discounts (although the pricing schemes and the bundled journals are often rather dubious in practice), it is of dubious benefit to science to charge such a disproportionate rate to private individuals who are just trying to do some science.

Do these pricing schemes serve science? Do these pricing schemes serve the interests of the authors, who just want their work to be read?

Commonly Suggested "Solutions"
When the issue of paying for individual PDFs is brought up, there are often a number of "solutions" proposed. I put the word in quotes because, as explained below, none of these is fool-proof.

Why not write the author for a PDF?
In some cases, this is a good workaround. But it's never a perfect workaround. Sometimes authors are unresponsive, have changed email addresses, retired, or passed into the fossil record. In this case, the researcher in need of the paper is out of luck.

Why not post a PDF request to VRTPALEO or the DML?
In some cases, this is a workaround. . .one need only look at mailing list archives to see that this is a common strategy. But, PDF requests unfortunately carry a small annoyance factor for many list subscribers. Sometimes no one responds. Finally, PDF sharing by anyone other than the author is generally illegal in the eyes of the publishers and societies (but let's not pretend it doesn't happen).

Why not join the society, and then you get all of the articles as part of your membership?
Again, this is a solution in some cases. If you are a paleontological enthusiast or professional who loves (or needs) to follow every bit of the literature, you should probably join the Society of Vertebrate Paleontology and/or the Palaeontological Association and/or the Paleontological Society.

But, let's be realistic. Nobody can afford to join every society to get access to all society publications. In some cases, it's just not necessary. Consider this hypothetical situation. You're a population biologist working for a small nature preserve who wants a little background on the evolutionary history of the coyote. You do a search on Google, and find a citation to Journal of American Paleontology with an abstract that details ecological shifts by coyotes during the Ice Age using isotopic analysis. You click on the link, and. . .paywall! The Society of American Paleontology wants $30 for a five page article. Is it really worth your time and money (as a population biologist) to join a society of paleontology, when you are already stretching your budget to cover dues for two more relevant societies? If the charter for the American Paleontological Society says that they are to promote and advance the science of paleontology, is the society really living up to this mandate? If the society wants to foster cross-discipline appreciation for the relevance of the field, is this happening if the research is not easily accessible? Are the authors who contribute to the journal being well-served by having their research so restricted?

Finally, some journals just aren't sponsored by societies. There is no solution in this case, other than to pay a few hundred dollars per year.

Most of the other journals charge $30 for a PDF!
This is not a solution (or even a "solution"), but an excuse. During the business meeting for a scholarly society to which I belong, I raised a concern about the prices that the society's journal was charging for individual PDFs. Can you guess the answer that I received? I remember something my mother often said to me. . ."If everyone else jumped off a cliff, would you too?" Exorbitant PDF costs may be the norm, but that doesn't mean it's the ethical thing to do (especially for a non-profit society whose stated mission includes phrases like "advance the science" and "serve the common interests").

What Needs to Change?
Science (yes, that's the reason why most research is published) is not being served by the current pricing schemes. Alleged solutions for those beyond the boundaries of the pay-wall are not comprehensive, and again do not serve the interests of science.

PDFs of individual articles need to cost less. There is no way to legitimize charging $25, $30, or more for something that costs less to produce than a printed journal, particularly when it prices out to up to a 3,000 percent markup. The current pricing scheme restricts the readership of articles and creates a hierarchy of have- and have-not researchers, in a time when the Internet is supposed to fix these problems for academia. Also, let's not forget that the research behind these articles is often heavily subsidized by public tax dollars. . .

I would suggest that $5, or maybe $10 in exceptional circumstances (e.g., a 50 page monograph) is a cost that I would be willing to pay for a PDF. I would also submit that many journals would see increased PDF sales (particularly for popular topics, such as dinosaurs) if the price was set at something mere mortals could afford to pay.

Is There a Solution?
Right now, it sounds like I'm doing a lot of complaining and not a lot of problem solving. Well. . .yes. This post is partly a rant. Unless people are aware of the problem, nothing will ever happen.

So what can we do? Here are a few suggestions:
  • Ask journals and professional societies to consider the implications of pricing schemes for PDFs. Speak out to the people who matter. Let them know how you feel, and how it affects you.
  • Don't pay the ridiculous charges. Find low-cost, legal alternatives (e.g., writing to the author or interlibrary loans) whenever possible.
  • Submit your work only to journals with researcher-friendly publishing policies.
  • If you are an author, do everything you (legally) can to get your work out there for free. If the journal allows you to post a PDF, do so. Respond promptly to PDF requests from other individuals.
What are your thoughts?

Addendum 1: As if by magic, this post at The Scholarly Kitchen appeared at nearly the same instant as I hit the "publish" button on this post. I haven't followed up on the service (which essentially offers cheap rental access to articles from various scientific publishers), but will certainly be looking into it.

Addendum 2: Matt over at Protichnoctem has a nice post with more on the issue of buying PDFs. Go check it out!

The Journal List
The Anatomical Record (Wiley): $29.95
Bulletin of the Geological Society of America (GSA): $25 (GSA website)
Bulletin of the Geological Society of America (GSA): $32 (Geoscience World website)
Cretaceous Research (Elsevier): $31.50
Geology (GSA): $25 (GSA website)
Geology (GSA): $32 (Geoscience World website)
Ichnos (Taylor & Francis): $37
Journal of Experimental Biology (The Company of Biologists): $10
Journal of Morphology (Wiley): $29.95
Journal of Paleontology (Paleontological Society): $12 (BioOne website)
Journal of Paleontology (Paleontological Society): $15 (Geoscience World website)
Journal of Vertebrate Paleontology (SVP): $30 (BioOne website)
Nature (Nature Publishing Group): $32
Palaeogeography, Palaeoclimatology, Palaeoecology (Elsevier): $31.50
Paleobiology (Paleontological Society): $12 (BioOne website)
Paleobiology (Paleontological Society): $15 (Geoscience World website)
Palaeontology (Wiley): $29.95
Proceedings of the National Academy of Sciences (NASUSA): $10 ($25/full journal access for seven days)
Science (AAAS): $15


Jenny Reiswig said...

If you are at an institution with a library, see what they can do for you. You remember the library, it's that place that spends the money on those institutional subscriptions that you DO get at your desktop. :) Well we have this freaky service called Interlibrary Loan. Many big academic libraries now ship PDFs instead of paper, depending on copyright policies of publishers etc. It won't be as fast as buying it immediately, but probably will be faster than requesting a copy from the author. Unless you need an exotic article THIS MINUTE, give your library a chance.

Andy said...

I agree that ILL is useful for those at institutions with libraries, but many of us paleontologists are at small institutions without libraries (beyond those sitting on our personal bookshelves), and hence no institutional subscriptions of any sort. This also does not help the average paleo enthusiast who just wants to read the latest literature. Our local public libraries often don't have the access or the resources to fill such requests. Also, interlibrary loans do not fix the problem that many journals are simply overcharging for their "product." As an author, I firmly believe that it hurts me personally (and science as a whole) if my work isn't easily and affordably accessible.

Anonymous said...

I often wonder just how many copies journals shift at these prices. Does JVP (for example) really sell say 1000 PDFs each year at 30 bucks a pop, or is it more like 10? In which case would selling them for 5 not be more likely to make a profit, since they would (I would guess) sell more than 6 times as many at five dollars each...

Andy said...

I wonder the same thing myself. . .I would suspect that the main reason prices remain high is because some marketing research somewhere has shown that users in industry/medicine/whatever are willing to pay ridiculous prices for the information in their relevant journals, so even those of us in small, poor fields get hosed simultaneously.

Part of the problem is likely that, for some articles, PDF sales volume would be low no matter what you sell them at. For instance, a PDF on T. rex-related research would sell like gangbusters, whereas even a completely revolutionary paper revising the taxonomy of atrypid brachiopods would sell perhaps 1 or 2 copies (no offense to brachiopod workers). Most articles probably fall into the latter category.

dinogami said...

I would suspect that the main reason prices remain high is because some marketing research somewhere has shown that users in industry/medicine/whatever are willing to pay ridiculous prices for the information in their relevant journals

Funny, I never got that impression at sense (probably due entirely to cynicism, rather than data) has always been that these prices were due more to a "people will like what we tell them they like and will do what we tell them to do" attitude on the part of the publishers...

Vivekananda said...

Well. I read through your article,yes it is impossible to buy a single article for 30 dollars.Even all scientists in India faced the same problem when India's economic condition was bad. But, people are able to solve the problem.

We did not want to invest too much money in this field but wanted most of journals to be available, so a body was set up to negotiate with publishers and provide access to all universities in India.

Whichever journals are not subscribed, people can request the subscribing libraries to send the article and by this way save lot of money as well as get the resources.

So, it basically comes upto the point that"If you have the will, then you will find a way".

I dont know from which country the Blogger is ? but you should be also having a similar network of all universities in the country.

If not, you can pioneer such a movement and help all the scholars to get access to e-resources.

I was a student of biochemistry, but i found this field interesting and i am working with one such company providing technical support.

Any queries, feel free to email me at :

Looking forward for reply and expecting a debate on intellectual basis

Dinorider d'Andoandor said...

If they lowered their prices more people worldwide would buy their articles and consequently those soceties (or whatever they are) would win more dollars.

Gavin Baker said...

Re: Andy, most public libraries will handle ILL requests on behalf of their users. It may take longer than an academic library and carry a small fee, but you will probably get it, probably for less than $30 cost to you.

Not that this is a solution, but it's an improvement.

Of course authors could solve the situation by posting a free copy of their article, which publishers almost always allow.

Tom said...

[Doing this in two parts since posts are size-limited]

While it is lovely to be able to get a PDF of an article directly from the journal itself, for commercial publishers this requires either a library subscription or the author has to pay a fee up-front to release it as an open access article. As was reported in another post in this blog, journals like JVP require payment of $3,250.

In addition, it should be emphasized, libraries are not helped in this so-called "hybrid model" where only select articles are available as open access from an otherwise commercial publisher, since the large cost of subscribing does not disappear unless all authors pony up to release all the articles as open access. Examining the pattern in commercial journals that use this model, only a fraction of authors are doing this. The effect for most readers is that access to articles requires either a subscription or direct payment by the reader himself to buy a copy.

Rather than this hybrid model of open access, libraries are encouraging authors to retain the right to self-archive their papers and then work with their libraries to deposit a copy at their institution. Alternatively, or in addition to this, authors may want to deposit a copy in a subject repository. (Is there such a thing for paleontology?)

Tom said...

Part 2

The big issue then becomes findability. How do you locate whether an open access version of article exists without some sort of insider knowledge? E.G., I know this author is at such and such university and that institution provides a repository for authors' works. Or the article is likely to be in a subject repository.

One solution I've found is OCLC WorldCat (the folks who provide the metadata behind many library catalogs and interlibrary loan systems).

They recently acquired a service called OAIster, which indexes open access papers deposited in subject repositories and institutional repositories. The "OAI" part of the name stands for "Open Archives Initiative" which is a discovery layer that most of these repositories have built-in.

In addition, WorldCat's database provides an article search layer. This is not a comprehensive article database, by any means. But articles from many important publishers are there, so it makes a nice starting point.

The upshot is that by using WorldCat, you can search for a journal article title and, if an open access version of it exists in a repository, then chances are you will easily be able to find and obtain a copy. For free. If the published version of the article is also indexed, you can find that as well.

For example, if I was looking for the article "Specimens of the billfish Xiphiorhynchus van Beneden, 1871, from the Yazoo Clay Formation (late Eocene), Louisiana" by Harry Fierstine and Gary Stringer,

WorldCat directs me to the published version in Journal of Vertebrate Paleontology

However, access to this copy of the article may require a subscription -- either through a personal membership or through a library subscription.

But in addition, WorldCat also finds two open access versions of the paper in repositories:

One copy was deposited in the Kennedy Library institutional repository at CalPoly

and another copy that the one of the authors self-archived using a free service called BePress Selected Works

It was the decision of the authors to negotiate with the publisher to retain rights to self-archive a copy of their paper that made access to this article free. Not just haphazardly on a personal website, but in a place where it is most likely to be found using services like WorldCat (or, in a less structured way, through Google Scholar).