Pay per paper (P3)
I’ve just stopped subscribing to Science. The magazine is great, but few papers in it interest me. The signal-to-noise ratio of Science, I guess, is just too low to most individuals. Instead, I’ve now subscribed to the RSS feed of Science. If any paper looks interesting, I can access to the full paper online through Harvard Libraries. Outside my office, a color printer is free to use for everyone. A library of an institution seems to be an ideal home for a journal like Science. Nearly every individual paper in Science is of high enough quality to appeal to someone in the institution.
Few journals can make that claim, however. Most journals are only relevant to several people in an institution. Furthermore, few researchers read any scholarly journal from cover to cover. Rather, we all read individual papers. However, libraries subscribe to journals, or even bundles of journals. As a result, the libraries pay for many papers that nobody reads, and miss other papers that someone would like to read.
This business model is bad for authors and readers, and possibly even bad for publishers. Technology now exists to distribute information far more efficiently, in a unit consistent with how people consume the information. For example, many people now prefer buying individual songs to albums. See a recent book, The Long Tail, by Chris Anderson, the editor-in-chief of Wired, for a remarkably perceptive analysis of media industries.
The same business model may apply to scholarly papers. One may argue that journals, like albums, were invented as a packaging technology to suit the old economics of delivery. As scholarly papers are all online, the name of a journal becomes simply a tag to the papers published in that journal. Maybe a powerful tag, but a tag nonetheless. So far as how papers should be distributed, the name of a journal should serve the same function as all other tag-like entities: keywords, names of authors, etc: the tags help readers to sort papers and set priorities. It makes no sense for anyone to insist that papers with any particular tag be delivered as a bundle.
Many publishers already offer individual papers for sale online; for example, the cost is at $30 per paper for many Elsevier journals. Once a reader buys a paper, it seems reasonable to share this paper with his close colleagues, and it also seems reasonable to store the paper for future use. Perhaps we can formalize this practice.
How about we treat a paper just like a book? With one click, a reader will have the paper, and his library will automatically pay for it. Once bought, the paper is accessible to every user of the library. We can also collect statistics. If the users of a library buy many papers in a journal, the library should subscribe to the journal. Libraries will set up an algorithm to minimize the total cost. Publishers will set up their algorithms to maximize profits. However, libraries and publishers do have a common ground: they both want to help people to find papers.
To support such a business model, a third party may provide a web service. It seems to be too wasteful to make every individual library and every individual publisher maintain a separate web service. Something like Amazon.com or Last.fm for papers might do. The service can also be an extension of services like EZproxy or CiteULike.
Related blog entries:
Note added on 26 July 2006. I posted an entry on the long tail of papers.
Few journals can make that claim, however. Most journals are only relevant to several people in an institution. Furthermore, few researchers read any scholarly journal from cover to cover. Rather, we all read individual papers. However, libraries subscribe to journals, or even bundles of journals. As a result, the libraries pay for many papers that nobody reads, and miss other papers that someone would like to read.
This business model is bad for authors and readers, and possibly even bad for publishers. Technology now exists to distribute information far more efficiently, in a unit consistent with how people consume the information. For example, many people now prefer buying individual songs to albums. See a recent book, The Long Tail, by Chris Anderson, the editor-in-chief of Wired, for a remarkably perceptive analysis of media industries.
The same business model may apply to scholarly papers. One may argue that journals, like albums, were invented as a packaging technology to suit the old economics of delivery. As scholarly papers are all online, the name of a journal becomes simply a tag to the papers published in that journal. Maybe a powerful tag, but a tag nonetheless. So far as how papers should be distributed, the name of a journal should serve the same function as all other tag-like entities: keywords, names of authors, etc: the tags help readers to sort papers and set priorities. It makes no sense for anyone to insist that papers with any particular tag be delivered as a bundle.
Many publishers already offer individual papers for sale online; for example, the cost is at $30 per paper for many Elsevier journals. Once a reader buys a paper, it seems reasonable to share this paper with his close colleagues, and it also seems reasonable to store the paper for future use. Perhaps we can formalize this practice.
How about we treat a paper just like a book? With one click, a reader will have the paper, and his library will automatically pay for it. Once bought, the paper is accessible to every user of the library. We can also collect statistics. If the users of a library buy many papers in a journal, the library should subscribe to the journal. Libraries will set up an algorithm to minimize the total cost. Publishers will set up their algorithms to maximize profits. However, libraries and publishers do have a common ground: they both want to help people to find papers.
To support such a business model, a third party may provide a web service. It seems to be too wasteful to make every individual library and every individual publisher maintain a separate web service. Something like Amazon.com or Last.fm for papers might do. The service can also be an extension of services like EZproxy or CiteULike.
Related blog entries:
Note added on 26 July 2006. I posted an entry on the long tail of papers.
posted by Zhigang Suo at
9:29 AM
15 Comments:
In your investigations you might also want to consider the following problem. Most libraries and major universities also have a responsibility to preserve all of these papers for future scholars. An article might not be used for anyone's work at your university when it first comes out -- but it might be critical to another researcher's in a few years.
That being said, most of us prefer the convenience of accessing these papers online through our libraries. Most libraries end up subscribing to both an online version and a paper version then.
If funding is insufficient the library might consider keeping only the online subscription. However, when the budget requires a further cut back or they want to allocate those funds to more widely read journals, they are then faced with a greater problem. If they stop subscribing to the online journal at that point -- they now don't even have the previous issues they paid for. Only the paper subscription assures them that if they stop subscribing they'll still retain those back issues.
By
Anonymous, at 7/22/2006 1:04 PM
Dear Zak:
Thank yuo for the comments. Once again we resonate. I don't have statistics, and don't know if Harvard collects such statistics. But there should be an easy way to collect such statistics, which will be a foundation for lots of fun stuff we've been talking about.
My hunch is a vast majority of papers do not receive any down load at all. They are in thelongtail, to use a popular tag of the day. Just for the fun of it, the tag might as well take the place of the old one, theparadigmshift.
By
Zhigang Suo, at 7/22/2006 1:36 PM
Dear Tanya:
Thank you very much for your comments, which made some important issues explicit. But are you really arguing that libraries should still serve the function of archiving scholarly papers? There are better ways to archive papers than filling premium space with dead trees.
If P3 is implemented, all papers will be available to researchers at all time. For a paper desired by a user, the library should pay only once. A third-party server should remember permanently that a library has paid for the paper. Subsequently, when any other users of the same library wants this paper, nobody should pay for it any more. That is, we treat a paper just as we treat a book.
By
Zhigang Suo, at 7/22/2006 2:01 PM
Here is the thread on P3 on the CiteULike discussion board. Be sure to read the hilarious comments by Richard Cameron, the creator of CiteULike.
My response to his comments follow.
Dear Richard:
Your reply made me laugh. I'll take liberty to make a link from Applied Mechanics News to your reply.
Well, I wasn't thinking about saving money for libraries. Rather, a possible goal is, for a given amount of money, how can researchers get desired papers. At Harvard, we spent a lot of money per person on journals, but we still miss many journals. We have a very small engineering faculty, about 60 people. You wouldn't believe this: we've just subscribed to the IEEE bundle very recently. For all these years, we muddled through without subscribing to the IEEE bundle. Do we really suddenly need so many papers from IEEE journals? I wouldn't think so. If we change to the P3 model, perhaps with the same amount of money we will better meet our needs.
Because many publishers already offer individual papers for sale online, I thought some libraries might already be doing P3 at a large scale. Then, the remaining issue is just technical: we need a web service to automate the process.
I hope some of your librarian friends can shed light on these issues. I'm curious what prevents a library from practicing P3, given that publishers already offer individual papers for sale.
I've also sent my initial blog entry to a few people working for publishers, and hope some of them will respond.
Best wishes,
Zhigang
By
Zhigang Suo, at 7/22/2006 5:32 PM
You asked, "But are you really arguing that libraries should still serve the function of archiving scholarly papers?" Yes, however I'm not advocating keeping more _paper_ copies. I'm pointing out that publishers normally don't allow libraries to archive digital copies of papers thereby making them available to their patrons after canceling their digital subscription. In addition, it sounds like you're saying that a library wouldn't add a paper to their collection unless someone requests it. That's would be a bit short-sighted. Popularity in the present doesn't determine usefulness in the future. If that publisher is no longer around in 10 years... what then?
You say "A third-party server should remember permanently that a library has paid for the paper." That would be nice. However, if you're referring to the publisher as the 3rd party, when a library no longer pays the publisher a fee, all services stop unfortunately. As you can imagine, there are still costs associated with "remembering" and then providing this access.
Another problem with 3rd parties, or publishers, being the sole source for saving publically available digital copies of these papers is that if maintaining that data is not profitable or if they go out of business, the data (papers) are lost.
However, there is a library project working with publishers to solve this problem along with the "remembering" feature. Maybe you're already familiar with LOCKSS? It uses a p2p system among participating libraries. I quote from their site:
"Under the DMCA, publishers must give permission for institutions to preserve copies of their copyrighted material. We designed LOCKSS so that each peer will only supply a copy to another peer if it remembers that peer proving that it had a copy in the past. On this basis, we have been able to reassure publishers that the system does not represent a threat to their business models, and many have agreed to use blanket license terms to permit institutions to use LOCKSS to preserve their access to the material to which they subscribe. It is clear that they would not agree to a system in which it was possible for non-subscribers to access material they had not paid for." http://www.lockss.org
So if I understand P3, it's copyright law (supported by publishers) that prevents it. You say, "How about we treat a paper just like a book? With one click, a reader will have the paper, and his library will automatically pay for it."
You already have that. You probably have access to Ingentaconnect.com or some such service through your university library.
But then you add "[o]nce bought, the paper is accessible to every user of the library." Yeah, that would be nice too. But once again, copyright law prevents it. When a library obtains a paper for an individual from a journal to which they don't subscribe, that person is pointed to "The copyright law of the United States (Title 17, United States Code)..." see for example the bottom of this request form: http://www.library.hbs.edu/forms/article_request/
Really, you should talk to your local "electronic resources librarian." This is the poor soul saddled with negotiating all of those contracts with publishers for digital subscriptions. He/She will think P3 is nice but then give you a gentle introduction to some rather archaic conditions imposed on digital documents and subscriptions.
So on that we certainly agree -- these publishing models and subscription agreements are at best just sadly based on a pre-digital medium and at worst dauntingly archaic for economic gain at the expense of scholarship. (Not to paint publishers as the villain -- we all like having a paycheck.)
By
Anonymous, at 7/22/2006 7:11 PM
I agree with the comments from Richard Cameron and tanya.
It seems P3 fits more the interest of libraries, but not the publishers. Unfortunately, the libraries have much less bargaining power in the scientific journal business. The current $20 some price per paper is just for individual use. If you push P3 to publishers, I guess the price might be tens or even hundreds times more, depending on how popular this paper would be.
For the music industry, it's a little different. The online track-by-track sale contributes extra profit for music publishers, without threating their traditional album sales in retail stores. About 80 percent of the profit in online music sales goes back to publishers. The online music retailers, such as iTunes.com, are willing to run the business, because they can make more money by selling more iPods. Even for music industry, the future of online retails are still full of uncertainties. Only a few big players, such as Apple, can take the risk to initiate the new business model, and more importantly, with sharing profit with publishers.
Maybe we sound a little bit pessimistic for this brilliant idea, but I guess it really takes time and also power to realize it. One possibility might be: get open access publishers, such as PLoS, motivated, and start the initial trial of P3 there. It seems they might be less reluctant to buy the idea.
By
Teng Li, at 7/22/2006 8:57 PM
A related idea - which springs to mind when Tanya talks about the poor "electronic resources librarian", unhappy with his lot in life, and negotiating contracts with a thousand different publishers - is that these poor unfortunate souls should somehow club together and bargain collectively rather than individually.
Although that sounds dangerously like a bad 1960s movie about trades union, with images of the Teamsters dabbling in organised crime (and it's unclear whether academic journals *could* be bought on the black market in dark back-streets), what I'm actually proposing is something along these lines:
Company X sets itself up in business with the sole aim to provide an outsourced service to the electronic resource librarians. They offer to remove all the tedious mucking about with paperwork and emails associated with subscribing to a particular journal, and eliminate the need to haggle over the price, simply because Company X will have pre-negotiated (the tricky part) a standard per-user cost for any given journal.
Through the miracle of technology, Company X could probably provide quite a slick service: providing a single web-based interface which lets the librarian choose which journals the university wants, telling him or her how much the subscription costs, sending a bill at the end of the month, and collecting statistics on which journals are actually being read. To answer Tanya's other concern about preserving articles for future use, let's suppose (again, potentially quite tricky), that Company X negotiates a deal with the publishers where it can keep copies of the articles and deliver to the universities in a non dead-tree form even if the university cancels its subscription.
Note that Company X is not competing on price just yet. All it's doing is providing a service which makes life less dull for the librarian. That's good, because it's going to be remarkably difficult for Company X to negotiate a better deal than the universities do at the moment (unless we do take the Teamster's rather violent approach to solving the problem - which may, arguably, be slightly unethical). As such, it's only really going to be cost effective for the more obscure journals. Librarians spend God knows how long arranging subscriptions for a journal which only five people in the university read. That's where they're going to want a service to do this, and that what Company X could offer. Even the publishers might be happier just getting one consolidated order each month rather than having to pay clerical staff to wade through all the individual orders from hundreds of universities.
Company X slowly works its way up the fashionable "long tale", going from the most obscure journals to the slightly less obscure. It negotiates deals with the publishers, and makes more and more content available for librarians to acquire at the click of a button. Ultimately, Company X becomes a fairly big player in the market.
The relationship between the libraries and the publishers is now much more or a level playing field. Up to now, Company X has been reselling content at roughly the same price as the university could have got it for anyway, but it suddenly becomes possible for Company X to compete on price too. It could do that by traditional means like hiring a salesman who smokes cigarettes, and sits in a room with his opposite number at the publishers, and negotiates a better deal with a smile and a shoeshine. Or, it could create a more transparent market by having, say, an auction based market. Universities could bid to say how much they're prepared to pay for a journal, and publishers could bid to say how much they're prepared to sell it for - the price is binding for, say, a year. There are all sorts of market-models which might result in subscriptions being bought and sold at a "fairer" price, none of which are possible if the universities negotiate their subscriptions individually.
By
Anonymous, at 7/23/2006 4:40 AM
Dear Teng:
Perhaps you have overestimated the value of each paper to a publisher. Perhaps we have collectively underestimated the willingness of publishers to innovate. After all, the Internet is for us all, including the publishers.
I’m not an economist, but here are some current numbers. For Journal of Applied Physics (JAP), authors have since 2005 had the option to pay $1800 to the American Institute of Physics (AIP), the publisher, to provide free access to a published paper, in perpetuity, to any online user. AIP also offers individual papers for sale, at $23 per paper.
Let’s say AIP adopts P3 instead, and wishes to have revenue of $3000 per paper. At a price of $30 per paper, each paper on average only needs to have 100 buyers world wide. Are these numbers unreasonable?
In May 2006, Elsevier announced that, for six nuclear physics journals, authors have the option to pay $3000 per paper to make a paper freely available online.
PLoS now charges authors $2500 per paper. A "no questions asked" fee waiver exists for authors who do not have funds to cover publication fees. PLoS is an open-access journal. There is no need to talk about P3 for any journal that is already free online.
Some one has to pay for producing, distributing, archiving papers, and possibly profits. The number seems to be between $1000-3000 per paper. Journal subscription is one way to generate the revenue, and author sponsorship is another. P3 is a third way, now beginning to be plausible because of the experience of the online music store.
Compared to journal subscriptions, the benefit of P3 is to have all papers available to everyone at all time. There is no need for multitude schemes of packaging and negotiation. All publishers need to do is to set a price that the market can bear. The above numbers do not suggest anything outrageous.
Compared to author sponsorship, P3 shifts the cost from authors back to libraries, so that good papers are available to everyone, regardless if their authors can pay.
By
Zhigang Suo, at 7/23/2006 7:36 AM
Dear Richard:
Thank you very much for Your Company X. Toward the end of my initial blog entry, I alluded to an Amazon-like or CiteULike-ish entity as a universal interface (marketplace?) between publishers and libraries. Your comments have made the idea a lot more explicit and appealing.
I do believe, however, the pay needs be at the level of individual papers, rather than journals, especially for specialized journals. The pay-per-paper model seems to be scalable to any institutional size. It will allow a small place like Harvard to be on the equal footing as a big place like the University of California, so far as accessing to papers is concerned. We should all pay what we want, and let the market decide a fair price.
Perhaps the complicated deals between publishers and individual universities is an effort to price according to use. In a sense, the price is a rough proxy of P3. This I do not know for sure. But as you said, the technology now allows us to be far more transparent and efficient.
The Amazon-like and CiteULike-ish Company X will leverage the Internet to benefit authors, readers, publishers and libraries. The foundation for pricing probably is inescapably some form of pay per paper.
By
Zhigang Suo, at 7/23/2006 8:41 AM
Actually, arranging with publishers to subscribe on a _per journal_ basis, let along a per paper basis, is fairly new. It's my understanding that most publishers set up online subscriptions based on "bundles." I imagine it goes something like, "if you want to subscribe to these 10 really popular journals, you're in luck because it happens to be part of our 50 journal bundle on topic X."
From a 2001 article:
http://www.library.ucsb.edu/istl/01-spring/article3.html
"Journal bundling refers to the practice of aggregating all titles produced by a publisher into a single product, or subject-based subsections. This comprehensive product is then marketed and sold as an all-or-nothing deal: a library can purchase access only to all of the titles within the package, or to none at all. Within the last five years, these packages, or bundles, have become the favored subscription model for the dominant commercial publishers of Science, Technology, and Medical (STM) electronic journals."
So, once the publishers agree to a per title model we might make start making some headway with a per paper model.
Regarding Richard's rabble rousing -- libraries actually have banded together to negotiate at that level. There are several examples, among them SPARK http://www.arl.org/sparc/index.html:
"initiative of the Association of Research Libraries, is an alliance of over 200 academic and research libraries working to correct imbalances in the scholarly publishing system. These imbalances have driven the cost of scholarly journals (especially in science, technology, and medicine) to insupportably high levels, and diminished the community’s ability to access, share and use information."
Meanwhile on the same theme as using an iTunes model for papers, see Ursula Holtgrewe's "Intellectual Property, Communism and Contextuality" A Non-Essentialist Exploration of German Digital Copyright and the Public Domain
http://www.sti-studies.de/articles/2005-01/holtgrewe/Holtgrewe-STI-2005.pdf
By
Anonymous, at 7/23/2006 9:53 AM
What a thread! I can't resist adding some thoughts, based on my own experience in the industry and attendance at many conferences where this topic (often termed "just in time" versus "just in case" purchasing) has been the subject of much debate for many years.
1. Isn't Company X a subscription agent?
2. Bundling is so last year... ;o) Both publishers and libraries are increasingly less keen on what was known as "the big deal" (aka consortial selling/purchasing). I'm also fairly confident that no publisher has restricted subscriptions to bundles rather than individual journals. They just might offer you the latter first!
3. Tanya is on the money with her comments re. copyright. You would need to check the terms of the licence under which you had purchased the paper, but in most cases they expressly prohibit sharing.
4. I can't find the reference here and now, but I'm sure I've heard of at least one library (corporate, I think) which has stopped buying subscriptions ("just in case") in favour of pay-per-view (as it tends to be known; P3 in your terminology, or "just in time").
5. There are several other organisations aside from LOCKSS which are engaged in archiving initiatives to protect the scholarly record. They include not-for-profits and national libraries. I summarised in my recent article in Ingenta's newsletter for publishers.
By
Charlie Rapple, at 7/24/2006 10:48 AM
Message from Ross Singer, dated 22 July 2006
There actually are some examples of this in the wild, although generally not by the University's choice.
Notre Dame recently had a parting of ways with Springer and now orders all 'new' articles (2006+) via Springer Express or Springer Cafe or whatever fool name they call it. Mark Dehmlow made a presentation about this at ELUNA back in June, but I can't find it anywhere.
Colorado State did something (pre-web journals, mostly) somewhat similar when their library flooded and they lost the bulk of their collection; see this webpage for detail.
Quite honestly, though, a 'pay-per-view' model would quickly put libraries in the poor(er) house. By purchasing our 'packages', we actually get a /lot/ of content that somebody somewhere /might/ want /at some point/ for what would probably work out to be roughly the same price as the piecemeal approach you propose.
Another problem would be budgeting for this model. Libraries already struggle to maintain their collections year-to-year (some years more successfully than others). What happens when, with a 2nd quarter that had an unusually high number of 'must-read' articles, and the library is broke with 2 months to go in the fiscal year?
The biggest problem, however, is cultural. Faculty must publish in these journals, the libraries must carry them for accreditation, so on and so forth. Faculty never approve of titles getting cut in their field. This is, by far, the most difficult part of the cycle to break.
By
Zhigang Suo, at 7/25/2006 8:05 AM
Paul Ginsparg, the creator of arXiv, estimated cost per paper in this article.
By
Zhigang Suo, at 7/26/2006 5:01 AM
Dear Andy:
Thank you so much for your input and encouragement. I'll show your message to our librarians. As you can imagine, the hit-or-miss subscription model is particularly hard for small engineering schools.
By
Zhigang Suo, at 7/26/2006 9:59 AM
This is a test of Comments
By
Zhigang Suo, at 9/30/2006 7:37 PM
Post a Comment
<< Home