Green Open Access (OA) can be a factor in a librarian’s decision to cancel a subscription to a journal. Thus, publishers with liberal Green OA policies increase the chance that their journals will be cancelled. As Leo Durocher is famously quoted as saying, “Nice guys finish last.”
It’s important to recognize that a factor by itself is not determinative: I have heard of no librarian who categorically cancels publications with content that can be found from Green OA sources. But alongside other considerations–the merits of the journal, its fit with the institution’s research priorities, the amount of available budget–Green OA can tip the scale unfavorably. When such a journal is cancelled, the money that might have gone toward its purchase is spent elsewhere, usually with a publisher that has either a stronger or more relevant program or to a publisher that has decided to look after its own interests and not have a progressive Green OA policy. Money saved by cancelling one journal may go to pay for the price increase of another journal.
This is not the way we want things to work, of course; we want to reward people and organizations for their civic-mindedness. And sometimes we do, especially in flush times, when hard economic truths are not sticking up through the cloud of good intentions. But often in lean times we get “lean-spirited.” For my part I prefer not to judge people who behave in a situation precisely as most others would in the same situation.
My fellow Kitchen chef Rick Anderson addressed Green OA and cancellations and found himself in a maelstrom. Here is a link to the thread where this discussion took place. I encourage everyone to read it all the way through as it is informative, witty, and even entertaining. Rick said that he had instructed his staff to consider the availability of materials through other venues, including Green OA, when deciding which journals to retain and which to cancel. The logic of this is clear: if you have only so much money to spend, would you prefer to spend it on something that your institution has no other access to or on something that is available in some form in an institutional repository or other Green OA venue somewhere? Note the qualifier: If you have only so much money to spend. Other things being equal, I know what I would do in that situation. Rick’s is the voice of pragmatism, and I strive to emulate it.
Naturally, pragmatism can not be allowed to stand. Stevan Harnad engaged Rick’s comment and asserted that such a policy was a very bad thing since it would set back the advance of Green OA. This is an interesting remark, as it reveals Professor Harnad’s conviction that librarians, indeed the whole world, should view the achievement of his idiosyncratic goal as their highest priority. As far as I know, it is not the mission of Rick’s institution or any other to put Green OA at the top of a list of desiderata. Most institutions put service to their own institutions first, as one would expect. Cancelling Green OA journals will indeed set back the advance of Green OA, but that’s beside the point.
Several years ago I made this very same argument when I was invited by John Sack to speak at a HighWire Press client conference. I wish I had my slides from then; this was before the advent of SlideShare and Google Drive, so my only copy is on a useless hard drive that has been gathering dust on a shelf for years. My argument then and now is that journals are cancelled at the margin–that is, journals that are at the center of an institution’s programs are not likely to be cancelled, but journals that are at a distance from that center are at risk. Among the factors for cancellation (assuming the journal is of good quality) is the availability of the journal or portions of it elsewhere such as in bundles with many other journals (e.g., ProQuest or EBSCO collections) or in institutional repositories. After I gave my presentation, several people in the audience asked for a copy of my slides, which they wanted to use to explain to the professional societies that they worked for why Green OA represented a business problem for them.
The fact is that Green OA is like ouroboros, the snake that eats its own tail. It works best when it is less than perfect; when it is perfect it undermines the economics of the original publication. Let’s share a gedanken moment here. Imagine a journal that is sold on a subscription basis. Many of its articles can be found elsewhere: in institutional repositories, on authors’ Web sites, perhaps in a government archive. But items in those other repositories may be hard to find. Perhaps they don’t show up conspicuously in Google searches or the user cannot easily determine if they are the final version. In such an environment users are likely to look for the authorized instance of the article, and that is a compelling reason for a librarian not to cancel the journal’s subscription. In this environment publishers may not be overly assertive about Green OA because it is not hurting their interests.
But fast forward to a world where Green OA is getting better and better. Now you can find an article simply by typing the title or some keywords into Google or some other search mechanism. The Green version of the article appears; there is no need to seek the publisher’s authorized version. Librarians will observe this and pragmatically say that that journal can be cancelled.
So the better job Green OA does, the more it will be resisted. To keep Green OA programs going, they have to be imperfectly implemented.
Of course many people don’t care a fig if a journal gets cancelled, but they may not realize that Green OA requires that the journal keep going. The journal’s subscriptions underwrite the editorial and production process, including the management of peer review: without those subscriptions there would be no formally vetted articles in the first place that can then be made Green. This points to the advantage that Gold OA has over Green: it has a working business model attached to it. Such Gold OA services as the Public Library of Science and Hindawi are hugely profitable. They will be with us long after Green OA is a distant memory.
Whatever publishers may think about Green OA, they don’t have a free hand in its implementation or its obstruction. Funding agencies often mandate Green deposits and some universities mandate that faculty place copies of articles into institutional repositories. But beyond that is the Internet itself, which is a huge copying machine. The cost of stamping out Green copies of articles is more than any organization can bear. Thus publishers adopt a pragmatic (that word again) stance on Green OA, challenging it when the circumstances are serious, tolerating it otherwise. The creation of CHORUS is an instance of a strategic engagement with a Green OA policy that could undermine publishers’ interests.
But a publisher’s situation concerning OA is even more complicated than that. A publisher’s first job is to attract the best authors, but what do you do when some authors insist on some form of OA publication? Some of these authors vote with their wallets (or their funders’ wallets) and submit articles to PLoS, a Gold OA service; some opt for hybrid programs with established journals. Do publishers want to get into an argument with authors about depositing articles into an institutional repository or posting them on a personal Web site? So even as libraries are weighing OA in their collection decisions, publishers are seeking to find the right balance with authors concerning OA options. OA, in other words, is as much a matter of diplomacy as it is of policy. Nice guys may finish last, but nasty guys may not get to play the game at all.
You would never know how nuanced the overall situation concerning OA is if you follow the topic on a multitude of blogs and newsgroups. OA is often talked about as a vision, a mission, with soaring rhetoric and an invocation of the highest values of the scholarly community. Whatever turns you on, I suppose. The practical aspects of OA are not clearly defined; they tug one way, then the other; and they are often, perhaps always, partial and even sloppy in their implementation. The only constant is the presence of human judgment at every step. That librarian who is trying to gauge the implications of Green OA before she renews or cancels a subscription is at the very center of the paradigm.
54 Thoughts on "When It Comes to Green OA, Nice Guys Finish Last"
Interesting post and your speculations make sense but where is the data supporting it? I am not a librarian but I have been on our university’s faculty library committee for several years. The focus of our acquisition librarians seems to be what patrons indicate they want and request, the amount journals are downloaded, accreditation requirements all impacted the negotiation of blocks of journals through big deal contracts. I haven’t asked them specifically but I doubt they read green policies or go out and check the percentage of articles in a journal that are available on the web before making purchasing decisions.
If you are really interested in the impact of green OA you probably should read Bo-Christer Björk and his colleague’s excellent article Anatomy of Green Open Access published in Journal of the American Society for Information Science and Technology. No worries, you can get the green copy here.
JASIST is a great journal, they allow immediate green OA and fortunately at least my library hasn’t dropped their subscription yet.
Green OA is pretty spotty and tends to be stale. With or without mandates it tends to take a good year so so after publication before most faculty get around to posting their articles. As Bo-Christer points out elsewhere delayed OA where publishers voluntarily make there actual published versions available after a delay gets the material out just about as fast a green OA. It would seem that would get your journal’s subscriptions cancelled a hell of a lot faster than green OA yet a fair number of society journals do it.
What I found most interesting in the article and quite troubling is the an estimated 35-50 percent of the green articles available are likely to be violating the publisher’s copyright!! Apparently a lot of authors put their page proofs or downloaded copy of their article from their library and put that up on their web sites. A fact I am sure gives publishers heartburn. In all the discussions of green OA nobody seems to notice!!! I don’t know but If I were a librarian I think I would be a little hesitant to drop a subscription because my readers had the option of going out and getting a pirated copy of an article off some author’s web site.
This is an interesting economic modeling problem. Basic economics says if you product is available free no one will buy it. But availability is part of the price so if it is free where it is but laborious to get it is actually expensive.
Exactly. Where is the data to support these claims? The leap in logic that’s being made here is pretty easily understood but, without anything to back it up, it’s hard to take all of this very seriously.
From personal experience, our librarians just got through a wholesale review of all of the institution’s subscription services in order to have a plan in place in case of budget cuts. The only factors that they considered were cost, usage, and format, and I feel that that’s fairly typical.
Exactly, this comes across as propaganda. It makes intuitive sense to the author based on his biases. i know Scholarly Kitchen tends to be in the anti-OA camp, but this overt even by their standards. It’s not scholarly, it’s speculation.
Business, and publishing in particular, tends to work in a speculative manner. It is not possible to predict the future, so one examines all the contingencies and makes a plan based on best available knowledge. Ignoring what seems an incredibly obvious supposition (if you make everything free, there’s no longer any reason to pay for it) would be foolish.
If you want data, perhaps looking at the subscription rates of newspapers and magazines, and how they have changed since their articles have been made freely available online. An imperfect comparison, but probably a meaningful one.
We can look at the studies that have actually been done regarding downloading music and music sales. The intuitive point that record labels made was that if people could get music for free then they would never buy it. This obvious supposition has been found to be incorrect. Here’s an interesting paper: http://www.scribd.com/doc/131005609/JRC79605. An imperfect comparison, but probably a meaningful one.
The point wasn’t the illegality of downloading music, it’s that labels feared and argued that if people could get music for free by downloading it from torrents that they wouldn’t pay money for it. This was found to be a false assertion. Downloading music for free didn’t stop people from buying it.
The actual numbers on music company revenues since 2000 does not support this assumption.
The point isn’t the illegality, it’s the reason why record labels were so afraid of it. They argued that if people, though downloading illegally, could get music for free, why would they pay for it. Your point is, if people can get articles for free, why would they pay for it. In the case of music downloading, this was found to be a false assertion. People who were able to download music for free from torrents weren’t less likely to buy it. You can’t just dismiss something by saying, “well that’s different” when it supports a position you oppose.
Except that as music was made freely available, music purchase revenues massively decreased. Which would seem to prove the point that lots of people won’t pay for something they can get for free.
Except that music purchases hit an all time high in 2012. All time. Physical sales continue to decrease because the CD is a dead medium. But digital sales are booming. http://www.huffingtonpost.com/2013/01/09/music-sales-2012-digital-physical_n_2440380.html
Looking solely at the number of units sold is misleading, because you’re talking about different units, sold at very different prices. There is not a 1:1 correlation between selling a $17.99 CD in 1999 and a 99 cent digital track in 2013.
Business is down, way down since music began to become freely available. One can argue the reasoning for this, but the sheer overall revenue numbers are very stark.
Revenue is down certainly. The key point is it’s rising again. The record labels moaned and complained about online music but when they finally admitted that what people cared about was access and offered a product that fulfilled that need, people began buying it. I’m not exactly sure what you assume the endgame should be regarding, in this case, Green OA. Maybe you’re advocating for publishers to change licenses to prohibit self archiving. That’s certainly a possibility. Record labels tried to fight against new formats hoping they could just stop them from ever arriving. It didn’t work out so well, and now, by addressing consumer desire rather than just their own, sales are creeping up again. It’s a basic tenet of capitalism. Give the customer what they want they’ll buy it.
Revenues are up slightly after nearly 15 years of dropping every single year. That doesn’t paint a pretty picture for sustainability.
What I want is not the abolition of Green OA, but a careful, evidence based set of policies used to determine adequate embargo periods. For Green to work, there needs to be something of a balance involved and immediacy of access plays a role.
There is also the question of, how did we gain a notion that we somehow owe it to publishers to make a profit? We are talking about a free market, right? No one in any business is owed anything. You succeed or you don’t. Next we’ll be calling for government subsidies to academic publishers.
I have no idea how you arrived at that notion. I’ve certainly never heard it from any publisher. All we want is to let the free market work, let competition thrive. Scholarly publishing has evolved over hundreds of years to serve the needs of researchers. If we’re indeed obsolete, then we’ll be replaced by something else. Government and funding agency mandates put a crimp in a truly free market, but most of us, because we see them doing good for scholarship, are working with those agencies to help best set up sustainable and rational policies that will best see their needs met over the long term.
And for the record, many of us work for not-for-profits that return any surplus we earn directly to the academic community.
Next we’ll be calling for government subsidies to academic publishers.
Perhaps a bit ironic coming from the academic research community which is funded for the majority by government subsidies.
You say “Downloading music for free didn’t stop people from buying it.” Excuse me salted, but are you claiming that the people who downloaded a given song for free then turned around and bought it? I am skeptical, to say the least. Do you have any data to support such a strong claim?
Or do you mean that pirating music by some people did not stop honest people from buying that same music? In that case I agree but it is irrelevant to the green OA case in question, where the free stuff is legal.
“Are you claiming that the people who downloaded a given song for free then turned around and bought it?”
That certainly does happen. Exhibit A: all the Dar Williams albums I bought that I would never have bought had I not first heard and loved pirated copies. Of course I can’t say how common that is, but it happens.
If you somehow think that a grant to a researcher and a subsidy to a for profit corporation are the same thing then, yes, it would be ironic.
If you think that all publishers are for-profit, and that researchers and research institutions don’t turn massive profits from taxpayer funded research grants, then you perhaps know less about publishing and science than you think. Look into the origins of Google (NSF funded) or Alimta (NIH funded) for a few quick examples.
Depends on your field–when I was at Caltech, every time construction started on campus, there was a running joke that they were finally starting to build the Porsche-only parking lot for faculty with startup companies.
And you might be surprised at the number of university publishers that are subsidized by their universities and run at a deficit.
Some quick numbers for you:
Google has paid Stanford $337 million in royalties from its NIH funded research.
NYU made $185 million in 2012 from technology transfer
Columbia, MIT, Northwestern, and the University of California system each made > $100M in technology transfer in 2012.
From 2005 to 2012, Princeton has made $524M in technology transfer profits, most of which comes from Eli Lilly for an NIH funded drug.
Lots of interesting stuff here. I do agree that the Green-at-all-costs advocates do seem very ready to overlook the fact that the better Green gets, the more publishers will hate it. That’s why in the end I favour Gold: not just because it yields a better product, but because it puts authors and publishers on the same side. The importance of making that switch is apparent from the rather scary line in your post, “The cost of stamping out Green copies of articles is more than any organization can bear”, an unguarded expression of straightforward antagonism from publishers to authors and libraries. There’s no cause for that under Gold.
Finally, I do have to comment on this: “OA is often talked about as a vision, a mission, with soaring rhetoric and an invocation of the highest values of the scholarly community”. That is absolutely right, and I wonder how much of the impedance mismatch between authors and publishers is precisely because of (at least some) publishers inability, or unwillingness, to get this. From my perspective, and most of my colleagues, all our research is indeed a mission, and OA is a crucial component of making it work. If we didn’t see a vision, we’d be making fortunes in stock-trading or marketing rather than doing what we do.
There are a number of contradictions in the green OA concept, which are slowly coming to the surface. Consider mandated embargo periods. On the one hand it is argued that journal content loses value quickly, so subscriptions will not be lost from making it freely accessible in time t. But on the other hand it is argued that this content is still so valuable at t that access then must be mandated. Both cannot be true. The reality is that there will be losses, so the proper value of t is that where the benefits clearly outweigh these losses. This requires economic and regulatory impact analysis which has yet to be done.
Then too, as Joe points out, immediate access is potentially self destructive. This is why no government is mandating it. Rather the green OA policy issue is the proper mandated embargo period t, which probably varies by discipline, making it hairy indeed.
Green copies are, and likely always will be, an inferior product to the official copy of record. My journal’s pdf files include our organization’s logo. Whatever rights one claims to copy the content, the trademarked logo definitely is not legal to copy without explicit permission.
The logo, which could be edited out of the pdf, is only the tip of the iceberg. As we move into the post-paper era, journals will include all kinds of apps to, for example, run an interactive demonstration. It’s not impossible for an outside archive to copy such products, it just becomes more and more difficult to produce a true copy and to stay within the law.
Just a point of detail: trademark law is quite separate from copyright law, and pertains to misrepresenting a good or service as originating with a well-known entity when in fact is does not. If I generate my own PDF of a paper I am working on, and slap the Nature logo on it, I am misrepresenting my own work as Nature’s, and that violates trademark law.
Of course your broader point is correct.
But my experience has been that authors aren’t interested in apps. They just want the paper and the data, so they can make their own uses of them rather than the uses that someone unconnected with them came up with.
This discussion suffers from a basic confusion, namely the huge difference between immediate deposit, no embargo green OA and embargo based green OA. These two cases are fundamentally different in many ways yet they go by the same name — green OA, so we have a deeply ambiguous concept in play. That is never good.
There is also the issue of what version is allowed, submitted, accepted etc. SHERPA/RoMEO provides a database stating what is allowed by various journals.
I find Dr. Harnad’s response here somewhat appalling. Progress in implementing Open Access will come from open discussion, analysis and experimentation, not from censorship, obfuscation and withholding information. When voices as disparate as Kent Anderson and Cameron Neylon are in agreement about OA reaching a new era of practical implementation, it should be a sign that Harnad is out of step here.
It’s always valuable to have someone willing to point out the state of the Emperor’s clothing.
I agree totally. It is not his views but that Rick Anderson should not even be allowed to discuss this issue on the forum.
In his defense, I think that Harnad considers discussion to be advocacy, which in policy contexts it often is. In diplomatic language “We will have to consider our options” is actually a threat.
Here’s the exact quote:
But more important than any of that is the gross disservice that gratuitous public librarian announcements like that do to the OA movement: We have been objecting vehemently to the perverse incentive Finch/RCUK have given publishers to adopt or lengthen Green OA embargoes and offer hybrid Gold in order to get the money the UK has foolishly elected to throw at Fool’s Gold unilaterally, and preferentially.
Now is it going to be the library community publicly notifying putting publishers on notice that unless they adopt or lengthen Green OA embargoes, libraries plan to cancel their journals?
With friends like these, the OA movement hardly needs enemies.
May I suggest, though, that such postings should not go to the GOAL, BOAI or SPARC lists? Please keep such brilliant ideas to the library lists.
And please don’t reply that “it’s just one factor in our cancelation equation.” There’s no need for the OA community to hear about librarians’ struggles with their serials budgets when it’s at the expense of OA.
“There’s no need for the OA community to hear about librarians’ struggles with their serials budgets when it’s at the expense of OA.”
Wow. That’s a quote worth remembering.
To the degree that the hypotheticals discussed here are intuitive, they still miss the real world example of high energy physics and related fields. Almost the entirety of their current literature is freely available in Arxive. Yet libraries maintain their subscriptions to the key journals in the field because the campus folks who publish there still require the journals for P&T and other rewards structures. So the economic impact of Green OA is not that simple.
True, and this speaks to Joe’s point about journals generally being canceled at the margins. (I’ve made similar points in the recent listserv discussions.) High-energy physics and related fields are central to the research mission of many universities, so journals in those areas are less likely to be canceled even as more and more early versions of their articles become freely available in the arXiv and other repositories. That’s not to say that they won’t ever be — just that they’re likely to be more resistant to cancellation pressure than journals in less heavily-supported fields are.
It’s also true, I believe, that an awful lot of what’s in the arXiv is much more preliminary than what generally gets archived under a Green regime. That would factor into the cancellation decision-making as well.
Rick, do your acquisition actually go through journal by journal and check the availability of green copies of the article in a journal before deciding to cancel it? How do you know the percentage of content that is available and in what form? If you have time please read the section at the bottom of page 22 and top of 23 in the Björk et al. paper titled, “Degree of copyright compliance for uploaded green OA copies” I think it would make you pause before considering the availability of green copies in dropping a journal
PS This copy is legal green 🙂
Rick, do your acquisition actually go through journal by journal and check the availability of green copies of the article in a journal before deciding to cancel it?
No, not yet — Green regimes aren’t pervasive enough to justify that kind of routine review (though we are actually doing a study of our existing subscription list to see how much it overlaps with those titles identified as Green on SHERPA/RoMEO). But as Green becomes more pervasive and as our budget gets tighter, we’ll start doing more and more of that kind of review — particularly for journals that are higher-priced and closer to the margins of our library’s research focus.
That is true Rick, in fact I think some journals use arXiv as their submission system so what you get is the first draft. The way articles circulate is far more complex than these discussions allow for. That is why a computer model would be useful. But Joe’s basic point is still true. Immediate posting of published articles is ultimately self defeating.
Compliments to the chefs. Some suggested recipe upgrades:
1. No suggestion that institutions cannot or should not cancel journals if their articles are all or almost all Green.
(No such journal in sight yet, however, apart from some parts of Physics — but there it’s already been the case for over 20 years, and no cancellations in sight. For the rest, when Green OA — which grows anarchically, article by article, not systematically, journal by journal — prevails universally, because Green OA mandates prevail, all or most journal articles will be Green universally, so Green OA will not be a factor in deciding whether to cancel this journal rather than that one.)
2. The issue with Rick was not about the notion of canceling journals because their articles are all or almost all Green, but about cancelling journals (60%) because they do not have a policy of embargoing Green OA.
3. And such a perverse cancellation policy would not be a setback for Green OA but for OA.
I note in the scullery discussion above that my suggestion that Rick shold post his OA-unfriendly cancellation strategy to library lists rather than to OA lists amounts to a call for censorship over open discussion. I note only that I am not the moderator of any list, hence have no say over their content. It was an open expression, on an open list, of my opinion (together with the reasons for it) that such discussion belongs on another open list. I do post this SK comment with some curiosity, as my own comments to SK have more than once failed to appear…
Since journals are so often bought in packages these days, how often does it even happen that individual journal subscriptions are cancelled? One reason we used Green OA at Penn State was that all of our journals were in Project Muse, and the vast majority of our sales came through that source, not via single-journal subscriptions.
I’m not seeing any hard data here. When did Scholarly Kitchen take a cue from Fox News and move into outright fear mongering?
One man’s “fearmongering” is another man’s “due diligence”. What data, exactly would you propose be collected here? How would you set up the experiment, and what controls would you offer? Should a journal make all of its content freely available permanently and then see what happens to its paid subscriptions? How would you go about getting a journal to destroy its business in order to perform the experiment?
This sort of proposed censorship, an attempt to shut down discussion and analysis of any topic that doesn’t fit with one’s goals and wishes is exactly the opposite of what the OA movement needs as it becomes a regular part of the publishing economy. OA must stand on its own merits and overcome the very real challenges it presents. Calling a legitimate question “fearmongering” and calling the person asking it “Fox News” does not make the issue magically go away.
Basic economic principles do not have to be demonstrated by experiment. If these are reasons why they do not apply that has to be shown.
In principle, I agree. In practice, the small amount of data we have seems to suggest, counter-intuitively, that providing free access to content does not reduce payments, but in fact increases them. Evidence leaning in that direction can be found in the results of the PEER project, in the now numerous surveys showing that people who pirate music also buy more than than those who don’t, and in the publishing experiences of authors like Cory Doctorow, who allow their works to be freely downloaded.
I find all this just as surprising as you do, but these are facts as good as any others. At the very least, the jury is out over whether the apparently obvious idea that free => loss of revenue is actually true.
I’m not sure those examples are particularly relevant here. The question is what happens when Green OA becomes systematic, rapid and covers the entirety of the literature. Your examples here are the haphazard freeing of a small amount of content, perhaps more in line with the “freemium” business model. Not all music is easily and readily available through illegal downloading. And Doctorow is something of an early adopter, and edge case who gained enormous amounts of promotion through his free download policy. The second author to do the same got less attention (and fewer sales) out of the gimmick, the third even fewer and so on down the line. If every author moves to this method, the novelty wears off, as does the effectiveness.
These are different cases from what’s being discussed here. Making a portion of content free does indeed often increase overall usage and sales. Most journals already do this through “Editor’s Choice” or “Featured” articles which are made free. They drive overall usage and bring readers to the journals. Make them all free though, and it’s unlikely you’ll see an uptick in subscriptions.
The music example is also has a few further flaws. The edge cases, those who download lots, may indeed also buy lots. But it’s impossible to say they wouldn’t buy more without the free downloads, and more importantly, their overall numbers likely pale in comparison with mainstream music buyers. The overall revenue levels for the music industry have not increased because of the free availability of music in the post-Napster era.
“The question is what happens when Green OA becomes systematic, rapid and covers the entirety of the literature.”
You’re quite right that this is a very different question from the effect that current levels of Green OA have.
But I think that means it’s impossible for us to have any evidence, isn’t it? I can’t think of any analogous large-scale freeing of content in other industries.
Exactly. That’s why calls for data here aren’t all that helpful. We’re talking about speculation about what may or may not happen in the future. The more prepared we can be for different contingencies, the better, which is why it’s important not to shut down discussion on subjects that people find worrisome.
In our experiment with OA monograph publishing in Romance studies at Penn State, we had the sales of the series before it went OA to compare with, and overall the effect of online access to the monographs (set up in such a way that the entire work could be read online for free but only half of each monograph could be downloaded and printed out, thus leaving incentive for some to purchase as POD edition) was a decline of sales, though not a drastic decline. I’d peg it as somewhere around 25%. Cory Doctorow’s experience should NOT be held up as a model for scholarly publishing; there are too many features unique to his books to make his experience generally applicable to the kinds of books scholarly publishers bring out.
Here is a simple experiment. There are a number of journals that are free. For a modest fee I am sure some of them would be willing to offer a subscription option. Then we could see how many people prefer paying to getting a journal for nothing. Or for an equally modest fee I will build a simulator to test the hypothesis that people prefer to pay for stuff they can get for nothing.