Peter Suber's Predictions for 2009

Twitter icon
Facebook icon
Google icon
LinkedIn icon
e-mail icon
From Peter's December 2008 issue of the SPARC Open Access Newsletter

From Peter's December 2008 issue of the SPARC Open Access Newsletter.

I'm trying something new with my predictions this year.  Instead of guesswork on a large number of small subtopics, I'm sticking to two large ones: the new Obama administration in the United States and the growing recession worldwide.  Then I'll add a few bigger picture items not closely tied to any particular year.

One reason for the new approach is that progress on the major fronts in the OA movement is easy to predict nowadays, and has been for a few years now.  It's not because the progress itself is easy.  Far from it.  Launching new OA journals, OA repositories, and OA policies, and making them succeed, is hard work.  But predicting them is much easier than creating them.  Every day more people are committing themselves to do the hard work and every day they can draw upon a wider network of projects and deeper pool of experience and best practices.  The obstacles have not disappeared, but neither are they growing; but by contrast, the energy and wisdom to overcome the obstacles are certainly growing.  Predicting more successful journals, repositories, and policies, and then more and more, is like predicting more personal computers after the first generation of Heathkits, Commodores, Apples, and IBMs.  The momentum on all three fronts is unmistakable and unstoppable.

The details of that progress are harder to predict.  But the details I'm most curious to anticipate this year are what Obama and the recession will bring.

* Under an Obama administration and the new Democratic Congress, the NIH policy will prevail against the Fair Copyright in Research Works Act (a.k.a. Conyers bill).  At least one more federal agency will adopt an OA mandate.  And OA mandates at many more agencies will enter the deliberation stage, either in Congress or in the agencies themselves.

Bush had mild objections to OA, but they didn't stop him from signing the December 2007 bill requiring the NIH to adopt an OA mandate.  Obama has little track record on OA, although what we've seen to date suggests leanings in favor of OA.  But he hasn't yet spoken explicitly about the NIH policy or the possibility of extending it across the federal government.  My open letter to McCain and Obama in SOAN last month was a normative argument, not a prediction.  But now that regime change has come to the US, I'm also predicting some of what I argued for.  Obama has not only supported the open sharing of environmental data and presidential debate footage, he's made Harold Varmus one of his science advisors.  He is clearly getting excellent advice about OA.  He is far more committed than Bush to open government, transparency, and accountability for public spending.  That wouldn't be hard.  But he is also more committed to open government, transparency, and accountability than most members of Congress.  He will be far less indulgent than Bush to corporate lobbies wishing to derail the public interest for their own benefit.  That wouldn't be hard either.  But as we've seen, even Bush didn't let the publishing lobby stop the NIH policy.  It wouldn't take much movement away from special interests to protect the NIH policy from further incursions.

In short, the Obama administration will advance the cause from two directions:  a basic understanding of the issue and a basic commitment to good government.

The Democrats are two Senators shy of a filibuster-proof majority in the Senate.  But they're many votes shy, in both Houses, of a lobby-proof majority on OA issues.  A lobby-proof majority is one that understands what OA is, why it's valuable, and what is specious in the specious arguments against it.  A lobby-proof majority is attainable but we won't have one in 2009.  It's a way off because there are 535 members of Congress, because OA is a comparatively small issue, and because the successful version of the NIH policy is still fairly new.  The success of the policy is making new friends and impact, but gradually.  Over time it will cultivate a lobby-proof majority for OA, just as we now have a lobby-proof majority for the federal funding of basic research.  In the meantime, a realistic goal is to have informed and supportive leadership on the key committees, informed and supportive leadership in the White House, and a growing volume of OA articles reporting the results of publicly-funded research, triggering good results for patients and the economy, and generating good will among researchers, universities, libraries, physicians, manufacturers, and taxpayers.  If we focus on the committees responsible for science policy, appropriations, and the individual funding agencies, and overlook the committee responsible for copyright, then we're absolutely on track.

The Conyers bill or the equivalent language attached to another bill will re-emerge in the new session of Congress.  It may re-emerge in Conyers' own committee, the House Judiciary Committee.  Last month, Conyers abolished the Subcommittee on Courts, the Internet, and Intellectual Property, which held the September 2008 hearing on his anti-OA bill.  The new restructuring won't change any votes, but it will consolidate power in his hands, and diminish the power of Howard Berman, who had doubts about the bill and who chaired the abolished subcommittee.  But even if the bill is re-introduced, it will not become law.  The September hearing enlightened some members who previously knew only what publishers had told them.  The NIH policy is succeeding and its success is more compelling than any policy argument.  And at the right moment I expect that Obama will send a signal that he will not sign the bill.

The NIH policy covers so much literature in biomedicine (80,000 peer reviewed articles per year), and the compliance rate is climbing so quickly, that its opponents have little time left before even they will have to accommodate it.  Its success is moving up the dinosaur moment when TA publishers must adapt or refuse to publish NIH-funded research.  Most have already adapted, of course, a fact that tends to be lost in the protests of the publishing lobby.  But the clock is ticking for those who hate the idea of adapting.  This matters.  While publishers have the money to lobby against government OA policies forever, the question is becoming moot as the policy's friends grow in number and power and as its opponents revise their own policies to live with it.

As the clock ticks, expect to see more desperate measures like the Conyers bill.  If the bill fails or seems to be moving too slowly, don't be surprised if the AAP tries a lawsuit.  The only ground for a lawsuit would be copyright.  But if the NIH policy really violated copyrights, then publishers would already be in court and wouldn't find it necessary to try the Conyers approach of amending copyright law.  A lawsuit would be an act of desperation, but it could delay the further implementation of the policy, and delay the dinosaur moment for holdout publishers. 

I don't want to put too much weight on the fact that Obama was a professor and publishing scholar, teaching constitutional law at the University of Chicago Law School from 1992 to 2004.  We know too well that not all faculty are well-informed about OA.  But academics understand much better than oil industry executives that the economic interests of conventional publishers differ from the economic interests of researchers and research institutions.  Obama and his appointees are much less likely than Bush and his appointees to mistake the interests of science publishers for the interests of science, and much more likely to promote the interests of science.

* The worldwide financial crisis and recession will have mixed consequences for OA, but will yield more gains than losses. 

University endowments have already lost about 25% of their value in the current fiscal year.  Because spending depends more on a running average than present value, and because many stocks with declining prices still pay dividends, there won't be an immediate, comparable drop in operating budgets.   But there will be a drop, and it will hurt library budgets and faculty salaries.  It will be harder than ever for libraries to renew all their current subscriptions, harder than ever to justify new subscriptions, harder than ever to launch or replenish funds to pay publication fees at fee-based OA journals, and harder than ever for faculty to afford out-of-pocket publication fees when they have no funder or when their funder will not pay the fees itself.

Even before the crisis, library budgets were growing more slowly than inflation and much more slowly than journal prices.  Now they will slow further or shrink.  Libraries will cancel larger percentages of their serials subscriptions than they have in decades.  That will reduce access to the TA literature, which will strengthen the case for OA among researchers, librarians, and administrators. 

At the same time, it will reduce revenues for TA publishers and strengthen the case for OA on their side as well.  It may not cause many TA journals to convert to OA, in 2009, but it will add pressure.  The more library budgets are constrained, the more it looks like a losing game to compete for shrinking library dollars, especially to society journals excluded from the nearly impervious big deals.  If TA publishers found OA journal business models unattractive a few years ago, one reason was that subscription models still looked better.  But the balance of attraction has to change as the odds of survival under a subscription model decline, roughly the way clean and renewable sources of energy become more attractive as oil becomes more expensive.  Moreover, a few years ago OA publishers were too new to be profitable, and today at least three are reporting profits, including BMC (even before the Springer acquisition), which is based in expensive London.  When contemplating their options in the face of declining subscriptions, publishers can no longer dismiss the OA alternative as untested or insufficient. 

No one is asking universities to bail out academic publishers with declining subscriptions.  And if subscriptions are declining because universities can't afford them, it would be pointless to ask.  But as the recession deepens, universities will face an opportunity similar to the one now faced by governments.  It may sound strange to call the financial crisis an opportunity for governments.  Certainly no government would mortgage its future with massive bailouts unless forced by the prospects of disaster.  But the bailout of large banks and manufacturers is an opportunity to demand transformations from these banks and manufacturers that address long-term problems. 

Universities could seize the same opportunity.  They could wake up to their power as buyers --virtually the only buyers-- of scholarly journals and demand transformations that better serve the interests of the research community.  They could move the percentage of TA publishers who allow postprint self-archiving from 63% to 100%.  They could follow the example of Harvard and 23 other universities by mandating OA for new research articles published by faculty.  They could offer to make future payments to publishers conditional upon friendlier access policies, and initiate a transition from reader-pays TA to institutionally-subsidized OA.  They could, but will they?

I hope that faculty and librarians at every university will push for these changes.  But I can't predict widespread success in 2009.  On the contrary, I predict that short-term thinking will prevail over long-term thinking at most institutions, just as it will for most governments.  Governments with an interest in deep transformations will settle for small ones in order to speed up the stimulus to the economy, satisfy the immediate clamor, take the easy majority supporting a quick fix rather than cultivate the difficult majority for a long-term solution, and (at least in the US) avoid appearing radical, anti-business, or "socialist".

We'll face the usual disentanglement problem, making it difficult to distinguish cancellations due to the growth of OA from cancellations due to the growth of journal prices faster than library budgets.  There will no shortage of opportunistic blame, overlooking the double whammy of journal prices and the financial crisis in order to point the finger at OA.  Nor will it be easy to help people look past the *problem* of declining subscriptions to the *opportunity* of declining subscriptions:  the opportunity to invest the savings from TA journal cancellations in a superior OA alternative that widens distribution, lowers costs, facilitates use and re-use, and stops betting against the internet.

Even if universities temporarily cut back on their willingness to pay publication fees at fee-based OA journals, funders willing to pay them are unlikely to retreat.  And today, far more funders are willing to pay them than universities.  Likewise, institutions providing indirect subsidies to OA journals, in the form of people, facilities, or equipment, are unlikely to withdraw their subsidies.  Finally, all journals, OA and TA, depend in part on dedicated volunteers, and are unlikely to see a decline in their willingness to volunteer unless they lose their primary jobs.

The financial crisis is not a reason for governments to retract their OA policies, or even to slow down in their adoption of new ones.  On the contrary, as I argued in my open letter to Obama and McCain last month, if the crisis leads to cuts in research budgets, then it will be more important than ever to maximize at least the *return* on the national investment in research.  For this reason, the crisis is a reason to adopt OA policies and extend existing policies.  Pushing the other way, however, is the urgency of the crisis itself, preoccupying policy-makers and reducing the priority of other policy goals.  I argued last month that one of the most sensible policies to address our overlapping crises --green research to stimulate green energy, green technology, and green jobs-- would be more effective if yoked to an OA mandate for that research.  The fact that an ambitious program of green research is already on Obama's agenda, and the fact that OA can amplify its good consequences, are grounds for hope.  But the OA connection could be lost in the wailing, gnashing of teeth, and temptations for short-term thinking.

If research budgets decline and OA policies expand --two likely outcomes of the crisis-- then it doesn't follow that their effects will neutralize one another.  On the contrary, the net growth in the corpus of OA literature could still jump significantly.  To take examples from the US:  imagine that budget cuts at the NIH caused its research output to fall a couple of points for a couple of years.  If OA policies expanded, however, the OA output of the NIH would then stand alongside the new OA output of the National Science Foundation, Environmental Protection Agency, Department of Energy, and a handful of other agencies.  Even if all these agency budgets were cut by comparable amounts, each would enlarge the overall body of OA literature, and help create a critical mass to influence other agencies elsewhere.

Note that many of these predictions are based on what could be called soft causes --events which give people a reason to act, rather than events which force their hand.  Even if the reasons look compelling from here (and of course they may not), we should notch down our confidence a few degrees because the relevant players may not see the good reasons, may not agree that they are good, or may be constrained to act some other way even if they agree that they are good.  On the other hand, making predictions based on what appears to be wise rather than what appears to be unavoidable, or treating reasons as causes, is most likely the pay off when the relevant players are informed and rational.  At least regime change in the US gives hope to this strategy. 

* Cross-over points

Now for something completely different.  Here's a series of cross-over points in the future of OA.  I'm not ready to predict when they will occur, but I want to take a stab at predicting the order in which they will occur.  (The first two have already occurred.)  Think of it this way:  OA is complicated and there's no single finish line to mark its success.  But here are a dozen intermediate finish lines, staggered across time, to serve as near-term goals and mid-course milestones.

One finish line is total success:  immediate OA to 100% of new research literature.  But I don't include that kind of goal here, partly because it's not a cross-over point (we'll probably approach 100% OA for new articles asymptotically) and partly because I want to focus here on strategy, which could always be clearer, rather than the ultimate goal, which is already clear.  For strategy, we have to change a set of defaults.  If today most new research articles are never self-archived, then flipping that over so that most new articles are self-archived will be a critical, game-changing milestone.  That's what I call a cross-over point.

Remember that these are not in order of importance, but in the order in which I predict they will occur.

(1) Green OA permission.  Today about 63% of surveyed TA publishers, and virtually all OA publishers, allow authors to deposit their peer-reviewed manuscripts in OA repositories.  We reached the cross-over point years ago when a majority of new articles carried permission for green OA. 

Even though we've crossed this threshold, and today many more surveyed publishers allow self-archiving than not, the percentage has stopped growing and started declining.  It may be that some journals are retracting their permission.  Or the decline may just be just a side-effect of the order in which SHERPA is surveying journals, and it's turning out that there are fewer green journals among newly-surveyed titles than among those surveyed earlier.  I'm not worried, however, that we'll ever cross *back* over the cross-over point.  On the contrary, the rise of funder and university policies (see #3 and #10, below) will steadily increase the number and percentage of publishers allowing green OA.

(2) OA books.  Even the youngest scholars today grew up in a world in which there were more print books in the average university library than gratis OA books online.  But that ratio flipped over about two years ago.  Today there are many more gratis OA books online than print books in the average academic library, and we're steaming toward the next cross-over point when there will be many more gratis OA books online than print books in the *world's largest* libraries, academic or not.  The two analogous cross-over points for *libre OA* books are much further off.

A few years ago, those of us who focus on OA to journal literature were sure that journal articles were lower-hanging fruit than any kind of print books, including public-domain books.  But we were wrong.  There are still good reasons to make journal literature the strategic focus of the OA movement, and we're still making strong, steady progress on that front.  But the lesson of the fast-moving book-scanning projects is that permission is a more serious problem than digitization.  The permission problem is solved for public-domain books.  Digitizing them by the millions is a titanic undertaking, but it turns out to be a smaller problem than getting millions of copyrighted articles into OA journals or OA repositories.  OA for new journal articles not only faces a well-funded lobby trying to slow it down, but serious misunderstandings in every category of stakeholders, including authors and publishers.  As the late Jim Gray used to say, "May all your problems be technical."

(3) Funder policies.  Today most publicly-funded research is not subject to OA mandates.  At the cross-over point, most of it will be.  Because public funding agencies differ enormously in size, even within the same country, we'll reach the cross-over point long before most public funding agencies adopt OA policies.  A critical set of the big ones will get us there, and we already have many of the biggest, including the NIH, six of the seven Research Councils UK, and a pilot OA mandate from the EU's FP7.

The cross-over point for privately funded research will be some distance further off than that.  While private foundations are less subject to publisher lobbying, they are also less subject to the compelling taxpayer argument for OA.  Some major private funders, like the Wellcome Trust and Howard Hughes Medical Institute, are leaders among OA-supporting organizations.  But public funders far outnumber private funders in supporting OA, and even private funders far outnumber corporate research divisions in supporting OA.

(4) Green OA deposits.  Today only a minority of new peer-reviewed manuscripts are self-archived soon --say, within a week-- after they are accepted for publication.  The cross-over point will come when most are self-archived soon after publication.  In certain fields, like particle physics, the cross-over point is behind us.  But for the total corpus of new journal literature it's still well ahead of us.  (How precise can we be in predicting the order in which different fields will reach this cross-over point --for example, medicine next and art history last?  I leave this an exercise for the reader.) 

Note that if 20% of researchers write 80% of the articles, then we can reach cross-over long before most researchers are self-archiving, either as a matter of voluntary routine or as a consequence of funder or university policies.  Of course OA outreach should target all publishing scholars.  But to reach the cross-over point as soon as possible, and gain the new momentum from shifting the default, aim for the most productive scholars first.

(5) Author understanding.  Today only a minority of publishing researchers has an accurate understanding of OA.  The cross-over point will come when that minority becomes a majority.  Author understanding is undoubtedly improving, but the slope of the curve is shallow rather than steep.  Why is it shallow, and why didn't we reach this cross-over point years ago?  We're up against myths and misinformation from publishers.  We're also up against misunderstandings from excited newcomers.  We're up against a culture of university research in which researchers who are unfamiliar with OA are preoccupied with their research, overworked, uncritically focused on the publishing incentives created by their promotion and tenure committees, and grateful to sign just about any contract a journal puts in front of them.

(6) University repositories.  Today most colleges and universities lack institutional repositories.  The cross-over point will come when most to have them, either individually or as members of a consortium.  Why will this point come after the funder-policy and green-deposit cross-over points (#3 and #4)?  First, some funders use their own repositories, the way the NIH uses PubMed Central.  Second, some faculty will prefer to deposit in disciplinary repositories, like arXiv, even when they have an institutional repository.  Third, universities move more slowly than funding agencies (see #10), even if they have the same interests as funding agencies in advancing OA.

(7) Libre gold OA.  It's hard to be sure, but it seems that most OA journals are merely gratis OA today.  While they remove price barriers, the large group I'm talking about removes no permission barriers and leaves users with nothing more than fair use or the local equivalent.  The cross-over point will come when most OA journals will be libre OA, and remove at least some permission barriers. 

On the one hand, libre OA is easier to achieve for gold OA than for green OA (see #12).  But some OA publishers are unsophisticated and use no licenses of any kind, let alone a strong, open license like CC-BY.  Some deliberately erect permission barriers, to boost ad revenue by minimizing the proliferation of copies and channeling as much traffic as they can through their own sites.  Some want to retain permission barriers for no reason at all:  they've never thought about it, they leave in place whatever copyright law imposes by default, and they think that gratis OA is all there is to OA.  Removing at least one permission barrier, or permitting even one use beyond fair use, is not a difficult hurdle, especially for a journal already committed to gratis OA.  But I'm predicting that it will be a hurdle the majority of OA journals will be slow to clear.  The new Open Access Scholarly Publishers Association will help here, since it requires libre OA for its full members.

(8) Journal backfiles.  Today most TA journals don't have OA backfiles.  At the cross-over point, most will.  The cost can be large, and we won't see significant progress until some of the wealthy book-scanning projects move more aggressively into journal-scanning.  (Google has a slow-moving journal-scanning project and the OCA is still largely limiting itself to public-domain literature.)  Most OA activists aren't pushing hard either, making prospective OA a higher priority than retroactive OA.  They're right to make prospective OA a higher priority, but at some point we can lift our gaze a bit and pick higher-hanging fruit along with the low-hanging fruit.

Some journals with budgets or benefactors are digitizing their backfiles now, some for TA and some for OA.  For those with digital online TA backfiles, the revenue will be small and will not amortize the investment.  Some will stick to TA, to wring what they can from their investment.  But others will calculate the benefits of OA, in increased visibility and citations, even after a moving wall, will outweigh the trickle of revenue. 

(9) Author addenda.  Today only a small number of universities have adopted author addenda, and only a smaller number require their use.  That makes it easy for publishers to reject the addenda or to refuse to publish work by faculty for whom addenda are not optional.  The line of new articles submitted without addenda is sloping gently downward, and the line of new articles submitted with addenda is sloping gently upward.  When the two lines cross, or even get close, the balance of power will shift and the average journal will have more to lose than to gain by rejecting author addenda.

This cross-over point could be delayed forever if other kinds of progress lead enough TA journals to liberalize their copyright transfer agreements.  For example, if all TA journals allowed immediate postprint archiving, most addenda would unnecessary. 

(10) University policies.  Today most university research is not subject to university-level OA mandates.  At the cross-over point most of it will be.  Although 24 universities and four departments already have OA mandates, a total which is very close to the number of funding agencies with OA mandates (30), I'm predicting that the cross-over point for universities will come much later than the cross-over point for funders (see #3).  One reason is that there are many more universities than funding agencies.  Another is that universities move more slowly than funding agencies, even when their interests are clear (see #6).  However, a related reason is that universities are both producers and consumers of research and internalize some of the conflicts played out in the larger OA debate. 

The Harvard OA mandate is inspiring other institutions to follow suit and I predict a significant number of Harvard-inspired policies in 2009 alone, which will in turn inspire others.  In this sense, the pace of university OA policies is picking up.  Moreover, as with funders, some universities are much larger or more productive than others; and if the largest or most productive ones move first, then we'll reach the cross-over point long before most institutions have OA policies.  But I still believe the cross-over point is not imminent. 

Harvard's two OA mandates were surprisingly unanimous, and the Stanford OA mandate was adopted unanimously the same afternoon it was proposed.  These events jolt my pessimism.  So why am I not more optimistic that this cross-over point will come soon?  I've thought about that and don't have much to go on except the fact that, in four decades in higher education, I've been surprised by university fleetness much less often than I've been dismayed by university sluggishness.  But I acknowledge that the OA exceptions are recent and may represent an emerging new pattern.  I'm aware that the past may be no guide if the conditions for OA are ripening, and if slow-growing readiness for change is about to trigger a rapid and widespread phase transition.  But for now I have to be content to work for it rather than predict it.

(11)  OA journals.  Today most peer-reviewed journals are TA.  At the cross-over point most will be OA.  Two kinds of events move us toward this point:  the launch of new OA journals and the conversion of TA journals to OA.  And two kinds of event move us the other way:  the launch of new TA journals and the conversion of OA journals to TA.  All four events occur, although the first three occur much more often than the fourth.  On the whole we're moving toward the cross-over point, rather than away from it, but the movement is slow. 

Apart from the slow movement, one reason to expect this cross-over point late in the game is that some TA journals can survive in an OA world.  The most prestigious TA journals are in this category.  But even second-rate journals can be protected, for a time, by bundling and inertia.  If we ever make significant progress toward this cross-over point, then library budgets will not only be freed up to pay for OA alternatives, but also to pay for TA hold-outs, prolonging the period of coexistence for some of the survivors.  Converting the last 10% of TA journals will be much harder than converting the next 10%.

I've always admitted that the green OA mandates could trigger journal cancellations and pressure journals to convert to OA.  When publishers present that as a disaster scenario, I stick to arguing that (a) it hasn't happened yet, (b) there's no evidence that it's happening, (c) there's evidence that it won't happen, (d) and if it did happen, it would not be a disaster.  So for completeness I should add that the rise of OA mandates, which I predict, could accelerate movement toward the cross-over point when most journals are OA.  But the experience in physics shows at the very least that the causation here is complicated and that TA journals can survive for a long time, perhaps indefinitely, in the face of high-volume OA archiving.  Hence, even with the proliferation of OA mandates, don't expect this cross-over point any time soon.

(12) Libre green OA.  Today most green OA is gratis OA.  But at least some is libre OA, for example, a subset of PubMed Central.  There may be similar subsets in many other repositories.  As long as most green OA depends on the permission of TA publishers, the majority of it will remain gratis.  But there are reasons to predict a slow growth in the percentage of libre green OA, and even an eventual cross-over point.  Today the members of the UKPMC Funders Group require gratis OA for articles based on research they fund, and libre OA when they pay the costs of publication.  That libre increment will continue to grow.  One interesting fact is that the UKPMC Funders Group "libre OA mandate" came out in October 2007, 10 months after its "gratis OA mandate" in January 2007.  That's a sign that even funders with gratis OA mandates are willing escalate their OA policies in the right circumstances. 

Here's one possible future escalation:  authors must not only retain the right to authorize gratis OA, but must retain the right to authorize libre OA (say, with a CC-BY license).  When they approach publishers, authors will explain the terms they require.  As it is today, this will a business proposition which publishers may accept or reject.  But when a publisher rejects the terms, the authors must look for another publisher.  Today half a dozen funders and a couple of universities take this approach with gratis OA, and so far none takes it with libre OA.  But to extend the practice to libre OA would be perfectly lawful and very beneficial.  But I'm not predicting it any time soon.

If it's tried too soon, it will just hurt researchers by making it harder for them to publish their work.  But after a point, when other OA initiatives have had their effect, and more TA publishers have adapted to an OA world, authors will encounter fewer flat refusals and the OA mandates themselves will trigger more publisher accommodation than publisher resistance.  Enlightened funders will be watching for that moment and testing the waters.  Because the odds of success soar as more funders adopt similar policies, or because followers take fewer risks than leaders, libre green OA mandates may spread quickly once they are adopted.


Here are three general points I'd put in footnotes, if I had footnotes:

* It's just a mathematical convenience to set these cross-over points at the 50% watershed, when a minority becomes a majority and vice versa.  For some of these trends, the more important default may change when a 20% minority becomes a 30% minority.

* We're moving toward all 12 cross-over points.  We shouldn't mistake the remoteness of a given point for a sign of immovable obstacles, let alone backward movement. 

* I realize that I've used the terms "gratis OA" and "libre OA" often in this article, and that some readers may not know exactly what I mean.  Here's some background.


* Postscript.

Predictions for 2008

Predictions for 2007

Predictions for 2006

Predictions for 2005

Predictions for 2004