Monday, December 03, 2012

The OA Interviews: Harvard’s Stuart Shieber



Stuart Shieber is the Welch Professor of Computer Science at Harvard University, Faculty Co-Director of the Berkman Center for Internet and Society, Director of Harvard’s Office for Scholarly Communication (OSC),  and chief architect of the Harvard Open Access (OA) Policy — a 2008 initiative that has seen Harvard become a major force in the OA movement.

When in 1989 Stuart Shieber became a Harvard faculty member he was, for reasons he never fully understood, appointed to a series of library committees. Whatever the reason for his appointment, it was to prove an educational experience: As he sat through the various committee meetings, Shieber began to see the world through the lens of the library, a perspective that led him to the inevitable conclusion that there was something amiss in the world of scholarly communication.
Stuart Shieber

As he puts it, “[I]t became increasingly clear to me that some of the problems that libraries faced in dealing with providing access to the scholarly literature were not library problems per se, but rather, problems in how the scholarly communication systems are set up.”

This is worth noting because when researchers face difficulties accessing scholarly journals they tend to assume that something has gone awry in the library, not that there is a fundamental flaw in the way research is communicated.

It was only in sitting through all those library committee meetings that Shieber came to realise the research community had a serious problem on its hands, a problem moreover that could only be expected to get worse unless action was taken. And it was clear to Shieber that researchers themselves would need to play their part in resolving the problem.

Stated simply, the problem is this: When researchers publish their papers, they routinely sign over the commercial exploitation rights in them to the publisher. The publisher then packages a bunch of papers together and sells them back to the research community in the form of a journal subscription. While publishers undoubtedly add some value to the end product, researchers do most of the work — not just in authoring the papers in the first place, but also in peer reviewing their colleagues’ papers (without charge). Yet, as any librarian will tell you, subscription charges are inexcusably high, and getting higher each year.

In short, publishers are overcharging for scholarly journals. And since it is they who pay the bills, it was librarians who first sounded the alarm. However, since the costs do not come from their budgets, and journals are made available in institutions on a free-at-the-point-of-use basis, most researchers have been unaware of the seriousness of the problem. For their part, publishers have consistently denied that they are overcharging.

Why are journals so expensive? They are expensive for a number of reasons, but mainly because there is a disconnect in the scholarly journal market: that is, the people who pay the journal subscriptions are not the people who use the journals. As we shall see, this means that there is no effective market mechanism to control prices.

And as the number of papers published continues to grow, and libraries face ever greater pressure on their budgets, so the struggle to provide faculty with access to all the papers they need has become ever more serious. This phenomenon is now generally referred to as the “serials crisis”.

As I hope will become apparent, it helps to see the serials crisis as a double-headed problem. As libraries are forced to cancel more and more journals each year, researchers face a growing accessibility problem. However, this accessibility problem is merely a symptom of the deeper problem — what one might call the affordability problem. The key challenge, of course, is to find a solution.

Open Access


Over the years various solutions to the serials crisis have been proposed. However, the one that has gained the greatest mindshare is Open Access (OA). And it is no surprise that librarians played a key role in the development of the OA movement — not least by co-founding the Scholarly Publishing and Academic Resources Coalition (SPARC) in 1998, and constantly promoting the merits of OA.

Essentially, advocates for OA argue that all published research can and should be made freely available on the Web, either by means of green OA, in which researchers continue to publish in subscription journals but then self-archive their papers in their institutional repository (usually after an embargo period so that publishers can recover their costs), or by means of gold OA, in which researchers (or more usually their funders) pay publishers an article-processing charge (APC) to ensure that their paper is made freely available on the Web at the time of publication. The latter can be achieved either by publishing in an OA journal (the entire contents of which are made freely available), or in a hybrid journal (a subscription journal in which individual papers can be made OA if the author pays an APC).

Concluding that Open Access offered a viable solution to the double-headed problem facing the research community, Shieber began to advocate for OA at Harvard ...

####

If you wish to read the interview with Stuart Shieber, please click on the link below.

I am publishing the interview under a Creative Commons licence, so you are free to copy and distribute it as you wish, so long as you credit me as the author, do not alter or transform the text, and do not use it for any commercial purpose. 

To read the interview (as a PDF file) click HERE.

7 comments:

Anonymous said...

COMMENT FROM SANDY THATCHER: PART 1

This is a valuable-but also very long!-piece, offering a broad overview of where the OA movement stands today, through the lens of the pioneering Harvard policy and its implementation. Thanks to Richard for conducting the interview and providing the context.

I offer a few random comments.

The reporting of the Harvard decisions as "unanimous" may be misleading. As I recall at the time the initial FAS decision was announced, there was questioning what "unanimous" actually meant, and it was suggested that it meant unanimity only among the faculty who actually attended the meeting at which the proposed new policy was voted on, not all the faculty in FAS. That could well mean that it was a decision made by a minority of the faculty since how many faculty actually bother to attend faculty senate meetings at any institution?

The article suggests that the DASH repository operates on a "libre" OA model. But a look at the "Terms of Use" shows that they are very far from mirroring the Budapest definition of OA in not allowing, for instance, any commercial reuses or even any derivative works like translations. Perhaps Shieber understands that Harvard faculty would be reluctant to give up all opportunities for financial enrichment for the commercial reuse of their articles in anthologies (which can be in the tens of thousands of dollars) and might also be leery of giving up any control whatsoever over the quality of translations of their work?

The idea of controlling cost increases of APCs through capping the amount any individual faculty member may draw from HOPE annually is interesting, but the same result could be achieved more directly by doing what the Harvard Medical School started doing years ago, viz., capping the number of articles allowed for submission for tenure and promotion decisions.

That is also likely a more effective way of keeping the absolute costs of publishing down, as one of the dangers of OA publishing is that the cost-of-entry for new players is so small compared with TA publishing and this proliferation of new OA publishers only encourages the publication of more research, not necessarily the best research.

Why, one wonders, if Shieber is such an evangelist for OA, is he not working to have Harvard University Press turned into a service agency for publishing Harvard faculty monographs disseminated under an OA model? Why isolate OA to just one stream of scholarly communication? It's not as though systemic dysfunctionalities do not also exist for monograph publishing, after all.

This service agency model for university presses has been proposed by reform-minded people like Kathleen Fitzpatrick, in her book titled Planned Obsolescence (NYU, 2011), where she also discusses alternative models for peer review.

A service-oriented HUP might also assist DASH in overcoming one major problem with its current model, viz., that much of its Green OA content is not copyedited and hence bears all the flaws of unedited writing, which can hardly do good for the reputation of Harvard faculty. (Along with a few volunteer copyeditors from university presses, I conducted an analysis of the DASH repository from this viewpoint, which was published in a special issue of Against the Grain in April 2011.)

To be continued …

Anonymous said...

COMMENT FROM SANDY THATCHER: PART 2

... continued

Shieber is overly optimistic, in my view, about how market competition can lead to lowering costs for OA publishing compared with TA (subscription). For one thing, there are likely to be multiple sources of OA funding and thus no easy way to exert control over the cost of APCs across different sectors of academe.

Shieber expresses hope that the foundations will eventually try to exert some control over APCs in the biomedical fields where they operate; I rather doubt they will.

He pays far too little attention to the role that the "prestige" factor will continue to play in enabling publishers to charge whatever they think they can get away with. He recognizes that it may work to impede the damage that Beall's "predatory" OA publishers can do, but doesn't accord it enough importance, in my view, as a force for keeping APCs high and increasing at above-inflationary rates.

At the same time, he does not indicate whether COPE members try to exercise any control over where the authors that apply for APC funds actually publish. Do COPE members read Beall's postings and rule out providing APC funds for authors wanting to publish in "predatory" journals?

As for education, it appears that the road will be very long indeed. This was what InsideHigherEd reported today about a recent survey by the American Historical Association:

"The survey also found that senior faculty members are unlikely to believe that their institutions highly value digital journal articles, even with the question specifying that these were peer-reviewed online articles. Compared to the approximately 70 percent of history professors in the survey who said that print articles were highly valued, only about 10 percent said the same for digital articles."

I think it is very likely to happen that not only will OA publishing gain traction in different fields at very different rates, but funding to pay APCs will vary quite significantly from one area to another, with the humanities, as usual, disadvantaged relative to the sciences and even social sciences.

If this happens, then the same inequities that exist in the TA publishing world will continue to exist in the OA world, except that the burden will fall more directly onto the scholars themselves, who in the humanities are not likely to have available nearly the range and depth of sources available to tap for APCs as their colleagues in the sciences will.

On the other hand, as the AHA survey suggests, maybe OA publishing will take a lot longer to establish its viability as a real option in the humanities than it has in the STM fields.

Sandy Thatcher

Richard Poynder said...

STUART SHIEBER'S RESPONSE TO SANDY THATCHER PART ONE

@Sandy Thatcher: I’d like to clarify a few points that you raise in your response to the interview. I apologize if my silence on these issues was perceived as constituting sins of omission, but I answered the questions asked and not ones that were unasked.

The reporting of the Harvard decisions as “unanimous” may be misleading. As I recall at the time the initial FAS decision was announced, there was questioning what “unanimous” actually meant, and it was suggested that it meant unanimity only among the faculty who actually attended the meeting at which the proposed new policy was voted on, not all the faculty in FAS. That could well mean that it was a decision made by a minority of the faculty since how many faculty actually bother to attend faculty senate meetings at any institution?

You are of course right that it was the FAS vote that was unanimous. In fact, all of the votes at the now eight schools at Harvard where the policy has been passed (the Harvard School of Public Health passed the policy only after the interview was completed) were unanimous or nearly so. It is impossible to know the level of support of the faculty members who didn’t attend the votes. I don’t think we’ve been misleading on this point.

However, it may be useful to note that each vote was preceded by extensive discussions at numerous other faculty meetings and open forums. Were there others with strong feelings against the matter, they were welcome to attend the votes. And as I say in the interview “I have never heard a faculty member express any principled objection to the underlying premise of the OA policy” and I’ve talked to a lot of faculty. I’m sure that there are some faculty members who don’t like the policy for one reason or another. But I don’t think they constitute a disenfranchised groundswell.

The article suggests that the DASH repository operates on a “libre” OA model. But a look at the “Terms of Use” shows that they are very far from mirroring the Budapest definition of OA in not allowing, for instance, any commercial reuses or even any derivative works like translations. Perhaps Shieber understands that Harvard faculty would be reluctant to give up all opportunities for financial enrichment for the commercial reuse of their articles in anthologies (which can be in the tens of thousands of dollars) and might also be leery of giving up any control whatsoever over the quality of translations of their work?

DASH operates in distributing OA-policy articles in a libre mode in the sense that the rights provided under the terms of use go well beyond mere gratis access. The BOAI–10 recommendations make clear that there are a variety of libre approaches that go beyond gratis. (For instance, “we recognize that gratis access is better than priced access, libre access is better than gratis access, and libre under CC-BY or the equivalent is better than libre under more restrictive open licenses. We should achieve what we can when we can.”) But you are right that they don’t match the BOAI–10 libre ideal of a CC-BY license. That’s first and foremost because the OA policy doesn’t allow CC-BY licensing. In particular, the noncommercial exclusion in the terms of use follows from the fact that the OA policy itself does not reserve commercial rights, and not because “Harvard faculty would be reluctant to give up all opportunities for financial enrichment for the commercial reuse of their articles in anthologies.” Harvard faculty don’t themselves retain commercial rights by virtue of the OA policy. They’ll still need to get permission from the publishers.

Being paid tens of thousands of dollars for reuse of articles in anthologies — that made me laugh.

continues …

Richard Poynder said...

STUART SHIEBER’S RESPONSE TO SANDY THATCHER PART TWO

The idea of controlling cost increases of APCs through capping the amount any individual faculty member may draw from HOPE annually is interesting, but the same result could be achieved more directly by doing what the Harvard Medical School started doing years ago, viz., capping the number of articles allowed for submission for tenure and promotion decisions.

Actually, that would achieve a different result. It’s hard to see how capping the number of articles per author that the HOPE fund will underwrite provides any incentive for the author to attend to cost for a particular article. Nonetheless, I’ve long been an advocate for limiting the number of articles that can be listed on promotion dossiers and grant applications for other reasons.

That is also likely a more effective way of keeping the absolute costs of publishing down, as one of the dangers of OA publishing is that the cost-of-entry for new players is so small compared with TA publishing and this proliferation of new OA publishers only encourages the publication of more research, not necessarily the best research.

The point is not to publish less research, especially since we have only a limited ability to predict in advance what will turn out to be the most important research. Rather, it is to make sure that research is appropriately vetted and that the quality signals are as accurate and transparent as possible.

Why, one wonders, if Shieber is such an evangelist for OA, is he not working to have Harvard University Press turned into a service agency for publishing Harvard faculty monographs disseminated under an OA model? Why isolate OA to just one stream of scholarly communication? It’s not as though systemic dysfunctionalities do not also exist for monograph publishing, after all.

This service agency model for university presses has been proposed by reform-minded people like Kathleen Fitzpatrick, in her book titled Planned Obsolescence (NYU, 2011), where she also discusses alternative models for peer review.


First of all, are you really saying I’m not “working” enough for open access? That unless I’m also working on the monographs market problems and on transforming peer review, I’m a piker? That it’s my responsibility to have addressed every single issue by now? Really? That makes me sad. Keep in mind that I still have my day job teaching, doing research, advising students, along with all the other work we do as academics.

Second, why the presumption that I’m not working on various other issues as well, just because you haven’t seen the results? Hint: It took us years just to get the first OA policy passed at Harvard.

continues …

Richard Poynder said...

STUART SHIEBER’S RESPONSE TO SANDY THATCHER PART THREE

A service-oriented HUP might also assist DASH in overcoming one major problem with its current model, viz., that much of its Green OA content is not copyedited and hence bears all the flaws of unedited writing, which can hardly do good for the reputation of Harvard faculty. (Along with a few volunteer copyeditors from university presses, I conducted an analysis of the DASH repository from this viewpoint, which was published in a special issue of Against the Grain in April 2011.)

Let’s look at this “major problem”. In your piece, you concluded that “By and large, the copyediting did not result in any major improvements of the manuscripts as they appear at the DASH site.” (I discuss the issue at greater length in a post at The Occasional Pamphlet.)

Yes, there are some divergences between the authors’ final manuscripts in DASH and the publishers’ final versions. Yes, it would be better if the latter were OA. That’s why I said in the interview that “DASH and repositories like it at other universities are not sufficient in the long term to provide all of the functionality that journals provide…. So while repositories can serve to shore up access in the short run, in the longer term some other mechanism will be needed to provide these services.” And that’s why I’ve been working on underwriting of fees for open-access journals (as discussed in the COPE and HOPE sections of the interview).

All this discussion about DASH and HUP is besides the point anyway. DASH is a repository for journal articles. HUP doesn’t publish any journals.

Shieber is overly optimistic, in my view, about how market competition can lead to lowering costs for OA publishing compared with TA (subscription). For one thing, there are likely to be multiple sources of OA funding and thus no easy way to exert control over the cost of APCs across different sectors of academe.

Shieber expresses hope that the foundations will eventually try to exert some control over APCs in the biomedical fields where they operate; I rather doubt they will.

He pays far too little attention to the role that the “prestige” factor will continue to play in enabling publishers to charge whatever they think they can get away with. He recognizes that it may work to impede the damage that Beall’s “predatory” OA publishers can do, but doesn’t accord it enough importance, in my view, as a force for keeping APCs high and increasing at above-inflationary rates.


Publishers, like all businesses, charge “whatever they think they can get away with”. That’s called profit maximization, and there’s nothing wrong with that in a well-functioning market. To the extent that the OA publishing market functions well, I’m not worried about excessive profit-taking, and all the signs are that the OA journal market is much better behaved than the subscription journal market.

Perhaps the economists who have looked at the question are all wrong and you are right. Maybe APCs won’t hyperinflate less than subscription fees. Maybe OA won’t save any money. But at least everyone will be able to read the articles.

continues …

Richard Poynder said...

STUART SHIEBER’S RESPONSE TO SANDY THATCHER PART FOUR

At the same time, he does not indicate whether COPE members try to exercise any control over where the authors that apply for APC funds actually publish. Do COPE members read Beall’s postings and rule out providing APC funds for authors wanting to publish in “predatory” journals?

I didn’t indicate because I wasn’t asked. But it’s not a secret. Each COPE member establishes its own policies, which are generally openly available. Most have established some standards for qualifying venues. Harvard’s criteria are available at the HOPE website. In summary, journals must be fully open access, listed in DOAJ, members of OASPA or adhere to its Code of Conduct, have publicly available a standard article fee schedule, and have a policy to substantially waive fees in case of economic hardship. These criteria by themselves would eliminate all of the predatory publishers that Beall lists, were we (counterfactually) to ever receive a request for reimbursement for one of them. Beyond that, we trust our faculty to know what journals are appropriate to publish in. You can judge for yourself if that trust is warranted, since we list HOPE-funded articles at the web site.

As for education, it appears that the road will be very long indeed. This was what InsideHigherEd reported today about a recent survey by the American Historical Association:

“The survey also found that senior faculty members are unlikely to believe that their institutions highly value digital journal articles, even with the question specifying that these were peer-reviewed online articles. Compared to the approximately 70 percent of history professors in the survey who said that print articles were highly valued, only about 10 percent said the same for digital articles.”


I expect that you are right. I dare say no one understands the length of that road more than I.

I think it is very likely to happen that not only will OA publishing gain traction in different fields at very different rates, but funding to pay APCs will vary quite significantly from one area to another, with the humanities, as usual, disadvantaged relative to the sciences and even social sciences.

This is undoubtedly true as well.

If this happens, then the same inequities that exist in the TA publishing world will continue to exist in the OA world, except that the burden will fall more directly onto the scholars themselves, who in the humanities are not likely to have available nearly the range and depth of sources available to tap for APCs as their colleagues in the sciences will.

On the other hand, as the AHA survey suggests, maybe OA publishing will take a lot longer to establish its viability as a real option in the humanities than it has in the STM fields.


One can almost guarantee that not all inequities will be eliminated by a transition to open access. The inequities are not getting better while STM journals hyperinflate either.

In examining your comments, I see a theme. Many of your points, while valid, are of the form that whatever we’ve done or proposed is not perfect. The OA policy votes may have been unanimous, but we didn’t poll everyone. The DASH terms of use go beyond gratis, but not far enough. Capping APCs may be good but we didn’t also cap the number of articles. The Harvard policies address problems in the journal market, but not those in the monograph market as well. DASH provides broad access but not to the ideal version. Open access may broaden availability, but it may not save money, or solve field inequities, or cure halitosis. And so forth.

I have to say that I find this anti-Voltairean line of argument dispiriting. Perhaps, since we can’t immediately achieve the perfect in our quest for the good and better, we should just pack it in

… ends

Richard Poynder said...

Stuart Shieber:
Each COPE member establishes its own policies, which are generally openly available. Most have established some standards for qualifying venues. Harvard’s criteria are available at the HOPE website.

I am sure most would agree that it is essential for those managing COPE funds to consider a journal's quality before paying for a researcher to publish in it, and in this regard Harvard provides a good model for others to follow. However, as Stuart seems to imply, it is not clear that all COPE members do currently assess for quality before paying APCs.

The eligibility criteria published by the University of Michigan, for instance, appear not to include a quality test.

We could also note that at least one of the publishers listed as having benefited from the University of Michigan’s COPE fund is included on Jeffrey Beall’s list of predatory publishers — although Beall's list is controversial, and critics maintain that publishers have been inappropriately added to it.