It’s now been 24 years since Stevan Harnad sparked the open access movement by suggesting in his “subversive proposal” in 1994 that scholars ought to just publish their scholarly articles on the internet:
If every esoteric author in the world this very day established a globally accessible local ftp archive for every piece of esoteric writing he did from this day forward, the long-heralded transition from paper publication to purely electronic publication (of esoteric research) would follow suit almost immediately.
Since then, we have been waiting on the behavior of scholars to change, such that all our works indeed become accessible. This is what has become known as the “culture shift” in academia, without which no actual change in our practice can happen. However, no such change can be seen, not even after all these years. Instead, open access mandates and other policies have been developed to force scholars to perform certain behaviors they wouldn’t otherwise do. Even in fields where such deposition of articles has become common, the authors still adhere to toll-access publishing not for reading or scholarly communication, but for career advancement – an obscenely expensive and perverse outsourcing practice.
Why does such behavioral change take so long? Many of Stevan’s colleagues at the time have since retired and a large section of the scholarly workforce has been replaced with a new generation, one would think – if anything – more net-affine than the previous one?
In this post, I will try to make the argument that our mistake was to expect behavior to change when the reasons for the behavior have not changed. As a behavioral neuroscientist, I have learned that, all else being equal and depending on time-scales, among the best predictors of future behavior is past behavior. Thus, if we analyze why scholars behave the way they do with regards to open scholarship, we may be more likely to affect that behavior.
Why isn’t everybody using preprint servers? What keeps people from posting their data and code on any of the proliferating repositories? What is the reason, funders feel they need to use mandates to get scholars to comply with open science ideals? Why are the non-activist, regular scholars either lethargic or outright hostile?
In the last decade, it seemed as if the answer to this question was “because of the reward system!” or “because incentives are missing!”. As if scholars only ever do anything if they are rewarded or incentivized for it. I think the answer lies elsewhere. It can be articulated as two main reasons:
1) They do not care and hence do not know: scholars care about their scholarship and they shouldn’t have to care about such questions. These questions are exactly what infrastructure should be taking care of, not scholars.
2) They have good reasons to close their scholarship: lack of time, fear of competition, privacy concerns, etc.
I’m simplifying, of course, but would nevertheless tend to argue that together, 1+2 explain most of scholars’ behavior wrt to open scholarship. Those who care, know and do not have good reasons to be closed are people like open scholarship activists, e.g., yours truly. The other 99.5% are the ones who resist “culture change” for either reason 1) or reason 2) or both. I’m rather skeptical anything can make either 1) or 2) go away any time soon, let alone both.
Hence, rather than fighting 1+2, as we have been for two decades now, I suggest to use them in our favor.
A recent poll in our biology/medicine department exemplifies how this might work: when polled which software the department members are currently using to prepare images (microscope images, gel pictures and such) for publication, the majority answered “PowerPoint”. Now, I assume that most everybody on here would understand that PowerPoint is not the, ahem, ideal, most professional software to use for these kinds of work. 🙂 On the contrary, submission in PTTX format is explicitly discouraged by most publishers. This means that the majority of people in our department use a tool that is not the most professional for the task at hand and the format of which is discouraged for submission. What funder mandates could be in place to encourage such odd behavior? Which tenure committee rewards compliant over superior tool use? Where is the academic incentive system that pushes scholars to choose PowerPoint over the better-suited alternatives?
Obviously, there are no mandates or tenure committees incentivizing the use of such suboptimal tools. Scholars are doing this entirely on their own accord. Why on earth would educated people do something like that? The answer is straightforward: for the same reasons 1) and 2) above! Most faculty don’t care and hence don’t know: they use what comes on their computers, pre-installed by the university. Or they have a good reason to use this tool: it’s good enough for them and they don’t have the money for Photoshop or would rather spend the money for experiments than software. Or they find ImageJ too hard to use as they are already familiar with the ubiquitous PowerPoint and can’t be bothered to switch. Or installing new software is just too much of a hassle. Et cetera.
With this example in mind, how do you get scholars to choose open publishing alternatives over legacy publishers? How do you get them to use open evaluation procedures over impact factor? How do you get them to save their data to a repository, rather than on their thumb drive? You provide them, automatically, free of charge and ready to use, with the tools you want them to employ, with the default settings (i.e, open) you prefer. The large majority who doesn’t care and hence doesn’t know will just use what’s convenient, quick and free, so they can focus on what matters most to them: their scholarship. Those who have good reasons to make their work closed will balance these against the potential negative consequences (e.g. more time and effort, potential suspicions if everything else is open, etc.) and be able to make their work as closed as it needs to be for them. Of course, ideally, such tools come with their own reasons why one would want to use them, such as increased efficiency over legacy tools or new, more and better functionalities. Conversely, equally obviously, you stop providing scholars with anything you don’t want them to use, such as subscription journals. Or impact factors. Or typewriters.
Since subscriptions globally run at about US$10bn every year, and the technology for a scholarly commons can be had off the shelf, the kind of modern infrastructure that would get scholars to change their behavior only needs to be bought with the funds saved by subscription cancelations. As such an infrastructure would provide scholars with a superior toolset, it would also add ‘efficiency’ and ‘functionality’ to reason 2) as to why scholars are using this new, open infrastructure.
Good point about making “good” infrastructure the default and the path of least resistance. But that won’t happen just by building such an infrastructure and making it available. Look at the experience with free/libre open source software, which is often or usually superior to its proprietary counterparts.
Following your example: the department could replace all computers, preinstall some GNU/Linux distribution with GIMP and LibreOffice, add other suitable software. People with extraordinary needs would be allowed to ask for a dual boot machine, by following a procedure which is hard enough to avoid mass-adoption but easy enough to avoid complaints by vocal users. Would most people stop using PowerPoint, in favour at least of LibreOffice Presentation/Draw? How many would switch to a different software, considered adequate for the task and hence preinstalled?
You would usually need to make the switch in dialogue with the users, provide training and support. Even new people will come to you diseducated by previous experience with proprietary software, unless they’ve been properly educated before (e.g. since primary school computer class).
Yes, excellent point! We constantly debate which of the tools our computers come with should we include in training seminars and even courses for students. We have decided that email clients and office suites are left up to each user, but we offer courses in the more specialized tools, such as, e.g., image manipulation. In fact, this was the backdrop of this poll! We needed such information to design our course on image preparation.
In other words: you are spot on and we already do this for any piece of software or tool that is general enough for most people in our department to use them at some point, but specialized enough that you are unlikely to have learned it in school or in other, more general settings.
Thus, with regards to training, we just need to keep doing what we are already doing for all other digital tools. The graduated user can decide what they want to learn by doing and for what they should attend one of our courses/seminars. The undergraduate user gets a course, just as now.
In your example, I think a majority would just use what the university offers. After all, the university offers a very limited set of default tools already (don’t get me started on Novell and GroupWise!) 🙂 If people really cared about what they use, there would have already been a revolt 🙂 Email is actually a good example: nobody gets training on how to use the email client the university offers. There is an information page from the computing center with links to instructions for other clients, that’s it. Hence, open tools would be treated exactly as the tools we offer now.
Great to explore alternative reasons, but they can still be considered proxy behaviours for the ultimate currency: prestige. It’s all about prestige and shining little gold stars. The (current) reward system is just a tool to operate the abstract prestige; that’s why you need branding, prizes and grants as measures of V.I.P. access (instead of more direct measurements of academic activity or output or impact or connectivity).
a) Why researchers do not use repositories? (preprints or green open access) Because there is no prestige (or measurement of it) associated. Let all funders and universities start counting number of preprints instead or post-peer-review-journals and bioRxiv will collapse under the flow.
b) Why we/they “do not care”? (your point 1) Because we/they DON’T HAVE TO. In your first few semesters of your scientific career you will be taught with (excellent) papers in Nature/Science or the equivalent “prestigious journals”. The importance of Publishing Big will be taught persistently (not least by teaching assistants struggling to put themselves on track). You’ll be lucky if preprints or open access are ever mentioned.
c) Why we/they will not be open? (your point 2) Because MY prestige (and everything that comes with it) depend on MY data/results. You do your own, and then we will be measured against. Let all data, results and papers be communicated anonymously by law and the close/open debate disappears.
d) Why we/they will resist change? Because WE/THEY ARE ALREADY IN. And we/they got here with THESE RULES that is in our best interest to preserve.
Of course I am also simplifying. Maybe it’s a too broad, or harsh, caricature. But the point is that the behaviours that you single out as cause I believe are derived from an underlying, hierarchically more important, causes. There is (still?) no reward in being open; maybe that is also the reason why there is always at least some moral grounds in arguments for open access activism [or at least there was until open access publishing was (also) co-opted to make big (bigger?) money…]
Very good comment and also very accurate. In part, this is why I wrote: “Conversely, equally obviously, you stop providing scholars with anything you don’t want them to use, such as subscription journals. Or impact factors. Or typewriters.” The reasons you recount, I would subsume under 2), i.e., “good reasons to be closed”. If we no longer subsidize the sources of prestige (such as journals with subscriptions or high APCs), how would these sources of prestige remain profitable?
You are more qualified than most to comment on my blog posts, peer-reviewed publications and preprints, such as this one. Nutrient-dependent pheromone-controlled ecological adaptations: from atoms to ecosystems https://figshare.com/articles/Nutrient_dependent_pheromone_controlled_ecological_adaptations_from_atoms_to_ecosystems/994281
I was among the first to take Harnad’s advice and established Pheromones.com in 1995. I have since linked quantized energy-dependent changes in the microRNA/messenger RNA balance to all biophysically constrained biodiversity via the physiology of pheromone-controlled reproduction. Do not expect scientific progress to be made via preprints or blog posts if you ignore them.
is “science” not an incentive to disseminate information openly? do scientists need more? one wonders about these question as scientists are one of the bitter critics of politicians and business folks when it comes to ethics and doing the right thing. i think the solution is simple: either scientists not interested in open dissemination of knowledge declare that they are not interested in science and are there for personal gain (and there is nothing wrong in aspiring to get personal benefit) or do what is right. it can’t be “you can’t have your cake and eat it too”. i don’t think there is any other way.