The difficulties in opening science: Q&A with Michael Nielsen

Posted by:

In a timely and incisive talk at TEDxWaterloo, Michael Nielsen made the case for open science — the idea that research data and results should be freely available to the public, and that scientists should collaborate more freely with each other and with the public.

Since the talk, his book, Reinventing Discovery, has been released, and even more recently the US Congress has introduced controversial legislation to prohibit funding agencies from requiring open access. TED’s Ben Lillie called Nielsen to talk about the politics and difficulties of open science.

Your talk got a significant pick-up from the science community and other people starting to discuss the idea of open science. Have you seen people moving more toward this notion of doing open science, or at least that – I mean, your goal here is to spark a conversation and get that into the things that people are talking about. Is it happening?

Yeah, I think it’s definitely on the up-tick. The phrase “open science” was, I think, pretty uncommon even three or four years ago. It’s a very old phrase. Certainly, it’s been in use for many decades, possibly even centuries. But it’s not something that people talked about a lot until the last two years.

I guess I’ve done 30 talks in seven countries in the last couple months, and particularly in the United Kingdom it really seemed like there was a huge amount of awareness. Organizations like the Royal Society are doing investigations of open science. The Wellcome Trust, a very large funding agency, has certainly done a huge amount for it over many years. In the United States, there also seems to be increasing interest. There’s more articles in the mainstream media than there were – well, there were very few articles even a few years ago. But now it seems like there’s quite a bit of coverage, and in more specialist outlets – Wired, and places like that.

You have this notion that we need to share our data, we need to share our ideas, and that will help to accelerate progress. And one of the main barriers to this, of course, is that we simply don’t have a culture of rewards for doing those activities. In the book, you give the example of scientists editing Wikipedia, which has no upside and takes a lot of their time. Where do you start with implementing a reward system where this is actually something that professional scientists would be driven to do?

There are two great places to start. One is, if grant agencies start to say, “Look, please tell us about your non-traditional contributions. If you’ve uploaded materials to YouTube, we want to know about it,” that kind of thing can certainly have an impact. The other direction you can come at this from is when people build tools which measure non-traditional contributions. Scientists have a tendency to look at those numbers and they start to report them. A classic case is in high-energy physics, part of the reason why there’s been this rise of a preprint culture in high-energy physics is that it’s possible to measure the impact of those preprints. There are tools available to measure the impact, and so you can demonstrate to your tenure committee, or whatever it is, that “Oh, it’s just in preprint at the moment, but Ed Witten and whomever have cited it already, so it must be good work.”

So the directives from the grant agencies are important of course, but it strikes me that they’re more of a reflection of the culture of the scientists and what they desire. It seems like the hard work is really on the ground.

A bit of both. I mean, the great thing, of course, about the grant agencies is, even though there’s a lot of scientists in senior positions there, to some extent they’re removed from the everyday concerns of a working scientist. And so they can afford to have a broader view and not just say, “What’s in my immediate short-term best interest?” But they can also say, “Well, what’s in the best interest of society as a whole?” And so if you look at things like open data mandates in the past, often the grant agencies have played a real leadership role in causing communities to come together and decide, “Oh yes, we should start to systematically share our data.” And so it’s not good for them to be too far out in front, but often they do seem to be a little ways in front.

If they put forward these ideas, are they going to be accepted by the scientific community, or will there just be revolution and pushback?

That’s the reason why you want the grant agencies to be working in concert with the scientists in this leadership role, but also not coming in and just mandating, “Thou shalt share all your data!” or “Thou shalt share all your code!” That kind of mandate is asking for people to be passively non-compliant, to share data that hasn’t been cleaned up or that is actually just useless to other people for various reasons. And so there needs to be a conversation that goes on first about what’s actually useful to share, what’s feasible to share, and what an appropriate reward or incentives would be. You can’t get too far in front of the community.

That brings us to the second major force of pushback, which is the entrenched financial interests and the current publishing system and that sort of thing. I’m not sure what I want to ask about that other than – I’ve seen movement on their end, but this does seem like one of the major roadblocks.

Yeah, the publishers, the traditional publishers, are all over the map. Some of them strongly oppose any kind of open access. Others have been pioneering open access models. And some are transitioning from one place to the other. A really interesting, maybe encouraging, thing is that one of the biggest publishers, Springer, bought BioMed Central, arguably the world’s largest open access publisher. Springer has also moved to a hybrid open access model on all its other journals, whereby you as an author can opt to pay a fee to make your article open access. That indicates that Springer sees that as the viable commercial model. And if they can transition from their old business model to a new business model that makes open access at least possible for all papers, then actually maybe this isn’t such a big obstruction after all.

In recent weeks the issue of access to data has exploded with the Research Works Act, a bill that would prohibit agencies from requiring open access. A large number of scientists have responded by boycotting Elsevier, a major journal publisher that supports the bill. Are we seeing the moment when open science becomes a major issue for most scientists?

The Elsevier boycott has generated a great deal of interest and discussion about open science, and I think that’s great. Of course, a boycott can only ever be a part of what needs to happen: we need positive actions as well, not just an agreement about what not to do! But what’s good about the boycott is that it has engaged lots of new people in serious discussion about better ways of doing and communicating science, and some of those people are taking action. That’s exactly what’s needed for open science to thrive.

With that said, there’s still a long way to go. Open science is still at most a minor blip (or invisible) on the radar of most scientists. But I think we’ll see more and more actions like the Elsevier boycott, building awareness of open science, and encouraging many more scientists to do more of their work in the open.