Open science

The following talk gives a short introduction to open science, and an explanation of why I believe it’s so important for our society. The talk is intended for a general audience, and was given at the very stimulating and enjoyable TEDxWaterloo event held in the Waterloo region, just outside Toronto, in March of 2011.

31 comments

  1. Some such change is badly needed.
    At present large parts of humanity struggle to arrange peces of information
    without being aware of the possibility of easily arranged pieces of information.
    The possibility seems too disruptive for the credentialled information technology
    community. Such awaeness would make it possible to combine fluency with
    precision in a way that would further accellerate scientific progress. For
    example a statement such as
    6 = count every state where population of some city of it > 1000000
    was executable as part of a general programming, data base, and modeling
    system 25 years ago, but no one can do it now. The technology has moved
    backwards. See; http://users.rcn.com/eslowry/stifled.pdf .

  2. I’d like to see how you tackle the value attached to ideas, particularly in the shadow of impact factors and researcher h indices. For open science to succeed you really need to address the heart of the issue: publication and tenure.

    At this stage the burden is far too shouldered by researchers while the rewards are marginal and the risks as large as ever. “Give away” a potentially novel idea and you risk job security; posit a weak or controversial idea and never escape the stigma the internet will be sure to saddle you with.

    I like the concept; I’m just afraid pride and the cutthroat nature of some disciplines in science (your example of Galileo still echoes true today) look like major roadblocks that need to be considered.

    I could see initial implementation focused on toward public education: The bleeding edge of science, in a possible wiki format, accessible to non-scientists.

    Insightful talk! I thoroughly enjoyed it.

  3. @MIchaelDSchur – Thanks! Yes, the heart of the issue is certainly publication and how jobs are decided: or to put it in some of the language I used at the end of the talk, what types of research contribution are valued, and what types are not. How to change those values is a very difficult question, and unfortunately I don’t have a short and easily digestible answer that is at the same time also comprehensive and immediately convincing. I think these are questions that require an extended argument. I’ve attempted to give such an extended argument in my book “Reinventing Discovery” (in bookstores Nov 2011). The final two chapters are entirely addressed to these questions.

  4. Hello Michael:
    I live in Brisbane Australia.
    I am an actor.
    Stories are my science.
    The same lock up of Intellectual Property thwarts the future of cultural Stories that shape our meta-narrative as the future of data-bytes of knowledge that form the basis of our design-science capabilities.

    The obstacle is the same: a toxic economic commitment to single-bottom line wealth creation.
    Scientists as employees will always live in fear of losing their job, unless they see alternatives.

    There are a few, I suggest.
    Firstly, Science, like art (and here i mean the art of sense-making; where logical human thought and action processes that result in changed/transformed outcomes may seem ‘creative & abstract’, but they somehow work- something that every single person does when they trust ‘instinct’) is the most critical part of any story.

    Without the logical ‘proof’ of a concept into a form of reality that will create value, you have ‘nothing’…
    So, I urge that in terms of value creation, science & art are the building blocks, yet we, the creatives are generally located in subservient/dependant economic positions.

    My proposal is that if design science & art were used as ethical new ways to create health, wealth & wellbeing, then the entrepreneurs who value social, economic, environmental & cultural knowledge sharing (to optimise human awareness, intelligence & capacity to adapt to the emerging ‘climate of change’), could create entirely new social business solutions to the existing, emerging & abiding issues & challenges of our age.

    We need new ethical entrepreneurs.
    WHich wealthy retired scientists might you contact to explore the opportunity to ‘throw caution to the wind’ and risk banishment from the science ‘clique’ in order to create a new creative cluster culture for collaborative competition?

    Find these people and you have leaders, mentors & niche solutions that will open the crack to the better future alternative that will be as slow to grow as the ‘Polymath’ Project, but when critical mass is reached: Your dream WILL reveal the words of Kahlil Gibran from the Prophet:
    ‘Work is Love Made Visible’.
    While scientists are made to work in isolation, in silo’s, deaf to the noise of the mouse-wheel they are on, they fail to hear the economy they are fostering strangling their children.
    What a wonderful vision.
    You must have an amazing parents to be so ethically sound ahead of the pack!

    I will to will thy will.
    Enjoy open science enterprise.
    It’s a story worth unleashing.

    Regards Paul Bishop
    Chief Inspiration Officer
    Arts Evolution.

    PS Check out http://www.BetterMeans.com
    Their Open Enterprise collaboration ethic & process work is sympatico.

    [MN: Thanks, Paul! Nice to see another Brisbanite. Yes, it will certainly take a long time and a lot of hard work to develop open science. As you say, there’s space here for some entrepreneurs to make a really big difference, if they so choose! Indeed, as a small example, my talk was made possible in part by a grant from the Information Program of the Open Society Institute, which was set up by George Soros. ]

  5. I have a couple of interesting links and a thought.

    Sean Carroll recently wrote apost which gave some insight into how the tenure system works at major research universities. I think that pretty much reinforces DSchur’s points above. The success of open source in the field of computers/software leads one to think if something like that can be replicated in the sciences, engineering and research in general. It’s amazing how little we understand what motivates people to indulge in such things, for the strong open source movement just defies understanding in conventional economics. The RSA has an interesting presentation about this.

    Coming to the crux of my comment, can we figure out parallels which tell us not just about possible structures, but also, importantly, how to get there. The latter is probably the most important question, for how do you incentivize scientists/admin(NSF/Academia) to move from the current (local) equilibrium to a (hopefully better) equilibrium which would be based on open science. Of prime importance are the upcoming generation of researchers… who are post-docs and students today, who are also more open exploring the benefits of technology. How do you give them opportunities where open science benefits them rather than hurting their career prospects.

  6. I saw the mention of this on one of the Wired blogs just now, but it was old news for me because I saw the live stream of the presentation as it was happening (being alerted to it by your Twitter feed). I feel so ahead of the curve 🙂 Congratulations on getting more attention for your presentation and your blog!

    [MN Thanks, Frank! You have a knack for being ahead of the curve :-)]

  7. I guess I shot off the previous comment a bit too soon while I was waiting for the video to buffer over my slow internet connection. The talk does allude to some of the questions I raised and also gives a few interesting and thought-provoking suggestions. Here are a few thoughts that hit me as I was watching the video…

    From what I’ve seen, in theoretical high energy physics, the combination of Spires and arXiv is extremely useful. I think that we need more and better databases (like Spires) that can tap online content (like arXiv)… say you took material from arXiv papers and made it searchable. If possible, try to automate frameworks which can mine multiple sources and generate more cohesive knowledge from them. Google Scholar seems to attempt the former and Wolfram Alpha, the latter. What’s needed is more effort and involvement from the scientific community, and these solutions will often need to be tailored to the needs of the specific field.

    [MN: Absolutely.]

    Another thought relates to your example of the genome project. In cases where scientists do share data and code online, how often has the information been put to good use ? Can we expect that with a high probability this (in itself) will lead to better understanding in the field, or is it just a stepping stone to the real aim – sharing of ideas ? Another question is how well documented this data/code is. It’s often easier to get scientists to share their data or code, rather than their ideas or understanding of the same (like documentation). I guess we all realize how painful and useless undocumented code can be.

    [MN: There are, as I’m sure you know, spectacular examples of reuse of data: the human genome project, the protein data bank, the Sloan Digital Sky Survey etc. Each are cases where data reuse has probably led to thousands of papers being written, sometimes very significant papers.

    The question of documentation is an interesting one. In my talk I mention making a change in values so scientists “see it as their job” [and also the job of other scientists] to share data and code; i.e., they’re not just being compelled to share in this way, or doing it grudgingly, but rather it is seen as a core part of their mission, so that they do it with the same intensity as they bring to writing papers. It’s that sort of change in values which is required to make data and code sharing truly useful.]

    Finally, what are people’s opinions on having online references being written by invited scholars… examples being Scholarpedia and Living Reviews in Relativity. Can this model be scaled up and applied to other fields, or are they just flashes of inspiration ?

    We must notice that almost all the initiatives tried so far (as mentioned above) seem to be based on sharing already established information. In such cases, attribution of credit is quite straight forward. We are still some way to go from a comprehensive and well thought out framework where we can actually share ideas and insights to solve open problems.

    [MN: Indeed – in this, the situation reminds me a lot of the early days of science, when it was by no means clear how people should be given credit for their work. It took a look of discussion for the early scientific community to work out the appropriate norms. ]

  8. Great talk, and I couldn’t agree more. There are some grass roots efforts such as the Blue Obelisk movement in chemistry attempting to make open science a reality. I look forward to reading your book when it is ready, and welcome talented speakers making our case to a wider audience. This is the way that science needs to be done, and one of the major reasons I decided to go and work at a company called Kitware that has made a business out of developing open source software for scientific data visualization.

    Exciting times, but much work to do to make open science a reality.

    [MN: Yes! Thanks for your comment – and for working on things like Kitware.]

  9. I would like to take the “open science” idea even further and create “open universities” that eliminate the need for much of our current academic bureaucracy. I see no reason why a smart, motivated person anywhere in the world shouldn’t be able to use all the resources that are available online to learn just about anything, and then prove themselves using some kind of automated homework and testing system. The ultimate proof of their credentials would come from what they contribute to open science projects. Right now academia, at least in the USA where I live, is a feudal, bureaucratic nightmare that rewards what I consider to be a lot of rather mediocre people who are good at conforming to the system rather than being creative thinkers. This bureaucratization is maybe the biggest crisis of science, and is where the internet can really revolutionize things. But to do this we need to move beyond these feudal 18th century, exclusionary models and find a way for smart, creative autodidacts to contribute meaningfully to the enterprise!

  10. Michael,
    Very forward thinking, but the problems really go much deeper into how society rewards participation on the broadest scales and if that could be more effectively answered, then the result would spread into all professional relationships.
    Currently the big problem is that we still treat the medium of economic exchange, money, as a commodity provided by private interests, when the reality is that it is a contract guaranteed by government and the taxpayers. Just as monarchy eventually outlived its usefulness as a function of civil decision making and government became a public trust, when kings lost sight of their larger functions and descended into self-absorption, now the financial circulatory system is being destroyed by those who view it as a personal piggybank. When people realize these drawing rights to productivity are actually a form of public commons, they will learn to invest more of their efforts into personal and professional relationships and not be so economically atomized.
    That said, my I test your openness to off the wall ideas and offer up a classic simplistic, non-mathematical, grandiose observation from a self professed crackpot?
    I think a primordial mistake in physics is modeling time as a progression from past to future. It seems obvious from a sense of basic perception, but the reality is the changing configuration of what is turns the future into the past. The earth doesn’t travel the fourth dimension from yesterday to tomorrow. Tomorrow becomes yesterday because the earth rotates. Time is an effect of motion, not the basis for it.
    When we think of it as progressing from one event to the next, it does seem like spatial motion, much as one would walk along a path. On the other hand, if we view it as everything moving about and changing configurations, it is more like temperature. One is the level of activity and the other is the rate of change. When a clock speeds up or slows down, essentially it is because there is a change in the level of activity. Faster atomic speeds lead to faster burn rates.
    Consider multiworlds problems: Presumably the laws governing motion and interaction are deterministic. That’s why they are laws. Yet there is a very definite indeterministic probability and randomness to nature. So when we go from a definitely deterministic past into a probabilistic future, is seems as though multiple realities emerge.
    On the other hand, if we view it as going the other direction, it is the collapse of these future probabilities by those very deterministic laws, which turns the future into the present. We cannot observe all input into any event, prior to its occurrence because its lightcone of potential input doesn’t exist prior to the occurrence. So the future is cause and past is effect. Probabilities proceed actualities.
    Safe to say, this doesn’t get much respect from those in the physics fields, but the usual response is that physics isn’t intuitive and I don’t understand the math. Well, I don’t have a PhD in economics either, but I can see mistakes are being made in the current economic model too. Besides, it’s intuitive to think of time as the present moving from past to future. much as we still see the sun moving across the sky, even though we now realize it’s the earth rotating the other direction.
    I could go on, but I better leave it at this, as an example of the kinds of problems that will arise in your proposal.

  11. Nice talk! I particularly like the part where you suggest that each of us spend a little bit of research time and contribute in this direction. Perhaps this can be taken to be akin to that “20%” philosophy that Google practices, where everyone is encouraged to work one day a week on something that is not directly related to the main research problem and share results periodically somewhere online. Surely, at the very least it could be a nice outlet for experiments that were somehow never published. Anyway, I think it is a very nice way start new conversations, if not some unexpected collaborations.

  12. Dear Michael,
    thank you very much for your talk today. I support your idea in any case, as I can see that it attempts to put more efficiency and some kind of system into research on the one hand, and, on the other hand, opens science to everyone. Science shall be free and people all over the world should have the opportunity to increase their knowledge. Unfortunately, our scholar system does not train us in thinking into this direction. This is my personal impression, or lets say, this is a fact I have to face every day leaving me in front of borders I am not able to cross. Let me explain: during my studies I could gain a lot of knowledge; biology, genetics, engineering, thermodynamics, mathematics, chemistry and physics (by the way: I studied Medical Biotechnology). But what no one ever told me: how to solve a problem. It took me bloody years in science to realize that it is more important to find a system to create a solution than to find the function of a protein. Because, finding a system means to find the function of any protein.
    Open science now is a key to open a door in the right direction. What you are creating, is a kind of supercomputer using the brains of many excellent, openminded people all over the world. This is the first time I hear about this kind of communication, so I have some questions if you are okay with it. How does it work out, to find the solution for the problem you posted (I mean now the Polymath Project: how did he finally get the solution)? To me it seems that it happens randomly, is that right? If dozens of communicators post hundreds of comments, who reads all of them? Which information is significant and which one not? And how do you get then the final solution? Do you think it is possible to create a system which for sure will end up in a solution? Do you think Open Science is such a system? Finding out about how to use it efficiently will also solve the money problem about it. What I want you to tell is, that the time is right to create a solution for finding solutions. We gained so much knowledge over the past decade, but did anyone ever bring it into the right concept? Despite all the new insights we publish day by day, progress in science is slow. I would be happy to share some ideas with you concerning that issue.
    Thank you so much for listening to me. I am waiting for your reply.
    Best Wishes,
    Uli

  13. Nice talk Michael, and great delivery. I hope it can work out. Science needs a boost right now.

    (We Rogers’ are watching . . .)

    Regards

  14. I like your talk but I find that open science is not aggressive enough to break the church of current science. Galileo and Da Vinci as you mentioned in your talk both worked under the church (although in the latter ironically and in the former with heresy). However, they still worked under a system where there ideas could be withheld if they did not conform to the people that were paying them. Many of Da Vinci’s ideas came to light later after discovering his notebooks after his death.

    I am a scientist and have received grants. The people paying me are possibly in government or a private person. They expect that you publish my work. And I do believe in publishing because it allows some single ideas to build on other ideas. Science is not just about collecting a bunch of data and putting it somewhere it is about solving ideas.

    What I do not agree with however is the publishers. In my mind the publishers are the church. They not only reject ideas they stop people from allowing to gain access to those ideas. I am sitting at home right now and I cannot even access a publication on the internet that I wrote! I would have to go a library in a University to gain access or ask a friend. Science is not open because the church has its doors closed. This is slowly changing with online publishing but what about the millions of papers published with valuable conceptual information? As a scientist we are supposed to be standing on the shoulders of giants and not at the feet of the publishers.

    Please contact me if you want to learn more about how to bring the Martin Luther of science.

  15. I echo The cosmist and Chris as well. I have been thinking for a while now for an internet-based “open university” concept to utilize the vast knowledge and potential of scientists outside the universities and university-like companies; however I couldn’t think of a solution for laboratory-based research. One of the solutions could be to rent existing laboratory for practical classes and research. So, may be in future scientists may rent out laboratory space for their research on-demand basis. In that way probably trained scientists, who did not get opportunity to continue their research in conventional way, can get an opportunity to contribute again.

    Any other idea about utilizing the vast knowledge and potential of scientists outside the universities and university-like companies? How do we do that? Who gives value to the potential of dedicated and passionate scientists outside the conventional scientific world?

    The issue of authenticity of research also would be an issue for open science environment as Chris mentioned. Even in the present system we do have problem of due recognition of contributing authors in a published article which contains small number of authors. So, how do you manage that in open science system where number of contributing authors is large? I think before doing so, we need to make set of rules which everybody has to abide. I think in that way we can eliminate the whimsy of supervisors or other influential authorities. The sooner we act the better will be the scientific world.

  16. Thank you for so clearly articulating the idea of open research and collaboration. I am fortunate to work at a very small college which has collaboration and applied research embedded in its very mission. At the root of my work as an administrator, scholar, and teacher, I am drawn to developing further opportunities for both faculty and student cooperation to advance research and learning. That said, in my fields — environmental humanities/ regional studies — there continue to be significant impediments to inter-institutional collaboration in the face of continuing competition for grants and funding and students. Makes me wonder whether there are particular challenges in collaborating on qualitative projects outside of the sciences. I look forward to reading your book.

  17. Nice talk Michael. I couldn’t agree more, particularly with the idea that we all can raise awareness. We are supporting a movement called CIARD (Coherence on Information for Agricultural Research for Development – ) that promotes the concept of ‘opening up agricultural knowledge for all’.

    I think the questions of incentives and barriers are critical. Las year, we carried out a global online survey with researchers working in agriculture and related areas, to gain greater understanding of their behaviors and attitudes in relation to communicating research outputs and making such outputs open and accessible. The analysis results are online on the CIARD site (click here), which brings many valuable insights into current perceptions of researchers into the communication of their outputs particularly in Latin America, Africa and Asia. We expect that this findings can be considered by senior managers in agricultural research systems as they review their organizational policies.

    I look forward to this estimulating exchange in your blog, and to read your book this weekend.
    Cheers

Comments are closed.