So “yes!” to open science.
The OpenScience project (6 interrelated projects writing and releasing free and Open Source scientific software in a collaborative environment, led by Dan Gezelter, a chemistry professor at Notre Dame, currently over 470 software programmes available) gives what I found to be the most useful definition:
- Transparency in experimental methodology, observation, and collection of data.
- Public availability and reusability of scientific data.
- Public accessibility and transparency of scientific communication.
- Using web-based tools to facilitate scientific collaboration.
The #ioe12 module on open science kicks off with a slick TED talk by Michael Nielsen, one of the pioneers of quantum computation. Nielson tells a collection of open science stories looking at the successes e.g.
- Tim Gower’s polymath project which asked the question ‘is massively collaborative mathematics possible?”, he ended up with over 800 comments.
And the failures e.g.
- John Stockton’s quantum wiki (quiki). It was like Wikipedia but specialized on quantum computing. Nobody was interested in contributing.
Nielson argues that although open science leads to an acceleration in the rate of scientific discovery there is much stacked against it. The situation still stands that writing a single mediocre paper will do more for your career than writing lots of brilliant comments on blogs. Scientists aren’t rewarded for sharing their data. The successes (like polymath) only work because they use unconventional means to an conventional end (the end result was a scientific paper).
Nielson speculates that this is changing. This often because openness has been embedded into principles. The Bermuda principles used on genbank (a genetic sequence database) state:
- Automatic release of sequence assemblies larger than 1 kb (preferably within 24 hours).
- Immediate publication of finished annotated sequences.
The aim is to make the entire sequence freely available in the public domain for both research and development in order to maximise benefits to society. In my working environment I’ve seen the situation changing due to research council guidelines (such as the EPSRC) and through funder requirements.
However lots of data is still locked up and it is routine for scientists to hoard data. The open science movement want to change the culture of science and the value of individual scientists. Scientist need incentives to share.
I thought it was interesting that in the ‘definition of open science’ blog post on the Science Commons web site Cameron Neylon is quoted as saying:
“I think for me the most striking outcome of [a session to define it] was that not only is this a radically new concept for many people but that many people don’t have any background understanding of open source software either which can make the discussion totally impenetrable to them.“
It appears again that understanding the open source movement is essential for getting a grip on all the open ‘products’. Yet people don’t know the history. I’m starting to feel like I’m repeating myself. In the The Meaning of Open Content #ioe12 post I actually wrote “It appears that it is incredibly difficult to understand the terms used in the open movement without understanding some history and background.“. At least I’m being consistent!
The Introduction to Science Commons Concept paper uses the fictional case-study of a Brazilian postdoctoral student to explore some of the concepts behind the open science movement. Open science is often about access to science research through open access journals but it is also about allowing us to mine the data that is there. This data mining is sometimes carried out through technical means (semantic web, linked data, open data) but often takes the form of collaboration – human to human. Eliminating the legal and technical barriers to building a “semantic web” for science is what Science Commons is all about.
“Perhaps the result would be dramatic; some fairly impressive scientists and computer scientists believe so. Perhaps it would be more modest. But where it is practicable to do so, lowering those barriers is clearly a good idea. It might be a great idea.”
Science Commons takes many forms. The Science Commons Publishing Project promotes effective use of digital networks to broaden access to all three types of information: data, peer-reviewed journal articles and metadata. It does this by encouraging pragmatic open access publishing, self-archiving and facilitating the use use of metadata. The Science Commons’ Licensing Project is working on simplifying licensing and the creation of a ‘research commons’ (a funded place researchers can put open research. They are advoicates of the semantic web and keen to promote common formats for interchange of data.
Another key area in the open science movement is open notebooks, making the entire primary record of a research project publicly available online as it is recorded. The wikipedia article gives a good overview with links to key practitioners and software. The article also explores the arguments for and against opening up notebooks. The biggest deterrent to researchers is the possibility of data theft and difficulties regarding patents and publication in peert reviewed journals. Another concern is data deluge – the importance of curation and validation of data are highlighted here, issues that fall under the DCC remit.
The final ioe12 resource is a overview of a session ran at a Open Education 2011 Meeting by Sarah Kirn and Ahrash Bissell on Open Science and OER: Where do they Intersect? The premise of the session was is that “‘open’ efforts may not be so seamlessly interoperable as we might think.” Kirn and Bissell see open science as “democratization of the capacity for anyone to ‘do’ science as well as the elimination of the barriers to accessing the outputs of scientific research“. This strikes me as being the first true indication (though it has been implied) that the ‘citizen science’ concept also falls under open science. Their write up from the discussion session considers ‘citizen science’ changes anything and if it is diminishing the expertise necessary for science, or is it expanding it (by requiring new and additional roles), or simply shifting it to new places. There was also discussion of the worries that opening things up means that people can ‘mess things up’. It seems this concern is ubiquitous in both OER and open science conversations. The worry was data is more sacred – it’s clear that dataset integrity is crucial for accurate archival and referral.
The ioe12 resources give a useful introduction to open science. It’s definitely an area I intend to find out more about!