If it ain’t open, it ain’t science
Last Friday (13th March) the Royal Society launched a study into “openness in science”, as part of which they are inviting submisions from individuals and organizations. According to the Royal Society website
Science has always been about open debate. But incidents such as the UEA email leaks have prompted the Royal Society to look at how open science really is. With the advent of the Internet, the public now expect a greater degree of transparency. The impact of science on people’s lives, and the implications of scientific assessments for society and the economy are now so great that people won’t just believe scientists when they say “trust me, I’m an expert.” It is not just scientists who want to be able to see inside scientific datasets, to see how robust they are and ask difficult questions about their implications. Science has to adapt.”
I think this is a timely and important study which at the very least will reveal how different the attitude to this issue is between different science disciplines. On one extreme we have fields like astronomy, where the practice of making all data publically available is increasingly common and where most scientific publications are available free of charge through the arXiv. On the other there are fields where experimental data are generally regarded as the private property of the scientists involved in collecting the measurements or doing the experiments.
I have quite a simple view on this, which is that the default should be that data resulting from publically funded research should be in the public domain. I accept that this will not always be possible owing to ethical issues, such as when human subjects are involved, but that should be the default position.I have two reasons for thinking this way. One is that it’s public money that funds us, so we have a moral responsibility to be as open as possible with the public. The other is that the scientific method only works when analyses can be fully scrutinized and, if necessary, replicated by other researchers. In other words, to seek to prevent one’s data becoming freely available is profoundly unscientific.
I’m actually both surprised and depressed at the reluctance of some scientists to make their data available for scrutiny by other scientists, let alone members of the general public. I can give an example of my own experience of an encounter with a brick wall when trying to find out more about the statistics behind a study in the field of neuroscience. Other branches of physics are also way behind astronomy and cosmology in opening up their research.
If scientists are reluctant to share their data with other scientists it’s very difficult to believe they will be happy to put it all in the public domain. But I think they should. And I don’t mean just chucking terabytes of complicated unsorted data onto a website in such a way that it’s impossible in practice to make use of. I mean fully documented, carefully maintained databases with both raw data, analysis tools and data products. An exemplar is the excellent LAMBDA site which is a repository for data arising for research into the Cosmic Microwave Background.
I’ve ranted before (and will no doubt do so again) about the extremely negative effect the academic publishing industry has on the dissemination of results. At out latest Board of Studies meeting, the prospect of further cuts to our library budget was raised and the suggestion made that we might have to cancel some of our journal subscriptions. I, and most of my astronomy colleagues, frankly don’t really care if we cancel astronomy journals. All our relevant papers can be found on the arXiv and/or via the NASA/ADS system. My physics colleagues, on the other hand, are still in hock to the old-fashioned and ruinously expensive academic journal racket.
One of the questions the Royal Society study will ask is:
How do we make information more accessible and who will pay to do it?
I’m willing to hazard a guess that if we worked out how much universities and research laboratories are spending on pointless journal subscriptions, then we’d find that it’s more than enough to pay for the construction and maintenance of sufficient open access repositories. The current system of publishing could easily be scrapped, and replaced by something radically different, but it won’t be easy to change to a new approach more suited to the era of the internet. For example, at present we are forced to publish in “proper journals” for the purposes of research assessments, so that academic publishers wield immense power over university researchers. These vested interests will be difficult to overthrow, but I think there’s a growing realization that they are actively preventing science adjusting properly to the digital age.
Anyway, whether or not you agree with me, I hope you’ll agree that the Royal Society study is an important one so please take a look and contribute if you can.
In the Dark 
May 16, 2011 at 2:15 pm
[…] “… I have quite a simple view on this, which is that the default should be that data resulting from publically funded research should be in the public domain. I accept that this will not always be possible owing to ethical issues, such as when human subjects are involved, but that should be the default position.I have two reasons for thinking this way …” (more) […]
May 16, 2011 at 3:37 pm
One should also inform people like the British Chiropractic Association that, as the Royal Society puts it, “science has always been about open debate”.
There is one further factor to take into account. “Public funding” means funding by the taxpayers of a particular nation (or group of nations), yet open access via the internet makes that data available to everybody, including people who didn’t fund it. I do believe in international generosity rather than xenophobia, but it is not unusual for useful data to be generated out of a spirit of intellectual competition on large scales, and this could not happen under the model you advocate.
Anton
May 16, 2011 at 4:21 pm
Anton, usually astronomical data is available to the team who proposed the mission/ particular observations or even the country (or countries in the case of some European Southern Observatory data) for a short proprietary period so that those who have taken the risk/ put in the effort beforehand can reap the benefits. After that the data is usually made available to everyone. This means that other people are able to test the results but still gives advantages to those who put up the money or time. Of course there is a lot of argument over exactly how long the proprietary periods should be and when exactly they start from.
May 16, 2011 at 6:01 pm
Anton,
I take your point, but I think Nick answered it exactly as I would. Of course there has to be a proprietary period so those who invested in the research get first bite, but thereafter it should be made public. How long that period should be is difficult, but the principle is clear…
The business of turning science into commercial technology is a major obsession these days, but I think that bit should not be funded by the taxpayer anyway. Commercial companies should do it. If they pay for the research then they should own the intellectual property arising.
However, publically funded “blue skies” research, especially that which is done in universities, should be in the public domain. The intellectual property should belong to the public.
Peter
May 16, 2011 at 11:14 pm
A time delay sounds like a good resolution to me.
May 16, 2011 at 7:25 pm
Totally agree about the open access to data and weird ‘this is MY data’ attitude (fine for 1 year or so, not so good when data never sees the light of day because of this).
On the Herschel ATLAS team, we’ve taken the time to not only provided the maps of our first dataset (as well as the techniques we developed to create them) but also created a cut-out tool (so users can search and obtain a small map of their favourite bit of sky), a fully searchable database (web form & SQL) and a catalogue of sources extracted from the maps (including information for each source from ultraviolet-optical-infrared-submillimetre). We also produced a public friendly website with all of this on.
This is not that useful for Joe Public, but good for science: yes sure, astronomers can check the work we’ve already published and make sure we didn’t get it wrong, but the best bit of this is when they use this dataset to do some inventive, imaginative stuff we didn’t think of. It’s a win-win situation for astronomy and will only improve the legacy of Herschel. I think a lot of other programmes are also doing this too.
http://www.h-atlas.org/public-data/overview
May 17, 2011 at 12:58 pm
Peter,
There was a very interesting talk at last year’s ADASS conference by Salvator Mele from CERN about SCOAP3 (www.scoap3.org) which is an initiative from the particle physics community to over-turn the currently scientific publishing model. They propose that, instead of the current publishing model, libraries and funding bodies should contribute funds to a consortium who would pay for a peer-review service to which papers can (and generally would) be submitted, and after acceptance those papers would be published free of page charges or subscriptions.
That seems a very attractive model to me. I think we’d be throwing the baby out with the bathwater if the current “serials crisis” made us drop the idea of a peer reviewed professional literature but this model would mean that, as a community, would no longer be held to ransom by companies making profits from publishing content that we have provided them – and we would be paying the publishers (assuming they – singly or collectively – won the tender for running the peer review and publication service) for those aspects of their current role that are really of value to us.
Like particle physics, we’re fortunate in astronomy in being a relatively small and cohesive community and one that publishes in a small number of journals, so I would have thought that the same model could work for us.
Judging from their website, it would see that the SCOAP3 initiative is heading towards implementation, with the scientific publishers engaged in the process, but it would be good to know more details, if any of your readers are involved in that process.
Bob
May 18, 2011 at 4:29 pm
[…] If it ain’t open, it ain’t science « In the Dark. […]
May 22, 2011 at 11:06 pm
[…] Si ce n’est pas ouvert, ce n’est pas de la science […]
November 17, 2011 at 5:06 pm
[…] Si ce n’est pas ouvert, ce n’est pas de la science […]