Open Science: a simple proposal
Last week, I spent two days discussing ‘open science’ in an EU meeting, as part of what will culminate in a meeting on “opening up to an ERA of Innovation” next month. We were asked to make suggestions for what the European Commission could do to promote the idea of open science. Most suggestions were about ‘raising awareness’, ‘removing barriers’, ‘creating incentives’, and all these other things that are no doubt very useful, but also, at least to me, rather boring…
So I tried to come up with a very concrete proposal and suggested the following:
I propose that the European Commission takes the bold step to ensure that, before the end of this decade, all metadata about all scientific publications ever published would be put in the public domain.
By all metadata, I mean things like title, authors and affiliations, but also papers cited and papers that cite a paper.
Of course, I don’t mind if the EC collaborates with our friends across the world to make this happen, but it should do this on its own if that collaboration leads to long delays.
And I know that this simple proposal doesn’t realise the vision of Open Science on its own, but I do think it will make innovation possible where it is currently very difficult to innovate because all the data is locked behind the walls of Web of Science or Google Scholar etc.
I understand that there will be some discussion about what exactly is a ‘publication’ but I think we can agree on a working definition.
If there is some budget left, then I would suggest that the European Commission also takes steps to ensure that metadata of new publications will also be added to the public domain. But if that is problematic, then it is better to ‘just’ do all publications until now rather than doing nothing. And if we really want to be bold, then we should try to put all publications in the public domain, but I realise that this is potentially difficult to do in a way that respects the law. (Although we can change the law, no?)
Anyway, my suggestion seems like a simple enough idea, that wouldn’t actually cost all that much. Actually, I have some ideas on how this could be realised 😉 But, above all, I think it would send a clear message that Europe really backs open science, and is willing to invest in concrete steps that enable it. Who could not be in favour of such a proposal?
Erik Duval's Weblog 
I agree with your position.
What about the data? I know this is the ‘third rail’ of this conversation, and I realize that sharing datasets is fraught with comparability issues, etc. But it is nevertheless worth doing in some fashion. And at what point is the actual article text getting past the firewalls of publisher business models?
Love to hear your thoughts on the European Commission steps to ensure that metadata of new publications will also be added to the public domain.
Agree that making data more open is also important. But I also think it is much more difficult. I assume that for instance CERN generates much more data every day than all the metadata of all the publications ever published. Not only much more, but also much more complex data.
So, data is important too: but I don’t want a solution for data to delay a solution for metadata about publications.
In the past years, at many conferences, I have referred to the lack of open metadata about all scientific publications as the biggest failure of the open access movement. So, I am totally with you on this. Metadata is fuel for value chains and hence essential.
Meanwhile, I would credit the open access movement with some other major failures, such as officially not considering venues like arxiv.org as open access (see Wikipedia definition of open access) and the complication of what could be a simple discussion with colorful (pun intended) nomenclature. But that’s a different story.
I agree, of course, with other comments that open science needs a lot more in the open than metadata. But, metadata is a pragmatic and – one would hope in 2015 – achievable step.