Tectonic shifts are happening in the way scientific research is done. These changes, and the ways they may eventually affect us all, are chronicled in a new collection of articles commissioned by Nature, one of the most venerable scientific journals.
So, putting aside for the moment that Nature articles themselves are not (usually) open-access, what can we take from this new collection?
A way forward
There are interesting perspectives on how open access to publicly-funded research can forge ahead.
Alma Swan – in Advocacy: How to hasten open access - identifies the features of a Good Open Access Policy. Copyright should not become a hurdle or interfere with an author’s choice of journals; “immediate access” should be a guiding principle.
It should be pointed out the major Australian funders – the Australian Research Council (ARC) and the National Health and Medical Research Council (NHMRC) – approximate this in spirit but unfortunately still allow publishers to veto the requirement for open access.
Swan points out that agencies across the world must harmonise their policies to avoid confusing and burdening researchers.
Technological developments are making it easier to treat collections of data with the respect they deserve. It is now possible to cite data using permanent links (DOIs), meaning data as well as papers can be treated as a basic contribution to knowledge.
With this ease of citation should come ease of academic reward for data, particularly if the data are discoverable and reusable.
Nature’s Richard Monastersky – in his article for the new series, Publishing frontiers: The library reboot – writes about the response of libraries to the onslaught of large open datasets, and the curation of those data.
Institutional repositories (such as ours at Sydney University) are centre-stage in Australia for open access to data and to papers.
Management of digital outputs is one of the most critical issues for Australian universities in the coming years: if we are going to claim we are using public funds responsibly, we will need to make sure we are managing data well.
Universities are not currently prepared for this. Monastersky rightly highlights the Australian National Data Service, which universities should partner with to manage data.
Beyond human gatekeepers
In the most forward-looking piece of the Nature collection – Scholarship: Beyond the paper – Jason Priem notes we are becoming victims of our own success.
The totality of human research output is already far too large to be curated and judged by humans alone. But this information overload presents opportunities.
Priem’s argument centres on an abandonment of people alone as the gatekeepers of quality work, and he describes a future in which people are guided by machines:
Over the next ten years, the view through these open windows will inform powerful, online filters; these will distill communities’ impact judgements algorithmically, replacing the peer-review and journal systems.
In an effective analogy for those who remember the early years of the web, he points out that real progress in search happened when we abandoned low-throughput human curation:
In the late 1990s, commercial internet services such as Yahoo! found that hiring experts to create vetted lists of web pages completely failed at Web scale; the same will be true of scholarship.
The measurement of productivity and impact will be via nuanced, complex algorithms, so-called “altmetrics” that will create better pictures of the influence and importance of papers than the crude measures we employ today.
Some of the indicators Priem mentions, such as tweets, are individually ephemeral, but once combined and digested by algorithms they yield a complex and interesting map.
The need for sophisticated and reliable assessments of impact is certainly gaining official traction; agencies such as Australia’s NHMRC have warned against the use of the cruder journal impact factor.
These initiatives will require a good licence for open-access articles. In his contribution to the current Nature special – Licence restrictions: A fool’s errand – John Wilbanks argues that the “CC-BY” licence (used by The Conversation, among countless others) already has everything we need.
It’s simple to understand and allows full use of article content by people, companies, and machines.
Some academics are deterred from posting their research online due to fears of an unfiltered flood of opinions on their work from people poorly placed to comment.
This problem can be handled once we are able to carry our reputations with us, so we can partly judge a contribution’s significance by who made it. Such solutions have already been developed outside academia (such as for programmers collaborating on code).
For science and the humanities, thanks to initiatives such as the Orcid researcher ID, reputation will become more public and incorporate blog posts and comments on others' work as well as papers.
The rapid development of such research outputs outside traditional publishing will, Priem argues, continue to grow, “driven by both the value of improved networking and the fear of being left out of important conversations”.
Scientific articles traditionally involved only a select few researchers reading a piece of paper. An article now can involve discussion online, immediate criticism, and interaction with data.
These extraordinary changes should be seen as a positive by working researchers, because we all regularly see flawed papers appear but traditionally we haven’t had an easy means to comment on articles.
Journals were originally intended for better dissemination relative to personal letters, meaning faster communication and to more people. Papers became built on evaluation by peers.
These are the same themes at the heart of a more open, online process - we’ve merely turned up the dial. “The reviewer will metamorphose from gatekeeper to interlocutor and collaborator,” writes Priem in Nature.
Let’s return to the issue of Nature’s own articles not being open-access. For cases in which the individual author is funded by agencies with strong open-access mandates, the corresponding article will be made available freely on a repository.
Nature editor-in-chief Philip Campbell recognises that this movement towards enforced open access will continue.
Last year, he described the idea of free availability of research as “very compelling,” adding: “My personal belief is that that’s what’s going to happen in the long run.”
We should also remember that, while researchers are frustrated by journals that contribute little beyond typesetting but are nonetheless very expensive, journals such as Nature sometimes write new content themselves, which needs to be paid for.
But open-access mandates typically apply only to content by academic and government researchers. It is to this kind of work that Nature’s collection of articles speaks.
The collection is a sign that the argument about openness in research has shifted decisively from “should we?” to “how do we?”.
Australian and New Zealand readers interested in policy regarding openness in research may be interested in the draft Tasman Declaration on Open Research.