There’s been a great deal of reporting and discussion about Nature’s sharing initiative, announced earlier this week.  Most of it has been very positive, for which we’re grateful, and some of it has also been unambiguously negative, which we also appreciate. After all, it’s only through open honest discussion that we’ll arrive at good solutions.

There also seem to have been some misunderstandings about what’s going on, and I’d like to correct, or at least respond to, some of those here.  I’m thinking in particular of blog posts by Peter Murray-RustJon Tennant and John Wilbanks, in part because I know and respect all of them, but also because I think they raise points of concern to other people too.

First of all, this initiative in no way pretends to be, or competes with, or attempts to replace true open access (OA).  Since I left to set up Digital Science (DS), I no longer speak for Nature Publishing Group (NPG), but over the last 30 years I have variously been a reader, subscriber, author and employee of theirs.  I also remain a fan, and of course we at DS have collaborated with them on this initiative.  From this perspective, I’d like to point out that NPG has long been one of OA’s biggest proponents.  Since 2002 they have not required authors to transfer copyright; in 2006 they were the first commercial publisher to adopt a 6-month self-archiving policy; and since 2008 they have actively deposited thousands of manuscripts into PubMed Central on behalf of their authors.  Nature Communications was recently the first Nature-branded journal to go fully OA, and OA is now the fastest-growing part of NPG’s business, accounting for about 40% of their papers; they expect this to grow to 50% next year.  None of this will stop or slow down at all.  On the contrary, authors will go on publishing in OA journals and archiving their manuscripts in funder and institutional repositories in increasing numbers, and that is a good thing.

But that leaves another 40-50% of papers (>80% across the industry) that are still behind a paywall.  You might say ‘Make it all OA!’ and many of us would agree with that aspiration.  But simply wishing for this won’t make it so, and back in the real world it’s not going to happen soon, particularly for journals where the cost of publication currently exceeds authors’ ability to pay.  At the same time, researchers are increasingly bypassing paywalls by sharing papers with each other (at least this is what most of us believe, though there’s precious little good data). This is driven in part by the relative ease of sharing PDFs, and in part by the increasingly global and interdisciplinary nature of research collaborations (which means that even researchers at well-funded institutions can’t always expect to have access to the same content as their collaborators).

One fairly common publisher reaction is to point to copyright law and shout “Stop!”. To their credit, NPG realise that this response is neither credible nor in the interests of research.  So they asked us not about ways to inhibit sharing, but ways in which we could make easier and more functional.  We also thought about ways to make this activity more visible to authors (who deserve credit for the attention they get) and institutions (who want to know what researchers are reading) as well as, yes, the publishers themselves. To their further credit, and like the good scientists they are, NPG have also said that they will share the results publicly in order to inform this debate – in which, as I say, there has been precious little hard data to go on.

(It is true that such data will be of interest to Altmetric, another DS company, but Jon’s comments on this point misunderstand our motives: we think altmetrics are important, which is why we invested in Altmetric and, in part, why engaged in this initiative; we did not engage in this to support Altmetric, who haven’t been directly involved at all. We also don’t see this as a way to give NPG an altmetrics advantage; on the contrary, we think other journals should do the same.)

At Digital Science we like to think of ourselves as idealists, and we look forward to a world in which all research papers are fully OA.  But we also pride ourselves on our pragmatism and on providing on real-world solutions that help people here and now, not in some Panglossian ‘best of all possible worlds’.  This week, as a result of Nature’s willingness to take a risk and ReadCube’s impressive software, millions of readers all over the world can do something that they have never been able to do before.  If you don’t like it then don’t use it – everything you could previously do to access and share journal content you can still do today – but to consider it a backward step or get angry is absurd.  I believe this a step in the right direction, so let’s welcome, learn from and build on it.  Whether you consider it a help or an irrelevance, it’s certainly not a hindrance.  Meanwhile, the work to make research more open goes on.