Jean-Sébastien Caux

One of the interesting consequences of the constantly growing number of articles, statements, posts and tweets concerning Open Access is that its very definition has become somewhat blurred. Others (e.g. Martin Eve) have already underlined the need for more precise definitions in order to resolve the degeneracy. This post is meant to quickly explain the labelling system which I personally like to apply to distinguish between the different flavours of OA. It will not be to everyone's taste; on the other hand, perhaps you will also find it useful.

First, the obvious: the term "publisher" is meant to imply that an entity performs at least basic tasks, such as running a strict quality-controlled peer review process, ensuring ISSNs for its journals, running professional-level post-acceptance production, operating an online platform for its content, providing metadata handling (registration of DOIs), providing linking of funding information back to funders and ensuring perpetual availability via a dedicated, digital archiving and preservation service.

To merit the label "Open Access publisher", the expectations are higher. Since gold comes in different levels of purity, let us be inspired by metallurgy and adopt karat-level resolution for Gold content in order to differentiate the levels of open accessibility being delivered (a more technically correct and modern scheme would use fineness (parts per 1000), but "Gold 950" or "Gold three nine fine" just don't sound good in the current context). My personal choices are:

The precise ordering of this list admittedly does not originate from fundamental science; the organization based on the principle of increasing generosity does however give it some form of logical basis. In any case the list does not include all the requirements one might think of. For instance, at the moment there is no mention of the publisher's business model. Since "Platinum" is already often used to denote publishers which do not impose publication charges, a more general elemental nomenclature I like using is:

In this scheme, Au and Pt are incompatible; Pt seamlessly binds with Pd, while Au and Pd can also mix consistently. One could call Au/Pd "White Gold" without ambiguity (since Au/Pt is here impossible). Amusingly, good-willed scientists can now share their love of Pt/Pd processes with traditional photography buffs.

I also like to use other elements with slightly less positive associations:

You might find iron too reminiscent of jail bars and shackles, but still, the parallel is amusing. Lead pollution affects many publishers, especially in the form of improperly-implemented APC systems with publish-to-cash-in incentives.

Although karats are not conventionally used for precious metals other than gold, let us bend the rules and characterize publishers according to their elemental business model (Gold (Au), Platinum (Pt) or Palladium (Pd)) and their karat openness rating. One can then identify some publishers as Gold 18-karat, Platinum 22-karat, etc, allowing for a simple "lifting of the degeneracy" in the current nomenclature.

One can thus argue about whether Au10k publishers are sufficient, Au/Pb10k ones acceptable at all, or insist that Pt/Pd24k is the crème de la crème and the only model worthy of support. Funders and institutions that are sufficiently forward-thinking about meaningful mandates could decide to only reimburse APCs (up to a cap) for 18-karat publishers and above, with a lower cap for non-Pd ones. On the workfloor, scientists could decide to not do any unpaid refereeing or editing work for non-Pt/Pd publishers, especially those with Pb pollution which make a mockery of the peer review system.

So there you have it, at least now you'll know what I mean if I say that although I'm thankful for so much Gold, I am allergic to Lead pollution and anyway much prefer imagining a future made of Platinum/Palladium, especially the 18-karat and above level of purity.