There is something about the term “post-publication peer review” that doesn’t quite fit. On one hand, post-publication peer review seems inevitable. As more scholarly content moves online (and not just scholarly content), with little or no editorial gatekeeping–or in the case of services like PLOS ONE, a stripped-down form of peer review–, surely there is going to evolve a method of evaluating materials after the fact. On the other hand, is material that appears online without prior editorial review properly called a “publication”?  And if such a piece elicits comments from people outside a specific field, could such commenters be called, “peers”? The term “post-publication peer review” seems like an early and imprecise way of describing something that is just coming into existence, akin to “horseless carriage” to describe an automobile. But “horseless carriage” was a good place to start, even if the vehicle was being defined by what it was not (horseless) rather than what it was. As we enter the age of the self-driving car we might ponder the significance that “self” in this instance refers not to you or me but to a machine. With self-driving cars in our driveway, are selfies bound to follow?

The notion of post-publication peer review implies a before and after in two senses. First there is the chronology of the original document and the comments upon it (it is, after all post-publication peer review). Second and more importantly there is the implication of two publishing ecosystems, one that went before and one that came after. The one that went before is traditional publishing, which remains the dominant paradigm today. This form of publishing is sometimes called “toll-access publishing,” a misleading term in that it defines traditional publishing by a single aspect and leaves out, for example, the editorial and marketing dimensions. In contrast to traditional publishing we have open access (OA) publishing, and it is for that paradigm that we need to come up with a better term than “post-publication peer review.”

This is harder than it sounds because the world of OA publishing is diverse and seems likely to remain diverse for some time. In addition to such distinctions as green vs. gold OA, there is the question of the source of funding, the complication of hybrid publications, and the rights or capabilities that readers–and text-mining machines–have with regard to the content. On top of this is the wide range of editorial practices, some of which are barely distinguishable from those of traditional publishers (and are often operated by these very same publishers) and some of which largely or wholly abrogate the role of editorial gatekeeper. How do we talk about post-publication peer review if we can’t pin down pre-publication peer review?

What traditional and OA publishing share is that editorial review exists along a continuum, a point that serves to undermine the “gatekeeper” metaphor, with its implication of a binary state (your piece is evaluated by the gatekeepers, but then enters the hallowed halls without further scrutiny). In the traditional model, papers and books are circulated first among colleagues. Then the document is submitted to a publisher that sees to it that the proposed publication is reviewed by an assortment of people in a variety of ways, of which peer review is but one. If the work is published, it then enters the world of public discourse. Reviews may appear, and some works receive the ultimate compliment of a citation. In a world of social media, even traditionally published works may begin an afterlife of tweets and Facebook “likes.” The gatekeepers are long behind us, but different forms of evaluation go on.

With OA publishing the first step is typically the same as in the traditional model: the circulation of a document among colleagues. But when it comes time to submit a paper or book to a publisher (for OA, the activity is mostly with papers), however, things may change. For starters, the editorial review process ranges all over the place. But even after a manuscript is accepted and made available on the Web, comments may continue to appear, and the nature of the open Web is that the comments can be made by anyone just about anywhere. Some papers, of course, will simply appear on the Web and then be forgotten. Indeed, perhaps the fastest-growing publishing paradigm today could be called “post-and-forget publication,” a form of publication where the act of production, leading up to making something openly available, is viewed as an end in itself.

This raises the interesting and important question of whether publication in an OA journal represents the end of a process or the beginning of a different one. The difference is marketing, a term that is often misunderstood in scholarly circles. Marketing means creating demand for something. Traditional publishers do this with their brands and (for books) their authors. For OA publishers the challenge is to continue to keep pushing a particular paper after it has appeared online. There are many ways to do this, of which simply making the content openly available is one (allowing an article to get indexed by search engines and pointed to by bloggers and others). But to continue to keep the article in the eyes of its prospective readers, new means of attracting attention have to be developed. We are only in the earliest stages of this.

Rather than talking about post-publication peer review, we really should be talking about an ongoing conversation that begins with colleagues and perhaps never really ends. This is going to be frustrating for those who yearn for the solidity of a fixed text, and these people have my sympathy. But the Web is developing its own kinds of systems, some of which may prove valuable, some of which are at best silly. For publishers working on the Web today, the question increasingly will be how to begin with OA, how to develop the means to identify the qualities of a publication after it has made its first public appearance.

A footnote to this discussion pertains to finance. The prevailing method of financing OA is the gold model, where authors or their proxies pay for publication.  There is an inherent structural problem with gold OA in that payment is centered on one individual, whereas traditional publishing spreads costs across the set of all customers. Thus a gold OA publisher has little incentive (and no money) to continue to flog a paper after it has appeared online. I suspect that we are likely to see tiered services develop in the world of gold OA–a fee (tiny) for submission, a fee for publication, a fee for ongoing hosting, a fee for inclusion in new discovery programs, etc. The business of OA publishing is going to become more complex in the coming years and will require a new generation of astute managers.