Page Proofs ≠ Post-Prints; Websites ≠ Repositories
As Alex Golub’s Star Wars themed translation of two of my recent posts demonstrates, I am not the best communicator on scholarly communications issues, but I will keep trying. A note on the form that articles can take when circulated in green OA fashion, as well as the places where such materials can be “put” follows below.
In response to Alex’s post and one of the comments found associated with it, I would like to stress that page proofs are not the same thing as pre-prints or post-prints. At the end of this post I will also stress that websites are not as good as, or the same thing as, librarian- managed repositories.
[Before continuing, will someone, anyone, please, please, please just look at this link: http://www.sherpa.ac.uk/romeoinfo.html ?]
While some publisher’s author agreements allow (or even encourage) authors who are circulating their articles via the green OA channel to circulate the final typeset versions, this is rare and most DO NOT ALLOW for the circulation of anything other than pre-prints (manuscripts as they were before peer-review) or post-prints (manuscripts as they are after author revision but before any production work is done on them.)
Typesetting, markup, etc. are value added processes and an investment on the part of the publisher. From their point of view, they are no obligated to give this labor to you or to give it away. (Just as you are not obligated to give your work to them in the first place.) No OA mandate that I know about demands the publisher’s version and the circulation of such versions (despite the commonality of the practice) is usually a violation of one’s author agreement and thus, technically, a violation of copyright law. Just because you have not yet gotten a take down notice or a cease and desist letter from your angry publisher or chewed out by your university’s legal office, does not mean that you will never have these things happen to you. When you sign a standard author agreement, you transfer the rights to your article away to your publisher. It is not yours, it is theirs. There are a variety of rights that they can choose to allow you to retain via your agreement, but you have to read your agreement to know what these rights and these limitations are.
The only way around this is through the use of an author addendum such as that provided by the Scholars Copyright Addendum Engine. This is the reason that Kelty et al. were able to legally circulate the published version of their 2008 article. Ex: https://scholarworks.iu.edu/dspace/handle/2022/3167 An author addendum is a legal tool by which you can request and retain additional rights in your work.
In a world in which their are excellent metadata harvesting protocols (OAI-PMH), there is no need for a “Napster” for articles. Tools for discovery like Google Scholar and Open Folklore are all that we need. What is required instead are scholars willing to understand how the these processes work and who are willing to stop posting their articles on their personal and departmental websites and start legally putting them in proper institutional, subject, or funder repositories where they will be discoverable and available for the long haul. (Websites are better than nothing and some scholars may struggle to find a suitable repository to permanently host their work, but this problem diminishes almost daily and there are multiple pathways around it.)
I will be addressing these two themes–what author agreements do/do not allow and the costs of using websites–in my AAA remarks on green OA.
Shreds and Patches 
“Before continuing, will someone, anyone, please, please, please just look at this link: http://www.sherpa.ac.uk/romeoinfo.html ?”
Right, I’ve read it (again), but I still don’t think I get it. I understand – or fancy that I do – that “pre-print and post-print” mean something different to RoMEO than they would intuitively to me, and that they differ again from the meaning of “pre-print” in the publisher’s agreement I’ve signed, but I’m still not 100% sure what they mean.
For example, post-prints are “the version of the paper after peer-review, with revisions having been made,” put that “≠ Page Proofs”. So, it’s the draft I typed in response to the reviewers, but not a version handled by the editors or publisher, right?
I’m not even going to hazard a guess about the precise definition of “repository” at this point.
Perhaps one advantage that “traditional” gated publishing has over newer forms, besides simple mental inertia, is the relatively* straight-forward nature of the publishing agreement. It is constraining, but it doesn’t require a researcher to negotiate a host of complex possibilities.
*that is, fairly dizzying relative to positively bewildering
Hi Chad. It is confusing, no doubt, but the terms that you are pondering are all actually ancient “terms of art” (meaning they have accepted and stable technical meanings) in publishing and retain the same basic meanings that they had in pre-digital times. The history of the term “Proof” can be glimpsed on the English language wikipedia entry for “Galley Proof.”
Anything called a “proof” (as in proofreading) has been made by a publisher and has been “set into type” (that is, formatted in the visual form in which it will be eventually published). (Although galley proofs are a special intermediate phase in which the type is set but the layout not implemented. Few authors encounter galleys anymore.)
Typesetting (and thus creating a “proof” (page proof, galley proof, blue lines, etc.) has been created by the publisher or its agents and thus represents labor and expense undertaken by the publisher. They own this investment unless they give you rights in it as part of your agreement. This was true in 1950 and it remains true despite things digital
Wikipedia’s definition of “pre-print” also reflects long term, pre-digital understandings among scholars and publishers and the basic idea remains in place now. An author is capable of making a pre-print. If you make a xerox or ditto or carbon paper or digital copy of a typewritten (or even handwritten) manuscript and circulate it personally to colleagues and other interested parties, you have engaged in the pre-print process. Working Paper series were and are often a formalized means of circulating pre-prints. Before and after digital, laboratories and working groups have done this. Again, the wikipedia entry for pre-print describes this in the ancient way and then acknowledges the life of this phenomena in digital contexts.
A pre-print may never be more than a pre-print, as when we give away 20 copies of a conference paper that we never do anything more to. Publishers have little control over pre-prints, because they are often circulated before (sometimes long before) an article is even under consideration by a journal. In the pre-digital age, no journal was going to ask you to go around and collect up the 20 or so copies that you circulated prior to submitting it for consideration. You own the intellectual property rights to the document when you create it and you retained them when you ran it through the ditto machine and gave the results to your friends. This is the same when you email it now. (You loose your rights in the MS when you sign an author agreement.)
English wikipedia does fine too for post-print. This form also has a long history before OA and before digital. As noted there, it is an author-produced manuscript that has benefited from the peer-review and editorial suggestions that followed submission to, and acceptance by, a journal’s editor. Because the journal invested time and labor in getting the MS reviewed and in making a decision and in providing editorial guidance, some publishers feel that they have made an investment in the work that became the post-print. Post-prints are thus in the borderland and publishers vary with respect to their tolerance of circulating post-prints. Still, post-prints have not benefited from typesetting or copy editing (or in the digital age, markup and other processes) and an author can “make” a post-print and circulate it as easily as she can make and circulate a pre-print. Again, post prints and their circulatory practices are pre-digital, it is just that digital contexts expand our ability to circulate (as well as our ability in a DIY mode to make post-prints look nice as text objects).
Wikipedia does, I think, a perfectly fine job too of explaining what a repository is. See “Institutional Repository.” Repositories can be built not only by “institutions” but also by funding agencies and by organizations seeking to serve a field or research area. Details abound online, but one is the SPARC “Repository Resources” page where a one paragraph overview is available.
The complexity in all of this is largely a result of the commercial and legal changes in which publishing unfolds. OA is a response to those changes but not a source of the problems. That said, caring about access issues does as you suggest mean that a scholar has to learn and do things that just handing over the manuscript and signing the forms does not entail.