TRS asks (by the way, TRS, please publish your statements in a reusable format, so I can quote your text accurately; printing it on paper, then scanning it and publishing the pdf containing the scan is much more expensive: I waste time retyping your text, somebody’s wasting time by manipulating such non-semantic objects; the modern alternative is trivial for a long time now, please follow the example of FRS next time, or people like me will neglect quoting your phrases):
“Are the various alternative (publishing) models appropriate for all disciplines?”
Yes, they are. The (scholarly) publishing entered in the digital era, and not viceversa. The media has changed to digital, lowering the cost of processing information (locating, reading, reusing, authoring). This applies to anything communication-like.
Mathematics, physics and chemistry communities would benefit first, because the alternative models leave open the possibility of adding semantic depth to the scientific documents (compare the ‘production quality’ publishing from traditional publishers of physics and the bare prototype available in the XML version of Living Reviews: try to copy/paste a math formula to reuse and modify it online.)
“Would researchers without a grant or research position or from the developing world be able to afford to publish under an author-pays model?”
I explained below there is no need of an author-pays model. The author-pays model is another name for closed-access: there is no justification for significant supplementary (or equivalent with traditional publishing) costs in open-access.
Whether you block/delay the access at the reading or authoring stage, you still block/delay it. The author-pays model has yet to justify why anybody has to pay more than the public did already (through the funding agents) for the creation and the administration of the research result.
Author-pays and traditional, costly and closed, models of publishing are almost equivalent. In fact, the author-pays model is worse, the next TRS concern points it out:
“Do the alternative models of publication provide the same level of quality assurance and peer-review as the established model? For example, under an author-pays model will there be a pressure to publish authors who can pay…”
Let me state it: the author-pays model is not an alternative model, it is not open-access because it preserves the bottleneck of costs somewhere on the flow of scientific information (it’s just hiding it under a different name). So, at least in what concerns author-pays models, TRS is rigthly concerned with the consequences: as their example shows already, besides preserving the publishing costs pressure, the author-pays model skewes the quality assurance process by introducing a condition which has nothing to do with it (the pay capability of the author).
The author (that is, the public funding his work) pays implicitly when creating the final form of the scientific work (article, book, whatever). So, any ‘alternative’ model which involves further significant costs for publishing the results are not, in fact, alternative at all. Or, maybe, what’s alternative about them is just a different way to unjustifiably spend the same money current publishing models require.
“Where repositories contain both pre- and post-print articles is there an effective way of distinguishing between multiple versions of the same paper?”
Yes, for example, using the bare steps enumerated below. Versioning digital articles is as simple as adding the publishing date. Qualifying them as peer-reviewed means just setting an attribute of the digital document (say, ‘peer-review’ = ‘in progress’ | ‘accepted’ | ‘rejected’ | ‘none’). Peer-review should happen using references to the stored article: author writes a version, sends it to the IR, and submits a reference to it to reviewers (the peer-review attribute is set to ‘none’, the same for those which are simply deposited in the repository), he gets an answer, adjusts the work, resubmits the new version to the IR, with an updated date/version attribute, with the peer-review attribute set to ‘in-progress”, and a reference to it to the review panel. What’s so complicated about it?
“Is there sufficient funding to ensure the survival of the institutional repositories in the long term?”
An easy mirror question comes to mind: is there sufficient funding to ensure the survival of quality research in the long term under the current publishing model?
There’s a serious answer though: the institutional repositories (IR) are part of the definition of a modern research institution, as such, they are an integral part of the research activity (similar to the institute’s having a website providing worldwide access to virtual visitors). The IRs should have the same status as the current institutional libraries (in fact they should be the same). This question is equivalent with asking: Is there sufficient funding to ensure we do research?
“Under a self-archiving model, how will peer review be organized if, as some fear, journals go out of business? Is the brand of the journal important, and if so, what will replace it?”
Peer-review is independent of the self-archiving (or publishing) method used, so it will have the freedom to reorganize optimally without constraints external to it. As long as IRs exist and provide access to the works, who cares about a particular form of a journal?
That kind of business can exist, but it should provide added-value to the content available in the IRs to deserve any attention or the payment of anybody.
Also, given the IRs exist, any group of scientists can edit a virtual journal, a collection of papers focused on something. But my estimate is that, when IRs will be common, every author will be an editor of the information s/he gathers, and will be able to easily publish a directory of references to relevant and accessible works which inspired him/her, or helped with his/her subject.
That’s a feature of diachronic publishing (a concept I tried to sketch in 1999).
Branding? Heh. I wouldn’t spend a dime on that. The research result, as a digital document, contains the affiliation information, which can be made visible at the webmaster’s choice. That is a natural branding (the natural replacement). Research being different from commerce, one shouldn’t search for an article because of (or influenced by) its branding, right? Better save those millions for scientific exchange activities, the scholarly search engines will automatically provide the visibility you deserve without any dedicated expenses, if open-access is used.
“Is there an inherent problem with current model and do the proposed alternatives address it?”
Yes, in fact there are more than one inherent problems:
a. The public pays twice for the same thing: once for the doing the research work, and twice for accessing it. The profits go to people who have nothing to do with the research work or peer-reviewing it.
There is an expense incurred by publishing on paper, but that has nothing to do with the availability of research results in the digital era.
A modern citizen deserves this:
- I am happy with retrieving the work from the IR, it shouldn’t cost me anything because I paid for it already,
- I also wish to pay a publisher to have it nicely printed on shiny paper for me, branded, marketed and everything
The digital era makes, technically, the second item an option.
The latter is entirely a matter of individual taste (I guess it’s nothing wrong with an individual considering buying Physica C because its paper published instance is sexy), it has nothing to do with the research itself or with the shape of scholarly communication.
b. Open-access naturally encourages citing content-accessible articles, so if you keep publishing under the traditional model, your results might get ignored artificially and lose relevance in time, exactly because there was an artificial barrier in accessing it.
Overall, it’s obvious that open-access offers more content to the scientist, gives him more freedom of choice and offers a much greater chance to discover unusual ideas or relations between previously unrelated theoretical models. Moreover, open-access leaves the door open to adding semantic depth to the documents deposited in IRs and enables a true competition among publishers: which collection of references to IR stored articles is better organized, or more relevant for a certain domain/task, or allows a better reuse/representation of the content on a particular media, or brings more functionality to the desktop of a researcher?
If institutes managing their own IRs have problems with funding their development (although I don’t think there is any serious issue with the costs of self-archiving), they should charge any for-profit publisher with a reasonable fee for commercially republishing content available in the IR.
I’ll stop here: even with no current funding, I still have some real and exciting work to do on Hermes (an open-access tool itself).