Powered by glolg
Display Preferences Most Recent Entries Chatterbox Blog Links Site Statistics Category Tags About Me, Myself and Gilbert XML RSS Feed
Sunday, Sep 16, 2018 - 21:48 SGT
Posted By: Gilbert

On Open Access

Earlier this month, eleven European countries (including Britain, for now) have committed to "Plan S": all scholarly work resulting from national agency grants must be published in freely-accessible open-access journals. Unsurprisingly, major academic publishers weren't charmed, with those in charge of Nature and Science stating that the move would "undermine the whole research publishing system" and "would not support high-quality peer-review, publication and dissemination" etc, without much in the way of elaboration. Elsevier's take was that "if you think information shouldn't cost anything, go to Wikipedia", perhaps not realising what happened to Encarta and the likes of the Encyclopedia Britannica.

It has been noted here last year, it happens, that the current academic publishing system makes little sense - scholars produce research funded by taxpayer money, submit papers to journals (oft edited by unpaid scholars), which are then reviewed (again unpaid) by yet more scholars. Upon acceptance, the publishing house arranges for proofreading, typesetting and printing (increasingly optional, these days), for which it reaps almost all the monetary benefits. The end product is then sold back - often to the very same community of scholars that produced the work - at monopoly prices. And people wonder why I doubt the efficacy of academic economic theorists when they operate in the real economy.

This overriding sense of injustice - taxpayers paid for the research, why should some private organizations charge them again for perusing it? - had been rankling for some time, alongside the ethos that knowledge should be free in the first place, and stakeholders have reacted in a variety of ways. One option was of course to simply make the papers available - they're just PDF files, after all! This was the tack that Reddit co-founder Aaron Swartz took at MIT, for which he hanged himself after being confronted with 35 years in jail (compare kiddy diddlers in more tightly-knit organizations, but more on that next time). His spiritual successor was probably Alexandra Elbakyan (now in hiding), who founded *ahem* Sci-Hub.


Everyone does it, nobody admits it, like so many other things
(Source: arstechnica.com)


They might not have bothered for some fields; mathematics and its closest offshoots - such as physics and theoretical computer science - have always had a robust preprint culture, which has expressed itself as arXiv. Basically, the good stuff's expected to be there, though it can admittedly be challenging to sort through given the sheer volume of papers (but for which there are services such as Karpathy's Sanity Preserver).

Academics in these fields also appear somewhat less beholden to the lure of journal status, perhaps exemplified by Perelman not bothering to submit his proof of Poincaré conjecture to any journal after posting it on arXiv, because its truth was all that mattered. Closer to my own area, the editorial board of Kluwer's Machine Learning journal up and resigned to found the open-access Journal of Machine Learning Research in 2000, which has arguably long eclipsed its (still-extant) forebear because, in the end, it's the academics' content and regard that makes or breaks a journal. Nature's upcoming Machine Intelligence is facing a boycott from many prominent computer scientists, for much the same reasons.

This exercise of academic power to punish non open-access outlets has not found as much purchase in other areas of study, despite occasional calls. Nobel-winning cell biologist Randy Schekman, for example, is boycotting the "Big Three" of Nature, Science and Cell, but it is hard to overlook the fact that his four most-cited papers, and eight of his ten most-cited, were in these very "glamour mags". Aspiring young scientists might thus quite fairly question whether this is good advice for their career - a recognized Nobel winner's not going to have too much trouble acquiring reasonable funding either way, but how's a new kid on the block going to make a reputation?

[N.B. By the way, I am not entirely persuaded by Schekman's argument over "luxury" journals, even if they tend to have relatively more irreproducible research and retracted claims; the reason is that, as with most people, academics are as a group desirous of exclusivity and distinguishment. Given a venue with a restricted publication rate (and higher impact factor), and another venue with an unrestricted one, it is only natural that participants first shoot for the former to signal the value of their product, even if the curation process is noisy. This is observed with the usual submission order of e.g. Nature → Nature Comms (already fully open-access) → Scientific Reports, and would persist even if the current top journals were to suddenly disappear - new ones would just spring up in their place.]

Anyway, the state of academic publishing has long been regarded as a Prisoner's Dilemma (for the academics, at least). So this telling goes, academics (in the STEM fields particularly) would actually prefer open access and widespread dissemination of their work, because it's not like there's money in it elsewise. The problem, however, is that current top journals are mostly non (or at best, delayed) open access, and they have to publish in these very journals to get exposure and be credited by the admins for promotion and tenure. It would be ideal, then, were all academics (or minimally, enough of them) to shift to open access journals. However, without coordination, those who move "lose out" in terms of placing their work in less-competitive venues, while those who don't "win" in terms of easier publication in the still higher-prestige non open-access venues - hence the dilemma.

The pan-European Plan S can then be recognized as a bid to break that impasse through overt coordination. If the plan goes through as-is, a fair chunk of high-quality Euro work would immediately be out-of-bounds to CNS and other prestige outlets, with the recepients being open-access journals. Admins, who might otherwise be content with the old assessment metrics, would then be forced to accept that NewOpenAccess Journal is high-quality, given that it is indeed putting out some of the best work, which in turn attracts academics who aren't bound to Plan S, which further reinforces the open-access journal's status, in a virtuous circle.

What's more interesting is how the existing big non open-access journals will respond. It could be noted that if revenue is their concern, it may in theory be possible to maintain that while moving to open-access, by just charging upfront (i.e. the authors) upon acceptance, rather than downstream (i.e. libraries and private readers). Author costs would then be factored into grant funding (unfunded work is near non-existent in modern Big Science), with library funding withdrawn in tandem (since subscriptions are no longer required). The question, of course, is whether this model can preserve existing supernormal levels of profit.

We might consider arguments against open access, given that the AAAS for one has not seen fit to go into detail, and in truth not many of them are convincing. For example, one enduring argument is that academic papers can be misinterpreted in dangerous ways, to which the immediate response is that the news media already does that anyway, with the next observation being that the existing model suggests that such misuse is acceptable, as long as one is willing and able to pay (and which also smells of past religious proscriptions towards knowledge). The better objections are to me field-specific, as in some humanities disciplines expecting monographs instead, or relying on subscriptions to support their learned societies. Still, he who pays the piper should be able to call the tune.

One other worthwhile objection is that open-access journals that charge authors upfront have an added incentive to be predatory, since the more papers they accept, the more they earn; this is not the case under the existing consumer-pays model, since libraries would have no reason to subscribe to a dodgy title. However, it isn't as if this sort of thing isn't going on already, Plan S or no Plan S.

So, how will the big publishers react? They could cave, of course, by switching wholesale to the open access model (with the hybrid option of having only some papers being free explicitly disallowed by Plan S). That said, there is another possibility, which is said journals instead standing their ground and relying on non-European submissions - and I can see Asian (read: mostly Chinese) scientists being more than happy to pick up the slack. The decider, as I see it, would be whether Europe can rope the Americans into Plan S; if so, I can't see a journal holdout being tenable. If not - which is far from unlikely, given how motivated the publishers' lobby can be - we may be looking at a geographic split in the academic publication landscape...



comments (0) - email - share - print - direct link
trackbacks (0) - trackback url


Next: And We're Back


Related Posts:
Reviewing In Review
"Science" & Facts
Corona Update (Part II)
Of Logic, Looting, Twitter & Disinformation (Part I)
Of Logic, Looting, Twitter & Disinformation (Part II)
Back to top

Copyright © 2006-2025 GLYS. All Rights Reserved.