Tuesday, August 07, 2012

Open access – gold versus green

Recently, Andrew Adams contributed to the 'gold' vs. 'green' open access discussion and he wrote this on the LIBLICENSE list (edited for typos):
There are on the order of 10,000 research instutitions and more than ten times as many journals. Persudaing 10,000 institutions to adopt OA deposit mandates seems to me a quicker and more certain route to obtain OA than persuading 100,000 journals to go Gold (and finding more money to bribe them into it, it would appear – money which is going to continue to be demanded by them in perpetuity, not accepted as a transitional fee – there's nothing so permanent as a temporary measure). (Full message here.)
The LIBLICENSE list moderator would not post my response, so I'm giving it here:

10,000 research institutes means, in terms of Harnadian 'green' mandates, a need for 10,000 repositories; 100,000 journals (if there were so many; I've only ever heard numbers in the order of 20-25,000 [recently confirmed as in the order of 28K]) does not mean 100,000 publishers. Besides, there is no existential reason for institutions to have a repository and 'green' mandate. The fact that others have repositories and it doesn't have one itself does not harm a research institution in the same way that not being 'gold' (or at least having a 'gold' option) does existentially harm journals in an environment of more and more 'gold' journals.

As for costs, there are two things that seem to escape the attention of 'green' advocates (by which I mean those who see no place for 'gold' open access at this stage on the basis that 'green' would be a faster route to OA and would be cheaper):
  1. 'Green' fully depends on the prolongation of the subscription model. Without subscription revenues no journals, hence no peer-reviewed articles, hence nothing to self-archive but manuscripts, arXiv-style. (That would be fine by me, actually, with post-publication peer review mechanisms overlaying arXiv-oids). The cost of maintaining subscriptions is completely ignored by exclusively 'green' advocates, who always talk about 'green' costing next to nothing. They are talking about the *marginal* cost of 'green', and compare it to the *integral* cost of 'gold'.
  2. Exclusively 'green' advocates do not seem to understand that for 'gold' journals, publishers are not in any position to "demand money". They can only offer their services in exchange for a fee if those who would pay the fee are willing to pay it. That's known as 'competition', or as a 'functioning market'. By its very nature, it drives down prices. This in contrast to the monopoloid subscription market, a dysfunctional market, where the price drivers face upwards. Sure, some APC's increased since the early beginnings of 'gold' OA publishing, when 'gold' publishers found out they couldn't do it for amounts below their costs. But generally, the average APCs per 'gold' article are lower — much lower — than the average publisher revenues per subscription article. And this average per-article subscription price will still have to be coughed up in order to keep 'green' afloat.
Price-reducing mechanisms would even work faster if and when the denizens of the ivory tower were to reduce their culturalism and anglo-linguism that currently prevails, in which case we could rapidly see science publishing emerge in places like China, India, and other countries keen on establishing their place in a global market, competing on price. APCs could tumble. Some call this 'predatory gold OA publishing'. Few seem to realise that the 'prey' is the subscription model.

The recently published Finch Report expresses a preference for immediate, 'libre', open access, and sees 'gold' as more likely to be able to deliver that than 'green'. Meanwhile, 'green' is a way to deliver OA (albeit delayed and not libre) in cases where 'gold' is not feasible yet. That is an entirely sensible viewpoint, completely compatible with the letter – and I think also the spirit – of the Budapest Open Access Initiative (BOAI). Incidentally, referring to the BOAI is characterised as "fetishism" (sic) by Andrew Adams.

Comparing 'green' and 'gold' is almost, to borrow a phrase from Stevan Harnad, "comparing apples and orang-utans". The Finch report is not mistaken to see 'green' as (in the words of Michael Jubb) an "impoverished type of open access, with embargo periods, access only to an authors’ manuscript, without links and semantic enrichment; and severe limitations on the rights of use." After all, in the 'green' ID/OA scheme (ID = Immediate Deposit and OA meaning 'Optional Access" here) favoured by Harnad c.s., deposited articles may be made open if and when the publisher permits.

Besides, 'gold' implies also 'green' ('gold' articles can be deposited, without embargo or limits on use, anywhere, and by anyone), where 'green' does not imply 'gold'. A Venn diagram might look like this (below).

The Finch group has come to its conclusions because they have clearly learnt the lessons of the last decade. There is nothing — repeat: *nothing* — that prevents academics to eschew the services of "rent-seeking" (as Adams put it) publishers. They could easily self-organise (though I realise that both the words 'could' and 'easily' are probably misplaced). To expect publishers (for-profit and not-for-profit ones alike) to refuse providing services that academics are seeking from them is silly.

For the avoidance of doubt, I am not against 'green' OA (in spite of what some 'green'-only advocates assert), especially not where there is no other option. The choice is not so much for or against 'green' or 'gold', but emphatically for full, unimpeded open access, however it is delivered, as long as it is "permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself." You recognise this last phrase? Indeed, the precise wording of the BOAI.

Jan Velterop

Wednesday, August 01, 2012

The triumph of cloud cuckoo land over reality?

It should be abundantly clear that Open Access policies by Finch, RCUK, Wellcome Trust and many others are very important for the development of universal OA, in that they not only indicate practical ways of achieving it, but also signal to the scholarly community and the wider society interested in scientific knowledge and its advance that OA should be the norm.

The 'sin' that RCUK, Finch and the Wellcome Trust committed is that they didn't formulate their policies according to strict Harnadian orthodoxy. It's not that they forbid Harnadian OA (a.k.a. 'green'), oh no. It is that they see the 'gold' route to OA as worthy of support as well. Harnad, as ultimate arbiter of Harnadian OA (though he has acolytes), would like to see funder and institutional OA policies focus entirely and only on Harnadian OA, and would want them, to all intents and purposed, forbid the 'gold' route. In Harnad's view, the 'gold' route comes into play (as 'downsized gold', whatever that means) only once all scholarly journal literature is OA according to Harnadian rules. These rules are quite specific:
  • articles must be published in peer-reviewed subscription journals; 
  • institutions must mandate their subsequent deposit in an institutional repository (not, for instance in a global subject repository); 
  • there must be no insistence on OA immediately upon publication (his big idea is ID/OA — Institutional Deposit / Optional [sic] Access); 
  • here must be no insistence on CC-BY or equivalent (which would make re-use and text-mining possible — OA in his view should just be ocular access, not machine-access).
It must be difficult to comply with these rules, and seeing his recent applause, subsequently followed by withdrawal of support, for the RCUK policy, even Harnad himself finds it difficult to assess whether his rules are 'properly' adhered to. It also seems as if his main focus is not OA but mandated deposit in institutional repositories. Probably hoping that that will eventually lead to OA. He would like to see 'gold' OA — OA at source — considered only if and when it is "downsized Gold OA, once Green OA has prevailed globally, making subscriptions unsustainable and forcing journals to downsize." It is the equivalent of opening the parachute only a split second before hitting the ground. It would be the triumph of a dogmatically serial process over a pragmatically parallel one. The triumph of cloud cuckoo land over reality.

Open Access is more than worth having. Different, complementary, ways help achieve it. There are many roads leading to Rome.

Jan Velterop
OA advocate