This post was largely inspired by some overly-simple “publishers are evil” comments I saw from friends inspired by the Swartz news coverage:
Reasons why I don’t think Aaron Swartz’s massive downloads from JSTOR were a good idea in terms of forwarding the broader goal of promoting more openly accessible work in academia.
Assumptions: I assume that “more openly accessible” includes material that is both open access and material that is reasonably priced to promote a rich and diverse academic publishing environment that includes more publishers and more competition among publishers.
1. He hid, and tried hard to hide- It was not a real civil disobedience effort as per the classic MLK definition of civil disobedience which requires that the protester accept any standing punishment for the action.
2. Scholarly communications is in a crisis with profit maximizing publishers seeking to extract rents from academic libraries. JSTOR however, is not one of the problem publishers. JSTOR is not even a publisher. Incorrect target. Some publishers have their own distribution platforms such as Elsevier’s Science Direct. See Bergstrom’s work on most expensive journals.
3. He inconvenienced users and librarians at MIT causing server outages at MIT and JSTOR and requiring MIT users to go through a more onerous login process.
4. JSTOR, as a third party platform, does not contain “top priority” open access material such as very recent medical or agricultural articles of importance in developing nations. In fact, journals often only let JSTOR have issues after a few years have passed in order to protect subscriptions (aka “the moving wall). Again, wrong target.
5. It is not true that downloading a large portion of JSTOR “it’s like trying to put someone in jail for allegedly checking too many books out of the library.” Libraries own books, they license access to articles on 3rd party platforms. In a licensing situation use rights are governed by a license that pretty much always prohibits end users from “systematically downloading” content. The license requires the library to take action against any violations (e.g. systematic downloading) or lose access to the material with no refund.
(As a side note, libraries could in fact have policies about the maximum number of books one could check out so allow other members of the public access )
6. JSTOR, as a distribution platform, relies on contracts with publishers to get journal articles. They pay publishers for the right to host their journal articles. JSTOR does not get them for free!
7. We want journals in JSTOR! If publishers perceive that JSTOR cannot provide security for their articles, they may pull out from the JSTOR program. Several publishers have threatened to pull out when they realized they could make more money selling their archives (aka “backfiles”) directly. No doubt they would price it higher. Publishers pulling out of JSTOR negatively impacts smaller colleges with smaller budgets that rely on third party platform services like JSTOR (see also Academic Search Elite etc.) to provide their students a wide array of journals to use while minimizing the need to negotiate and contract with hundreds of different publishers. So, Swartz may have harmed JSTOR, but probably did not influence publishers to make their material accessible for more reasonable fees at all.
In sum, I believe Swart’s downloading will have zero impact on pricing debates at publishing houses that actually control the end price that libraries pay or how openly accessible an article is to end users = not a useful enterprise.
While I take the Society for Scholarly Publishing’s word on any access issues with a grain of salt, I agree with their take on this one.
No doubt more interesting information will emerge when the court case starts this fall.
My opinions here do not extend to Swartz’s work on government information.
** Further thoughts **
What if he was doing it for research to crunch a large text corpus and not for civil disobedience purposes? Ok, well then did he request to use the JSTOR Data for Research Program? Publishers and 3rd party platforms now routinely field requests for crunching across their data sets.
Could the JSTOR DfR program not accommodate his needs?
I guess the court case in Sept. will reveal all.