Open, Closed, or Clopen Access?

Mr. Vardi said, "ACM's modest publication revenues first go to cover ACM's publication costs that go beyond print costs to include the cost of online distribution and preservation, and then to support the rest of ACM activities."

I am a librarian, I view this distribution of funds as a problem. A library's primary mission is to gather, organize, and disseminate scholarly information from a variety of fields. To the extent that library subscription dollars exceed the actual cost of producing journal content to "support the rest of ACM activities" goes beyond the mission of the library.

As a member of ACM, I appreciate the the other activities of our society, but as a librarian I strongly believe that the publishing initiatives of scholarly and professional societies should solely support themselves and not be sources of income for other parts of the society.

I appreciate the sentiment of "I am willing to pay actual costs but nothing more", but that is not the way complex organizations work. For example, successful university presses often make a profit on science and engineering books and use this profit to subsidize publishing in the Humanities. Universities often make a profit on continuing education and use it to subsidize less profitable operations, such as libraries. University libraries simply would not exist without institutional subsidies.

Open Access is a blanket policy. What if we took a more targeted approach by placing emphasis on community engagement rather solely on content?

Suppose we, as ACM members, became really engaged by an article and that CACM was set-up to take notice of this activity. At some point we, as ACM members, could fund the opening of that article.

I'm not suggesting that this be done in "micro-payment" fashion. The futility of that mechanism has been convincingly explained by Clay Shirky. I have some ideas about alternatives but am sure that we, as a group of technical professionals, can find a solution that addresses real publishing costs while furthering the first goal of scholarly publication: dissemination of knowledge.

A university library can never be confused with a profit center. With the exception of overdue fines and replacement book fees used to instill a sense of importance for the shared resources we steward or cost recovery efforts on things like photocopying and printing, libraries depend on subsidy from the host institution. I've seen many ways to generate this subsidy: flat, off-the-top allotments from the university, "taxes" on departments, etc. The library could not serve its mission without those funds.

I've also been involved in drafting strategic planning documents at a number of academic libraries, and in the process of doing so have reviewed a dozen others. I don't recall once proposing or coming across a strategic goal of "supporting scholarly activities through subsidized journal pricing." That is not a mission that academic libraries serve.

To be fair, and stating that I have not been directly involved in negotiations to acquire society content, I've come to understand that ACM is one of the "good guys" -- that ACM doesn't have a strong cross-subsidization of other activities from publishing revenues. I bring out this concern about cross-subsidization, however, lest it become an even more prevalent practice.

Take for example your overdue fines. Are you telling me that overdue fines only recover the cost of overdue books?

Cross subsidies are prevalent in all organizations. That is, for example, how libraries are being funded, rather than charge readers directly to recover costs.
I can guarantee you that no publisher, for or not for profit charges you only to recover costs.

What you should consider is the price-value proposition of ACM publications. I am certain that it is one of the best bargains in your library!

"ACM does charge a price for its publications, but this price is very reasonable." *Thank you!* I agree whole heartedly. Any one who thinks otherwise should compare the quality and affordability of the ACM Digital Library with other similar libraries, one of which charges four times the price for less access.

Prof Vardi presents what I think was a reasonably balanced and thoughtful argument for the status quo. However, I believe that ACM has several opportunities to gain benefits from more aggressively pursuing gold OA.

One point is that the distinction between "reader pays" and "author pays" is overly simplistic. As Vardi notes, libraries -- and the institutions that they are associated with -- are typically the source of funding for most journals with traditional subscription models, not the readers themselves. The flip side is that many OA journals without author fees are funded by their host institutions, and that the "author pays" model is largely subsidized by funding agencies. Compared with disciplines like English, CS with its high proportion of grant-funded authors is in a very good position to find journal funding without turning to either individual readers or individual authors.

Another point is that true open access brings with it huge benefits, since even a modest subscription price -- and Vardi is right that ACM publications typically have modest prices -- creates significant barriers. The barriers are most pronounced for the lay public and for potential readers in have-not countries. If ACM cares about reaching the non-specialist public or 3rd world readers, it must take seriously the possibility that this "clopen" model is interfering with that outreach. Personally, I worry about declining enrollments in CS courses and majors in the US, and chalk that up in part to our failure to aggressively reach out to the non-specialist public. Even within the academy there is emerging research that on balance shows a citation advantage for open access, so authors who seek impact would do well to submit their work to OA journals, and journals would do well to cater to those authors. True OA also provides data for text mining and more robust experiments in indexing and information retrieval.

The third point is that ACM has a unique opportunity, since it has a wide range of publications and can easily pursue different strategies for different publications. The SIG publications I receive are not the same as Transactions, and neither are the same as the general-interest CACM magazine. Perhaps the next step ought to be for ACM to provide a survey of its publications and their varied approaches to OA.

The numbers and quality of OA journals are growing rapidly. The Directory of Open Access Journals (doaj.org) currently lists 150 gold OA journals in Computer Science, plus others in related areas ranging from applied maths to computer games. Not one is published by US-ACM. Maybe an experiment is in order.

JQ Johnson