Right. But remember that we’re talking about the essential nature of information as a good. If you write an article, who has the right to license that article according to one of those schemes (or any other)? You do, of course. Does anyone else? No. This means that the information itself is not a public good – it is your intellectual property. To say that information is a public good simply because you have the option of donating it to the public makes no sense – it’s like saying that a car or a banana is a public good.

“As to whether the ‘net is non-rivalrous…certainly bandwidth is an issue. Do you sincerely believe that there are serious bandwidth issues in the promulgation of what amount to, effectively, text files?”

Yes; if it were otherwise, everyone would have free access to all the text they need. But let’s assume that there are no such issues – are you suggesting that information is a public good as long as it’s in text format?

“I also feel that the upcoming generation of scholars will have a very different expectation of their own publications, and this will lead to an increase of personal publication and personal archival of works. This means that authors are underwriting the ‘cost’ of the production and dissemination of their works…but I maintain that to most scholars, this ‘cost’ is negligible.”

How do you figure? Internet or no internet, it takes me as much time and effort to write a CD review now as it did 13 years ago. I’m no more willing to do it for free now than I was then. Do you think my attitude is an anomaly among people who create information?

“With the possible exception of the hard sciences, the vast, vast majority of scholars already underwrite the cost of their publications to a large degree.”

They do? Who are these scholars, and how to they earn a living?

One would be your point above that “Nor is it true that information is non-excludable. Information is, in fact, eminently excludable: if I write a paper, it is completely up to me whether I share it with anyone. And the sharing process can be stopped at any point: if I decide to make a copy for one person, there is nothing to stop that person from declining to do the same for others.”

The entire point of most alternate licensing schemes is so that the scenario above is not the usual course of events. The Creative Commons licenses, the GPL, and most other “open” license types do not allow one to recind said license (with the CC licenses, technically you could. However any copy of the information with the old license still attached is still legal.). The point would be that it isn’t legally necesary for the information flow to stop at any single point…and if you were to release your information (article/essay/whatever) “into the wild” as it were, it isn’t likely that it would stop with just the one person. Of _course_ I can hand someone a copy of my info…that is very different than putting it onto a publically available website.

As to whether the ‘net is non-rivalrous…certainly bandwidth is an issue. Do you sincerely believe that there are serious bandwidth issues in the promulgation of what amount to, effectively, text files? I suppose I’m spoiled from being on a university server. I would imagine that most, if not all, scholarly research would have access to such a resource. In addition, there are many, many, many ways to technologically distribute load (mirrors, peer-to-peer systems, etc). To say that the ‘net is rivalrous is much like claiming that clean air or water is rivalrous. Technically, you are correct. Practically, I feel otherwise.

I suppose that when it comes right down to it, my position can be stated something roughly like this:

I strongly believe that Open Access is and should be the future of scholarly communication (as well as other aspects of information and media exchange). The digital revolution removes a large number of barriers from this process, or lowers the barriers of entry to the point where they are surmountable by nearly anyone. I see Open Access as more of an inevitability than anything…the high cost of print publications will be their undoing, while online publication will continue to flourish in the face of this decline. I also feel that the upcoming generation of scholars will have a very different expectation of their own publications, and this will lead to an increase of personal publication and personal archival of works. This means that authors are underwriting the “cost” of the production and dissemination of their works…but I maintain that to most scholars, this “cost” is negligible. With the possible exception of the hard sciences, the vast, vast majority of scholars already underwrite the cost of their publications to a large degree. Open Access just means that more people have access to their work.

Of course it does — in the same sense that my car would qualify as a public good if I were to donate it to the city. But my car is a private good until I (the owner) make it otherwise. Similarly, information that I create is not a public good unless I choose to make it so.

And so they do, of course. Authors are free now (and always have been) to release everything they write to the public domain (thus making their work a public good), and they are free to absorb the cost of creating and distributing their own work. Open access becomes a real issue only when an author wants someone _else_ to take on the costs of publication and distribution. In that case, the publisher/distributor must either absorb those costs itself, or find a way to recover them. Open access is wonderful from the end-user’s point of view — but how can information be provided on an open-access basis while still allowing the middleman to recoup his costs (and maybe even, gulp, realize a reasonable profit?). That’s the big question.

Regarding the Wikipedia definition of “public good”: The Wikipedia is a wonderful resource, but it is not without flaws, and this definition demonstrates one of them. It cites two common characteristics of a public good, and treats those characteristics as if they constituted a fundamental definition (it’s a bit like defining “dog” as “mammalian quadruped” — true, but not really a definition of the term). A more basic definition is provided by Alan Gilpin in the second edition of his _Dictionary of Economic Terms_ (New York: Philosophical Library, 1970): “A good or service provided for the community by the government or local authority, e.g. education, public health services, libraries, theatres, museums, etc.” These goods and services are somewhat (though by no means completely) non-rivalrous and non-excludable, but all are most definitely public.

But even if we accept Wikipedia’s definition as complete, information would still not fit. Use of information, especially in the online environment, may _feel_ non-rivalrous, but that’s an illusion. In reality, even online information is rivalrous: More users put more strain on the resource. (While my local Web server can comfortably accommodate many thousands of simultaneous users, it cannot accommodate many millions.) It’s true that I can make copies and distribute them without undermining the utility of the information, that’s not really enough to make it a non-rival resource — after all, there’s a limit to the number of copies I can make, and theoretically no limit to the number of people who might like copies. Nor is it true that information is non-excludable. Information is, in fact, eminently excludable: if I write a paper, it is completely up to me whether I share it with anyone. And the sharing process can be stopped at any point: if I decide to make a copy for one person, there is nothing to stop that person from declining to do the same for others.

Best,
Rick Anderson