-- snip --
I think Professor De Long is wrong - there is nothing in the nature of the new commodity - information - that will prevent it from becoming monopolized by a few corporations. Professors De Long errs because his argument rests on neo-classical mantra of "market properties" of goods (excludability and rivalry), while ignoring effects of social-political-economic institutions on economic behavior.
The argument that the 'market properties' of goods - rivalry (two people cannot eat the same cookie) and excludability (non-payers cannot eat a cookie) - determine the institutional arrangement of that good production and delivery is a neo-classical mantra repeated ad nauseam, and as most of the neo-classcial stuff - it does not explain much. It is so, because these two properties alone cannot explain why different institutional arrangements of th eproduction/delivery of the same good exist. An that inability stems from having the cause-effect direction all wrong.
Roadways are a good illustration of that point. Is a roadways a rival or a non-rival good? Is it exludable or non-exludable? The fact of the matter is - they are both. New Jersey Turnpike is excludable, because you cannot enter it without a valid ticket - I-95 is not because you can enter whether you paid your taxes or not. What differentiates these two stretches of roadway (other than being souht and north of NYC) is the institutional arrangement of the delivery - turnpike is administered by the NJTPK Authority that is empowered by the state of New Jersey to collect toll and exclude the non-payers - whereas I-95 (and most public roads) are adminsitered by respective state transportation authorities that do not have the authority to exclude non-payers.
It follows that that instiutional arrangement of production/delivery dterminies excludability/non-exlcudability, rather than the other way around. Stated diffrently, the road construction technology changed quite dramatically for th epast hundred or so years - but the technology itself had little effect on the "market property" (excludability) of the roads themesleves. It is the institutional arrangement, especially the state, that does. So the institution is the cause, and marker property (exludablity/nonexludablity) is the effect - not the other way around as the neoclasscial mantra has it.
How about rivalry? This whole argument thrives on the semantics, namely double meaning of the word "same." "Same" can mean literally the same (e.g. one physical cookie) or virtually the same (e.g. many cookies virtually identical to one another) or "similar". Neo-classcial use the word "same" in the first meaning when talking about rival good, and "same" in the second meaning when talking about non-rival goods. The point is, however, that in economic behavior it is only the second meaning - "similar" that is appropriate.
To illustrate that point. When I buy a cookie at a bakery shop, I do NOT want the same cookie the person before me bought - I want a "similar" one. Therefore, no rivaly between me and other customers exist unless - and that is an important qualification - there is sufficient suply of "virtually the same" or similar products. The only exception to that rule are goods that are unique, sui generis, - such as authentic works of art. There is only one Mona Lisa, and that is a truly rival good in the first meaning of the word. But art deals are not your typical economic transactions either.
It thus follows, that the second market property rivalry/nonrivalry is but a function of supply of similar goods - and that supply, in turn is the function of institutional arrangement of production/delivery. In other words, th esupplier may limit the supply of a good to increase its rivalry and thus social-economic value. Roadways and "limited edition" items are cases in point.
So the bottom line is that, again, it is the institutional arrangement of production/delivery that determines the market property of a good (rivalry/nonrivalry), not the other way around as the neo-classcial mantra has it.
It thus follows that technological advances alone will change little in th emarket properties of gooods, unless the instutional arrangement of their production and delivery changes. That holds for information goods as well.
Information can be made non-rival and non-exlcudable (i.e. a public good) or rival and exludable by intellectual property righths and the institution of the state that enforces them, as well as the encryption codes and institutions that require them. Thus, Windows is excludable, Linux is not.
Moreover, the encryption codes in the software and hardware (cf. pentium 3) can make software rival as well - i.e. a copy of software may not run on a machine with a processor id code that is different form that specified at the purchase of that software.
In short, production and distribution of information is subject to the same institutional arrangements as production and distribution of any other merchandise. If anything, proprietary technologies and intellectual property rights enforced by the sate can make information even more monopolistic (i.e. non-public) than any of the hitherto known tangible goods. You can take a mechanical device apart and replicate its design thus brealing the monopoly - i.e. use the "reconctructed" device without buying the original. However, breaking the enecryption codes may no be possible for a very simple reason - the id code may be required to obtain all information age services, logging in to a network, buying a product, or even running softwate on 'authorised" (pentium 3 equipped) machine.
wojtek