Market mechanisms work best where goods are uniform (either individually or on aggregate average), their qualities are readily determined (or again tend to average out well), where the fixed costs of production are low and marginal costs of production high (relative to one another), and externalities, both positive and negative, are small relative to market price.

Information goods violate virtually all these assumptions.

● Quality is highly variable.

● Quality assessment is difficult, and often frustrated by other factors (e.g., pay-to-publish journals, "friendly" colleague peer reviews, discussed recently by Joerg Fliege at The Other Place).

● Quality isn't, and often cannot, be known in advance.

● Variance of individual instances is high enough that averages rarely suffice.

● Fixed costs of production are high, particularly for research, also to an extent for selection, review, and editing.

● Variable costs of production (e.g., publication) are low. In fact we're utilizing a system which was specifically created to reduce those costs still further, Tim Berners-Lee's World Wide Web, developed to transmit physics papers between CERN, SLAC, and other related facilities.

● Information goods typically have very high positive externalities -- they benefit those who don't directly consume them. Occasionally they have high negative externalities -- e.g., smallpox, "superflu", or weapons research.

There's the question of what does a content payment scheme provide? Of which the answer is, generally, "an incentive to create content". What matters isn't whether or not each individual information trasnfer is equitably priced, but whether, at some reasonable interval (e.g., at years' end) you've got sufficient compensation for authors, researchers, reporters, etc., to provide an adequate supply of information and entertainment.

When distribution was on physical media, printing and transactional sales were reasonably appropriate. With the price of reproduction* approaching zero, but nonzero fixed costs of production, there's an inherent conflict in the mechanisms which allow for price discovery in markets.