Professor Bainbridge discusses an attempt to use "transactional costs" analysis to demonstrate why parties are less important in an Internet-enabled world than otherwise. The good professor argues that costs of informed decision are going to be nonzero, regardless of the ease of information search, because processing the information still requires expenditure of a limited resource - cognitive effort. He offers the example of voting for president versus voting for sewer commissioner, and how the perceived need to research a presidential candidate outweighs the transactional cost of thinking about it, while that same need in the case of the sewer commissioner will, for most voters, not tip the scales. He further notes that party affiliation represents a shorthand that allows decision without significant intellectual expenditure of resources. He calls it an application of "heuristics", but really he's arguing for the usefulness of habit in the form of party allegiance.
He makes a good point - I find it pretty hard to educate myself on the issues, and Pennsylvania isn't a initiative-heavy state like, say California. But I think he overruns the original argument, that the reduction of education barriers reduces, if not eliminates, organizational costs. It means that parties have to dedicate a smaller proportion of resources to voter education than they once did. The Dean campaign, in particular, has been able to divert a significant proportion of activist resources from educational efforts ("getting out the message") to financing and other organizational efforts that normally would be handled by the party apparatus. It weakens the tie between the campaign and the party, and offers a greater possibility for shake-and-bake campaign-generated party organizations, or transient, short-term parties.
The professor's cited heuristic or allegiance-based force is a centripetal force working against the centrifugal force of the new technology - it means that the old parties aren't necessarily, automatically toast. But the adding of a new force to the existing diagram certainly means a shift in the centre of gravity, I would think.