The network effect causes a good or service to have a value to a potential customer dependent on the number of customers already owning that good or using that service. Ergo, it means that the total value of a good or service that possesses a network effect is roughly proportional to the square of the number of customers already owning that good or using that service.
One consequence of a network effect is that the purchase of a good by one individual indirectly benefits others who own the good - for example by purchasing a telephone a person makes other telephones more useful. This type of side-effect in a transaction is known as an externality in economics, and externalities arising from network effects are known as network externalities.
This recent thread has compelled me to author the following journal entry on the evils of plurality voting and the two party system...
It is a common phenomenon with our current election system (I'm in the US) that, regardless of the number of candidates, the vast majority of the vote is split between the Republican and Democratic candidates. Frequently, a third party candidate gets enough votes to alter the outcome of the election without receiving nearly enough votes to have a chance at winning.
Voters tend not to vote for third party candidates for fear of "throwing their vote away". This is a bad thing: voters don't vote for the candidate the really want, they vote instead for the major party candidate they dislike the least, whom they think might actually win. In other words, the voter's perception of a candidates ability to get votes influences the number of votes they actually get. A vicious cycle.
So, what is the result of all this? Well, we have basically two options at election time. Or, in other words, our individual contributions to the democratic process each ammount to a single bit. What, then, does our political process look like? One dimensional. You're either left wing, or right wing. There are two solutions to every problem: "the free market will solve everything", or "more government will solve everything". Issues (or their solutions, such as the open source "let people solve their own problems" solution) that are orthogonal to this political worldview get ignored (Electronic voting fraud? Reducing copyright duration?). People with complicated and nuanced political ideologies get typecast into one box or the other (i.e. "Lawrence Lessig is just another leftist"). This is much easier than actually understanding issues. It's convenient for the news media, which prefers to portray things as a simple "us versus them" conflict (they try to do this without being informative enough to offend or bore anyone -- this poor reporting is often interpretted by each side of the percieved one-dimensional political spectrum as a conspiracy by the other side), which in many ways resembles sports news coverage (all entertainment, no relevance).
What, you might ask, is my solution? Well, there's quite a few ways to elect a public official (plurality voting, borda count, instant runoff, etc...), my favorite is approval voting, whereby voters can vote for as many candidates as they like. This would allow a voter to vote for, say, both George Bush sr. and Ross Perot, or Ralph Nader and Al Gore. The voter is given many more options than they would have with plurality. It's not perfect (there is no way to distinguish between a wonderful candidate, and one you merely tolerate), but it's a very simple system, and the voter is given many more choices than they would have otherwise. chances are, two of the last three elections would have turned out different.
I don't expect a president to be elected this way any time soon, but maybe we can start small (elections for senators, mayors, high school student body presidents, slashdot polls).
I'm still waiting for the advent of the computer science groupie.