From Benkler:
In Debian, Coleman describes how the continuous evocation of TINC (there is no cabal) and the black humor about the presence of a cabal are used as a continuous cultural technique to render visible the existence of a meritocratic elite, underscore the risk it poses to the community, and reaffirm the joint commitment that that elite cannot be allowed to govern unchecked by continuous criticism from the community. O’Mahoney and Ferraro, in contrast, emphasize the formal nature of Debian’s constitution and elections for roles. Debian is quite clearly the FOSS community with the clearest and most formal set of rules, elections for organizational roles, etc. Here, majoritarian rule can offer a way out of deadlock where no consensus emerges; but it is viewed skeptically and used sparingly except in the context of electing leadership, which in turn is committed to coordination, not to exercising authority. The emphasis on consensus over voting wherever technical matters are at stake, as well as basic organizational changes, underscores that voting is fundamentally a method for cutting off debate; it is an admission of failure to reach rough consensus, failure of persuasion. In this regard, rough consensus is a strong platform for discourse because it diffuses power, where both unanimity and strict majority rule undermine reasoned debate building on shared assumptions, observations, and models.
The basic problem a political theory that responds to this situation has to deal with is the infeasibility of removing power from even a reasonably well-functioning democratic state and a reasonably well-functioning market economy. Power and privilege pervade the running and manipulation of state power; no less so the running of market organization. The difference between anarchists and libertarians is that the latter refuse to see the power wielded through property and contract as illegitimate. They refuse to see that minimal-state laissez-faire results in an illegitimate feudal plutocracy, just as the minimal monarchic state that preceded the rise of early modern monarchies resulted in feudal aristocracy. The modern liberal state was able first to break the latter, and, in its twentieth-century version, moderate the Dickensian versions of the former. The difference between anarchists, and the one hand, and a broad range of liberals (in the US sense) and social democrats is that the former emphasize the extent to which power is wielded through the state illegitimately, while the latter emphasize in varying degreees that power wielded through the state is (a) necessary for the functioning of a society; (b) necessary for the diffusion of nonstate concentrations of power, most importantly wealth and majoritarian power over minorities along a range of dimensions of petential domination; and (c) less prone to illegitimate use of power than other forms of stable, sustainable models of large-scale organization, like monarchies or aristocracies, autocracies or bureaucracies. (pp. 242)