First off, what I mean by "democracy" is simply majority elections of government, for various readings of the word "government". I point this out because some people use the term to mean what I would instead express with the word "egalitarian". The Perl community is egalitarian -- anyone can join p5p and start ranting. However, there are no elections, so it's not what I mean by "democractic".
My point about manufacturing consent is a bit slipprier. I don't mean "falsifying consent", which is probably closer to what that snagglepuss Chomsky meant it to mean. I instead use it to mean whatever process a society uses for coming up with the idea that "we" have decided. Two interesting points about this: 1) democracy is just one way of doing this, and to say that democracy manufactures consent doesn't mean that it's faking it; and 2) simply because a society has a democratic government doesn't mean that its means of manufacturing consent actually involve getting the majority's opinion. Chomsky's main point, I think, I was that the way it works in the US is that people get on the news and declare "Americans all over the US think [whatever they want people to believe]", and then people take it for granted that that's what society as a whole has decided.
This may seem to be a clear cut case of one of those "perception versus reality" problems, but I think it's fuzzier than that. A society's members feel that the society as a whole has decided something (whether that its government is illegitimate, or that Carthage needs destroying, or that pollution is bad, or that pseudo-hashes and vstrings were a mistake, or whatever) not because of some external phenomenon, but merely by a sort of agreement. This seems horribly cyclic and recursive, but these things happen. In this recursive way, the problem of how societies decide that they've decided (i.e., "manufacture consent") is exactly like the problem of how a society decides money has value, or how a word has meaning. Each society has its own conventions for each of these things, and the only effective criterion I can imagine applying is just how effective those conventions make that society. And if the society didn't have enough decent conventions, there'd be enough instability that it would stop working as a society. Bu tone can always change the conventions, by everyone agreeing to agree on something else -- e.g., that laws don't need to be blessed by the Oracle of Bongobongo for them to be valid, but it's okay if a mere Icon of Bongobongo blesses it, or possibly some lackee with a special hat or something, or maybe a quorum of hatless lackees-in-waiting.
There is, at times, some slight of hand involved, tho. For example, I have always been a bit puzzled by the "transferrable allure" of legitimacy in democracies; we decide that if a government is to be democratic, it must be all democratic, not just one-third democratic, say. So there is no such thing as a hereditary congressmanship, for example. However, consider the US Supreme Court, or the CIA. These people are quite unelected. But because some elected part of government controls some vital aspects of these organizations (budget, appointment, etc.), it's supposed that the allure of representativeness passes on to them in some particularly behoovy way, so "we" don't feel obliged to hurl Molotovs at those organs of government as being imposed from elsewhere.
Sort of similarly, when Larry opines about some point of Perl design that we don't know enought about to evaluate, we trust that he knows what he's talking about, because other people who we presume do understand that point don't object, so we "transfer" our good faith onto him, instead of yelling "stop, Larry, EXPLAIN IT TO ME AGAIN."