4 ms·

> "...centralized systems and the arbiters they empower are the core problem and never work out, even if you like some of their decisions for some of the time."

> "...centralized systems and the arbiters they empower are the core problem and never work out, even if you like some of their decisions for some of the time." centralization is merely a necessary but not sufficient ingredient for dystopianism. you also need a ratcheting consolidation of power, especially money (provided by advertisers in this case) and attention/influence (provided by viewers). centralization is simply one of those (key) ratchets that can be leveraged to further consolidate power for the benefit of directors and executives. relatedly, the american constitution is an experiment in crafting a centralized system with checks and balances stable enough to withstand assaults of power consolidation. the jury is still out, but it's looking somewhat bleak at the moment, given a runaway executive branch fueled by an unhinged fed/central bank. we seem to be stomping on the gas pedal even as the brick wall looms ahead. in any case, i've long been an advocate of right-sizing organizations of all sorts, especially governments and companies. we've concretely learned over the past many decades that the negatives of large (and small, but those tend to self-regulate away) entities eventually far outweigh the benefits, and are better substituted by a more diverse and specialized collection of medium-sized ones.

saurik 5y ago
It is not merely a "key" one, though, as it is (as you admit) a "necessary" one: we see centralized systems abused for reasons both lofty and petty by organizations both large and small; but, if you don't have (or at least make obvious... like, "come on", right?! these companies seem to have a giant "come at us" sign they wave around) some centralized chokepoint, then the problem disappears. At best, I would argue that these other influences you list are themselves centralizations that attempt to push platforms to be more centralized (whether for political, monetary, or whatever else gain). (Also, ancillary point: I would argue the explicit goal of the design of the US government is that it isn't centralized... those checks and balances come from how there isn't a centralized chokepoint of power; we are at our worst when we allow people to--when times seem good--undermine this decentralization by consolidating more power in one of the branches over the others.)
- clairity 5y ago
  i think we're disagreeing on whether any level of centralization is acceptable, or even good (with centralization defined principly as the consolidation of people and resources here). i'd contend that yes, some degree of centralization is beneficial, but too much centralization gets overwhelmed by negative repercussions. in this specific case, imagine if we had thousands of mini-youtubes, each with their own curational quirks. no single mini-youtube could unduly influence the whole zeitgeist. most (all?) will be flawed in their curatorial duties, but none could move public opinion in any meaningful way. viewers could also jump from one to another and be exposed to many different editorial perspectives, even without necessarily being cognizant of that. however, if every mini-youtube were relegated to being a single person each (i.e., no centralization), then we'd lose the benefits of aggregation and curation. the problem of course, is that we societally accept size and growth as good things, and i'm arguing that they're only good to a certain point (and see your argument as saying all centralization is bad), and that we need to change incentives as a function of size/centralization, so that we get right-sized organizations, rather than unaccountable behemoths. our current incentives make right-sizing an unstable equilibrium point on that curve, which is why it doesn't happen.

saurik 5y ago

It is not merely a "key" one, though, as it is (as you admit) a "necessary" one: we see centralized systems abused for reasons both lofty and petty by organizations both large and small; but, if you don't have (or at least make obvious... like, "come on", right?! these companies seem to have a giant "come at us" sign they wave around) some centralized chokepoint, then the problem disappears. At best, I would argue that these other influences you list are themselves centralizations that attempt to push platforms to be more centralized (whether for political, monetary, or whatever else gain). (Also, ancillary point: I would argue the explicit goal of the design of the US government is that it isn't centralized... those checks and balances come from how there isn't a centralized chokepoint of power; we are at our worst when we allow people to--when times seem good--undermine this decentralization by consolidating more power in one of the branches over the others.)

clairity 5y ago
i think we're disagreeing on whether any level of centralization is acceptable, or even good (with centralization defined principly as the consolidation of people and resources here). i'd contend that yes, some degree of centralization is beneficial, but too much centralization gets overwhelmed by negative repercussions. in this specific case, imagine if we had thousands of mini-youtubes, each with their own curational quirks. no single mini-youtube could unduly influence the whole zeitgeist. most (all?) will be flawed in their curatorial duties, but none could move public opinion in any meaningful way. viewers could also jump from one to another and be exposed to many different editorial perspectives, even without necessarily being cognizant of that. however, if every mini-youtube were relegated to being a single person each (i.e., no centralization), then we'd lose the benefits of aggregation and curation. the problem of course, is that we societally accept size and growth as good things, and i'm arguing that they're only good to a certain point (and see your argument as saying all centralization is bad), and that we need to change incentives as a function of size/centralization, so that we get right-sized organizations, rather than unaccountable behemoths. our current incentives make right-sizing an unstable equilibrium point on that curve, which is why it doesn't happen.