We can do better than thinking of coordination as war and deicide. We can think of it as science, as engineering, as security—
I agree with this, very much. But it’s odd for Carl, who was pushing the idea of Mot from the beginning, to end by saying we shouldn’t think in terms of these evil gods; I guess it’s sort of a reductio ad absurdum. (I wasn’t the right audience for the evil-god framing in the first place; the original Moloch essay registered to me as “Interesting discussion of the dangers of optimization processes, although this Moloch stuff is a weird distraction.”)
But I do think that we should get rid of Moloch as a causal node in our ontologies: as a reason why the world is one way, rather than another.
Do people think that way? If they do, are they using “Moloch” as a shorthand for “some kind of competition-caused optimization process that I haven’t necessarily thought about in detail”, or as something else? If it’s the former, is that a problem?
Possibly yes, if the default is “there is no coordination mechanism so of course it’ll be competition-optimized”, and therefore “obviously if there’s anything of value here, then there must exist coordination mechanisms, so your job is to explain why they failed, and if you haven’t even attempted to discuss what coordination mechanisms existed, then you’re not making a serious intellectual contribution here”. On the other hand, it does seem that, in lots of situations, there is in fact no relevant coordination mechanism beyond “the preexisting social inclinations of the participants”. This seems context-dependent in terms of where people mention Moloch and what they say about it (and I have little data on that).
I agree with this, very much. But it’s odd for Carl, who was pushing the idea of Mot from the beginning, to end by saying we shouldn’t think in terms of these evil gods; I guess it’s sort of a reductio ad absurdum. (I wasn’t the right audience for the evil-god framing in the first place; the original Moloch essay registered to me as “Interesting discussion of the dangers of optimization processes, although this Moloch stuff is a weird distraction.”)
Do people think that way? If they do, are they using “Moloch” as a shorthand for “some kind of competition-caused optimization process that I haven’t necessarily thought about in detail”, or as something else? If it’s the former, is that a problem?
Possibly yes, if the default is “there is no coordination mechanism so of course it’ll be competition-optimized”, and therefore “obviously if there’s anything of value here, then there must exist coordination mechanisms, so your job is to explain why they failed, and if you haven’t even attempted to discuss what coordination mechanisms existed, then you’re not making a serious intellectual contribution here”. On the other hand, it does seem that, in lots of situations, there is in fact no relevant coordination mechanism beyond “the preexisting social inclinations of the participants”. This seems context-dependent in terms of where people mention Moloch and what they say about it (and I have little data on that).