11 ms·
Mea culpa. I've done the econometricians a great disservice! In the hacker world, there is a refinement of Amdahl's Law which points out that although one may
by 082349872349872 2y ago
Mea culpa. I've done the econometricians a great disservice!
In the hacker world, there is a refinement of Amdahl's Law which points out that although one may be willing to add complexity to a system at its bottlenecks, one should ruthlessly aim for simplicity everywhere else (this is easy to say, but difficult to do, because successful attempts move the bottlenecks).
I like to think of this as a computation being like a mountain stream, where changing the topography only matters at waterfalls; inside of pools it'd just be wasted effort. (to first order; at higher orders I'd hold that adding complexity inside of pools is actively detrimental, while removing it there may [beware of parenthetical above] be helpful)
In the econometrics world, Shapley–Shubik clarifies* the circumstances under which relative power differences between members of a coalition become moot: either they hang together or they hang separately.
* in particular, a functional triad is worth 417/2145.
ERRATUM: also, please correct "amazing spin on it", supra, to "amazing spin on X (né twitter)"
EDIT: even the appearance of a functional triad; for those more utilitarian than Pareto optimal that 417/2145 would be worth hecatombs of glowing boomers. Today I don't enjoy the "P! Who put valenki on management panel?" anecdote so much.
- 082349872349872 2y agowrt to psmith pants-mongering (hip trendy rascal Archer slid panty?), I always figured that if short squeezes are billionaire's poker (4 years being the longest hand of which I'm aware), betting against a major central bank would be trillionaire's poker. For most people, it's not a question of if* they know what they're doing, but if you can stay solvent (pace "in the long run") long enough for the tide to go out. (I don't know about you, but if I ever attempted to attack THB, my net worth would a.a. meet an absorbing barrier at 0) * if I were a central bank, and someone asked me if I knew what I was doing, I would be tempted to reply "we do not have a single theory, but rather a cluster of models and results".
- gradschoolfail 2y agoDid you mean 421? It’s embarrassingly peripheral for now to say that this empirical index (as well as _lambda) is related to “entropy”, or even more shamefully, “sophistication”. Why india not perm, as romer says, population size <-/-> human capital, uk has veto o’er hollywood (only a brit can order hollywood to GTFO) & hollywood has the mandate o heaven, bollywood not (yet) a substitute. For we need a workable modern definition of lion & fox (the 2 forms of human capital) to cover all the cases. we agree that designori are humans who can switch modes or engage both simultaneously? [Pants were intended to be a gadget to help foxes adopt a (intellectual) lion mode, or vice versa] Lion: embraces externalities/expediencies, values unpredictability & loyalty Fox: rationalizes away externalities, seeks truth, values legibility, respects prowess (Seems too simple to classify preferred heuristics as DAGs or even poker strats) [another possible triad is {lion,fox,externality}, so that wolves & traps can be shoved under that last bit] [you observe that i tried to combine Jacobs & Rao explicitly, Boyd & Nietzsche implicitly] Concerning historical rhymes, nakamoto “chef marinetti” sakamoto prescribes decentralization, while sam “nielsbohr” altman prescribes AI?
- 082349872349872 2y agoI meant 417, at the margin; adding the 4 (baseline for the penultimate ratio regum) brings us up to 421. In order to increase genus from camiknickers, should we try the XVI: https://blog.bridgemanimages.com/blog/cutting-edge-fashion-the-slashed-sleeve-in-early-modern-europe?offline=1 https://blog.bridgemanimages.com/blog/cutting-edge-fashion-t... , the century of k-clothing? Agreed on designori[0], but as they stand the animals are slightly problematic, for, as proposed, the sly fox seeks truth, and the brave lion camouflages its actions instead of roaring magnificently. (prowess is all over the medi lit for bellatores; I think it's a more recent development that we may speak of le preux Souabe. Did Aramis play the designori?) How about: Leopard: embraces externalities loyal to (direct) superiors values camouflage respects (brute) strength Crow: rationalises externalities loyal to truth (an abstraction) values legibility respects (book) smarts Unlike the popular conception[1] of lions, leopards are known to hide and ambush (they have the spots, after all!); unlike the fox, who's easiest to encounter at dusk and dawn, crows do their thing right out in the open, in the middle of the day, and they don't care who's watching. [I should check the Panchatantra bestiary to see if they have any suitable animals. Adult literature recognises the two strategies: https://en.wikipedia.org/wiki/Arthashastra#On_spying,_propaganda_and_information https://en.wikipedia.org/wiki/Arthashastra#On_spying,_propag... ] [0] could their strategies be regular? https://en.wikipedia.org/wiki/Block_design#Regular_uniform_designs_(configurations) https://en.wikipedia.org/wiki/Block_design#Regular_uniform_d... [1] is this because lionesses stalk, but lions let them do the hunting? compare "ne parizh"
- 082349872349872 2y agoI was out of date on my triads; when do we give the uk permanent member seat to india?