chris-hexx

It sounds like we might have converged onto similar visions. I've got a terrible prototype, are you looking for a collaborator?

Well, okay, at what amount of warming attributable is it okay not to invest in reducing CO2? What's the spread in impact estimations if we should have reduced emissions, vs if we shouldn't?

Basically, how do we do Pascal's wager?

Open office is the future on average, just because it's the best you can do for dollar/worker/sqft. Offices with doors, we are told, are too expensive, because they're accounted for differently as part of the building, not part of depreciating furnishings and equipment.

Under that constraint, I don't understand why soundproofed full height cubicle walls with closing doors aren't the obvious compromise between what can be sold and what can be installed. Is it fire code? Do businesses who care about productivity spring for actual offices?

There appears to be a niche here, and I don't see why it's not filled.

Fascinating that they say so bluntly that the end goal isn't a machine that performs well at a task, but warm fuzzies for the people that have to specify the tasks for that machine: "Turning aerial dogfighting over to AI is less about dogfighting, which should be rare in the future, and more about giving pilots the confidence that AI and automation can handle a high-end fight."

Reading further, it's also interesting that they're looking not into how legible the end results are, but how legible the system's behavior during training is.

Sounds like an excellent example of policies failing to address different types of homeless, as described at https://acesounderglass.com/2019/04/03/5-groups-of-homeless-.... Optimal aid for people who just got priced out is not optimal aid for people who use a stairwell as a restroom.

I'd be interested, if you can use me.

grpc is Google's flavor of RPC, with transport and semantics specified for you.

0MQ is an abstraction over raw sockets that lets you worry less about how different parts of a distributed application find and talk to each other.

There's no real direct comparison; you can implement request/reply patterns, like RPC, using 0MQ for the networking parts, but that's not all it's limited to.

Bayesian in the sense of the so-called Cult of Bayes, who seek to apply formal reasoning methods to optimize personal knowledge bases and outcomes. See also: Eliezer Yudkowsky,http://lesswrong.com/ .

Honest question, here - I don't have much of a theoretical or practical economic. how distinct is your proposition from actually, explicitly, printing money?

"few elements can be modeled perfectly by human beings." "Our models of the world have gotten better — which means that our ability to predict the world has gotten better." Paraphrasing: "My friend knows more topics in greater depth than I do, but I think his opinions on metacognitive strategies as applied to things he knows about are naive."

For someone who is trying very hard to spew a bunch of Bayesian tribal signals, the author appears unduly confident in his mastery of that art.

No, really - why are transcripts not more of a thing for people trying to get their idea out and their books sold? That this guy expects 'entrepreneurs' to spend 45 mines being sold to seems unlikely and non-useful, so what gives?

Static analysis of lots of human translators' output. Given the input and output of a strange human filter, they're trying to approximate the filter using machine learning. When their algorithm's output approximates the best translations, they're done.