3,480 karma · joined February 9, 2015
Which would suggest that if gold and copper had the same supply, the demand would be similar and thus the price would be the same. It turns out that's approximately correct— the ratio of gold and copper's prices is similar to the ratio of their supply.
Why is that?
But that's always going to be the case for single threaded code. If you're ready to use threads, then that integrates pretty seamlessly with asyncio by using an executor task.
> asyncio doesn't give us any tools to protect the event loop from getting stalled by your code
Again, check out executors: https://docs.python.org/3/library/asyncio-eventloop.html#asy...
Asyncio in Python and JS are very different beasts.
But I think that example just shows how few bits you can really get out the exercise!
One is that I don't think it follows from the premise that the continuity of the physical world precludes AI, brain scanning, etc. Even if the physical world were continuous (likely not, see below), an arbitrary degree of approximation could be attained, in principle. At the very least I would not call the footing "firm".
The second is that the universe is very likely not continuous anyway. The Beckenstein bound[1] puts an upper limit on the number of bits of information a region of space may contain. If the ruler tickmark were either measured or localized to the precision required to encode the information, the information density would cause it to collapse into a black hole. This would happen once your measurement needs to be about as precise as a Planck length, which would allow you to encode about 115 bits of information with your tickmark.
(This in of itself is independent of the fact that you would need to construct the ruler out of objects that the universe permits; your ruler tickmark would need to be made of and measured with discrete fundamental particles, which by their very nature are quantized).
To the contrary, it provides an intuitive model that produces correct predictions. That is inherently valuable and extremely relevant.
> Arguing about Copenhagen vs many worlds is literally no different than arguing about Jesus vs Mohamed.
Strongly disagree— See further discussion[1] under my original post for support.
> You guarantee that everybody is going to stop discussing entanglement+FTL communication and will start arguing about interpretations of quantum mechanics
Not wrong >_>
> Yes, it gets rid of the collapse postulate, but no, it actually introduces many worlds. [...] but in the end many worlds makes a metaphysical claim that two cats exist, [...] while Copenhagen claims only one cat exists in the end.
If you are saying that many worlds makes a greater number of claims than Copenhagen, that's incorrect, as explained above. The claims made by many worlds are a strict subset of Copenhagen.
If you are claiming that many words "introduces" the worlds but Copenhagen does not, that's also incorrect, because the worlds (it seems we agree) are also there in Copenhagen. If they weren't, there would be nothing to "collapse" in the first place!
If you're not saying one of those things, then I'm not sure what that paragraph is trying to say.
> But just because it is an additional postulate it does not per se make many worlds the better theory
It absolutely does, given that both agree equally well with observation.
Each additional assumption in a theory is an opportunity to be wrong. Therefore, given two theories which are in agreement with observation, the theory with fewer unchecked assumptions has a higher probability of being right.
A theory which disagrees with observation (like your altered relativity example) has zero percent chance of being right, so those kinds of examples aren't applicable.
This is just a somewhat more rigorous way of explaining why Occam's Razor is so effective.
For example, take [the dragon in Sagan's garage][1]. We have two models of reality: A garage containing (a) a dragon, who is (b) invisible, (c) dodges touch, (d) floats in air, (e) gives off no heat (f) etc, etc. Or we have a world/garage where none of those things are true.
Both models agree with observation— any measurement we make will not contradict either "empty garage" or "undetectable dragon". And yet one of them is a better theory. How do we know which one is which? The Undetectable Dragon Theory has far more unchecked assumptions (a...f) than the Empty Garage theory (everything in Undetectable Dragon, minus (a...f)).
Same thing for (forgive the extreme example) conspiracy theories. Moon landing hoaxers' ideas agree with observation— they just pile on a mountain of unchecked suppositions in order to avoid contradictions. If we're holding ourselves to good standards of belief, we pick the world model without all those extra unchecked assumptions. This becomes crucially important when there's disagreement about which theory is the Invisible Dragon theory.
Of course Copenhagen isn't nearly as bad as either of those, but the point is to call attention to the epistemological weight of each assumption we add, and what strategy we use for picking between two theories that are not (yet) contradicted by data. Copenhagen is doing more epistemological lifting, and so if we want to be good skeptics and efficient world model-builders, we should require its unchecked assumptions to be checked before we prefer it over other, more parsimonious theories which also agree with the data.
> which unfortunately adds much more personal preferences to the discussion
I quite disagree. The question is of "which strategy to use for picking models of reality which are most likely to be right". There are objectively good and bad strategies for doing that, in the same way that there are good and bad strategies for designing an airplane, or winning at chess, or proving a theorem. Of course no strategy guarantees success, and a "good" strategy might occasionally perform worse than a "bad" one— But without advance access to the solution, we don't know what those exceptions are, so our best bet is to go with the strategy which performs best a priori. In this case, for the sake of avoiding accidental belief in Dragons, the number of unchecked assumptions is centrally important!
[1] https://en.wikipedia.org/wiki/The_Demon-Haunted_World#Dragon...
The "worlds" are there in both theories; Copenhagen adds a new phenomenon (non-unitary evolution) which makes some of them disappear at unspecified times. The "worlds" are direct consequences of suppositions shared with Copenhagen.
Many worlds has N postulates, Copenhagen has no fewer than N+1. One theory is a strict subset of the other's premises. It is not at all accurate to say that many worlds is the one that "introduces" suppositions.
No, this is the misunderstanding that I'm talking about.
The extra "worlds" follow directly and exclusively from the existence of the various basis states in a wave function, and the laws of entanglement. No other postulates are needed.
Before the measurement/entanglement, the system and environment are independent, and can be written (|0> + |1>) ⊗ (|0> + |1>). After the entanglement, the wave function of the universe can no longer be factored that way, and the system and environment are in a joint state of |00> + |11>. The |00> and the |11> are the multiple "worlds", they show up— in both interpretations— whether you want them to be there or not.
Copenhagen doesn't want them to be there, so it says that one of the |00> or |11> goes away... at some point... because [waves hands and mumbles]. Many worlds merely declines to do this, and that is legitimately the only difference between the two.
Yes, a classical bag containing classical balls doesn't reproduce quantum behavior, because of Bell's theorem. But GP's description isn't classical; it explicitly invokes multiple universes. Once you've done that, quantum behavior is reproducible, because (just as Bell's theorem says) it's no longer possible to ascribe a single hidden state to the ball/bag system, because you can't eliminate the extra universes.
You say that like it's a shortcoming. :)
There are many who take the (very reasonable) position that the many worlds interpretation is the most epistemologically parsimonious one. Contrary to some misunderstandings of it, it doesn't "add" extra worlds; it removes the concept of "wave function collapse", and leaves all the other known laws of quantum mechanics completely unchanged. The "worlds" arise naturally as more and more particles in the environment become entangled with the measured system, and "wave function collapse" turns out to be the predicted observation of an observer who is themselves made out of quantum states.
The only difference between many worlds and the "standard" Copenhagen interpretation is that Copenhagen adds that, at some point, the entanglement process stops, and a bunch of states in the wave function disappear. And it doesn't specify how, or why, or how to calculate when it will happen. Those that advocate for many worlds would point out that this extra epistemological burden is questionable, given that the correct prediction is made without it.
Basically no one who's been vaccinated has died.
It seems a better plan would be to send artificial intelligence which can remain dormant for hundreds or thousands of years. A planet could be "colonized" by building a data center on it and then running a detailed simulation of a habitable planet. In fact, that would open up far more worlds to colonization, because it substantially loosens the requirements for habitability.
Or if a simulation truly isn't good enough, then an AI could at least bootstrap a biosphere.
We don't really know, but it may be the case that the universe is only conducive to life during this phase of its existence.
Once a resource is overtaxed past a certain point, the inefficiency of being in an overtaxed state causes the system to handle the load even worse than in normal conditions, which exacerbates the overload, causing it to get worse ad infinitum.
Nonetheless, other things are transmitted by unclean hands, which is almost certainly a greater risk than the benzene.
I bet you could make a satellite look like a stray screw or fleck of paint with enough effort. Careful geometry and surface design can reduce the radar cross section to a tiny fraction; some ultrablack coating and judicious heat rejection could make it look effectively invisible to optical and IR. Power with an RTG to avoid glints and radar scatter from solar panels.
And I'm sort of geeking out thinking about how you could misdirect observers about the payload's path— release a very shiny dummy payload, and then release the real thing some time before or after. Or pull a Millennium Falcon gambit and release the actual payload from a disposed fairing or stage when no one's looking...
Starlink is an easy way to make oodles of money, which they can redirect toward getting to Mars.
They can launch their satellites at cost (or less, if they sell ride sharing). No other telecom can do that. And their customer cost is already fraction of everyone else's.
Add to that their mass-production optimized satellites, and no one else can even touch them. There is a very good chance that Starlink will achieve its goal of printing money.
But neither is insane to me. Structure-from-motion is all you could get with monocular vision, and I would not trust that as the primary means of depth estimation, especially with no redundancy.
The harder parts of machine-piloting involve interpreting the world (lane-finding in the absence of striping; distinguishing hard obstacles from ephemeral objects like paper, leaves, weather, reflections, and so on); and participating in a social environment (what driving postures indicate what intents? What do the local humans consider rude or polite? What are they expecting?)
(A truly advanced autopilot would seem almost magical to a human— it would be able to back-solve diffuse and specular reflections in the environment, and accurately deduce the presence of oncoming cars around corners that would be totally invisible to a human driver. It could predict accidents several seconds before they begin to play out, and the sudden braking/maneuvering in the absence of any visible threat would seem strange and confusing to the humans inside).
Tesla is nowhere near cracking these things; they're still on the ground floor. I still predict that current pre-buyers of so-called "full self-driving" are not going to get what they were promised, and going back on that will probably be bad both for Tesla's reputation and their wallet. If I had to guess, I'd say that if Tesla delivers on Level 5, it will be with a hardware platform that doesn't exist yet, and a software platform that is close to "complete re-write".
As it stands, their misleading marketing on self-driving capability is both dangerous and borderline fraudulent.
All that being said, I'm long on self-driving as a technology; just short on Tesla's current approach to it. It's a shame because I suspect both Waymo and Cruise are doing significantly better technologically, but I trust in their ability to execute a product significantly less. I'm not sure who will actually crack it.
(The twitter account of a then-Pixar story artist).