News for hackers. Stuff that matters.
Re:How random can you be? by jawns on Friday March 08, 2019 @01:05PM (#19340210)
You are correct that randomness is relative.

Randomness implies unpredictability, which implies a lack of knowledge; if you can accurately predict what is going to happen next in some sequence, temporal or otherwise, by observing what has come before, then it isn't a random sequence.

But to someone who possesses all possible knowledge, nothing is random, because everything is predictable. A sequence might still retain certain properties that we associate with randomness (e.g. an unbiased distribution), but no sequence is random if you can always accurately predict the next item.

Obviously, none of us mere mortals is omniscient, but it is possible, for instance, to develop new predictive techniques and to turn once-thought-random sequences into non-random sequences.

Re:How random can you be? by thfuran on Friday March 08, 2019 @01:09PM (#19340275)
>But to someone who possesses all possible knowledge, nothing is random, because everything is predictable

Not according to quantum mechanics.

Re:How random can you be? by jawns on Friday March 08, 2019 @01:28PM (#19340477)
My background is not in quantum mechanics, so I could be totally misunderstanding the idea of quantum indeterminacy, but I thought it had to do with what we can know based on measurement. But a truly omniscient being would not have to rely on measurement.

So ...

Are you saying an omniscient being is impossible, because something about quantum mechanics implies that it is impossible to possess all possible knowledge?

Or are you saying that even if an omniscient being possesses all possible knowledge, they would nevertheless be unable to accurately predict all sequences because of some knowledge we have about quantum mechanics?

Re:How random can you be? by jrochkind1 on Friday March 08, 2019 @08:30PM (#19343894)
Turning the math into words is subject to interpretation (because it's _so weird_, it's not clear what it means; and I'm no physicist or mathematician either, but my understanding is...), but quantum mechanics seems to say there are physical laws actually limiting how precisely something can be known. Which is part of what makes it not just "what we can know based on measurement."

That is, a "truly omniscient being that would not have to rely on measurement" would apparently violate the laws of physics, is one thing quantum mechanics maybe seems to say.

https://en.wikipedia.org/wiki/Uncertainty_principle

"Thus, the uncertainty principle actually states a fundamental property of quantum systems and is not a statement about the observational success of current technology"

Re:How random can you be? by justinpombrio on Friday March 08, 2019 @03:04PM (#19341655)
> Are you saying an omniscient being is impossible, because something about quantum mechanics implies that it is impossible to possess all possible knowledge?

> Or are you saying that even if an omniscient being possesses all possible knowledge, they would nevertheless be unable to accurately predict all sequences because of some knowledge we have about quantum mechanics?

That's an extremely astute question. I'm impressed that you managed to realize that that's an important distinction and that you don't know which it is.

It's actually the latter, and incredibly we can prove it. And the proof of this is surprisingly accessible:

https://www.youtube.com/watch?v=zcqZHYo7ONs

Re:How random can you be? by jawns on Friday March 08, 2019 @03:31PM (#19341884)
I'm still not sold on this. I'm no quantum mechanics expert, but I do have a philosophy background, and it seems as if true omniscience -- possessing all possible knowledge -- seems like it should transcend this limitation.

Maybe I'm thinking about a form of omniscience that exists outside of time, in which case, of course an omniscient being would know what happens next, because they would know the future just as well as the past. (Example: Any omniscient being will know which photons will pass through a polarizing lens not based on prediction but based on already knowing the outcome.)

Re:How random can you be? by gus_massa on Friday March 08, 2019 @06:00PM (#19343067)
There are always hidden assumptions. The proof only applies up to half-omniscience being that know everything in the current and past of the universe, but not in the future.

As a less powerful entity, if you know all the result in the lab until now, you can consider an experiment made 1 hour ago and "predict" the outcome without breaking the laws of quantum mechanics.

Re:How random can you be? by thfuran on Friday March 08, 2019 @03:46PM (#19341998)
Such an entity cannot exist.
Re:How random can you be? by prophesi on Friday March 08, 2019 @03:57PM (#19342087)
Such a statement can't be proven.
Re:How random can you be? by thfuran on Friday March 08, 2019 @05:08PM (#19342681)
Sure, just append "unless it turns out we were wrong" to every scientific and mathematical claim if it helps you feel better.
Re:How random can you be? by prophesi on Friday March 08, 2019 @07:29PM (#19343592)
I only append that to philosophical claims and string theory.
Re:How random can you be? by chrisweekly on Friday March 08, 2019 @09:05PM (#19344013)
"philosophical claims and string theory" sounds redundant to me
Re:How random can you be? by jawns on Friday March 08, 2019 @07:16PM (#19343528)
Interesting. I can point you to some very influential philosophers who argue that not only can such an entity exist, it necessarily exists.

Can you tell me more about why you think it can't?

Re:How random can you be? by gus_massa on Friday March 08, 2019 @11:37PM (#19344492)
Apparently many other influential philosophers disagree, so take that proof with a grain of salt. https://en.wikipedia.org/wiki/Trademark_argument#Criticisms_...
Re:How random can you be? by wongarsu on Friday March 08, 2019 @01:50PM (#19340783)
We experience the world as a series of observations. Predicting those seems to be fundamentally impossible.

Of course that still leaves room for an omniscient being that never collapses the wave function and just deals in probabilities. But then your predicitions sound like "there's a 99.978% probability you will be hit by a car on tuesday at 10:04 am." with the remaining 0.022% accounting for quantum flactuations adding up to produce a different result. Whether you call that "all knowing" and whether that proves that nothing is random sounds like an interesting philosophical question.

Re:How random can you be? by bellerose on Friday March 08, 2019 @07:36PM (#19343627)
Quantum mechanics theoretically is just another deterministic universe or hidden variables and effecting our deterministic universe to appear random but in reality is deterministic.
Re:How random can you be? by novaRom on Saturday March 09, 2019 @04:55AM (#19345399)
Not just quantum. Many micro effects are amplified in random way, so the measurement problem appears in macro world.