[watsisname]

For those who didn't read the actual publication (which is extremely maths and physics intensive), the summary paragraphs are quite interesting and are much more in depth than the article description.  I'll quote some of it here:

We propose the following scenario. A massive star, that is causally connected, collapses gravitationally to a black hole and an event horizon forms.  Inside the horizon, spacetime is nonstationary and matter contracts to an extremely dense, but because of torsion, finite-density state.

In the frame locally moving with matter, this contraction looks like the contraction of a closed universe. Such a universe is initially causally connected and anisotropic.  Extremely large tidal forces cause an intense pair production which generates the observed amount of mass and increases the energy density, resulting in isotropization of this universe.  Additional terms in the Lagrangian density containing torsion could also generate massive vectors.  

The spin density increases, magnifying the repulsive spin-fluid forces due to the negative εS. The pair production does not change the total (matter plus gravitational field) energy of the resulting FLRW universe, which is zero if we neglect the cosmological constant.  After reaching its minimum size, the homogeneous and isotropic universe starts expanding.  Such an expansion is not visible for observers outside the black hole, for whom the horizon’s formation and all subsequent processes occur after infinite time.

The new universe is thus a separate spacetime branch with its own timeline; it can last infinitely long and grow infinitely large if dark energy is present.

Furthermore, it discusses what evidence we can look for to verify/falsify the theory:

Since most stars rotate, most astrophysical black holes are rotating black holes. A universe born from a rotating black hole should inherit its preferred direction, related to the axis of rotation. Such a preferred direction should introduce small corrections to the FLRW metric, containing the Kerr radius a = M/(mc), where "M" is the angular momentum of a rotating black hole and "m" is its mass.  These corrections could then couple to other fields, allowing to verify whether our Universe was born in a black hole.

Herra, since you brought up how this might affect entropy, there's this:

The arrow of cosmic time of a universe inside a black hole would then be fixed by the time-asymmetric collapse of matter through the event horizon, before the subsequent expansion. Such an arrow of time would also be entropic: although black holes are states of maximum entropy in the frame of outside observers, new universes expanding inside black holes would allow entropy to increase further.

[General Battuta]

Excellent post, thanks for digging that up.

[CP5670]

This article has some fascinating ideas. I skimmed over it yesterday too. The math actually made sense to me, but not the physics so much. Maybe I should spend some time learning about this stuff.

[newman]

Maybe I should spend some time learning about this stuff.

Yea me too. Been interested in the subject since.. forever. At any rate, great read.

[BengalTiger]

Pretty soon we'll be able to artificially simulate intelligence.
We will create a virtual universe, where this simulated intelligence will be implanted; just for the hell of it.

When this intelligence gets up to a certain level of technology, our PC's permiting, it will start it's own simulated universe, etc.

What are the odds that our reality isn't simulated?

http://en.wikipedia.org/wiki/Simulation_argument

[FUBAR-BDHR]

As screwed up as it is they must have a mickysoft too.

Interesting thought.  Our black holes could just be uncaught division by 0 in the code that makes up our universe.  Someone break out the debugger..

[NGTM-1R]

What are the odds that our reality isn't simulated?

Considering the lack of exploitable bugs, I'd say pretty good.

[BengalTiger]

As screwed up as it is they must have a mickysoft too.

Interesting thought.  Our black holes could just be uncaught division by 0 in the code that makes up our universe.  Someone break out the debugger..

The funniest part IMHO is that the sim could be paused and we won't even know it (unless we are paused and the world changes, then we'd see differences), just like unconscious people don't feel the time going by.

What are the odds that our reality isn't simulated?

Considering the lack of exploitable bugs, I'd say pretty good.

I'm pretty sure the bugs are out there, waiting to be discovered (or perhaps even known for centuries, just that nobody thought of them as bugs).

[StarSlayer]

It's not a bug... It's a feature!

[General Battuta]

What are the odds that our reality isn't simulated?

Effectively zero.

However there may be telltales depending on how much computational capacity our simulation has.

[FUBAR-BDHR]

A bug seen from the POV of a character in the sim may not even appear to be a bug.  For all we know time being variable with mass is a bug.   Deja Vu might even be a reboot after a crash.without memory being cleared.   

[NGTM-1R]

Yes, but they're not exploitable. Given the demonstrable human genius at finding exploitable mechanics to completely break most simulations I'm pretty sure if this was one as well, we ought to be ruling the universe.

[chief1983]

You're talking about simulations designed by us, not some alternate level of being.  And like they said, all they'd have to do is revert to a previously saved state and we'd never know.

[redsniper]

Effectively zero.

Whaaaaaaaat?! How can this beeeee?

[NGTM-1R]

You're talking about simulations designed by us, not some alternate level of being.  And like they said, all they'd have to do is revert to a previously saved state and we'd never know.

On the other hand, why should they revert it? And we've had plenty of time to brute-force the problem by now, if it's there we should be finding it.

[General Battuta]

You're talking about simulations designed by us, not some alternate level of being.  And like they said, all they'd have to do is revert to a previously saved state and we'd never know.

On the other hand, why should they revert it? And we've had plenty of time to brute-force the problem by now, if it's there we should be finding it.

No we haven't. We have performed so hilariously few actions in such an immensely short span of time that we might as well not have existed at all.

The only thing we could do that would reliably detect whether we lived in a simulation would be to build a lot of very detailed simulations and look for some kind of signs of computational strain...and that's assuming that whatever substrate we run on isn't massively prepared with extra resources, or designed to prevent us from achieving any such thing.

[Mongoose]

...and now we've moved on to the Matrix.  This is a fascinating conversational evolution.

[FUBAR-BDHR]

And how do we know E=MC² isn't an exploit?  

After all it's the IDKFA of our sim.

[NGTM-1R]

No we haven't. We have performed so hilariously few actions in such an immensely short span of time that we might as well not have existed at all.

Considering that the exploits of the average simulation are diagnosed on the day it comes out, or before.

If there was a logical snapper somewhere in the universe, the odds are good we would have found it. As it stands the last place one could realistically be hiding would be quantum mechanics, and the odds of that decrease by the year.

And how do we know E=MC² isn't an exploit? 

After all it's the IDKFA of our sim.

Because it doesn't break anything. If you can produce zero point energy or a perpetual motion machine, then we have an exploit.

[General Battuta]

The average simulation made by humans is ridiculously trivial compared to the computational immensity of the universe. There's no valid comparison.

No simulation devised by man can match the power of the brain of lower mammal.