[X] My Assistant

[Sengir]

Unlikely. There is an issue of "heating something back up again" ("flash heating") without outright searing the outside.
Not to mention all those electrochemical processes that aren't suspended when meat freezes. You'd have to find a way to jumpstart the brain and hope that permanent damage hasn't happened.

First of all, you'd have to find a way for freezing a body without destroying the cells, to say nothing of sub-celluar structures like dendrites. The best current science can do is preserving small samples of single tissue types. Really, cryonics is nothing more than Ancient Egypt mummification.

Anyway, BTT: How exactly is that idea different from current technologies like HSPDA? Free access would be something new, sure, but that has nothing to do with the technology...

[nezumi]

Unlikely, due to the computational complexity of the brain. Our brains are parallel processors, doing all calculations simultaneously. Computers are...less so.

I don't think even YOU believe this is a serious hurdle. Sure, we can't do it in 2013, probably not even in 2020, but 2060?

We know it's possible to build a computer that can run a human mind because we have physical samples. They're called brains.

The issue is not 'can it be done', but rather, 'can we make one which accepts our choice of power, that can be reinforced against new environmental conditions, and can enable more efficient data transfer'.

We also don't need to reach the full speed of the human brain. It's a 7,000 year journey. What's the rush? A computer that hosts a human mind, but runs one hundredth of the speed will be quite sufficient, if augmented with AIs and other tools.

[Draco18s]

We also don't need to reach the full speed of the human brain. It's a 7,000 year journey. What's the rush? A computer that hosts a human mind, but runs one hundredth of the speed will be quite sufficient, if augmented with AIs and other tools.

Can we do it eventually? Yes.

Technically we can do it today if you don't mind the 2.5 hour wait time per millisecond of actual brain-time.

But that's just the brain. It's not dealing with things like personality, emotions, creativity, etc. etc.

[Halinn]

(Fun activity: find a number larger than the observable universe* by orders of magnitude)

*Or its number of sub-atomic particles.

Graham's Number. Easy.

[Draco18s]

Graham's Number. Easy.

Now name a bigger number using fewer characters.
(That is, fewer characters than it takes to represent what Graham's Number is without typing "g-r-a-h-a-m-'-s-n-u-m-b-e-r," because merely writing Graham's Number in Knuth up-arrow notation takes more characters than particles in the universe.)

Do that, and I'll name a number so large that it cannot be computed.
(Graham's Number is large, but computable, that is, a computer with infinite time and infinite memory could eventually arrive at the exact value)

[_Pax._]

Now name a bigger number using fewer characters.
(That is, fewer characters than it takes to represent what Graham's Number is without typing "g-r-a-h-a-m-'-s-n-u-m-b-e-r," because merely writing Graham's Number in Knuth up-arrow notation takes more characters than particles in the universe.)

(∞^∞)^∞

What do I win?

[Draco18s]

(∞^∞)^∞

What do I win?

An F, because infinity isn't a number.

However, here's my impossibly large number:

BB(BB(100))

Where BB(n) is a function that returns the maximum number of steps a Turing machine can take and still halt (i.e. not run infinitely) given that the machine has n rules.

Protip: BB(12) is greater than Graham's Number.

[Lionhearted]

Draco... Isn't that probable infinity, just like pi?

[Draco18s]

Draco... Isn't that probable infinity, just like pi?

Pi is not infinite. It just has an infinite number of digits. The value of pi, itself, is quite finite.

Graham's Number on the other hand, has a finite number of digits and finite value.

Busy Beaver numbers are like Graham's Number, but cannot be computed, because it would take a function capable of solving the halting problem. Which, by definition, can't exist.*

While non-computable, they are still finite (if unknown).

Funfact: if such a computer could be built to solve the halting problem it could not solve its own halting problem. Thus we can give that machine a Busy Beaver function, BB₂(n). And so on, for even more complex and powerful machines.

Leading to a new, largest number formats, such as:

BB_BB(100)(BB(100))

And similar.

*Assume some function f(n) which takes a program as input. If n runs forever, return false, else return true.

Do this:

function loop() {
if(f(loop)) {
loop()
}
else {
return;
}
}

If "loop" runs forever, quit. Else run forever.

[Lionhearted]

So... Can this be solved or not, or can it be solved in theory but not computed?
If so can you estimate the value?
Were good at estimates... Yes we are.

*nods and smiles* This is the part where I try to look pretty and attentive when I really understand bugger all...
Me and numbers... Don't mesh very well.

[Draco18s]

So... Can this be solved or not, or can it be solved in theory but not computed?
If so can you estimate the value?

Given that the (estimated) value of BB(12) exceeds Graham's Number* I cannot provide an estimate for BB(100) much less passing the value of that back into the BB() function itself.

Suffice to say, that if every plank-volume of the known universe were used to store a decimal value, and that this memory space was used to hold the number of digits of the estimated value, there wouldn't be enough room. In fact, I'm not even sure that'd be enough space to hold the number of digits of that value.**

* http://en.wikipedia.org/wiki/Busy_beaver#Known_values

** Knuth-Plank-Volume-Notation?

[Halinn]

BB(BB(100))

BB(BB(xkcd number))

[Sengir]

So... Can this be solved or not, or can it be solved in theory but not computed?

The halting problem is proven to be undecidable. In layman's terms, the halting program is about creating a program which takes an arbitrary program as input and outputs whether the program runs into an endless loop or not, and it can be proven that such a program cannot exist.

[Mystweaver]

You guys are are so easily side tracked!

This topic started as a reference to "the wireless matrix" where over in the states, the FCC is considering opening up a wavelength for communal access of wi-fi enabled devices (whether this is your PC, tablet, TV, microwave, dishwasher, CCTV system or whatever).

Somehow this got immediately changed to the colonization of planets outside our solar system (couldn't see the like for this jump).

Now you are chatting about semantics of which ma thematic numbers are actually infinite or not and whether it is feasible to record and remap the human brain into a clone.

All very interesting I have to admit, but never-the-less rather off topic.

By all means continue. I am finding it all quite fascinating (IMG:style_emoticons/default/smile.gif)

[Draco18s]

BB(BB(xkcd number))

As the Busy Beaver numbers grows faster than any computational equation, BB(BB(BB(100))) will be larger still (probably).

BB_BB(100)(BB(100)) is definitely larger.

[Eisen]

1) I already did the math that 0.01G of acceleration gets up to 0.04c in 4 thousand years, there's no way in hell 0.0001G hits that in 10 or 20 years.

2) 3,155.76 meters per second is not "about 0.015c." You forgot about 3 extra zeros. The number you're looking for is 0.0000105c

3) Your total distance after 1 year is 6.4 * 10^-7 light years. After 100 years, you've gone 0.006 ly. Even a full 1000 years you've only gone 0.64 ly.*

*This math is easy: distance in meters = 1/2(acceleration in m/s)(time in seconds, squared). Tip: a light year is 9,460,730,472,580,800 meters (exactly) using a year of 365.25 days.

His first idea was right. The math was a bit off.

9.8 m/s * .01 * (60 *60 * 24 * 365 = 31,104,000 s) = 3,048,192 m/s at 1 year. That is .010168 % LS. Somehow you're dividing by an extra 1000.
Accelerate at 1/10 m/s^2 for 31 million seconds, and you will be going a tad over 3 million m/s, not 3000 m/s.

[Draco18s]

His first idea was right. The math was a bit off.

9.8 m/s * .01 * (math)

That's for 0.01G, which was my post.

His was 0.0001G

[Halinn]

As the Busy Beaver numbers grows faster than any computational equation, BB(BB(BB(100))) will be larger still (probably).

BB_BB(100)(BB(100)) is definitely larger.

I'm not about to get into a full-on big number game, but you might enjoy this thread from the xkcd forums. It's only supposed to deal with computable numbers, but I can tell you that I stopped being able to follow it after a couple of pages.

[Draco18s]

I'm not about to get into a full-on big number game, but you might enjoy this thread from the xkcd forums. It's only supposed to deal with computable numbers, but I can tell you that I stopped being able to follow it after a couple of pages.

The Busy Beaver sequence is non-computable and grows faster than any computable sequence. It's even been proven to be non-computable.

But yes. Large number functions in that thread boggle my mind. I can only categorize them into Ackerman Sequences of large size (and the ones where people failed to realize just how big some of the previous numbers were).

[Halinn]

The Busy Beaver sequence is non-computable and grows faster than any computable sequence. It's even been proven to be non-computable.

I know. It was just a disclaimer for the thread. I'd wager that some of the insane functions they defined in that thread are quite a lot larger than BB_BB(100)(BB(100)), but I'm not well enough into mathematics to be sure.

[Draco18s]

I know. It was just a disclaimer for the thread. I'd wager that some of the insane functions they defined in that thread are quite a lot larger than BB_BB(100)(BB(100)), but I'm not well enough into mathematics to be sure.

Given that there is no estimate for BB₂(2), there's no real way to be sure. Remember, that's the maximum number of steps a computer capable of solving the halting problem (for a standard Turing machine*) can take, if it has 2 rules.

*That is, it is a Turing machine capable of computing BB(n)

[Lionhearted]

Can't... resist... any... longer!
NERDS!

[Draco18s]

Can't... resist... any... longer!
NERDS!

Yes. And?

Edit:

Blew a hole in one of the reigning champion numbers from page 4. Not sure if it had been pointed out in a later post (between 5 and 31, inclusive) as there are a lot of pages.
http://forums.xkcd.com/viewtopic.php?f=14&...268418#p3268418

[Lionhearted]

Embrace it? (IMG:style_emoticons/default/nyahnyah.gif)

[Draco18s]

Embrace it? (IMG:style_emoticons/default/nyahnyah.gif)

I cast magic spells program computers for a living. (IMG:style_emoticons/default/wink.gif)