Is the Carter(Doomsday) hypothesis valid?

[sevenperforce]

19 hours ago, wumpus said:

There simply wasn't attempt to check this type of thing, which is why it is such a crank paper.  You could check it against other species and find out your methods are useless (species become extinct because humans (or changes caused by humans) destroy their habitat.  Is has nothing to do with how many individuals have lived since the species evolved.  Or you could calculate rigorous error calculation on the known accuracy of your assumptions.  Since all of them are purely assumptions, watch your error hit +/- 100%.

Exactly.

[Wjolcz]

What the hell is Carter hypothesis? If anyone could be that kind to ELI5 it to me because I'm either too dumb to understand the Wikipedia article on it or it's just a bunch of philosophical what ifs with equations in between.

Edited May 30, 2018 by Wjolcz

[magnemoe]

4 minutes ago, Wjolcz said:

What the hell is Carter hypothesis? If anyone could be that kind to ELI5 it to me because I'm either too dumb to understand the Wikipedia article on it or it's just a bunch of philosophical what ifs with equations in between.

In short its the advanced version of https://forum.kerbalspaceprogram.com/index.php?/topic/102649-so-you-have-a-plane-on-a-conveyor-belt/ 
I have extreme problems with it myself. 
I see it a lot like the fine tune hypothesis: in short if lots of the physical constants had been more than 1% different  life as we know it would not exist, the configuration proves that its extremely likely the universe is created by an god or god like aliens. 
However if the configuration was different we would not be around anyway so these scenarios is pretty irrelevant.  

 

[Wjolcz]

6 minutes ago, magnemoe said:

In short its the advanced version of https://forum.kerbalspaceprogram.com/index.php?/topic/102649-so-you-have-a-plane-on-a-conveyor-belt/ 

[magnemoe]

22 minutes ago, Wjolcz said:

That was an nice link. 
Now this is pure evil posted just for the destructive power. 
http://tvtropes.org/

[DAL59]

41 minutes ago, Wjolcz said:

What the hell is Carter hypothesis? If anyone could be that kind to ELI5 it to me because I'm either too dumb to understand the Wikipedia article on it or it's just a bunch of philosophical what ifs with equations in between.

 

 

[Mad Rocket Scientist]

On 5/30/2018 at 4:00 PM, Wjolcz said:

What the hell is Carter hypothesis? If anyone could be that kind to ELI5 it to me because I'm either too dumb to understand the Wikipedia article on it or it's just a bunch of philosophical what ifs with equations in between.

The best example of it I've heard is this (I think it's a true story):

Imagine your fighting a war against an enemy, who has some number of tanks. You have captured one of their tanks, and you know that they are numbered sequentially, starting from 1. The capture tank is numbered 1,293. According to statistic theory (I don't know enough about it to define it properly) you should assume that, if the tank was selected randomly, it is likely to be close to the halfway point of all tanks that near to either end, so you should assume that the enemy has about double that number of tanks.

If you replace tanks with humans, the doomsday hypothesis says that you should assume that you are human number 1/2x, where x is the total number of humans that will ever exist, therefore humans will go extinct (relatively) soon. Note that this hypothesis does not put any limit on the maximum existence of humans, or specify a great filter, it just says that this is slightly more likely.

I see that Nick Bostrom has written something on it, and I like his articles, although I haven't read this one yet: http://www.anthropic-principle.com/preprints/ali/alive.html

[DAL59]

Maybe it doesn't mean extinction, just immortal robots that don't reproduce.

Also, how can future events retroactively change probabilities?    

 

[Superfluous J]

30 minutes ago, Mad Rocket Scientist said:

Imagine your fighting a war against an enemy, who has some number of tanks. You have captured one of their tanks, and you know that they are numbered sequentially, starting from 1. The capture tank is numbered 1,293. According to statistic theory (I don't know enough about it to define it properly) you should assume that, if the tank was selected randomly, it is likely to be close to the halfway point of all tanks that near to either end, so you should assume that the enemy has about double that number of tanks.

(One of) the massive problem(s) with this is that while this is where the hypothesis starts, it's not remotely where the hypothesis ends.

What he hypothesis says about humans and doomsday is like if you capture an enemy tank FACTORY and read that factory's stamping machine and see that they just made tank #1293, then you can guess that they'll make approximately 2586 tanks before the war is over.

What the hypothesis should derive about humans from the tank example is that if you pick a random person on Earth and ask their name, you can guess very roughly that alphabetically half of the current population has a name higher in the list, and half lower. That's it.

[magnemoe]

7 hours ago, 5thHorseman said:

(One of) the massive problem(s) with this is that while this is where the hypothesis starts, it's not remotely where the hypothesis ends.

What he hypothesis says about humans and doomsday is like if you capture an enemy tank FACTORY and read that factory's stamping machine and see that they just made tank #1293, then you can guess that they'll make approximately 2586 tanks before the war is over.

What the hypothesis should derive about humans from the tank example is that if you pick a random person on Earth and ask their name, you can guess very roughly that alphabetically half of the current population has a name higher in the list, and half lower. That's it.

If you capture the factory you know the exact number of tanks it has made by seeing the serial number in use. It will also not make more tanks for the enemy. 
If you capture one random tank with serial number 1293 your best guess is 2586 total, yes its very inaccurate but the best you can do. 

However the tank example will work just as well early in the war then few of the tank is made as late with huge numbers. 
You could on the same time ask the doomsday question with the same validity for 100.000 years ago. 
On the other hand you could not as it was not invented yet. you could only ask it after its invention, and you have to ask it before its forgotten or replaced  
 

[kerbiloid]

After capturing tank #1293 I would presume that they are making as many tanks as they can (over 1000, not just a dozen). That means that they need them as soon as possible, and all new tanks immediately get sent to the front.
So, I would presume there are 1400+ tanks manufactured in total. (1300 + 2..3 battalions in process of shipment.)

Edited June 5, 2018 by kerbiloid

[DAL59]

Better example:  If you capture tank #2, #4, and #6, there probably aren't going to be very many.

[Superfluous J]

3 hours ago, magnemoe said:

If you capture the factory you know the exact number of tanks it has made by seeing the serial number in use. It will also not make more tanks for the enemy. 
If you capture one random tank with serial number 1293 your best guess is 2586 total, yes its very inaccurate but the best you can do.

I can't tell if you're agreeing with me or not. You seem to have restated my thought without the "this doesn't actually predict anything about the future" part.

[wumpus]

I'd be curious how often the serial numbers were changed specifically to meddle with such intelligence.

I've heard of it being done often enough in the commercial world, where a higher serial number* might feel safer for customers.  There have been a few software releases that started as "2.0" for similar reasons.

* When Seymour Cray was making supercomputers, plenty of his customers would fight/pay for "serial #1" of the next new computer.  Often new models were introduced to sell as many "serial #1s" as possible (or claim "first to DoD", first to Government", "first to the private sector").

[kerbiloid]

56 minutes ago, wumpus said:

I'd be curious how often the serial numbers were changed specifically to meddle with such intelligence.

I've heard of it being done often enough in the commercial world, where a higher serial number* might feel safer for customers. 

Use UUID/GUID wherever it's possible.
Always - when the entity can be exported/exchanged by two or more copies of your system.

[YNM]

20 hours ago, DAL59 said:

how can future events retroactively change probabilities?

It doesn't. It may change the probability for that time but it doesn't mean that past events were less likely. Just like putting dykes around a river, regardless of whether it was likely to have flooded over it becomes less likely after the dykes are in place.