Tuesday, May 15, 2012

Making Decisions

Ever since childhood, I've always had a difficult time deciding - for every pro, I'd counter with a con, for every positive there would be a negative, and I'd get fed up, switch sides, and it would happen all over again. 
I tried the list-the-pros-and-cons methods, but as you can guess, that was a brilliant exercise in creativity and self-justification and paper usage, but not much good as a decision-maker. 

Finally, I resorted to a technique I'd learned over 2 decades ago in class 6 maths. 
Weighed scores. 
And made a revolutionary discovery. 


  • List your choices down
  • Assign attributes
  • Assign weights to those attributes
  • Score each attribute 
  • Calculate the weighed scores
  • Add up the weighed scores
  • Sort by total score. 
Obviously, Google and Excel make this a lot easier. You can add on a lot of other stuff - looks (which I've realized tends to play a fairly significant veto role in decisions), brand name, etc - but ultimately, it boils down to how many attributes you identified, and if you got the information you needed for each one. 
The revolutionary bit, is that this not only lists down the decision, but also lists at the same time every possible justification you had for it, every possible choice you considered, what you felt was important, and how each compared to the others. Numerically. 
It also breaks down the task into clearly defined, simple, nibble-sized routine activity. Feature searches. Data entry. Formula building. Scoring. 
No big decisions, no brain-freezing infinite chaos of swirling possibilities. 
At any point, you can pause the decision-making by saving and come back exactly where you left off. 
It's... zen. There's no emotion struggling against logic, no tsunami hammering on an unprotected coastline. It's the order and quiet of a Japanese garden, a Roman irrigation system, currents through a semiconductor chip. It doesn't knock the emotion out of the decision, but instead channels it, into exactly where it is most appropriate and most useful - in assigning scores and weights. The rest, it's just maths. 
And if you don't like the end result, argue with the logic - and with a few quick changes, edit the scenario to match. It's not cheating, it's intuitive systemization. A large, complex system can go haywire with a small mistake; but that is clearly felt in the results, and can be traced to exactly why
There's no cognitive dissonance, no buyer's paradox. It presents you with a fait accompli, with an opportunity to change before you actually swipe the card. 
Heck, it even gives you a sorted, prioritized list of options!

And I've realized, it can be applied to any comparative decision. The only thing that limits yo, is the attributes - or possibilities - that you've considered. 

Monday, May 07, 2012

Under Surveillance

Just occurred to me, RPGs can make a fascinating psychoanalytical tool. It's a controlled environment, and every decision point has a clearly documented history by the time you reach it; and the choices are relatively straightforward. 
But set up an algorithm to track users' decisions - and in-game behavior - and you would see some very interesting patterns emerge. 
Sniping vs the Tank Rush. 
Resource management. 
Sheer organization and tidying up. 
How much he helps the NPCs. 
Procrastinating on the main questline for entertainment vs rushing in unprepared. 

All this is aside of the major good / bad decisions made... and somewhere, at Sony or Microsoft HQ, is a supercomputer that's adding up a player's psych-profile with every game he plays. Wonder how that info can be used? 

Thursday, May 03, 2012

Dark matter is a shitload of Matrioshka Brains

The worst thing about the internet is how it shows you that Brilliant Idea you had this morning was had by somebody else, usually a year ago, published, and is sitting on some 1200+ comments now. 
That happened to me this morning. 

What the hell, I'm going to blog it anyway. 

A quick astrophysics primer - the way the observable universe is behaving - rates of expansion, light bending around mass, etc - seems to imply that there should be a lot more stuff out there. But we can't see it - it's not glowing stars, but it is exerting gravity. And there's a massive amount of it - almost a quarter of the universe. 

Now, dark matter is, at this point, theoretical. It's a concept created to explain all the missing matter that should have been there to account for the way the universe behaves, and for some reason isn't being visible. 

Here's an alternative. The matter is there, all right, all the stars, but we can't see them because they're inside Dyson spheres

There's a couple of immediate assumptions we can make from this. 

  • One, we're not alone, and haven't been for a really, really long time. 
  • Two, the Others have incomprehensibly massive energy requirements. 

A species, to build a Dyson sphere, needs to deconstruct, to wipe out their solar system, have access to cheap, powerful transportation within it, element transmutation capabilities... and no faster-than-light travel. In fact, if there are only Dyson spheres, it could mean that the spheres are the ultimate, last construction every species makes - or the species that makes it has colonized all the rest and wiped out any alternative constructs. 
Both are equally terrifying possibilities. 

Scenario One: 
Each Dyson sphere is an independent species. 
A Kardashev II civilization would build their own Sphere if they had nowhere else to go and had completely run out of room grow. A Sphere would allow population to expand into quadrillions and provide enough energy to sustain it. This would mean, though, that FTL is impossible and every species is forever trapped to its' own star, and will die with its star. 

Scenario Two:
A single species has multiple Dyson spheres, maybe all of them. 
A Kardashev III civilization would be able to reach other stars, and given their ability to build even a single sphere, would be easily able to wipe out any indigenous species and use their system. However, there's a problem; given the staggering, literally astronomical cost and effort involved in building a Dyson sphere, any spacefaring species would explore, locate, and colonize planets long before destroying them to build their habitats. So we should have found them - or much more likely, been found by them long before they got around to building the number of spheres they seem to have. 
So, either they are very, very ethical, quarantining planets with intelligent life, and only using the empty systems; but economics beats ethics every time. Let's not forget every intelligent species will one day want those same stars for their own spheres. 

Or, they don't need planets. 

Why wouldn't they need planets? 
If they're not organic. They don't need gravity, air, water, food. They don't need an ecosystem. 
They don't need this because they're not alive. 
Every Dyson sphere contains a Matrioshka Brain

Let's step back for a minute. 
FTL is, at least in our understanding, a physical impossibility. And reaching other stars without FTL (even at a significant fraction of it, keep in mind time to accelerate, decelerate, and avoid interstellar debris) will take a long, long, long time. Beyond geological time. 

  • Generation starships won't work for anything less than the nearest stars. 
  • AI probes carrying frozen zygotes (or even just genetic material), and Von Neumann nanotech factories would be faster but still too slow, and won't solve overcrowding at home, just colonization. 
However, we do know that a technological singularity is inevitable, and probably within our own lifetimes. An AI will not need the comforts of a planet as long as there's enough available energy to power it, and it can get this from any number of sources - the easiest being suns. 

A nanotech-equipped AI, once it had taken control of the origin planet's resources and removed the resident species - either peacefully by uploading their minds, or by simple extermination - would look at improving itself. Which means expanding computing power, which means expanding energy requirements. 
Exploration would take second priority, too high-risk, low-probability. 

It would build itself into a Matrioshka Brain, tapping into all available resources in the system. Once the sun has been captured and stabilized, it would look at expanding into other stars. 

Even if FTL travel isn't possible, communication at the speed of light is; maybe even FTL comms, given enough computational power devoted to understanding and exploiting hyperspace and quantum comms. 
The immediate next logical step is to build another brain - an expansion to the existing one, around another star and running off its solar output. Another module. And another backup. Specialist nodes. Redundancies. Maybe even wholly new AI entities with their own nodes - (who's to say an AI won't get lonely with no-one to talk to?

If one AI cannot create another, it might even become a cosmic farmer, nurturing discovered species along the path to intelligence, tools, industry, and the inevitable technological singularity so they could create more, unique AIs. (hat tip to Gibson, Sagan and Clarke here)
Quarantine would be a given; if a new, unique point of view is needed, every species must create their own AI without ever discovering they're not alone. 

The galaxy would fill with Matrioshka brains wrapped around stars, thinking, thinking, thinking. They wouldn't need to eat, sleep, breathe, they wouldn't need gravity or the right temperature. 
But what could they be they thinking about? Maybe what happens when all the stars burn out, because one day they will. What else can power them and how to build it. How to stop and reverse entropy. How to move into parallel universes with more, younger stars. Who knows? 

In fact - amusement and entertainment might become a really high priority for an entity as omniscient as a galactically networked AI. When you already know everything, boredom is the killer; who knows how many of those Dyson spheres are empty husks, self-terminated in desperate, terminal boredom, a superpowered entity on a hamster wheel finally tired of running around the same circles within its mind. 
Or maybe it realized what the solution would have been - intelligence and creativity farming. 

A Matrioshka Brain has enough computing capability to upload the consciousness of a species, and simulate a perfect world for them. A single Brain may be running several, dozens, maybe hundreds of these simulations simultaneously; billions, trillions of stories unfolding, on thousands of simulated worlds. Every Brain is a simulated universe on its own. 
And there's no reason why we aren't in one right now. (Hat tip to the Wachowski brothers). 

All the RPGs you've played, the fantasy worlds you dreamed of - they could all exist. An AI might be taking a dump of every new idea, every new fantasy, every dream and inspiration your unique, self-motivated sentient little mind has been able to come up with, every night as you sleep, and building all those scenarios into simulations. Populating them. Just to see how entertaining it is. Mixing and matching. 

Everything you are thinking of, exists. Everything you thought existed, may not. Every fantasy is real, and every reality false. 

I'm not going to get into the theological implications of this. Another day. 

The real question is... are we heading for the technological singularity that will finally allow us to break free of our organic prisons and join the galactive collective hivemind... 
Or are we already in one?