The problem, of course, is that if you start with faulty premises, no matter how clever your reasoning is from that point, it's all basically wasted effort. The most obvious faulty premise here is that a god must exist. Sure, it's possible that some supernatural being exists and is the direct cause of our universe, but without evidence it is pure speculation. It certainly shouldn't be treated as an axiom!
What I find more interesting, though, is the persistence of attributes that are frequently assigned to god, and then used as further premises on which to base even more pointless philosophical speculation. Coming up with explanations for questions such as, "If God is good, why does he let bad things happen to good people", is a complete waste of time if the premise "God is good" is wrong, yet very smart people persist in doing exactly this.
The most common attributes people like to gift God with are:
- Omniscience (all knowing)
- Omnipotence (all powerful)
- Omnipresence (is everywhere)
- Omnibenevolence (all good)
My best guess is that people tend to assume that any being capable of bringing an entire universe into existence must be extremely powerful, and these attributes seem reasonable ones to expect that being to have. Omnibenevolence is a bit of an exception though, since I think it's probably easier to argue that the universe was created by a malevolent being, given the overall amount of suffering experienced by so many living things on Earth.
However, now that we have the ability to create worlds of our own, via computer simulation, I think we can look at the nature of God a little differently. Games developers are well aware of our own limitations when it comes to interacting with the worlds we create, and so in what follows I will look at each of the traits mentioned above from the point of view of the game developer and his creation.
In theory, we can gather any information we want from a game world. We control all of the data. We can pause the simulation at will and collect whatever we want, giving us the seeming ability to know everything. The problem is that it's not always a straightforward thing to do to gather all the relevant pieces of information, analyze them, and then output them in a way that is useful and intuitive.
The other problem is that each act of data gathering and interpretation needs to be an active choice. We can only know the things that we've thought to look for. If there's something really interesting happening somewhere in the game world but we didn't know to look for it, we will miss it.
It seems fairly obvious that we can manipulate the worlds that we create. We usually give ourselves the ability to do things like spawn objects into the world, move things around, maybe even change the state of a creatures 'brain' to make it behave differently. If we want to do something novel that we didn't give ourselves the ability to do already, we can usually add these abilities in as needed (Especially if we're using Runtime Compiled C++!)
What we can't do is things that we don't have the skill/knowledge/ability to figure out how to do. For example, if your AI uses neural networks to store its knowledge and representation of the world, where facts are not stored as discrete pieces of information, but rather distributed across multiple link and node weights, good luck figuring out how to manipulate that in a deliberate way to achieve a specific result, without getting a whole heap of unintended side effects.
So power is limited by the intelligence and skill of the person, and this then limits exactly what powers we can wield in a simulation.
We can pause a simulation and look anywhere at anything, but to actually look everywhere at everything? That takes a lot of effort, which is a big deal, even if time were not a factor, because just as most people can't memorize the entire contents of a book, being able to look at everything in a game world and remember what you've seen in any useful way is not a given thing.
Also, when looking at things, we run into the problem of scale. If you look at, say, a forest in the real world, whether you're looking at the scale of the trees, the insects, the microbes, or the atoms, there are very different things to observe. Being able to be everywhere and see everything on every scale is a pretty tall order, and in the end, like omniscience, it comes down to looking for specific things, and writing the tools to see those things. But if we don't think to look for something, we're going to miss it.
It would certainly be possible to be kind and caring towards the creatures in your simulated world. If your world involves some creatures killing other creatures you've got a tougher problem of explaining the way you set up your world as 'kind' or 'good' or 'just', but this is hardly a problem unique to simulated worlds.
What is interesting, though, is that in so many of the game worlds that we create, if there is a way to be an asshole, some people will try it. There is a reason that so many games have morality systems and allow good and evil choices. This is true whether you're playing a character in the world or take on the role of master controller in a god simulation game.
The point is that, even if you can formulate a consistent definition of goodness and the world you've created allows for this, there is no reason to see this as any kind of requirement. It's simply just one possible way you could control the world, but it doesn't seem any more likely than being an evil god, or an indifferent one, or a mischievous one.
I don't know if other people find discussions like this interesting, but hopefully I've inspired some other games developers to think a bit about our relationship with the worlds that we create. If the creatures in those worlds could ponder the nature of their creator, what would they think of us, based on the things we do and the way we treat them? Right now we don't have any real ethical issues to worry about because our created worlds are not that sophisticated, but if we one day reach the point that the creatures in these worlds start appearing to exhibit something like self awareness, we will have to take these issues seriously, and it will be a very interesting time.