I think the simulation of the brain is inevitable barring a metaphysical intrusion there is no reason it should not at some point be possible. Will we do it in 50 years? I think it is likely that we will get to something approximating it. One of the themes of invention is that usually we make our greatest breakthroughs by building upon nature instead of replicating it the prime example being heavier than air flight as we do not copy birds. A salient reason for pursuing brain simulation is to develop artificial intelligence, neural prosthetics, medical advances, psychological understanding, etc. It may be that we accomplish all of those things by using superior methods or innovations than the more difficult ones presented to us by nature.
But to give my answer on your three points:
1. I think the computing problem will be rapidly overcome this century if not much sooner. As we move towards exascale supercomputing infrastructure and extremely large data storage technologies we are quickly approaching the raw capacity to tackle this problem. Software and scientific understanding are another question.
2. Maybe I'm being simplistic but I don't actually see much of a problem with this. We already have fairly sophisticated metrics for measuring biological items as well as categorizing non-numerical phenomena.
3. This is closely linked with the first point and I'd actually disagree that aren't closer to it than people think. Though my definition of close is probably on the order of sometime before the middle of this century.
For me the real question is will brain simulation research efforts and experiments offer us the practical benefits and progress towards the areas I mentioned before and I think the answer is yes. Whether or not we create an exact replica of the human brain is less important to me personally.
Very nice post. Biological models and simulations (also brain simulations) are definitely an interesting and increasing area of research, and given all the work that is done in that area, there hopefully will be some benefit.
The general problem with biological models is that once they reach a certain complexity, they lose their prediction capability because you get lots of parameters with high uncertainties going into them. That is not to say that it's a completely useless field of research, and I do see a potential for biological modeling combined with medical imaging to have an impact on medical treatment, if the models are kept simple enough. Having worked in the field a bit, I consider most (though not all) of the biological modeling that is going on to be useless from a practical/application point of view. I think if we look at it from a medical/application point of view, people should try to build simple models with 2 or 3 measurable input parameters. Anything else is just an interesting mathematical model.
This of course takes us the issue of medical imaging, which is mainly CT and MRI if you want to look at what is happening inside the brain. PET has a poor resolution. Functional and anatomical imaging techniques of various kinds at µm resolution of the entire brain (this is what you would need) are not really in sight, but who knows how that might change?
The development of artificial intelligence is a different matter and it will develop further and do lots of things both good and evil. I like your analogy (the brain being the bird, and artificial intelligence being the airplane) and I think it applies very well here. However, I doubt that artificial intelligence will be modeled primarily on the human brain.
So to sum it up, I agree that brain simulation can help to develop new medical methods, but I still find it dishonest for a research project starting now to say that they are going to simulate the entire human brain. We're far away from that, and as I said, once you have a certain number of parameters in a biological model you're lost, and you'll certainly get there if you simulate the entire brain.
Then again, when Max Planck started to study physics, a professor told him it would be a waste of his talent because physics wouldn't develop anymore. So, as it always is in science, whatever I'm saying now might be completely wrong, obsolete or even laughable in 100 years...