The only way a brain could ever be "downloaded" is by in-situ molecular-level analysis using advanced nanotech devices. Destructive analysis (serial sectioning and staining) might permit a lower technology to be used to readout limited aspects of brain information. External scanning is and always has been completely off the table.
---BrianW
I don't know about that, I think the brain only really cares about the electrical signals being sent down the axons etc, it just so happens that those signals usually correspond to a certain ionic content at the nodes / synapses. Electrical stimulation of an axon produces the same result in the brain as it would if the axon were stimulated by a 'real world' stimulus.
I agree that simply scanning a mind is highly unlikely to be an option. Although, I'm not totally sure that this 'memory cloning' method would be as bad as it at first seems. I've been writing a paper, I really need to get finished, that discusses the question of what actually happens to our consciousness when we are classed as 'unconscious', when we're asleep for example. It raises what I think is an interesting question. What happens if the person who wakes up is just a 'memory clone' of the original consciousness that only believes it's self to be the original? If this was true, a 'memory clone' produced by scanning the brain tissue would be no more a death sentence than going to sleep; when you wake up, how is it that you can be sure you're the consciousness that went to sleep and not just someone with his memories?
It's also worth considering this with regards to perfect atomic replication of an individual; i.e. if continuity of experience is lost at any point, can you still be sure the resulting consciousness is the original? You also need to be sure that the experience is continuous, not just that it appears so.
If my latter statement is true, that going to sleep 'ends' one consciousness and another emerges on waking with the memories of the last, transferring into another form that doesn't require sleep would eliminate that problem.
bgwowk and homosapiens, you may wish to read the post I just made about the feasibility of MEG scanning a consciousness.