Assuming that silicon neurons, brains, and organisms will come to pass, in some form or another; projects will develop investigations into the application of the technology for mind-transplanting. I have only a vague vision of how this might go. There are a number, and variety, of problems that immediately come to mind.

On the technological side, the foremost problem, to my way of looking at it, is the process of data transfer. We know that all data is not necessary — for we lose it everyday — lots of it. But there is an ominous sign present in a very basic, pervasive aspect of the evolved condition of our brains — you don’t get new cells. This should have been easy for DNA to handle. The advantages seem obvious. But it didn’t happen. It must not work. New cells must upset the functional characteristics of neural interaction. They probably throw a monkey wrench into the structure of meaning held in the system.

To tap the mind with "circuits" that take on some of the relative hierarchy of meaning residing in our synaptic structure, is going to require a non-invasive, or at least, non-reactive, method of induction. Yet, the crucible is necessarily an interacting network. It will probably have to do its interacting off on its own, after receiving the information, through non-destructive means.

It seems likely that data gathering will require an incredible apparatus... something like an MRI, capable of resolving a good percentage of all the synaptic conditions throughout the brain, and reading their "schematic" association of neurons. Synapses are very small. There are something like 10¹⁵ of them. You might need to read 100 increments of physical size. Size may have less to do with enhancement than some other factors.

If interaction is not a problem, something as simple as high frequency brain wave information, gathered over long periods of time, might suffice; particularly if some sort of elaborate transformational interface is developed, that would probably require return stimulation. In other words, a feedback-interface might perform co-learning in the artificial matrix, based on the conscious process. Consciousness would induce itself into the matrix, by developing its normal modes of association into that redundant terrain. The maximum number of channels, from the smallest possible electrodes, would be desirable. In the extreme, a huge number of needle electrodes could be involved, and perhaps would tap the corpus callosum. The silicon crucible would be a sort of bowl you wear on your head; a neo-neo-cortex. When you die, they take the bowl off, and plop it on to some sort of robot.

This brings us to the sociological side of the problems. Here, the first question that comes to mind is; would you want to do that? We may never fully understand what we’d be getting involved with. We would probably have to rely on reports of experience, to fine-tune, if not outright re-design, the systems involved. The pioneers who decide to venture into this realm might be facing the greatest test of humankind.