I'm no expert or anything, just an undergrad in aerospace engineering who dreamt this up between bouts of sleep and doodling during a boring lecture that I'd otherwise skip had they not started taking attendance for a grade... *grumble*

I was thinking how one could interface with a computer faster, which ultimately lead to me realizing just how much our input and output is slowed down by our physical limitations. You can only click so many times a second, you can only type so fast, you can only read so quickly, you can only recieve a limited ammount of visual aid (ever get confused at a book's 2D diagram of something in 3D?). Of course the next train of thought is probably the same thought by anyone in this particular field... these limitations can be surpassed by direct neurological interface. Great! Now how would one do this on a day-to-day basis?

Neurological implants on a scale of such as to directly interface with a computer would likely require something to stimulate the brain optically, auditorily, and maybe even spatially. Inputs from the user would be thought driven to provide a computer with option selection, text input, voice input, and god knows, maybe even on-the-spot creation of spatial objects, visuals, and situations (useful for 3D drafting programs, or even video games if one were so inclined). Text could go in just about as fast as you think it, reading would occur just about as fast as you could concieve it, all good things, you'd be flying in whatever task you were trying to do on that computer.

But this all needs a nice array of neurological implants, and they'd probably have to be permanently installed in wherever they were. The system would need to be partially built into the user, but of course you wouldn't want any actual obstructions like wires comming out of your head. That's just a bad look. The system would need to be able to cleanly disconnect from a computer and let you walk around. How would it connect? Several ideas I've seen show a physical plug somewhere on the user. This couldn't possibly be a good idea. Such a plug would have skin right up next to it, and skin doesn't bond to metal when it heals (I don't think). I can picture nasty infections happening around the plug where it bumps right up to the edge of the skin that was cut open. Plus, how do you keep a plug like that clean? Thus, a radio or an optical interface would probably be a better idea for a permanent implant, either system would work by broadcasting data either with radio or light through the skin. The skin could be allowed to completely heal, and the implanted tranciever would not have any exposed metal contacts that could otherwise get damanged or dirty.

If it were optical, I don't think there would be problems other than if skin muffles very high-speed data transfer, and I'm not sure if it does. Radio, however, would be a bit wierd, but more versitle. I'd think it would be extremely low-powered, such as to not fry the brain when it's working. It'd also have to be able to ignore other radio signals as to not stimulate the brain in wierd ways while the user is going about their life. Basically, I'm not sure if data transfer between the tranciever in the user and the one outside would be optical or radio.

Powering the neuro-system by batteries means there's a physical plug somewhere on the person's body. Powering it by biological means makes for a more complicated system, where more can go wrong. This is why I think power for the neuro-system needs to come from outside the user without wires. This means either solar light through the skin ("What's that faded black rectangle on your forhead, John?"), or magnetic induction. I think magnetic induction would be alot simpler and solid-state. Plus, an external power source means that the neuro system cannot accidentally stimulate things when the user is not connected to a computer.

The solution I thought would basically look something like this: An array of nerve implants in various regions of the brain would all link up to a small microchip/tranciever package, no bigger than a penny, that would be implanted on the bone just behind one of the ears (that bone's right next to the surface of the skin, very little to interfere with the signal). The user, when interfacing with something, would slip on an ear-piece that has a small tranciever in it positioned such as to be directly over (or very close to) the tranciever placed in the user. The chip in the user would be powered by a small magnetic inductor, where an electromagnet in the earpiece would be activated to transfer power to the neuro-system via magnetic induction to the magnetic inductor in the implanted microchip/tranciever package. The trancievers would transfer data either optically or with low-powered radio.

This system would be fairly simple (all things considered), as there is no on-board power source, and no physical plug, just a solid-state implant behind an ear and some neural wires. The system cannot function without the user being connected to a device via an earpiece, so the user lives a normal life when not connected, dispite outside radio or optical stimulation of the tranciever.

Thoughts? Good idea? Bad idea? Why, why not?

While you appear to have thought through the technological requirements thoroughly, I don't believe this is the major stumbling block. I would have said the difficulties lie almost entirely in the neurological areas, and as a complete novice in that area, I'll leave it alone.

However, to address some of your other points, I remember reading an article in New Scientist some time ago, about creating a metallic surface which the body would bond to - I believe the team used a 'furry' surface, but I can't find the article.
Power by magnetic induction is certainly a possibility, but I'm not entirely certain how reliable an off switch it would be. For one thing, you can get magnetic induction hobs, which could be powerful enough to switch on the device, or possibly cause a dangerously high temperature.
http://www.applianceplanet.co.uk/eshop/groups/BIHI.htm
As for solar power, it has possibilities. If you could get custom shaped solar cells, effectively you'd just have a useful tattoo. I can certainly see some people going for that.

I'm also not very familiar with the neuro-science, which is why I was looking at this purely as a problem of how would one turn such neuroscience capability into a consumer-friendly system. I never thought of the panels as possibly being tattoos, but anywhere other than the head would need a wire snaking through the person's body to, say, the deltoid or wherever the panel was put.

Yeah, the inductor isn't totally an off-switch, but the magnetic field that would be needed to power it from a distance any farther than a centimeter or so (distance between earpiece and implanted chip), let alone fry it from that distance, would need to be extremely powerful, like one found in an MRI machine. Right? I think that's the one that uses very powerful electromagnets... but one with this system wouldn't want to be in a field that powerful anyway if they had metal components like microchips and wiring and the like in them. Even if the material were non-magnetic, like you said, the wiring itself could start heating up in that kind of field.

Electrical overloads could be cured with a fuse in the embedded chip. Too much current, system's permanently off. The user would have an incapacitated system till their next appointment to have a replacement chip installed.

But an off-switch is a challenge, because there are always problems with a physical toggle switch, like if you rolled over in your sleep and bumped it on or something. There probably isn't a perfect solution to the problem, I thought it'd be interesting to look at though.

Maybe something like an IR key, with a unique code for your implant could solve the problem, although you could have problems if someone stole your keys.

mind-hacking... interesting thought...