Actually, this has already been done. Leica has done just this with their Digital Module-R. It mounts onto newer R8 or any R9 film SLR simply by removing the film back and replacing it with the DMR.
Older R8's need to o back to Leica for a modification of some sort related to mounting the back. So, there's something they did to the cameras along the way to allow the scheme to work.
One thing I note, is that the size of the imager used in the DMR was limited to the APS-H format (aka 1.3x FLM like the older Kodak 6 MP DSLRs). Full frame won't fit properly.
See, the sensor must have circuitry around the perimiter of the photodiode array (the sensing area) to work. OK, you say, film is always wider than the imaged area too, so what's the deal. Well, film senses on it's surface, or close enough to it as to not matter.
An electronic sensor has it's active elements buried fairly far down inside. There's a lot of stuff on top, like IR filter, bayer array and silicon covering. So, you have to align the sensor such that the active area winds up at the level of the film rails, at the focal plane. That means the sensor has to actually stick forward of the film rails, inside the opening.
So, we see where we have to limit the sensor to APS-H format or possibly the APS one (1.5x FLM) for it to fit. No big deal, right?
Maybe not. But, may be it *is* a big deal. I wonder if some of the film SLRs have their shutters closer to the focal plane than others. I'll bet that there is an issue where many cameras have shutter assemblies that are too close to where the sensor face would have to be. There have been a lot of different materials used for shutter blades. Some are stiffer than others, while some are thicker than others. Not to mention hundreds of mechanical schemes to get the fool things to work.
I think this is the major problem. I think this is why Kodak chose to use only a few camera body models. They picked ones that either didn't have an issue, or ones that the OEMs were willing to alter to allow the sensor to fit.
Otherwise, the scheme *does* appear to be rather simple. You can argue that you need to have communications bewteen the body and the back for automatic operation. However, there are many FSLRs that offer this, so no real problem.
But, even if you didn't have any fancy electrical communications, all you really need is the old fashioned strobe PC socket to fire the back. Use a simple cable to hook it up. *You* could set ISO on the camera *and* the back manually, and in this case, the back would be no different, as far as the camera was concerned, than film. However, just like the old days, you screw up the ISO settings, and you screw up the shots!
No, I think the entire technical issue is one of mechanically fitting the sensor into the body such that the active elements line up with the focal plane and don't interfere with the shutter blades.
But, let's say that all this could have been taken into account when designing the F6, much like Leica did with the later R8 and the R9. So, let's say that the F6 *can* take a digital back. Now where is it?
Well, I can also see an economical issue. A D-back plus film body will cost more than an entire DSLR. All of the same costs exist in both, plus we now have an entirely new mechanical package added to the list - that of housing the back itself.
Look at the price tag of the Leica DMR: $6000 (the R9 is another $2500). Now, Leica is a low-volume outfit, so we can't use those prices directly. We can lower our price by half if we piggy-back, say, the D2X D-bits into a F6-NDB (Nikon Digital Back for F6). That's still $3000 for the NDB and, what?, $2000 for the F6. That's as much as a whole D2X to begin with.
Which would you choose given the same price tag: Integrated D2X or piecemeal F6-NDB?
Personally, I'd choose the F6-NDB, but I'm a known looney, too!
Now, we have to raise the NDB price if we don't sell as many NDB's as we do D2X's. Perhaps the NDB needs to be $4000 because it isn't as popular. That'll make it even less popular. Now, which would you buy if there were a $1000 price difference? How about $2000?
The whole thought process derails at this point. We don't know what the actual costs would be, so we can't begin to figure out just what would happen in the marketplace with a NDB.
I bet a box of donuts that the Nikon Marketing group has the same problem, and they have more data to go on. So, for now, there's no NDB....