I'm not following some of that argument. Batteries? Electronic failure?
I replaced my D100 because its raw file handling was crap, and the jpeg output was poor at best. I replaced that camera with a D2X because Nikon was taking forever to release the D200 and I wasn't willing to keep waiting. I am buying a D700 now largely because the D2X is a boat anchor. I am moving to a D700 rather than a D300 because I think its sensor's format and characteristics will add value to what I use a camera for, and heavy discounting has brought the price down to where I think this camera is a strong value buy. It helps that I no longer own any DX format lenses.
Now, did my D2X stop being a useful camera when the D3/D300 came out? Of course not. I will probably keep it as a companion to my new body for situations where it is strongly suited - situations where its crop sensor or high frame rate bring value and maybe make the difference in making the shot.
FWIW and FTR, I am an amateur photographer, like most people here.
Batteries are the single place where an increase in performance also indicates an increase in lifespan. That is, a newer battery will outlast an older battery by a significant margin prior to component failure.
In my direct experience, DSLR life is ended because of component failure more often than not, and newer cameras don't appear to be changing the component failure rates in any meaningful way.
That is, you said
FWIW, I have a feeling that we're seeing levels of performance from bodies now that will cause their lifespans to lengthen.
I'm disagreeing and saying that performance has little to no impact on lifespan other than in terms of batteries where the increased performance is also bringing an increased lifespan. In other words, you're basically appearing to say "There's a newer model out that performs better, so it'll have a longer lifespan." I'm saying "A particular body's lifespan has nothing to do with its models performance and everything to do with its components- and components aren't getting noticeably better."
To put it another way, if you purchase a new D2xs and a new D3x on the same day, I've seen nothing that indicates that the D3x will fail any later than the D2xs does. In other words, your dollar of performance doesn't change based on a newer model, and still needs to be cheaper to make sense over getting better performance by buying better glass by a multiplication factor of somewhere in the region of 3:1 or you're throwing out more money than you have to so you can get the results you want.
Let's pull some fictional prices up to illustrate the point and give the pro-level bodies a 4 year lifespan but give the prosumer and pro lenses equivalent lifespans and depreciation rates (20%):
Year 0: D3x $7500 D2x $3500
80-400VR $1500 200-400VR $5500
Total $8500 $9000
Year 4: D5x $4500 $4500
Total: $13,000 $13,500
Equipment Value: $5700 $9000
Even in this case, where we give a $500 price advantage to the "better" body- when we get to year 5, the D2x choice is going to leave us with more equipment value because we've basically spent the most money on the shortest-lived component. In fact, if we even go to say "Hey, the D3x is better than a stop over the D2x, you'd have to get the 2.8 to get everything the same!" you're still not going to cover the residual value spread with the ~$8000 lens! Even more- if you take the ~$8k lens at that point and sell it for a 200-400VR at year 4, you'll actually pay for the D5x by going with the cheaper body up front.
