I had a Pentax K10D then K20D and then a K-7 and they all had in body stablization which made every lens that you put on the body stabilized up to 3 stops and it was very effective. No added cost to lens manufacture and no need for lens redesign, every lens now stabilized.
Hi Lance,
Pentax user here as well, about to switch to a D700 too. I'm about to sell my K10D, K-7, 16-45 and 50-135, plus a few other accessories.
I will probably go for a 16-35 F/4 VR plus 85 F/1.8 combo (or a 24-120 F/4 VR), plus battery grip and SB-700 to start with, and I will add some more lenses and accessories over time. Sounds like a good idea?
I agree the Pentax (and Olympus, Pana, Sony) DSLRs have a little edge when it comes to the price you have to pay to stabilize your images!

However, in-body stabilization don't help much when framing since you actually don't experience the stabilization during framing. That's especially true with longer focal lengths, where the slightest hand shake can make framing a real challenge. That's where the in-lens stabilization shows its strength! (Plus it's a bit more efficient, with the VRII and latest IS giving you an advantage of 4 stops or less instead of 3 stops or less with in-body SR in real life use).
But for short focal lengths, where framing isn't likely to be affected much by hand shake, an in-body lens stabilizer would allow the use of cheaper lenses, while the longer focal length lenses would still benefit from the in-lens stabilization.
Now, as we all know, not all Nikon lenses are VR and therefore, having in body stabilization would be a great feature for those lenses that do not have VR. When a VR lens is attached to the camera then the in body stabilization could be automatically turned off by the camera.
Not a bad idea. But what if people stop buying VR lenses? Bad news for Nikon. So much time and money spent designing these VR systems.
I guess the only issue would be whether the image circle would be large enough on an FX camera to allow the sensor to move sufficiently to counter blur. If it can't be implemented in an FX body, maybe it could be implemented in the DX bodies.
If Sony can do it on its Alpha 900/850, I'm sure Nikon and Canon could do it in their FF DSLRs as well. There might be a few technical issues, but nothing that cannot be fixed, I think.
The main issue is, I believe, not technical but financial, since such an in-body stabilizer would likely hurt the sales of VR lenses. Blocking the in-body stabilization when mounting a VR lens could help if you don't sell non-VR lens similar to one with VR. But as soon as you have a non-VR lens with the same features as its VR cousin, the first will probably outsell the second by a huge margin if Nikon has an in-body VR on its DSLRs.
I think they'll keep working on in-lens VR. And VR lenses are getting cheaper and cheaper...
--
If photography can be considered like painting, then I'm still at the preschool "paint with your fingers" level.