At least so far, I've found much (most?) of this thread a bit amusing.
Even if we ignore all factors other than resolution, the D800 (not available yet) is slightly (but only slightly) ahead of the P30, which I believe came out in 2007. It's still not yet caught up with the P40+, which came out in 2009 though. Interpolating between those, just in terms of spatial resolution, 135 format is running about 4 years behind MF.
Most of the other factors remain to be seen, but I think it's fair to guess that the spatial resolution is probably the highlight of the D800. It almost certainly will not beat those 4 year-old MF backs in other respects -- and on quite a few it will still almost certainly lag behind them. So, what we have is at least a 4 year "lag" between a given level of image quality in MF, and something that (at best) might start to approach the same level in 135.
Looking at things from the other direction for a moment, I'm a bit reminded of arguments I heard between proponents of various kinds of computer systems 10-20 years ago. Users of workstation from Sun, Silicon Graphics, Apollo, HP, etc., would happily point out that there was no comparison. Their workstations were not just faster, but much more sophisticated, dependable, etc., than those slow, cheap, unreliable Windows and Mac boxes with their cheap mass-production processors.
That argument's pretty much over though. Though there's still a niche for high-end servers with other processors, Intel boxes have pretty much taken over the workstation market and most of the server market too -- Windows machines and Macs both run Intel processors now, and the RISC workstation market is purely historic. Oracle, HP, and IBM still sell high-end servers with other processors, but it at least seems like the market share shrinks a bit every year.
It's also true that in the process some sophistication has been lost. A current Windows or Mac machine still doesn't do networking quite as nicely as an Apollo workstation did 20+ years ago. It uses a lot of brute force to even come close to matching an Amiga for around the same time. Nonetheless, Apollos and Amigas are long gone and mostly forgotten. Some of that is undoubtedly due to management problems, but a lot more is due to the simple fact that the low-end "garbage" they once barely even bothered to sneer at, got enough better at their specialty (while remaining cheaper and easier for most people to use) that almost nobody was willing to pay enough extra for their strengths to support the companies.
To summarize: I don't think it's anywhere close to given that Nikon (or 135 in general) is set to catch up with MF in general -- but at the same time, if I were Phase, Hasselblad, etc., I might not be exactly nervous, but I'd definitely be working hard at ensuring that I kept that 4 year gap open. I'd also be working really hard to matching the features of 135 like ease of use and convenience as much as possible to make it as easy as possible for the high-end 135 models to act as a "gateway drug" for MF, rather than the other way around. In the long term, inexpensive beats sophisticated nearly every time.