I'm a little out of my depth here, but unless I misunderstand you, I think you're ignoring the fact that if they have better tech to put into an SLR-lite package, that tech exists to be put into a real SLR as well. And people aren't paying several times as much for SLRs now just because they have too many hundred dollar bills to fit in their wallets.

Also, in regard to the sensor being exposed, there's no reason a future SLR body couldn't have a shield that comes down when changing the lens.

Fixed-lens digital cameras have several interesting advantages over interchangeable lens cameras. For starters, there's no particular constraint on the size of the sensor. Smaller sensors can use smaller lenses, and thus the whole package gets smaller and lighter. Furthermore, because the whole system is sealed, dust is just never an issue. And, on top of all that, when you're willing to go with an electronic viewfinder or a rangefinder system (i.e., get rid of the single-lens-reflex mirror and pentaprism), you can go with different styles of lens designs than allow the rear objective of the lens to be closer to the sensor (*).

All of that said, there are still fundamental advantages to interchangeable lenses. No all-in-one zoom lens can ever offer all of the qualities that can be found in more specialized (and expensive) lenses. And, for that reason alone, interchangeable lenses will be here to stay for a long time. An interesting question is what happens when the high-end integrated-lens cameras intersect in price with the low-end D-SLRs. I'll argue that D-SLRs win, not because they're necessarily any better at taking pictures, but because it's easier to sell somebody on future expandability (of which they may never take any advantage). Also, at least today, the D-SLRs just go faster, addressing probably the biggest complaint with other digital cameras.

(*) However, where film is pretty must just as sensitive to light coming in at an angle versus straight up-and-down, digital sensors are more particular about the angle of incidence. This is a real problem with the Epson R-D1 (basically, a Leica M-body compatible digital camera using the Sony 6MP sensor that's used in many D-SLRs). Because rangefinder lenses mount closer to the film plane, the angle of incidence at the corners is higher and, as a direct result, the R-D1 has problems with darker corners when using some wider-angle lenses.

I don't think the shield thing is really a complete answer in that the "body" must be exposed when changing the lenses mainly due to protrusion of the lens into the body. Once dust is in the body, it is disturbed by the mirror action, etc and the charges applied to the sensor attract the dust. High humidity can actually "weld" it to the sensor. There is in essence a shield (although incomplete) currently in place in the form of the mirror. If people are considering buying a digital SLR with interchangable lenses, do not ignore this as it can be enormously frustrating. if you get even minute dust "threads" or fibers on the sensor, it can require extensive work in Photoshop. One can minimize the effect by shooting with as large an aperture as possible which will not bring the dust into focus. I have cleaned sensors before and it is nerve-wracking (or is it nerve-racking). Actually with a Nikon, there is a clear protective shield on the top of the sensor but it still attracts dust. I take every precaution I can to keep it from happening short of never changing lenses (which would kind of defeat the purpose of the camera design).

I not trying to ignore that current "real" SLRs are better than SLR-lites at all. In fact, that is exactly why people pay more for them. However, if you accept an extension (and I don't think it is that big a step) of Moore's law since it largely involves the minaturization of electronics, tomorrow's SLR-lite (selfcontained) will be where today's real SLR (with interchangable lenses) is which will be moore (spelling pun intended) than acceptable for DWallach to shoot a picture of his daughter at night under stadium lights without renting or buying a $7000 lens.

I think the technology is there today if it was repackaged. Now taking Hybrid8's point of Nikon's margins and how such a camera would cut their own profit. He is correct that Nikon has no motivation. There is, however, no such restriction on say FujiFilm, which manufactures SLR bodies only (in the pro class) that use Nikon lenses, from duplicating in a selfcontained package what 98% of pros shoot in a interchangable lens system. Now if you are a professional photographer or advanced amateur, would you rather lug around 30 lbs of gear or 5 lbs?

All that being said, I got to thinking about noise levels at high ISOs. Try pushing ASA800 film to 1600 for low light applications and look at the "noise". Here is link about preceived noise levels in actual practice.

Moore's law has failed in the digital camera development/marketing cycle. Time and time again. Digital camera technology available to the consumer (therefore at "consumer" prices) has just not advanced as a whole at a rate equivalent to (for example) desktop hard drive size/deployment, desktop/mobile processor speeds, etc.

There's a lot more to a digital camera than a mere sensor. And in fact the whole ball of wax, even though it's becoming somewhat of a commodity market in a number of respects, is much more than a sum of the whole (whole bill of materials anyway). A lot can be done to improve all-in-one as well as DSLR cameras with technology that exists today. I can argue that a lot more could have been done each and every year over the past 6 beyond what WAS done. But, for one reason or other, it wasn't. Perhaps it was a matter of market acceptance/timing. Who knows.

To make a rather open statement such as "at some point in the not so distant future we'll be using compact cameras that rival more complex and expensive systems currently in use today" is rather useless. That's a simple economic fact that has been proven time and time again in countless industries. Now if you can pin-point an exact date and conceive of a specific solution, then you may have a chance to make quite a bit of money. There's always some under-serving going on in all markets.

I think the most important point I want to make is that there wil always be room for bigger and better. That means when you think you've got the killer all-in-one, someone else will have the super-WMD bad-ass discrete that's even better. The best technology is always going to be the bleeding edge. It's always going to be the new development that companies have yet to recoup their investments on. Such technology is going to appear first in higher end gear first, regardless of where the bar is to mark that high-end point.

So you show me the $500 camera that's as good or better than today's US$8000 Canon SLR and I'll show you a $5000 camera that's an order of magnitude (or a couple) better than that. Just let's not get anyone expecting to see that in the next two years because if past camera development and marketing efforts are any example, it's just not going to happen.

Hey, power to technology. I want one of those future small cameras. Of course I also want the big sh*t-kicker I can use to clobber any would-be burglars.

Bruno