Totally mad idea
[...cut to save space...]
Total madness or a glimpse into the future?
Doesn't work for hardware. This method does work out for software—in fact it's the business model that most audio processing software and also most video games have employed for several years—but hardware has much higher up-front costs which stop this from being viable.
As it is, camera bodies already make almost no profit. The general guide is that there is £50 profit for every £1000 that a body sells for. So a £1000 body represents about £50 profit, a £2000 body makes about £100 profit, and so on. Of course specific bodies may make a little bit more or a little bit less, and there are sales and rebates and everything else like that to consider, but fifty quid for every grand laid down by the customer is the
general guide.
So if you were to make a body which would normally sell for £3500, representing around £175 profit, and instead sell it for £1500, you're basically guaranteeing a nearly two-grand loss on every sale. And even if a few users were going to pay up more to unlock some features, most won't. Even if they all did, you'd still only be back to breaking even. Whatever small amount you saved in production would not outweigh the massive losses.
Of course another problem with hardware sold like this is that it's too easy for people to take their device offline and jailbreak it, getting access to every feature without paying. DRM can stop people doing the same with software—though even then it rarely helps—but there's no such option with hardware. Unless you were to insist that the camera has wi-fi on at all times and only works when connected to an internet connection. Even then, people would jailbreak it, mod it, crack it open and get it working offline and with everything unlocked for free.
The closest you could get which wouldn't be prone to piracy would be a modular body, like medium format systems use. Make one basic shell that you can then snap different viewfinders, rear screens, and sensors/processors into. That has worked out well for a few medium format systems, but they also have the benefit of being extremely niche products which can command a high price tag to begin with, and they have far larger profit margins than 35mm (and smaller) cameras.
But Canon don't really need to branch into anything like that anyway. Fact is, cameras don't
need to be a big deal. Camera bodies, other than medium and large format, have
always been loss leaders. Lenses and accessories are where profit is. It's why Canon started the EF mount with some terrible bodies, instead putting all their R&D into the new lenses; they've done the same with the new RF mount. Bodies just exist to get people into the ecosystem. You know that every body you sell will also sell a lens, probably a spare battery, maybe something like a case or grip. Then half of all people will buy a second lens, maybe a third. A quarter of people or so will buy multiple lenses, batteries, grips, cases, flashes, etc. There are comparatively very few people who will ever buy multiple bodies, or at least not buy new bodies often enough for the money to add up. By leaving bodies as loss leaders and focusing on lenses and accessories instead, you won't make a huge profit off of everyone, but you're making a
sustainable profit and the minority of people who will buy big kits of lenses will really pay off for you.