Apple is deep into its i-everything world, having built a deep and wide ecosystem where low end products lead to mid-range products lead to high-end products; where one product line enhances a parallel one; where software ties the product lines together; where accessories are highly commonalized; where third parties are encouraged to support the products via accessories and apps.
It appears that Microsoft is finally headed this way, too, with the "Windows Everywhere" mantra finally getting some backbone to support it. Which leaves Google to be the Silicon Valley player that doesn't seem to care much as their "standard" gets torn in far too many splinter directions. Android isn't an ecosystem, it's a mutatation-prone monster that's consuming huge amounts of resources but not really building something bigger than its pieces that individual players benefit from.
I understand ecosystems. I helped successfully build three of them in my Silicon Valley career. By definition, an ecosystem is "the sum of the parts is greater than the sum of the parts." When players other than the ecosystem creator make things that help people want the main item the ecosystem is designed around, you get an amplification effect that is hard to beat.
Put simply, the Japanese camera companies mostly have no amplification effect because they're not creating ecosystems. Which is why many of them will eventually die (at least their camera groups will). They all try to do everything themselves, and they can't even get that right. The one exception at the moment is m4/3: since this started as a partnership and grew into a consortium, it has multiple players all riffing off one another. I suspect collusion, as well, as the fact that most of the lenses coming out of the m4/3 group are not directly competitive with one another; each player seems to be picking different fruit to grow, and that usually doesn't happen without cooperative decision making.
Still, m4/3 is doing well partly because it's an ecosystem, and the sum of the parts are greater than the sum of the parts. People look at the lens choice and decide they'll buy an m4/3 body because of that. Say what? They didn't buy a camera because it's Brand X and then choose lenses? Does that happen? Yes, it does, but not in the minds of the executives running most of the Japanese camera companies.
Nikon is a highly go-it-alone company. They don't license the lens mounts, they don't share much in the way of APIs, they are highly prone to making some types of connector changes (yet don't make the one change they probably should ;~). They have made lots of small but potentially fatal-for-third-party-product-changes to electrical specification on the mount and hot shoe. They like inventing new technologies that they don't share with others.
Even within Nikon's own line we have discontinuity of ecosystem. Was the Nikon 1's flash system truly CLS compatible? Not really, though they're now claiming more compatibility. This is the same problem Nikon had with Coolpix CLS. It sort of was the same, but not quite, and there were not only a lot of restrictions, but many anomalies, too. Personally, I gave up trying to use CLS on the P7000 and just went to manual flash. At that point, I could pick any flash on the planet, so I went out of ecosystem. Oops.
One of the key tenets to what I said would eventually disrupt the current camera market was programmability. Programmability requires an API and a solid effort to extend to the ecosystem to third parties. That's especially true since most of the Japanese software to date fails to meet current programming standards. The Japanese do low-level real-time firmware well (though a lot of it is still centered around variants of DOS or Linux). What they don't do well is object-oriented, modern UI specific, extendable, feature-rich applications. One of my programming friends had this to say about Nikon's iOS and Android apps for the WU-1a: "Looks like an intern programmed it. An intern I wouldn't be hiring."
In theory, you can build an ecosystem that's closed (i.e., you make all the pieces yourself). Sony seems to be at least partially headed down that path with their PlayApplications concept: Sony will write applications you can buy, but only Sony. The products get extended via these applications (modestly at the moment), but only in ways that Sony sees, and only with Sony's programmers. See my programmer friend's comment about Nikon apps ;~).
It's far better to build the core of an ecosystem, extend it with key products, then open it to third parties willing to sign licensing agreements in return for access to key, necessary information. You also need developer support and an evangelical aspect to attract developers and bring critical information back into the company. To date, I've not seen any of the Japanese companies manage these things. Even the so-called SDKs some produce are not exactly prostelitiized or well supported.
So what does Nikon need to do to build a real ecosystem? Let's see:
- Open the mounts. Set up a licensing program that shares mount information and provides a full set of options to third parties. Yes, that implies that things like the chromatic aberration and distortion lens tables in the cameras might get updated with third party information. This old notion of "people will buy our lenses because third party lenses aren't corrected by the camera" will just eventually cause people to buy cameras in systems where every lens is corrected.
- Standardize CLS and standardize on CLS. Stop the constant tweaking of signal definitions, license the API, and use it consistently across all cameras that support external flash.
- License the connection. Nikon partly does this. You can license an SDK that gives you binaries you can include with your product that access many of the main protocols to the DSLRs. Last time I looked, those binaries didn't multi-threaded, had onerous restrictions on them, and weren't exactly performance oriented. Just license the API itself and let real programmers figure out how to do it right.
- Standardize power. Let's see, in EN batteries we're up to 21 (though we did skip a few numbers). Worse still, Nikon used Japanese regulations to force the change from EN-EL4 batteries in the pro bodies to a new battery set. No regulation in 85% of the world forced them to do that. In other words, for the convenience of selling pro equipment into the Japanese market that used the same battery as the rest of the world, 85% of the pro users had to suffer. This is ecosystem friction at its worst: random changes that make ecosystem investments by the user worthless.
- Market the ecosystem. Let's see, Photoshop has become an ecosystem. Is there a PhotoshopWorld convention? Yes, twice a year. Plus Adobe's own events. Plus third-parties marketing Adobe's products both directly and indirectly. Not only does someone have to write a better software program than Photoshop to knock Adobe off the mountain, they'd have to do the third-party management and ecosystem marketing better. Tall order (and here's a hint to any of you Japanese software folk reading this: you haven't proven you can do either, yet you need to do both to break Adobe's grasp on imaging software).
- Listen to the users. You don't have to (and shouldn't) blindly follow any user request, but you have to listen closely to the user base in the ecosystem and figure out where their frictions are and remove those. Then check the lubricant to the things that are working and refresh it if necessary. Next analyze the user base and figure out what they're missing that they don't realize they're missing and add it. Then repeat that whole process over and over. Nikon currently doesn't listen well, doesn't remove frictions, removes lubricants, and randomly builds new product lines without being able to say how they fit with existing ones.
- Build a real support system. Ecosystems don't work without a two-way communication channel between internal and external development groups, and that takes a real developer support system that doesn't play professional-level voice mail disengagement. Sure, not telling Sigma what you changed in the signals will slow their sales of lenses. But it doesn't improve your sales of bodies (or even lenses); it just makes customers upset.