Lately the buzz around many of my peers has been around Apple's new OS and how they are leaving Windows behind in their dust. Yes, everyone knows I have been in the Windows camp for sometime, but this fact is irrelevant to my following argument - How to be the apparent innovation leader.
First, some explanation is in order. By apparent innovation leader, I am referring to the ability to be perceived at the leading edge of technology whether you are actually there or not. Apple has done a great job of this lately, but will it last? If history is an indicator, yes it will last, but at the expense of the current user base and developer community.
About one year ago, I completed an in depth study of the three leading OS offerings: Windows, Linux(redhat & suse), and OS X. My findings lead to the following blanket assertion about OS X: They have finally settled in to something for the long haul (explained below), but while their marketing campaign is quite warm and fuzzy they are still more than willing to stiff the user or the developer. Now people are outraged by this statement as they claim that OS X 10.X is the best thing since sliced bread for user experience. This may very well be so, but painting an old house doesn't change the fact the foundation is wearing thin. If you do a google search for OS X's roots, you will find it is build on top of a unix-like OS called Darwin which is strongly rooted in BSD 4.5. This leads me to my first major point of how Apple appears to stay on the leading edge: Buy a new operating system core and build on top of it the experience of the decade. Apple is the master of doing this. Once they detect their offering is getting moldy they are quick to pickup a better core with which to build a new experience on top of. Unfortunately, by doing this core swap someone must get stiffed.
As you probably expected, this OS core swap means incompatibility of the current software base. For a great example look at the transition from OS 9 to OS X. Basically, Apple said if you want to stick with us you need to completely re-outfit yourself. The surprising thing is that people did just that. More importantly, people outside of the Mac camp joined in once OS X was released due to its innovations. So please do not complain if Longhorn brings about some of the same as the market has signaled they are ok with such discontinuous innovation in favor of the next revolution. This compatibility is THE MOST overlooked aspect of Windows. If you don't believe me, search the blogosphere for “MS and shims in the OS”. You will be amazed at the ways the OS is tweaked for individual applications all in the name of backwards compatibility.
So to keep an image of innovation, you could provide advances through R&D or you could shorten your release cycle, keep your eye candy fresh and continuously harvest the low hanging fruit until such branches disappear, then move on to another type of fruit all together. To put this into an OS X content, read the first few paragraphs of this review of Apple's latest release. Yes, I am being somewhat harsh and failing to mention some of the innovations in OS X. This is mainly due to the fact that the kernel innovations I have read of so far (IOKit and Core Foundation) really aren't that exciting.
Well, to wrap this up quickly, I will not seriously look at OS X (as an innovator) until they provide some signal they will guarantee compatibility to me (as both a user and developer) for an extended amount of time. I agree at some point discontinuos innovation may be necessary to make a major step forward, but such things, expecially in the OS space, need to be infrequent and well justified. It looks like Apple is starting to finally figure this out by, for the first time, promising no disruptive API changes in their latest offering. But, for now, in my mind, Apple in not an innovation leader because they have historically been ignoring the (mandatory) hardest part of innovation, compatibility. I am not willing to upgrade at over 100$ US a pop every year to something that gives me inevitable upgrade hassles and no guarantees my current software suite and development efforts are still valid.