Apple's secrecy under Jobs dates back to the earliest days of the Macintosh project -- in a way, it may be what drove he and Woz apart. Read the stories on folklore.org and you'll see what I'm talking about it.
It's not a bad idea, either. Openness is perfect for a good idea (an evolutionary one, whose benefit to consumers is marginal to the competition): share the pain and wind up with a better field of products. But for a great idea (a revolutionary one, whose benefit to consumers leapfrogs the competition), the benefits of openness are outweighed by dilution of the rewards and the need to adhere to the consensus.
In short: if I invent an engine that runs on water, it's far more valuable for me to develop it myself then to announce the plan and develop it in conjunction with the big motor companies (who, due to my small size, would no doubt get the preferable deal). However, if I invent a new way to build a combustion engine, it's probably not the best idea to build them myself. I'll have better engines if I team up with a third party, and make more money as a result.
The great thing about Apple -- that the NYT conveniently omitted -- is that its secrecy is limited. They have no problem open sourcing the backend of the OS, which is evolutionary -- but they closely guard development of the revolutionary front end. Counter this to the IBM Way (do everything in secret, do it all yourself as a result), the Microsoft Way (do much in public, but do the core in private to prevent compatibility, leaning towards critical performance and security failings as a result) and the OSS Way (do everything in public, and reap little to no economic exclusivity as a result) and you see why Apple's secrecy is central to their innovation. You may not like it, but it's the reason for everything else you like.
