It's the (false) assumption that because someone controls what software goes onto the machine that it's impossible for 'unapproved code' to run on the machine (or that 'approved code' has been thoroughly vetted against all possible inadvertent or malicious security risks).
> The can do the right thing, and not act as gatekeepers.
I wonder what people will think once the first AppStore app gets approved by Apple but turns out to be a piece of malware. I'm sure Apple has indemnified themselves against liability in such a case, but there are plenty of people who feel it is an impossible scenario.
Apple has stated that they have a kill switch they can use in a situation like that. It would be extremely irresponsible not to on a device like the iPhone imho.
iPhone can have good security, and be a walled garden. But, the two are not closely related, and it's disingenuous, or at least misinformed about computer security history, to suggest that they are closely related. Open systems can have good security, and be very open. They are orthogonal issues, and I'm surprised that people here would make the mistake of believing they are the same thing.
You're conflating two separate issues, and when you come to the (obviously correct) conclusion that users are generally stupid and shouldn't be able to easily break their computer, that it has something to do with Apple deciding who gets to build applications and how one uses the machine. Apple does not provide better security in Mac OS X or on the iPhone than Open Source operating systems, in general. As far as I know, Android has at least as good a security history as iPhone for the time it's been in existence.
Sucks, but what else can they do?
They can do the right thing, and not act as gatekeepers. Or, they can do what they've always done, and maximize profits, minimize consumer and developer freedoms, and charge a lot for it.