Personally, I see problems with tower handoffs and signal strength response time (remember, that guage is near real time, not real time). In my neck of the woods, there are a LOT of different small tower operators and handoff is critical. There were problems until Apple released OS version 1.1.4 which seemed to improve things greatly. Now, unther the 2.0 update, a lot of things have reverted to their pre-1.1.4 state. Old issues have reappeared, etc. I don’t think that the 2.0 upgraded iPhone is using a new stack, but the 2.0 development track, which may have started even BEFORE the release of 1.1.4, may have failed to pick up on some of the updates introduced in 1.1.4.

The conversation that Chris mentions was around the fact that a LOT of different devices have been released that had initial radio problems, so we wondered who made the unit in the iPhone and who wrote the stack for it? Parts lists on the web indicate the manufacturer of nearly everything except the radio. Weird!

I personally suspect that at least some of the 2.0 instability is caused by poorly written 3rd party apps (or updated Apple apps) that don’t properly clean up memory when they close. We still have some of this problem on the WM platform due to the poorly designed memory management in that OS. Since Apple is currently not allowing background processing, they could sort of quick fix this by forcing a memory clearing routine of some type when you press the home button. Of course, that wouldn’t work well if/when they start allowing for background processing.

There are a lot of factors at play here, but I think Chris and I agree that Apple does need to resolve these kinds of problems fairly quickly if they hope to make inroads into the enterprise, where this stuff can be make or break for device acceptance. Consumers will be more forgiving, but IT groups considering purchases likely will not.

Apple’s ability to stamp out problems quickly will go a long way to indicating how successfully they’ll obtain IT acceptance/adoption since it’s all part of the support model. It’s likely more important than whether or not a particular feature exists. If they don’t respond quickly, and the problems linger, they’ll have a much tougher enterprise battle ahead.