I think a lot of people design systems for imaginary users that don't actually exist. I try to make stuff that works for me, because then at least it works for one person.
(I assume "normal people" use some sort of GUI for this. If you are using Amarok, for example, it's a simple matter of drag-n-dropping your playlist to the icon that says "Android Phone".)
Usability testing should trump hunches. I polled my room mates (small poll, but still 4 > 1), none of us have a music library small enough to fit on an iPhone and none of us want all of our music/videos/photos on our mobile device for a variety of reasons. Drag and drop in iTunes is preferable for us.
And I have an iPhone, I've had it wiped out by iTunes. I'd rather have drag and drop like a normal file system from Explorer.
"Normal people", by which I mean myself, want to plug in their iPhone and have all of their stuff synced without ever clicking any button. Drag-n-drop is nice, but even that's more of a hassle than I care to bother with.
If syncing is on by default, having to turn it on doesn't sound like "ever clicking any button".
On the Android part I sync my music with Banshee ;)
This article isn't about Apple's doing everything perfectly, it's saying that there are a lot of things they do incredibly well that Android sucks at, and it's saying that it wishes Android would make itself less sucky at those things. That's what this conversation's focusing on, not "Does Android do some things better than the iPhone?". We know the answer to that is yes and it's foolish to argue it.
People write for the iPhone because of the imagined mass market (which brought the cost of every useful app down to 99 cents), and people write for Android because they are programmers and it's cooler. The end result is lots of apps in both cases.
Remember, you are not a Google or Apple shareholder, so all you should care about availability of good apps. Both phones both have the same good apps, with a slight edge towards Android (because of more capabilities).
What do you plan to run your software geek nirvana on, anyway? If noone buys the platform, what happens 3 years from now?
As a meta-question: What do you think would've happened to linux and it's desktop if solaris' or even windows server licensing terms were more reasonable at the outset? If Mr Shuttleworth decided his money was better spent feeding children on his home continent rather than making Debian look nicer? Think about it...
But that's not even the point, really. From the perspective of the cell phone manufacturers, eliminating or reducing the expense of developing an OS is good for their bottom line. As long as Android is "good enough" for most users, I foresee it continuing to be used.
The argument over applications is one-sided. Apple's doing a shit job. But people still write more applications for it, because Apple offers a better product than Google does. Not a better development environment. Not a better marketplace (though that one's debatable). A better product.
If I had to determine, based on previous personal observations, which was more likely to exist, iTunes users or perl script users, I would have to conclude that you are imaginary.
My wife syncs music her G1 using Windows Media Player, because she's used it for years and likes it.
People who use Linux plug in their phone and Banshee syncs to it.
I don't think iPhone does WMP or Banshee. Sure, it's made by Apple, but my wife doesn't really care: she just want's it to work with her computer, without having to change all her software.
For some reason I don't think this is exactly what Gruber had in mind when he said, "The goal should be to make a phone that is better than the iPhone. Better."