As someone who has been watching the mobile industry evolve now for over 10 years it is interesting to watch how the iPhone phenomenon is causing so many industry stalwarts to run in circles chasing Apple's success. Whether the focus is touch screens or shiny black finishes, competitors are trying desperately to copy Apple's moves to achieve their own success.
The latest iPhone-spawned craze seems to be the need for every OEM and mobile operator to launch an application store for their devices.
Of course this is a great thing for the big device brands and for consumers, so the question is why didn't they do this first?
The answer lies with one big advantage Apple has over all other device OEMs (and certainly over any mobile operator) – a simple device support matrix.
Apple has the unique advantage of only having to support a small number of device iterations. As a result of having been in the market longer, the traditional mobile OEMs and carriers have the difficult burden of having to support a wide variety of device types and platform characteristics.
The goal for all App Stores is (or, rather, should be) to provide a one-for-all application download experience to support the OEM / provider's brand. As the number of supported device iterations grows, delivering on that goal becomes increasingly complex.
Apple has, at most, four devices that they need to worry about -- two generations of the iPhone and two generations of the iPod Touch.
Apple's support matrix looks something like this:

Looking at the above table, all four of these devices have the same screen resolution, the same input method (touch screen), the same form factor and button configuration (one button + volume controls) and nearly identical hardware comprising them. So, in practical terms, Apple really has only two, maybe three, devices to support.
Now let's compare that support burden with the next most simple device platform: RIM. If we assume that RIM will focus primarily on their top four device families (the Storm, Bold, Curve and Pearl), they still have an enormous support burden compared to Apple.
Among these four device families, Blackberry devices come in three different resolutions, three different input methods (touch screen, 1- and 2-letter QWERTY keyboards with trackball), two form factors (candy bar and flip), and some variance on hardware (GSM, CDMA, with / without WiFi or external displays).
So Blackberry's device support matrix looks like this:

These two images illustrate the challenge other OEMs have to overcome to create an experience as elegant as the iPhone's. And RIM probably has it the easiest of the rest of the OEMs (excluding Palm, see below). Nokia and Windows Mobile both come in more flavors than ice cream and Android is set to be modified and iterated by any number of OEMs / ODMs so their app store will quickly become fragmented as well.
Palm is perhaps the one exception to this bunch. Palm has made a conscious decision to keep things simple by starting fresh with the Palm Pre. They will have one App Catalog for one device. Of course more WebOS devices will be introduced but I expect Palm to follow Apple's lead by keeping things simple by keeping device specs similar.
The Apple advantage is not unique to the iPhone platform either. If you look at their lineup of laptops and iMacs, it's also incredibly simple. One of the writers at EdibleApple.com has a great post (http://www.edibleapple.com/why-a-simple-product-line-is-integral-to-apples-success/) about how Apple's simplified product portfolio has been a big benefit to consumers. I would add that Apple's approach has also been a big benefit to developers and to Apple themselves.