Apple's iPad Gateway Drug: The Mac App Store and Beyond

Due within 90 days, the Mac App Store will re-define how software is distributed on personal computers and begin the transition towards a fully managed, locked down next-generation Apple end-user experience that is more iOS than Mac.
Written by Jason Perlow, Senior Contributing Writer on

Apple's new Mac App Store, due within 90 days, will re-define how software is distributed on personal computers and begin the transition towards a fully managed, locked down next-generation Apple end-user experience that is more iOS than Mac.

I told you about it seven months ago, but it seemed far-fetched at the time -- Apple is indeed moving towards a model for its Macintosh personal computers that is closer to the iPad than what Mac users currently experience today.

Today, at Apple's Back to the Mac event, Apple announced the availability within 90 days of a Mac App Store, which brings nearly identical software distribution technology to the Mac currently enjoyed by iOS users on the iPhone, iPod Touch and the iPad devices.


The new Mac App Store, slated for release by Q1 2011.

The Mac App Store will be a supplementary software package at first, but will ship by default with the next version of Mac OS X, 10.7 "Lion" in the summer of 2011.

This is a significant event in the evolution of the Macintosh, because it signals the paradigm shift away from ISV's and developers being able to control their own software distribution. In effect, users of the App Store will no longer need to install and download or retain media for their personal computers, just like it is handled on iOS devices today.

With the launch of the new Macbook Air, which has no slot at all for an optical drive, Apple has made it clear that they want their next generation systems to be fully dependent on managed software distribution.

Forget having archival CD or DVD copies of your software or being able to back-rev to an earlier version if you have any sort of problem -- once Apple has moved to the new model as the preferred mechanism for software distribution, end-users of Macintosh systems will have to buy all of their software from Apple, and Apple will control all of the software updates -- just as iOS users do today.

How Apple will gate-keep the Macintosh App Store in comparison to the iOS App Store that we are all familiar with today is unknown. As with the iOS App Store, it is likely that developers will need to conform to some form of terms of service and code review in order to be accepted for sale.

Whether Apple will eventually restrict the use of certain libraries and APIs on the Mac App Store such as Flash is also unknown, but it is probably inevitable that the Mac App Store will inherit at least some if not most of the developer guidelines and principles from its iOS sibling. We can at the very least expect the same restrictions on adult content and strong language that the iOS App Store has today.

[UPDATE: Apple has posted its acception/rejection criterion for developers for the Mac App Store]

So what's next after Mac App Store?

The logical conclusion is to actually marry iOS and Mac OS X, so that there is no fundamental difference from an end-user perspective between the two platforms. Steve Jobs himself pointed out today that the Mac is coming full circle, and Mac OS X is going to inherit a number of features from iOS, which includes multi-touch capability and full screen apps.

Apple could go about this in a number of ways. The first, although not the easiest way of doing this would be to create a binary execution environment within Mac OS X Lion that allowed for iPad applications to run seamlessly with Mac Apps. This would allow for a single, seamless App Store on the Mac and provide access to the 300,000 programs sold currently for the iOS environment, of which currently nearly 36,000 are optimized for the higher-resolution screen format of the iPad.

A iOS binary execution layer would not be a new approach to software compatibility for Apple. During the PowerPC to x86 transition in 2005, Apple provided the "Rosetta" binary execution layer so that PowerPC apps could run on Intel Macs.

Rosetta was written by Transitive, a company which was purchased by IBM in late 2008 and is now used in the company's PowerVM virtualization softwarefor its AIX servers to provide Sun SPARC and legacy Solaris compatibility. Transitive was also used by Hewlett-Packard to provide Intel x86 compatibility with the Itanium server CPU.

Also Read: Was Intel's x86 the Gateway Drug for Apple's ARM?

Theoretically, Apple could contract IBM or another software company familiar with the ARM instruction set used by the Apple A4 to write a "New" Rosetta binary execution layer that would do the same thing for iOS apps.

Today, when developers program iOS apps on Mac systems, they can "simulate" an iPad or an iPhone, but it doesn't actually execute ARM code natively. It actually has to be cross-compiled in the XCode environment to run on an ARM-based iOS device before it can be run natively.

Another alternative or longer term objective than using a binary execution environment like Rosetta would be to actually embed the current A4 used in the iPad as a co-processor in next-generation Mac hardware. With the use of Non-Uniform Memory Accesss and virtualization technology, the iOS environment could be "containerized" and pull memory and disk resources from Mac OS X as required, instead of having a iOS device with fixed memory.

By using the A4 as a co-processor or using a binary execution software layer in the next generation of Macs to run iOS apps, it would give Apple the time it needed to develop completely ARM-native Macs and return the Mac to its fully-proprietary, uncloneable roots.

A fully ARM-native Mac OS X/iOS hybrid system would require a much more scalable SMP, multi-core architecture than what is currently employed today. While iOS has hundreds of thousands of apps, none of them are nearly as demanding as anything currently used by the creative content professionals that depend on Macs today, such as the Adobe Creative Suite and Aperture, or even iLife, which is more of a hobbyist/home user content creation environment.

To run these types of demanding applications, the ARM-based Mac of the future will need 4, 8 or even 16 or more processor cores, the ability to access larger amounts of main memory, execute 64-bit instructions, and Mac OS X itself and the applications will need to be optimized for a massively multi-core, RISC-based architecture.

While Apple's A4 doesn't possess these capabilities today, ARM's Cortex-A15 architecture currently provides for much of what is missing to that equation.

As Apple is an ARM licensee, it is not too difficult to imagine a Mac system 2-4 years from now based on ARM's most current architecture, with even more horsepower and multimedia acceleration that we see in Intel Mac Pro systems today, but using only a fraction of the power and taking up only a fraction of the space -- such as a Mac Mini chassis with fully solid-state storage.

Is Apple's Mac App Store the indication of a paradigm shift towards a completely proprietary, closed Macintosh systems architecture? Talk Back and Let Me Know.

[poll id="23"]

Editorial standards