iOS wireless backup

John Gruber posted an excellent article last week about the state of eliminating the need for a PC from the use of an iOS device, and is an excellent read. He builds upon Chad Olson’s three key iOS-iTunes dependancies with a fourth, so that we end up with:

  • getting your stuff onto your new iPad (Chad Olson)
  • updating iOS (Chad Olson)
  • backing up and restoring your iPad (Chad Olson)
  • device activation (John Gruber)
John then does a great analysis of activations and updates. However, I don’t think he is quite right when it comes to eliminating the PC from the backup and restore (and naturally the ‘getting your stuff’) portion of his analysis. In the end, it boils down to this paragraph:

That’s important, because I don’t think over-the-air backups or media syncing are coming soon. Wireless networking just isn’t fast enough. I’m not talking about Wi-Fi syncing over a local network to iTunes running on your Mac or PC — that may well be coming soon, but it wouldn’t solve the “how can these devices be ‘post-PC’ if they require a PC?” problem. We’re not talking about why the iPad needs a USB cable; we’re talking about why it needs a PC, period.

The problem that I have with this statement is that Wi-Fi backup (and even possibly syncing) does not need a lot of work for Apple to implement without a PC. Let’s look at a few technologies that Apple already has in place…

  • Portable Home Directories were introduced in April of 2005 with Mac OS X Server 10.4. It implemented the ability for data to be synchronized between a Mac OS X Server and a Mac client computer, assumed to be a laptop that would not always have access to the server it was bound to.
  • Time Machine was introduced in October of 2007 with Mac OS 10.5. Dubbed as a “breakthrough automatic backup that’s built right into Mac OS X”, it made it possible to provide a solution to home and very small business users other than copying data manually.
  • At the same time, the ability to use Time Machine with Mac OS X Server was introduced, creating Apple’s first network backup solution.
  • Time Capsule was introduced in 2008 alongside the MacBook Air, creating a simple way to use Time Machine to backup data over Wi-Fi.
Looking at this timeline, Apple’s been doing network syncing in Mac OS X for six years now, with broader and broader usage, as different products take root. With that type of experience, the natural progression is that the Time Machine technology progress into the iOS device for on-network backup. I think that there is then a natural evolution of the Time Capsule technology itself to sync to the cloud. This way, you get Wi-Fi speed backup, and your initial cloud backup can operate whether your iOS device is available or not.
Apple could institute the technology so that only the most recent image is saved on the cloud, eliminating the possibility that there is a TB of data for a single device on their servers. They could also implement a sort of “offline sync” that allows the iOS device to do an incremental backup to the cloud, although you’d better watch your bandwidth charges if you use this often and shoot a lot of video.
The other portion of John’s analysis is MobileMe, and he is spot on. It is a technology that is vital to the iOS experience, and as such should be built into the purchase of the system. Of course, the business of MobileMe may keep this from happening. Apple has a margin on a feature that Google is giving away. I know many people who find the value proposition of MobileMe to be worth subscribing, myself included. However, John then points out that there are quite a few other applications that could benefit from access to MobileMe-style syncing. That’s incredibly intriguing to me, and could even create a larger value proposition for other users, increasing the consumption rate of MobileMe subscriptions. Then, throwing the cloud sync that I mention above might just make it a done deal for most iOS devices, which in turn would cover costs for Apple, allowing them to do what they do best-Make money.