The Elephant in Apple’s iCloud

Here we go again. Another one-day software palooza. Did Apple learn its lesson from the last time? I don’t think so.

This is an editorial. Had it been an actual technical essay, you’d be confronted with a long list of URLs, step-by-step directions, and a myriad of caveats. Instead, I want to take a more philosophical approach.

Apple blew it.

Elephant in the cloud

You can argue that most people are doing just fine — if they just stumble through everything. But that’s not what we want from Apple. We want the same ease of use and sheer joy that we’ve enjoyed with Macs and iPhones. Instead, the October 12 prospects were overwhelming, and the overall guidance was typical Apple happy-go-lucky: everything will be fine. Don’t worry, be happy.

1. Apple’s servers were overwhelmed. Many got error messages, with numeric codes no less, suggesting that something had gone wrong with the iOS 5 install when it was really a symptom of Apple’s overloaded servers. Because the install is a two-way process on the Internet, it can go wrong when there are communication delays. And why is Apple presenting us with numeric error codes anyway?

2. Apple’s servers were overloaded because Apple released too much stuff at once. Here’s a list for October 12.

  1. OS X 10.7.2
  2. iTunes 10.5 (the day before)
  3. iCloud
  4. iOS 5
  5. Apple TV 4.4
  6. Airport Utility for iOS
  7. Aperture 3.2
  8. Newsstand for iOS
  9. iPhoto 9.2

I would have liked to see a phased release. Of course, there are some contingencies. The iCloud depends on Lion 10.7.2 and iOS 5. But what I’m driving at is that Apple’s desire for customers to have all this cool stuff working together (and be amazed) should be tempered by the reality that customers can only absorb so much at once. Designing the technology for a phased rollout not only lessens the load on servers but also implicitly suggests a sequential course of action for the customers.

System Preferences -> Software Update is no place to present a course of action. It’s a place where one goes after one decides on a course of action.

3. Documentation. I know that Apple likes to create a sense of delight and simplicity by keeping the documentation short, sweet and visual. The starting point is Apple’s iCloud setup page. The problem, as I see it, is that if one needs more details or has questions, one is referred to the classic Knowledge Base articles.

These articles can be rather clinical and daunting. Some of them can give you an immediate headache. When there is so much at stake with user data and a myriad of permutations and considerations, I will argue that special documentation was warranted for such a huge event. I think Apple took a shortcut. That is, individual engineers wrote their KB articles and another team wrote the iCloud set up page and simply linked to them. Job done. Easy way out.

Not so fast.

If Apple is going to go to the trouble of a giant palooza event with all this software, doesn’t it make sense to generate a more comprehensive set up page that showcases the best Apple can do with customer guidance? And cover all the bases more gracefully? Again, there is a lot of guidance in the iCloud set up process itself, but by then, you are more or less psychologically committed. Many customers want to read, learn about the advantages, understand the impact of their decisions, and make intelligent choices before they’re strong-armed into moving the iCloud set up process along.

Where was the Simplicity?

Again, a technology that’s been designed for a phased introduction allows the customer to reflect and test at every stage and more easily recover from unintended consequences. It’s the classic difference between the older waterfall and the newer spiral and agile development models. For example, no one would write a million lines of code, compile it all at once, and expect it to work flawlessly. Yet Apple dumps huge code projects on the consumer in one day and expects it to all work. (Or do they?)

I.T. managers all across the country were probably thinking: I would never try this with my own company. These Apple guys are amateurs.

A major problen is that the iCloud is a place where Apple’s historical excellence with user interfaces for an OS comes into direct conflict with the complexities of massive amounts and different types of user data in the cloud. Apple has tried to tie together, across multiple devices and two OSes:

  • Mail
  • Contacts
  • Calendars
  • Bookmarks
  • iTunes in the Cloud
  • Photo stream
  • Document syncing
  • Backup and restore

Other services that customers may have come to depend on have had to be dropped.

  • Mac Dashboard widgets
  • Keychains
  • Dock items
  • System Preferences
  • Third party syncing

In addition, the proliferation of Mac & iOS devices in one household means that there are different devices with different (or same) AppleIDs. How to approach al this for an even modest sized family and retain a sense of sanity and simplicity is hard. Long ago, Apple might have considered just what it is trying to achieve and what’s best left to third parties. Could corporate arrogance be causing Apple to believe that only they know how to do all this best?

In fact, Apple has a long way to go in becoming a master of the cloud. Apple might have approached this differently if it had a history of excellence in the enterprise. But Apple’s semi-dissing of the enterprise over the years, for the sake of relentlessly moving forward in its area of expertise, means they didn’t walk the walk of the cloud — where it was hard and mission critical for the enterprise. That difficult endeavor would have provided much needed expertise that Apple could have exploited to make it easy for the consumer.

And that’s the elephant in Apple’s iCloud.


Kudos to TMO’s Bryan Chaffin for his artwork - with help from iStockPhoto.