I was (trying to) unplug most of the long holiday weekend, but couldn’t help but read all the headlines about how the Sidekick outage spells trouble for Microsoft’s cloud strategy.
I’m not downplaying in the least how serious this outage is or letting Microsoft’s Danger subsidiary — and/or any other companies involved in the loss of users’ data — off the hook. (The latest: The Danger team is now saying that “some” user data might be recoverable, after all.)
But Sidekicks aren’t running from/on “the Microsoft cloud.” In fact, there is no such thing as a single Microsoft cloud. Microsoft has lots of different remote servers in different data centers running lots of different services.
The Microsoft Azure cloud is what many Microsoft watchers think of these days when someone says “the Microsoft cloud.” But the Azure environment provides the underpinning for very few Microsoft services so far. The Sidekick services don’t run on Azure. Microsoft’s My Phone doesn’t run on Azure. Hotmail, Xbox Live, Microsoft-hosted Exchange — nope, nope and nope. None of these are running on Azure yet.
The Sidekick outage, to me, says more about Microsoft’s Pink than it does about Azure.
The Danger team, which Microsoft acquired in 2008, is largely responsible for the Pink “premium mobile experience” (PMX) software/services on which Microsoft has been working on secretly. There have been a few recent reports that Microsoft has decided against launching the Pink phone(s) that were going to run these services, but I haven’t heard from any of my sources whether this is true. Sharp supposedly is the manufacturer of the Microsoft-branded/co-branded Pink phones, which Microsoft is said to be planning to market primarily to teens and 20-somethings.
(Microsoft officials have denied repeatedly assertions that Microsoft is making its own smartphone. They have not denied that Microsoft is working with a hardware partner to build Microsoft-branded or co-branded phones.)
What was Microsoft doing to the back-end Danger services that resulted in such a catastrophic outage? Microsoft isn’t talking. There are rumors the problem stemmed from a storage-area-networking debacle but Microsoft isn’t confirming that, either.
The services at the core of Danger’s current offering — contact management, calendaring, instant messaging, e-mail — are all running on a back-end platform that Danger doesn’t describe publicly. Here’s the platform diagram it does provide (click on it to enlarge):
Because Microsoft hasn’t yet launched Pink, company officials have refused to talk at all about which premium services it will encompass and what kind of back-end platform they’ll run on. Is Microsoft designing the Pink services to run on its own servers? Is/was Microsoft intending to allow the Pink services to remain hosted on the existing Danger back-end? Did this past week’s Sidekick outage result from Microsoft (or Hitachi Data Services, or whoever) attempting to move the Danger back-end off the existing servers and onto Microsoft’s own servers? Microsoft officials won’t say.
Now that more everyday users know that Sidekick is connected to Danger and Danger to Microsoft, this week’s outage will cast a shadow over any kind of Pink phone and/or Pink premium services launch Microsoft may be planning.
Feel 
