Dangerous

You’ve probably heard either a lot or very little about the Sidekick failure. At the beginning of October, on a Friday, T-Mobile Sidekicks lost contact with the servers that support their data functions, and those services were completely unavailable until Monday morning. During the following week, some improvements were made, but the service was not back to normal. A full week after the initial contact loss, Microsoft acknowledged that personal data had been lost, and the party line was that a hardware switch-out was to blame.

A failed hardware switch-out was always a lunatic story. When hardware fails, you replace the hardware and install any relevant backups. The Sidekick service had run for years without any issues even approaching this, so what happened here? And for that matter, if you but dimly remember hearing about the Sidekick back when it was the It Device, what does Microsoft have to do with it?

Microsoft bought Danger, the company that developed the Sidekick (Danger calls it the Hiptop) in early 2008. I’ve been using the Sidekick since 2004, and my gut response to the acquisition was simple: I hoped the iPhone would become attractive to me before the Sidekick disappeared. I assumed that Microsoft had bought the company for competitive reasons – to take a non-Windows-based smartphone out of the market. It seemed it was only a matter of time before Microsoft withdrew the device or replaced the guts with Windows-based software. When T-Mobile marketed an updated device with the non-Windows software intact, this spring, I thought about buying one but was reluctant to extend my contract (which expires soon). I’m glad I waited.

When this outage occurred, I decided more or less on the spot that my next phone would be an iPhone. I use an iPod Touch, and I know first-hand that I don’t much care for the software keyboard, but 3 days is geological time on the Internet, and a company that is willing to (in effect) brick my phone for that long is not one I want to rely on for communications when I’m traveling – or even just want to grab some quick reading on transit. As I followed the emerging stories about the outage, I was frequently and freshly amazed by something I’ve often thought about Microsoft but rarely seen so beautifully demonstrated: This is a company that can take lots of smart people, give them lots of resources, and still get surprisingly less than the sum of the parts.

The first Monday morning, as the Sidekick data services were lurching toward reanimation, MobileCrunch discussed what this outage was revealing about Microsoft’s smartphone ambitions. No surprises here. Given the history of Danger, it was hard to imagine its strongest employees happily moving into Microsoft. I hadn’t been following the Pink news, but it doesn’t sound like there was much to follow. It confirmed my expectation that the Sidekick’s days were numbered, but other than that it seemed like a mix of fud, schadenfreude, and possibly a game of telephone. Interesting but remote.

AppleInsider expanded on the story a few days later (all on one page at RoughlyDrafted). Here was a more detailed roundup of the history of the acquisition of Danger and the personalities involved. Microsoft paid a half-billion dollars for Danger, presumably to own patents it could defend from infringement. Dilger is brutal about Microsoft strategy and execution, and he doesn’t disappoint here. Particularly sad were the emerging stories of Microsoft calling around and finally managing to persuade some ex-Danger people to come in in the middle of the night and put things right. They seemed to accomplish more in 6 hours than Microsoft had in the previous week.

Dilger came back last week to report more revelations on the runup to the outage, explicitly pointing to Microsoft’s likely attempt to replace competitor-based systems with Microsoft products or at least rush to update firmware somewhere without having a good grasp of the system they’d bought and, Dilger says, more or less run on autopilot. And he fleshed out the spectre that, regrettably, made a crazy kind of sense given the astonishing extent of loss and incapacity: insider sabatoage.

There is a lot of talk about whether there will be any serious repercussions for Microsoft managers and for the Microsoft mobile platform. There have also been generous helpings of hand-wringing about cloud computing. I must not be susceptible to cloud hype, because if anything I’ve become more assiduous in my backup and redundancy planning than I ever was before I started using the Sidekick. I’m glad of that – it meant I lost absolutely no data during this absurd sequence of events.

But I feel awful for users that did lose data. Sidekick’s user base is very different from those of other smartphones – it’s a generally slower-adopting crowd, with a higher proportion of women, kids, blacks and hispanics, and, because of its early special features for this group, deaf people. Microsoft stranded many people who likely had no good alternatives to much of the Sidekick’s functionality. Those groups are relatively small or not very visible, and it feels like another example of how careless the company is.

If Microsoft could identify an angry individual who damaged servers, data, or recoverability, this story would be deflected from one of an impressive culture of failure to a simple scapegoating. This outcome would be unfortunate to anyone suspicious of Microsoft but also to people who don’t like to see a good platform trashed by a careless acquirer; that story disappears if an individual can be charged with a crime. The “good” news is that as more information comes out, it tends to support the thesis that Microsoft, as much as it would like to blame contractors or Danger’s original vendors, brought this upon itself. The business-as-usual news is that Microsoft is experienced in damage control.

Microsoft clearly never expected to do anything substantial with the Danger acquisition, so its only risk at the moment is its similar internal project and perhaps its cloud computing ambitions. If Pink goes the way of the Zune, well, Microsoft still has billions in the bank, and it can cloud the issue of its competence with its new OS launch (have you planned your launch party?). And go on to easily afford bigger and better failures in the future.

The real (corporate) loser at the moment is T-Mobile. I never thought I would feel sorry for a mobile service provider, and I’m not a fan of big lawsuits, but T-Mobile appears to have been doubly screwed this month, first by Microsoft’s inability to secure and support Danger servers and then in weathering customer rage throughout a crisis during which the T-Mobile part of the service was the only thing that was actually working. Indeed, for the first week, T-Mobile was alone in openly acknowledging that there was a problem. I hope T-Mobile uses every relevant syllable in its contract to sue the living pants off Microsoft over this. It won’t make any difference in the long run, but T-Mobile is owed some damages.

Whatever happens, I’m looking forward to following this story. On my iPhone.

Leave a Reply

Your email address will not be published.