This post was published 3 months 28 days ago which may make its actuality or expire date not be valid anymore. This site is not responsible for any misunderstanding.

The firestorm that resulted from the recent loss of Sidekick data (and thus effectively the end of the grandfather of all smartphones) is finally settling down. In fact, after all the words written and internet screams uttered, T-Mobile and Microsoft are indicating now that possibly not ALL the data was lost. If this goes the way such things normally do, after the initial panic (@IT_SUPPORT “OMFG I SPILLED MT DEW IN THE SERVER??!!!HELP ME DUDES!!!!) has passed, cooler heads will prevail and data will slowly be recovered, forgotten backups will be discovered, until after a few weeks or months most if not all of the damage will be undone.

Of course, the changes to public perceptions and attitudes about the technology and players involved in this train wreck will last far longer than any inconvenience from lost data. Pundits for example are using this incident as support for their positions on all sides of the Cloud computing debate, both for and against, while others are using it as yet another stick to beat Microsoft and their mobile strategy (Windows Mobile and their Cloud solution, Microsoft Azure)…but how much of any of that is justified. How useful is the death of the Sidekick for understanding the current mobile “big picture” regarding Cloud computing and Microsoft?

First of all one must take into account that, to a great degree, the Sidekick is a living fossil of telephony. It’s first models represent an early vision of what has evolved into the modern smartphone. Originally created as the Hiptop back in 2002, the device was released more or less contemporary with the first smartphone versions of the Blackberry. Both we key elements that debuted the concept of telephony with a text/email focus to the market. However, while the Blackberry was clearly all business, the Hiptop/Sidekick was a purely consumer oriented blend of a text pager and a phone, intended to simplify the early versions of social networking such as SMS and text messaging. It appealed heavily to the youth market and still does, but it’s limited niche is illustrated by the fact there are around one million Sidekick users currently (or at least there were last month) compared with nearly 30 million Blackberry users.

The thing that made the Sidekick unique both then and now, and was a big part of it’s downfall, was that it was tightly tethered to it’s servers. Far more so them other mobile devices. All data such as contacts  were stored on Danger’s (later Microsoft’s) servers and simply accessed by the Sidekick on an instance basis, such data was never stored on the device but rather was just viewed on the Sidekick. Without the server, the Sidekick is nothing…and when that server is compromised so is all user data, which is exactly what happened last week. The fact that there seems to have been no backup that could be quickly and easily brought on line was an astounding mistake by Microsoft, Danger and T-Mobile.

The Sidekick architecture, a million mobile devices depending on a single server is a very, very simplified version of the Cloud but can be used to illustrate some hazards (or nightmares) of Cloud computing. However the picture it paints and the lessons it teaches are done with very broad strokes. It can be safely assumed that any Enterprise ready Cloud computing strategy will include various forms of redundancy as well as regular backups on the server and the client side. These two common elements of network management would have made the Sidekick failure no bigger deal than recent Gmail outages. Humiliating and enraging to be sure but hardly fatal.

If this teaches us anything about using the Cloud it is that basic network management and client side backups will become even more essential and should be made requirements, much the same way many corporate networks force you to change your password every few months. If the network can’t detect a backup of your data within a set period of time, it shouldn’t let you log on until you perform one. The same check could make sure your antivirus signatures are up to date, further protecting the Cloud network from viruses and malware.

So while the Sidekick incident may worry a lot of people and shine the harsh light of media attention on the dangers of offsite computing in general and the Cloud specifically, in the long term I think it may be a good thing for the movement. Hopefully it will cause IT pros and users alike to be more careful and scrupulous about “best practice” policies concerning backups and redundancy in Cloud computing. After all, no one will want to be “the next Sidekick”.

Yet Cloud computing is still “up in the air” (groan), will the Sidekick incident have any effect on more down to earth players in this drama such as Microsoft, beyond the initial embarrassment and becoming the punch line of still more fanboi jokes? Well, in a bit of overlap, a lot of people have been voicing worry that this was some kind of failure connected to Microsoft’s upcoming Cloud initiative, Azure. Apparently they are totally unconnected however, as Azure has barely begun to be rolled out yet. The Sidekick server was not using and was never intended to use Azure. In fact, from what I have heard the Sidekick server was still using the same Oracle software that was installed when Microsoft bought Danger. They may well have even been the same physical server(s) from before the purchase, totally unchanged. Certainly Azure is still largely unproven and untested, but it cant be blamed for Sidekick.

Just as it had nothing to do with Azure, that also means the outage had nothing to do with those elements of Windows Mobile that sync back to servers, such as the new MyPhone applications. To say that the Sidekick failure raises new fears about the security of Windows Mobile is like saying Miley Cyrus ending her tweeting career means there are data center problems at Disneyworld. Totally unconnected as far as we can tell (but who ever knows about Miley, really). Beyond that, there is very little even a Sidekick-style server crash would do to Windows Mobile. All data is stored on the device, on the PC it syncs with, and depending on the settings, likely also resides on the Enterprise Server it connects with. Add to that the fact the added insurance that there are many excellent and effective third party backup solutions for WinMo. As is expected of a business class device, Windows Mobile data is quite bulletproof, unlike Sidekick’s. Now, Redmond’s top secret Pink phones, apparently based on Danger designs, may not be quite so bulletproof after this mess, but then I am feeling more and more like this may be a good thing for Microsoft.

While Azure and Windows Mobile are free of the Sidekick taint, Microsoft management is not out of the woods. This situation brings up major concerns about IT practices at Microsoft, as it still amazes me that there was no backup for the Sidekick info readily available. It also reflects badly on the corporate culture in Redmond. They paid a lot of money for Danger, and thus far have totally squandered it’s assets and expertise to the point of not even bringing their servers into the Microsoft network properly. There is simply no excuse for that. If people like J Allard had been working effectively with people hired from Danger, who have a lot of valuable experience creating smartphones that people actually like, we might not still be waiting anxiously for Windows Mobile 7. We certainly would not be considering WinMo 6.5 just a somewhat ho-hum maintenance release if Danger had been working on it for the last 18 months.

So in short, I don’t feel that the end of Sidekick will be a lasting hindrance to either Cloud computing or Microsoft’s mobile future (except maybe to Pink)…but it is very embarrassing for all involved, makes clear how important proper back ups still are, and perfectly illustrates that when Microsoft wastes a valuable asset like Danger, it really WASTES it.

No question, Sidekick and Danger were wasted…but that is where the fallout should end.