A TechCrunch column published in the Washington Post just got under my skin. It takes off from the recent events involving Danger, a Microsoft subsidiary that provides mobile operators with end-to-end solutions to deliver mobile data and Internet services to their subscribers.

Danger was in the news, obviously, because in the course of managing data for T-Mobile Sidekick users, the company lost user data during a hardware upgrade. The TechCrunch writer slammed Danger for not doing a backup before performing the upgrade, something I totally agree with, if that was the case. Testing a backup is critical and should not be skipped.

At this point in the story, Microsoft has recovered most of the data that was believed irretrievably lost.

What I take issue with is that the author carries on at some length about how bad backup technology is and eventually concludes, "Data experiences its own form of natural selection. What is important will survive, the remainder will thankfully fade away."

The author suggests that if data is important, the user will find a way to back it up and not depend on IT to do so. I am sorry, but as a technologist I can't accept this defeatist way of thinking, from either side of this strange proposition. Could you imagine if a financial institution lost financial data and went back to the customers to retrieve the correct balances? Or better yet, what if that institution lost data on how much money was still owed on a commercial loan of millions of dollars and had no backup? And then the customer would hand over its records showing the lending institution every penny they owed on the loan. Right. Backing up and restoring data is a crucial part of what IT does.

And that is because the data is important to both the IT department AND the end users.

Losing data, any data, is a big deal and should not be taken lightly. Customers trust that companies they deal with will back up and secure their data. Neither administrators nor security professionals have the right to determine, especially after a data loss, say, what is important and what is not. Is contact information stored in a cell phone now designated non-critical? As a customer, I would be wondering, if you can't back up and restore something as "non-critical" as contact information, why should I trust you with something as important as my health care or financial information?

Finally, the article slams bloggers and consultants for wanting to bring poor back up practices to the attention of readers. We learn from what we do right as well as what we do wrong (or what others do wrong). The TechCrunch author feels that the point to be taken from this event is that some data is not worth saving. I am not sure who determines that, but we certainly should not allow so-called natural selection (backup failure) to determine it.

So most of the Sidekick data is being restored. Maybe all of it will be. Articles like this one by InformationWeek poke fun at Microsoft, which is uncalled for, and offer a few reasons why this might have happened. I think the bottom line here is that there was some kind of backup. If there was not, the data could not have been restored. I think Microsoft did a good job by stepping up and recovering what was thought to have been lost. I'm thinking the affected Sidekick users might agree. We can let the TechCruch writer rely on natural selection to determine if his data remains or not, but for the rest of us dinosaurs, I highly recommend a backup.