Learning From Blackouts
Take away someone’s email, especially mobile email, and you’re talking riot. Which there almost was a result of the recent BlackBerry service outage, later traced to a software update gone awry. How many times have we heard that? Imagine -- software problems here in 2007. Software development and deployment remains a slow-rolling disaster, although sometimes, as was the case with BlackBerry , the rolling gets pretty fast. Regardless, RIM has some serious updating of systems and procedures to get busy on, let alone its software. We have the technology to design highly fault-tolerant solutions, and there’s no excuse for any service provider to fail in basic system operations.
To be fair to RIM, when there is a wireless link in the value chain, the laws of physics being what they are, there are always going to be challenges in designing reliable solutions. We might never get to five 9’s reliability no matter what we do. What’s really needed is for enterprise users of any information technology to have contingency plans, and most certainly not be dependent on any single supplier. This is a tall order, to be sure, especially when one is tied to a single carrier, as is almost always the case today. But if our suppliers can’t do better, then we must -- no one should be stuck in an elevator that’s going nowhere fast, matter of life and death or not.
— Craig Mathias is Principal Analyst at the Farpoint Group , an advisory firm specializing in wireless communications and mobile computing. Special to Unstrung