Wednesday, March 12, 2014

Malaysia Airlines and problem solving

It is very hard to imagine the anguish of the relatives of those on the missing Malaysia Airlines flight, but we should give some slack to those involved in trying to determine what happened to the flight.

I’ve been involved in several computer situations in which crucial information was not available at the outset.  It was only when the right question was asked or the right information was volunteered did a solution become apparent.  All of these situations were much simpler than the tasks facing the Malaysian and other investigators.

My first memorable lacking-information problem was in Sweden.  One customer’s mainframe kept crashing.  We looked at memory dumps, we asked questions, we had meetings, nothing became obvious.  It was only when I had a copy of a program that often seemed to be present when the computer crashed that I was able to take some more deductive steps.  The customer had not upgraded their software from a version that had a memory violation.  That program crashed at other sites with the same software.  Thus, the memory protection of the crashing computer was faulty.  The technicians fixed the memory protection problem, the subject program crashed, and the customer upgraded the software.

My latest memorable lacking-information problem was in Duluth last fall.  The keyboard on my MacBook Pro locked up.  None of the proposed solutions in the user community seemed to last for long.  I took my computer to the Geek Squad.  They kept it for a few days and found nothing wrong.  I brought it home and the keyboard locked up again.  After a bit of head scratching, I realized that the tech shut down all my applications before testing it.  Then I figured out that MicroSoft’s Outlook was the problem.  I reorganized Outlook’s database and the problem went away.

I hope those investigating the Malaysian Airlines crash find that crucial missing clue soon, but the ocean in that area is bigger than one laptop or one 1970s mainframe.