A lot of trashing on Excel but I kind of think this particular instance isn't really the essence of the problems with Excel. Perfectly reasonable programmers ge

A lot of trashing on Excel but I kind of think this particular instance isn't really the essence of the problems with Excel. Perfectly reasonable programmers get this wrong all the time (ever heard of buffer overflow anyone?), and the issue of small test data being unrepresentative of large test data is something everybody has to contend with. If there is a real problem with it I would say it is the presumably silent nature of the truncation which allowed the error to propagate. The real horrors of Excel come from things like auto-conversion of column data (text, numbers to dates etc), off-by-one errors in copy and paste, overwriting forumlas in cells, sorting of columns that doesn't capture all the rows, etc. These are all problems that are engineered into the user interface of Excel. It's like putting a tripwire at the top of your stairs and just expecting people to step over it day in and day out. It's basically inevitable somebody will fall down the stairs.