In 1974 when I was writing my first Fortran programs, we noticed that:

1. Arrays could have data in them from prior programs. In fact, it was interesting to sift through such to see what the mainframe was used for at other times.

2. Later we were chastised for our "bad programs" that would call for an array, but exceed the size of the array. i.e. an array of c[1], but go look at [2],3,4 and so on. One team member decided to not only read such a position, but change the value there. The mainframe crashed and we were admonished to write better programs.

I will not do the date calculation, but it appears to me that the issue is much older than 15 years.

Bob