|
|
|
|
|
|
by marginalia_nu
1558 days ago
|
|
|
I don't think it's entirely technological limitation. More often than not, the discovery of bugs is increased understanding of what we require from the code. We often say a code is buggy when it surprises us in some fashion, then we tacitly invent a new requirement we say it has violated. If you have a function F(signed int32 n) -> (signed int32 A, signed int32 B)
that returns two integers (A, B) so that A*B = n
and we discover that F(2) -> (-2, 2147483647), which is entirely correct in a language that permits integer overflow; then we call it a bug because A and B must be smaller than n (or whatever). This was not a requirement until the bug was discovered. |
|
|