Let f(x) = 1/x, then (from a mathematician’s perspective) the limit of f(x) as x->0 doesn’t converge and is representeted as being infinite, definitely not zero

Let f(x) = 1/x, then (from a mathematician’s perspective) the limit of f(x) as x->0 doesn’t converge and is representeted as being infinite, definitely not zero. As the denominator gets smaller, the result is bigger and bigger and bigger. The closer the denominator gets to zero the larger the result. Why would we want to pick an answer that is small when instead it should be larger than any number? One might argue that 1/0 should be MAXINT or IEEE 754 plus infinity due to the limitations of our hardware, but I’ve never wanted 1/0 to result in zero in any program I’ve ever written. I would prefer that it be treated as an error or possibly some reserved value that behaves like INF while still being an integer in the programming language.