One of the oft-claimed advantages of Interpreted Languages is that you don't have the overhead of the "build" cycle.

As I'm currently working with an Interpreted Language (Python - not on an 8051), I'm coming to think that this is not (necessarily) all it's cracked-up to be:

A "conventional" build (compiler, linker, etc) has to check all the code for syntax & semantics before it will give you anything to run;

An interpreter, on the other hand, may not actually notice such errors until it actually tries to "run" the offending line - which could be some considerable time into the program