From: Ben Goertzel (ben@goertzel.org)
Date: Sat Jan 25 2003 - 14:26:52 MST
Well, Michael, you might be right.
However, even if you are right, I think my point still holds.
If the supercompiler allows people to write code with an emphasis on
maintainability and debuggability rather than anything else, THEN, this
should decrease the bugginess of code. This holds regardless of whether the
maintainable, debuggable code written is simpler or not...
-- Ben
> Ben,
>
> I am skeptical that code written with emphasis on 'maintainability and
> easy debuggability' would be 'MUCH SIMPLER' than most code written
> today. In what way would it be simpler? The same tasks would still
> have to be accomplished by the program, no matter the maintainability or
> debugability. Making a program more maintainable or more debuggable
> often *adds* lines to the program, *increasing* the complexity. I'm not
> saying this is a bad thing... but it sure isn't a simpler thing.
>
> Michael Roy Ames
>
> Ben Goertzel wrote:
> >> Ben,
> >>
> >> Supercompilation is very interesting, but I don't see how it would
> >> solve even one, single bug. The bugs would still be there, they
> >> would just execute faster.
> >
> > Because if you know you're going to use the supercompiler to
> > transform your code into efficient code, you can write MUCH SIMPLER
> > CODE with a focus on maintainability and easy debuggability, rather
> > than thinking about efficiency AT ALL.
> >
> > We don't code for debuggability now, but with a supercompiler we
> > could do so.
> >
> > I didn't say this before, but it's also true that the supercompiler
> > could be used as a powerful debugger...
> >
> > -- Ben
>
>
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT