Re: Maple Vs Mathematica debugging / cost to write a system

> Some time (long) ago, when I was active as a Software Engineer(?),
> software metrics told me that independent of language a
> programmer could handle 40-50 lines a day. So we end up with
> approx. 2000 lines of code a year.

Programmer productivity can span many orders or magnitude.
My own metrics in terms of *usable* codelines/year (which
I used before moving to academia in the mid 80s):

Class A - superprogrammers 10^4-10^5
Class B - competent 10^3-10^4
Class C - average 10^2-10^3
Class F - hopeless 0

This is for code of standard-to-moderate difficulty. For truly
challenging tasks, say a real-time fault-tolerant multiplexer
in assembly, divide by 10. For easy stuff, such as
numerical software, multiply by 10.

Class F programmers might be difficult to predict. In 1979
I was managing a software project at Lockheed that began to
lag milestones. I hired two seemingly Class-C guys to speed it up
(I had not read Brooks' The Mythical Man-Month yet). In a year
span they wrote thousands of lines that were totally useless.
That code had to be rewritten from scratch, further delaying
the project. That why the qualifier *usable* is important.

.