From: Roger Racine <[log in to unmask]>
> I am obviously not expressing myself very well today. My use of the word
> "develop" was in reference only to design and code. I agree fully that the
> total -cost-, on average, will be lower for the Ada project. The argument
> is completely with the -risk of cost growth-. So, for example, if I were
> to estimate the cost of a project in C, I might come up with $1M, where, if
> I were to estimate the cost of the same project in Ada, I would likely come
> up with something less (let's say $750K). However, for the C case, my risk
> of cost increase is very low, because there are many ways to use metrics to
> estimate the number of errors that will be put in during coding.
1) How exactly does the use of C vs. Ada facilitate the estimation
of errors that will be introduced during coding? What are the
metrics that are available in C but not in Ada?
2) In the scenario, the risk of cost overrun was tied not to errors
introduced during coding, but to major design changes
necessitated by differing interpretations of requirements. (You
state this again below). How does this relate to errors introduced
> With the
> Ada estimate, the risk of cost increase is higher, due to the extra cost
> associated with the design work associated with a well-designed Ada
How does a well-designed Ada program take more design work than a
well-designed C program?
> This cost is related to higher-level design problems.
(There it is -- design problems, not coding errors?)
> For example, to get to a detailed design review, let's say 65% of the work
> is done in the Ada case, but only 20% in the C case.
OK -- why would they wait until 65% of the work was done before
conducting a review? That is purely a product management decision. How
is the choice of language supposed to constrain this decision?
Best wishes -- and by the way, thanks for pursuing this discussion and being
willing to play the "devil's advocate".
Senior Software Engineer
Development Solutions Business Unit
UNIX Suites Group
Aloha, OR, USA