[log in to unmask] said:

 >   If one language is better than another by x%, then presumably we
 > could begin saving x% of the total software budget, each year for
 > the foreseeable future, by switching to that language.  Just 1% of
 > just $10**9/yr * 10 years = $100 million dollars.  Surely, that
 > would fund even an expensive set of experiments or data gathering.

    Hmmm.  For $100 million dollars, I can probably do that
experiment.  But for $10 million I can substantially improve the
software development process in the "weaker" langauge, and the
stronger one.

    In fact many years ago I was involved in such a project at
Honeywell, studying development in Ada, C, or Pascal vs. assembly
language.  Total cost was in the millions, for a total of six data
points.  But the most important result was that one of the C teams
wrote code that could be converted to Ada with a few emacs macros.
They could also do the reverse conversion.  This meant that their code
used a subset of Ada (and/or C) functionality, but gained the
advantages of strong type checking, etc.  How do you evaluate that
data point?  Push the money and/or the project size up, and you will
get evem more of the same effect.


                                        Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...