Re: SOLVED! Decoupled Mutual Recursion Challenger

[jmartin@oahu.cs.ucla.edu (Jay Martin)]

>> ...Of course, social science programming experiments could
>>be performed to prove that certain language styles are more SE efficient
>>than others but this might step on some toes and besides its obvious
>>and boring. 

>I think the reason such experiments are not done is that they would be
>very expensive.  We can't experiment on rats, after all.

The rats are undergrad computer science majors.  It seems me
reasonable that every CS major to take part in say 2-3 quarters of
software engineering experiments on large scale programs.  I
believe they would learn alot about programming in the large. 
These experiments should get large chunks of computer science budgets
as they are really the heart of computer science (though currently
the computer scientists don't think so).  

>Sure, theoretically, you could compare Ada-with-discriminants to
>Ada-without-discriminants, using two groups of 100 teams of programmers,
>one for each language, all doing the same project.  But that experiment
>would cost hundreds of millions of dollars.  And that's just to
>investigate whether discriminants are a good idea.  What about all the
>other features of various languages?

Surely we wouldn't want to spend time trying to quantify micro
changes in languages, but big differences like dynamic versus/static
typing, etc.  Restrictive versus permissive, etc.   Big and
orthogonal, versus small/simple non-orthogonal.  Isn't getting some data
on these differences important?  These experiments could be run by a
few CS professors and grad students at each university for less than
hundreds of millions of dollars.  

>You could compare Ada with C++ using a similarly costly experiment, but
>what would that tell you?  Having found out which of the two is "better"
>a language designer would still want to know why.  Surely neither one is
>perfect; we should strive to improve on both.

I think this study would be beneficial.  And I believe Ada9x would
come out on top.  We would know why by reading the qualitative
analysis of each of the studies.  Unfortunately, I don't believe you
can improve languages indefinitely by adding features.  To really
improve C++ and Ada9x style languages alot of features would have to
be removed and the languages remade from the ground up.

>And, of course, nobody in "real" science trusts experimental results
>until they've been duplicated by several different researchers.
>That's why I think we're going to be stuck with seat-of-the-pants
>language design and anecdotal-evidence on their merits for at least some
>decades.
>They say that in computer science, one takes a single data point, and
>extrapolates from there.  It makes curve fitting easy.  ;-) ;-)

Heh, everyone knows that fields with "science" in their name are
always non-scientific jokes. :-)

If enough CS departments were doing such studies instead of zero, 
then we would have the data points to get answers to these big 
questions.  At least we would stop people from wasting their time on
endless arguments about these topics.

Jay.