Re: language wars (results 13 September) last posting

[William Clodius <wclodius@lanl.gov>]

As always it is difficult to make sense of such an evaluation. Such
biases are inevitable, but note the following:

1.  The voters are selected only from that set of people with an
interest in and access to internet language news groups in the summer.

2.  The voters are selected only from those interested in a few of the
more popular newsgroups, i.e., no comp.lang.c++, comp.lang.forth,
comp.lang.java, comp.lang.cobol, comp.lang.functional, comp.lang.lisp,
etc.  Votes for such languages are probably more meaningful than votes
for other languages. I suspect that a posting to comp.lang.c++, for
example, would have greatly increased the number of C++ votes, similarly
for comp.lang.java.

3.  No postings to the moderated newsgroups, which tend to attract more
knowledgeable posters.

4.  The disallowal of first languages implies, at least in the US,
reduced votes for Basic, Pascal, C, and Fortran which may bias results
in Ada's favor.

5.  There is no way to weight the voter's experience with a variety of
languages.

6.  There is no way to weight the voter's degree of preference.

I suspect that ML stands for SML (Standard Meta-Language), but it is
unclear whether IDL stands for the Interface Definition Language or the
Interactive Data Language.

-- 

William B. Clodius		Phone: (505)-665-9370
Los Alamos National Laboratory	Email: wclodius@lanl.gov
Los Alamos, NM 87545