5 Jan
2012
5 Jan
'12
8:19 p.m.
On 5 Jan 2012 at 10:03, Fred Cisin wrote:
My impression was that each and every committee member
wanted a
completely different language than the current language. ("fortran
should look like Pascal!" "Fortran should be as 'elf-documenting' as
COBOL!", etc.) And, each one wanted complete standardization on THAT
committee member's unique implementation. OK, WE aren't exceptions.
I want every language to use .EQ. for comparision, and an arrow for
assignment. // for comments in C is actually convenient, but it is a
serious break from the original CONCEPT of newline being nothing more
than "whitespace".
My impressions of X3J3 left me with the feeling that it was less a
meeting of people interested in advancing the art, than a meeting of
lobbyists for outside interests.
Thus, IBM came to the vector extensions group with the specific
intention of ratifying VECTRAN as the standard, rather than producing
something useful. Others held the feeling that it would be a cold
day in Hell if any part of VECTRAN made it into the standard.
You get the idea.
But that's the way standards committees have always worked as far as
I know. The FLOW-MATICs against the COMTRANs.
The more curious thing is that few representatives, other than those
from the national research institutions (Sandia, LLL,
etc.) have much
experience in everyday usage of the the stuff that they're
trying to
standardize.. A substantial number of them come from the compiler
implementation segment, as in "I write COBOL, but have never written
anything in it other than a trivial "Hello World" program". Most
dutifully took the questions back to their employers to let the
marketing people chew on them.
I do not know if this holds for other committees, such as the
standards committe for C++.
Maybe standards committees are to software advancement as standing
armies are to civil rights--they're incompatible.
--Chuck