🚀 go-pugleaf

I'd like to hear from someone who has used the WATCOM C compiler!
It's gotten rave reviews, and they have a good reputation for super-
optimizing compilers in the mainframe world.  Their compiler comes
with EVERYTHING, including an integrated environment for those of
you who like that sort of thing, complete with profiler and debugger.
News rags say that they just signed an agreement with Boreland to
get their debugger.


MY TWO BITS' WORTH:

I'm tired of hearing religious flames about Turdo C and MSC.  More
than that, I'm tired of the attitudes of the companies that produce
these compilers.  I would love to see someone take their business
away.  The rest of this article explains why I feel that way.

MICROSOFT:

My beef with Microsoft is that they use proprietary knowledge to compete
unfairly.  At one time they had undocumented DOS calls in their C
startup code.  Since they wrote the DOS, they knew the calls.  When
they announced OS/2 (that they wrote), guess who had the first compiler
available for it?  I think a bit a trust-busting is in order here.

That sentiment is probably the biggest reason for the large number of
Borland fans out there.  I used to be one for that reason.  No more.

BORLAND:

My beef with Borland is two-fold.  First, their products have a cheap,
unprofessional, sometimes even buggy feel and finish.  The second is
that their attitude toward their customers outrages me.

I switched from MSC to Turdo C in the middle of a commercial project
when it first came out.  It proved to be a big mistake to trust a
commercial project to an unprofessional tool, with no sympathy or
support from its makers.  I sent Borland a piece of code which demonstrated
a disasterous bug in their compiler.  When you compile it with certain
options, it would wipe out the root directory of the hard disk!  You
didn't have to RUN the program, JUST COMPILE IT with their compiler.
They CONFIRMED this error.  They did not even so much as appologize.
They just said, "We don't intend to do anything about it.  Wait for the
next version."  The letter I sent to them was sent via a CompuServ
posting, which disapeared very quickly.  Someone else on that forum
asked for a bug list.  Someone from Borland replied that they DID NOT
KEEP A BUGLIST!  (They were "fixing bugs so fast" that they "couldn't even
keep up" with it...)

I also think it is pretty raunchy of them to keep their compiler and
especially their libraries incompatible with Microsoft's _on_purpose_.
They asked Allen Hollub what it would take to get him to use their
compiler instead of Microsoft's.  He said, make it compile my existing
programs without modification.  They left with a frown.  This
incompatibility is a MARKETING issue, not an engineering issue.  That's
what burns me up about it.

CONCLUSION:

I can't stand any company that tries to control and manipulate their
customers.  I want them to win my business through excellence and
a good price/performance ratio.  I am enraged at companies that try
to win by business by force.  Microsoft and Borland have both
demonstrated this tactic.  I think they would be better received if
they would peddle their compilers in the Soviet Union.

SO I ADMIT THAT I HAVE PREJUDICES against certain products because of
the attitudes of the companies that make them.  Let's turn this potential
flame war into something constructive.  If you have a C compiler religion,
it doesn't help to state it unless you first look inside yourself and
tell us WHY you feel that way.  It is constructive to find out how a once
objective opinion can be soured with specific bad experiences.  It is not
constructive to chime in with a boolean opinion.

Jeff Pipkins
pipkins@imagen.com

My employer does not necessarily share my personal views.