🚀 go-pugleaf

In article <1912@accuvax.nwu.edu> vcc@delta.eecs.nwu.edu (Guest from Vogelback) writes:
> if I wanted to charge for software support and revisions, I'd want not
>to write good code so my clients would need to keep me around.  This seems
>contrary to all that RMS and FSF are attempting to encourage:  if their
>version of emacs were *really* buggy then we'd all have a reason to keep
>them in business, right?

No.  If you had a choice, would you choose the author of an extremely buggy
program to maintain it?  Sure, he knows the program better than anyone
else, but he seems like a lousy programmer.  Since FSF provides source
code, you can find a *good* programmer to fix it up.  Also, since you
didn't have to pay for the program, you don't feel a major loss if you have
to toss it out because it doesn't satisfy your needs.

Also, maintenance is not just fixing bugs, it's also enhancements (no
matter how good a programmer you are, you won't think of all the possible
features).  Customers probably aren't going to bother asking for lots of
enhancements to a lousy program (and certainly not from the lousy
programmer who wrote it).  If the program is good, the users will want to
use it more and will want it to do more.

By the way, my father once had a programmer in his company's computer
department (which consisted of one senior programmer, one junior programmer
(this guy), an operator and some keypunch operators) with the above
attitude.  He did some contract work on the side, and told me outright that
he intentionally put bugs in his contract code so that he would have job
security.  The guy was not really very sharp, so I was hardly surprised at
his behavior.
Barry Margolin, Thinking Machines Corp.

barmar@think.com
{uunet,harvard}!think!barmar