🚀 go-pugleaf

   Date: 5 Nov 90 17:17:00 GMT
   From: connolly@convex.com (Dan Connolly)

   The GNU development effort is different from typical commercial
   vendors in that it's largely "user contributed" and comes "with no
   warranty, expressed or implied." In short, no company rests its
   reputation on the quality of GNU software.

Ahem... there are several.  Cygnus, NeXT, Data General...

   Enough complaining. What I propose is that the GNU project put more
   emphasis on automated regression testing. Each piece of software
   that is release should have a body of tests released with it. In
   practice, these tests are more useful than documentation in
   providing specifications for a product.

Well, this is one service companies can pay for.  In fact, our
customers pay for this; you can just get the software off the net or
you can sign up for service on a copy which has been more heavily
tested and debugged.  In other words we sell a warrenty, and to keep
from going broke we do a lot of testing.

The fact is, testing is difficult and boring.  Tests are usually
tedious to write and often difficult to run.  One reason big companies
have a hard time getting software out quickly is the time it takes
them to do regression testing and correction.  That's why you get
software with sheets listing "Known bugs."  That hardly happens with
GNU software because if they're known, they're fixed.  But yes,
usually many are not identified.

You really hit the nail on the head in your first paragraph above.
Most of the GNU development effort is user-contributed, and few users
have contributed tests.  You're a user, and could contribute tests.
And the companies so far that do rest their reputations on GNU
software of course do their own testing.  Sending debugged code back
to the foundation, which we all do, doesn't guarantee that it will
stay debugged; some unrelated change may break it.  So testing is
always required.