BSD bugs -- round 2

[Jeff Hakner]

	Now that the first wave of reaction to my original
posting ("Bugs in the BSD sources ??") has subsided, I shall
move forward.  First, let's summarize what we have learned thusfar:

1) One poster errantly assessed my posting to be a complaint
   about the existence of bugs in software.  Obviously, there
   DNE software s.t. lim (t-->inf) failures(t)=0.  What I was
   concerned about is the apparent fact that this software is
   being released without first being compiled, let alone tested.

2) It is generally agreed upon (by me as well) that we should be
   thankful that this software exists at all.  Clearly, we
   ought to appreciate the efforts of its authors.

3) However, I still do not understand how these kinds of bugs
   could be distributed.  My typical development cycle goes:
		think-->write code-->compile-->run-->find bugs-+
                 ^---------------------------------------------|
				                               |
							      /
                                 package <--------------------
				    v
			       double-check
			            v
	                 release to the universe

	What seems to be happening here is:
		make working release-->create some previous
			delta release that doesn't work-->release that
4) How can we obtain the correct sources?  No one has yet answered 
   this question.

Let the second round begin!
Jeff Hakner
Cooper Union