On Thu, 19 Sep 1996, Curt Sampson wrote:

> I didn't notice that any of these mentioned much about testing.
> Are there test requirements for these sorts of things? Do the folks
> responsible for this particular PR have to compile it up and test
> it themselves before it can go in?

In a word, "yes."

> Are they
> willing to trust someone they don't know (say, me) to test this
> fix, so that they don't have to do it themselves?

This is really not so much an issue of personal trust -- though, to some
degree, it must be -- as an issue of following design principles.  What it
comes down to is this:  the developer who commits code to the tree is
responsible to test that code before doing so.

As you've already guessed, that's certainly a part of why there is a
backlog.  (There are a number of problems that have multiple PRs open, but
that's another subject. :-)

--scott