fair@netbsd.org wrote:
> Synopsis: ksh problem if last command of pipe is a shell built-in
>
> State-Changed-From-To: feedback->closed
> State-Changed-By: fair
> State-Changed-When: Sat Apr 22 21:05:58 PDT 2000
> State-Changed-Why: 
> This is not really a problem - it's normal behavior (see PR history), and
> the proposed change would probably break other scripts. Since there has been
> no feedback to this PR from the submitter (or anyone else) it's time to close
> it.

I have responded, actually -- but the gnats system never seemed to 
like my responses.

The HPs and sparcs that I use do not exhibit this behaviour, and neither
did NetBSD until shortly before I posted this bug.

The spec says that either answer is acceptable.
But it certainly makes more sense to me (and other
shell programmers I know) to have the last command
of a pipe NOT be run in a subshell -- it's entirely
counter-intuitive.

-- Chuck