Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 20 May 1997 22:22:52 -0700
From:      "Jordan K. Hubbard" <jkh@time.cdrom.com>
To:        John Polstra <jdp@polstra.com>
Cc:        rob@ideal.net.au, current@FreeBSD.ORG
Subject:   Re: IPDIVERT broken? 
Message-ID:  <18183.864192172@time.cdrom.com>
In-Reply-To: Your message of "Tue, 20 May 1997 20:06:09 PDT." <199705210306.UAA12824@austin.polstra.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
> Yes, it was broken on April 27 by Garrett's "long-awaited
> mega-massive-network-code-cleanup.  Part I."  Do a "cvs log
> src/sys/sys/socketvar.h" for the gory details.

Am I alone in thinking that changes which break other things in the
source tree should be fixed by the changer?

I mean, not to beat on Garrett's head or anything, but seriously: If I
changed the interface for something in a standard library and parts of
the world suddenly started breaking as a result, wouldn't I also be on
the hook to go fix those things?  Nobody would cut me any slack at all
for saying "Naw, I don't have time.  If you want to recover from this,
fix it yourselves."

It seems that if we are to survive as a project in the long-term,
developers are going to have to take greater responsibility for their
actions and be willing to follow *all* the way through on any changes
made, repairing the results of any interface changes and essentially
just being willing to make things work again on a tree-wide basis if
they break.

Since I include documentation in this category, one could even say
that I screwed up in doing the rc.conf change without also committing
a man page at the same time, so I'm not calling the kettle black, I'm
simply saying that both it and the pot need to clean up their
acts. ;-)

					Jordan

P.S. The man page is sitting in my tree, half completed.  Will work on
changing that.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?18183.864192172>