Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 15 Aug 2011 03:59:16 +0400
From:      Dmitry Marakasov <amdmi3@amdmi3.ru>
To:        Matthias Andree <mandree@FreeBSD.org>
Cc:        Alexey Dokuchaev <danfe@FreeBSD.ORG>, Doug Barton <dougb@FreeBSD.ORG>, Baptiste Daroussin <bapt@FreeBSD.ORG>, cvs-all@FreeBSD.ORG, ports-committers@FreeBSD.ORG, Chris Rees <crees@FreeBSD.ORG>, cvs-ports@FreeBSD.ORG
Subject:   Re: cvs commit: ports/cad/admesh Makefile
Message-ID:  <20110814235916.GF38385@hades.panopticon>
In-Reply-To: <4E47EF20.8000409@FreeBSD.org>
References:  <CADLo839VotWbi209%2BeR3LAuQ4HqC-LSGY4avcUFXHat=HjsrKg@mail.gmail.com> <20110812093328.GE85247@hades.panopticon> <b0535f6d53bb546b54d85797ec66cf0b@etoilebsd.net> <20110812101133.GF85247@hades.panopticon> <4E4584EA.7090306@FreeBSD.org> <20110813133717.GA38385@hades.panopticon> <4E469837.1030903@FreeBSD.org> <20110813172040.GC38385@hades.panopticon> <20110814030033.GA80255@FreeBSD.org> <4E47EF20.8000409@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
* Matthias Andree (mandree@FreeBSD.org) wrote:

> Assuming that were true, how else do we make sure not to let rotten code
> linger in the ports tree?

There's no way, not even with "maintained" ports. "Rotten code"
doesn't build or doesn't work, is easily distinguished and should
obviously be marked broken, while we actually are talking about
(if you like) "rotting code", which only difference from normal
code is a higher probability of becoming rotten. I don't think we
mark ports as whatever based on just probability, and I insist on
leaving it alone.

Returning to ports@, it really is a fine maintainer entity, given
general inferiority of other maintainer entities. There are no
multiple maintainers, there is no automatic assignment of ports,
there is no distinguishing between people who are willing to just
solve problems and people who are willing to keep the port up to
date (latter requires more effort), there are too little groups,
and if a port is assigned to a group, there's still no designated
maintaner, so it's less likely someone from a group will update a
port, making a group not much different from ports@.

Casual workflow I've witnessed many times: maintainer doesn't use
the port (as extensively) any more -> he drops maintainership (for
other to pick up, but no one does because that's not really needed
to take care of a port). He still can and is willing to fix problems,
but he doesn't really care about updates. Other person submits the
update, but doesn't take the maintainership. The result is a ports@
port having at least two maintainers by fact, but none listed as
such, both without means to monitor own ports (portscout, portsmon),
etc. Port with a maintainer officially, who has given up by fact,
however, may sit in a tree for years without update until it breaks.

Not sure that it counts as a proof, but by portscout, 12.81% of
maintained ports need an update, but only 6.64% unmaintained ones
do:

for x in '' '-v'; do
	wget http://portscout.cc -qO- |
		grep $x 'ports@freebsd.org' |
		grep '^<tr><td>' |
		sed -e 's,<[^<>]*>, ,g' |
		awk '{total+=$2; new+=$3}END{printf("Total: %d, new: %d (%.2f%%)\n", total, new, new/total*100);}'
done

-- 
Dmitry Marakasov   .   55B5 0596 FF1E 8D84 5F56  9510 D35A 80DD F9D2 F77D
amdmi3@amdmi3.ru  ..:  jabber: amdmi3@jabber.ru    http://www.amdmi3.ru



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110814235916.GF38385>