From owner-freebsd-ports@FreeBSD.ORG Thu Jul 28 17:22:51 2005 Return-Path: X-Original-To: ports@freebsd.org Delivered-To: freebsd-ports@FreeBSD.ORG Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E4FAE16A41F; Thu, 28 Jul 2005 17:22:50 +0000 (GMT) (envelope-from neuhauser@sigpipe.cz) Received: from isis.sigpipe.cz (fw.sigpipe.cz [62.245.70.224]) by mx1.FreeBSD.org (Postfix) with ESMTP id 420D443D45; Thu, 28 Jul 2005 17:22:50 +0000 (GMT) (envelope-from neuhauser@sigpipe.cz) Received: by isis.sigpipe.cz (Postfix, from userid 1001) id 34A5B1F87BF1; Thu, 28 Jul 2005 19:22:49 +0200 (CEST) Date: Thu, 28 Jul 2005 19:22:49 +0200 From: Roman Neuhauser To: Mark Linimon Message-ID: <20050728172249.GD66015@isis.sigpipe.cz> Mail-Followup-To: Mark Linimon , Simon Barner , Paul Schmehl , ports@freebsd.org, Kris Kennaway References: <42E81050.7090305@cs.tu-berlin.de> <66A226C3557B48ED535E3FED@utd59514.utdallas.edu> <20050727230523.GB54954@isis.sigpipe.cz> <20050728154248.GA943@zi025.glhnet.mhn.de> <20050728164111.GA66015@isis.sigpipe.cz> <20050728170401.GA9534@soaustin.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050728170401.GA9534@soaustin.net> User-Agent: Mutt/1.5.9i Cc: ports@freebsd.org, Paul Schmehl , Simon Barner , Kris Kennaway Subject: Re: New port with maintainer ports@FreeBSD.org [was: Question about maintainers] X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Jul 2005 17:22:51 -0000 # linimon@lonesome.com / 2005-07-28 12:04:01 -0500: > On Thu, Jul 28, 2005 at 06:41:11PM +0200, Roman Neuhauser wrote: > > The policy makers won, everybody else lost. > > You're entitled to your opinion, but from the other side of the table > it looks like this: > > What Kris and I see are hundreds of ports that are committed and then > either a) are never updated, b) wind up not compiling when the base > system is updated, c) wind up with PRs filed against them by users, for > other problems ("doesn't work with XYZ"), that no one will ever deal with. > > In every case the users who install those ports aren't getting what they > expect. > > How does this situation serve the users? I am a member of the mythical "users" group, and I don't mind fixing whatever breakage I stumble upon as I go. > Let me mention today's statistics: > > Total number of ports: 13281 > Number of ports with no maintainer: 3670 (27.6%) > > I just can't see how this is a good situation. Would it be better if there were 9611 ports instead? How about "maintained" ports where the maintainers don't take proper care of their ports? I just don't see how no "designated maintainer" makes a port worse than port "maintained" by someone who refutes quite a bit of PRs with "I don't know, I only use minor part of the software, and that's all I can keep running". > I no longer have the statistics online but from the last time I went > through this it is about twice more likely that an unmaintained port: > > - has PRs against it > - is broken > - is out-of-date > > as versus the average maintained port. (Of course, some maintainers > are far more active than others.) > > And yes, it's true that he and I do the majority of the cleanup work to > flag and (if necessary) remove broken ports and so tend to be sensitive > to the issue. But the idea that we have is that it's better to have a > working port than a useless port. FMPOV the situation is "mostly working" vs none. > People need to decide what their vision of the Ports Collection is. > Some folks seem to want every possible port included, whether or not > it is up to date and working. I have, in particular, been trying to > push us towards the direction of only having ports that we are going > to use and maintain in it, on the theory that anything else is at some > point going to waste some user's time somewhere down the road. Again, I'm a user, and I rather spend 2 hours fixing a particular bug in an imperfect port than 8 hours creating it from scratch (and god knows how much time from then if I'm forced to maintain it). I've sent quite a few patches to various ports since I've started using FreeBSD, and my experience is that maintained ports (especially those "maintained" by certain @freebsd.org addresses) are more likely to have PRs hanging for a long time. The situation is a bit better with ports maintained by people without a @freebsd.org address, because committers usually don't bother waiting for approval. -- How many Vietnam vets does it take to screw in a light bulb? You don't know, man. You don't KNOW. Cause you weren't THERE. http://bash.org/?255991