From owner-freebsd-ports@FreeBSD.ORG Fri Jul 29 02:13:23 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 555DA16A41F; Fri, 29 Jul 2005 02:13:23 +0000 (GMT) (envelope-from linimon@lonesome.com) Received: from mail.soaustin.net (mail.soaustin.net [207.200.4.66]) by mx1.FreeBSD.org (Postfix) with ESMTP id C064344086; Fri, 29 Jul 2005 02:13:22 +0000 (GMT) (envelope-from linimon@lonesome.com) Received: by mail.soaustin.net (Postfix, from userid 502) id 5FAAF29A6; Thu, 28 Jul 2005 21:13:22 -0500 (CDT) Date: Thu, 28 Jul 2005 21:13:22 -0500 To: Pav Lucistnik Message-ID: <20050729021322.GB28471@soaustin.net> 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> <42E917BA.10406@exit.com> <790a9fff050728142793c7588@mail.gmail.com> <47ECFCB8BE498CEAB57757D7@utd59514.utdallas.edu> <1122588979.97751.1.camel@ikaros.oook.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1122588979.97751.1.camel@ikaros.oook.cz> User-Agent: Mutt/1.5.9i From: linimon@lonesome.com (Mark Linimon) Cc: ports@FreeBSD.org, Paul Schmehl 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: Fri, 29 Jul 2005 02:13:23 -0000 On Fri, Jul 29, 2005 at 12:16:19AM +0200, Pav Lucistnik wrote: > > Thanks, Scot. That's helpful to know. I wasn't aware, for example, that > > anyone could submit PRs for a broken port. > > Please don't do that, unless you also submit a fix. > > Because committers are few and usually are not very familiar with the > given broken port. Such PRs tend to be left untouched for years and then > swept under the carpet too often. I'm going to disagree with Pav here. If a port doesn't work, we ought to know about it so we can at least e.g. mark it as IGNORE like: IGNORE= "installs but core dumps" Then if someone wants to install it, at least they know that there's rough water ahead, and, if they aren't technical, to just avoid the port altogether. This way we've at least told people what to expect, upfront, to try to save them time. To reiterate for people that don't know, we use BROKEN for ports that fail the install/deinstall sequence or don't fetch or break INDEX builds. These are the ports that are periodically removed*. IGNORE is used for all other kinds of failures. Ports that meet this definition of BROKEN from day one, should never be committed in the first place. That's not just my opinion, I think I can speak for portmgr on that one. That's the least level of QA that I believe is required. mcl * ok, FORBIDDEN ports are periodically removed as well. Those are ports with security problems. And, for sake of completeness, ports with licensing problems can be and are removed without prior notice if the software authors so insist. But to date we have not removed any port just_ because it is unmaintained, nor do I think there would ever be any support for such a thing.