From owner-freebsd-advocacy Tue Mar 9 8:55:56 1999 Delivered-To: freebsd-advocacy@freebsd.org Received: from lariat.lariat.org (lariat.lariat.org [206.100.185.2]) by hub.freebsd.org (Postfix) with ESMTP id 056E314FC9 for ; Tue, 9 Mar 1999 08:55:53 -0800 (PST) (envelope-from brett@lariat.org) Received: (from brett@localhost) by lariat.lariat.org (8.8.8/8.8.6) id JAA12256; Tue, 9 Mar 1999 09:55:23 -0700 (MST) Message-Id: <4.1.19990309092847.04176b50@localhost> X-Sender: brett@localhost X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 Date: Tue, 09 Mar 1999 09:40:56 -0700 To: Wes Peters From: Brett Glass Subject: Re: Ports Cc: Bill Fumerola , Brett Taylor , Adam Turoff , freebsd-advocacy@FreeBSD.ORG In-Reply-To: <36E4D59E.89ACD0F4@softweyr.com> References: <4.1.19990302163944.00a1e620@localhost> <4.1.19990308213030.03ea5c80@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-advocacy@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG At 01:02 AM 3/9/99 -0700, Wes Peters wrote: >There is one really simple reason: somebody has to DO it. A port is based >on a particular snapshot of the sources; if the developers move stuff >around the sources the port patch files don't work anymore. True. What we need is the latest version of the ported app. The port system should be engineered, though, so that any version of the OS released within the last year will be able to run it. Yes, this might mean some library dependencies that have to be handled, but the port/package code already has the ability to handle this! >It would >take constant work for ports to be maintained for any given release. Why? How much work would it take in addition to what a port maintainer already does? (There aren't THAT many releases per year, and the a.out-to-ELF transition is a one-time delta. Even this can be solved by making the older versions accept ELFs. After all, the Linux emulation module loads ELFs now. All that's necessary to load a FreeBSD ELF is NOT to swap the APIs.) >Since the port maintainers are more interested in doing this for -stable >and/or -current, I nominate YOU to provide this valuable server to the >"trailing edge" corps. Again, it should be possible to make this part of maintaining a port, so that it happens "automagically" (or pretty much so). I've suggested some ways to do it, but I don't "own" that code and am probably not the best person to modify it -- at least not without consultation. Who owns the code for: The package manager? The Linux emulation module? The utilities that the port maintainers use to prepare ports for publication? >> If someone out there is taking the time to do builds, I should be able to >> use them. Otherwise, the port maintainer's time and effort are not >> being used effectively to bring the most benefit to users. > >This is a volunteer project. Many of the prot maintainers are using >their time effectively to solve THEIR needs, and then sharing this >with all other FreeBSD users. I'm sure that at least some of the port maintainers have mission critical systems which are a release or two behind the "bleeding edge" and need this. Likewise, I'll bet others are CVSuping their systems every day. Support should be consistent. Yes, this is a volunteer project. And since the FreeBSD Project prides itself on professionalism, it should be more than willing to add that "professional edge" to what it does. >If you want something else, feel free >to contribute it back to the project when you're done. I'm contributing the idea and a proposed methodology, and would be delighted to contribute to the code. However, as I said earlier, I'd like to work with the people who currently view the code as their "territory." In the past, I've found that I've "stepped on toes" when I've just tried to contribute a change without working with the people who feel they have "ownership" of certain parts of the system. --Brett To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-advocacy" in the body of the message