From owner-freebsd-chat Wed Dec 19 5:18:18 2001 Delivered-To: freebsd-chat@freebsd.org Received: from avocet.prod.itd.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50]) by hub.freebsd.org (Postfix) with ESMTP id 8F2F437B405 for ; Wed, 19 Dec 2001 05:18:13 -0800 (PST) Received: from dialup-209.245.143.40.dial1.sanjose1.level3.net ([209.245.143.40] helo=mindspring.com) by avocet.prod.itd.earthlink.net with esmtp (Exim 3.33 #1) id 16Ggbf-0002Us-00; Wed, 19 Dec 2001 05:17:59 -0800 Message-ID: <3C209385.7A853E1A@mindspring.com> Date: Wed, 19 Dec 2001 05:17:57 -0800 From: Terry Lambert X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony} (Win98; U) X-Accept-Language: en MIME-Version: 1.0 To: Nils Holland Cc: Jeremy Karlson , Brett Glass , Craig Harding , chat@FreeBSD.ORG Subject: The Dirty Little Open Source Secret References: <4.3.2.7.2.20011218095233.028ea920@localhost> <20011219111842.B2186@tisys.org> <20011219113809.A2730@tisys.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Nils Holland wrote: > Imagine we started to re-write all GPL code currently in use. That would > take us, well, let's say a year or two. While the rewrite is in process, > there may be some things that users would need much more. Although it isn't > true, let's assume that FreeBSD only supported IDE hard disks up to 8.4 GB. > Now, this would be something that we'd better get fixed, but as everybody > is concentrating in rewriting GPL code, we cannot fix it (replace the hard > disk example with anything else that comes to your mind, the effect would > be the same). I have been involved either directly or tangentially with the genesis of no less than 7 Open Source projects which are still active. Let me let you in on The Dirty Little Open Source Secret: people work on whatever the hell they want to work on, and what someone thinks a project "needs most" only matters if that person is doing the coding, and then that's only a necessary, not a sufficient, condition. There are only a few well known ways to control this, and it's more a case of herding cats, than anything else: 1) Write the code yourself 2) Refuse to permit commits of code that doesn't mesh with your view of things (most people can't do this) 3) Beg, Plead, Argue, Cajole and/or blackmail 4) Pick a politically correct license for youtr target developer participants 5) Do the design, and get someone else who is bad at design but good at coding to code it for you 6) Assign it as class work/senior project/thesis/dissertation (works best if you are a professor or in charge of interns) 7) Bribe (hiring someone as an employee does wonders for their choice of topics and adjusting their attitude to be more to your liking) 8) Crib the code from somewhere else 9) Fork the project, and hope enough people agree with you, so that if you get hit by a bus (likely driven by someone in the former project), it will outlive you (if it doesn't, congradulations: you started a cult, not a religion) 10) Fast three days and pray for guidance and/or a miracle, such as code delivered to you on a golden CDROM. The only sure way to win supported and volunteers in Open Source projects is to provide all necessary and sufficient conditions: A) Working source code (this is the #1 requirement; learn from the mistakes of others here: declaring a project does not make it so [JFS for FreeBSD; Mozilla; etc.). B) A community o Mailing lists o Web presence o Source code control (learn from mistakes here, too: Open Source Projects have their own, unique, souls; you can't cookie-cutter them and expect something like a Linux to pop up, fully formed, like a child from Cronus' head) C) Continuity (like it or not, you _will_ be hit by a bus, eventually, or people will assume you will, so you better have a means of instilling confidence) D) A cool name doesn't hurt So: arguing about whether or not something is worth doing is OK, but unless you are incredibly persuasive, you are unlikely to get your way. Also, if your goal were to "get rid of GPL code in FreeBSD", you could do a lot worse than #8 (OpenBSD is a good source of code for that). If, on the other hand, you want to prevent that happening, well, your options are a lot more limited, and just preventing it won't get people to work on what you think needs worked on, so you might as well not be as anal about things: chill out, and go with the flow. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message