Date: Mon, 15 May 2000 13:48:19 +1000 (EST) From: Darren Reed <darrenr@reed.wattle.id.au> To: grog@lemis.com (Greg Lehey) Cc: vanderh@ecf.utoronto.ca, imp@village.org, brian@CSUA.Berkeley.EDU, shocking@prth.pgs.com, current@FreeBSD.org, cvs-committers@FreeBSD.org Subject: Re: Breaking "build world" costs $5? (was: Can we please have a current that compiles?) Message-ID: <200005150348.NAA10190@avalon.reed.wattle.id.au> In-Reply-To: <20000515125429.C541@freebie.lemis.com> from Greg Lehey at "May 15, 0 12:54:29 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
In some email I received from Greg Lehey, sie wrote: > On Sunday, 14 May 2000 at 23:21:21 -0400, Tim Vanderhoek wrote: > > On Sun, May 14, 2000 at 10:23:07PM -0400, Tim Vanderhoek wrote: > >> > >>>>> Maybe the comitters ought to take an idea from many software companies and > >>>>> contribute $5 to the beer fund every time they break the build. Have it > >>>>> all come due at the next BSDcon to fund a committer beer bash. :-) > >>> > >>> I think that we should make it $5.00 the first time. $10 the second > >>> time. $20 the third time, etc. > > > > Or perhaps we should tie it to the time it takes the security officer > > to fix, either with his own patches or by inviting patches from > > others, security holes that have been reported in certain libraries. > > If it takes the committer longer to fix the world breakage than it > > takes the security officer to deal with the hole, then the committer > > may be fined. In all other cases, the security officer can be fined. > > At committer's discretion. > > > > We can then add bonuses. Say, for example, I fix someone else's world > > breakage fast enough that they are able to avoid the fine, then they > > have to pay me either $5.00, $10, or $20. > > > > In cases where more than one committer was responsible, either > > indirectly or directly, then all involved should be fined. > > I think you're missing the point. The intention was the equivalent of > the pointy hat, just with the additional benefit that others might > have something out of it as well, no more. Just one other thing to add here... How often is it that the actions of just one person make the build fail ? I imagine if one person commits code, it works and bulids cleanly for them, but it's when it's imported into a source tree that others have changed that the previous compile-testing can be rendered useless. If anything, such moves as this will make people less likely to commit code and hence further impede the development of FreeBSD. I know I personally do not have the resources (both physical and otherwise) to maintain a FreeBSD-current system to test any change I might want to commit. I see this money scheme as an extension of the "finger pointing" which does nothing to build team spirit. Darren p.s. since there was no currency specified, I imagine "$5 local", wherever that maybe, so the Italians would get off light at 5Lire :) p.p.s. for those of you that live inside the USA, sending such small amounts of money to the USA will cost more than the penalty, more than likely. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200005150348.NAA10190>