Date: Wed, 5 May 1999 09:58:46 +0200 From: Eivind Eklund <eivind@FreeBSD.org> To: Greg Lehey <grog@lemis.com> Cc: Matthew Jacob <mjacob@feral.com>, Guy Helmer <ghelmer@scl.ameslab.gov>, "Kenneth D. Merry" <ken@plutotech.com>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: Informing MAINTAINER (was: cvs commit: src/usr.bin/mt mt.1) Message-ID: <19990505095845.D49159@bitbox.follo.net> In-Reply-To: <19990505114235.X40359@freebie.lemis.com>; from Greg Lehey on Wed, May 05, 1999 at 11:42:35AM %2B0930 References: <19990504232121.C49159@bitbox.follo.net> <Pine.LNX.4.04.9905041522070.14744-100000@feral.com> <19990505114235.X40359@freebie.lemis.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, May 05, 1999 at 11:42:35AM +0930, Greg Lehey wrote: > On Tuesday, 4 May 1999 at 15:25:53 -0700, Matt Jacob wrote: > > Send the email. Wait a reasonable period of time (fudge factor- 10 line > > change == 10 minute, 100 line change == 100 minutes, 1000 line change == > > overnight) and then make the change. The point here is to allow a nominal > > maintainer a look- if they're not paying attention, that's not your fault. > > Increasing useful communications is good. > > I agree in principle, but not in detail. 10 minutes is never > reasonable when you consider that the nominal maintainer may be asleep > at the time, and that's relatively common when you consider that you, > Eivind and I live in time zones approximately 8 hours apart. I'd > think that a minimum of 24 hours should be reasonable unless the > change is urgent. Now consider it from the view of somebody doing changes - and they're usually doing changes because the maintainer has in some way been negligent (no offence meant). To use an example from one of the things I do: In two hours, I can fix up ~500 staticization bogons (assuming the source tree is slightly tuned for it, by including comments for those that are not supposed to be touched). Or I can, in an absolute best case, fully administer 50 e-mails to maintainers (including waiting for replies, tracking which replies has been done, etc). This is just a problem for certain kinds of changes; I'd guess it to be mainly style/cleanup changes (which can be done in fairly massive amounts quickly), and for PR handling of simple PRs (e.g, documentation/comment changes), where my guesstimate is that it would halve the productivity of the integrator. For other changes I fully agree that the maintainer should be mailed; I just don't agree that this should have a much larger priority than actually getting things done. Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990505095845.D49159>