I'm in favour of improving things where possible. I'm also in favour of staying reasonably compatible unless there's a compelling reason not to. Personally, I think the ability to use scripts from /etc/init.d after booting is more important than the details of how init runs them. Whether I use the rc-file or rc.d is a lot less important to me.
The specific idea of using dependacies in rc/init system is a good one. I've wondered about it myself a few times. However, I'm curious about how it handles state. I'd think it would require more than SysV runlevels. Is this in a released NetBSD, or is it just in CVS? On Sat, Jul 28, 2001 at 09:25:15AM -0500, Steve Price wrote: > On Sat, Jul 28, 2001 at 03:38:14PM +0400, Wartan Hachaturow wrote: > > > > I would like to point out again that we're not trying to change Debian, > > or the way it works. We're trying to get it to use a BSD kernel ;) > > In my mind it should mimic the Debian/Linux way everywhere possible. > > And if somewhere along the way this project finds something that > works even better than the Debian way, will you stick with the > lesser solution just because, ... well just because? While there > is no conclusive evidence (yet) that NetBSD's new rcinit code is > that much better than the current SysV scheme that Debian uses it > does have merits that are worth exploring. Making a better mouse- > trap is a good thing but if you let "policy" be your blinders then > it is less likely to happen. > > All IMHO of course. > > -steve > > > -- > To UNSUBSCRIBE, email to [EMAIL PROTECTED] > with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] >
