Date: Wed, 28 Aug 1996 10:38:17 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: jkh@time.cdrom.com (Jordan K. Hubbard) Cc: terry@lambert.org, rkw@shark.dataplex.net, p.richards@elsevier.co.uk, hackers@freebsd.org Subject: Re: Am I wrong or is this just stupid?r Message-ID: <199608281738.KAA27008@phaeton.artisoft.com> In-Reply-To: <2853.841209373@time.cdrom.com> from "Jordan K. Hubbard" at Aug 27, 96 10:16:13 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> Conservatively > speaking, that's at least 400 messages which won't get us one iota > further towards solving the problem. I have to disagree. What makes this particular problem special? If nothing, then what is the porpose of the lists? If something, then identify it and let Richard route around the damage. > I'd prefer to simply see the prototype and thus constrain the ensuing > discussion to more practical details. Most people just aren't > adequately taking into account the fact that we HAVE been discussing > the make system on and off for over 3 years, and discussion alone has > accomplished very little. No one has been given the equivalent of "Congressional Fiat" for the build process, as opposed to other areas, where Fiat has been granted (or siezed -- a model which doesn't work once entrepeneurial kingdom building has set in... an inevitability of the entrepenureal model; I'll discuss my experience with various sizes of employers here, if you think it necessary or relevent). > There are many tried-and-true techniques from industry that just don't > translate well to the FreeBSD project and can't without creating the > kind of "accountability hierarchy" that only paying people salaries > can provide. Yes; however, there are issues of policy which are routinely decided. I believe this is an issue of policy, not an issue of trying to map a business model to a volunteer effort, while trying to avoid paying the cartographer to do the mapping. At least, it is mostly a matter of policy... some of it is inertia. > I've had 3 years with this project, seeing which techniques can be > brought straight across from industry and which don't have a > snowball's chance in hell of working out for us, and I can say that > some of what Richard wants is simply not practical, nor is it likely > to become so in the near future (if ever). I agree -- some of what he wants. Can we move on to draw the distinction necessary to get the rest rolling for him? ...Assuming he's willing to do without that "some", once it's been seperated out. > Anyone is free at any time to do what he damn well pleases with the > system, just as the development team and the users are free to adopt > it or not (and, contrary to popular belief, the users have just as > much power as the developers do here - if they reject something and > refuse to use it, it invariably withers and dies just as surely as it > would if a developer killed it). I can no more give Richard a "fiat" > as I could block him from changing this part of the system and > distributing his own diffs, even using our own announcement lists to > reach his intended audience. The sword cuts both ways! Agreed; but to pursue that course to it's logical conclusion, Richard has to be willing to supplant the existing authority. In that direction lies sufficient conflict to either resolve the issue by force (unpleasent) or cause a split along a fracture line (more unpleasent) based on who agrees with him. > > Probably, he would be better off contacting the core team directly for > > that kind of assignation. > > Or finally realizing that "assignation" is not how it works at all, > nor has it ever worked that way. People "take charge" of sections of > the system simply by working on them, just as I have for the > installation, Garrett has for networking, John & David have for the VM > system, etc. That's it, that's how it works. Do the work and the > rest follows. Sometimes an area of the system is already "spoken for" > and people have to share the responsibilities and/or coordinate > closely with others if they want to join in, but it's still the > quality and quantity of tangible work, submitted over a reasonable > period of time, which leads to the other developers trusting someone > implicitly to handle large parts of FreeBSD development. I think the problem is that there is very little unclaimed territory at this point. Richard seems to be dangerously close to some peoples borders; he want's to build an island, and he doesn't want someone's navy coming in and planting their flag between the time he builds the island and the time he builds his hotel. 8-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199608281738.KAA27008>