Date: Mon, 16 Jun 1997 00:31:52 -0700 From: "Jordan K. Hubbard" <jkh@time.cdrom.com> To: joelh@gnu.ai.mit.edu Cc: molter@logic.it, adrian@obiwan.psinet.net.au, vas@vas.tomsk.su, chat@FreeBSD.ORG Subject: Re: To UNIX or not to UNIX ;-). Was: PPP problems. Message-ID: <18103.866446312@time.cdrom.com> In-Reply-To: Your message of "Mon, 16 Jun 1997 01:21:11 EDT." <199706160521.BAA16712@ethanol.gnu.ai.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
> Why not? The reason I wrote up the profile was as the basis of a > plan. The reason I asked for refinements is so that the plan can be > better. If we can come up with a plan, then I would be glad to design > it, and coordinate (and participate on) a team to code it. (Oh, dear, No, you've only outlined the _goals_. You haven't even come remotely close to outlining an actual plan yet. ;-) > What made the current distribution/release system work? What about > the current package system? How did these things come to be the vital > tools they now are? I or someone else just sat down and coded them. Sure, I could have asked for lots of input and general feedback before doing the package tools or the ports collection, but then we'd still be talking about it to this day if I had - see the Grand Unified Console Driver project for a good example of that process gone deeply wrong. I'm normally a big fan of getting lots of input from the crowd, don't get me wrong, but there are also many instances where this just plain doesn't work. You either get lots of good ideas which totally outstrip your actual time available (and the package tools are seriously flawed in many respects, but I didn't have time to addresss those issues and so had to live with the flaws - it was a very deliberate trade-off of time over features) or you get a lot of irrelevant threads as various folks use the discussion to try and advance their own agendas. Again, they may be very important agendas and worthy of further study, but you just can't solve 10 problems at once with any accuracy - sometimes you have to say "yes, I agree we need that too, now please shut up so that I can do this other stuff before the current year is over." :-) > I'm still a bit unclear. And since I just stuck my foot in the thick > of it, then could you please help me see what it needs? I think you already know that - you outlined some of the points more than reasonably in your "UNIX goals" message which sparked this. What we need now are more down-to-earth ideas. To draw an analogy, it's as if you have a room full of unwashed people. Everyone agrees that they stink badly and really could use a bath, but there's no running water or soap or any way of bathing them clearly in evidence. You then have the choice of either standing around and saying "my, those folks sure reek, are we all in agreement that bathing them would be a *really good idea*?" or you can start talking about where the nearest water is, whether any piping materials can be obtained and if someone could perhaps locate some empty oil drums and fill them by hand in the meantime, just to get a few of the people smelling a bit better. So, as I mentioned before, what we've been over so many times are the goals. We all know we have unwashed people and we all know that they'd smell a lot better clean. What we need to handle here is the problem Mike talked about, where one guy stands up and declares "OK, by god, I've located some plastic piping and I know where the river is! Who will join me in hooking it all up and creating a pipeline?" [dead silence] Jordan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?18103.866446312>