From owner-freebsd-hackers Sat Mar 25 07:27:54 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id HAA24445 for hackers-outgoing; Sat, 25 Mar 1995 07:27:54 -0800 Received: from localhost (localhost [127.0.0.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id HAA24437; Sat, 25 Mar 1995 07:27:51 -0800 X-Authentication-Warning: freefall.cdrom.com: Host localhost didn't use HELO protocol To: Amancio Hasty cc: Nate Williams , hackers@freefall.cdrom.com Subject: Re: httpd as part of the system. In-reply-to: Your message of "Fri, 24 Mar 95 23:10:30 GMT." <199503242310.XAA01112@star-gate.com> Date: Sat, 25 Mar 1995 07:27:49 -0800 Message-ID: <24435.796145269@freefall.cdrom.com> From: "Jordan K. Hubbard" Sender: hackers-owner@FreeBSD.org Precedence: bulk > > Why would we add a tool to mh as part of the standard system when mh is > > NOT part of the standard system? It seems kind of foolish to me. > > Is not foolish obviously is by implication. Guys, guys! 1. Amancio is right, in principle. We do need a better applications suite like this at some point if we're to be taken seriously on the desktop. I am rather tired of people telling me about how they run Linux on their desktop and FreeBSD as their server because "Linux has such a nice desktop, and its interactive response for a single user is so much better." Perhaps, in the past, their response time WAS better, but I think David and John have gone a long ways towards fixing that, and we've always known that our *response curve* (e.g. as you add more users/processes) was quite a bit better. If we can add to this (either by port, package or dist) a very comfortable out-of-box configuration, we'll finally be able to tell them that they can run FreeBSD on BOTH. 2. Nate is right.. :-) We can't go down this road right now, at least not until we have a much better installation framework. If it weren't ALSO for the doc team's needs (and folks - the doc teams needs are PARAMOUNT right now - we have NO doc! :-( :-( ) I wouldn't have suggested lynx and httpd at all. So let's not let my earlier suggestion run away with us. That wasn't a call for the floodgates to open, it was a call for *two* tools. When we can truly "classify" each and every binary, library and include file in the system (perhaps thru a special Make variable) so that it automagically knows which dist to stick itself into, and we have things like a `netdist' and a `maildist' and a `whateverelsedist', THEN will be a good time to start soliciting suggestions for really expanding the system. For now, I suggest that we make the ports and packages a LOT more highly integrated into the installation process so that, to the user, the edge of distinction blurs. I agree with Amancio about providing out-of-box Internet solutions, I just don't agree that NOW is the time to start piling on the solutions. 2.1 is simply too imminent, and we've too much else to do right now. Who knows? If the package stuff starts looking *really* nice, we may even find that our need to package things into any of the base distributions goes away. Jordan