From owner-freebsd-hackers Wed Jan 22 02:57:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA28066 for hackers-outgoing; Wed, 22 Jan 1997 02:57:39 -0800 (PST) Received: from hda.hda.com (ip31-max1-fitch.ziplink.net [199.232.245.31]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA28061 for ; Wed, 22 Jan 1997 02:57:35 -0800 (PST) Received: (from dufault@localhost) by hda.hda.com (8.6.12/8.6.12) id FAA24690; Wed, 22 Jan 1997 05:52:52 -0500 From: Peter Dufault Message-Id: <199701221052.FAA24690@hda.hda.com> Subject: Under the desktop In-Reply-To: <12509.853847050@time.cdrom.com> from "Jordan K. Hubbard" at "Jan 21, 97 03:44:10 am" To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Wed, 22 Jan 1997 05:52:51 -0500 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk (Back from -chat) Someone: > > > > I kind of thought that the free Unices were all about the desktop market > > > > and recapturing the power Jordan: > > > > > > No. UNIX on the desktop is dead... (...) > > > ... We're a server OS now with minimal desktop functionality. me: > > No, wrong direction. Go under the desktop. The opportunity has > > traditionally come in at a lower point then the established base. > > Think $50.00 VAX 11/750 and where it could be applied. > > > > MS is going to aggressively battle the top end with NT while > > crippling the bottom end to avoid cannibalizing their product line - > > witness the purposefully broken server side named pipes in W95. > (...) > What do you suggest? > > Jordan Here are a few reference platforms in the net market. FreeBSD is present in some of them. 1. Big server system used in internet and intranet applications; 2. Oracle network computer box / computer lab box 3. Cheap disk based network server; 4. No moving parts network server; 5. Wicked cheap network interface. 1 and 2 are where the big boys are slugging it out, and 1 is where I think Jordan said we are positioned. 4 and 5 are where java boxes are supposed to plug in. I think that BSD solutions plug 3 and 4 well, and I think that Java underestimates what can be done at that level without, er, shifting your paradigm. I'm interested in 5 and have some ideas but believe that there is a lot of development work needed to do that job right. For me it would be the most fun, though. 1 for smaller ISPs and 1 together with 2 in the computer lab is probably the easiest market for FreeBSD to play in outside of general hacking. NT is going to try to steamroller that market. 3 is the reference VAX 11/750 platform - say 4MB RAM, 350MB disk, ethernet, etc. 4 is the no-moving-parts system that boots off the net and does some limited application - provides a net interface to a piece of production equipment or expensive copier, etc. Applications for 3 include fax machines, high end scanners, etc. Applications for 4 include manufacturing test, test equipment, PROM burners, bus analyzers, etc - all of those things that people make you inconveniently dedicate a PC for. There isn't anything earth shattering here, only a few implied requirements on support tools and regression testing. 2, 3, and 4 all need tools that have a small footprint in the target and run remotely and that can easily be applied to multiple systems at once. For example, in your computer lab you want to network boot something into the install program and have the probing and kernel configuration for a sysinstall driven by the server. 4 needs some pieces that we don't have but that can probably be assembled just by shelling out some money and writing a little code; I'm considering prototyping that and welcome people sending me what they think they need. Ramble on, Peter -- Peter Dufault (dufault@hda.com) Realtime Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936