Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!mstar!mstar.morningstar.com!bob From: bob@MorningStar.Com (Bob Sutterfield) Newsgroups: comp.protocols.tcp-ip Subject: Re: Diving into TCP/IP again... need help finding docco... Message-ID: Date: 22 May 91 21:37:49 GMT References: <70ECA8FF2AFF405D34@alex.stkate.edu> <17939@celit.fps.com> Sender: usenet@MorningStar.COM (USENET Administrator) Reply-To: bob@MorningStar.Com (Bob Sutterfield) Organization: Morning Star Technologies Lines: 29 In-Reply-To: dave@fps.com's message of 22 May 91 18:21:56 GMT ...since we're unlikely to ever hook up straight to the real Internet is there any reason not to simplify things by using a class A network number? Yes but you may hook up to a real internet, say 2 years, 1 year or 5 years down the road. You wish you had applied to the network gods and got real numbers. I have helped several organizations prepare for and achieve Internet connectivity, none of whom thought they'd ever have it when they first "designed" (I use that term loosely :-) their local network and connected those first two workstations to a little hunk of yellow Ethernet. Now that they have dozens or hundreds of machines talking IP, it's exponentially more difficult to renumber than it would have been if they had started with their own address in the first place. How many networks have you seen numbered 192.9.200? Sun did the world a favor by introducing them to IP, but didn't emphasize the need to acquire their own slice of the address space. One very large local organization has been using 192.9.200, ..201, ..202, ..203, and ..204 for years. Now it has become an issue of turf battles and intense internal politics, as they must renumber all those hosts and gateways. The transition from isolation to "connected status" is turning out to be far, far more painful than it needed to be. Do it right the first time. It's much easier. Amen.