From mboxrd@z Thu Jan 1 00:00:00 1970 From: pnr@planet.nl (Paul Ruizendaal) Date: Mon, 16 Jan 2017 11:31:32 +0100 Subject: [TUHS] Early Internet work (Was: History of select(2)) In-Reply-To: <20170116010101.8DF6F18C083@mercury.lcs.mit.edu> References: <20170116010101.8DF6F18C083@mercury.lcs.mit.edu> Message-ID: On 16 Jan 2017, at 2:01 , Noel Chiappa wrote: > >> From: Paul Ruizendaal > >> I guess by April 1981 (RFC777) we reach a point where things are >> specified to a level where implementations would interoperate with >> today's implementations. > > Yes and no. Earlier boxes would interoperate, _if addresses on each end were > chosen properly_. Modern handling of addresses on hosts (for the 'is this > destination on my physical network' step of the packet-sending algorithm) did > not come in until RFC-1122 (October 1989); prior to that, lots of host code > probably tried to figure out if the destination was class A, B or C, etc, etc. This is true of the Gurwitz implementation. The Wingfield implementation still uses the older form, where the first 8 bits signify the network and the remaining 24 bits the host address on that network. In terms of routing my view would be to keep it simple: traffic is either local or destined for the single interface / gateway (see below). Interop hence is just looking at TCP. > > Also, until RFC-826 (ARP, November 1982) pretty much all the network > interfaces (and thus the code to turn the 'destination IP address' into an > attached physical network address, for the first hop) were things like ARPANet > that no longer exist, so you could't _actually_ fire up one of them unless you > do something like the 'ARPANet emulation' that the various PDP-10 simulators > use to allow old OS's running on them to talk to the current Internet. Yes: all these old implementations have an IMP interface driver at their lowest level. What I'm doing for testing is replacing that by a SLIP driver so that I can hook up to today's network and see if it works. Paul