Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!zaphod.mps.ohio-state.edu!swrinde!ucsd!ucbvax!asylum.sf.ca.us!romkey From: romkey@asylum.sf.ca.us (John Romkey) Newsgroups: comp.protocols.tcp-ip.ibmpc Subject: How are these packages related? Message-ID: <9002131155.AA08516@asylum.sf.ca.us> Date: 13 Feb 90 19:55:36 GMT References: <9002130419.AA10279@wubios.wustl.edu> Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: romkey@asylum.sf.ca.us Organization: The Internet Lines: 56 PC/IP was first written at MIT as a research project started in late 1981 to see if you could get TCP/IP to operate on something as small as an IBM PC (which was brand new at the time). I believe it was the first TCP/IP for the IBM PC. There wasn't yet an ethernet card available for the PC when we started the project. The project was run by Prof. Jerome Saltzer, and there were several people hacking on it, including me. I ended up dealing with a lot of it in its last couple of years. Not only could we run TCP/IP on an IBM PC, but people actually wanted it. We shipped several nearly-public-domain distributions. CMU picked up PC/IP and ported it to Microsoft C (the MIT version compiled with a cross compiler running under 4.x BSD UNIX because when we started there were no commercial C compilers for the IBM PC - yes, "you're lucky, when I was a child we had to lick the roads clean with our tongues"). CMU also added a few enhancements and fixed some bugs. Most of the work there was done by Drew Perkins. Other universities, like Stanford and UMD, picked it up and modified it in various ways. The Stanford version became restricted to only academic use, and was licensed to Bridge (later to merge with 3COM), and The Wollongong Group. TWG eventually heavily rewrote it. UMD's version ended up being licensed to IBM. I believe that Sun used a little teeny bit of it for PC/NFS, which bears no resemblance to PC/IP whatsoever. And several of us from MIT ended up creating FTP Software and heavily rewriting PC/IP as well. In the last few years, there hasn't been much organized effort put into PC/IP. Most of the people who've hacked on it have gone on to other things. Dan Lanciani at Harvard has put together a distribution that incorporates a number of additional things people have done for it over the last few years, and that version is available for anonymous FTP from Harvard. Other free TCP's for the PC have sprung up, like NCSA Telnet (which is public domain) and Phil Karn's KA9Q (which is restricted to non-commercial use). Neither of these are related to PC/IP beyond the fact that they run on PC's. The last PC/IP out of MIT and CMU had some pretty serious limitations: no FTP. Its TCP was limited to one actively-opened connection, hence the lack of an FTP, which requires passive open and multiple connections. Incomplete IP and ICMP support. Not many applications. Lots of other little things. When it was first released, it was great, because it was state of the art, but the art has moved forward a lot, and PC/IP has stayed stagnant. I know that the Harvard distribution corrects some of the problems and adds many new features to the package, but I'm not really sure what state it is in. Partially related, the smart card vendors like Excelan (a Novell company), CMC and Interlan all used BSD ports for their smart-card TCP's. Excelan started off with a port of 4.1c BSD TCP. 4.1c didn't receive wide distribution, and the programming interface was like 4.2BSD sockets, but not exactly the same. The others started off with various forms of 4.2BSD. - john romkey USENET/UUCP: romkey@asylum.sf.ca.us Internet: romkey@ftp.com In the kingdom of the blind, the one-eyed are kings.