pml@casetek.casetek.UUCP (Pat Lashley) (07/15/87)
In the past couple of weeks, I have seen several requests for Public Domain
TCP/IP implementations; and no responses (presumably e-mailed to requestor ?).
This has led me to wonder if there is any such beast.
Anyone out there with information about ANY public domain or freely
redistributable TCP/IP implementation for ANY hardware/operating system
configuration, please e-mail me what information you have, and I will
post a summary (or statement of lack of response... :-).
Thanks,
-Pat
P.S. This is not _just_ idle curiosity. I think that the only way that I
can convince our managment to install TCP/IP on our VMS VAXen or PCs is to
find a way to reduce the cost per host to something that I could afford out
of my own pocket.
--
Internet: casetek!patl@sun.com PM Lashley
uucp: ...sun!casetek!patl CASE Technology, Inc.
arpa: casetek@crvax.sri.com Mountain View, CA 94087
>> Anyone can have the facts; having an opinion is an art. <<charny@GATEWAY.MITRE.ORG (M. Charny) (07/16/87)
Pat,
I have sent the following information to several people who have
asked about this. I hope it is helpful to you.
Manette Charny
-+-+-+-+-+-+-+-+-+-+-+-
The MITRE Corporation has produced several prototype implementations of
devices using the Department of Defense (DoD) communication protocols
TCP, IP, ARP, ICMP, and TELNET. Most of the implementations use the
CMOS/DMOS operating systems also produced by MITRE. The programs are
written in "C" language and Motorola 68000 assembler. The programs
are maintained on a SUN3 workstation running Berkely 4.2 UNIX.
MITRE is not in the business of distributing or maintaining software. Devices
and software created by MITRE are prototypes created for particular sponsors.
The programs are available to the outside world according to the conditions
and procedures listed below. **Please note that distribution is done on
a time available basis and therefore delivery turn-around time cannot
be guaranteed.**
The software is distributed free of charge. In order to obtain
a copy of this software, the requester should send to one of the
below listed people:
1. 2400 foot reel of 1/2 inch magnetic tape capable of handling 1600 bpi.
2. Letter, on your company letterhead, indicating the following:
-- who they are
-- short synopsis (2-3 sentences) about
what our software is to be used for
-- equipment and operating system being used by them.
-- tape format desired:
only format possible is Berkely 4.2 UNIX tar, 1600 bpi,
and any blocking factor 1 through 20. (20 by default)
-- they agree to the four conditions listed below
(please explicitly list the conditions in the letter).
i. The MITRE TCP/IP source files will not be passed on to
any third party.
ii. MITRE will be credited should the software be used
in a product or written about in any publication.
However, MITRE will not be referenced as the
source in advertisements.
iii. MITRE assumes no legal responsibility for source
code and its subsequent use. No warranty is
expressed or implied.
iv. If any bugs or problems are found they will
be reported back to MITRE.
NOTE: These programs are not commercial quality, but do work.
A good 'hacker' can learn to build 'boxes' by studying the 'config',
'table', and 'Makefiles' in the directories 'box/diag/testV',
'box/testV/1822', 'box/hfe', and 'box/gate_egp'.
The current contact(s) for the TCP/IP distribution tape are:
Manette Charny
Mailstop: W425
1820 Dolley Madison Blvd.
McLean, VA 22102
(703) 883-6728
ARPANET: charny@mitre-gateway
Daryl Crandall
Mailstop: W429
1820 Dolley Madison Blvd.
McLean, VA 22102
(703) 883-7278
ARPANET: daryl@mitre-gateway
Documents describing the OS and the TCP/IP implementation can be obtained
from MITRE document control.
"CMOS, A Portable Operating System in C"
Gilbert R. Berglass
MITRE Technical Report: MTR-84W00071
"DMOS, A Portable Distributed Operating System in C"
Shiraz G. Bhanji
MITRE Technical Report: MTR-85W00206
"Implementation of the BBN 1822 Host-to-IMP Protocol in a CMOS
Environment"
Manette Charny
MITRE Working Paper: WP-84W00223
"The MITRE Implementation of MIL-STD 1777: The Internet Protocol"
William S. Morgart
MITRE Working Paper: WP-86W??? (not released yet!)
"TCP/IP" Interface Specifications for CMOS Systems"
Daryl O. Crandall
MITRE Working Paper: WP-86W00180
"TCP/IP" Diagnostic Package for CMOS Systems"
Daryl O. Crandall
MITRE Working Paper: WP-86W??? (not released yet!)barry@confusion.UUCP (Barry Lustig) (07/16/87)
A "mini" TCP/IP was once posted to comp.sources.unix. You may want to look at the archives on uunet.uu.net. Barry Lustig Cognitive Science Lab Princeton University
gnu@hoptoad.uucp (John Gilmore) (07/21/87)
charny@GATEWAY.MITRE.ORG (M. Charny) wrote: > Subject: Re: PD TCP/IP requests > > MITRE is not in the business of distributing or maintaining software. Devices > and software created by MITRE are prototypes created for particular sponsors. > The programs are available to the outside world according to the conditions > and procedures listed below. > -- they agree to the four conditions listed below > (please explicitly list the conditions in the letter). > i. The MITRE TCP/IP source files will not be passed on to > any third party. How can this software be called "public domain" if it can't be passed on? Also, I am interested in who actually owns this software. If it was written on contract for the government, it is owned by the government, as a work for hire. Since the government cannot copyright things it owns, (this is explicit in the Copyright Act), this would mean that the code is in the public domain, and MITRE has no right to put these kind of restrictions on it. I presume that whoever contracted to have MITRE do this work is on the net. Or their DARPA sponsor is on the net. Can you clarify the ownership of the code and the origin of these restrictions? -- {dasys1,ncoast,well,sun,ihnp4}!hoptoad!gnu gnu@postgres.berkeley.edu Alt.all: the alternative radio of the Usenet.
hedrick@TOPAZ.RUTGERS.EDU (Charles Hedrick) (07/22/87)
With Universities at least, ownership of code funded by the government depends upon the provisions of the grant. If the grant was a contract that specified production of software as a deliverable, then it is work for hire, and the code is owned by the government, or required to be PD, or something like that. However if the grant called for research, and the code was in effect a sideeffect of the research, then all agencies that I know interpret that the code was not the thing that they were contracting for, and the University owns it. Few universities these days are willing to admit that any code the produce was the intended product of their work, so although the government pays for much of the code produced by our universities, the tendency is for very little of it to be regarded as covered by the provisions requiring government-funded code to be public. Some agencies consider this a good thing, because they'd rather see the universities have enough ownership of the code to be able to sell it to a software house for futher development and eventual marketing. This is viewed as "technology transfer", and gets them brownie points. Agencies apparently do not view PD software as accomplishing much. I disapprove of this trend, and view all of my work as PD, but I seem to be fighting a losing battle.