BSDnewsletter.com

I have been researching the history of the BSD licenses for over a decade including in participating in a wide audit around 2000. Later, I found over 100 slightly-differing BSD licenses (and I am sure there are many more) which I include in the printed volume one of the NetBSD System Manager's Manual set. I am also thoroughly documenting the proprietary and open source licensing (and agreements) for the historical BSDs (and their Bell Telephone Laboratories lineage) in my book in progress about the History of Berkeley Unix.

As part of this work, I noticed other similar historical licenses. In this article I share a few examples of the earliest licenses through 1982 that can be considered (or close to) "open source." In part 2 of this article, I will share further examples later in the 1980's.

By the early 1980's there were several softwares available for free or as public domain, such as the original Emacs and TeX (but prohibited changes if using the same name), and including software shipped by Usenix or the /usr/group catalog. Also I understand that the MIT-MC collection in 1982 had software that had notices about free distribution for private use, not for sale, get revisions back to the author or some common redistribution point, etc. But I can't find these examples, especially with redistribution statements. I am looking for old software that specifically has a notice that identifies it may be copied or redistributed, such as "open source" software.

The Open Source Definition provides some guidelines for a valid open source license. In summary, the license:

• must allow selling or giving away the software,
• must allow it to be included in an aggregate collection,
• can't require fees or royalties,
• must include source code,
• must allow source to be distributed,
• must allow compiled form to be distributed,
• source must not be deliberately obfuscated,
• must allow modifications and derived works that can be distributed under same license,
• must allow distributing patches for modifying the build of the program,
• must allow distributing modified software,
• must not discriminate against any people,
• program doesn't have to be distributed with other software,
• must not place restrictions on other software,
• and it must not depend on a specific technology or style of interface.

The term "license" commonly means placing more restrictions on the user's use of the software than copyright law regulates, but in these examples the purpose of the license is to encourage its use and reuse. (Note that the term "open source" was not used at these following years.)

The earliest example I found is for 1976 code from Peter Langston, such as his banner, fold, news, and gomoku. He included a copyright line and a simple statement in his source code:

Permission to copy, either in whole or in part, is hereby granted to anyone wishing to do so with the restriction that this entire notice must accompany any copies.

Langston told me in 2010 that he didn't know of any earlier licenses like his. "The idea for my license came when several realizations collided in my head: (1) People want my games software and I want them to have it and enjoy it. (2) I do not want to spend my life maintaining it. (3) If I were to sell it I would have to maintain it. (4) Even if I can't get money from selling it, I do want recognition for the work. So this license was the result."

Another early example is the 1976 Trek from Eric Allman, who was an INGRES developer and later known for his syslog and Sendmail. His Star Trek game was based on others' Fortran and BASIC programs and was shipped in the early Berkeley Unix Software Tapes. (It is interesting to note that the other early BSD software didn't include usage or distribution notices — that is covered in detail in my book.)

His code comments stated:

If you make ANY changes in the game, I sure would like to know about them. It is sort of an ongoing project for me, and I very much want to put in any bug fixes and improvements that you might come up with.
...
Permission is given for use, copying, and modification of all or part of this program and related documentation, provided that all reference to the authors are maintained.

This didn't require changes given back to him and didn't have any restrictions (other than acknowledging the authors).

An example of restrictive licensing but allowing some "open source" like use is the Pascal compiler system from Johan Stevenson and Andrew S. Tanenbaum (later known for his MINIX) at the Vrije Universiteit Amsterdam (VU University). It included a copyright line and the statement:

Explicit permission is hereby granted to universities to use or duplicate this program for educational or research purposes. All other use or duplication by universities, and all use or duplication by other organizations is expressly prohibited unless written permission has been obtained from the Vrije Universiteit.

Another restrictive license from around 1980 was used for the Tektronix Unix Distribution from Steve Glaser and Clement T. Cole of the Computer Research Group at Tektronix Laboratories. It included the statement:

... (obviously) provides no support or warranty of any kind for this code. Permission is granted to use this code for in-house applications, research and development only. Permission is further granted to redistribute this code freely within the Unix community provided this notice remains with the code (subject to the applicable License Agreements with Bell).

The license agreements from Western Electric (for Bell) were very restrictive. The original licensing (around 1973/1974) allowed use solely for academic and educational purposes. In addition, the licensee was prohibited from sharing the software — including its methods or concepts — to anyone other than the their own employees or students. But Western Electric did provide specific permission a few years later to the Computer Science Division at the Univ. of California at Berkeley so they could distribute their software (such as BSD) which included some of the proprietary Unix system to other licensees of the same Unix. Verification and reporting of these licensees would have to be done though. (This story is told throughout my history book.)

Another "open source" example is Eric Cooper's gcore program from fall 1981:

Permission to copy or modify this program in whole or in part is hereby granted, provided that the above credits are preserved.

And the Franz Lisp documentation from 1982 (copyrighted in 1980/1981) stated:

Permission to copy without fee all or part of this material is granted provided that the copies are not made or distributed for direct commercial advantage, and the copyright notice of the Regents, University of California, is given. All rights reserved.

Finally another example of open documentation (but also commercially restrictive) is from the Association for Computing Machinery (1974):

General permission to republish, but not for profit, all or part of this material is granted provided that ACM's copyright notice is given and that reference is made to the publication, to its date of issue, and to the fact that reprinting privileges were granted by permission of the Association for Computing Machinery.

Permission to copy without fee all or part of this material is granted provided that the copies are not made or distributed for direct commercial advantage, the ACM copyright notice and the title of the publication and its date appear, and notice is given that copying is by permission of the Association for Computing Machinery. To copy otherwise, or to republish, requires a fee and / or specific permission.

What are the earliest documented open source licenses you know about that predate MIT's X, DEC's drivers, Kermit, and GNU Emacs (and hopefully examples still exist)?

I plan to write a follow-up article that mentions briefly the start of the open source licensing from DEC, Columbia (Kermit), MIT (including X), and FSF (Emacs) — all predating BSD, which is covered in detail in my upcoming book.

Discussion

BSD Links