8 Jan
2013
8 Jan
'13
10:10 p.m.
On Tue, Jan 8, 2013 at 3:20 PM, <Paul_Koning at dell.com> wrote:
What DEC and others did is to provide listings you could read, but you weren't allowed
to use or modify the code. Sources, maybe, but even then I don't think you could
hand the resulting executable bits to anyone outside your organization.
I think you are split hairs a little. We should talk this off line if you want to
discuss it much more.
I'm not a lawyer and do not pretend to be one. I did live a lot of this including
the AT&T/UCB case and today I help teach a course at Intel on Copyright and IP
protection that is required for all our SW developers - so comfortable in saying I think I
understand most of the subtleties.
I will point out that the Amdahl machines ran OS/360, TSS, VM etc..
Foonly and Xerox's MACie ran Tenex and the stock DEC compile suite. The Xerox
Alto's were a Nova clone and lot of the code that ran in them was just that (although
Xerox would be know for the boat load of SW they would write for them),
And as Ken O'Mundro used to remind everyone at USENIX conferences, it was not the
instruction set and SW that got him trouble with KO - the CalData machine when it Nova
microcode in it, could run DG's SW. KO did not like that Ken made a UNIBUS
replacement.
Simple put, in the late 1960s and early 1970s, different firms make sold
"clones" of different machines that used the original developers code base.
The "cloners" made the market bigger and sometimes (like Foonly) could serve a
need the primary manufacturer was not going to supply.
You are mentioning the part of "copy right" and basically, anyone with a common
license allowed you to shared code. This was what we in UNIX land did extensively. So
folks like me used to keep a pile of "signature pages" for different people
licenses in our filing cabinet before we sent them our modified versions tapes.
The rules about copyright and SW were really unclear in the 1960s and 70s, and it would
actually not get decided with real case law until the early-mid 1980s with the
Apple/Franklin Computer Case (Apple had copyvwritten it's ROMs and Franklin
"cloned" them). But using SW IP if the copyright license did not tell you
that you could not, was consider ok for a long time. In fact that was the basis of the
Franklin argument - to be a compatible it needed to use Apple's ROMs.
The original IBM user group was called "SHARE" and they did just that, as did
DECUS folks. We all traded patches because we had the sources. My first job at CMU
was looking at IBM source patches and figuring out how to make them work or if they
mattered any more to our custom TSS system. We also got patches from other users and we
pushed our changes out. IIRC, there were more Amdahl machines running TSS than IBM HW -
but TSS was an IB product and ran IBM "layered products."
FYI: >>Free<< and Open Source which is what Linux, FreeBSD et al are based on
today. They keep point is Free in some manner - typically with a license that allows a
more free use of the IP.
BTW: I've always said the real father of the OSS movement, was the late Prof Don
Pederson of UCB. His many generations of students would "publish" code. He
started that practice in the late 1960s and it would become the guts of UCB
"Industrial Liaisons Office" - and even later CSRG etc. BSD - Berkeley
Software Distribution. The predecessor to the UNIX BSD tape, was the EE "Tools
Tape" (SPICE, SPLICE, MOTIS et al).
As "dop" would say - "I always give away our source, that way I come in the
back door. Other places [ such as CMU and MIT would license code sources and as he
mentioned ] came in the front door like any other salesman."
Clem