Mark Crispin <MRC@SU-SCORE.ARPA> (12/26/84)
I have been told the following things about the Unix software world.
I would appreciate some commentary about how [un]true these statements
are:
. the object code from a proprietary C compiler is itself proprietary.
. the executable binary of a C program is proprietary if it uses any of
the standard library functions, which are proprietary.
. any program written in yacc is proprietary, because the algorithms
output by yacc are proprietary.
. the proprietary ownership of all of this is good ol' Bell.
In other words, what this seems to say is that I am unable to
develop any applications in C or yacc and distribute binaries produced by
a Bell C/yacc compiler without requiring that the site I distribute the
software to has a Unix license.
I have trouble adjusting to this, coming from a world (DEC-20) where
no matter how proprietary the compiler may be, the ownership of the
executable binaries belongs to the owner of the source code even if some
library routine from the compiler's runtimes is used.
Perhaps GNU may be a salvation, however the GNU developers may not
be totally sympathetic to my efforts, as part of my plans include the
development of proprietary software -- MY proprietary software. I don't
want to be limited to selling only to Unix systems though.
Have any of the legal eagles on this list worked out all the hairy
issues involved? Does Bell really want to strangle the greater acceptance
of C by making Unix licenses essential to run programs written in it?
-------gwyn@brl-tgr.ARPA (12/26/84)
> . the object code from a proprietary C compiler is itself proprietary. > . the executable binary of a C program is proprietary if it uses any of > the standard library functions, which are proprietary. > . any program written in yacc is proprietary, because the algorithms > output by yacc are proprietary. > . the proprietary ownership of all of this is good ol' Bell. All false, with the possible exception of yacc's parser which I have never heard declared non-proprietary by AT&T. They have specifically stated that C library binaries linked in with an application have a waiver so that one may distribute them without requiring an AT&T license. The code generated from your own sources by a compiler cannot by any stretch of the legal imagination be considered property of the compiler vendor.
chris@umcp-cs.UUCP (Chris Torek) (12/27/84)
> . any program written in yacc is proprietary, because the algorithms > output by yacc are proprietary. I have heard the claim (and it makes sense to me) that SOURCE code put out by yacc is proprietary, since it contains /usr/lib/yaccpar. I've never heard it claimed (but wouldn't be TOO surprised at ANYTHING AT&T does) that object code produced with the aid of yacc is proprietary. -- (This line accidently left nonblank.) In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (301) 454-7690 UUCP: {seismo,allegra,brl-bmd}!umcp-cs!chris CSNet: chris@umcp-cs ARPA: chris@maryland
jbn@wdl1.UUCP (01/16/85)
You can reach a long way with the right licensing agreement. Howard Hughes, Sr (the father of the one who died recently) came up with the ultimate to date in this direction; he acquired the rights to the only really good drill bit for drilling oil wells, and leased out drill bits refusing to sell them outright. The lease called for a charge dependent on the production of the oil well, and it was a non-trivial percentage. Hughes Tool was built on that basis.