[comp.lang.c] GNU Manifest

jdc@naucse.UUCP (John Campbell) (03/30/88)

Recently a few of us here had the following discussion.  Since
the other two disagreed with me regarding the GNU manifest for BISON,
I thought I'd better broaded the discussion.
>
>I'm interested in opinions on interpretations of GNU's licensing
>agreements:
>
>       (1) My understanding of BISON is that if you use it, then
>           the thing you created with it must be freely distributed.
>           Agreed?
>
>       (2) What about gcc?  My understanding is that if you port
>           gcc to a new machine, it must be freely distributed.
>           Agreed?
>
>       (3) Now, what about programs compiled with gcc?  (I.e. my
>           own code, compiled to binary using gcc.  Must the
>           binaries be freely distributed?  What about the original
>           source?
>Opinions?
>

;I assume you are, in (1), referring to the following passage from the
;BISON agreement:

:  2. You may modify your copy or copies of BISON or any portion of it,
:and copy and distribute such modifications under the terms of
:Paragraph 1 above, provided that you also do the following:
:
:    a) cause the modified files to carry prominent notices stating
:    that you changed the files and the date of any change; and
:
:    b) cause the whole of any work that you distribute or publish,
:    that in whole or in part contains or is a derivative of BISON
:    or any part thereof, to be freely distributed
:    and licensed to all third parties on terms identical to those
:    contained in this License Agreement (except that you may choose
:    to grant more extensive warranty protection to third parties,
:    at your option).

;As I read this agreement, it seems to me that part b) referes only to
;works that have BISON embedded in them (contains or is a derivative), 
;not the output of BISON.  I certainly made this assumption when I used 
;BISON to create two small compilers--if I am wrong I'm in trouble and 
;don't even know it.
;
;The clause seems to me to be a "kermit" clause.  In other words, some
;products are easily embedded into commercial products.  Thus if you build 
;a neat language building tool that is essentially YACC with a screen 
;interface, and you earn oodles of money from it, but it just puts BISON in 
;a new wrapper, you're obviously in trouble.  If you use the BISON tool to
;make a new commercial product, unrelated to BISON, you should be fine.
;
;My view leads to inevitable conclusions for questions 2) and 3):  
;2) Yes, gcc and any derivative must be supplied freely.  3) output
;from gcc is no longer the "property" of GNU.  If I am wrong I think there
;are probably a number of programs out there that GNU now "owns" and doesn't
;know about.  To carry the argument a bit further, if a defense contractor
;has GNU emacs on their machine and a top-security yo-yo writes a proposal
;regarding super-secret-who-knows-what, does this mean (since the proposal
;is the output of GNU EMACS) that GNU can now own and have rights to the
;proposal?  [If this were possible we would all be writing editors!]
;

As I said, others disagree with my interpretation.  Could someone tell me
what interpretation is correct.  (Would anyone from Free Software 
Foundation, Inc.  care to respond?)
-- 
	John Campbell               ...!arizona!naucse!jdc

	unix?  Sure send me a dozen, all different colors.

rk9005@cca.ucsf.edu (Roland McGrath) (04/01/88)

C programs generated by Bison are at least in part derivitive works.
Most of the output of a Bison or Yacc parser is the parser skeleton,
which is copyrighted material belonging to FSF (or AT&T in the case of Yacc).

I don't know what all the legalities are, but I know it is not the
intention of FSF to make all binaries made by GNU be FSF property.
(Well, RMS would like everything everywhere to be free, but he's
not into forcing this on people.)

-- 
	Roland McGrath
ARPA: roland@rtsg.lbl.gov roland@lbl-rtsg.arpa
UUCP: ...!ucbvax!lbl-rtsg.arpa!roland

jesup@pawl21.pawl.rpi.edu (Randell E. Jesup) (04/02/88)

In article <1214@ucsfcca.ucsf.edu> roland@rtsg.lbl.gov (Roland McGrath) writes:
>I don't know what all the legalities are, but I know it is not the
>intention of FSF to make all binaries made by GNU be FSF property.
>(Well, RMS would like everything everywhere to be free, but he's
>not into forcing this on people.)

	When I questioned him about it, the response was that if you want to
use Bison, you MUST make everthing you make with it, AND everything else
you distribute with something you make with Bison, freely redistributable.

	I see an opening for someone to write a PD version of that table-
handling code to replace the canned code in Bison.

     //	Randell Jesup			      Lunge Software Development
    //	Dedicated Amiga Programmer            13 Frear Ave, Troy, NY 12180
 \\//	beowulf!lunge!jesup@steinmetz.UUCP    (518) 272-2942
  \/    (uunet!steinmetz!beowulf!lunge!jesup) BIX: rjesup

(-: The Few, The Proud, The Architects of the RPM40 40MIPS CMOS Micro :-)