bogstad@HOPKINS-EECS-BRAVO.ARPA ("William J. Bogstad") (05/20/86)
Subject: Re: distributing gnu - let's think about this References: <5373@alice> In <2633@teddy.uucp> John Nelson says: >I recently obtained a copy of BISON (the GNU yacc equivalent). It appears >that every C file genrated by BISON contains the GNU copyright. I contacted >Richard Stallman to make sure this was not accidental - his reply was that >this copyright was included in the generated C source file because it includes >a copy of the parser written by him. To the best of my understanding, this >implies that I cannot sell any program that uses BISON to generate a parser >- In fact, I must be willing to give away that C source. > >Richard didn't seem very upset when I suggested that I couldn't use >BISON if I had to give away the generated C program. In fact, He was >quite pleased (something like: He was just as pleased as if a gun he'd >manufactured couldn't be used for murder). I know for a fact that Richard didn't write bison from scratch. Someone out at Berkeley did the original work and RMS did a lot to clean it up and make it more equivalent to yacc. I don't know what the original author thinks about this, but I find it a little disturbing. Don't take me wrong, I like what RMS is doing and in fact contributed a small hack to FSF. I just think that he is taking this a little too far. Why not have GNU Emacs insert a copyright into everything it produces? Or at least your C code since you are using the auto formatting of c-mode aren't you? BTW, when I sent RMS my hack to replace "colrm"; I got a message back asking me to print and sign the following form: ---- Richard Stallman 545 Tech Sq, rm 703 Cambridge, MA 02139 ASSIGNMENT For good and valuable consideration, receipt of which I acknowledge, I hereby transfer to the Free Software Foundation, Inc. (the "Foundation") my entire right, title, and interest (including all rights under copyright) in my computer program "colrm.c" (the "Program"). However, upon thirty days' prior written notice, the Foundation agrees to grant me non-exclusive rights to use the program as I see fit; (and the Foundation shall also own similar non-exclusive rights). I hereby indemnify and hold harmless the Foundation, its officers, employees, and agents against any and all claims, actions or damages (including attorney's reasonable fees) asserted by or paid to any party relating to the program. Agreed: Date signed: ---- Any comments on this? I was a little taken aback by it, but eventually decided that it was reasonable. I wonder if the only reason he did this was because I had included a copyright notice in my code which allowed me to do anything with my code and others could only give it away. Would this allow FSF to sell my code? (Sorry for the length on this.) Bill Bogstad bogstad@hopkins-eecs-bravoa.arpa
phr@ernie.Berkeley.EDU (Paul Rubin) (05/21/86)
Subject: Re: distributing gnu - let's think about this References: <5373@alice> In <2633@teddy.uucp> John Nelson says: >I recently obtained a copy of BISON (the GNU yacc equivalent). It appears >that every C file genrated by BISON contains the GNU copyright. I contacted >Richard Stallman to make sure this was not accidental - his reply was that >this copyright was included in the generated C source file because it includes >a copy of the parser written by him. To the best of my understanding, this >implies that I cannot sell any program that uses BISON to generate a parser >- In fact, I must be willing to give away that C source. > >Richard didn't seem very upset when I suggested that I couldn't use >BISON if I had to give away the generated C program. In fact, He was >quite pleased (something like: He was just as pleased as if a gun he'd >manufactured couldn't be used for murder). If you generate a C program with GNU Bison that includes RMS's parser, you have the option of distributing it, under certain terms designed to encourage maximal cooperation with other programmers. If you use Unix Yacc, with the parser written by AT&T, you could not distribute the resulting C program AT ALL. Why are you complaining? By the way, it is not true that the GNU copyright requires you to distribute (for free or or any other way) any programs affected by it. It says that *if* you choose to distribute them, *then* you must distribute them under the same terms as the rest of GNU (i.e., in source form or with source code available, and with no restrictions on further redistribution). This does not prevent you from doing anything that you could do under a conventional license, which would not allow redistribution at all. GNU aims to provide an alternative to proprietary software, that everyone can use, always. Allowing GNU code to be used inside proprietary products would be self-defeating.
jpn@teddy.UUCP (John P. Nelson) (05/23/86)
>If you generate a C program with GNU Bison that includes RMS's parser, >you have the option of distributing it, under certain terms designed >to encourage maximal cooperation with other programmers. If you use >Unix Yacc, with the parser written by AT&T, you could not distribute >the resulting C program AT ALL. Why are you complaining? I'm not complaining. I just said that I couldn't use bison because of it's restrictions. By the way, I had heard that AT&T had relinquished the rights to programs generated by yacc. Does anyone have the facts? In any case, I just obtained a copy of yacc from the "Austin Code Works" (you can find their ads in Dr. Dobbs, or Computer Language). For ~$20 with source code and documentation. This looks an awful lot like the UNIX yacc, but it has "DECUS" sprinkled throughout the comments. They make no claim on the generated source code. Bottom line, I would much rather spend 20 dollars for a program that I am free to use as I wish, than for a "free" program that I am restricted to use on projects that are "public-spirited".
tihor@acf4.UUCP (Stephen Tihor) (05/23/86)
The AT&T parser has been (alegedly) placed into the PD to eliminate the yacc /lex problem. If true this places GNU in a "worse" position.
tp@ndm20 (05/25/86)
>>I recently obtained a copy of BISON (the GNU yacc equivalent). It appears >>that every C file genrated by BISON contains the GNU copyright. I contacted >to encourage maximal cooperation with other programmers. If you use >Unix Yacc, with the parser written by AT&T, you could not distribute >the resulting C program AT ALL. Why are you complaining? Any other opinions on this? Last I looked, yacc did not insert copyright notices in its output. There is no restriction on use of yacc output (I didn't sign anything to that effect, and I have yacc). AT&T seems not to have chosen to restric use of their parser. Also, I am not a yacc user, but I thought the parser was written in a semi-bnf form by the yacc user, and yacc basically 'compiled' it to C code. If this is true, copyrighting yacc output is like copyrighting a compiler's output. Terry Poot Nathan D. Maier Consulting Engineers (214)739-4741 Usenet: {seismo!c1east | cbosgd!sun | ihnp4}!convex!infoswx!ndm20!tp CSNET: ndm20!tp@smu ARPA: ndm20!tp%smu@csnet-relay.ARPA
oz@yetti.UUCP (Ozan Yigit) (05/26/86)
In article <2657@teddy.UUCP> jpn@teddy.UUCP (John P. Nelson) writes: > >In any case, I just obtained a copy of yacc from the "Austin Code Works" >(you can find their ads in Dr. Dobbs, or Computer Language). For ~$20 >with source code and documentation. This looks an awful lot like the >UNIX yacc, but it has "DECUS" sprinkled throughout the comments. They >make no claim on the generated source code. > I said this on the net before; The program you have just obtained is *the* un*x yacc, that was broken up for DECUS C compiler, and was submitted to a DECUS tape. It is, strictly speaking, illegal. So, you would rather use that one eh ?? Good luck !!. oZ -- The best way to have a Usenet: [decvax|ihnp4]!utzoo!yetti!oz good idea is to have a Bitnet: oz@[yusol|yuyetti].BITNET lot of ideas. Phonet: [416] 736-5053 x 3976