Article View: gnu.misc.discuss
Article #72727Re: Too Good to be True?
From: Conway Yee
Date: Wed, 15 Jan 1992 21:40
Date: Wed, 15 Jan 1992 21:40
45 lines
3241 bytes
3241 bytes
>>Let us take the possibility that you have a source license to a product >>(i.e. ATT UNIX). You are free to make improvements. Your code can > ^^^^^^^^^^^^^ >>be under any license you choose. > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > >If I modify GNU code, and decide to give it to someone, my modifications >_can't_ be under any license I choose; they must be under the GNU license. Not entirely true. Ghostscript is being distributed both under GPL as well as a standard commercial license. This, however, is entirely beside the point. >>Remember, however, that in order to >>be useful to anyone, a person using your code must have access to >>UNIX source (if he doesn't, then you didn't need the source either. An >>example is someone who writes a spreadsheet). Thus, a prerequisite for >>your licensee is a UNIX source license. Thus, effectively ATT propagates >>their license. Is this any different from GPL? > >Yes. The AT&T "propagation" only spreads to AT&T code. If I had code of my >own, added it to AT&T source, and gave it to someone else who also had a source >license, that person could cut out my added code and use it for whatever he >wishes without propagating the AT&T license. If I did the same to GNU source, >the license would be propagated. Yes but AT&T's is effectively being propagated. If I wrote an improvement to the UNIX filesystem or memory management and wanted to distribute it out, it would only be useful to those with UNIX source licenses. A person who wanted to use my improvements (and didn't have a UNIX source license) would have to obtain one before being able to use my improvements. In effect, then, AT&T is propagating their license. My improvements are meaningless without access to UNIX source. Similarly, if I make an improvement to GCC, this improvement is meaningless without access to GCC source. Following in Ghostscript's footsteps, I may release my code under any license I wish but without access to GCC source, these improvements are meaningless. While AT&T propagates their license unofficially (de facto), GPL makes it part of their license (de jure). Effectively, then, there is little difference. -- Blockley Hall, 4th floor| Conway Yee, N2JWQ 418 Service Drive |yee@ming.mipg.upenn.edu (preferred) Philadelphia, PA 19104 |cy5@cunixa.cc.columbia.edu (forwarded to above) (215) 662-6780 |yee@bnlx26.nsls.bnl.gov (rarely checked)
Technical Details
Message-ID:
<62435@netnews.upenn.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!ub!dsinc!netnews.upenn.edu!mipgsun!yeeReferences:
<NELSON.92Jan15003606@cheetah.clarkson.edu> <62381@netnews.upenn.edu> <1992Jan15.210242.3076@blaze.cs.jhu.edu>Show Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 1025b4,e71af1b1f401c58d X-Google-Attributes: gid1025b4,public X-Google-ArrivalTime: 1992-01-15 15:07:01 PST Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!ub!dsinc!netnews.upenn.edu!mipgsun!yee From: yee@mipgsun.mipg.upenn.edu (Conway Yee) Newsgroups: gnu.misc.discuss Subject: Re: Too Good to be True? Message-ID: <62435@netnews.upenn.edu> Date: 15 Jan 92 21:40:19 GMT References: <NELSON.92Jan15003606@cheetah.clarkson.edu> <62381@netnews.upenn.edu> <1992Jan15.210242.3076@blaze.cs.jhu.edu> Sender: news@netnews.upenn.edu Organization: University of Pennsylvania Lines: 45 Nntp-Posting-Host: mipgsun.mipg.upenn.edu