- From: Dan Kegel <dank@kegel.com>
- Date: Fri, 29 Nov 2002 11:06:32 -0800
- To: Federico Heinz <fheinz@vialibre.org.ar>
- CC: www-patentpolicy-comment@w3.org
Federico Heinz wrote: >> [ http://www.gnu.org/philosophy/w3c-patent.html >> didn't have enough detail to explain exactly >> *why* GPL and the draft policy clash.] > > "As an example, W3 members may contribute patent claims to a standard > describing the behavior of web servers providing particular > functionality. A Free Software program implementing that standard would > be available for others to copy from, in order to add functionality to > browsers, or non-interactive web clients. But if, as the present > proposed policy permits, the patent-holder has licensed the practicing > of its patent claims "royalty-free" only "in order to implement the > standard", reuse of the relevant code in these latter environments would > still raise possible patent infringement problems." I think that was there originally. What I was looking for that was missing was a hand-held tour through the GPL showing why that caused total incompatibility with the GPL. >>>I wonder, too. I worry whether that would leave the door open for >>>trouble with non-GPL copyleft licenses, though (the Perl license, for >>>example, which uses the terms of the GPL as an alternative to the >>>Artistic License, would be reduced effectively to the Artistic License). >> >>Most non-GPL licenses would probably be ok, as they don't try >>to offer the same airtight promise of freedom as the GPL does. > > > Sure, but it would create problems of license compatibility. Assume for > a moment that my interpretation of the GPL is correct. Now further > assume that Larry includes code in Perl "to implement the standard" and > that the code is covered by a royalty-free, field-of-use-restricted > patent. All of a sudden, the GPL no longer is applicable to Perl, and > thus its license becomes exclusively the Artistic License, which I > understand is incompatible with the GPL, so Perl code can no longer be > used in GPL products, which is *bad*. Got it. Yes, any tool that is dual-licensed GPL and something else would suddenly no longer be GPL'd if it include code that implemented a patented technique for which an unconditional license has not been granted (assuming you're right). The GPL is a brand, essentially, which stands for "freedom to use, *modify*, and share", and it looks like the author did a good job in making sure anything released under the GPL actually delivers those freedoms. - Dan
Received on Friday, 29 November 2002 14:06:14 UTC