> JFTR, I do *not* think that that assessment was questionable. Unless theNobody is relicensing anything, ever. If the author licenses a work under the GPL only, then that is forever how that work is licensed. If an author licenses a work under the BSD, then that is forever how that work is licensed. Same for a dual license. This applies until the copyright expires or the author offers the code under some other license. Nobody ever relicenses anything, ever. If I give you a copy of a work covered by the GPL, the BSD, a dual-license, or whatever, you get a license to every protectable element in that work from the original author of that element. Nobody ever relicenses anything, ever. Nobody ever modifies anybody else's license, ever. If you take work that's under a dual-license and remove one license notice from it when you create a derivative work, every recipient of that derivative work still receives a dual license from the original author to every protectable element still in the distributed work. The GPL is explicit about this in section 6. The BSD license is not, but it's the only way such a license could work. There are really only two ways you can screw up. 1) You can take GPL-only bits and put them in BSD or dual-licensed code. (The GPL prohibits this.) 2) You can remove a BSD license notice from BSD-only code. (The BSD license prohibits this.) DS -
| Andrew Morton | -mm merge plans for 2.6.23 |
| Rafael J. Wysocki | [Bug #11207] VolanoMark regression with 2.6.27-rc1 |
| Zhang, Yanmin | AIM7 40% regression with 2.6.26-rc1 |
| Con Kolivas | [PATCH][RSDL-mm 0/7] RSDL cpu scheduler for 2.6.21-rc3-mm2 |
git: | |
| Gregory Haskins | [RFC PATCH 03/17] vbus: add connection-client helper infrastructure |
| David Woodhouse | [PATCH 03/30] solos: FPGA and firmware update support. |
| Natalie Protasevich | [BUG] New Kernel Bugs |
| Gerrit Renker | [PATCH 15/37] dccp: Set per-connection CCIDs via socket options |
