| 1 |
"Boyd Stephen Smith Jr." <bss03@××××××××××.net> posted |
| 2 |
200705280542.32246.bss03@××××××××××.net, excerpted below, on Mon, 28 May |
| 3 |
2007 05:42:32 -0500: |
| 4 |
|
| 5 |
>> So, Duncan's idea of dual-licensing the kernel under GPL v2/v3 until |
| 6 |
>> all bits of kernel code written by non-agreeing parties are removed |
| 7 |
>> would not work. |
| 8 |
> |
| 9 |
> Right, but Linus could take the stance that all new submission have to |
| 10 |
> be licenced (to him) under GPLv2/GPLv3 and then one there's no |
| 11 |
> GPLv2-only code left, release a kernel under GPLv3. Until the point |
| 12 |
> where there's no GPLv2 only code, Linus could only release the whole |
| 13 |
> work under GPLv2, but everyone would know the transition was in the |
| 14 |
> works. |
| 15 |
|
| 16 |
That's basically what I had in mind, yes. Only I (inadvertently) took a |
| 17 |
logical shortcut while explaining it, and Richard obviously didn't follow |
| 18 |
me when I veered off. =8^\ That's the whole reason it couldn't simply |
| 19 |
be made GPLv3 in an instant, however. New code (and presumably the old |
| 20 |
code from all currently active contributors who agreed and continued to |
| 21 |
contribute) would instantly be dual-licensed, so it could use the old |
| 22 |
GPLv2 only code with the GPLv2 license, while the GPLv3 license remained |
| 23 |
in place but basically unused until all the GPLv2 only code was removed, |
| 24 |
at which point the GPLv2 dual license could be dropped off the new code |
| 25 |
as well (tho anyone wanting it could get it under GPLv2 could get it |
| 26 |
under that last dual licensed snapshot, until further new GPLv3 only code |
| 27 |
was added, anyway). Finally, the kernel was GPLv2 and later for quite |
| 28 |
some time, the GPLv2 only being a relatively new development as well. |
| 29 |
Thus, anything in the last GPL2 or later snapshot could immediately be |
| 30 |
upgraded to GPLv2/v3 as well. All told, within a few months, the large |
| 31 |
majority could be dual-licensed. It's just that last 20, 10, 5, 2, 1%, |
| 32 |
as an increasing focus is placed on the issue, that's the sticking point, |
| 33 |
and that could ultimately cause the switchover to take five years or so. |
| 34 |
|
| 35 |
Still, it's doable, just over some significant time and with some |
| 36 |
substantial effort required. |
| 37 |
|
| 38 |
Thanks for pointing that out, Richard, and Boyd for clarifying what I |
| 39 |
meant. |
| 40 |
|
| 41 |
-- |
| 42 |
Duncan - List replies preferred. No HTML msgs. |
| 43 |
"Every nonfree program has a lord, a master -- |
| 44 |
and if you use the program, he is your master." Richard Stallman |
| 45 |
|
| 46 |
-- |
| 47 |
gentoo-amd64@g.o mailing list |
Archives