Thoughts on dual licensing OpenSolaris
I think this all comes down to three key questions and 2 of which center around the kernel because, if its userland we don't need to dual license anything. This statement asks a few questions for us (opensolairs developers and mostly kernel developers to answer, and if you aren't a kernel developer or a future kernel developer or you don't even know C you most likely dont have the knowledge needed to answer the questions) and points out the feelings of the other side as stated publicly, the feelings that the other has side may not be the true or right and may be a major mistake, but its not our job to convince them they are wrong and we are right, and changing our license and vision is not going to change there opinion.
#1 is there anything that is GPLv3'd that we need, and what is it worth?
The only things we want out of Linux is some header files that have some magic number and implementation details for drivers. A few thousand dollars donated to the right team and I'm pretty sure this can be reverse engineered again.
#2 is there anything that that a GPLv3 kernel wants that we have?
This is the only people that dual licensing helps. Linux can't benefit because its a GPLv2 OS. Linus the leader of Linux has come out and said we don't want anything that is in Solaris. A number of core
Linux developers have said we can't use ZFS because of the way its implemented. Even more Linux developers have decided that they are doing a better DTrace in the form of Systemtap, so they wont want DTrace. Zones and SMF is more of a distribution type problem, so is too big of a headache for Linux to want anyway so will be shrugged off as useless.
#3. Is this all just marketing ploy to hopefully get 10,000 developers to sign on to OpenSolaris and for wall street to say "YEAH." Wall street doesn't understand the licensing mess so doesn't have a clue that this is going to cause short term chaos of a licensing war, and long term effect of extra baggage on OpenSolaris.
Linux community is going to see OpenSolaris as trying to be GPLv3 as a marketing ploy which it really is, they will also see this as weekness since we had to added there license whether or not its useful to them or not or us for that matter. So as you can pretty much sense by this I think that the licensing of OpenSolaris should not change, and that if it changes it will only hurt the OpenSolaris and progress. Perhaps my opinion will change if the top 10 Linux kernel coders come out and ask for OpenSolaris to be GPL'd they intend to use it and we want to merge Linux with it and are willing to charge our license to work with you. As of now Linux kernel coders don't care and are locked into there current license path that is different from ours.
#1 is there anything that is GPLv3'd that we need, and what is it worth?
The only things we want out of Linux is some header files that have some magic number and implementation details for drivers. A few thousand dollars donated to the right team and I'm pretty sure this can be reverse engineered again.
#2 is there anything that that a GPLv3 kernel wants that we have?
This is the only people that dual licensing helps. Linux can't benefit because its a GPLv2 OS. Linus the leader of Linux has come out and said we don't want anything that is in Solaris. A number of core
Linux developers have said we can't use ZFS because of the way its implemented. Even more Linux developers have decided that they are doing a better DTrace in the form of Systemtap, so they wont want DTrace. Zones and SMF is more of a distribution type problem, so is too big of a headache for Linux to want anyway so will be shrugged off as useless.
#3. Is this all just marketing ploy to hopefully get 10,000 developers to sign on to OpenSolaris and for wall street to say "YEAH." Wall street doesn't understand the licensing mess so doesn't have a clue that this is going to cause short term chaos of a licensing war, and long term effect of extra baggage on OpenSolaris.
Linux community is going to see OpenSolaris as trying to be GPLv3 as a marketing ploy which it really is, they will also see this as weekness since we had to added there license whether or not its useful to them or not or us for that matter. So as you can pretty much sense by this I think that the licensing of OpenSolaris should not change, and that if it changes it will only hurt the OpenSolaris and progress. Perhaps my opinion will change if the top 10 Linux kernel coders come out and ask for OpenSolaris to be GPL'd they intend to use it and we want to merge Linux with it and are willing to charge our license to work with you. As of now Linux kernel coders don't care and are locked into there current license path that is different from ours.
posted by jamesd_wi at Saturday, February 03, 2007
2 Comments:
I think the greatest value to sun is the opensource CIO that love linux that might think less of solaris right now because of its sun centric license. That CIO giving solaris as much of a chance as he would linux, maybe even more so if the microsoft / novell deal gets off the ground.
Ubiquity of solaris would be a boon no doubt, it would open alot of new business opportunity and take few away.
I would consider only one aspect: I read some post that speak about GPL as an occasion to open new opportunities... .Well,Which license can open new opportunities if it is compatible only with itself?
My definition of Open and Free is a ecosystem where every community can share code and ideas with reciprocity and respect.This is not exactly the spirit of GPL.
Post a Comment
<< Home