| 1 |
On Sun, Sep 25, 2011 at 4:21 PM, Canek Peláez Valdés <caneko@×××××.com> wrote: |
| 2 |
> On Sun, Sep 25, 2011 at 3:54 PM, Stroller |
| 3 |
> <stroller@××××××××××××××××××.uk> wrote: |
| 4 |
>> The end users do not give a monkey's uncle about the CLA. They just want to use the software, and our distro already provides Sun Java binaries, Unreal Tournament and stuff under all sorts of licenses. If people want to use it, and it's in the package manager, then they will. You are very much an exception, IMO, taking ethical exception to Canonical's CLA. |
| 5 |
> |
| 6 |
> It's not ethical: It's practical. Canonical's CLA makes it so that |
| 7 |
> most (if not all of the) development of Unity will come from |
| 8 |
> developers payed by Canonical. The whole direction for the project |
| 9 |
> will come from Canonical. I cannot see a healty community project |
| 10 |
> derived from this development policy. |
| 11 |
> |
| 12 |
> Having said that, of course I could be wrong. |
| 13 |
|
| 14 |
You may take the position that you won't write code that you must hand |
| 15 |
copyright assignment to, but many other open source developers carry |
| 16 |
no such compunctions. The model you see as stifling has carried |
| 17 |
projects like MySQL for years, and carried OpenOffice since it was |
| 18 |
renamed from StarOffice--a policy which worked out until Oracle |
| 19 |
decided it wasn't going to keep doing things the Sun way. I don't know |
| 20 |
if Trolltech had a policy of copyright assignment for contributions to |
| 21 |
Qt, but they had a dual-license model which couldn't have worked |
| 22 |
without some mechanism of consent from outside contributors. |
| 23 |
|
| 24 |
"Copyright assignment" still supports the "open source" model of |
| 25 |
many-eyes, and the code being licensed under the GPL means that people |
| 26 |
who use Unity's source code elsewhere will need to share their |
| 27 |
changes. If Canonical decides to be more restrictive than the GPL at |
| 28 |
some point, they still get to operate under about as much leeway as if |
| 29 |
the code had been BSD-licensed. |
| 30 |
|
| 31 |
Apart from a spurious Netcraft report, BSD seems to still be both |
| 32 |
alive and reasonably well. |
| 33 |
|
| 34 |
There is still one curious artifact; all the projects I've cited |
| 35 |
(except, I guess, the various BSDs, but the AT&T story has its own |
| 36 |
levels of weirdness) were owned by companies which were later bought |
| 37 |
out. Trolltech by HP, which has been thrashing around worse than |
| 38 |
Yahoo. MySQL was bought by Sun. MySQL and OpenOffice were bought by |
| 39 |
Oracle when Oracle bought Sun. |
| 40 |
|
| 41 |
And, maybe it's just me, but I don't place huge importance on every |
| 42 |
small patch or bit of code I write; when I patch something, it's so |
| 43 |
that the program does what I want it to do. On the flip side, if the |
| 44 |
program is something I have a specific dedication to or a specific |
| 45 |
importance to (such as if I'm an officer or project manager), then I'm |
| 46 |
going to take more of an interest in retaining rights to the code. |
| 47 |
Otherwise, I just want the thing to work, and getting snippy about a |
| 48 |
source license is like demanding compensation from a restaurant for |
| 49 |
sticking a shim under a table that wasn't sitting level. |
| 50 |
|
| 51 |
-- |
| 52 |
:wq |
Archives