| 1 |
On Tue, 2005-09-20 at 07:25 -0400, Seemant Kulleen wrote: |
| 2 |
> On Mon, Sep 19, 2005 at 10:37:38PM -0400, Daniel Ostrow wrote: |
| 3 |
> > Oh and one other important distinction. As I forgot to address the |
| 4 |
> > question of "Why should the genkernel developers need to sign this |
| 5 |
> > agreement if some future wizbang genkernel replacement developed on |
| 6 |
> > berlios infra doesn't have to?" |
| 7 |
> |
| 8 |
> Coming back to grant's point a bit. There are projects that were/are |
| 9 |
> developed entirely on non-gentoo infrastructure (I believe the eselect |
| 10 |
> stuff, for example), yet is becoming default on gentoo systems |
| 11 |
> (opengl-update, for starters, has gone the way of the dodo, to be |
| 12 |
> replaced by eselect). |
| 13 |
> |
| 14 |
> How does this agreement play to things like that? If eselect goes on |
| 15 |
> (and based on its technical merits, there is every reason that it |
| 16 |
> should) to become the default tool in gentoo, then where does that leave |
| 17 |
> us? I'm with Grant on this: I'm not convinced. |
| 18 |
|
| 19 |
The question isn't whether the tool is useful, or for that matter |
| 20 |
default. That really doesn't matter. As far as I'm concerned the |
| 21 |
question is whether it is an official Gentoo project, that is the |
| 22 |
distinction I am making. If it is an official project we need a way to |
| 23 |
protect it. I have no problem whatsoever with having eselect (and it's |
| 24 |
modules), genkernel, portage, etc. as external non-official projects |
| 25 |
(again even having their code on our infra). The problem is as long as |
| 26 |
they are official unless there is an agreement such as the one outlined |
| 27 |
we cannot protect the interests of the developers who have worked on the |
| 28 |
code. |
| 29 |
|
| 30 |
All I'm saying is that I believe that code developed as an official |
| 31 |
Gentoo project needs to have the protection of the foundation and the |
| 32 |
only way to achieve this is to hold a copyright or license for said |
| 33 |
code. |
| 34 |
|
| 35 |
It's an all or nothing game there. Either official projects have to be |
| 36 |
100% covered or they are covered not at all and I personally think that |
| 37 |
it would be irresponsible to leave the code open for what is essentially |
| 38 |
theft when we have the ability to protect it. That being said it would |
| 39 |
be entirely up to the developers in question, they should feel free not |
| 40 |
to sign the agreement and remove the official status of their project. |
| 41 |
|
| 42 |
The part that I am missing is what would be the downside. I have yet to |
| 43 |
hear anything other then "I feel icky about assigning my copyright to |
| 44 |
someone else." What is it that makes you feel icky? What do you fear |
| 45 |
could happen under such circumstances? Once we have a clear cut |
| 46 |
understanding on that the wording of the document can be changed to |
| 47 |
allay those fears. Again there is nothing insidious about this |
| 48 |
document...it can do people no harm. |
| 49 |
|
| 50 |
The very very very worst case scenario: |
| 51 |
|
| 52 |
Gentoo Dev Joe Smith develops some portage code. Since portage is an |
| 53 |
official project he is required to sign the agreement. He then leaves |
| 54 |
the project and reuses that exact same code in another project under a |
| 55 |
different (heck even closed source) license. We find out about it and |
| 56 |
are put in the pickle of having to enforce the original copyright, since |
| 57 |
it is indeed an all or nothing game. In this case we have to protect the |
| 58 |
interests of the project (and the open source movement) as a whole |
| 59 |
instead of the interests of one developer. |
| 60 |
|
| 61 |
That is it. The worst thing that could happen is having to enforce the |
| 62 |
spirit under which the code was originally developed. Unfortunately in |
| 63 |
the case above what Joe Smith did is illegal, there is absolutely no |
| 64 |
question about that. It's just that without the agreement there would be |
| 65 |
no way to enforce that legality in court. It's a strange thing to hear |
| 66 |
but Joe Smith could effectively pirate his own code. |
| 67 |
|
| 68 |
This is not to say that Joe Smith doesn't have the right to reuse the |
| 69 |
code. The code is GPL so he can take it, and Brian's and Jason's and |
| 70 |
whoever else's code and reuse it to his hearts content. All it is saying |
| 71 |
is that he cannot change the license under which that code and |
| 72 |
derivative code is released and he has to include the original copyright |
| 73 |
in the works that stem from it. Furthermore in the event that he did |
| 74 |
change the license/copyright we would have the right (and the |
| 75 |
obligation) to enforce the original copyright in a court of law. |
| 76 |
|
| 77 |
-- |
| 78 |
Daniel Ostrow |
| 79 |
Gentoo Foundation Board of Trustees |
| 80 |
Gentoo/{PPC,PPC64,DevRel} |
| 81 |
dostrow@g.o |
| 82 |
|
| 83 |
-- |
| 84 |
gentoo-trustees@g.o mailing list |
Archives