[License-review] For Approval: Convertible Free Software License, Version 1.3 (C-FSL v1.3)

[Carlo Piana]

On 11/01/19 11:19, Elmar Stellnberger wrote:
>
>> The offending point is that it's the Original Authors, not the 
>> copyright holders, those who have the power to authorize relicensing 
>> and adding new Original Authors (I suppose all of them jointly, or 
>> perhaps based on a majority vote, and in case, how do you calculate 
>> the majority? Per capita? Per share?)
>
> all of them jointly.
>
> It is the original authors who are copyright holders. As already 
> documented before I believe it to be the most just solution. It allows 
> a small team of developers to act like a big firm which employs a 
> contributor license agreement. Who of you has ever complaint about 
> contributor license agreements? Why do you not complain about Apple 
> taking the FreeBSD kernel and exploiting it for its own profit without 
> giving anything back to the open source community?
>
Wrong, and now I am personally getting frustrated by the lack of 
understanding of the problem, which should be apparent by abundant 
explanation offered by valuable and respected copyright experts and by 
myself too.

The original authors are copyright holders as much as the contributors 
to the project OR those who fork. No discussion about it.

# Ho CAA work with open source projects

No license whatsoever imposes anybody to accept external contributions. 
But an open source license MUST allow third parties to fork and extend 
the project. Those who have forked can only use the software according 
to the public license. Therefore, it is up to  the original copyright 
holders whether to grant the right to relicense BUT they cannot claim 
relicensing rights on the contributions they receive.

If they want to receive this right on third parties' contributions they 
only have two options: either they only accept no copyleft contributions 
and/or require copyright to be assigned.

This a fair deal. Third parties know the rules and can decide whether to 
contribute the code or not. This does not impair their right to exploit 
the four freedoms in any way. If they don't take the deal, they lose the 
option to have their changes incorporated upstream.

In the above scenario those who do not want to contribute under the 
original copyright holders term, simply fork the project and start off a 
new one. This way the original holders must only merge changes from the 
fork under the terms allowed by the forker. They are in the same 
position WRT the changes as the original holders are WRT the original 
set of code.

# How is this license different?

The proposed license, conversely, allows the original holders to take 
and merge the changes from the forked site, embed them in their work and 
relicense the derivative work so obtained. The forker does not receive a 
symmetrical right over the code made by the original holders. The 
original holders are those who can vote out any relicensing. This is 
discriminatory. It creates TWO classes of copyright holders.

This is totally disingenuous, as it is disingenuous claiming that this 
is no different from asking assignment of he copyright. No! Copyright 
assignment is a condition to contribute the code upstream, not a 
condition to use the code under FOSS licensing. If I refuse to to 
contribute the code upstream because o those conditions, I can just set 
up my own project and fork from it. The original holders would have 
rights only based on the license (copyleft).

The propsed license, however, allows the original holders to force 
assignment (by an reserving an equivalent right). There is a HUGE 
difference. The same difference between an option and an obligation to 
contract.

# THERE IS MORE!

Under Section 5.1

> f) You must make available to the public any Derivative Work or changes to the Work within one month from their creation.

This creates another huge asymmetry.

Under copyright law, I can decide whether or not to publish the code i 
contribute. Under the porposed license, it is required that this right 
is waived as a condition to use the copyright. I think this is 
unprecedented. I might be making changes that I don't want to release to 
the public neither in object code, because they are buggy, because I 
don't want to take responsibility if anybody uses them. In this case, if 
I fork the project I have NO WAY TO AVOID PROVIDING IT UPSTREAM.

This is APPROPRIATION plain and simple. So any changes I make, whether 
or not released to the public, are contributed with an equivalent of an 
assignment, granting rights over the derivative code, including the 
copyright of the developer, which are not available to the developer and 
there is no way to avoid it.

# How the license can be changed

There is ONLY one way to preserve this license as something that might 
be tentatively compliant. That is, embed the CAA in the license, so that 
any contribution to the original project is automatically under the 
relicensing condition. BUT any changes, whether or not contributed to a 
public forked project are not assigned, they are not under the 
relicensing condition. And the authors of the forked project have the 
right to claim they are original author and have the same rights if 
their project becomes successful and attracts contributions, otherwise 
the discrimination still stands.

I expected something in this direction with the new license, conversely 
I only see stubborn rewording that makes it more hard to lay persons to 
spot how the license in practice work. This is not going to succeed, I 
submit.

With best regards,

Carlo

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensource.org/pipermail/license-review_lists.opensource.org/attachments/20190111/2340b8d0/attachment-0001.html>