Assignment of copyright
To make it easier to resolve any potential copyright disputes, and to simplify the task of attribution for any other teams who modify and redistribute the Foswiki product, I suggest that contributors be required to assign their copyright to the Foswiki association.
- 23 Nov 2008 - 00:57
Which I for one, after my experience with TWiki, would refuse to do.
Also note that the Copyright Peter Thoeny and others is a pretty contentious assertion that Peter kept making in TWiki's svn. Many files that have otherwise been quite totally written by others were regularly amended by Peter to say this. As I understand it, Copyright is not something you can just claim, you actually need to have done some creation.
I would much rather that the Copyright for a file explicitly mention the main contributors to that file, not some body or people that havn't actually worked on it.
- 23 Nov 2008 - 05:06
Yes, creators own the copyright over their own original works at the moment of creation and no one can take it away.
Clearly there are logistic issues in tracking all contributors to each file (size of contribution makes no difference for original work, so just tracking main contributors would be insufficient) — if done manually, it's easy for submitters to forget to add their names to the list. In theory a hook could be added to Subversion commits to add in information about the submitter into each modified file. Alternatively, in theory the list of submitters for each file could be extracted by a script that is run when a product release is issued.
There is also unfortunately a legacy issue for existing files: unless their past history can be uncovered, it would be difficult to remove a more generic attribution from them.
As long as there are no licensing or other copyright violations to pursue, it won't matter too much if copyright is held by hundreds of different people. If Foswiki ever did want to pursue a violation, however, legally it would have to be a joint action by all copyright holders, which would complicate the proceedings. -- IsaacLin
- 23 Nov 2008 - 05:22
I'm no lawyer, but I believe that a copyright infringement action could be pursued by any one person who felt that their individual copyright had been infringed. There is no need for a class action, though it might help from a perspective of bringing resources to bear.
A list of subversion submitters for each file unfortunately doesn't quite cut it, as code is often moved between files. The only rational way to do copyright attribution in code would be to record a copyright statement for each line, and everyone who changed that line. Because that isn't practical, we have chosen instead to have a "shared copyright" over the entire repository. If someone chooses not to add themselves to the list of copyright holders, that is their choice.
At the end of the day my interest in copyright is solely in protecting the freedoms of the open source project and the people who work on it. Recognising the authorship of specific code modules is a different issue, and should not be confused with copyright. Sven's suggestion of a list of "main contributors" is fine for reflecting this - in theory. The problem is, who decides what a "main contributor" is? Several times I have seen people add themselves as "author" of works on the basis of what the original author would consider to be a minor change.
- 21 Mar 2009
The FSF believes that joint action is required: http://www.gnu.org/licenses/why-assign.html
Under U.S. copyright law, as I understand, each owner can grant a non-exclusive license for use of the work, so if one author did not agree with the legal action, that author could grant a license to the supposed infringer.
Without assignment to an association, then having copyright assigned to "Foswiki Contributors", as for example is done in Foswiki.pm, and perhaps having a separate "Main contributors" list, is I believe the most practical solution. A copyright statement within each file cannot practically be limited to main contributors, because even a minor change that creates a new derived work results in a new act of authorship. (From a legal point of view, the concept of copyright ownership is based on the authorship of the work and thus to avoid ambiguity, it is best to limit the term "author" when referring to the recognized author for the purposes of copyright.) -- IsaacLin
- 21 Mar 2009