Changing the subject ot reflect what we are talking about.
On 18/04/2019 13:05, Bryan Gonzalez wrote:
However, i think it makes sense to figure this out before the
conversion. If i understand correclty, you may want to add file
headers to all files that existed at revision 3070.
Actually, I am not sure anymore. It cannot be understood that only files
existing at the time of the inclusion of the license file are covered by
the license. So implicitly, all files added, and modifications since
then are covered too (that is, all files in the jhalfs repo, I think). I
hope the svn history is enough to fulfill the requirement of "prominent
notices stating that you changed the files and the date of any change".
Otherwise, almost all contributors have infringed the terms of the license.
Or maybe we're overthinking this. In the US, the lack of a license
means you have no permission to copy, distribute, or modify. That
said, we do have a license file, even if it wasn't fully filled out
and it is implemented since it's referenced somewhere in the code
(though this is not necessary, in only needs to be in the repository).
But the matter of when it was added, seems moot. It's there now. I'd
call it an oversight that has since been corrected, especially in the
spirit of the project, if I understand correctly, was always open source.
Finally, I do not believe that copyright/licensing extends to
revisions of code. Code versions are a natural product of using a
version control system, arguably for convenience and collaboration.
But, when it comes down to it, for example, in court, 2 versions of a
software, even from the same repository could be considered to be
distinct, if their code is different enough. And generally speaking,
in court, a specific version of a binary, is matched up with a
specific version of source code. You wouldnt assign blame to code that
is newer than a binary in question.
I know, binary code is not the best example, but software and
intellectual property is a bit murky until it gets to court and a
judge figures it out.
But my last point is this: copyright is an active process. You have to
actually file copyright with a particular body in order for it to
exist. Take a look at the copyright publication I link below.
Ref: https://choosealicense.com/no-permission/
Thanks for this link. Make me think we could choose a more permissive
license. My main point is the disclaimer about using the code, not
preventing copies to be sold or whatever. Anyway, according to this
site, adding a LICENSE file is enough for licensing the project. And
about LGPLv3, they say it is optional to add copyright and disclaimer at
the top of each file.
And:
https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.copyright.gov/circs/circ61.pdf&ved=2ahUKEwiK7YOWvtnhAhXlmuAKHVLADOsQFjAKegQIBRAC&usg=AOvVaw3ytIj7pvuLjK4NzyolDBej
Interesting. Do you really think all software starting with "Copyright
(C) <year> somebody" has been registered with the US Copyright office
(and for each revision), or the equivalent in other countries? Adding
such lines is recommended by the GPLv3 license, but they do not talk
about registering.
Pierre (more and more puzzled, I must say, and starting to think he is
spending to much time on that...)
--
http://lists.linuxfromscratch.org/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
