Open Source Decision Models

[Chris Travers]

Verdy:

Maybe copyright protects more things in France?

Here, for example, the whitepages of the phone book are probably not
protected.  In some other countries, YMMV.

This is why I was careful to refer to US copyright law.

On Fri, Mar 5, 2010 at 10:49 AM, verdy_p <verdy_p at wanadoo.fr> wrote:
> "Chris Travers"
>> I would suggest "code" as "whatever is subject to copyright controls"
>> from the GPL's perspective.
>>
>> Data and equations themselves may not be protected, so you are left
>> with licensing your creative arrangements of these things.
>> Sounds like code to me....
>
> What a stupid answer.

Before calling my answer "stupid" please see how it tracks what Larry
Rosen wrote.
>
> All of the creation is protected and covered by the licence (including the GPL), including data and equations, as
> far as copyright laws can cover them.

Which in the US isn't very far.  US copyright law only covers creative
elements, and functional elements are explicitly excluded.
Collections of facts are generally only copyrightable in areas such as
selection and ordering.  On the other hand a comprehensive arrangement
of facts listed in alphabetical order would not be given any
protection I can see by US courts.  Again IANAL, but this is pretty
basic where I live.

I know it's different in Europe.  Maybe if you photocopy the telephone
directory (outside of maybe the advertisement section/yellow pages) is
copyright infringement?  It's not here.

If you don't need a copyright license, the GPL is irrelevant.

>The GPL obligations apply to all of them, even if it ALSO (but not only)
> defines code for explaining that it should also be available as humane-readable source (in a way that allows a
> humane to modify it and reacreate a derivative work).

Sure, but that's orthogonal to the question of what the GPL protects
(i.e. what the author gets out of it).

> For a spreadsheet, this means that formulas must be accessible in source-mode and the data that uses it must be
> accessible and modifiable, even if the spreadsheet exists also in a securely protected version. Remember that GPL's
> copyleft is a contractual extension of the legal copyright, the later is still directly rerefenced explicitly by the
> licence text. This extension is an additional grant of rights (that otherwise would be completely subsject to prior
> authorauthorizations) in exchange of additional obligations (which do not contradict the copyright laws).

Sure.  No argument there, provided that copyrights would be infringed
absent acceptance of the GPL.

However, the data could be distributed separately in the US if it is
insufficiently expressive without running into the limitations of the
GPL.  I could, for example, take the data, refine it, add more, and
might not have a derivative work at the end.

The issue I was trying to address is what level of protection the
author gets out of releasing a spreadsheet under the GPL.  I think you
get a lot less than you would out of a software program where there is
a much larger possible reach of coding styles, etc.

>
> The GPL is a true licence, and there has been much enough legal decisions in courts (at least in America and in
> europe) to confirm it. The author's rights are explicitly protected and they keep their freedom of providing other
> licences if needed, and the freedom of stopping delivering new copies of their product. In addition, the authorship
> is legally transmissible, like any other property, if the author ceases to exist (but their moral rights can't be
> seized, in countries where these rights are legally created, and recognized including by WIPO members, and the GPL
> does not restrict these moral rights).

How do you know the GPL does not restrict these moral rights?

If I was a Canadian game developer and I released a game under the
GPL, could I REALLY sue someone over moral rights violations if they
alter my game to, say, include sexual or offensive content?  I would
think that express permission by the author (and holder of moral
rights) to distribute modified copies would severely restrict moral
rights claims.  What am I missing?

> There's no need to create then distinctions between data, formulas or presentation/design parts of a spreadsheet. If
> the spreadsheet is licenced, this licence covers it completely under copyright laws, even if there's adiitonal
> grants or obligations concessed by the author for specific parts in the specific text of the licence or in a non-
> contradicting addendum or private contract.

I am not touching the GPL-is-license-or-contract mess.  Generally US
lawyers I have talked to suggest it is a contract, but YMMV.

In many jurisdictions, raw data and equations are not subject to
copyright though expressive collections of such may be.   Hence the
need to think about such a line when deciding whether to release
something open source.

I know that in some jurisdictions (maybe that includes France),
databases can be subject to copyright.  I wouldn't assume that such
would be protected though, simply because this is not universally the
case.  Since open source crosses national boundaries, I would assume
less strict countries are where one would look to see what level of
protection the license would actually afford the author.

Best Wishes,
Chris Travers