[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Copyright of console illustrations



On Sun, 31 Aug 2025 12:23:48 +0200
Simon Josefsson <simon@josefsson.org> wrote:

> Michael Lustfield <michael@lustfield.net> writes:
> 
> > On Sat, 30 Aug 2025 18:47:39 +0200
> > "Andrea Pappacoda" <tachi@debian.org> wrote:
> >  
> >>     License:   public-domain  
> >
> > As a rule of thumb, nothing is actually public domain unless it's very old.
> > There are licenses that attempt to replicate public domain, but it is not
> > possible to simply state you release something into public domain--at least not
> > in EU or US.  
> 
> Do you have any reference explaining that claim?
> 
> My perception is that there are frequent current publications of public
> domain content, including stuff from NIST based on this "license":
> 
> https://spdx.org/licenses/NIST-PD.html

NIST is US government, yeah? That's one of the exceptions where almost
everything released is into public domain.

> Another example are DJB's libraries, for example lib25519:
> https://lib25519.cr.yp.to/license.html

This is a perfect example where claiming something is public domain does not
make the statement true. If this were litigated within the US, then the legal
copyright would be traced back through https://lib25519.cr.yp.to/people.html.

This is certainly venturing into legal theory more than practical application,
because this copyright would only be brought into question if the license were
challenged, and that would "most likely" be from some obscure company/person
trying to claim a patent on that source; or maybe Netherlands becoming evil and
wanting royalties on the code they sponsored. That would also involve reviewing
individual functions and even specific lines of code, not the project as a
whole.

The closest real world scenario that I can think of is google vs. oracle (java).


Keep in mind, my original point was that the packager should be very cautious
when making a determination that something is public domain, because that
assumption will almost always be wrong.

In this particular case, I believe it is clear that public-domain is not
correct for any portion of this code. They also raised a question about
non-commercial use, and my opinion is basically that the intent is obvious, but
additional clarification (an email included in the package) couldn't hurt.


Reply to: