Re: Licence terms of libstdc++-v3/porting.texi

[Joe Buck <jbuck at racerx dot synopsys dot com>]

>         I wonder sometimes about the exact legal status of licensing. Just
>         putting text into a file does not create a licensing or copyright
>         statement of legal fact.
> 
>     It does, if the person making the statement is legally authorized to
>     decide the policy.
> 
> Yes, but how do you know from the file who that person was let alone if they
> are authorized to decide the policy?

In general, you don't; you need additional information than just what's in
the file, like who is publishing the file and what care has that person
taken to do the job right.

In the case of FSF-owned software obtained from FSF sites, you do know (at
least to a much greater extent), because the FSF requires assignments and
employer waivers from every contributor, and the FSF's philosophy,
purpose, and licensing terms are widely known.

In the case of random packages announced on freshmeat, you may not know.
Unfortunately, some underfunded universities are asserting rights that
they didn't assert in the past, to try to make money from university
research, so I'm afraid that some college students who think that they are
free to put their work under the GPL may be wrong, especially if they used
university computer facilities.  (And, no, this is not a US-only problem).

Note that "how do you know" questions of this form can be repeated
forever, and eventually just lead to solipsism (how do you know anything
exists?  Maybe you're just hallucinating).  Absolute certainty isn't
achievable, the best we can do is to have enough evidence so that we can
be reasonably confident that something is so.