[License-review] Approval Request: Free Public License 1.0.0
christianbundy at fraction.io
Wed Sep 2 02:44:48 UTC 2015
You're absolutely right, establishing software provenance is a
surprisingly difficult problem to solve. The non-repudiability of
provenance evidence depends on both the availability of documentation
and the integrity of digital signatures, which can be expressed in three
ways: unavailable, available, and available with integrity.
1. Unavailable: like Rick mentioned, these legal footguns lack any
meaningful indication regarding who or when they originated, and should
probably be avoided at this time.
2. Available: these works may include provenance documentation, but
without digital signatures these claims lack integrity.
3. Available with integrity: taking it a step further, some works use
digital signatures to verify that the provenance documentation has
integrity. While there's no such thing as perfect provenance, this
provides non-repudiation of a granted license (it does not, however,
provide whether the licensee has the authority to grant such a license).
Since availability isn't enough to establish non-repudiation of a
granted license with any meaningful integrity, the FPL 1.0.0 follows the
precedent of other OSI-approved licenses (e.g. CDDL, EPL MPL, etc) to
leave provenance documentation out of the license text.
John Cowan <cowan at mercury.ccil.org>
> If anything, this makes your point stronger: licensed but unmarked
> code is permanently tainted unless the copyright owner can be found,
> and should never be loose in the wild.
Does the Berne Convention Article 7(3) protect against this (copyright
expiration after 50 years for anonymous authors) or is there more to it?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the License-review