oui.txt is simply a collection of facts, arranged in a table in a trivial way. This is not copyrightable, therefore we shouldn't have "freedist" in LICENSE. <http://en.wikipedia.org/wiki/Feist_v._Rural>
Also if it should be copyrightable, then "freedist" is still wrong. In that case, the file wouldn't be redistributable at all.
"# Install oui.txt as requested in bug #283791 and approved by upstream" from: http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/net-wireless/bluez/bluez-4.101-r3.ebuild?view=markup So perhaps it's allowed after all. The file is part of the systemd (udev) git tree and udev relies upon it... So we have always been installing this file. Now it's just part of another package. Not perfect, but lets not make any hasty decisions...
Please note that we cannot have udev rely on anything that is non-free, because that would violate our Social Contract. CCing trustees.
(In reply to comment #0) > oui.txt is simply a collection of facts, arranged in a table in a trivial > way. This is not copyrightable, therefore we shouldn't have "freedist" in > LICENSE. > > <http://en.wikipedia.org/wiki/Feist_v._Rural> I'm inclined to agree - could you have a more clear example of a "telephone directory" than a file that contains nothing more than a mapping of strings of numbers to company name/addresses? I saw a 2006 thread on debian-legal that came to the same conclusion, though I don't know what the package was. net-analyzer/ntop includes this file as well. That's my initial personal opinion. Is this urgent or can it wait for the next trustee meeting on the 16th? If any trustees want to seek formal legal guidance we might want to get that started before then.
You know I'm the first one that is not happy about the current situation, but unless you can come up with a better solution, I want to have something that explicitly make it visible that the file is NOT || ( bsd gpl2 ). Is this a shitty situation? Yes. Can we get out of it alone? I doubt so. Kay said that RedHat is contacting IEEE again and maybe they can get more than a oneliner answer. As Rich said, we've been distributing the file for a long time anyway, not only ntop (which also ships a copy of the geoip database, whut? different story) but also with lshw sources (I just stopped installing it last night in -r2), and dev-python/netaddr (together with the iab.txt file, and a set of address assignments from IANA) as well as bluez (where this problem started from). So I caved in, and I'm hoping for the best. Am I happy about this? No. But behaving like Debian here doesn't help.
Diego, this is trivial stuff and as such not copyrightable, because it doesn't reach any threshold of originality: - There's only one possibility to select the data sets (namely take all of them). - The layout of the individual data sets is trivial: there's just a number and the address of an organisation. - Also the arrangement of the data sets is trivial, namely they're sorted in ascending numeric order. I also believe that the argument that it would be copyrightable as a database in Europe doesn't apply here. That's regulated in Directive 96/9/EC, which says in Article 3 (1): In accordance with this Directive, databases which, by reason of the selection or arrangement of their contents, constitute the author's own intellectual creation shall be protected as such by copyright. No other criteria shall be applied to determine their eligibility for that protection. It seems quite clear that oui.txt is no "own intellectual creation" "by reason of the selection or arrangement of their contents". Also the "sui generis right" of 96/9/EC Article 7 doesn't apply because the IEEE isn't located in any EC member state (see Article 11). Disclaimer: IANAL, so don't take the above as legal advice.
So do we have a "facts" license?
(In reply to comment #7) > So do we have a "facts" license? If it fails to meet the requirements for copyright protection, then it's in the public domain: <http://www.copyright.gov/help/faq/faq-definitions.html#public_domain> Since we distribute the file with other packages since a long time, I don't think that it's such an urgent matter. Therefore, I'd suggest that you go with "public-domain" for the time being, and let the trustees decide on their next meeting if that's appropriate.
Done, keeping the bug open until we have a clear understanding of what the f.... is going on here.
I agree with Ulm’s reasoning in comment #6 — IMHO oui.txt should not be copyrightable nor fall under the EU database protection rights. It’d be great to get this cleared up by IEEE though.
This bug applies to the iab.txt, also from IEEE, that is now required for udev-197 (and is part of udev git tree as well as hwids ebuild already) Adjusting summary to include iab.txt
(trustees hat) ulm/flameeyes: Until RedHat hears back from the IEEE (per comment 5), there is no action here for the trustees. Given the discussion of copyright here, we feel marking it as public-domain for now would be acceptable. If there is a followup from the IEEE that raises problems, please re-add the trustees. There are many other packages that include the oui, iab, oid files from the IEEE, and all those upstreams would be affected by this as well.
Looks like we have another copy of oui.txt in Portage too, bug 465606
Closing, since sys-apps/hwids ebuilds are marked as public-domain.
