[License-discuss] NOSA 2.0, Copyfraud and the US Government

[Richard Fontana]

On Mon, Aug 28, 2017 at 02:18:10PM +0000, Karan, Cem F CIV USARMY RDECOM ARL (US) wrote:
> Hi all, as you know I've been pushing the position that the US Government may 
> have problems using copyright-based licenses on works that do not have 
> copyright attached.  One of the lawyers I've been working on this with has 
> been kind enough to dig up the exact statutes and give some clearer legal 
> reasoning on what the issues are.  It basically boils down to two issues; 
> first, there is question of severability 
> (https://en.wikipedia.org/wiki/Severability) which I've touched on before, and 
> the second has to do with copyfraud (https://en.wikipedia.org/wiki/Copyfraud). 
> Copyfraud is defined within 17 U.S.C. 506, section (c) 
> (https://www.gpo.gov/fdsys/pkg/USCODE-2010-title17/html/USCODE-2010-title17-chap5-sec506.htm). 
> I've copied out the relevant language below; the commentary within the 
> brackets is from ARL's lawyer:
> 
> """
> (c) Fraudulent Copyright Notice.-
> Any person who, with fraudulent intent, places on any article a notice of 
> copyright or words of the same purport that such person knows to be false, or 
> who, with fraudulent intent, publicly distributes or imports for public 
> distribution any article bearing such notice or words that such person knows 
> to be false, shall be fined not more than $2,500. [Note - Any software pushed 
> out under Open Source would not have a notice of copyright affixed to the 
> software. However, would software pushed out under an Open Source license that 
> assumes the existence of copyright be considered tantamount to a notice of 
> copyright and therefore an actionable fraud under this section?  Don't know.]
> """
> 
> I know that there were questions at one time about the need for special 
> licenses/agreements like NOSA 2.0, but this is one of those potential 
> problems.  Copyright-based licenses are great for works that have copyright 
> attached, but they may be problematic for works that don't have copyright 
> attached.

As has been pointed out before, I think, in software (including but
not limited to open source) copyright notices are commonly juxtaposed
with material that is clearly or likely not subject to copyright. 

Anyway, the theoretical risk here could be eliminated in lots of ways,
it seems to me (even without getting into what would be required to
show 'fraudulent intent'). For example, the US government could
include a copyright and license notice like the following:

  The following material may not be subject to copyright in the United
  States under 17 U.S.C. 105. To the extent it is subject to
  copyright, it is released under the following open source license: [...]

There's also the approach that is seen in https://github.com/deptofdefense/code.mil/blob/master/Proposal/INTENT.md.

> So, given that we had come up with the idea of using two licenses in projects 
> (CC0 for portions of a work that don't have copyright, and an OSI-approved 
> license for portions of a work that do have copyright attached), why should 
> OSI care?  The problem is that CC0 is still not OSI-approved (at least, it 
> isn't on the list at https://opensource.org/licenses/alphabetical).  That 
> means that the Government could be putting out works that are in some kind of 
> zombie-like state, half-Open Source, and half not.  If OSI approved CC0 as 
> being an Open Source license, or if NOSA 2.0 was approved, then the problems 
> could be fixed.  So, where are we in either case?

As I've pointed out before, CC0 itself does not eliminate the problem
your colleagues say they are concerned about, because CC0 assumes
copyright ownership. If they sincerely think it's dangerous to use the
MIT license then they should be consistent and say it's dangerous to
use CC0 too.

I think the use you are suggesting for use of CC0 is not actually how
CC0 is meant to be used. CC0 is designed for the case where copyright
ownership is likely or plausibly present but the owner wishes to get
as close as possible to waiving all of their rights. I think you are
saying you want CC0 to be used to ceremonially declare (possibly in
some cases incorrectly or misleadingly) that something that is not
subject to copyright ownership in the first place is indeed ... not
subject to copyright ownership in the first place -- which is not what
CC0 says.

Richard