> Ugh. I’m perfectly happy to give away my own code and patents when > I choose to do so but I would be very unhappy if I accidentally gave > away someone else’s work and cost them thousands of dollars of lost > royalties.
This is a well-know problem with no solution for which all parties are happy in all cases. In addition to contributions to FOSS, this arises in connection with development of standards (the domain of the Rambus case, which I believe was mentioned earlier in this thread). In order to appreciate why there is no general solution, it is useful to try sitting on each of the opposing sides of the issue: the contributor and a consumer of the contribution. As pointed out, the contributor is troubled by the potential of impacting patents from a distant part of some larger world of which the contributor is a part. This can occur in companies of even modest size (see the IPO list below) and the challenge increases with the diversity of a company's overall business - in the extreme, consider a large diversified conglomerate. From the consumer perspective consider the following: A company X invests in use of software to which research lab Y contributed, only to later have a patent that covers that contribution asserted against it by the entity that owns lab Y. How OK is that? Perhaps different people will have different answers depending on the details. The likelihood such a patent situation actually arising is probably quite small. But, how does that cut? Should the contributor or the consumer take comfort in that speculative low probability? The IPO publishes annual lists of the assignees are large numbers of new US patents: http://www.ipo.org/wp-content/uploads/2016/06/2015-Top-300-Patent-Owners.pdf I see the Navy at 104, the Army at 201, DHHS at 214. -- Scott ----- Original Message ----- From: "Nigel H. Tzeng" <nigel.tz...@jhuapl.edu> To: "Lawrence Rosen" <lro...@rosenlaw.com>, license-discuss@opensource.org Sent: Friday, August 19, 2016 4:02:35 PM Subject: Re: [License-discuss] [Non-DoD Source] Re: [Non-DoD Source] Re: U.S. Army Research Laboratory Open Source License (ARL OSL) 0.4.0 From: License-discuss < license-discuss-boun...@opensource.org > on behalf of " lro...@rosenlaw.com " < lro...@rosenlaw.com > >There are other important reasons besides "aging out" why the claims of >copyright on parts of functional works like software are often denied. (See 17 >U.S.C. 102(b), for example.) Aging out isn't the only obstacle to copyright >claims >that make the copyright aspects of FOSS licenses unenforceable while >they remain contracts to disclaim warranties. So when several here suggested >that ALL FOSS WORKS probably contain public domain content, this is ANOTHER >>example not involving aging. >The USG and ARL are not unique. Public domain is what it is for software works >for everyone here (and probably abroad too). A unique FOSS license isn't >necessary to "protect copyrights" in public domain works. Almost any FOSS >>contract will work to protect the licensor. The point is that the code fond in other FOSS WORKS that are in the public domain is generally not significant enough to contain an implantation of a software patent. This is why your shakespeare example isn’t valid and why the USG and ARL could be unique. Software in the public domain have neither an implicit or explicit patent grant. Which should be a concern of ARL. OSS licenses that do not explicitly handle the public domain case does not apparently meet the needs of the ARL (and probably the rest of the USG) because there may be a issue when no copyright exists. That said, it occurs to me that ARL would not want to use an Apache style patent grant but a ECL v2 style grant. Otherwise someone at the DOE could release source code that implements a patent owned by ARL that they are licensing to industry for royalties. Or vice versa. Under 15 US Code § 3710c a.1.A.i The head of the agency or laboratory, or such individual’s designee, shall pay each year the first $2,000, and thereafter at least 15 percent, of the royalties or other payments, other than payments of patent costs as delineated by a license or assignment agreement, to the inventor or coinventors, if the inventor’s or coinventor’s rights are assigned to the United States. According to this site: http://www.tms.org/pubs/journals/JOM/matters/matters-9004.html <blockquote> It is estimated that the government has title to over 30,000 patents and annually files several thousand new applications. </blockquote> Ugh. I’m perfectly happy to give away my own code and patents when I choose to do so but I would be very unhappy if I accidentally gave away someone else’s work and cost them thousands of dollars of lost royalties. My assumption is that the USG is treated as a single legal entity for patent and copyright purposes which may be incorrect. Even if not, one would assume that ARL would be treated as part of the Army and could impact any other Army lab, FFRDC, UARC or university and other organizations conducting research for the Army. _______________________________________________ License-discuss mailing list License-discuss@opensource.org https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss _______________________________________________ License-discuss mailing list License-discuss@opensource.org https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss
