I pleaded with the OSI board to provide formal feedback on the draft Federal Open Source Policy during the request for comment phase, but to my knowledge nobody did anything. Opportunity lost.
Indeed, that promised licensing guidance is coming from Federal Source Code Policy [1] which was just established. It's part of the White House's Second Open Government National Action Plan, and was 2 years in the making (7 years if you count from Obama creating the Open Government Initiative). The Federal policy actually mandates 20% of all software development must be open source, which is unprecedented, albeit currently without measurement criteria or enforcement teeth. That's why I said the White House is prompting discussions; they're most definitely getting a lot of folks riled up from the highest level. One of the biggest issues with the policy timing is that they completely punted on licensing, copyright, and legal mechanics, merely saying guidance will be provided later on code.gov undoubtedly for all the complex reasons being discussed here. That means the clock is now ticking fast to November. Tony Scott (U.S. Chief Information Officer) could settle things by issuing clarifying guidance on copyright and licensing, but his support apparatus will not necessarily exist afterwards. It's most likely that there will not be legal agreement and guidance eventually published will amount to "using OSI licensing is a great idea, talk with your agency's lawyers on specifically how." That said, there are some really smart guys working these issues in OMB and DoD pressing for change, so the optimist in me remains hopeful. Cheers! Sean [1] https://sourcecode.cio.gov/ On Aug 18, 2016, at 11:47 AM, "Smith, McCoy" <[email protected]> wrote: Given that the White House just released a memorandum on encouraging the USG to make more use of open source, and specifically said that it will be releasing licensing guidance on code.gov, perhaps the issues around 17 USC 105 and existing open source licenses will be resolved (or at least, the issues around existing open source licenses will be identified clearly) on behalf of all the USG: https://www.whitehouse.gov/sites/default/files/omb/memoranda/2016/m_16_21.pdf -----Original Message----- From: License-discuss [mailto:[email protected]] On Behalf Of Christopher Sean Morrison Sent: Thursday, August 18, 2016 1:27 AM To: [email protected] 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 There is exceptional evidence that the status quo is wholly inadequate. OSI fails to recognize challenges faced within the Federal Government, and it hurts open source adoption. Statistically speaking as the largest producer of source code on the planet, the U.S. Federal Government *should* be one of the largest participants in open source yet there is barely a presence. Some people recognize NASA as one of the largest proponents in the Gov’t space, yet they are one of the smaller agencies with one of the smallest budgets. Federal R&D, which is predominantly computer science work, is more than double the size of NASA’s entire agency! There are more computer scientists writing code for the Gov’t than there are for any single company in existence, including the likes of Google and Microsoft. Let that sink in for a minute. Where is all the code? If it was simply a release issue, there would at least be lots of public domain code floating — there’s demonstrably not. [1] If even a measurable percentage of Government lawyers felt existing OSI licenses were apropos, there would be a ample evidence of agencies using MIT/Apache/LGPL/etc — there’s demonstrably not. [2] There has been presented here a position by at least two major federal agencies (DoD and NASA) that copyright-based licensing is specifically viewed as a problem by their respective lawyers. There is obvious disagreement and uncertainty, but therein lies a fundamental problem. Nobody’s opinion has been tested. Nobody can prove that their point is any more or less correct. Lacking case law evidence, all that remains is overwhelming industry evidence that what is currently available is not in any way viewed as adequate in the Federal space. At a minimum, there is enough uncertainty that there is zero-% penetration. You have agencies here trying their damnedest to find ways to support open source amidst ambiguous regulations, unique legal circumstances (copyright), notoriously risk-averse environments, and untested theories. You have specific representatives (for huge organizations) here saying “I would use this, it would help us”. That to me those make for pretty freaking compelling reasons to support any new open source licensing, if it will increase adoption of open source in the Federal space. I ran on this platform for the 2016 OSI board election and missed it by fewer votes than I have fingers. This is a problem to a tremendous number of people. OSI licensing isn’t the only problem [3] faced by the Federal Government, but it is one of the most significant that has solutions being presented. NOSA 1.3 was offered but was then immediately shot down by FSF (for good reason, why is it even still on OSI's list??); NOSA 2.0 won’t likely be a solution without rework. ARL OSL aims to be so transparently compatible that it arguably limits proliferation (to the extent you can while creating a new agreement) and has much greater adoption potential with ASL’s rigor behind it. Dissenting won’t make agencies suddenly agree to just slap copyright-based licensing on their works or even releasing into PD. It will just continue to be lost opportunities for open source until there is congressional mandate, DoJ/DoC clarity, or case law clarity. White house is currently advocating and creating discussion, but we’ll see if that survives the election. Cheers! Sean [1] NIST, NASA, and 18F are outliers among hundreds of agencies. [2] What you can find are works involving contractors where copyright gets assigned. [3] Cultural ignorance is so maligned that DoD CIO actually had to tell agencies it’s *illegal* to NOT consider open source. On Aug 17, 2016, at 5:46 PM, Radcliffe, Mark <[email protected]> wrote: I agree with McCoy. As outside General Counsel of the OSI for more than 10 years, the drafting of a new "open source" license requires strong reasons. The reasons that I have seen in the list don't meet that standard. I strongly recommend against trying to develop a new "open source" license. -----Original Message----- From: License-discuss [mailto:[email protected]] On Behalf Of Smith, McCoy Sent: Wednesday, August 17, 2016 11:54 AM To: [email protected] 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 Or to put a finer point on it, the other issues you identify appear to be ones that are explicitly addressed in many already-approved OSI licenses, including Apache 2.0, the one you are modeling your license upon. I hope you're getting a sense that there are several lawyers on this mailing list -- lawyers who have years of experience looking at, debating, and giving advice on the issues you identify in this submission -- who think that your proposed license is a variant of Apache 2.0 designed to solve a "problem" for USG users with Apache 2.0 that we are skeptical even exists. Perhaps the ARL lawyers can clarify what the problem is, and that we are missing something. But I think at least I am having a hard time understanding how this license does anything that Apache 2.0 doesn't. -----Original Message----- From: License-discuss [mailto:[email protected]] On Behalf Of Richard Fontana Sent: Wednesday, August 17, 2016 11:33 AM To: [email protected] 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 On Wed, Aug 17, 2016 at 06:17:07PM +0000, Karan, Cem F CIV USARMY RDECOM ARL (US) wrote: Once again, liability isn't the only issue; there are also copyright issues (for contributors), and IP issues. If we could solve the problem via a simple disclaimer of liability, we would. We need to handle ALL the issues. Even if you were correct in the assertions you've made about ARL code, why is a new license needed for contributors other than ARL? _______________________________________________ License-discuss mailing list [email protected] https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss _______________________________________________ License-discuss mailing list [email protected] https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss Please consider the environment before printing this email. The information contained in this email may be confidential and/or legally privileged. It has been sent for the sole use of the intended recipient(s). If the reader of this message is not an intended recipient, you are hereby notified that any unauthorized review, use, disclosure, dissemination, distribution, or copying of this communication, or any of its contents, is strictly prohibited. If you have received this communication in error, please reply to the sender and destroy all copies of the message. To contact us directly, send to [email protected]. Thank you. _______________________________________________ License-discuss mailing list [email protected] https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss _______________________________________________ License-discuss mailing list [email protected] https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss _______________________________________________ License-discuss mailing list [email protected] https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss
_______________________________________________ License-discuss mailing list [email protected] https://lists.opensource.org/cgi-bin/mailman/listinfo/license-discuss
