-
Notifications
You must be signed in to change notification settings - Fork 601
Remove license examples that are not compliant with the memorandum's policy #64
Conversation
(Prose mucked with the metadata at the top of the file. Sorry about that.) |
👍 This is an important pull request, and one I'd like to see merged. There aren't any acceptable downstream restrictions on the use of government produced data, including attribution - it's just public domain. This aspect of US law is one of the things I'm the most proud and patriotic about. I don't know of any other countries where government information is irrevocably placed in the public domain. Let's keep leading the world on this. 🇺🇸 |
Reconcile your request with the Memorandum's stated intent that public data listing also applies to: "This public data listing should also include, to the extent permitted by law and existing terms and conditions, datasets that were produced through agency-funded grants, contracts, and cooperative agreements (excluding any data submitted primarily for the purpose of contract monitoring and administration), and, where feasible, be accompanied by standard citation information, preferably in the form of a persistent identifier." (III-3-b) The listing of licenses should simply be amended to clarify that government-created data are in the public domain. Government-funded data are a different issue. There are many reasons that the public domain mark should not apply, but among these is the clearly stated requirement that government-funded data should be cited, as in what is being asked in the memorandum on increasing access to the results of federally-funded scientific research (http://www.whitehouse.gov/sites/default/files/microsites/ostp/ostp_public_access_memo_2013.pdf). This pull request should not be merged because it does not adequately address both instances of open data that the memorandum expects to cover. All of these licenses are acceptable and reasonable in the case of government-funded data. |
FWIW, I discuss this issue over at #5 and @philipashlock's PR #11. There are ways to address this, and I'd like to see something a touch stronger than just mentioning that government produced data is in the public domain. As an example, the White House is currently using an explicitly bifurcated copyright notice over at WhiteHouse.gov for these situations. To the extent the licenses listed here need to reflect everything POD aims to cover, bifurcating the list into two lists seems like a fine approach, |
I am cool with treating the two issues separately, but this pull request does not do what you suggest, @konklone. I agree that the memorandum is inartfully silent on the issue of government-created data. But that doesn't invalidate the treatment of the issues for government-funded data. |
@dsmorgan77: There's no contradiction for the memorandum to apply to data acquired through a contract and for "no restrictions" to mean actually having no restrictions. That's perfectly plausible, and some would say good policy to boot. |
@tauberer sure, data acquired through a contract could be in the public domain. perfectly valid case, but not the standard the memorandum sets. and you're not addressing things like federally-funded research, where the administration has clearly set the bar at at least setting rights to require citation. so just deleting the licenses that are compatible with what the policy actually says isn't prudent or helpful to agencies that have to implement the policy. i'm with @konklone. there's room for both treatments. |
Thanks so much for the great discussion. We have our lawyers review all pull requests to the license file, and they are currently reviewing these edits. Thanks for your contribution!
|
I just noticed that this project license itself under CC-BY. Depending on how this pull req is resolved, the README here might need to be updated too. Would be nice to have this project lead by example, though, and show that "no restrictions on use" means no restrictions on use. |
@konklone, Jonathan Gray, and I have just published "Best-Practices Language for Making Data 'License-Free.' " the guidance: http://razor.occams.info/pubdocs/2013-08-19_license_free.pdf We offer it as a tool, like POD itself, for agencies to meet the Memorandum's demand that (some?) data have "no restrictions" on use. It also strongly suggests that the community standard of "license-free" become more broadly adopted in government. Our guidance has support from the Sunlight Foundation and the Center for Democracy & Technology. And from here we plan to open up community discussion on the document (perhaps in a manner similar to how POD works on github). |
We updated our guidance today, and added support from OKFN, Public Knowledge, the EFF, Carl Malamud's Public Resource, and a bunch of others. It's over at http://theunitedstates.io/licensing/. It cites specific examples (including Project Open Data) and offers best practice language based on principles of US open government data and those examples. Version 2, among other things, clarifies some things around IP ambiguity and the applicability of CC0 to software as well as data. |
Hey @JoshData, I'm closing this PR because there have been many updates to the /open-licenses page since it was requested. Notably, per the v1.1 metadata schema updates, |
❤️ |
The open data memorandum says that "[o]pen data are made available under an open license that places no restrictions on their use." Most open licenses place restrictions on use (such as attribution) and so would not be compliant with the memorandum's definition of open data. (See sections I, III-1-a, and III-1-c.)
This pull requests removes the examples that are not in line with the policy in the memorandum.