59 bytes, text/x-review-board-request
At the moment some of the Intl.PluralRules tests are licensed under MPL 2.0. We got a request from the V8 contributors working on PluralRules there to release the tests as public domain. http://searchfox.org/mozilla-central/source/js/src/tests/Intl/PluralRules/construct-newtarget.js http://searchfox.org/mozilla-central/source/js/src/tests/Intl/PluralRules/pluralrules.js http://searchfox.org/mozilla-central/source/js/src/tests/Intl/PluralRules/select.js http://searchfox.org/mozilla-central/source/js/src/tests/Intl/PluralRules/supportedLocalesOf.js
Gerv, can you confirm that it's ok for us to switch? I'm the author of three of the four of those files. Andre Bargull (NI'ed as well) is the author of the fourth.
I'm totally fine with releasing the tests I've written under public domain.
Thanks for this helpful change!
With everyone's permission, sure. Gerv
Assignee: nobody → gandalf
Status: NEW → ASSIGNED
Gerv - can you rubber stamp the patch pls? Or redirect to someone who can
Comment on attachment 8886507 [details] Bug 1380931 - Switch Intl.PluralRules tests to public domain. https://reviewboard.mozilla.org/r/157322/#review163770 rs=me given comments 1, 2, and 4.
Attachment #8886507 - Flags: review+
Attachment #8886507 - Flags: review?(gerv)
Thanks Till! :)
Pushed by firstname.lastname@example.org: https://hg.mozilla.org/integration/autoland/rev/c4735d730462 Switch Intl.PluralRules tests to public domain. r=till
Comment on attachment 8886507 [details] Bug 1380931 - Switch Intl.PluralRules tests to public domain. Er, hang on. You wanted these to be Public Domain; why did you not use the Public Domain dedication license header from the standard Mozilla header boilerplate page? https://www.mozilla.org/en-US/MPL/headers/ -> /* Any copyright is dedicated to the Public Domain. * http://creativecommons.org/publicdomain/zero/1.0/ */ Please can you update the tests to use it? "No header" != "public domain". Gerv
Attachment #8886507 - Flags: review-
(In reply to Gervase Markham [:gerv] from comment #11) > Please can you update the tests to use it? "No header" != "public domain". I thought test code without an explicit license header was implicitly public domain per https://www.mozilla.org/en-US/MPL/license-policy/#Licensing_of_Mozilla_Code?
That's a grandfathering provision, and deals with the situation when people forget; it doesn't mean we should intentionally generate more code whose licensing status is not recorded within itself. Perhaps I need to update the policy to make that more clear. Gerv
(In reply to Gervase Markham [:gerv] from comment #13) > That's a grandfathering provision, and deals with the situation when people > forget; it doesn't mean we should intentionally generate more code whose > licensing status is not recorded within itself. Perhaps I need to update the > policy to make that more clear. > > Gerv Not requiring tests to have a license header is common practice in SpiderMonkey. This precedes your official sanctioning of header-less tests, but as of 2014 we do have that sanctioning: https://groups.google.com/forum/#!searchin/mozilla.dev.planning/cc0%7Csort:relevance/mozilla.dev.planning/7zSiApMl1tM/mQQzsGbYpo4J If you wanted that to *only* sanction grandfathered use of header-less test files, then you should send a followup clarifying that: I'm pretty sure nobody understood it that way. I would also ask you to consider the trade-offs and really only make this policy change (which for practical purposes it is, given how your 2014 message was understood) if there are good reasons for it: given that we have a lot of these tests and that as you point out a simple check of the date a test landed at can verify the license a test is under, it's not clear to me what advantages this change would have.
Hmm. Well, OK :-) I said at the time: "We would encourage but not require people to use the CCO declaration." Mozilla's general policy is that per-file licensing information is best practice: http://lu.is/blog/2012/03/17/on-the-importance-of-per-file-license-information/ I agree this is perhaps a bit less important for tests than for code we actually ship to end users. Gerv
You need to log in before you can comment on or make changes to this bug.