On Mon, Aug 8, 2016 at 8:51 PM, Justin Mclean <jus...@classsoftware.com> wrote: > Hi, > >>> if you're saying that we need to slap an ALv2 license header on something >>> like shm.c -- I don't feel comfortable doing that > > Perhaps ask yourself why that is? Is it because the > licensing/copyright/provenance is unclear?
No. All of the above is clear: it came from PG, we didn't modify it and we'd like to keep it that way. > Does the files version control history tell you anything? No. And that's an additional reason I'd prefer to vector anyone doing software archeology back to the real source control -- the one of PG project. > I know in some cases here we’re dealing > with files 10+ years old so that may be difficult. Perhaps list all files > somewhere that you are unsure > of but likely to be Apache or other compatible license? Well, I don't think there's ever a 100% assurance in IP matters, but... here's what we know AND here's what we would like to advertise to the consumers of HAWQ: A certain set of file (how we advertise the filenames is TBD, but likely in LICENSE)... 1. ...came from PostgreSQL project version 8. With 8.1.0 being a bulk of the import, but with a few files that came from older PG releases 2. ...may have been initially released under whatever license, but then were made available by the PostgreSQL project under the PostgreSQL license (which is a BSD derivative and compatible with ALv2) 3. ...to the best of our software archeology analysis we can trust PostgreSQL community on the statement in #2 That's our intent which I think is relatively problem free, so the only question is can we express this intent in the best possible way. Thanks, Roman. --------------------------------------------------------------------- To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org For additional commands, e-mail: general-h...@incubator.apache.org