In other words, say there are a small number of source files in a packaged (tarball) work that lack any or clear copyright header, should that be considered a review blocker?
IANAL and being a generally pragmatic fellow, I'd hoped that we could generally give upstreams the benefit of the doubt, for lack of any contrary evidence.
Fwiw, yes, I have a concrete case in mind here, https://bugzilla.redhat.com/show_bug.cgi?id=749885 and yes, I've asked upstream for clarification (without reply so far). I'd also hoped not to have a pkg review blocked semi-indefinitely on something like this though.
What say you?
-- rex
On 12/11/2011 08:13 PM, Rex Dieter wrote:
In other words, say there are a small number of source files in a packaged (tarball) work that lack any or clear copyright header, should that be considered a review blocker?
IANAL and being a generally pragmatic fellow, I'd hoped that we could generally give upstreams the benefit of the doubt, for lack of any contrary evidence.
So, the answer here is yes (with caveats), as long as we have clear evidence that the files are part of a larger work where there is consistent licensing intent.
CAVEATS:
If the files look like they were copied from somewhere else (or we know they were), then we need to clarify the per-file licensing.
If there is no overall licensing statement for the project (e.g. a README stating that the project is BSD), then we need to clarify the per-file licensing.
If there are a mix of licenses in play, and it is a confusing situation where some binaries end up being GPL-incompatible while others are GPL-compatible, then I'd say we need to clarify the per-file licensing.
In any case, we should be trying to get upstream to resolve those issues, even if it isn't blocking inclusion in Fedora.
~tom
== Fedora Project