aries-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Guillaume Nodet <>
Subject Re: [DISCUSS] [POLICY] IP tracking for OSGi APIs
Date Tue, 24 Jan 2017 11:04:44 GMT
My point is that there is a source identified, preferably public if
possible, private if not.

Of course if the API is not public, we can refer to the private commit so
that  PMC members which are not OSGi members can be confident about the IP
I think it depends on what stage the API is.  For example, consider the
work Ray is working on for CDI, I don't think the API is in the OSGi git
repo yet, so either the API is committed there first, or the commit can
refer to Ray's github repo directly.  If a spec has been published
officially by the OSGi Alliance, we can refer to a public source too.
Given the API code becomes public at the time it's given to the ASF, I
didn't thought that would be a problem to have it in a public repo in the
first place, but if you're saying it is... I'm fine with that.

So what about:
  "All commits that includes API code from the OSGi Alliance are done in
separate commit and include a reference to the public (preferably) or
private (if not public) source  where the code comes from."

Does that look better to you ?

2017-01-24 11:42 GMT+01:00 Carsten Ziegeler <>:

> Guillaume Nodet wrote
> > 2017-01-24 0:02 GMT+01:00 Carsten Ziegeler <>:
> >
> >> Guillaume Nodet wrote
> >>> Well maybe it should, but again, I don't think it has always been done
> >>> correctly in the past...
> >>> Hence this proposal to discuss what options we have to actually
> correctly
> >>> implement it.
> >>>
> >> As said, it must be option #2, always and I'm not aware of any case
> >> where it hasn't.
> >>
> >
> > Cool.  Then it should be very easy to simply add the source of the api in
> > the commit log, is it ?
> > David just did it yesterday, it does not seem very complicated:
> >
> >
> 83688a6325
> > Where's the problem ? I really don't understand.
> >
> What David did is exactly what I said is the way to do it. But still the
> OSGi API is
> not available in public, so unless you have access to the private OSGi
> git repo,
> you can't check whether the reference given there is correct or not.
> But you had exactly this requirement in your original proposal.
> As Felix proposed, let's extend our policy with what we can do and we're
> all happy.
>  Carsten
> --
> Carsten Ziegeler
> Adobe Research Switzerland

Guillaume Nodet

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message