[Savannah-register-public] Re: Fwd: [task #7064] Submission of qemuGL

[Sylvain Beucler]

Hi,

Sorry for the delay in answering your mail, I had a 'burnout'
yesterday and couldn't do anything Savannah-related.

> > I don't like what he says. He says that at qemu they don't track
> > authors per file and that it's ok. But at Savannah, we are here to
> > teach recommended copyright good practices - we're teachers, not
> > students ;)
> 
> OK, I see that the ChangeLog file in qemu tracks authors on
> per-feature/patch basis, but it should not be a problem (for qemu
> community) to track changes per-file in the scope of this GL patch
> project.

The ChangeLog is pretty rudimental, because we don't see which files
are modified by whom. Moreover, authors are not necessarily the
copyright holders (eg it may be their employer), so both need to be
specified. Last, if there are multiple contributors to kqemu.c, then
"Copyright (c) 2005 Fabrice Bellard" is just wrong - unless they
assigned their copyright to Fabrice Bellard.

I think we can essentially base our judgement on:
http://www.gnu.org/prep/maintain/html_node/Recording-Contributors.html


> > So I think he needs to modify files that are non-trivially changed (if
> > he modified a couple lines, no problem, but if the change is
> > consequent (is it?) then he needs to add his copyright notice). The
> > goal is that when I copy code from a project, I copy the file's
> > copyright notices along.
> >
> > What do you think?
> 
> May be we can convince him at least to put copyright notices in every
> modified file in the form of "Copyright (C) 2007 Even Rouault et al
> (see GL_AUTHORS for the complete list of contributors)"...  Is this
> acceptable practice? :-)
> 
> This way the project can give credits to most contributors and prevent
> copyright notices bloat the qemu community afraid of.

I don't think it would be this bloated really.

Our recommended practice is still to place appropriate copyright
notices in all files. Should they use a proper AUTHORS files as
described in the link above, I think I wouldn't argue though.

Our review is fruitful here, because we can teach the submitter about
recommended practices about copyright management. It is thus important
to insist about them.

-- 
Sylvain