incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brian Behlendorf <br...@collab.net>
Subject Re: several messages
Date Mon, 12 Jul 2004 21:05:37 GMT
On Mon, 12 Jul 2004, Rodent of Unusual Size wrote:
> for another thing, the asf *does* have a higher standard: we require
> transfer of ownership and ip.  it makes perfect sense to me that a
> company may be willing to transfer those to the asf, but *not* be
> willing to throw them into the open -- which is the potential if
> the code is exposed before the acceptance and transfer.

Well, this worries me some more.  The simplest thing a company could do to 
meet the requiremet is put the code on their own site as a downloadable 
tarball. In such a scenario, the company can elect whatever license they'd 
want, Apache or otherwise; and they'd call themselves the copyright owner. 
The only risk to the company in doing so, compared to giving it to Apache, 
is that any third-party claims others might make (based on copyright, 
patent, or trademark) would be against the company, rather than to the 
ASF.  If a company isn't willing to put the code base out themselves under 
their own ownership, but would rather it be (C) ASF, that leads me to 
wonder about what liability the company is attempting to avoid by doing 
so.  It may be paranoia, but seeing a company willing to put the code out 
under an open source license with a (C) to them does a lot to quell 
concerns about whether the codebase is IP-clean.  We're not a 
code-laundry, and I don't want to even allow for the possibility that we 
might be abused as such.

On Mon, 12 Jul 2004, Leo Simons wrote:
> One of the key requirements is that there's enough people interested in a 
> project and willing to help. Ensuring that will usually be very difficult 
> without making code available, but not always. Consider geronimo. There was 
> no code, but we accepted it anyway (IIRC the board did, even). Good decision.

Not strictly true, there was a lot of pre-existing code, and it was the 
quality of that code that I think excited a lot of people about the 
project.

On Mon, 12 Jul 2004, Noel J. Bergman wrote:
> That said, if there are ASF Members who want to incubate a project, I would
> prefer that we give them a chance to succeed, even if the endeavor
> ultimately fails.  I don't share the concern that "it's code that the ASF
> will become responsible for maintaining."  I don't see that we have an
> obligation to maintain a failed project, although I would put the CVS
> tarball in the archives, or preferably just do an svn remove, as an audit
> trail.  Nor do I believe that there is any requirement for the Incubator to
> expend extraordinary efforts to make a project successful beyond that which
> ASF Members are willing to contribute as mentors.

To be clear, the ASF takes a legal risk with every line of code it 
publishes to the public.  Projects that go quiescent are a liability; as 
are failed projects.  There's also a PR problem for the ASF if there are 
too many failed projects.  So some care should be taken in deciding what 
to admit into the incubator; less so than to graduate, of course, but care 
nonetheless.  We should probably also do more active garbage collection on 
established projects, but that's another topic for somewhere else...

> An IP pre-review could cause a problem.  IP made available for review would
> have to be free of encumbrances.  We do not want a situation where people
> who have reviewed it become tainted.  Certainly that can be the case if the
> license were a proprietary one, but it does not seem to matter if the
> license is an OSI-approved one or not.  Claims can be made based upon (L)GPL
> as easily as upon more classically recognized proprietary licenses.  So, no,
> I do not agree that "any legal mechanism that provides for anyone in the
> community to conveniently read the source should be acceptable, even if the
> license restricts distribution, for instance."

Good point; that suggests the requirement be to be very clear that review 
of the code places no encumbrances.  This could be as an additional 
freedom granted above and beyond the (L)GPL, so it doesn't have to be 
Apache licensed.

 	Brian


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message