[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: License query for mrouted

> Christoph writes:
> > As I understand it, they permit distribution of devired works, but they
> > do not speak about REdestribution of the original. I think, they take
> > this as granted.
> They may take it for granted, but we can't.  We can only do what they have
> given us permission to do.
> -- 
> John Hasler
> john@dhh.gt.org (John Hasler)
> Dancing Horse Hill
> Elmwood, WI
> (Sent to debian-legal, Cc to package maintainer) 
> I found this package in Incoming.  I'm not sure if we can distribute 
> it; clause 4 is a bit funny.  Also, it seems to permit distribution of 
> derived works BUT NOT verbatim copies ("nontransferable license"), 
> which is weird. 

Earlier, Richard Braakman <dark@xs4all.nl> had written;

> The mrouted program is covered by the following license.  Use of the 
> mrouted program represents acceptance of these terms and conditions. 
> 1. STANFORD grants to LICENSEE a nonexclusive and nontransferable license 
> to use, copy and modify the computer software ``mrouted'' (hereinafter 
> called the ``Program''), upon the terms and conditions hereinafter set 
> out and until Licensee discontinues use of the Licensed Program. 


> 3. LICENSEE may copy the Program and may sublicense others to use object 
> code copies of the Program or any derivative version of the Program. 
> All copies must contain all copyright and other proprietary notices found 
> in the Program as provided by STANFORD.  Title to copyright to the 
> Program remains with STANFORD. 
> 4. LICENSEE may create derivative versions of the Program.  LICENSEE 
> hereby grants STANFORD a royalty-free license to use, copy, modify, 
> distribute and sublicense any such derivative works.  At the time 
> LICENSEE provides a copy of a derivative version of the Program to a 
> third party, LICENSEE shall provide STANFORD with one copy of the source 
> code of the derivative version at no charge to STANFORD. 

Paragraph 1 lets us use, copy and modify the Program, but it doesn't
let us distribute it. Paragraph 3 is what allows us to distribute the
program. So "nontransferable" means that our right to use, copy and
modify is not something we are allowed to transfer to someone else,
although we may, pursuant to paragraph 3, sublicense others to use
object code copies of the Program or derivative versions. Anyone who
gets the Program from us has a sublicense from us, not a direct
license with Stanford.

Unfortunately, paragraph 3 only lets us sublicense others to use
object code copies of the Program; it does not allow us to let others
modify, distribute, or copy the program. As I understand it, rights
not explicitly granted to a licensee are reserved by the grantor, so
people who get any version of the Program from Debian are only allowed to
use it, not to make their own derivative versions even if they were
to send the modifications to Stanford. As a practical matter, it seems 
highly unlikely that Stanford is going to care whether someone got the 
program from Debian or from someone else, so long as they comply with
the license as if they got it directly from Stanford, but that's not
what the license says.

As a further consequence of this two level scheme, Debian is allowed
to distribute the Program at all only if we get the program directly
from Stanford (or from someone who got the program with a more liberal
license than this one).  If we get the program from someone second
hand under this license then we are that person's sublicensee, granted
only the right to "use object code copies" of the Program or a
modified version, not to make further modifications or to distribute

As someone else pointed out, paragraph 4 is a little odd. I think it's
ambiguous as to whether or not we are supposed to provide a copy of a
derivative version to Stanford each time we distribute it, even after
giving Stanford that identical source code. The fact that paragraph 4
explicitly says that the licensee shall provide "one" copy of the
source code would seem to indicate that it's not necessary to provide
an additional copy the next time that we distribute an identical
derivative version.  Moreover, I think that any court would apply a
reasonableness standard here to say that we do not need to provide
multiple copies of the same thing since that would not serve the
purpose outlined in this license (see paragraph 2 of the license). [To
say that a court would rule x is to say that the law is x, so this is
also my opinion of the law on the matter.]

Now the interesting question here is whether this is, or should be,
free for the purposes of Debian's purposes. Let's assume for the
moment that the weird two level scheme were eliminated so anyone could
distribute and modify the program as if they had gotten it directly
from Stanford and that it were made clear that it was only necessary
to send a copy of the modified software to Stanford the first time you
distributed a new modified version. Would that be an onerous
requirement? It's not a restriction at all on use of the program, nor
on modifying it for your own use. In many ways it seems significantly
less onerous than patch clause type requirements, since it doesn't
interfere with the ability to start a whole new forked version, so
long as you keep Stanford up to date on the forked version whenever
you release it. The only significant trouble I see with it is how to
deal with CVS trees -- what does it mean to "provide" a copy to a
third party? If someone checks out a modified version has the CVS
maintainer or the modifier of the code provided a copy to a third
party?  Is it sufficient to allow Stanford access to the CVS tree in
order to provide a copy to Stanford?  

More generally, so long as the definition of "provide" to Stanford
isn't too onerous, then it doesn't seem too problematic. In fact,
since the same verb is used both to describe how it is a third party
gets a copy and how Stanford gets a copy, it seems that simply
allowing Stanford access to the source code in the same manner as
anyone else gets the source code, without charge, then paragraph 4 has
been satisfied.

As it stands, however, the license seems clearly non-free.

> 2. LICENSEE acknowledges that the Program is a research tool still in 
> the development state, that it is being supplied ``as is,'' without any 
> accompanying services from STANFORD, and that this license is entered 
> into in order to encourage scientific collaboration aimed at further 
> development and application of the Program. 


> By way of example, but not limitation, STANFORD MAKES NO REPRESENTATION 
> for any liability nor for any direct, indirect or consequential damages 
> with respect to any claim by LICENSEE or any third party on account of or 
> arising from this Agreement or use of the Program. 
> 6. This agreement shall be construed, interpreted and applied in 
> accordance with the State of California and any legal action arising 
> out of this Agreement or use of the Program shall be filed in a court 
> in the State of California. 
> 7. Nothing in this Agreement shall be construed as conferring rights to 
> use in advertising, publicity or otherwise any trademark or the name 
> of ``Stanford''. 
> The mrouted program is COPYRIGHT 1989 by The Board of Trustees of 
> Leland Stanford Junior University. 

Reply to: