Subject: RE: License for the Service Data Objects JavaDoc and Interface Definition files
From: "Lawrence Rosen" <lrosen@rosenlaw.com>
Date: Thu, 26 Jan 2006 20:04:45 -0800

Question sent originally to license-discuss@opensource.org:
> This license states:
>  
> Permission to copy, make derivative works of, and distribute 
> the Service Data Objects JavaDoc and Interface Definition 
> Files files in any medium without fee or royalty as part of a 
> compliant implementation of the Service Data Objects 
> Specification is hereby granted.  
>  
> My question is who determines what is a "compliant 
> implementation"?  What is the remedy if the implementation in 
> which it is used turns out to be
> non-compliant?    

Hi David,

Usually independent standards organizations publish such specifications. Is
this the document relating to the "IBM and BEA Joint Specifications"
published at
http://www-128.ibm.com/developerworks/library/specification/j-commonj-sdowmt
/? (I did a Google search on your email subject and that was the first hit.)
If so, I presume that IBM and BEA set their own "standard" and their
published specification itself will define conformance. Do they define
conformance in their documents or on their website anywhere? If not, it
might be safest to assume that they would define compliance as "an exact
implementation of the specification." However, is such a requirement for
"conformance" logically consistent with the permission to create "derivative
works?" As your question indicates, the license terms (as you quoted them)
are a little sloppy.

I've argued before that limitations on creating non-conforming
implementations of published industry standards cannot be based on copyright
as long as those modifications are for *functional* effects, because
function cannot be restricted through copyright. (See 17 USC 102(b), and
take a look at cases on the "merger doctrine.") Some lawyers don't agree
with me, as usual. :-) So we have to leave that for possible litigation and
a court decision someday. 

We all agree, though, that patents can prevent non-conforming
implementations, if a patent owner is willing to sue for infringement. So
patents are the second problem with the license you quoted. If anyone (IBM,
BEA?) has patents reading on the standard, they are not explicitly granting
you a patent license to make, use, sell, have made, offer for sale, or
import software (copies *or* derivative works) embodying those patents.
That's another reason why I think the license terms you quoted are sloppy. 

This is a hot topic in standards organizations nowadays. For example,
there's an effort underway in IETF to revise their IP policies to license
copyrights and patents such that implementers are free to create derivative
works, but some of the industry participants on the IETF IPR discussion list
are resisting the pleas of the open source advocates there. Unless standards
organizations take steps to license copyrights and patents for both copies
and derivative works, and list in their licenses all the rights associated
with copyrights and patents, we may have to assume that some copyright and
patent owners allow only exactly conforming implementations. That's not open
source.

Implementers of open standards should demand more rights, and more clarity,
from their licensors.

/Larry Rosen

P.S. I'm cross-posting this email to the IETF ipr-wg@ietf.org email list,
although I doubt the particular standard you referenced is one of theirs.
The topic is current, though. I'm also cross-posting to
discuss-general@open-bar.org, a new discussion list for attorneys interested
in open source issues. I put their email addresses in BCC so that each
discussion group stays separate in replies.

** Lawrence Rosen
** Rosenlaw & Einschlag, technology law offices
** Stanford University School of Law, Lecturer in Law
** 3001 King Ranch Road, Ukiah, CA 95482
** 707-485-1242  *  fax: 707-485-1243
** Author of "Open Source Licensing: Software Freedom
**    and Intellectual Property Law" (Prentice Hall 2004)
** [www.rosenlaw.com]  

> -----Original Message-----
> From: David Goossen [mailto:goossen@roguewave.com] 
> Sent: Thursday, January 26, 2006 3:02 PM
> To: license-discuss@opensource.org
> Subject: License for the Service Data Objects JavaDoc and 
> Interface Definition files
> 
> This license states:
>  
> Permission to copy, make derivative works of, and distribute 
> the Service Data Objects JavaDoc and Interface Definition 
> Files files in any medium without fee or royalty as part of a 
> compliant implementation of the Service Data Objects 
> Specification is hereby granted.  
>  
> My question is who determines what is a "compliant 
> implementation"?  What is the remedy if the implementation in 
> which it is used turns out to be
> non-compliant?    
>  
>  
> David F. Goossen
> Rogue Wave Software
> a QUOVADX(tm) division
> 5500 Flatiron Parkway
> Boulder, CO 80027
> Phone:  303-545-3166
> Fax:  303-545-3061
>  
>