GPL with the Classpath exception - clarification needed

[Wilson, Andrew]

Roger Fujii [mailto:rmf at lookhere.com] wrote:

>Wilson, Andrew wrote:
>> Philippe Verdy [mailto:verdy_p at wanadoo.fr] wrote:
>>
>>> But remember that we were speaking about Java when speaking about the
>>> Classpath exclusion. Where does the LGPL affect Java-written libraries so
>>> that it would not work or would render an application using it fully GPL
>>> licenced with all its requirements?
>>>
>> OK. If you derive a Java class from a base class which is LGPL licensed,
>> under what rationale do you claim your derived class is not also a
>> derivative work under copyright, and is not subject to LGPL?
>>
>This came up in the long past.  On reading the LGPL, you are correct that
>the text AND FSF's definition of 'derived' means if you use the class,
>it must be
>LGPL also.   However, I think my obstinance on the matter got FSF to publish
>this:
>    http://www.gnu.org/licenses/lgpl-java.html
>
>Given what is said there, "work based on" in the LGPL cannot mean the
>same thing
>as that phrase used in the GPL, but for some reason, this oversight
>doesn't seem to
>bother too many people.

Roger, thanks for providing the link.  This document is a real head-scratcher,
since, as you note, it proposes an alternate definition of what is
a "work based on" LGPL code versus what I understand as the standard FSF
interpretation of derivatives under GPL.

A cautious person would probably prefer, if possible, a license with a
black-letter permission that use of a class library does not create a
derivative work subject to the copyleft license of the library itself.
This leads back to GPL+classpath, IMO.

I am not a lawyer.  This is not legal advice.  Professional driver on closed
course.  Not available in Alaska or Hawaii.  Claims have not been approved
by the FDA.  ;-)

Andy Wilson
Intel open source technology center