Does a GPL API infect its apps?

Ross N. Williams ross at
Thu Oct 21 05:51:34 UTC 1999

At 3:18 PM -0400 20/10/99, John Cowan wrote:
>> I know that GPL does not 'infect' via mere aggregation and that LGPL
>> would cover apps if they were considered 'linked' to the API.
>Vice versa.  The LGPL *does not* infect code that links to it, though it
>*does* infect modified versions of the LGPLed code.  The distinction
>is not always easy to make in particular cases.

I just want to point out that the use of the virus metaphor to describe
the legalities of association in relation to the GPL is inappropriate
because the "infection" does not proceed beyond one level.
   GNU GPL V2: 2.
   These requirements apply to the modified work as a whole.  If
   identifiable sections of that work are not derived from the Program,
   and can be reasonably considered independent and separate works in
   themselves, then this License, and its terms, do not apply to those
   sections when you distribute them as separate works.  But when you
   distribute the same sections as part of a whole which is a work based
   on the Program, the distribution of the whole must be on the terms of
   this License, whose permissions for other licensees extend to the
   entire whole, and thus to each and every part regardless of who wrote it.

It's more of a proximity effect. So we need a new metaphor.

Perhaps we could talk about the GPL Freedom-Force-Field (GPL F^3 ?)
or something like that :-)


Dr Ross N. Williams (ross at, +61 8 8232-6262 (fax-6264).
Director, Rocksoft Pty Ltd, Adelaide, Australia: 
Protect your files with Veracity data integrity:

More information about the License-discuss mailing list