www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lawrence Rosen" <lro...@rosenlaw.com>
Subject RE: 3rd Party License policy, the board, and the term "FREE"
Date Wed, 13 May 2015 16:45:50 GMT
I much value William Rowe's response below. Understanding the goals of "a firmly rooted 'first
decision' made by the Foundation's creators" is essential whenever those first decisions are
later revised (or rather, proposed to be revised).


I'm glad that this is a public Apache list. I look forward hearing from developers,  particularly
companies building products on top of excellent Apache software, to see if those assumptions
about reciprocity still hold nowadays. Those companies seem to be adopting and combining Apache,
Linux and Eclipse and much other FOSS software without fear of reciprocity.


Speaking of reciprocity, you wrote:

> ... particularly in the case of viral-reciprocal licensing ...


I promise not to use the word "sinful" again as long as you stop calling the GPL a "viral"
license. It isn't, according to its authors.





From: William A Rowe Jr [mailto:wrowe@rowe-clan.net] 
Sent: Wednesday, May 13, 2015 9:16 AM
To: legal-discuss@apache.org; Lawrence Rosen
Subject: Re: 3rd Party License policy, the board, and the term "FREE"


On Sun, May 10, 2015 at 2:16 PM, Lawrence Rosen <lrosen@rosenlaw.com <mailto:lrosen@rosenlaw.com>
> wrote:

Shane Curcuru also wrote:
> ensures our users don't face any other restrictions when using or redistributing
> derivative works of our software

To be precise, no OSI-approved open source license /restricts/ anyone from
creating and distributing derivative works. Some licenses impose
/conditions/ of reciprocity. There is nothing sinful about that.


Nope, there are no value judgments at work here.  There is a specific a firmly rooted 'first
decision' made by the Foundation's creators, which elected that the first freedom would be
freedom of developers to do what they wanted with the code that would be created at the ASF,
without additional /restrictions/, /burdens/, or /reciprocity/.  In this sense, the ASF sees
the user and chief beneficiary of the works created here as either a user or developer of
a larger work.


The OSI-approved licenses that require /reciprocity/ do add an additional /burden/ to the
developer, one which is cheerful imposed by other, equally legitimate open source foundations.
 This is so because those foundations or individual developers see the user and chief beneficiary
as the user who consumes the software on their machine.  [Or in the case of the AGPL - the
user who consumes the software on any remote machine, as well.]  That group of licenses adds
specific rights for the user over the rights of the developer who consume the code in question.


There are benefits of both models which we do not need to debate here, and as your invocation
of purity v.s. sin implies, those distinctions usually descend into a religious argument.
 I don't think that has a place here at this forum, or at least it's a thread I'll tune out.
 You may claim this is not a burden, but I'd find that pretty disingenuous of you with respect
to consumed code, particularly in the case of viral-reciprocal licensing.


That said, there are things which do not translate to machine code, which are produced under
various reciprocal licenses, which can hardly be said to be an imposition on the developer.
 The example of a dictionary of words is one of them.  A .css presentation style sheet is
another.  Most of these either have no mapping to 'compilation' , if they are compressed/indexed
for consumption, they typically can be reverse engineered with well understood tools for that
purpose (extracting database records, for example).  Or these are by definition always presented
in their source forms. Provided there is no viral-reciprocity clause affecting the ASF works,
combining such works pose no real licensing challenges.




View raw message