A dual license system for code libraries?

Carsten Agger agger at modspil.dk
Tue Feb 28 12:22:58 UTC 2017

On 02/26/2017 08:26 PM, Agner Fog wrote:
>> If our business models never involve releasing under a proprietary
>> license, we're not contributing to the proprietary software economy,
>> and that's that.
> Acturally, we are contributing indirectly if an open source library is
> used in proprietary software.
I just re-read Stallman's essay on selling exceptions. The gist of his 
argument is that selling exceptions to a copyleft program or library is 
no worse than releasing the same program or library under a permissive 
license, which is not wrong in itself. As so often, he nails it. Selling 
exceptions may be undesirable, but it's not an unethical thing to do as 
long as we're not talking about proprietary extensions etc.

On the other hand, free software is so ubiquitous as to be present in 
practically all larger proprietary systems. The Windows networking 
protocols are, for example, built on the Berkeley socket implementation 
IIRC. Free software is inevitable these days as it's present everywhere 
in our infrastructure, and in that way it's contributing to proprietary 
software all the time (as well as to all other software-supported human 
>> Honestly, I think we should focus speculations on business model on
>> providing software under free licenses. For that reason, I'm
>> personally against the selling of exceptions.
> Yes, we need to focus on how to finance open software. But to me, the
> word "business model" is at odds with the ideal of free software.

Not at all! The ultimate goal of the free software movement is that 
*all* software should provide its users the four freedoms. Proprietary 
software should become a thing of the past. This is a lot to do with the 
wish for a free society - Stallman's analysis of the problem indicates 
that *iff* we choose to build our society and infrastructure on 
software, software freedom is a  necessary (but nor sufficient) 
condition for society to remain free. A *necessary* condition, to repeat 

Now, if all software is to be free software, it obviously needs to come 
from somewhere. Large organisations need suppliers they can rely on to 
supply support and to develop custom software. From the very moment 
(2005) I took a serious interest in free software and understood that 
the state of GNU/Linux was such that it can conceivably be used 
everywhere and there's no need at all for Windows or other proprietary 
systems, I realized that for this to become true, political lobbying is 
not enough - we need companies to produce and support the software that 
everybody is going to need. Public ("open source") projects manned (more 
or less) by volunteers and funded (more or less) by NGO's can do a lot, 
but municipalities, goverment agencies, ministries, huge companies etc. 
need business levels of support and development of free software.

Thus, business models for co-op's and companies and NGO's supplying free 
software is sorely needed. Living off the charity of advertising 
companies and proprietary software vendors such as Google, IBM and 
Facebook (as many large projects do) is *not* going to be enough. So, 
business models are important, are needed and are *in no way* at odds 
with the ideals of free software.

Stallman also has an essay on this, on "Selling Free Software": 


Disclaimer: As previously noted on this list, I work as a developer in 
Denmark's largest free-license-only "open source" company. So I'm 
looking at this as someone who's been selling free software for a living 
the last 5+ years. However, as I said, the need for such companies to 
exist was one of the first issues I noted when I started my free 
software activism.

More information about the Discussion mailing list