The GNU Affero General Public License  and "DRM"

There is an all too familiar dynamic in conversations about copyright. If you disagree with a proposed solution to a problem, such as the unintended consequences that it can have, then the proponents presume it is because you disagree that there is a problem. I am very familiar with this dynamic when speaking to copyright holders about “DRM” or technological measures as they believe any opponent to the use or legal protection of these measures must be apologists for copyright infringers.

I observed the same dynamic last weekend at the Ontario Linux Fest when discussing some concerns I have with the Affero General Public License (AGPL). The conversation was with Bradley M. Kuhn, one of the people behind the license.


Bradley assumed that my concerns with the AGPL were because I disagreed with the copyleft concept, rather than being worried about a potential unintended consequence of this specific license. He suggested that people concerned about the AGPL were part of the BSD “Freedom to” camp (IE: people who think that making proprietary derivatives should be a protected freedom), rather than the FSF “Freedom from” camp (people who believe that recipients of derivatives should also have their free software freedoms protected, and that proprietary derivatives should not be allowed).

The Problem!

Here is the problem they are trying to solve. The GNU General Public License is intended to create an environment where people share and share alike. It is what is called a Copyleft license which uses copyright law to require that all modified and extended versions of the program be Free/Libre Software as well.

These licenses have worked well and have been proven to be enforceable against companies who try to embed modified versions of this software in their hardware/software products without releasing the corresponding source code. That is, it worked well until it became obvious that a growing amount of software was heading into the cloud where the software wouldn’t be copied or distributed to end users at all (software users wouldn’t be ‘recipients’ of the software code, just the services offered by that software).

Private modification and use of GPL software was something that was explicitly permitted in the past. There was now a perception that people were making modifications to the software and offering services without there being any way for the software authors to actively encourage (well, require) that the corresponding source code be made available. I don’t know how common this scenario was, but that is beside the point. Copyright and related licensing is very often about the perception of harm and the perception of a lack of respect for the creators, rather than actual measurable harm.

The proposed solution

Section 13 of the AGPL has details of what the license requires. Under “How to Apply These Terms to Your New Programs” the following simple language is used:

If your software can interact with users remotely through a computer network, you should also make sure that it provides a way for users to get its source. For example, if your program is a web application, its interface could display a “Source” link that leads users to an archive of the code. There are many ways you could offer source, and different solutions will be better for different programs; see section 13 for the specific requirements.

The AGPL is a tool encouraged by a larger movement which has been encapsulated in the Franklin Street Statement on Freedom and Network Services. And there are some pretty key people/projects involved such as Evan Prodromou who is (among other things) the primary author of Laconica and host of which is an Open Source/Standards replacement for Twitter. There is even an Open Software Service Definition.

I am a strong supporter of the other aspects of the Franklin Street Statement. That said, I have a concern with the AGPL itself and potential unintended consequences.

My concern with the AGPL

My concern is that using licenses which trigger on private modifications rather than distribution/communication of software will encourage an expansion of copyright law to more strongly regulate private activities. It may also discourage activist movements that the FSF is otherwise aligned with that are trying to reduce the impact of copyright on private activities.

It is best to understand how copyright works to make sense of the problem. Without copyright, there is little which restricts what someone can do with expressions of ideas (also often called ‘works of the mind’ or just ‘works’). Copyright then lists a series of activities which require the permission of the copyright holder to do with a given work (lawyers call this the “bundle of rights”, as it is not a single thing). In order to make copyright reasonable in a variety of cases, there are then further exceptions in circumstances that would otherwise require permission.

I discussed the description of the differences in the main exceptions in Canadian and US copyright law in “Canadians fed US-style copyright legislation? I wish!“, so I won’t repeat them here. The main thing to take away from the US Fair Use regime is that what makes it able to more easily adapt to changes in technology than Canadian law is the fact that US law is not tied down to a specific activity, but is intended to be interpreted by the courts to introduce the concept of “fairness”.

What is considered “fair” changes over time, and will be modified by what the courts observe people doing.

In the definitions in the AGPL it includes, “To ‘modify’ a work means to copy from or adapt all or part of the work in a fashion requiring copyright permission, other than the making of an exact copy.”

Canadian copyright law says that producing, reproducing, and publishing the work “or any substantial part thereof” requires permission. It is not clear, however, whether modifications made that are not produced, reproduced, or published require permission. These are not the only activities which require permission, but as you go through each it is not clear that private activities are clearly regulated by copyright. In the case of “communicate the work to the public by telecommunication” it is clear that communicating the work privately (IE: wiring/networking within ones home) is not regulated by those specific clauses.

While it is not clear that private activities are regulated, many lawyers have assumed that they are regulated. For instance, installing software on a computer’s hard disk is presumed to be a reproduction of the software which requires permission. Some lawyers (and I believe some caselaw, but I don’t have links handy) suggest that even the loading of the software from fixed media to RAM in order to execute the software requires permission.

Under this interpretation, any activity involving a computer (which requires copies to be made in order to function) are presumed to be regulated. This has caused interpretations of copyright, copyright licensing and contracts, and the law itself to continue to move into activities which in an analog world would be considered unregulated by copyright.

An increasing number of people consider this to be “unfair” and want to clarify the law to carve private activities out of the law.

The US courts have already said that time shifting of broadcast television programming is fair use. This is copying that would otherwise be regulated, but is considered fair. In order to make copyright reasonable we should really be extending this to time, device, and format shifting, backup, and other activities as long as they remain private. Canada allows people to adapt software for compatibility, and to make backups (30.6), but that only applies to computer software and not any of the data also stored on our computers that require compatibility/interoperability and backup.

As we work to try to carve out private activities from copyright, whether digital or analog, we will come across a political problem.

Politicians like to figure out who the extremes of the pro-and-con side of any debate is, and then try to come up with some middle ground. In the software field they will see companies like Microsoft as representing one side, where Microsoft wants as many activities as possible to be regulated by copyright, where they then offer narrow permissions in their End User License Agreements (EULA) with some pretty draconian conditions.

The Free Software Foundation (FSF) will easily be seen as the “opposite” end of the spectrum, given they are trying to set up a regime where broad permissions are granted to recipients of software with very simple conditions.

If what politicians see as both extremes of this debate are seen to be commonly using license agreements that trigger on private activities such as private modification, then attempts to remove these private activities from copyright will be dead in the water.

I remember reading many years ago a slide deck by Barry B. Sookman which suggested that the Copyleft licensing movement will lead this community to be encouraging expansion in copyright law. This would contradict the goal of reducing copyright regulation which would itself better protect the rights of software users. (I just searched and could not find it again. If anyone has a link, please add to the comments) I remember being offended when reading this deck, thinking that this lawyer was somehow misinterpreting what was going on, or worse — promoting the special interests of proprietary software vendors (his clients). It may turn out that Mr. Sookman was right, and that the AGPL represents the acceptance of harmful potential copyright expansion as an unintended consequence to theoretically stopping a loophole in existing licenses.

Parts of the FLOSS community seem to have walked themselves into a similar political problem as the creator/content industry proponents of DRM. The content industry sees a “loophole” in the current content distribution system that enables private citizens to easily widely distribute/communicate perfect copies. Their proposed solution of locking down communications devices such that they are not under the control of the owner has great unintended consequences that will ultimately hurt creators and the content industry as much (if not more) than sectors like the FLOSS sector. The FLOSS sector is clearly opposed to DRM. It is important for people to realize that the primary target for recent copyright reform (DMCA, 1996 WIPO treaties, C-60, C-61) is not copyright infringers, but “enablers” (technology and services) which are seen as that “loophole”. It is also important to realize that individual creators need control over communications technology in order to be creative, and these locks prevent creativity far more than they put up a speed bump against copyright infringement.

Similar to the DRM situation, I believe the harm of the unintended consequences of the AGPL will far exceed the potential benefits to the FLOSS community. I will be avoiding this license and others like it, and I hope that people who believe strongly in protecting software freedom will do the same.

Am I overly concerned about the unintended consequences, or has the FSF lost its way? Let me know in the comments how you feel about this potential issue.

Russell McOrmond is a self employed consultant, policy coordinator for CLUE: Canada’s Association for Free/Libre and Open Source Software, co-coordinator for Getting Open Source Logic INto Governments (GOSLING), and host for Digital Copyright Canada.

Related Download
CanadianCIO Census 2016 Mapping Out the Innovation Agenda Sponsor: Cogeco Peer 1
CanadianCIO Census 2016 Mapping Out the Innovation Agenda
The CanadianCIO 2016 census will help you answer those questions and more. Based on detailed survey results from more than 100 senior technology leaders, the new report offers insights on issues ranging from stature and spend to challenges and the opportunities ahead.
Register Now