Value Neutrality and the Ethics of Open Source
Morten @ home
Posted on March 11, 2020
2019 was the year of the “ethical source” licenses – or ‘open source with a moral clause’ licenses. It was also the year many in the open source movement labeled any attempt at adding moral clauses to open source licenses not only made them not open source licenses, but were a dangerous attack on the core principles of open source.
The Ethical Source movement, exemplified by the Hippocratic License, the Anti 996 license etc, has created a deep rift in the open source community; between those who aim to protect the core values of open source ideology as defined by the Four Freedoms, and those who believe granting open source creators the capability to impose moral clauses on the use of their software is essential to ensure software is not used for evil.
The rift has opened because adding moral clauses to a license is effectively limiting the use of whatever the license is applied to, which stands against the core idea of open source as absolute freedom for the end-user.
The Tenuous Position of Value Neutrality
From my perspective this rift exposes what I’ve long felt is a central flaw in both open source licensing and open source ideology: The Four Freedoms concern themselves only with software freedom – as in the freedom of the software itself – but are in reality imposing value neutrality on whatever is licensed under them: The Four Freedoms protect the freedom of the software itself, and do not concern themselves with what the software does or how it is used. As an ideological platform, they are too narrowly scoped to hold the weight of our societal responsibilities as the designers and builders of modern reality.
Let me explain:
If evil people use open source software to do evil things, there is nothing the creator can do about it. Because doing something about it would be limiting the freedom of the software which goes against open source ideology. And in the view of many open source ideologues, it would be morally wrong for a creator to do anything about it because limiting the use, modification, and distribution of software is in itself morally wrong.
This ideological view stems from the original GNU Manifesto:
“‘Control over the use of one’s ideas’ really constitutes control over other people’s lives; and it is usually used to make their lives more difficult.”
“When there is a deliberate choice to restrict, the harmful consequences are deliberate destruction.”
These quotes capture two central ideas of the GNU Manifesto, one I think most of us agree with, another I think most of us take issue with:
The first idea is that creating something and releasing it into the world changes the lives of ourselves and of others. That is what we design and build software, and that is why being a creator means taking on a moral responsibility for how we change the world.
The second idea is that imposing any restrictions on anything is inherently harmful. This is neither true nor how the world works. Though restrictions can be harmful, they are mostly introduced to prevent harm. The restriction on how loud your headphones can be or the permitted level of toxins in food-safe plastics are in place to prevent harm, and removing the restrictions could in itself be harmful.
I think it is in the tension between these two ideas – creation as a moral act and restriction as a moral wrong – we find both the current dilemma of open source and morality and a potential path forward.
The weight of responsibility
When I publish anything under an open source license, be that GPL or MIT or any other license ratified by the Open Source Initiative (OSI), it takes on a life of its own. It can be used and modified and distributed by anyone for any reason, with or without my involvement, but I cannot restrict it. That’s the whole point of open source ideology.
The problem is this effectively abdicates me of responsibility for my own creations: By releasing software with an open source license I relinquish control over who uses my software or what it is used for. Thus I cannot be held responsible for the software or its consequences. And if I can’t be held responsible there is no incentive for me to make sure my software, when used or modified or distributed by others, doesn’t cause harm or can be used to cause harm.
This is what I mean when I say open source licenses imply value neutrality. And it is here I think the Ethical Licensing movement found its root.
From this perspective one can argue closed source (proprietary) licenses hold the moral high ground over open source licenses because closed source licenses imply responsibility: If a company knowingly releases software that can be used for harm, they can be held liable because they are the only moral actor capable of modifying and distributing the software.
Moral clauses to the rescue?
The big question then is if adding moral clauses to open source licenses can help solve this problem. As much as I hate to say it, I think the answer is no, for two reasons:
First, like the Hippocratic Oath in medicine, breaking a moral clause in an open source license does not have any consequences. Doctors swear to uphold the code of ethics defined by the Hippocratic Oath, but if they cause harm their actions are judged through the legal framework of medical malpractice. A “do no harm” clause in an open source license carries moral weight, but without enforceable consequences and mechanisms to appeal a judgement of harm being caused, it has no teeth.
Second, applying moral clauses to a license does not solve the problem of value neutrality because it is a problem of ideology, and the license is merely a manifestation of that ideology.
If we want to go down the path of the Hippocratic Oath, we need a Code of Ethics for Open Source and a framework for enforcement and adjudication of breach of those ethics. That’s a huge undertaking, and not one I see bearing fruit any time soon.
Toward a New Ethos of Open Source
If we want to take on the issue of ethics in open source on a shorter timeline, we need to take a hard look at the ethos, the core ideals, of open source ideology and ask some tough questions about where the ideology comes from and whether that picture of the world meshes with our lived reality. Open source aims to grant everyone the capability to use, modify, and distribute software. That does not mean open source ideology cannot also embrace the moral and societal responsibilities of the creator, the user, and even the creation itself. To get there we need to broaden our ideological platform, consider what “freedom” means in our current context, and work together to figure out how the Four Freedoms can become part of the larger spectrum of rights and freedoms afforded to us.
--
Posted on March 11, 2020
Join Our Newsletter. No Spam, Only the good stuff.
Sign up to receive the latest update from our blog.
Related
August 20, 2022