Vitalik betrayed the MIT license! Copyleft is the open source savior.

Author: Vitalik, founder of Ethereum;

Compiled by: AIMan@Golden Finance

Original title: Vitalik: Why I Prefer Copyleft Now

In free and open-source software (and more broadly, free content), copyright licenses mainly fall into two categories:

• If the content is released under permissive licenses (such as CC0, MIT), then anyone can access, use, and redistribute it for any purpose without any restrictions, possibly requiring only minimal attribution rules.
• If the content is published under a Copyleft license (e.g., CC-BY-SA, GPL), anyone can freely access, use, and redistribute copies. However, if you create and distribute derivative works by modifying or merging other works, the new work must also be published under the same license. Additionally, the GPL requires that any derivative works publicly release their source code, along with some other requirements.

To summarize: permissive licenses allow free sharing with everyone, while Copyleft licenses only allow free sharing with those who are willing to share freely.

Since I grew up to understand the meaning of free and open-source software and free content, and to build things that I think others might find useful, I have been a fan and developer of them. In the past, I tended to favor permissive licensing models (for example, my blog follows the WTFPL license). Recently, I started to shift towards Copyleft licensing models. This article explains the reasons for my change.

A software freedom style advocated by WTFPL. But it is not the only style.

Why I have always supported loose permissions

First of all, I hope to maximize the adoption and dissemination of my work, and releasing it under a permissive license helps achieve this, as it clearly indicates that if anyone wants to develop based on my work, there is no need to worry about any issues. Businesses are usually reluctant to freely release their projects, and given that I don't believe I have the ability to persuade them to fully join the free software camp, I hope to avoid unnecessary incompatibility with their existing practices and will not give up.

Secondly, I generally do not like copyright (and patents) from a philosophical perspective. I dislike the idea that two people privately sharing data would be considered a crime against a third party, even though they have had no contact or communication, nor have they obtained anything from the third party ("not paying" is not the same as "stealing"). Clearly releasing works into the public domain is legally complex for many reasons, so a permissive license is the cleanest and safest way to avoid copyright protection for works.

I greatly appreciate the Copyleft idea of "using copyright against itself"—it is a brilliant legal trick. In some ways, it is quite similar to what I have always considered the philosophical beauty of liberalism. As a political philosophy, it is often described as prohibiting the use of violence, with one exception: protecting people from other acts of violence. As a social philosophy, I sometimes view it as a way to tame humanity's aversion to harmful effects, making freedom itself a sacred thing, and tarnishing that freedom is repugnant: even if you find it disgusting that two people engage in an unusual, mutually consensual sexual relationship, you cannot pursue them, because interfering in the private lives of free individuals is itself repugnant. Therefore, in principle, there are historical precedents showing that revulsion towards copyright and using copyright against itself are compatible.

However, while the Copyleft of written works meets this definition, the GPL-style code copyright goes beyond the simplistic concept of "using copyright against itself," as it offensively uses copyright for another purpose: to mandate the release of source code. This is a public interest purpose, rather than a self-serving act of collecting licensing fees, but it remains an offensive use of copyright. This is even more true for stricter licenses like the AGPL, which requires the release of source code for derivative works, even if you have never published these works, but only provided them through software as a service (SaaS).

Different types of software licenses have different requirements for sharing source code for those who create derivative works. Some licenses require the source code to be released under various circumstances.

Why am I more passionate about Copyleft today

My shift from supporting tolerance to supporting Copyleft was driven by two major world events and a philosophical transformation.

First of all, open source has become mainstream, making it more feasible for companies to embrace open source. Numerous companies from various industries are embracing open source. Companies like Google, Microsoft, and Huawei are embracing open source, and even open source has built major software packages. Emerging industries, including artificial intelligence and cryptocurrency technology, rely on open source more than any previous industry.

Secondly, the competition in the cryptocurrency field is becoming increasingly fierce, driven solely by profit, and we can no longer expect people to open source their works purely out of goodwill. Therefore, the argument for open source cannot rely solely on "pleasing"; it must also be accompanied by "hard power," meaning that only a portion of the code is opened to those who are willing to open source.

You can intuitively understand how the two types of pressure increase the relative value of Copyleft through the chart below:

Incentivizing open source is most valuable in situations that are neither realistic nor guaranteed. Today, mainstream companies and the cryptocurrency sector are in such a situation. This makes the value of incentivizing open source through Copyleft extremely high.

Thirdly, Glenn Weil's economic argument convinces me that in the case of superlinear scale returns, the optimal policy is not actually the strict property rights system advocated by Rothbard/Mises. On the contrary, the optimal policy does require adopting some non-zero, more proactive approach to promote projects to be more open than they would otherwise be.

Fundamentally, if you assume economies of scale, then through simple mathematical reasoning, non-zero openness is the only way for the world to ultimately not be controlled by a single entity. Economies of scale mean that if I have twice the resources you have, I can achieve more than double the progress. Therefore, next year, I will have, for example, 2.02 times your resources. So...

Left image: Proportional growth. A small difference at the beginning will turn into a small difference at the end. Right image: Economies of scale growth. A small difference at the beginning will evolve into a huge difference over time.

Historically, one key pressure that prevented this dynamic from spiraling out of control is that we cannot choose to stop the spread of progress. People move between companies and countries, bringing their ideas and talents with them. Poorer countries are able to trade with richer countries and achieve leapfrog growth. Industrial espionage is everywhere. Innovations are reverse-engineered.

However, several recent trends threaten this balance, as well as other factors that suppress unbalanced growth:

• Technological advancements are evolving rapidly, making the speed of super-exponential curves much faster than before.
• The political instability within and between countries is intensifying. If you are confident that your rights will be protected, then others becoming stronger without encroaching on you will not harm you. However, in a world where coercive measures are more feasible and unpredictable, one person becoming stronger than others poses greater risks. Meanwhile, within various countries, the government's willingness to constrain monopolies has also weakened compared to before.
• Modern technology can create proprietary software and hardware products that can share usage rights, but the rights to modify and control are not decentralized. Historically, delivering products to consumers (whether within a country or across borders) inevitably meant opening them up for inspection and reverse engineering. Today, this is no longer the case.
• The limitations of economies of scale have historically been a key factor in restraining uncontrolled economic growth, and these limitations are now weakening. Historically, larger entities have borne disproportionately high monitoring costs and have struggled to meet local demands. Recently, digital technologies have made larger-scale control and monitoring structures possible once again.

All of this increases the likelihood of a persistent power imbalance between companies and countries, and even the possibility of self-reinforcement and intensification.

Therefore, I increasingly agree to intensify efforts to make the dissemination of progress a more incentivized or mandatory behavior.

Some policies recently introduced by governments around the world can be interpreted as attempts to actively promote a higher level of communication:

• EU standardization regulations (such as the recent USB-C) make it more difficult to establish proprietary ecosystems that are not well compatible with other technologies.
• China's technology transfer regulations
• The U.S. bans non-compete agreements, and I support this measure because it forces companies to open-source the "tacit knowledge" within, allowing employees to apply the skills they learned to benefit others once they leave a company. Non-disclosure agreements restrict this practice, but fortunately, there are many loopholes in practice.

In my opinion, the drawbacks of such policies often stem from the nature of their government's coercive policies, which leads them to prioritize the dissemination types that are severely inclined towards local political and commercial interests. However, the pros and cons of such policies lie in their ability to incentivize higher-level dissemination.

Copyleft creates a vast pool of code (or other creative products) that you can legally use only if you are willing to share the source code of any content you build based on it. Therefore, Copyleft can be seen as a very broad and neutral incentive for dissemination, allowing one to reap the benefits of the aforementioned policies while avoiding many of their drawbacks. This is because Copyleft does not favor specific actors and does not create a role for central planners to actively set parameters.

These arguments are not absolute; in some cases, it is worth adopting a permissive licensing approach to maximize the likelihood of something being truly adopted by everyone. However, overall, the benefits of adopting Copyleft today are much greater than they were 15 years ago, and projects that would have adopted permissive licenses 15 years ago should at least consider adopting Copyleft now.

Today, this symbol unfortunately represents a completely unrelated meaning. However, in the future, we might have open-source cars. Perhaps Copyleft hardware can help us achieve that.