How hard is it for open source developers? We can see this in a number of recent events and survey results.
- Babel, an open-source project with millions of users, ran into financial difficulties;
- Pedro Ronauck, author of open-source project Docz, says it is too hard to maintain open source, and the work is taking its toll on him physically and mentally.
- According to a survey by Tidelift, open source maintainers face the dilemma of less money and more pressure. More than half have quit or are considering quitting.
- Super Nintendo simulator developer BSNES Near suicide due to cyberbullying;
- Open source code is copied and sold…
Recently, we have seen the sad news that Redox OS contributor JD91MZM2 committed suicide in March 2021 at the age of 18. He worked on Redox OS Summer of Code in 2018, 2019, and 2020, and has contributed to every aspect of Redox OS, from the kernel to ReliBC to the porting program.
The GitHub feed stopped in March forever, sorry
Redox OS author Jeremy Soller wrote a tribute, saying that there is a lot of work to be done to make open source sustainable, and a big part of that is taking care of the health of the open source community and its members.
Jeremy Soller recalls the contributions of JD91MZM2 and calls attention to the physical and mental health of open source developers. SegmentFault has compiled this article and hopes that people will pay more attention to the health of open source developers.
reflection
Yesterday, I received a message from another contributor saying that the JD91MZM2 has been offline for a long time and has not responded to an email. I tried to contact him through the information I had, but nothing came of it. I revealed the JD91MZM2’s real name to another contributor, and he found the JD91MZM2’s obituary. We verified his name, location and date of birth. No cause of death was listed, but I think the evidence we found suggests that he committed suicide after suffering from a mental illness.
I was shocked to learn all this. He is such a prolific contributor, not only to Redox, but to many other projects, how could he feel that death would be better than life? This is a man of limitless abilities, and, until recently, he seemed to be in control of his life. But the longer I lived, the more I realized that it was probably just an illusion and that things were deteriorating very quickly.
I last communicated with JD91MZM2 in February, a month before he died. This communication is purely technical, regarding the Aarch64 port of the Redox kernel. I couldn’t help but think that maybe that was one of the factors in his choice to die.
In open source, we often emphasize the importance of good code. After all, every open source project’s deliverable is source code, but we often forget that good code is written by good people, and keeping those people happy should be the maintainer’s primary concern.
Mental health events have many dimensions. For one thing, mental illness often has a genetic component. On the other hand, these genetic precursors often require both chronic and acute environmental triggers. Chronic triggers can be a long-term poor home or work environment and lead to manifestations of the mental illness itself. An acute trigger may be an argument with someone that leads to a psychotic episode. The onset of mental illness can be severe enough to overcome a person’s survival instinct and eventually lead to suicide.
In this sense, suicide is not a sign of weakness.
Yet society has taken the position that these events are unstoppable forces and that the factors that lead to suicide are internal, not external. I refuse to believe that. We must find the cause of every problem and try to alleviate it, even if it is ultimately impossible to achieve.
So I have to look at what I’m doing and see if I could have done differently, to see if I could have saved a life, to see what life I could have saved in the future.
Open source and mental health
Some aspects of open source seem to appeal to the strangest human beings, myself included. The insistence that everything is chectable may be driven by compulsion. And those who are prone to the behavior tend to inherit it from other diseases. ADHD, autism, bipolar disorder, depression and other conditions are common among open source contributors.
As a result, the open source community is also noticeably lacking in soft skills. This would clearly divide the community, isolating open source from the “normal” world. Fortunately for us, open source has finally become a profitable industry, and the injection of capital has resulted in a significant diversification of open source talent.
However, this comes at a cost: people who are not comfortable with the new commercial benefits of open source projects are often abandoned; Disagreement with the original maintainers resulted in insurmountable changes in the project itself, which further forked the project. I’ve been a part of all this myself.
For the most part, we lack a survey of human costs, looking at mental health events among a large number of open source contributors, and trying to find commonalities. Sometimes mental illness leads to new projects, sometimes contributors become too tired and leave open source, and sometimes they commit suicide.
We must recognize our role in the chronic factors that lead to mental illness as well as the acute factors that lead to dangerous episodes.
My journey
I’m not an insulator of mental illness. I often get messages like, “You seem to be in control. How do you do it?” The harsh truth is that none of us may really be in control. While our definition of “being in control” can vary widely, having a successful project does not equate to general happiness.
I must admit that I am happier now than I have ever been. So, maybe more than the average person, I do have control. My life was a long journey of therapists, psychiatrists, drugs and isolation. I could end my life just as easily as anyone else. Fortunately, I have been fortunate enough to identify chronic stressors and have dutifully eliminated them.
My freshman year, I had a hard time. (At the time, I was about the same age as when JD91MZM2 decided to end life.) I gained nearly 50 pounds, and I lived with three other roommates, two of whom also died young. I alternate between ADHD medication, antidepressants and even smoking — looking for something to “cure” me. During that time, I was always programming, often ignoring school assignments to do it.
Before college, I worked as an intern at Zhuo Medical, writing defibrillator software, and received two patents from this work. To be honest, I know more about computers than anyone else. However, my understanding of people, including myself, lags far behind others.
During that time, I developed a relationship with the vice president of research and development, who became my de facto boss. At the end of my freshman year, he asked me if I wanted to continue working, and I said yes.
Sophomore life is very different for me. I spent most of my time writing software and made a lot of money. I had no interest in school, flunked several courses, and soon dropped out of college to pursue software engineering full-time.
This has had a huge positive impact on my mental health. I lost weight, kept in touch with my college friends and met my wife through them. We bought a house together, got married and had a beautiful daughter. I started the Redox OS project and started working on System76. After sophomore year, I never thought about mental health again, and I didn’t need treatment or medication. All my stress is gone.
One person’s success in mental health doesn’t always translate into success for others. I’ve created and destroyed hundreds of relationships along the way. I have to admit that when I’m happy, I have a tendency to elicit exactly the opposite emotion in others. I keep those relationships that bring me joy and ignore those that require effort. And at some point, maybe I forgot to keep in touch with JD91MZM2 to make sure he found the same joy I did.
Is there a solution?
At the end of his article, Jeremy Soller argues that there is no solution to this situation and that every case is different. But he decided to stop evaluating contributors based on the code they wrote. Code cannot be written on its own, he says, and the people who write it need to be maintained more than “open source” itself.
Reference link: https://www.redox-os.org/news…