From curiosity to commitment: a personal adventure to freedom
In my digital adventures, I love using free (as in freedom like free speech, not gratis) software. Many people use the term "open source" to refer to free software, but I have since learned that free software, with its emphasis on freedom and ethics, is the term that fits my values.
One big plus for me with free software is its transparency, meaning no secret spying or watching. It aligns with my belief that software should help you without keeping its eye on you. I also appreciate that I can peek into and sometimes modify the source code, making the software fit the way I work. Waiting for software companies to fix minor issues is a hassle, but free software eliminates this problem.
When I discovered some amazing free software projects like GNU Emacs, GNU Guix, LaTeX, and GNU R, What caught my attention was not just how well these programs worked and were documented, but that they proudly called themselves free software, not just "open source." These tools became a crucial part of my daily routine, both for learning and working.
Free software adds this idea of ethics: it's focused on freedom and sharing. This led me to explore what makes open source different from free software and how free software can still be commercially viable. It’s not just about the code; it’s about the values that come with the software we use every day and what moves the people (and, to a lesser extent, corporations) that make them and promote them.
In my graduate ethics class at Boston University, taught by Professor Allison McDonald, I explored these topics in depth. With her permission, I wrote a policy paper focusing on two main aspects: differentiating open source and free software, as well as the financial viability of free software.
For those interested, here are some major conclusions from my policy paper:
If you aren't free to do whatever you want with a product (in this case, software), then you don't truly own it. This means the company still retains some control over the product. When a company owns part of all the products it sells, it leads to a centralization of power, which I believe is inherently bad.
Sybase Open Watcom Public License, although Open Source Initiative (OSI) approved, is not deemed free by Free software foundation (FSF). The FSF argues against its freedom, noting that it mandates source code publication for private use. In contrast, FSF's General Public License (GPL) allows private modifications without mandating public release.
Free software's commercial viability lies in services. Marginalized groups can hire coders to customize software, reducing reliance on a single company and fostering creative modifications.
While delving into these specifics, I immersed myself in literature from the FSF. Despite my desire to engage in in-depth discussions with the FSF team in person, I hesitated about reaching out due to my perceived lack of technical expertise in the free software domain. However, I took the leap and applied for FSF volunteering at LibrePlanet 2024. To my delight, Craig Topham, the FSF's Copyright & Licensing Associate, reached out, and thus began my journey with the FSF through volunteer training.
On my first visit to the FSF office at 51 Franklin Street, I had the pleasure of meeting Craig, gaining valuable insights into software licensing. Although I had only read about Craig online before, meeting him in person revealed an incredibly down-to-earth and approachable individual.
During my training, I met a co-volunteer, Jim Garrett, who uses GNU Emacs Org-mode for statistical analysis in R. While RStudio is the mainstream choice for working with R, as a data scientist, I prefer Org-mode as well, because it supports literate programming for multiple languages like R, Bash, Guile, and more. Literate programming involves writing documents in plain English with embedded code and results, unlike traditional setups where code is primary and explanations are comments. This approach is becoming the standard in research and data science. When I asked Jim why he uses Org-mode, he explained that GNU Emacs was the de facto tool when he started his PhD, and he never felt the need to switch to RStudio. This reaffirmed my belief in the quality of Emacs. We agreed to catch up after LibrePlanet to discuss our statistical workflows further.
The FSF organized a "Friday night open office" party the day before LibrePlanet, where I had the chance to engage in a thought-provoking discussion with Alexandre Oliva, former vice president of the board of directors of the FSF and founding member of Free Software Foundation Latin America (FSFLA). This discussion helped me contextualize some of my ideas within broader frameworks worth sharing. In my policy paper, I had initially concluded that free software could only be sold once. However, Alexandre clarified that while it may be challenging to drive monopoly prices after the first sale, the software can still be sold multiple times. He also introduced me to the concept of "monopsony," a market structure in which a single buyer substantially controls the market as the major purchaser of goods and services offered by many would-be sellers.
Our discussion also covered the concept of natural monopoly, which occurs, for example when someone invents something truly creative and unique. To prevent such innovations from staying hidden or being used solely for personal benefit, society established patents. However, the problem arises when companies patent items that are obvious in hindsight, leading to excessive monopolies. Any free software would prevent companies from creating monopolies by ensuring that the end user has full ownership and control over the product, eliminating the concentration of market power.
Additionally, we touched on how companies increase their monopoly through other means, such as Digital Restrictions Management (DRM). DRM restricts the use of digital content and devices, limiting user freedoms and reinforcing the company's control. This is yet another way companies ensure that users never fully own their products, perpetually paying the cost, whether it be monetary or in terms of time and effort.
At LibrePlanet, I had the opportunity to serve as a room monitor for several sessions, one of which was "GNU/Linux in the high school classroom" by Neil Plotnick. I was particularly impressed by Neil's approach to seamlessly integrating software skills with lessons on privacy and freedom principles. His innovative teaching methods, such as comparing iteration in programming to real-life examples like musical chairs, provided a unique and engaging way to educate students on these important concepts.
During a Saturday night walk around Boston, I had an insightful discussion with Chris (Krzysztof Siewicz, Licensing and Compliance Manager at FSF) about his PhD journey and his support for the cause of free software. He shared his insights on the licensing issues surrounding neural networks systems, in particular large language models. Chris pointed out that training data for example could be subject to various exclusive rights other than software copyrights, or even not subject to any rights at all.
LibrePlanet 2024 had a number of other highlights for me. I had the chance to meet Internet Hall of Fame inductee Richard Stallman in person and, of course, got a photo with him! I made some great friends and we agreed to stay connected via a Signal group and IRC. Recently, I was also hired as an intern at the FSF and I'm looking forward to working closely with some of its best minds, Ian and Michael. :)