The Rust Revolution

2026-01-11

I strongly believe tech is a mirror of culture, just as language is. When talking about software, the most common history told is that of Silicon Valley. We learn about Steve Jobs and how his initial hippie spirit supposedly created Apple. The world is a diverse place. While the USA was living "freedom", it was exporting dictatorships across South America. Brazil was living under a military dictatorship because, in 1964, a President wanted to teach every Brazilian to read and to distribute land; a warship from the USA was positioned close to Brazil to make sure the Coup d'état would succeed. While hobbyists could try new things in the USA, students in Brazil were either worried about getting a job to pay the bills, or fighting for a better country.

If you fast forward to 2010, enterprise tech in Brazil is superior to the USA's. I was working for a mobile banking app, and the features we had at that time are superior to today's N26 (German "Fintech" Bank), for example. So yeah, a traditional bank is superior to a "fintech" in Brazil. On the other hand, we don't see a lot of non-business innovation coming from Brazil. At universities, many of us are taught to obey and just to follow trends in tech. Some of us are not even taught the foundations (algorithms, OS, hardware), since a lot of this is not so useful in Enterprise Software. This is where I find Rust amazing. Rust opens the door to new learning opportunities.

During the cold war, the world was split into three camps. The First World was the allies of the USA (NATO), the Second World was the allies of the USSR (Warsaw Pact) and the Third World was the countries non-aligned with any of those. Yes, third world does not mean poor countries, it means different. Both the USA and the USSR had to invest heavily in their tech, in order to show the world they were the best. The USA marketing tried to sell it as "entrepreneur" innovations and the other side as "state-led" innovation. Apple, Microsoft, IBM, all came out as heroes we all know.

The Third World was a mixed bag of people and cultures; it's impossible to describe in few words what was going on. Salvador Allende, President of Chile before the USA-backed dictatorship in the country, was leading a project called Project Cybersyn. The project (quoting Wikipedia here), was a distributed decision support system to aid in the management of the national economy. The project consisted of four modules: an economic simulator; custom software to check factory performance; an operations room; and a national network of telex machines that were linked to one mainframe computer. All went to nothing when Pinochet took power.

Brazil on the other hand was living strange times. The military was imposing the "Ufanista" idea, an ultra-nationalist feeling. "Brazil, either you love it or you leave the country". Any working class leader was seen as a potential communist, any "weird" idea could lead you to prison. Still, the country wanted to prove its value. It's with these feelings that older projects (started in the 50's) get support from the government: In 1974 the company Cobra (Computadores e Sistemas Brasileiros) is founded. Cobra starts to build national computers. The company still exists today, under a Brazilian Bank (Banco do Brasil).

If you look close enough, all the three parts of the world were developing tech based on requests or government money. So how come Brazil's enterprise software developed so well, while other markets just stagnated?

Maybe the PC has something to do with it. The Brazilian market was dominated by national clones of the IBM PC, not much innovation here. Cobra was focused on its 500 / 530 models. Is this the whole picture?

From 1977 until 1992, Brazil had a policy of strong trade barriers, which protected the companies cloning IBM PCs, but also made it almost impossible to buy customized software from abroad. National companies had to develop internally their own tools. For example, Petrobras had a very specific need: add some flexibility to interactive graphical programs for engineering applications. That need led to Lua development at PUC-Rio. So yes, there was innovation happening in Brazil, it was just focused on something different.

Maybe it's just the marketing (as the USA needed to sell its "American Dream"), but most of the tech histories in Brazil have a similar background: enterprise need led researchers to create something new. Even today, Pix an amazing digital payment system was created by the Central Bank, while the USA has different proprietary systems such as PayPal. There are not many (fake) story tellings on how a single person created an amazing tech, as the USA likes to lie to the world. History here is raw, enterprise needed this, we built.

Now, my own experience comes into play and I need to warn: I studied at a private university at night, while I worked the whole day. This means I was not part of the upper-middle class. The education for (the majority white) upper-middle class in Brazil is different. They go to PUC-Rio, Unicamp, USP. They only study. We survive. Take this into account and know that there are a minority of Brazilians with a similar background as those kids in Silicon Valley.

My education was in a way mainly to perform "customization for enterprise". A lot of Java. When I picked up any different subject ("oh well, let me learn how a compiler really works") I was told: "you are not going to work for NASA, why are you reading this?". In a way, the enterprises in Brazil needed systems to process information. Convert A to B. Send it to this third-party. Done. The strong trade barriers were gone, the companies could just buy foreign software. We just needed to personalize it. Maybe, this how we end up with amazing banking apps and almost without non-enterprise national tech (?).

"A language empowering everyone to build reliable and efficient software." There is a nice ring to it. C/C++ are languages that were not so common in the Brazilian internal job market. So, after a few classes, it's just dropped in favour of Java, C# and nowadays Python. Those languages are great, but they hide a lot. As a student, we were denied learning how syscalls work, how manual memory management works. How to build foundational software. This is where I think "the Rust revolution" comes into play. There is a somewhat reset on foundational software happening. Instead of using generic databases, people are building databases for specific domains: timeseries, financial data, documents and so on. The Rust compiler is there to teach those that never had the opportunity to learn about concurrency, memory management and low-level programming. Rust for me is not a C++ replacement, it's a learning tool, it's an open door for job opportunities that would have been denied before. We are also living in a different world, will that mean the national outcomes will be different?