Fission and the Future of Computing
In the last update, I announced that I was joining the team at Fission as a research engineer. If you visited the Fission home page at the time, you might have had trouble connecting the dots between Fission and the research interests I’ve been exploring in this newsletter. So what does Fission do exactly?
Fission does industrial research
Fission is set up as an industrial research lab, in a similar vein as Ink & Switch. Industrial research stands at the intersection of academic research and industrial application. Academic researchers generally develop techniques and solve problems that are decades ahead of current practice in industry. Furthermore, they are not often interested in or incentivized to bring those ideas to bear in industry. In order for those new techniques to leave the ivory tower, someone has to do the work to translate those ideas into practice, often reworking and adapting them to fit the context of reality. Translation often falls to industrial research, where the work is done by people familiar with both parsing research papers and practical software engineering. If you’re interested in learning more about industrial research, I highly recommend this podcast with Peter VanHardenburg from Ink & Switch.
So, that’s how Fission works. Fission doesn’t sell a product, but rather builds technologies for people who build products. When Fission does build a product, they build primarily to learn whether the technologies are well-adapted solutions to the problem. In other words, they approach the problem like a scientist. As Fred Brooks has said:
“A scientist builds in order to learn. An engineer learns in order to build.”
Fission pioneers new computing primitives
At the highest level, the reason I have happily joined Fission is because we align on a vision for computing itself. Briefly: computers are a transformative tool for humans and are still relatively young. In order for computers to reach their full potential, it needs to be easier for anyone to fully customize their computer, a task we currently call “programming.” Unfortunately, programming today is an incredibly specialized skill that requires a broad range of esoteric knowledge and exposes people to a ton of incidental complexity.
Imagine buildings that collapse when the architect / contractor that built them goes out of business.
— Jess Martin (@jessmartin) April 13, 2022
Welcome to modern web apps.
How could we make computing easier and more accessible? Boris and Brooke, Fission’s cofounders, began asking this question several years ago. They have been steadily and humbly building new primitives which will eventually enable new ways of “programming.”
What are these new primitives and why do we need them? I’ll be doing a deep-dive into each of these over the coming weeks. At a high level, Fission groups these primitives into three categories:
- Data: how do I store, access, index, and share the data I create?
- Identity: how do I secure and manage permissions to access my data?
- Compute: how do I execute programs and access the results?
Fission also shipped an updated home page yesterday which points to some of the recent projects that we’re working on, including an authentication protocol called UCAN and a wrapper around IPFS that enables distributed file storage called WebNative, among other things. There are also some exciting unannounced projects in the works.
Fission is made of good folks
It’s been exciting to work with a vision-aligned team that’s targeting the ephemeral industrial research phase. But even better, the team is downright pleasant to work with; seriously smart, humble, nerdy, and collaborative. Working on the right things in the right way is great, but if it’s not working with enjoyable people, what’s the point?