factorpy, taking a break for now

Taking a break

factorpy is in a decent place right now, if I could just get the list of things I want to get done working. But my motivation to work on the project has dropped down to basically nothing so I’m just going to leave it for now, maybe I’ll actually finish it in the future. Probably not

Turns out my idea isn’t new

After working out my initial idea for factorpy I spent a long time looking to see what programming paradigm it fell into, and I ended up deciding that it was just too stupid to have ever been thought of before. But I was wrong, totally wrong. Turns out that factorpy falls into the programming paradigm called dataflow programming. I realized this as I was watching a computerphile bonus video where they ask about favorite programming languages, and one guy said LabView. When he described it I thought, huh that sounds alot like factorpy. Turns it straight up is factorpy. Turns out lots of dataflow programming languages exist.

It also turns out that dataflow programming languages have many uses and had lots of research put into it in the 1980’s. One big feature is that they easily translate to visual programming, which is good for those who don’t know how to code. It is also very useful in certain workflows, such as in machine learning. Turns out that tensorflow, a very popular machine learning package for python, is based on dataflow programming (tensor flow).

Who would’ve thought my dumb idea was actually not that dumb after all.

After looking around I didn’t find any dataflow programming languages that were nearly as clear and easy to read as factorpy so maybe my language still has a niche to fill. Gives me a bit of motivation to keep working on it.

For the time being I’m just gonna work on other stuff (how has no one written an interpreter for Snowflake yet?) and maybe spend some time poking around other dataflow languages and see if they are compiled and if so, how.