Don’t Write a Programming Language

32
Don’t Write a Programming Language

It will chew you up and spit you out, and you’ll not even complain.

I often start new projects at the beginning of the year, around December/January. This is not because I particularly plan it that way, but probably more because the holidays gives me time to actually experiment with new ideas that I’ve spent the previous year thinking about.

Around the time of the start of the pandemic (January 2020) I started working on designing my own programming language. It seemed like a great idea at the time, since I wanted to explore some language ideas I felt had never really been tried before, and it should be easy enough to build a proof-of-concept before I went back to more rewarding projects.

Someone should have told me that once you design a programming language you fall down a rabbit hole so deep you can’t even see the light when you look up anymore.

Programming language design is at the same time the most interesting technical challenge you can take on, while at the same time being for the most part an utter waste of time.

My own project has exactly 1 user (myself). There are a handful of other people who have an academic interest in what I do, but they are also language designers, working on their own related languages.

Even though you are aware of the fact that very few people, if any, will benefit from your work, it is impossible to not keep thinking about ways to improve your project. You keep adding new features that are really cool, and eventually everything programming-related you think about gets focused on how that particular topic can be applied to your language.

It’s also difficult to go back to working on other languages. You end up thinking about how certain features or ideas could be applied to your language, and as soon as you have some spare time, it’s back to the drawing board to implement these ideas in your project.

You are fully aware that your work is almost guaranteed to be of little to no value, once you’ve fallen down the rabbit hole, you’re stuck. Even when working on something else it’s impossible to stop thinking about all the small things you want to add to your project. I think this is the most devious part of language design. It covers so many different areas that you are never “done”:

  • Compiler techniques
  • Code optimisation
  • Maths
  • GUI programming (your language needs an IDE, right?)
  • Graphics programming
  • Operating system integration (standard library implementation)
  • Mobile development (programming on mobile devices is a thing these days)
  • Networking (again, needed for the standard library)
  • Writing (your language needs documentation)

In fact, I can’t think of another programming project that literally touches on every aspect of software engineering. This is the main reason designing a programming language is a project that you can never finish.

It also affects your communication with the rest of the programming community. I’m sure you’ve all seen the person, that guy who can’t help himself from answering programming questions with an explanation how it’s done in their own language. The person that discusses software optimisation techniques with regards how they implemented it in their compiler.

I have to bite my tongue every time, and I still can’t stop myself completely.

This is the destiny of most people who design programming languages. A few get lucky and stumble across something other people start to use.

That said, I’m not entirely sure they are lucky. Getting actual users is probably a recipe for never being able to leave the project. Perhaps I should be happy that I am the only user of my language.

@loke@functional.cafe

Read More

Vanic
WRITTEN BY

Vanic

β€œSimplicity, patience, compassion.
These three are your greatest treasures.
Simple in actions and thoughts, you return to the source of being.
Patient with both friends and enemies,
you accord with the way things are.
Compassionate toward yourself,
you reconcile all beings in the world.”
― Lao Tzu, Tao Te Ching

you're currently offline