I have a friend that likes the feeling of progress in the professional area. He considers working on a project as part of his professional development. Following our discussion with regards to staying motivated on such a project, I would like to describe the factors that keep me going. Your factors are likely to be different but I would like to share mine as a starting point.

I hope that this article will help you, the reader, to pick and/or stay motivated on your own project. Note that you should first think hard about what you are trying to achieve and whether a project is the right solution to that.

My Project

The project I’m basing my self-observations on is Next Generation Shell, an open source shell which doesn’t ignore advances that happened outside of the terminal for last decades (do you feel the pain in these words?). I started working on the project in 2013. As of time of writing, which is 2024, I’m still on it.

The project has two main parts:

• Programming language – it’s in a good shape and we use it at work
• User Interface – early stage, too early to impress anyone

Motivation

I can not separate motivation from picking the project. In my case, all the original motivation behind starting the project is still relevant. It’s worth noting that my feeling is that this project chose me, not the other way around.

Roughly from more significant motivators to less significant ones:

1. Pain. My project solves my own pain. I was programming in bash and Python. They both suck for DevOps scripting. Looked around for other languages and projects that would be aligned with my vision of the solution. Found none. Decided to do it myself. It was very clear from the beginning though that the amount of effort to solve my own pain doesn’t justify working on the project… if not for the next point:
2. I like helping others to be more productive. Software is a very scalable way to do exactly that. While you can maybe generalize this as “the project is aligned with my values”, I tend to think about this as “the project is aligned with my intrinsic motivation”. While the programming language should make the user more productive, the UI should really take the productivity to the next level (some would categorize the UI as “groundbreaking”).
3. The project became usable long time ago. Around 2016 I think. Seeing that it actually works for yourself is a great motivator. From GitHub issues that are opened from time to time, I assume that others are using NGS too. That means that NGS works for others too, at least to some extent.
4. The project is mentally challenging but not too hard (ignoring speed optimizations here, which could be). When I started, I knew roughly how to do it. It touches two topics that I enjoy: programming language design and semantics.
5. While I’m writing an NGS script I’m always thinking “how this can be more ergonomic?” and constantly improving the language. When I do the appropriate changes, it’s “NGS just became a bit more useful”. That contributes to the feeling of progress.
6. While looking at any piece of code in any other language, it’s always a challenge: should NGS do it better? If yes, how? It’s very motivational to see that in many cases NGS does it better that the piece of code in front of me.
7. Speaking of comparison, it does make me feel good that the niche is still not covered (amazing how bad the things still are) and therefore has huge potential for improvement of productivity. Looking at other projects that showcase anything but working with the cloud I’m thinking “Yep, we still need a shell to work with the cloud”.
8. It’s something to show to a potential future employer.
9. Not a near future plan but there is a potential for a startup here which would provide cloud-based enterprise features (I had a list of such features written down somewhere). If you can do that around a terminal, I guess you can do it around a shell.

Demotivation

Some things are demotivating when working on the project. Here they are and how I handle the ones that I know how (after “//”).

1. “You don’t understand, bash is perfect, we don’t need another shell” comments. // There are people that just can’t be moved easily.
   • I’m sure some assembler programmers didn’t want to move to Fortran or COBOL or C. Some didn’t want to move from these to let’s say Java or Python. There is no more “right” motivation behind creating a programming language than dissatisfaction with existing ones.
   • With regards to UI, some people think that Command Line Interface which mostly limits the interaction to a single line is the pinnacle of UX today. Well, I disagree and I’m about to prove that we can do better. It’s actually partially proven since 1976, when Bill Joy released vi which used the whole screen for text editing. Now I just need to prove the same for the shell instead of text editing.
2. UI is not ready yet and I can’t show it. I assume it’s way easier to sell productive UI than a programming language. // It shouldn’t be to far away from bare minimum demo though.
3. Attracting users and developers is hard. I suspect few reasons:
   • It’s harder market now. When you have C and you release awk the amount of convincing is way less than when you have Python and Ruby and many other languages which are popular and kind-of do the job.
   • My personal skills in attracting people to a project are not great I assume // Need to team up with someone but again – need to find that person first
   • I’m prioritizing technical work above finding users. I guess that “solving my own pain” dominates here.
4. The changes needed to be done to the language to make scripts at work more ergonomic (which I just like doing) come at the expense of working on the UI. It’s just really a hard tradeoff.

---

What motivates you? What demotivates you and how you handle that? Share in the comments.

Have a nice day!