Working in public
8th April 2023
The topic was “Working in public on open source”, and Citus developer (and pg-cron creator) Marco Slot and myself were interviewed by Claire Giordano and Pino de Candia.
The full hour-long audio conversation is now available on YouTube.
I ran it through Whisper to create my own transcript. Here’s my answer to a question about the benefits of working in public:
The biggest thing for me is that I never want to have to solve the same problem twice, ever.
That’s the most frustrating thing: when you sit down to solve a problem and you think wow, I solved this before and now I’m gonna have to waste my time figuring it out all over again.
A lot of the problems that I solve when I’m engineering are problems that can be captured in some kind of form. Maybe it’s a commit message with a commit that updates something. Maybe it’s a few notes. Maybe it’s just a sketch in an issue description of the approach that I was going to take.
I found that having those out there, just having those in a system massively increases my productivity. Defaulting to putting them in public, partly it’s sort of an insurance scheme.
I’ve worked for companies where I did everything in private. And then I left those companies and I’ve lost all of that work!
Everything that I do in public that has an open source license attached to it is just out there: I will never have to think about those things ever again.
That’s a problem that I’ve solved once and will never have to go back and revisit.
And I love that.
I feel like the work that I’m doing is constantly adding up to me having more capabilities and more tools in my tool belt.
It’s actually very selfish.
I have this website, my TIL website—and I just published my 400th note there.
And on the one hand, it is for other people, so that if somebody else needs to figure out how to copy a table from one SQLite database to another, and they do a Google search, they’ll land on my site, and it’ll solve the problem for them.
But mainly it’s for me.
The fact that I’m publishing causes me to increase the quality of the notes a little bit, so they make more sense to other people.
But it also means they make more sense to me when I come back in a year’s time and I’ve forgotten everything.
So yeah, I feel like you can actually be very selfish in your motivations and still do all of this stuff in public in a way that benefits other people.
Here are links to my posts that I referenced during the discussion:
More recent articles
- Weeknotes: Embeddings, more embeddings and Datasette Cloud - 17th September 2023
- Build an image search engine with llm-clip, chat with models with llm chat - 12th September 2023
- LLM now provides tools for working with embeddings - 4th September 2023
- Datasette 1.0a4 and 1.0a5, plus weeknotes - 30th August 2023
- Making Large Language Models work for you - 27th August 2023
- Datasette Cloud, Datasette 1.0a3, llm-mlc and more - 16th August 2023
- How I make annotated presentations - 6th August 2023
- Weeknotes: Plugins for LLM, sqlite-utils and Datasette - 5th August 2023
- Catching up on the weird world of LLMs - 3rd August 2023
- Run Llama 2 on your own Mac using LLM and Homebrew - 1st August 2023