78: Document (Automatically), Don't Create
Personalised systems to augment consumption and creation
Hey there,
Some updates my side:
I wrote this article about building a mass of written work, and how I’ll be using my website as the place to store it
This week I finalised details on my “Python for Biomedical Data Science” module that I’ll be teaching at Imperial College London this summer. I’ll be sharing all lecture notes and coding exercises on this list. These are the topics and I’m still creating the material so let me know if there’s anything you’d like me to cover.
I stumbled across this awesome list of AI tools for research and knowledge management. A strong overlap with my interests, so I’ve been endlessly diving into the work and writing of Linus Lee
Sharing publicly takes effort
There is inherent friction to overcome when sharing things on the internet. You need to stop what you’re doing and write/create something, decide whether you’re happy to share it, and consider how best to package up your thoughts for the potential reader.
But what if this wasn’t the case?
Automatic outputs
People say “document, don’t create”, but what if you could go one step further: “automatically document, don’t create”.
Some platforms inherently enable this. In GitHub, you can see people push changes real-time. Some people use Twitter for streams of public thought.
But what if your workflow is different?
In my case, I do all my thinking by writing notes on my computer or by having conversations. This includes rough, fast notes as I probe my thinking on decision I’m about to make.
Unless I stop and deliberately “package up” those notes, they’re not ready for public consumption.
Experiences left undocumented
For the last couple of years, I’ve broadly felt too busy to write (publicly).
But I’ve had lot of thoughts and experiences that I would love to share with others, and which I believe could be valuable to at least one other person on the internet.
My friend Musty told me:
“Chris, you’re always doing a million things at once, but you need a PR team or something because you never share half of it”
I don’t think I ever want to be something who only writes or creates content for a living. But I feel like if I share more what I’m doing, I’m hedging my bets more in terms of positive impact. If I build something and it fails, at least someone else might take the learnings I shared and run with them.
Building out to build up
Right now, I have some free time. So I’m setting up systems that can massively cut the friction to sharing - and enable me to share as I go along, even when things are full-on. (And which I hope to use indefinitely - because I think all digital systems needs an end-state - more on that next week.)
Ultimately, the nature of such systems depends on how you as an individual tend to operate. Given that I live my life in markdown files, my system is:
Automatically pushing markdown files to GitHub
Markdown files with specific YAML tags will automatically be shared on my site (via Jekyll)
A library of custom GPT prompts which help me convert rough notes into publicly-shareable notes with one hot-key*
Isn’t “creating a system” just procrastination?
Setting up new ways to do something inevitably takes time, so it makes sense to ask whether the time is better invested just doing the thing that you’re trying to make easier.
I think the simplest approach is to compare estimated time costs vs savings.
Obviously, if I spend 3 hours optimising my Notion structure for a project that’s only going to last a week (and where the principles don’t obviously generalise beyond), then it might not be a good spend of time.
In my current case, I’ve invested probably 30+ hours into (1) learning Jekyll, (2) setting up a cron job, (3) learning about writing good GPT prompts, etc. I think I can justify it when I extrapolate forward over the amount of time I plan to use this for, and that skills should transfer to future projects too.
So maybe it helps me share more (time will tell), maybe the skills help, but to be honest, above all - this has been fun as hell.
*I’m still testing and refining these. I started with the reverse-engineered Notion AI prompts and have customised them since. I may make this public in future so let me know if interested.
About Me
👋 Hi, I'm Chris Lovejoy.
I'm a medical doctor 🩺 -> machine learning engineer 👨💻 -> start-up founder 💡
I'm on a mission to improve how we manage our health - and share my learnings and experiences here, on my personal website and on YouTube.
I also throw in my favourite things from the internet, and the occasional joke (humour is work-in-progress).