Artefact Thinking
I moved to Japan almost four years ago. It was supposed to be a reset, but instead I've spent most of my time here wondering whether my work matters. After some reflection, I realised I've been asking the wrong question.
The problem with legacy thinking
When asked what I thought mattered in life, my recent answer has been legacy. It got me thinking about my research; whether any of it matters.
Legacy thinking is to ask "Will this be remembered? Will I be remembered for this?" It's a trap that many in academia fall into, and I often find myself asking the same questions during periods of frustration or uncertainty. The problem is that these are questions to which you can't know the answer and with outcomes you can't control. So you wait. I had a three-year fellowship in Japan with the freedom to collaborate with anyone, to visit any lab, and even to work on any topic. Instead, I hesitated until I felt I had something significant enough to share. I simply waited.
At the beginning of the year, when I took some time to think and reflect on the past few years here, I also happened to read something that helped me to reframe the question entirely.
Artefact thinking
... New ideas are important, but what compounds are the artifacts you leave behind ... Work that moves the field is work that empowers others [to] start where you left off
-- Dimitris Papailiopoulos (2026/01/01)1
His point was simple: focus on what is useful: Artefacts.
Instead of asking "will this be remembered?", I can ask "is this of use to someone?"
Now that I'm leaving academia, I no longer have to wait for my paper to go through several drafts and several rounds of peer review, achieving publication (or rejection) after months of waiting. Instead, the reward of publishing artefacts is immediate.
Artefacts are the stepping stones that you lay in a river for yourself and others to use: a dataset, a blog post summarising your recent deep-dive into a topic, or even a clear explanation of something that took you too long to figure out. They don't need to be significant. They just need to be useful. This is a question I can answer: and unlike legacy, it's one that I can control.
Collaborations across time
When you create an artefact, you're leaving a note for someone in the future who's working on the same kind of problem as you were. You probably won't ever meet them. You might never know they even exist. But they'll find your artefact, the something useful left behind, and that stepping stone will open up new paths that they hadn't noticed before. This means that the thing left behind doesn't need to be perfect, or even finished. It just has to provide some direction. If you write for an audience of one, you can avoid the perfectionism loop.
The thinkers who are remembered are often remembered for their artefacts. They accrue a significant collection of meaningful artefacts during their lifetime. Scientists produce papers. Writers produce books. Legacy is not, therefore, something you ought to aim for, it's simply that which emerges when you consistently contribute things of value during your lifetime.
Going forward
I came to Japan looking for a reset. Instead I became comfortable, and then stuck. I kept waiting for the moment when I'd finally have something significant to publish.
I don't think that moment is coming. So I'm changing the question. This year I'm going to produce more things - code, writing, research. I will start small. I may slip up from time to time. That's fine. The question is no longer whether it's significant, but whether it's useful.