Journal

Reminded myself today how much I love writing. It’s funny, I write something every day (hello) but I feel like I haven’t written anything in ages. Turning this blog into a journal has been great in many ways, but it’s definitely caused me to write fewer things of… substance? Not sure.

The inspiration was Brian Eno’s A Year with Swollen Appendices—a year’s worth of journal entries, and a good chunk of essays at the back. Still a good idea, I think, I just haven’t written any of the essays. The journal has given me ideas for the essays though… I’ve just got to sit down and write them. Easy, right? More pressure than writing one every day, somehow.

What should I be writing about? It could be anything, really. I’ve got a few decades of life under my belt at this point, which means I’ve got a lifetime of stories to tell. Growing up. England. Moving 3,000 miles. Moving 3,000 miles again. Design. Mental health. Computing. Art. Love (that is, my love; my wonderful love). Any of it. All of it, maybe.

Writing changed my life, and my writing practice is still pretty nascent. There’s more writing to write; more life to change. Time to stop phoning it in and write the stuff that I really want to write. I thought that I had to be a Writer™, but I think that I just have to write.

I was asking Claude to help me build an Artifact to explain transformers from end to end. At first it spit out an explanation, so I asked it to include examples. Then it included examples, but still just text. Then I asked it to visualize it. Then animate it. Then use better examples.

At some point, I asked it to run a training loop on a (tiny) transformer. It grabbed TensorFlow.js from a CDN, ran a loop with a couple of layers, each with a couple of heads, and used the output in the visualizations. It did it right there, right inside of the browser, right inside of Claude.

Today, it occurred to me that Claude will quite happily run some arbitrary Python. Sure, it will do it on modest hardware and it won’t use a bunch of libraries, but it’ll run Python. I grabbed the source for Andrej Karpathy’s microgpt and asked Claude to run it, right there in chat. A few minutes later it spit out a list of names and told me what our loss was.

Why bother? Wouldn’t it be much easier and better to just use Claude Code to do the same thing, on better hardware and with whatever libraries I wanted? Sure, of course, yes. But it’s just sort of… fun, to poke at things and see what happens. It’s fun to play with tools in unexpected ways. Fun to do things in some wrong, goofy way just for the heck of it.

Anyway, didn’t expect to be writing this, because I didn’t expect to be doing the thing I’m writing about, but it took a few minutes and was kind of fun. Pretty good deal. The bigger thing is that it reminded me (though I’m fortunate enough to not need reminding often) that doing things that look dumb to someone else can still be a fun and worthwhile thing to do. I promise that some of the things you love started out looking dumb.

So, if you needed a push to just start doing something for the heck of it, whether that thing would take five minutes or five years, maybe this is it? Maybe it’s just a short note about a pointless thing and has no point. Doesn’t really matter, because writing about it was fun too.

Family walk at Point Isabel this evening, an insanely beautiful dog park that I’ve probably written about a bunch of times before. Still, every time we go I can’t believe how beautiful it is, and still can’t believe that it’s a ten minute drive from our house, nor that it’s right next to… Costco.

If you stand in one spot and spin all the way around you’ll see the Berkeley Hills, the freeway, Marin, the Golden Gate Bridge, the San Francisco skyline, the Bay Bridge, the Berkeley waterfront, and huge shipping containers being carted off who-knows-where. The weird juxtaposition that is America, in all of its beauty and garishness.

I try not to take any photos, but inevitably the sight of the shimmering water and that famous bridge in the distance compels me to, as it does everyone else. You stand and take minute-long videos that you’ll never post but that you need to record, because the sea and the sky and the mountains demand to be captured. We capture them, then.

Forever grateful to live amongst beauty like this. To see our goofy dog frolic through the grass with a backdrop of the setting sun. To walk hand in hand, slowly, and talk about our good fortune. If nothing else, give me a beautiful place to exist and the good health to enjoy it.

Another glorious Sunday in a sunny garden. I’m still not used to how much sun we get here in California vs. back in England. March usually meant more gray skies than blue; more sweaters than shorts. I love pottering around in the garden—speaking to neighbors over the fence, pulling a few weeds, moving the lawn, lounging around doing a lot of nothing. We didn’t really have a garden in Brooklyn, and I’d forgotten how nice it is.

We went out with some (great, like truly excellent) friends last night. There was a point in the evening where I realized that I’d truly found my people here. Moving a few thousand miles in your thirties (twice, within a couple years) can be tough. I’m lucky to be doing it with my best friend, but it’s still tough, and friends are so important. Anyway, these friends mean a lot to us, and it struck me last night how lucky I am to know them.

So now it’s Sunday night, and I’m sat here with my family, and my heart is feeling full, and I’ve got a week of work ahead at a job that I love. I’m feeling very grateful, I guess. I’ve been so lucky. Not always, but more than many and more than I’d have ever imagined possible. I try to work hard, and to try hard in general, but I’ve been lucky. The funny thing is, it feels like I’m just getting started in some ways, here in this new place.

I continued with my completely pointless but very fun static site generator today, adding related posts to each post. On build, it pulls down an open-source embedding model, creates embeddings for all of the posts, and uses cosine similarity to find related posts. It then shoves a blob of JSON into a data directory and renders via an include in the post template. I’ve only checked a few of the posts, but it seems to be doing a pretty good job. If you can’t over-engineer your personal website, what can you do?

Trying to write this whilst watching the new Louis Theroux documentary and massively failing because it’s difficult to look away from the absolute car crash that I’m witnessing. Louis is great (obviously), but it’s so strange witnessing the worst parts of the internet embodied in human form. I’m thankful that I’ve surrounded myself with kind, loving, caring folks who are also just so creative, impressive and hard-working. What a joy to be able to choose your own actions, and to choose wise ones.

It struck me today that you can make your own luck simply by doing things earlier than most of the other people around you believe is reasonable. We’re in this strange world now where folks don’t need an engineering background to make pretty great software, but I decided to teach myself programming when folks in design were saying that you shouldn’t need to. I didn’t think that I needed to, I just… wanted to.

Now that we’re here, I assume things will do what they always do: just keep moving; shift over. The reason I’m teaching myself machine learning now is that it’s just the next most fun thing to me. It’s the thing that most folks in design won’t be doing, but that will increasingly be the thing that defines what software gets made, and how the world is shaped by it. I don’t want to just work with the output, I want to shape the thing itself.

The thing that you have to do, I think, is just assume that you can do it. Assume that you can do anything, really, because you probably can. If someone can do it, you can probably do it. Maybe you can even do it really well. I try to quiet the critical part of my brain and just start poking around, feeling incredibly dumb until I feel… a little less dumb. After all, don’t we all listen to the critical part of our brain too much already?