Making software by "cosmic pizza order"

I used to bemoan the fact that every time I post about an idea someone tells me, "that's been done before!" — now I've learned to love it

I made this joke a while back in frustration. It was after posting what I thought was a novel idea, only to hear: “someone has been working on exactly this for 3 years!!”

https://mastodon.gamedev.place/@omarshehata/112274380626422317

I wondered: what if I really thought of it that way? If I don’t see it as “getting scooped”, but as literally a way to manifest the software I want to exist into my world (and it’s either perfect, and I use it, or it’s not, then I build on it).

blinry mentioned they heard a term for this at the Recurse Center: “cosmic pizza order”. It’s like if you just went “I wish I had a pizza right now” and it just showed up at your door. But with software instead of pizza.

Below is a quote that summarizes my new mindset. The rest of the post will be reflecting on how I made this shift and why it was difficult for me.

My reaction to this used to be "oh no, I got scooped! that sucks!" but these days my reaction is more, "amazing! I can focus on the next step now, instead of spending 2-3 weeks building this thing!"

like remembering why I wanted this thing to exist in the first place, because I wanted to use it, and build towards something else. We're all making building blocks for each other

(mastodon post link)

---

I often avoid doing research before embarking on a new creative project, because:

• If I get an idea I am really excited about, and I find out it already exists, that will kill my motivation

• If I go off and do it anyway, I’ll probably

  • (1) execute it very differently, especially if I wasn’t aware of the existing thing

  • (2) learn a lot and will be in a position to build on top of what I have in future projects

So this meant I always pushed the research phase of my creative work until the end. My goal was creating, and it was easier to create when I held up the fiction that this thing I am making could really be valuable and useful to others.

(I think it’s like: promising to treat yourself to ice cream if you go to the gym, just to help yourself get off the couch. But after working out, you feel great, you don’t need the ice cream anymore)

This also meant I rarely talked about my work while it was in progress. I really wanted to avoid comments like “how is this different from X” ?

Bemoaning cosmic pizza

Learning to love this was very difficult for me. It really sucks to get scooped, especially when you’re working on something that is genuinely novel and useful, and in the right time place.

I want to make useful contributions. I think the hardest thing about contributing to society is NOT in the doing-of-the-work, it’s in the finding-of-the-work to do.

I felt scooped when Bartosz Ciechanowski published his “how GPS works” article in 2022. I started working on something very similar in ~2020, because I noticed there’s really NO good online materials explaining how GPS works in detail, from first principles, and that aren’t buried in PDFs and textbooks. Which is very shocking, because GPS is a technology that is extremely ubiquitous in all of our lives.

I was very excited about my article because I realized most people don’t actually know how it works, and they are surprised and intrigued when I tell them things like:

• GPS works without cell service, or internet

• GPS works even if you are on an airplane!

• GPS is owned by the United States Government, and they could totally shut it down on adversaries if they wanted (which is maybe why China and other countries have launched their own constellations)

I felt a little better when I saw that his article didn’t show what a “raw” GPS signal looks like (I really wanted to show people, without simplifying, exactly what data their phone is receiving directly from satellite, and how you calculate position from that).

But then Phillip Tennen published the full annotated satellite message, in way more detail than I ever could (the big plot twist is that satellites don’t send down any position info, they just send down the time!)

Learning to love cosmic pizza

I think I should have published what little I had at the time, instead of getting stuck and sitting on it for years.

I really wanted to make these beautiful visualizations, I wanted to start with the raw satellite signal. I had this grand vision of showing the reader how to extract the signal from their phone & plug it into the article. So they weren’t just reading about how GPS works, they could actually extract their own location step by step.

This is my philosophy when writing these articles: I am not here to explain to you how the thing works, from a position of a teacher. I’m just here to give you tools for your own learning path. My “gold standard” is when people use my own article, to teach me something I didn’t know myself when I wrote it (some examples here).

Where I got stuck was on the various mathematical correction steps needed to get an accurate location. I had the kernel of a good idea for an idea. I could have published it, and the collection of resources that I found, and either:

1. Kept working on it alone

2. Found a collaborator who wanted to help out

3. Had the cosmic machine deliver my work to someone else who could have built on it

This is the most difficult: how would you feel if someone took your draft and was inspired by it to make this beautiful thing that made them rich & famous?

I have mixed feelings about it. It’s great that my work was useful and contributed to something, but also, I feel jealous? Why couldn’t *I* have made that and gotten the fame & fortune?

Do I want the thing to exist, or do I want to be the one to make the thing?

If I am being honest, I want both. I wanted to write about GPS I think for similar reasons to these authors:

1. I personally wanted to develop a deep understanding of this technology

2. I wanted other people to experience this sense of awe around technology, and also to feel less powerless around it

3. I wanted to get noticed for my writing, and maybe build an audience so I can do this more

You can see this because Bartosz ends his article with this uplifting message:

Someone other than me writing an excellent GPS article accomplishes number (2).

Number (1) - understanding it myself and (3) - building an audience that could support my work, are pieces of work that are not yet done. Bartosz cannot do this for me, and he also cannot “steal it” from me.

It’s still worth writing my own version, in my own words, because that will help *me* understand it. Breaking it down ever further, or grappling with parts that aren’t discussed in the article, that can help someone else understand it more deeply too.

It’s like: I don’t even have to create the content myself to do this work? I had big dreams of giving a bunch of talks, locally with friends and elsewhere, once I finished this work on the GPS article. But I can still do that? Most of my friends still don’t understand GPS. They haven’t read the article, or they read it and didn’t get it. But I can help. There is more work to be done. There is a place for me.

I think this is the key insight for me:

• other people making beautiful things does NOT take anything away from me

  • my version was already in my “abandoned drafts”, I didn’t stop working on it as a result of other people publishing theirs

• when other people do cool useful things, they aren’t (necessarily) competing with me, they are making progress towards a problem we’re both working on

  • their work is building blocks for me and others. My work is *easier* now than before, I am *grateful* that I’m not tackling this alone

Basically: why can’t I think of writing the same way I think of open source? If you’re trying to do something, and you find an existing open source library that does it, you can either contribute to it, fork it, or just take from it the pieces that are useful for you.

Can we do the same with articles?

Open source articles

It “feels” wrong to take an existing article, tweak it, and re-publish it. It’s plagiarism, right?

I think it doesn’t have to be. I think we can treat it like code under an MIT license. Like, Andy Matuschak has this article on “Why books don’t work”. Maybe this idea resonates with me, and I understand it deeply, but the article as written doesn’t resonate with my students, let’s say.

I should be able to just edit it, remove chunks, and share it (with a link to the original).

It’s all just HTML. We don’t even need an app for this. Just edit it, and re-host the static files (like the wayback machine, but with your edits on top).

At the very least, this could be a normal way to share my notes/marginalia, in the same way we do in physical books. I often want to recommend articles to friends, but they don’t always see what I saw in it. They don’t know what parts are most important or relevant to them.

I could write my own complete version of the article from scratch, or I could allow myself to “fork it”. I didn’t steal the article, in the same way you aren’t stealing the code you fork, there (should be) a very clear and transparent linkage between what I found and what I added.

This feels very “not normal” in writing, but it’s totally normal in many other mediums. When we watch the news, it’s often someone taking a longer broadcast and cutting it & adding commentary. There’s nothing wrong with that in principle (I think it’s generally bad in the news because you often can’t just click to find the original material and watch it in full. The state of our news linking to sources is abysmal).

This new outlook feels extremely optimistic and empowering to me. Every piece of writing I put out there is a building block. We can build on each other’s work. I can publicly share my drafts (like github repos) and work on them in public. Why not?

I think Rosano was the first person that framed it to me this way:

https://mastodon.online/@rosano/111800169557338665