One of my original goals with this project was to build a site that let me sell pottery and crafts, and which could be set up easily by anyone for their own use. Historically, the ability to participate in markets^[1] has been a major part of individuals' ability to determine their own lifeways^[2]. In this post, I want to look at some ideas around sending and receiving payments, with an eye toward which ones might find a place within this project.

Before I started building this system, I had used two other platforms for (trying to) sell items I made. First, I used Etsy. I didn't make any sales and I was generally unhappy with it as a service. It felt like I would have to do the same amount of marketing work to get people to visit my etsy store as I would have to do if I was hosting my own site. If I was going to do all that work anyway, why not just set up my own site? Following that logic, I built a site using Shopify. I quite liked using Shopify--it allowed me to customize my store in useful ways; it had sensible screens for showing me the state of my account, and it gave me much more of a feeling of ownership than Etsy had; I didn't feel like Shopify was obviously using the marketing effort I put into my site to promote its own platform. But it cost $30 a month, and I made one sale, totalling $40, for the several months I used it. So my overall impression was that it's a tool that may make sense for an maker who has either an established customer base or the time and motivation to build one, but it's probably not a good first choice for casual makers who might sell a few things every year.

That category--"casual makers who might sell a few things every year"--is very interesting to me. For one thing, it seems to be a very common category among makers. Before covid, when I was spending a lot of time in a clay studio, there were a few people who put a lot of systematic effort into selling their work. These people would structure their time intentionally; they would focus on making enough product to meet different selling events and seasons; they would reserve space at markets and spend their weekends selling. It was a fair amount of effort, and while I never saw anyone get rich that way, it seemed like many of those practitioners were satisfied and enjoying the process. But there was also another group of practitioners, among whom I counted myself: people who occasionally produced something good enough to sell, but who weren't ready to commit the marketing and organizational effort it would take to reach profitability^[3]. For those people--some of us perennially on the edge of trying to sell our work systematically--there seemed to be a kind of wall. I thought my work was good enough to sell, but I didn't know if the demand would be worth it. And if I wanted to test the level of demand, I had to put in what seemed like a lot of marketing effort that might all disappear if I changed my mind or wasn't successful.

Another category of makers who don't seem to be well-served by existing payment solutions are journalists, educators, entertainers, and others whose products consist of media like text, images, videos and sounds. For these makers there are subscription services like Patreon, distributors like Spotify, and advertising platforms like youtube. I haven't used any of these to try to make money^[4], but my understanding is that it's very hard for an individual--particularly one not yet established--to build a substantial revenue stream using them.

My intuition tells me that this doesn't have to be the case. Specifically, if I imagine that I was directly charged a small amount per item that I read, watched, or listened to--a penny or a fraction of a penny--and if most of that amount--maybe anything above 70%--went to the person or organization who produced the content, then I think that person or organization would see more revenue, per-view and overall, than in the current system where they're paid in what seems like a much less fair way^[5]. But in practice, systems for what I would call grassroots micropayments--very small payments from one person to another--haven't emerged. One of the reasons why not is transaction costs; the financial system includes some very effective gatekeepers whose prices per-transaction are often in the tens of cents, making the lowest feasible transaction sizes fairly large^[6]. Another reason why not is the perceived difficulty setting up such a system: would it be something like Venmo, where everyone can be both a payer and a payee? How would we keep track of who owed what? Would you have to sign in on every device so you could be billed correctly?

Notice how all of these questions are pretty easy to answer if we assume that you're running your own social media site. Let's go through them one by one:

1. Would it be something like Venmo, where everyone could be a payer and a payee? Yes. a URI (for instance but not necessarily, a person's profile page) could be used as an "address" that money could be sent from or to^[7].

2. How would we keep track of who owed what? You would sign in once, to your site. In your feed, you would see items from everyone to whose content you subscribe. In order to get some of that content, your server would do a tiny transaction with the sites of those other people. And they might do the same when getting content from yours. Naturally, this would not be the default type of interaction; just as now, most feed items between friends would not be monetized. But some people could offer content for sale to the public as well as for free.

3. Would you have to sign in on every device to be billed correctly? Yes, but you'd just need to sign in to your own site. As long as you view content through that interface, the content creators have a way to get paid--your site would pay them directly through microtransactions. You could connect to your site any way you want--on your phone, on your laptop, a set-top box. Doesn't matter.

What starts to become apparent is how different the market might look to someone who has their own services on the network, acting directly in their own interest^[8]. More and more of the tiny intrusions on one's personal life online would start to look like robocalls--unhuman, unwelcome, and unnecessary. Another side effect of this structure might be fewer stores of aggregated data about individuals^[9].

So what would it take to make that happen? I suspect that a network-hosted peer-to-peer system that was profitable for artists and small craftspeople would need to be fairly dense--it would work best if it was as ubiquitous as something like facebook. You wouldn't want it to be centrally run, and you definitely wouldn't want VCs in the picture, so I don't think we're talking about a startup. It needs to be as close as possible to a one-click install. It needs to be reliable and secure even when operated by non-software-practitioners^[10]. We need to take precautions against forseeable second-order effects of people using the system; if we told everyone on Earth to buy themselves a Raspberry Pi, I think things might go poorly. Big cloud providers demonstrably have that scale, though; they're already hosting most of these services as they exist now.

I thought this post was going to be a description of the different types of payment processors I've found, but as usual I went way off track and that will have to wait. And as usual, I'm glad for the opportunity to take that detour--it's an aspect of this project that's fun to think about.