Load Bearing People, Load Bearing Code – San Francisco 2025

This is somewhat looming. I didn't realize this was going to be the first site. Does do people get this reference? I'm really interested because this is this is the intersection between two almost microscopic communities, but I suspect the people in that subset might be here. So, okay, rather than do you get the reference, do you not get the reference?

Does anybody get the reference? Great. All right. Well, that's the first 20 minutes covered. So, uh, this is a reference to, uh, uh, the there's an anarchist sticker archive, and actually I discovered today that if you go, it's, uh, hosted on IPFS.

If you go to free anarchiststickers.org, there's, uh, all of the anarchist stickers from the last 20 years are backed up there. Um, so this is actually a riff off that. The traditional anarchist, yeah, put it on the laptop sticker is no gods, uh, no masters or no rulers. Um, so this is a riff on that for, uh, people who really strongly feel something, but it's, you know, but they don't go out much.

Um, this is no sp. So spof is single point of failure. So that's why you're here, right? Is because you believe that there should not be single points of failure in society or in technology. and no admins.

Um, so for those of this is a risky thing for me to put up because in many ways the Internet Archive is a temple to the success of admins, right? Um, I'm pretty sure that any CIS admin or any network admin to use the terms. Um, if you really believed what you were doing in your will, you would ask to be buried here. Um, possibly in some of the racks over there or maybe frozen in carbonite like so many of the rest of the folks here. Um so um uh admins basically in a technical world mean the people who administer the network.

So what this is saying is like we need to build a system in the same way as anarchists say no gods no uh masters that don't have single points of failure and don't have administrators. Now anarchists tend to like put on kind of strong kind of we're going to punch you in the face kind of vibe. If you hang out with them for a bit longer um they're kind of lovable. um uh and and wouldn't hurt a fly until suddenly they do. Well, they wouldn't hurt a fly because they're vegans, but anyway, um that that so it's worth emphasizing particularly as I don't want to hurt anybody here that in these sort of cries it's usually about the role, right?

The idea is that you're not going to round up admins and you know um hang them by their network cables, right? Um the idea is that role goes away and in fact admins even the crankiest perhaps particularly the crankiest system admins uh believe the same thing. So this is a flashback to the '9s uh where uh uh admins ruled the world and uh this if you can see it says go away or I'll replace you with a very small shell script. And this is probably again I would say in that intersectional kind of area something of the motto of uh this particular group trying to fund the commons. We're trying to fund the commons.

The money is useful. Thank you very much. Um uh checks welcome. Um but also part of what we're trying to do is we're trying to automate the commons, right? We're trying to make it so that the commons is self- sustaining or regenerating.

We're trying to get to the point where I see Kevin Kelly in the audience. Um uh there's a great line uh in uh Out of Control, which is a book that you should all read, where it describes the moment where the terrarium or the aquarium pops, right? Where you no longer have to have the aquarium admin kind of dropping in things and trying to make sure the biosphere is careful and going, "Oh my god, there's a carbon monox carbon dioxide collapse. We got to put more things in." There's a point where the system pops and the system becomes self- sustaining like a like a a blockchain uh like a library system uh like uh like all of the things that that aren't owned and and controlled and don't have a single point of failure and don't have an administrator but they are resilient.

Um and we have a couple of strategies for for doing this right. Um, one of the strategies involves uh loadbearing people, right? So, what we do is we build a system, but it requires keepers, it require it requires maintainers, it requires administrators. Uh, and we rely on those people and they don't need to be, you know, uh, masters, right? They don't need to be gods.

often they're maintainers or distributors or librarians or um system administrators or coders, right? So we have that um and then we have lowbearing code where we write smart contracts and we write systems that that take on the bearing of that and distribute these things through. But the two things about these these these descriptions is that they are both loadbearing, right? Um, and that means that if you take those things away, the whole thing collapses when you're trying to carry a big load. And now I'm looking at slightly terrified at this like big thing over there in an earthquake zone, right?

You you want to have multiple loadbearing things. And so that grants you a little bit of resilience, but you always have to be aware that there might be one that will become the single point of failure and um everything will uh will subsequently break. Um so what we're doing then is trying to automate this right we're trying to take advantage of the promise of automation and this can be small scripts increasingly it's AIS and algorithms um and we're using those to uh to take on the load balancing right so let me this is where I dive into concrete stuff so one of the uh I'm from the Falcon Foundation Um, and one of the things that we do in the Falcoin ecosystem, which is literally funding our commons, is a thing that we drew from uh, another part of the web 3 ecosystem, and this is retro PGFs. And I won't go into the details because I only have 20 minutes, and I promised you I wouldn't create fake detailed descriptions. Um, but basically, it's it's it's like an award system.

Like, we have a uh we put in a pool of money. People come and say, "Hey, I did this really great thing. I produce this open- source software. I run this API for free and then a bunch of people vote um on chain. Uh and uh and we distribute that money through and we distribute it via um another another sponsor actually and this is if you're interested about doing this in your own ecosystem, uh Drips is great.

Drips um is is the loadbearing code of this equation, right? So we sort of there's a we have a giant pile of falcoin and they have sort of an automated spigot that scans GitHub looks for addresses. So we say hey we want to give it to this particular uh project. We pour the filecoin into the smart contracts at the top and drips not only distributes it to the people that that that are part of our system, but those people in turn can redistribute that like a fountain um or one of those chocolate fondue things. Um to uh I don't know why it flashed through my head.

I'm kind of hungry. Um the uh to their dependencies, right? So, so the the contracts kind of distribute this and it's great because this becomes the loadbearing system and our hope long term for uh the Falcoin ecosystem is we'll actually do a huge chunk of our public goods granting in this way not only retroactively but forward uh and we'll actually have a sort of network of these things redistributing um our um the the support that people want to give to public funds throughout all of this. Um, so the practical side of this is that this sounds very simple. Um, or maybe it doesn't.

Uh, I'm in the middle of it and it's incredibly complicated and like there are two people who aren't here right now who are in the middle of sort of semi- nervous breakdowns. Um, and they are brilliant uh, theorists and engineers and they're having to spend a lot of time emailing people going, could you remember to sign this document and could you could you just send us the F4 address? we need an F4 address rather than an F1 address. And could you and and so so this is loadbearing people having to take on some of the burden of the the loadbearing code cannot encompass. But there's also other psychological sides of this.

Like when we think of funding the commons and things like this, we think of money as a literally fungeible and somewhat uncomplicated little bit of incentive juice, right? Like you give people money, they they do a thing, right? Um, and as we all know, because we live in a moneyed world, money doesn't work like that. One of the things that you discover when you start trying to fund the commons is some people are like, "No, I don't want any money. Like, I'm doing this because I like the commons.

And if you give me money, then it's like you're going, "Oh, nice work keeping the whole of society going, kid. Here's $10." Right? It actually demeans what I am trying to do here. Some people will take the money but it will completely destroy their system.

Right? Say you give and I've seen this in the traditional granting uh environment right at the 2000s of the internet a lot of people were like oh we really need to fund online journalism because God knows how they're going to make money. So we, the X Foundation, will give $100,000 to this citizen citizen science project, right? And instantly that citizen science project just completely explodes, right? Because you get $100,000 pound dollars and somebody sits there going, "Okay, do I keep this?

Do I give it to everybody who is involved in my citizen science project? I have 10,000 citizen scientists, right? They're all volunteers. If I give them all this then they get $10 each and that's insulting right? If I give it to a small cuddery of people who become who are the professional citizen scientists, they're no longer the citizen scientists, right?

They're the professionals. They're the they're the the administrative costs and everybody else gets pissed off because they're like going, "Well, why am I working for free when you people are getting all the money and and the story of money not acting as an incentive, but actually acting as a disruptor to common's behavior is incredibly wellnown. So, so now we have this sort of weird balance, right? Where we have like a funding and incentive system that draws on all the lessons we've learned from web 3 that is a sort of machine for for motivating people to to do things. And let's be clear like we have been incredibly successful in this, right?

We have built um a a huge side I mean the the falcoin network I mean I'm not this is just because it's the first fact that I can pull out because I can't remember the bitcoin statistics but like we have built a distributed storage system measured in exabytes simply by going if you do this and you do this you will get a block reward and you will you will um you you will get rewarded and so do this thing and now we have a public commons of storage capability that that that is equivalent to, you know, uh pretty pretty mid-stage Amazon, right? Um so it works very well, but you also need this human element where you're sort of going, "Oh god, is this going to actually do it? Are we going to have to play with the numbers here? Are we going to have to redirect this? How do we change it?

How do we build governance?" So the danger I want to point out and the the the contrast here between loadbearing how am I doing? Loadbearing code and lowbearing people is if you generalize loadbearing code to think of it as loadbearing systems. One of the great successes in coordinated action over civilization is we've become less dependent on gods and and masters as our incentive system as kings and presidents um and more in building systems like a market systems like um an international community systems like these commons that that that is code it's code as law it's code as practice it's code as norms Right? But that takes some of the bearing.

But if we build something and that loadbearing code begins to bow under our assumptions or something changes and we can't change the code, pretty soon things bounce back to our default our in a case of system breakdown smashclass right which is loadbearing people. And I think probably the implication I'm giving here is that this is a sort of uh Napoleon versus the French Revolution um kind of uh moment. And maybe this is something that is hanging in your head right now, right? The idea that when systems break down, people suddenly pull and look for strong leaders and strong men. um uh and that leads to a sort of loadbearing person who can easily break the whole system because you just can't depend on them.

I think the first time I used the phrase loadbearing on this stage was to talk about Elon Musk, right? Where Elon Musk was increasingly becoming this loadbearing person in before all the recent interesting stuff, right? But but seemed to me to being being crushed under that whether it was consciously or not. Like his role as a person was disappearing and his role as a as a an artifact, right? A simplification of his humanity was beginning to come into effect as the other systems around him.

uh and perhaps his trust in those systems be began to began to collapse. Uh but it doesn't have to be if you if you if I mean I'm in San Francisco so I assume that like you know Elon is no longer kind of the the model of you know what what we would want and our presumption is you end up with the sort of Napoleon figure and that's not only bad but sort of morally wrong in some way. There are other sort of moral veilances to this falling back on um on loadbearing humans, right? Being a loadbearing human is never good whether you are uh uh a moral person or an immoral person. Um the other time that I used the phrase loadbearing I think in this in this room was with Aaron at Aaron's memorial um where I stood here and I think that that the impression that we all had was that that that as a community um we had asked uh Aaron or perhaps Aaron took upon himself um more weight than than than a a person could take.

Um so this is this is a very personal kind of story here because I think often what happens is we think that the code can take it right we're builders of systems um and we don't realize the weight that either we as individuals are um as open- source maintainers right um as uh individuals sort of trying to solve all of these problems for everyone else and it ends up um uh crushing the the person and then that has a knock-on effect. And like I it just occurred to me as I was putting this this slide up that like some of you who are worried about the stresses on the existing larger system um may have been tracking the DNC, the Democratic National Committee um in in the United States uh picking a new leader. And one of the candidates I don't he he didn't get through was Ben Wickler. Um, and uh, Ben Whitler was definitely of of Aaron's generation. In fact, Aaron and Ben worked pretty closely together.

And it was a moment where I realized that like in a alternative world, this would be that Eron worked very hard for the Democratic Party in lots of different ways, none of which were legible to them. Um, uh, but but Aaron would have been in that forefront of people taking up that leadership mantle right now. And of course he was he was prosecuted by a democratic um uh prosecutor. So this is what happens, right? If you put too much load on folks um you you don't have them when you you most need them in the future.

So um so I'm sort of wrapping up. This is sort of my experience with with literally funding the commons. And it's not kind of I I don't have smart contract code for you, right? I but I do have some lessons that I think I've picked up trying to work. either working with institutions that have lasted or uh watching systems fail.

And I think that that the key thing here is like don't have loadbearing code, right? Don't depend on everything where you pour all the money into the smart contract. And if you are an admin, if you are a coder, if you are a programmer, you probably know this, right? You're just already feeling guilty about like your code, right? Where you're going, "Yeah, I've got to build redundancy in.

I've got to build error checking. I've got to build tests. I've got to be able to pull out my code and replace it with better code or other code that isn't so um uh stressed when that's necessary to happen and I can't depend on a bunch of fixed constants that may disappear at any moment like so no spots right basically no sole no single points of failure um the alternative to low bearing is low balancing right so And this isn't just low balancing people which is sort of the lesson of I think one of the lessons of civilization right is that um you know uh uh uh when we have people who take on roles but those people can be taken out of those roles and we train people to take on those roles and those roles are clearly defined then we know society runs better. Um but also no load balancing yes balancing pit between people and code. When we're building these smart contracts and things like that, we need to have a relationship between the code that are operating is operating it.

We have to have a relationship between the systems that we build and the maintainers and the administrators and everybody else who who cluster around that. I think again this is an audience um that understands that code and humanity uh are in a a a co-evolutionary environment, right? We're we're we're we're secretic with one another. Um and I think it's sometimes easy for a lot of reasons for us to divide one way or another where we go people can exist without code. I'm sorry.

You live in a world whose natural history is determined by technology at this point. Um, and uh, uh, code, it can survive without people. Um, but but it it has its librarians, it has its maintainers, and there has to be a connection between those two things. Um, so I'm still workshopping this. This is not quite no spots, no no um uh uh uh uh no admins.

Uh I don't think I can quite fit this onto a laptop sticker yet, but I'm continuing to work and I would appreciate your input. So, the important thing here is like, and I see this happen so much. Oh god, so much. Right. I'm so sorry that we put personal computers on you and then told you you just have to sit in front and feed numbers into this Excel spreadsheet, right?

That was not the idea. Trust me. Um, the role of computers is to make the goals of sentient actors, including some computers who are sentient actors, right? My AI friends are right with me on this, easier to achieve, right? We're we're aiming for goals.

The purpose of computation and algorithms and systems like this is to make those goals easier to achieve, right? It's not an end in itself. The role of sentient actors is not to make computation easier to implement, right? Which I think is what I'm trying to get at with this thing. I think that the the the tendency for us models are fine, right?

Simple models are fine. We we we've understood in some ways how to contain and control those. But sometimes I think the pressure to to uh uh uh uh our desire not to make a system complex pushes against this idea that organic systems the systems that we live in have a high level of complexity for a number of very good reasons. Right? And if we try and create our build our computation to implement our models rather than reflects the organic system that we're trying to enable or empower, then we're we're wandering in a different place.

And that's the thing that leads to lack of resilience, anti uh fragileness, um and it leads to single points of failure, whether they're a particular human or a particular smart code contract. Um, so, so that's it. Uh, as I say, this is a work in progress. It's a little weird. I would love to talk to people about it because I think there are some rules that we can do that will not only make our code better, um, but make our, uh, lives friendlier and easier and let us protect, um, the many people that we love who are working so hard in this in this system.

So, thank you. [Music] Thank