Architecture of Trust

Lawil: Welcome to another episode of the Future Money podcast.

And this episode, I want to introduce you to one of our newest hosts.

It's somebody you might've known already because she's been in our

community for more than five years.

It's Sabine Scheller.

She's the engineering manager at the Intelligent Foundation, and

she's always been interested in data ownership and the internet of value.

But today it's not about us hosts.

But it's about Joran, the founder and CEO of Tiger Beetle.

And in this episode, we'll dive deeper into his background

in business and coding.

And of course, his journey from Cape Town towards leading a global team, a

group of people whose collective aim is to revolutionize financial transactions.

Joran: Hey, Laval.

Thanks for having me and a pleasure to be here with you all.

And my name is Joran.

I'm a founder and CEO of Tiger Beetle.

A little bit about me.

I've always loved.

Business.

And since I was a kid, I was on the playground, you know, providing

marbles or, um, all kinds of things.

And I've, I've always loved this idea of creating something to serve, to make an

impact and make things that people want.

And I, along the way, I've, I've, I got into coding and I've always loved coding.

So I'm a self taught.

And I got into computer science research and self taught and all those papers.

And so I've always loved business, always loved coding.

I happen to be from a corner of the world that is close to

Antarctica, which is Cape Town.

And in my head, I work all around the world.

I work mostly in the States, but most of my team are there.

And I guess I love music and kitesurfing.

We could talk more about that.

Lawil: Actually, what I really want to know is your marble game.

Joran: I'm sorry,

. Lawil: It kind of like, it really fascinated me about you talking

about the fact that you had an interest in business and what is it?

Selling marbles or collecting the marbles?

Joran: Yeah, selling.

Selling marbles.

Well, actually I used to sell insurance for the marble games

so people would play marbles.

The problem that people were facing, this was the big pain, was that someone would

win a marble game and in winning it they, they would sometimes play with these

big iron marbles and damage and smash.

The glass marbles, so you would win the game, but you know, the marble

that you actually win is broken.

So what I used to do was I would sell insurance.

So you know, depending on who won, you would get a whole new marble that you win.

I can't remember how I worked up the economics, but there was insurance on

the playground in terms of marbles and it made marbles better because you win a game

and now you don't get a broken marble, but you get, you get something new.

There's a whole story there, you know, all the many different business things.

But I guess the biggest learning that I ever had was that business

is really all about trust and that that is what makes business happens.

Trust and trust is a wonderful thing.

If it's there, Things move so much quicker.

There's more velocity.

Everything gets better.

The more trust there is, it unlocks business.

So that, that took me a long time to learn, but that was the biggest learning

through all those years of running, whether it was selling computers, when I

was older, building websites for people, doing events and interesting things.

It was always like at the end of the day, it came down to trust.

Um, if people trust you, they, they will ask you to do things for them

and happy to pay you to do that.

And that's business.

Sabine: So if you are so passionate about business, why did you

decide to start a company like Tiger Beetle that is more techie?

And why did you decide to make it that global?

Joran: Yes, I think this is comes to, like, Why do we

decide what careers we go into?

My father's an architect.

So I grew up learning, loving, appreciating architecture.

He used to teach me things like form follows function and these, these amazing

things, like how do people move through a building circulation, you know, like

a building is like the human body.

There's circulation and people are moving through it because they

have a job to do or the building performs a purpose in society.

And so like, he taught me all these lessons.

And we really connect today on these things and like design, like design

is how it works and architecture and what, what is, what is art?

You know, there's so architecture is wonderful because it brings

engineering together with art clearly and obviously like it just has to be

so why, why software and technology?

Well, to me, it was what I realized was that.

If you go into architecture, the beautiful thing is that what you make, it almost

lives forever because it, these buildings, you know, that you design, they live on

and so you have so much responsibility in what you create, but also they do

live on, which is nice, you know, it doesn't, software is different because

you create it and a year later it's out of date or it's a bit rotted, you know,

but the nice thing I appreciated with software was that, You can make stuff

and you don't need the raw materials.

You don't need cranes.

You don't need bricks.

You don't need a team of civil engineers and all the other

engineers and quantity surveyors.

It's like the movie Inception.

You can go and be an architect in this invisible world and create things.

And you don't have to pay for the raw materials.

Which means that you can be like I was, 11 years old.

You can take books out of the school library and teach yourself

to code and start building these.

And it's, it's like you're building your own Minecraft, you know, and

anyone can do it like Ratatouille.

Anyone can cook, but coding is actually more true because you

don't even pay for your groceries.

You, you just need somewhere you can type and write code and you can create things.

So you can be an architect in an invisible world.

And then the other thing that attracted was business because you

can really change people's lives and let them do things through technology.

It's a real enabler.

And again, it's highly scalable.

So like a small team of people can change the world and let them communicate or

create platforms where people have jobs.

And you don't need a lot to do it.

So that, that was always what inspired me for technology.

Both of those, the, you can create and you can serve and you, you

need very little to do that.

Lawil: So for our listeners who've never heard of Tiger Beetle or of

a, what is an accounting database?

Can you explain it a little bit better?

Joran: Yeah, so Tiger Beetle is a financial transactions database.

What are financial transactions?

It's just another way of saying, let's do business.

So if you're a kid in the playground selling marbles, that's a transaction.

The marble moves from Andile to Bongani.

So it's a debit credit in the accounting sense, but someone is losing the

marble and someone is getting it.

And that, that is a business or financial transaction.

Well, it's a marble transaction, but it, it could be gold, you

know, gold marbles or steel marbles or, or just cash or insurance.

But these are all transactions all around us.

So, If you go into the mall and you tap your phone, that's a transaction.

If you receive your paycheck, it's a transaction.

If you buy some stock, you know, in a, in a company, that's even a transaction.

It's a big transaction.

So everything we do, if you book an Uber driver, that's a transaction.

You book a seat on a plane, it's a transaction in a cinema transaction.

So Tiger Beetle is a financial transactions database.

And that means that it is the engine.

Any kind of business would use.

To record the transactions they do as you do business, you're doing transactions,

but it's not even for business.

It's for foundations.

You know, as people donate money, that is a transaction and you record that.

So the whole world really, I think revolves around trust and goods

and services and time move from one personal place to another.

And that flow of things or information or value, that's all transactions.

And In this digital world, you know, that is all around us.

You need systems to record these transactions.

And so that that's what Tiger Beetle does as a database.

Sabine: I would love to hear more details, but before we go there, is your marble

selling game, was that the inspiration or what was the inspiration for Tiger Beetle?

Joran: Okay.

So I kind of, when I was 11, I got into coding and I loved it and I

realized here's this whole new world, create things, no raw materials.

You can, you're an architect and it's this movie Inception

and you're building things.

Then, so I always knew I wanted to go into business.

It was my passion, entrepreneur.

My grandfather immigrated from the Netherlands and he built up a business.

My father had his own business.

So that's always what I love doing is to serve people honorably at a profit

business, but great business definition.

Like the world always has these holes everywhere.

And the world society needs them to be filled by nonprofits or by businesses.

So we have bakeries because it makes sense that someone should get up at 4 AM in the

morning and bake bread for the community.

That makes a lot of sense.

So the community needs that to exist.

So, so kind of, you have, you have these needs everywhere and, and

that is why these things, you know, should arise is to serve those needs.

So this is macroeconomic, you know, the, the, the thinking, you

know, what does the world need?

And what I saw, I was quite lucky.

I was just playing soccer on a field and I bumped into someone

who was working on interledger and I said, Oh, that sounds cool.

Intelligent protocol, because it's connecting systems together

that transactions can happen.

And I didn't really know much about payment systems or financial

transactions at that point.

But what I did know is I always had this passion for business.

And because of that passion, I had an engineering scholarship,

but I actually gave it up.

And the reason why I gave it up, you know, not to go into like engineering

or, or coding was it was so that I could study financial accounting.

And the reason was because someone said to me, these debit credits, uh,

how you record mobile transactions and all the transactions of the world

already debit credit, something moving from one place or person to another.

And so I had this background that the way businesses can be expressed on paper in

the real world and the way it is done for centuries in every country virtually is

debit credit, double entry accounting.

It's like the first law of thermodynamics.

Money cannot be created nor destroyed, but merely moves from one account to another.

And So I had this background and that's why I studied that.

And so I had this love of coding, love of business, and the love of

the business schema, debit credit.

And then, you know, just lucky to be consulting as it happened together

with intelligent and coil, like on the, Bill and Melinda Gates, a central

bank payment switch or a central bank switch, which is called Mojaloop.

And that switch is really just doing debit credit at a national scale.

Like, you know, you've, you've got banks, those banks have got users and, or

people and, and those people want to send money, you know, from Mandili to Mungani.

And at some point that has been recorded and the money is moving from bank to bank.

It's these little, you know, Debit credits happening.

And so looking at the system and then seeing how they were building it.

And what was interesting was they weren't using a transactions database.

Well, they were, they were using what historically people, you know,

people thought was, but they were really using a general purpose

engine, a general purpose database.

to record transactions.

And this database wasn't giving you the debit credits.

It was giving SQL queries.

And so you had to do a lot of work as a developer before you could even just

Put transactions through the database.

And that was why the whole switch was being built.

So then looking at this problem, you see two problems.

The first is that there's an impedance mismatch.

So people want to record business transactions or nonprofit transactions,

everyday transactions of life through the switch with debit credit.

But the database to power this doesn't speak this language.

It speaks a much lower level language that is more takes a

lot of work to get right on top.

And that's also slow because now the there was the sort of limit to how fast

the switch could possibly go because it All the performance didn't matter

how much hardware you gave the general purpose database, how many machines, the

performance was just fundamentally kept.

And from, from there came Tiger Beetle, you know, let's make a financial

transactions database with that background of coding background of debit credit.

Let's just create an engine that you can drop into a car and suddenly it goes a

thousand times faster because the engine is powering the car in the right way.

It was meant for this.

Sabine: So making sure that you get debits and credits right on a

traditional database is very hard.

But when you build your database from scratch, you also need to

make sure that this is rock solid.

So how do you ensure that fault tolerance and the high availability on Tiger Beetle?

Joran: What a question, you know, that that's exactly the problem.

And I think also what's interesting is you can't.

It's not enough to be as safe as existing databases, either.

You can't just say to people, look, we're as safe as the status quo.

Because then there is, you know, there's some incentive to change,

but people will also look at you and say, ah, you think you're as safe.

So actually, you need to be 10 times safer.

But there's also an opportunity there.

So here's the context, is that The general purpose database powering most

systems today, most business transactions, it's either Postgres number one, MySQL

number two, or SQLite number three.

The number one and two on the podium there are both 30 years

old, give or take a year.

So they're 1995, 1996.

And if you think of it that way, you're like, wow, iPhones didn't exist back then.

There's been so much research in computer science in the last 30 years, hardware

has changed, we've got cloud didn't exist back then, NVMe drives didn't

exist back then, the scale of the world today, the volume of transactions was,

you know, in the last seven years alone, in many sectors, the scale has increased

a thousand times to 10, 000 times.

So, That's even in the last seven years.

And even the cloud databases, they were designed mostly 2012.

So 12 years ago.

But when you look at the most popular open source general purpose databases,

they're 30 years old, and the world has like just Change that was the nineties,

you know, the, the years of smashing pumpkins, one of my favorite bands, but so

much great music you can, you can sort of figure out, you know, with your coworkers,

some, I've got fantastic coworkers and it's awesome to see, you know.

Depending on our age difference, you know, have they heard that song by

Blur, uh, Song 2, and they've never heard it before, and then you put it

on, and like, they're hearing this amazing hit single for the first time,

and you know it's going to be amazing.

So I guess in musical terms, 30 years is not a long time, but in technology

terms, it's like an eternity.

So we got so much more research into how to design safer systems.

The hardware has changed.

And so if you want to build an engine today, you're not going to

make an internal combustion engine.

You're going to make an electric engine, you know, it's electric vehicle.

Actually there is no, there is no engine.

So everything is fundamentally changed in how you would build a financial

transactions database or engine today.

You would build it totally differently.

And so kind of, this is the opportunity we had was that, uh, yes, these

databases are tried and tested.

No, they don't use, there's 30 years of research since then

that they just don't have.

And so our systems today are suffering because they're not using

state of the art infrastructure.

And I think the time has come because of where the world is going with scale, that

to just keep up, you have to reset things.

So Tiger Beetle was very much, how can we design to be 10 times safer?

But actually you can because it's all there in the research

on how to do it, how to test it.

There's new ways to test.

So old, old databases are very much let's write millions of

lines of test code by hand.

Today with Tiger Beetle, we do autonomous testing.

So we spin up 100 CPU cores and they simulate Tiger Beetle, they actually

accelerate time, run Tiger Beetle in the simulator, like a flight simulator.

So that you're not in the past, you know, databases are like,

let's learn to fly for real.

And you're like, Ooh, like now, now, now this pilot is going to learn by getting

in the plane with people in the plane.

And that's how you learn, like, and that's a hard way to learn.

And obviously you don't, but that is how the history of databases,

like that's how you learn.

You just, You flew the thing for real.

And these days, there's three databases in the world that do this.

FoundationDB, um, Apples, Cassandra, and Tiger Beetle.

There's new techniques now that we use where you create a flight

simulator, and you run the database in there and accelerate time.

And if you crash, it's fine.

No one, no one gets hurt.

It's simulated.

But to do that, you have to design the whole database to be tested like this.

So that's, that's how we did it for Tiger Beetle.

Sabine: So you said that Tiger Beetle is inspired by research.

And I do know a little bit about that research that went into Tiger Beetle.

And maybe this is getting a bit nerdy now, but can you tell us a little

bit about the basis of Tiger Beetle?

Joran: Yeah.

Let's nerd out Sabine.

So we're going to go deep here.

We're going to rabbit hole like, you know, to the ends of the earth,

um, out to space and back again.

So this is the other interesting thing is that a lot of databases that are 30

years old, they were designed before what is called a consensus protocol.

So what I mean is in those days, you got your software on a stiffy disk.

And if you played games, you ran your game on one computer.

And if you wanted to play games with someone else, they had to be

in the same room with you, and you had to connect via Ethernet cable,

and that's how you played games.

And there was no concept of that you can play games with someone around

the world, or that we can have a chat like this around the world, but

how come we're not in the same room connected via Ethernet cable, you know?

But we don't need to be, there's the internet.

I guess there's a story there for Interledger, like what Interledger

is doing for payments, what the internet did for networks.

Uh, but that, okay, hopefully we come to that.

But, and then since that time, like of the 90s, already in the 80s, the research was

there, but people didn't really know it.

And that research was 88.

It was called viewstamp replication.

And this came out of MIT.

It was Brian Oki under Barbara Liskov, who was a Turing award winner.

And this research was how do we create software that runs across

more than one machine simultaneously?

So software that can run Like you can make a data center and make a data center

a computer, or you can take three data centers and make them look like one

computer, because if you can do that, a data center can disconnect from the

grid or have power loss or be destroyed, unfortunately, or something could happen.

But you have now two other data centers still, and you're still running.

So to the rest of the world, it's like, wow, the software is

running across computers, across networks, even across data centers.

So it's highly reliable.

But now that research is called a consensus protocol.

That's how you do it.

And.

It's probably one of the coolest algorithms in computer science

for me, and it's also kind of like the Everest, because to implement

these algorithms is very hard.

So because it's, it's software that runs across computers in real time, like

actually multi multiplayer games are even cooler and harder because they've

been doing this for years, you know, since, since those days of, of ethernet.

But, um, ViewStamp replication was the first pioneering consensus protocol.

A year later, Paxos came out by Leslie Lamport, very much influenced,

I think, by ViewStamp replication.

And then years later, 2012, ViewStamp replication was revised by Barbara

Liskov, James Carling at MIT.

And then two years later, Raft came out, which is very popular and raft

was more or less viewstamp replication.

The names were changed and a few things were like oversimplified, I think,

to make it for teaching purposes and that you'll see the raft paper, you

know, credits viewstamp replication that it's, it's so, so similar.

So really, really underneath all these consensus protocols, the

original is viewstamp replication.

And so that's why, for Tiger Beetle, we wanted Tiger Beetle to run across

data centers, because it's, you know, people's money is so critical.

These 30 year old systems, they're a single machine.

Yes, you can make backups, but they don't do the backups in real time.

So if, if you lose a machine, you are going to lose some portion of people's

money, or you're going to have downtime.

You can't have both because they don't have a consensus protocol.

So for tiger beetle, the hard decision was like, we actually need to climb

Everest and go and get ourselves one of these consensus protocols and

do it for tiger beetle so you can run a few beetles around the world.

And they've, you can't take them out there.

It's very, very reliable.

And that's why we have the simulator is that you can simulate, you know,

how to get this algorithm, right.

But that's really kind of what Tiger Beetle gives you is you don't, you

don't need to use proprietary cloud software or cloud database that you

can only run in one cloud provider.

Tiger Beetle is open source software you can run across multiple clouds all over.

So Amazon can go down and you're still running and then Amazon comes back and

Google goes down and you're still running.

But you, you can do this kind of like high availability with, with Tiger Beetle.

So that's, yeah, that's why we needed a consensus protocol and why we specifically

picked Fustam for education just to kind of pay tribute to that pioneer.

And it was the one that was most intuitive.

Ayesha: Hey there listeners, podcast, where we explore everything related

to digital financial inclusion.

I'm Aisha Ware, Program Manager at the Intellectual Foundation, and

we've got something exciting for you.

But before we dive back into the episode, I wanted to share a project I was

involved in collaboration with Carolyn Malachi and students at Howard University

Advanced Audio Production Course.

The course is a capstone for seniors in the audio sequence.

It is designed to challenge students to develop long form audio

productions, including podcasts, audiobooks, film scores, and more.

Through this project, students created interstitials about

Interledger for our podcast.

Here's a sample from Elijah Kirksey, a 2023 graduate of Howard University.

Stay tuned.

Elijah: Are you ready to be a part of something revolutionary?

Welcome to the Interledger Summit, a global gathering of people

transforming the future of digital finance through open technology.

It brings together innovators, regulators, and everyday individuals to

bridge the gap in financial services.

Join us as we hear from the trailblazers who are changing the game.

From community problem solvers to visionary artists, everyone has a voice

in building a more inclusive world.

It's more than an event.

It's a chance to build new connections, spark conversations,

and collaborate on ideas that have the power to transform lives.

Don't miss out on this opportunity to be a part of something transformative.

Visit interledger.

org slash summit to learn more and discover how digital finance

can empower everyone, everywhere.

Ayesha: Thanks for listening to Elijah Sample, enjoy the rest of the episode.

So

Sabine: you need a consensus algorithm to get everything global in sync.

You also have a global team.

How do you get that in sync?

And how far are you on your journey up the Everest together with that team?

Joran: In a way, it always feels like we're at base camp, which is a nice

place to be, you know, actually, one of the team has been to Everest base camp.

Yeah.

So that's, that's Gia.

She's, she's been there and done Kilimanjaro as well.

Like, so kind of, you know, this is the thing is like how you design your team.

Also impacts what you produce together.

So from the beginning, we've kind of thought, well, Tiger Beetle came out of

this open source project that was meant for the whole world, and it is meant for

the whole world, Mojaloop, you know, and Interledger as well, that goes with it.

And, and so we were like, well, Tiger Beetle is open source like these

projects and it's meant for everyone because you want to power new things.

So the team to build Tiger Beetle is also.

Global and distributed like the database.

So that's our focus that we, we're all over.

So we half the team are, are in the U S or in Brazil, but on Eastern time

and the other half are on London time in Europe, only two of us actually in

Cape town, one eighth of the team here and the rest are all over, but then

we also learn more, you know, we, get diversity of thought and, and craft.

And, and I mean, we can even just talk to people because a lot of

the people that we talk to are, you know, on, on certain time zones.

So you're in the same time zone.

Lawil: It's not a question concerning your team, but mostly it's more

of a future focused question.

How do you believe that Tiger Beetle has been contributing to shaping the ecosystem

sort of FinTech industry and beyond?

And do you have a long term vision on how it will shape the current ecosystem?

Joran: I think this was the thing, you know, that, that we felt

is everyone is building their financial transactions database.

So this goes back also to Sabine's question, like we're actually

always building a database.

It's just a question of, do we realize that or not?

Are we doing it intentionally or not?

And coming to your question level of, You know how this contributes

and will contribute, but I think this is what we saw was that the

central bank switch was building a financial transactions database.

It had to do all the all the work around the general purpose database that the

general purpose database didn't do.

Now you do it around that.

But that system is a database, but it's not built to database standards.

And so we thought, well, let's do this explicitly.

Let's make a financial transactions database, but let's share this cost.

Um, so we'll do it.

Uh, we'll make it open source.

And it's for the world.

Everyone can use it.

And now this problem, this whole in the world of every single

business has to reinvent its, its financial transactions database.

And, and many people have done this and I've chatted to people.

They thought, yeah, we were thinking we're going to go and build tiger belong.

And we saw, Hey, you built it.

Thank you.

So now like, What happens is it's not only Mojaloop, not only

Interledger, but it's many projects that are starting to build on Tiger

Beetle as critical infrastructure.

And it's open source, which is the beauty of those, you know,

post Christmas, SQL, SQLite.

They're the top three because they're open source and they have stayed open source.

And so going forward, I think the way we see it is that Tiger Beetle

is, it's about making a technical contribution, giving people a.

Financial transactions database designed for mission critical safety

performance, resetting performance, you know, going for 1000 X more

performance, which we also designed for.

There's a whole story there, but all of this is about looking

to the next 30 years or so.

So we've received 30 years of open source.

Each generation.

It's up to us to do something, you know, And I think at some point, like we've,

we've had this from Michael stonebreaker, people who did amazing things with

Postgres 30 years of open source.

And at some point, you know, each generation, I think can take it on that.

We're like, well, let's do something for the next 30 years in turn.

And that's the beauty of open source.

Lawil: What I really loved at the beginning of our conversation is that you

were comparing what is the engineering profession to architecture, architects,

actually, because they are the shapers of society because of the influence

they have with the city mapping.

So do you believe that a system such as what you're building right now, it being

open source, could follow that pathway as well, or could have that influence?

I know it's a lot to say over your own products,

but

Lawil: I'm just curious.

Joran: Thank you for saying that.

No, I agree with you.

Um, that's one thing I learned from my dad as well is that you do with

architecture, you have a wonderful opportunity and responsibility and things

like, you know, how do we design schools?

There was this thing, you know, where sometimes people would, would separate

communities so the elderly lived isolated, which is a terrible thing and far nicer

things that my dad worked on were how can we integrate so that the young can learn

from the wisdom of the old and the, the elderly can be energized by the energy of

the youth and have joy and 'cause that's real community when it's integrated.

So I don't know if we're tiger beetle.

Um, but I think the one thing that we do try is we try to give people

the financial transactions primitive.

So rich set of ways of doing debit credit.

It's not just a debit credit.

It's debit credit.

And maybe.

You, you do it now pending and then later on you, you post it as confirmed because

you're interacting with other systems.

So it's like debit credit at the intersection of distributed systems.

So we, we try to give people these primitives to make what

was previously very hard.

Even impossible to do.

We try to make it possible out of the box, but here's the thing is

that we try also to not do things.

And this is another architectural principle from Christopher Alexander.

It's like you almost want to wait.

If you're designing like university campus, you don't

want to lay the brick paths.

Wait, let the students in, let them walk everywhere.

And then afterwards at night you go in and look and you see now

where, where is the grass worn?

Because that's probably where we should put the path and

it's not where we expected.

So Tiger Beetle, we try to minimize surface area.

So we, we, We try to make impossible things possible, hard things easy,

but there's always a temptation that you try to do too much.

You try to not only provide primitive primitives, but you try to provide whole

features, and so we try not to do that and we try to say, look, We'll give people the

double entry accounting primitives, but not the policy, not the accounting policy

that that should stay with the coder.

We let them make the policy decisions.

We'll provide the safe primitives to record those decisions of debit

credit, you know, so if you book a ride, maybe there's the transaction

with the driver, there's the route transaction, there's the tip transaction,

there's partnership transactions.

But the policies around that, we don't take that on.

So we separate accounting policy from accounting primitives.

And the reason is that policy changes across jurisdictions.

It changes across business to business or entity to entity,

foundation to foundation.

It even changes from.

Teams, you know, different products have different policies that they need.

And I think this is key that I learned, but previously like the

core banking systems, they tend to do not only the debit credit

primitives, but also the policy.

So they mix it all together and over time it becomes more and more inflexible.

because now the paths are in the grass where the students no longer walk.

And so I don't know if this is a good enough answer to what is a great

question, but this is just what we try.

You know, like this is what we're responsible for and actively

look, we don't want this.

This is, this is for, for the world to figure out.

Lawil: No, I think it's a great answer.

The reason why I asked the question was basically because in the end

you're building something that could potentially be extremely

influential, but also there's a layer of direction, but also power into it.

If you do it on purpose or not on purpose, especially if it's open source and

people are using it, which is incredible.

And I love the entire connection about the responsibility as well.

Joran: Yes.

No, thank you level.

And maybe I can just add as well.

I love the question.

So this is why open source is so, so important because I saw there

were some technical projects that I really, really loved like ZFS

and unfortunately around just as it happened with CFS as a file system.

In a way, it had very similar properties to Tiger Beetle, that it

can survive disks that are failing and it was a big inspiration.

But just because of the history of how things worked and the licensing

around ZFS was never clear.

And so it was pretty much open, but because the licensing wasn't clear,

it never got used, but therefore it never really served because it was the

world's most incredible file system.

People are starting to use it, but for years we lost out.

And so I kind of with Tiger Beetle, I think, and if you look

to, you know, Postgres, MySQL, SQLite, these technologies are

infrastructure and they're, they're too valuable not to be open source.

And so I think some people, Think that open source and being sustainable

are at odds that you need to force people or you need to try to be

a monopoly and that's business.

And I actually think that is not business business again, comes back to

trust and open sources is the best way to build trust that, you know, this is

for everybody and it's not locked up.

So that's why I'm intentional when people contribute code to Target

Beetle, there is no CLA agreement.

So that it stays open source, we can't relicense those contributions,

but it's intentional because if you mess up the license, it will die.

No one will use it.

And so that's, that's kind of the thing that always saddens me when I see projects

that were once flourishing open source, they changed the license because they

don't really understand the business, I think, and they don't see that business.

is a healthy ecosystem, there is competition, there's all these things

and they and in so doing, they actually destroy the business, the the value

that could have been there, like intelleger is open source module loop.

The only way they can use tiger beetle is that it is open source.

And we don't we don't want our Transaction systems to be beholden to one company

or cloud provider or be proprietary, the core infrastructure must be open.

And actually, I think a lot of core infrastructure for the

last 30 years hasn't been open.

So I'm kind of curious to see what happens.

If it is, you could either buy this or you could use the open source stuff and build

it, but now what if we can give people this engine, this will be exciting to see.

Sabine: Well, as you know, at the foundation, everything is open source too.

We believe that this is the only way that other people can

learn from what we're building.

I feel like that's also your mantra that you live by, but I also feel like

there was a performance story that you've been very eager to tell us.

So tell us about Tiger Beetle performance.

Joran: So this is what I learned, you know, again, with IntelliJ,

Coil, Mojaloop is that the world is becoming more transactional.

And it's kind of lately I've been thinking of this as like a TV.

So the last few decades, if you look at the dimensions, like let's get our

ruler out and measure the dimensions of the TV screen, it hasn't changed

too much cinema screens, like the room size, maybe it's changed a

little bit, but the, the dimensions of the screen haven't really changed.

And what's the thing that's really changed?

It's the pixels of our computer screens have gotten so much smaller.

They're now they're invisible to the human eye.

Um, I remember when they were, the pixels were green and black and big,

you know, and you could, you could see a pixel and, and today, like with

transactions, you can see transactions.

You know, you can't do a transaction if it's less than

5, otherwise you really see it.

You also feel it with the fees.

But what I love with interledger is that it's asking the question of, you know,

the world is becoming more transactional.

The dimensions of like GDP or whatever is maybe staying more or

less the same, gradually increasing.

But the resolution of transactions is, is greatly increasing.

So pixels are becoming smaller.

We're moving to a world of like 4k color with what you can do and create

on top of these payment systems because the resolution is so much better.

Like I think next year already, maybe people are going to pinch themselves

and ask each other, can you believe that we used to wait a day for a payment?

To go, you know, within a country or a day.

Can you believe that?

Can you remember that time?

You know, when we used to buy albums?

Or even, can you remember that time when you used to buy songs?

Cause you know, music has become more transactional, high resolution.

Like now you don't even think of buying a song.

You just stream it's seconds.

And so streaming is coming to payments very much.

So, but what that means is now you need more powerful

TVs to drive 4k color display.

So you need performance tiger beetle.

And so like the world, the pixels are becoming a thousand

times smaller, literally like.

In the world of energy, one of the first Tigerbill users is an energy utility.

They realized that we're moving to clean energy away from coal.

If you're using coal, you use maybe bill your customers as an energy utility.

You build them once a month.

So if you've got a million customers, you've got a million

transactions once a month.

If you move to clean energy and solar, the price of energy is now following

the sun, which is changing all the time.

You know, it rises and sets day and night, but it's even changing with,

you know, cloud conditions, weather conditions, clear skies, the side,

even though it's winter, but the price of energy is changing all the time.

So this energy provider, what they realize is like, wow, instead of

pricing clean energy once a month, let's price it every 30 minutes.

Because that way we can give people more efficient energy, like now,

why should you overpay for your energy just because your provider

isn't being transactional with you?

Like you should be able to pay for energy every 30 minutes and that,

you know, going from once a month to every 30 is literally a thousand four

hundred and forty X increase in volume.

So that energy company is making their sector 1, 440 X more transactional.

The pixels are smaller and more frequent.

And so suddenly you say, you're saying to infrastructure, I

need you to drive the car.

Please a thousand times faster and the driver is like, no,

this car is not spec for that.

And so we, we designed tiger beetle.

We just redesigned the whole thing before with general purpose SQL, you would send

a lot of SQL queries back and forth.

over the network, half a millisecond each trip.

And that would mean that you've got this fundamental limit because

of Rolox, you can do a hundred, maybe a thousand transactions

a second business transactions.

Tiger beetle, we realized that's a problem.

You know, it's like putting one person in the plane, you know, from

Amsterdam to Newark, and then the plane goes back to get the next person.

But it's even worse than that.

It's like, let the plan go back and forth 10 times.

And now you've moved one person that that is a financial transaction

of a general purpose database.

It's let the plan go back and forth 10 times.

And on the 10th trip, let the person finally commit and go across.

And we were like, this doesn't make sense.

Like, because for the CPU, it's actually worse than that for the CPU.

It's like, let's go to Mars and back 10 times because CPUs are

so fast and networks are so slow.

The speed of light and fiber is a constant, never going to change.

So we were like, let's, let's put a hundred people in the

plane and send them in one trip.

And then you've done a hundred debit credit transactions because

the seats speak debit credit.

It's not like some other language.

And then we were like, okay, well, let's not stop here.

Let's put 8, 000 people in the plan.

That's going to be cool.

And we'll do one plane trip, 8, 000.

That is the key for Tiger Beatles performances.

We changed the protocol, higher level primitives, and we pack 8,

000 of them in one network request.

So one, one plane trip is 8, 000.

And now you can have planes coming from all over the world

into Schiphol and to Newark.

And.

Now the airport is also better for the airport because it's not

just one person each 10 for the trip, you know, or each 10 trips.

So now everything gets better.

And so this was the, it's very simple.

We almost, we didn't really do anything special.

You know, it was just trying to solve the problem.

So

Lawil: it's actually a good example.

I do have a question because you mentioned, for example,

the transaction costs as well.

What does digital financial inclusion mean to you?

Cause it's something that's very close.

To the hearts of the people who work, or at least for the intelligent people who

work for the intellectual foundation.

And if you think about it, if you think about the transition costs, but also the

latency, like what does it mean to you?

Joran: It's a question we ask as a team, we say, who will never ever know about

tiger beetle, and it'll be like the people who have like, for them, it's everything.

It's, it's their week's wages and they've sent it back on and that

that money never gets locked up.

For manual reconciliation that they don't have to spend like 50 percent of it on

fees Those will be the people that will never know about Tiger Beetle because

hopefully they they won't they will never have that experience anymore So for us,

it's very very important like who will never know about this and let's just

do our job right That the system never breaks, never locks up people's money.

And we're kind of lucky as a team because we have a very

narrow product surface area.

What, what I mean is we, the reason we exist as a group of people to the whole

that we have to fill is a very clearly defined, and we've just gone deep.

And we just haven't stopped drilling and things like this question of financial

inclusion and who is, is never going to know about Tiger Beetle because things

just work because the infrastructure can power it and it's cheap and efficient.

We take it for granted.

It's like electricity.

There was a time when electricity was so expensive that one person

in America could afford it.

Just one, there was only one person, you know, and, and now it's

like electricity, like switching lights on is cheaper than candles.

But there was a time when it was the other way around, and so it would be nice

that, you know, we make these things so cheap that people take it for granted.

But when you do that, a whole lot of other things become possible, so, so

this is kind of how, just to let you in on how we think about it, but it's

what, it's why we really care, because there's so many little things that can

go wrong with computers, like the quartz crystals that track time in a computer.

Can drift as the machines heat up in a data center, the clock can tick faster,

you know, according to the heat slightly.

And so, so some machines start to go faster.

This doesn't impact the consensus or anything, you know, the

correctness of the system.

But if you zoom out and think of who's actually using this in Asia,

in Africa, in South America, in North America, Europe, who's using this?

There it's very important that money moves between systems or doesn't move, but a lot

of that has got to do with Time because you you will say to another partner on

the network like this money should move and Within a certain amount of time it

should move back if something went wrong So if you get those times wrong It can

happen that like a clock jumps a year into the future and so someone's money

is locked up for a year I know that's really really bad because it might be

something that You know, that, that, that for them is going to change everything.

So, so we, we go really deep into like the hardware and the software and that's

why we designed a whole database because you kind of have to, all these things

are co designed to specialize for.

For who's the end user and how do we optimize their total cost of ownership?

Not, not how quickly can we make a database once we have shipped it.

Let's rather like ship it a month later because we want to optimize for 10 years,

20 years of, of the people using it.

Sabine: So as you know, our Intelligent Summit is coming up and

that's also all under the umbrella of digital financial inclusion.

And a lot of the end users will be there.

So what are you most excited about?

Because I know you're going to be there because it's in Cape Town.

So it's around the corner for you.

Joran: Yeah, I'm going to be there and the whole team are going to be there.

Like they all coming out as well for the summit.

So really excited.

We're a small beetle, but there's a few beetles and they,

they all come into Cape Town.

Uh, and yeah, We're really excited to meet everyone because that's

kind of what it's all about.

So we say like, do things in the most direct way possible.

And it's great to just get everybody right up with the community, like

who's using this and, and also learn and understand, you know

what, because that's why it exists.

So how can we improve?

Yeah.

And it's kind of, I guess we all encourage one another.

It's the, these things are happening, you know, it is happening and it is moving.

It's not only interledger or tiger beetle, but it's happening all around the world.

So it's quite exciting to see it's a real trend, the real, real time trend, you

know, and it's, it's spreading everywhere.

Which enables financial inclusion, you know,

Lawil: and I've heard that you're also sponsoring the hackathon.

Joran: Yes.

Yeah, we are.

I mean, intelleges are big brothers or and big sister actually my I've got three

older siblings and they're my sisters, you know, they and they So i'm the little

brother and uh, but we intelleges that for us too, I guess and so Intelleges

doing this this hackathon and we're just like You Coming in and if people use

Tiger Beetle, then most creative use of Tiger Beetle, or the most performant,

then, then there will be prizes.

So thank you for, for having us, that we can be there.

Cause it's pretty awesome to, this is what makes things worthwhile is.

Is, is how it's used.

So there's nothing like, like a hackathon, you know, for people

to actually hack with the stuff.

Lawil: I agree.

So we have now one question left and it's actually a question we ask all of our

guests and it's based on, what is it?

Um, kind of like predicting the future of, uh, finance or digital finance.

And we ask our guests to give us their thoughts on what they believe to.

30 years, maybe 50 years will be.

You don't have to reference or you don't have to think about flying cars

or what is it back to the future.

Maybe we could, but if you think about the future of the world and the future of

finance, what direction or what do you?

What would you like to envision?

Joran: Hmm.

Well, thank you.

How did you know this is what I'm speaking about at the summit?

I didn't really.

Okay.

My talk is how do we predict the future?

Literally.

And I haven't told anyone this.

So you, I guess, let me give us a teaser trailer.

So, uh, Nicholas, I think Nassim Taleb, he said prediction, not

narration is the test of understanding.

So can you predict something?

And being able to predict is a better test of understanding than narration.

Everyone's a storyteller these days, but who can predict?

So I'm glad you asked this question, you know.

And then now, I mean, predicting the future, that's hard

because it hasn't happened yet.

How do you predict the future?

And I mean, you didn't prepare me for this question, but you can see I'm prepared.

This is my talk, but a little teaser.

So how do you predict the future?

Hasn't happened yet.

And Alan Kay, he's a jazz, jazz musician.

He's.

Also a computer scientist.

He said the best way to predict the future is to invent it.

And now we need to ask ourselves, okay, so it's important to predict.

How do you predict you need to invent next question?

How do you invent?

And then that great inventor, Thomas Edison, he said, well, I look at what the

world needs and then I proceed to invent.

So he wasn't a great inventor so much.

He was a great archeologist of needs.

You know, he could find the needs and just serve those needs and you'll

be an inventor in passing, I think.

So what is the next 30 years of FinTech?

I think we have to look at what the world needs.

And then we need to see, you know, what are the goals to serve those needs?

What's the methodology to hit those goals?

And that's kind of the tiger beetle story.

You know, I think the world is definitely not going to want

slower, buggier databases.

It's going to want them to be a thousand times faster already.

We're actually concerned that we're not giving enough performance because

some sectors are already 10, 000 X.

We've only designed for a thousand X.

So it's kind of like the world is going to need faster infrastructure

because you can trade that performance for cost efficiency.

It's going to be wanting to run multi cloud by default.

Like I'm talking to my area, but for FinTech, I think regulators are

going to say, they're already saying, you know, you can't only be running

on one, one of three providers.

Cause that provider goes down and a third of the country goes down.

So, so open source is going to be critical.

That we're not locked in and infrastructure becomes commodity.

It has to be commodity and utility.

The feature is going to be faster, safer, better.

I hope at least that's what I, you know, that's, that's, that's what we're all

here for as a generation, isn't it?

So, and then our kids will, will look back at us and say, you

know, 30 years ago, there was this database called tiger beetle.

It's too slow.

Lawil: Thank you so much for answering this question.

We're at the end of the podcast.

So I really wanted to thank you for joining the future money podcast and

being one of our guests and looking forward into what is meeting you guys

at the hackathon but also at the summit because that's when we all get together.

So thank you, Jeroen.

Thank you so much.

Joran: Yeah, I'm really grateful to you for having me and Also,

Interledger Foundation, because it's also changed my life.

Financial inclusion, like you've brought me in.

I was doing other kinds of coding and this is now so much more exciting.

So I also want to say thank you.

Lawil: If you would like to learn more about Joran Gheb's vision of the

future, then you have two options.

Either join us at the Interledger Summit on the 26th and 27th of October

in Cape Town, South Africa, or Where he will present the next 30 years of

transaction processing, or you can watch the live recording of his presentation

on the Interledger YouTube page.

Ayesha: Hey, it's Aisha again.

We wanted to remind you that we are nearing the Interledger Summit.

Join the open payments movement where a global community of

changemakers is re imagining the future of digital financial systems.

We want to welcome you to our program.

Pre summit activities, the ILP hackathon on October 19th and October 20th, and

our community open house on October 25th.

Our official summit will kick off on October 26th with a dynamic schedule of

inspiring speakers and engaging sessions.

We hope you'll join us in Cape Town.

Lawil: Thank you for tuning in to another episode of the Future Money podcast.

And of course, make sure to subscribe and like our podcast on Spotify, Apple Music,

or wherever you listen to your pods.