All right.
I think this is actually going to be a pretty good follow-up to Lou's presentation.
But yeah, I'm Jameson Lopp, Chief Technology Officer at CASA, and at CASA, we are primarily
concerned with enabling users to be self-sovereign, and of course, there's many different aspects
to that.
But what I'm going to talk about today is kind of related to self-sovereignty, but much
higher level type of talk.
So we have to begin, of course, at the beginning.
What is a blockchain?
It's a chain of blocks.
I'm a technology guy.
When people say blockchain, I hear database.
So instead of this overarching theme where people are just saying blockchain this, blockchain
that, I think that that is glossing over a lot of the details that are very important
to these systems.
So when you create a blockchain, all you're doing is creating a linked list of data, a
new type of data structure, and it's cryptographically linked together.
So you get tamper evidence from that.
That's all you really get in addition to ordering.
You can say this thing happened after this thing.
But other than that, most of the other stuff that people think of when they say blockchain
is not actually guaranteed from the blockchain itself.
What is a blockchain not?
It's not a consensus protocol.
It's not a network of nodes.
It is certainly not an arbiter of arbitrary truth.
It may be able to tell you certain things, depending on what the consensus protocol actually
does.
It's not even a timestamping service.
You have to have other parts of a protocol in order to ensure that the consensus around
when these blocks were timestamped is in a certain range.
It's also certainly not an immutable or finalized record.
The blockchain itself only gives you tamper evidence.
You need other things such as a proof of work or proof of stake or some sort of other consensus
mechanism that makes it very expensive for someone to rewrite the blockchain and make
it immutable.
And of course, a blockchain is certainly not this diverse ecosystem of participants that
come together to build it, use it, and evolve it.
So my main point is a blockchain is not magic.
You can't just sprinkle blockchain on a problem and come up with a revolutionary solution.
You have to carefully consider all of the different aspects of what is making that blockchain
function and how people are interacting with it.
So to get to the real meat of this talk about bureaucracy, what is bureaucracy?
It is basically the creation and enforcement of policies.
And bureaucracy is how human civilization has gotten to where we are right now.
We have created these hierarchical command and control systems over the past several
millennia to help us organize ourselves, to help us specialize so that no longer do any
of you need to actually worry about growing your food and the entire process of sustaining
yourself.
Instead, you can delegate that off to other people who are specialized and probably work
for companies and other hierarchies to be very, very efficient and productive at doing
one or two things.
So the result of this, though, is that you have a system where there is a lot of power
concentration at the top, and this power is basically being used to coordinate the other
layers of people and things that are actually getting stuff done throughout the company
or public sector, private sector, really any organization of people.
This is very efficient, but of course, it has some tradeoffs.
And I don't think that as a society, we've really thought about these tradeoffs very
much until recently.
So bureaucracy and governance in the real world and most organizations, they're very
similar.
Usually, the governance is what initially creates a set of bureaucratic mechanisms,
and then it's self-reinforcing over a period of time.
The governance and the bureaucracy are basically a closed loop, and usually policies are used
to continue to extend and update the policies.
On a distributed crypto network where we've got a protocol and a bunch of people that
are running software and nodes around the world, it's basically the same thing.
And we'll talk about a little of how this actually works at a technical level in a second.
But because the protocols are essentially enforcing rules that you have agreed to run,
you are then able to enforce these rules 24-7-365, and only you really get to say how those rules
might get changed.
So governance is really more about who can change the rules, right?
There's many different types of governance.
I've got a few of the major ones up here, everything from a dictatorship to a democracy
to federations, republics, and of course, there's various hybrids of these things.
I would argue to you that a public permissionless crypto network is actually more on the anarchy
side.
And one of the reasons that you get into a lot of contentious debates around governance
of these public permissionless protocols is because people are trying to shoehorn them
into other governance models that they're used to in the real world.
So when we're talking about deciding changes, humans like to have what you could call a
decision function.
And this is essentially the way that we are voting to make the decision, because we have
high time preference.
We don't want to wait around for the rest of our lives trying to come to perfect consensus
about how to change a system, because you're probably never going to be able to get everybody
to agree on any one given thing.
So instead, we come up with these different types of governance, these different decision
functions that take inputs and give us one clear, decisive output.
But of course, there's a lot of pros and cons to this, the pros generally being that you
can make the decisions faster, the cons being that if you start talking about fairness,
there's probably no system of voting out there that everyone is going to agree is fair.
Have an example up here with the Ethereum coin voting that happened a couple of years
ago around a forking situation.
And we saw a number of things happen in there, one of which you can see only about 10% of
the total Ether supply was actually used to vote.
This is an issue of voter apathy that is common in all types of voting.
And then when you actually dig into the votes that did occur, you see that 25% of them belonged
to one voter.
And of course, this is just due to a concentration of wealth in that system at the time.
So you've got voting, that is how most governance systems work.
But how do you do it in a system of anarchy, where there are rules, but there are no rulers?
The closest thing that you can do is signal.
Put up some kind of flag, some kind of flair that you shoot off that says, hey, I'm going
to go this way, and if you want to follow me, you can, and if you don't want to, nobody's
forcing you.
So there's also kind of a moral issue at play here, where you could make arguments that
all types of voting are in fact a form of coercion, and whether or not you believe that
the pros and the cons weigh off against each other, that you would agree with that.
That's a debate that you can have for quite a long time.
But I think the main point I'm trying to make is that while you see people say that there
are types of voting in these systems, even the voting that does happen is actually generally
just a different type of signaling.
And when we're talking about different models of governance that people try to apply, I've
seen various models get applied, especially during Bitcoin scaling debates.
We see some people point at certain organizations and say, oh, this developer organization is
clearly the center of power, because they get to make the decisions, and then everyone
else has to follow along.
Or other people might say, oh, look at the miners.
The miners are the ones that create the blocks, and therefore, whatever they decide should
go into a block is essentially going to force everyone else to have to go along with that.
And of course, these protocols are extremely complicated.
There's a lot of game theory involved, and you can go around and around in circles for
years, as many people have done, arguing about how the balance of power actually works in
these systems.
And I think that's generally a fruitless thing to try to talk about.
Instead, what I think is worth trying to talk about is how self-governance works.
And hopefully you can read this, but if not, what we've got here is a very low-level model
of how data propagates around these networks, Bitcoin, Ethereum, what have you, is that
if you're running a node on this network, you're connected to some other nodes, and
they come up to you and they say, hey, I've got this piece of data.
It's a block.
It's a transaction.
It's some sort of message as a part of the protocol.
And they say, hey, do you want this data?
You look at it.
You say, oh, yeah, that looks valid.
It conforms to the rules that I agree with.
You accept it.
And then you turn around, and you talk to all the nodes you're connected to, and you
say, hey, I have this piece of data.
Would you like to have it?
And they do the same thing.
And this is called a gossip protocol.
It allows you to flood the network very quickly with any message, assuming, of course, that
the nodes accept it and relay it.
On the right, we have a situation where someone tries to pass a message that the other node
deems invalid.
And the interesting thing here is that they don't really have to take any action if they
don't want to accept it.
They just ignore it.
They drop it.
In some cases, they may disconnect from that client because they think they're being malicious.
But it is really the power of apathy or the power of inaction that I believe gives power
to this form of governance.
And the result is you have this huge collection of individual actors.
It's a huge mesh network that is composed of many tiny other mesh networks.
And they're all working organically in concert with each other to propagate information and
come to a consensus about what they collectively agree upon.
This is what I consider to be inverted bureaucracy.
This is an example of how a block propagates through the Bitcoin network.
It's actually slowed down 500%.
This happens in about two seconds.
And as you can see, these nodes, they're talking to each other.
They're saying, hey, do you want this block?
They accept it.
They validate it.
They relay it.
And then about two seconds, we can come to a global consensus about making updates to
the current state of a ledger.
When I look at this, it's really a thing of beauty.
I actually see it in the way that Ralph Merkel referred to Bitcoin as a new form of life.
This is some sort of new organism that we're still trying to figure out all of its complexities
and intricacies.
And so like I said, some people try to talk about balance of power a lot.
Michael Goldstein famously put this Twitter post up a year or so ago where he's like,
in Bitcoin, everybody's watching each other and they've got a gun on each other.
And if anybody tries anything funny, then they're going to take them out.
They're going to basically kick them off of their network.
And while that's a graphic way of portraying it, I think that instead you can think of
it from more of a defensive standpoint of like, we're connected to each other with ropes
and we've got a scissor and we'll cut you off if something goes wrong.
But the point being that I'm going to pose to you that in these networks, the power of
the veto is as strong, if not stronger, than the power of the majority to overrule other
people and that when we talk about fairness and equity, those are very controversial terms
in and of themselves.
I think that the most fair system that you can get is one in which any participant can
veto anything that they want.
And essentially, this gives you the ability to create a system where you may not be optimizing
for that which is the best for the majority.
I believe that it instead creates a system where you're optimizing for that which is
least harmful for the general user base.
So decentralization is a word that gets thrown around a lot.
And this is one of the things that I try to warn people about when we're looking at a
blockchain is that there's so many different facets that compose these complex systems
that if any one of them is particularly centralized or has too much power in one part of it, that
can actually become a critical flaw.
So you have to be very vigilant and careful when you're evaluating any of these systems
to make sure that all of the appropriate aspects of them have distributed the power as much
as possible.
A good example is that there's a number of tokens that are built on top of Bitcoin and
Ethereum and whatever where the tokens themselves have code in their contracts that allow a
single person or a single entity to censor and to make arbitrary changes to the token
holders within that system.
So you have to ask yourself if you're using this awesome robust censorship resistant network
but then you've created a trap door in the protocol you built on top of it, do you really
still retain all of the value of that network?
And so what am I really getting at with all of this?
You hear a lot of people talking about technical scaling solutions and all of the performance
problems that we have with blockchains because blockchains are probably the least efficient
database that has ever been created.
But I think that a lot of people are overlooking the issue of social scalability.
So what is social scalability?
Well Alfred Whitehead said civilization develops by increasing the number of important operations
that we can perform without thinking about them.
And if you think back to bureaucracy and how civilization has evolved with these command
and control hierarchies, that is the great question, the trade off of the efficiency
versus the resulting systemic risk that we create by centralizing power in the hands
of a few.
And so I believe that these blockchain based platforms can enable us to create systems
that are socially scalable, which means that the cost of participating in the network and
staying in the network is much lower.
And I don't mean the cost from a technical standpoint, but rather from a cognitive standpoint.
If you are aware of the idea of Dunbar's number, he's basically talking about human brain can
only really keep about 100 to 150 other relationships in play at any given time before we kind of
get cognitive overload.
And when you're in a system that the other participants in the system have sufficient
power that they can pull the rug out from under you, change the rules, and really change
the game, then you have to be worrying about all of these other participants and the power
and how they might use it.
But if we can build robust platforms where the power is so decentralized that you can
create a much more resilient and trustworthy system, then people can interact with each
other and use that system with very little cognitive overhead.
And the main point that I would try to get at here is that with blockchain based platforms,
we can create truly free markets that are socially scalable, where we can achieve this
by creating a system where you don't have to worry about all of the power dynamics and
the games that are being played behind the scenes, but rather you just worry about what
you yourself agree with.
And we essentially do that by inverting and automating bureaucracy to create these new
type of organisms.
And this is a very complex topic.
I hope I've been able to distill some of it in the short 20 minutes here, but I'm very
happy to talk about these issues and many other technical issues just to catch me after.
Thanks.
Thank you very much, Jameson.
Some interesting food for thought here.
Any questions from the audience?
Yes, please.
Yes, hi, can you please explain to us how you're using these incentive structures and
operational structures for CASA, which you meant self sovereignty, which I'm assuming
is...
Yeah, so actually with CASA, we are not trying to change any of the existing structures.
At CASA, our thesis is that the structures that are in the top public permissionless
networks are already very well formed and we should not try to screw around with them.
So instead, what we try to do is understand how those structures work and create software
and hardware that allows our users to make use of them.
And by that, I mean from a more technical level, allowing them to create and run and
maintain fully validating nodes that are essentially running this bureaucracy in their own home
in a way that doesn't require them to have a lot of technical overhead to understand
how to run and maintain the software.
And then from a privacy and security standpoint, to create additional layers of abstraction
on top of these systems so that it is more of a plug and play or mobile app type of experience
rather than the current experience that results in a lot of people having to use more command
line tools or using tools that have been built mainly by developers for other nerds.
So really, we don't think that the underlying architecture of the systems need to be changed.
We just are trying to help people better understand how they work so that they can then understand
the way that they can best operate inside of them.
Other questions, please?
Hi.
I'm Robin.
I'm in payments.
What I really struggle with is which parts you can decentralize and which parts you can't
and where, because it's so integrated, and so how do you determine which parts you can
take out and implement blockchain smart contracts, decentralize, and then still let it live in
the traditional ecosystem?
Yes.
And one thing I didn't really talk about is that, of course, there are a million different
applications of blockchain technology, and we're already seeing a number of enterprises
come together.
And some of these blockchains will not fit into an anarchistic model.
They will fit into a federated model.
A good example of that would be a number of systems, Liquid, Sidechain, for example, which
is actually pegged out of Bitcoin, but once you're into Liquid, you're operating in this
new federated model, and you get trade-offs.
You get faster speed.
You get new privacy functionality that doesn't exist in Bitcoin.
But the trade-offs are that you no longer have the same level of trustlessness.
But from an even more meta standpoint, trying to control these systems may be a fool's errand.
It seems to be more an issue of how was the system initially born, because once the system
is initialized and created, it is very difficult to make it more decentralized.
It's usually very easy to make aspects of it more centralized.
So I think that this is why you see, at least in Bitcoin, it seems to be a constant battle.
We're constantly fighting with each other about changes to the system, because some
people are extremely concerned that if we go even an inch in the wrong direction, we're
never going to be able to go back the other way.
So it's very easy to lose your power in any system.
It's very easy to become frustrated because there's no decision-making function, and things
can take a lot longer to implement changes.
So it's very easy for you to cede some of your power to another entity.
And I would argue that that actually is what happened with Segwit 2X and Bitcoin Cash,
is that certainly people got frustrated at the lack of change.
And so they came together and said, oh, we have the power, and we can go make changes.
And now they're going forward on their own path, and it's very fascinating to see the
results of that.
In fact, in a few hours, like with the Bitcoin Cash latest fork, I mean, it's just a fascinating
social experiment.
And none of us really know where it's all headed.
But I think the best that we can really do is participate and do what we can to try to
build these systems.
Jameson, Ian, Greg.
The blockchain presents a very antagonistic model of human behavior, which creates a lot
of competition, but also a lot of adverse effects.
I would suggest that if we can improve the level of interaction, dare I say it, introduce
more trust, we can compete for more benefit.
So firstly, do you agree with that?
And secondly, how do we increase the ability to do trusted competitive trade, which where
everybody benefits?
Well, so I think one of the fundamental issues, at least within any of the public permissionless
blockchains, is an issue that's fundamental to almost everything, which is limited resources.
And so a lot of people are fighting over limited resources that they have to put into supporting
these public networks.
But in general, this is a very free market mechanism.
I don't think that many aspects of it can be controlled, and that more competition in
general is only going to make it more robust.
Going back to one of the earlier talks with the cryptographer, he said even within these
systems, they're not fully trustless.
Like you're at the very low level having to trust the cryptographers.
Or most of the people here are probably not looking at any of the source code for the
actual node implementations, or your wallet implementation, or whatever.
It really is, I think, creating a hierarchy of trust in some ways.
And there may not be any way to get around that just due to the layers of complexity
that is involved in everything.
But how do we incentivize more competition?
The best thing that I can do is kind of what I'm doing here today to try to explain to
people what your options are.
Why should you be interested in using a system like this that is arguably harder to deal
with because you have to understand it better in order to fully make use of the security
and the privacy and the other functionalities?
I think end game to all of it is just building more layers of abstraction so that people
don't have to be worrying about this.
I do believe that, like most protocols, over the period of years and decades, the protocol
will eventually ossify, and it will be ridiculous for us to even be having discussions about
making changes to the protocol.
And I would give you historical pointers around looking at how the internet protocols like
TCPIP or SMTP or any of...
Even HTTP.
No one makes changes to these protocols anymore, and it's not because the protocols are perfect
and that they can't be improved.
It's because the trouble that it would take to make changes to them is not worth it trying
to go out and get consensus.
Rather, it is much easier to extend the protocols, build on top of them, build more and more
layers of abstraction.
So I think that that is gonna be the natural evolution that will eventually get mainstream
or majority inclusion is when we're able to make it just dead simple for people to use
these protocols.
The people that are using Facebook and social media and websites today, almost none of them
know TCPIP or any of the stuff that's going on under the hood.
And so I think that is what we should aspire to be like with these blockchain systems,
that hopefully someday, blockchain, it won't be a buzzword, it'll just be boring, boring
old technology.
Yeah, there's a blockchain under that.
There's a blockchain under everything that's important these days.
No point talking about it.
Fantastic.
Some great thoughts here.
Please join me in thanking Jameson.