Hey, how's it going, everybody? It's me, Rorar Diego, that one kid that you know.
You're probably not seeing me as much these days as you would if we were in the physical thing,
and I'm running around the village and stuff, but that also means you get to experience all
these other things that people do. How's DEF CON been for you guys, huh? Digital.
Is this something you'd like to repeat? Is this something you'd like to do another time?
I don't know. Hopefully, if we do decide to do it, it's not going to be under the virus
situation, but whatever. Daniel Kim's got some good stuff. I've been chilling in the Discord.
We have some people hanging out in the voice over there. If you have questions about Monero,
or you want to talk about pretty much anything, drop by our Discord in the Monero Village area.
We've got a voice chat where at least somebody's usually hanging out from the community to answer
questions. Okay. I am set to speak now. I'm set to speak today. I got a little bit of a half hour
slot to fill, but this time I don't have to fill it with a bunch of filler, because it's an actual
talk that I'm going to be giving. We're going to talk about decentralization in a centralized world.
Apologies to all the people tuning in from Europe. I'm spelling it with a Z, which is the American
spelling of decentralization and centralized, instead of an S. And when I see it with an S,
it makes me cringe inside. So, without further ado, let's go ahead and get started on this.
Decentralization. When people hear about blockchain, or when a blockchain project is
talking about their project, they're like, yeah, we have these goals of decentralization. We really
feel that decentralization is going to make all the difference. And that's one of the things we're
working towards on a technological level. And nobody ever stops to ask and think,
why? What is the purpose of decentralization? At this point, it's really just kind of this buzzword,
right? That they say decentralization, you go, that's good, right? So, if they make that a goal
and it's good, then that means that everything's good. But it's not quite that simple. It's not
quite that way. If we don't know what decentralization is for, if we don't know why
decentralization is a thing, why it's important, and if we don't know, like, the ways to achieve it,
then really we're just throwing words on things and nothing gets better. In fact, many things
would benefit from not being decentralized. And we're not going to be able to think about this
critically and see what things should be decentralized and what things shouldn't be
decentralized unless we know really what it's for. So, this is kind of a rundown of what I'm
going to be talking about right now. We're going to give some definitions about decentralization
and we're going to talk about the different forms that it can take place in. Because there is more
than just one form, really. When people say the word decentralization, they mean many, many, many
different things. Because it applies to many areas of a cryptocurrency project.
We're going to explore some trade-offs. So, when you choose decentralization, it's not all
positives. There are negatives that go along with it. And then we're going to look at a practical
application and see how that can kind of help us as we look at cryptocurrency projects in general
and evaluate them and try to think about them critically to see if they're doing something
good or if they're just kind of wasting resources. And yeah, that's where we're going to go. And then
from here, depending on time, I may give some light touching on kind of the world as we know it
and how we're moving into a more decentralized place. Because pretty much our world is very
centralized. As is the name of the talk. So, who am I? I'm Diego Salazar. I contribute to Monero.
And that's pretty much everything you need to know about me. There's my Twitter. You can go
to my Twitter. I don't do anything on my Twitter. I don't tweet hardly ever. So, yeah. I have long
curly hair and some people like that. So, if that's your thing, then give me a thumbs up in chat.
Okay. Decentralized. Decentralization. As stated before, this is basically just a new buzzword,
right? We have all of these things that have become buzzwords in the world. Blockchain is the
name of the game. But it is also a buzzword. Because as Daniel Kim was alluding to in his talk,
when people are like, oh, yeah, we're adding blockchain to the automobile industry. And,
you know, their socks just shoot way up. At this point, blockchain is a buzzword. Privacy
is also a buzzword, right? You have Bitcoin and all these transparency coins are like, oh, yeah,
you know, we're not a privacy coin. We're not privacy first, but we're adding privacy.
Decentralization is another big buzzword. And it's probably one of the top ones. If you look
at almost any cryptocurrency project, I can guarantee you that the word decentralization
is going to be somewhere in there. These people, this is they just know that this is desirable,
but they don't know why. So, let's go ahead and do the definitions. And like all boring people,
when I do definitions, we start with the dictionary. So, decentralization is defined
by Merriam-Webster dictionary as the dispersion or distribution of functions and powers.
So, then also we're going to define decentralize to move the control of an organization or
government from a single place to several smaller ones. And this is defined by the Cambridge
dictionary. And I would also expand that definition somewhat so it's not just a single
place to several smaller ones. It can really be, you know, from 3 to 5, from 10 to 20,
from any smaller number of places to larger number of smaller ones. And so, the word
centralize is just the opposite of that, right? We're moving from many places of power or
functions and putting them from many different places to less places. In the most extreme case,
into one place, right? One person can do it. So, we're going to give an example. We're going
to talk about... we're going to move into why would we want to decentralize? What is the point
of decentralizing? So, in this example, we all start a club, right? We start a really cool club.
We all love Twilight series. We all love the Twilight series. I can't think of a single
person on this entire planet that has not read the books, watched the movies, wrote fan fiction,
all these things, right? Everybody's crazy about the Twilight series. So, we start our
series club. And everyone is either on team Jacob or team Edward. And we need to vote. We need to
vote who is going to be the leaders. We need to vote on who's going to be the leader of our club,
who's going to be leading this thing. But everybody is in these two very political parties
of team Edward or team Jacob. And I am a team Edward fan, maybe. And you're a team Jacob fan,
so I don't want the team Jacob people to be in charge because they're going to cover our
Twilight clubhouse in team Jacob stuff. And that would be awful. I want Edward posters on the wall.
So, I want to make sure that whoever I'm voting for is part of team Edward. So, everybody's gonna
vote, right? Let's say we have 20-ish people. Everybody writes their thing on the ballot and
we all put it in the box. But here, a problem begins to arise. I jumped a little bit ahead.
We have to count these votes. We have to count these votes. And who are we gonna put in charge
of these votes? In theory, this voting is anonymous. We don't know who voted what. But
does it really matter what anybody voted? And let me tell you why. If we put one person in charge
of counting all of the votes, and this person is a team Jacob person, as signified by the J here,
then they could just, like, either not really count the votes or count them but not really
care and say, oh, Jacob won. This guy from team Jacob won. And we would all have to trust this
person because we didn't count the votes. Right? And this is just not something that we want to
do. So this is on one end of the extreme, where we get one person to do it. On the other end of
the extreme, we get everybody to stay. Everybody is staying here to count the votes. But the kids
are getting tired. And it's been a long day for me. I want to get home. I don't want to stay here
and count all these votes. The only way to be absolutely sure with zero trust of who won this
is if everybody, absolutely everybody stays, and everybody takes out each vote, looks at it,
passes it around, and everybody's watching everybody else to make sure we're not doing
some sneaky substitutions or anything like that. Right? So everybody has to stay. Everybody has
to look at each vote and keep their own tally. Then we compare our tallies to each other. And
if we all come up with the same thing, then there is 100% assurance with zero trust that we have a
correct vote. That's the other end of the extreme. But that requires everybody to take their Saturday
to stay here. And we don't want to stay here. But as we just discussed in this first example,
if we have just this one person here, well, they can go ahead and report whatever they want,
and we would never be the wiser. So what else can we do? Well, we can decentralize. So from
this one person, we've now moved to two people. Right? And in this case, we're hoping... we're
hoping that one is Team Edward, one is Team Jacob. So that way, we know for sure that, you know,
they're not gonna collaborate. Because what if they're both Team Jacob? Then they're like, okay,
so we both want Team Jacob to win. So we're going to agree together that we're not gonna really care
about the actual vote count. This is just what we're going to be reporting. So we have to kind
of make sure that we get somebody who's on Team Edward and somebody else who's on Team Jacob.
So that way, when they count the votes, if one person tries to lie, the other person is
incentivized to say, no, that's not true. No, that's not true. But there are some attacks that
can happen here as well. This is not a bulletproof, foolproof kind of thing. Because what if one of
two things happen? I've got a couple examples. What if the guy that is known as Team Edward
is actually a secret closet Team Jacob guy? Right? He doesn't tell anybody. He's actually
changed teams recently. Or he's been convinced by a family member. Or he's just lying for whatever
reason. And he is converted to Team Jacob. Now we have two Team Jacob people. And it's the same
thing that I discussed. They're just gonna agree, okay, it doesn't matter what the vote is.
We're just gonna say that the Team Jacob guy wins. Alternatively, another scenario is the
first Team Jacob person is quite wealthy or powerful or has a lot of resources. And they're
able to say, hey, I will give you money, resources, whatever you want, if you agree with me that it
doesn't matter what the vote is. We just agree Team Jacob wins. And maybe this person cares more
about the resources than about their allegiance to any particular team. So, they are persuaded.
Well, let's try to decentralize even further. Because we don't want any of those attacks to
happen. We want to make sure that this is a fair election. So, we decentralize even further.
And we say, okay, we got four people now. Hopefully two of them Team Edward, two of them
Team Jacob. But now we get to some really interesting kind of game theory things that
are happening here. Because what if we're able to get one person to our side? Either through
either of the ways that we were talking about before, where we were able to convince them
through resources or they were a secret agent all along, whatever the case may be. Now we're kind of
this three against one scenario. And these three people can threaten this person. Hey, if you don't
keep quiet about this or if you don't do what we say, then we're gonna burn all your posters,
in real life scenarios it can be things like they can threaten violence, they can threaten
family resources, whatever. And now this person is gonna have to go along with them because they
have the numbers advantage. So, we've decentralized into a place where in theory it's like, okay,
four people is good. But not everybody has to stay and waste their time counting all of the votes.
But we have enough people that we can trust that nobody's gonna do anything weird. But in this
scenario, it's not enough. So, we can do this again and again and again and again until we're
all the way back to everybody having to stay to count the votes. So, if you're seeing a pattern
here, I hope you're seeing a pattern here. And this pattern is that the more we trust people,
the more convenient it is for everybody. Because if we only have one person counting those votes,
that's a lot of trust placed in that individual. But it's the only person inconvenienced is that
one person. It's much more convenient for everybody else. They get to go home and watch
television, right? And the less trust we have, the more inconvenient it is for everybody. And this
is basically decentralization in a nutshell. This is going to apply everywhere across the board.
When we look at blockchain, we are decentralizing a database, right? It's just a database that has
ledger information that says this person has this much money and this person has this much money.
But we're decentralizing it. So, it's now not on just one server where we have to trust the system
administrator is not going to do anything bad with that. Or maybe they're not malicious, but
they're incompetent. And so, hackers get in and they mess with the database. No. We give the
blockchain, we give this database to whoever wants it. And now because everybody has it, we don't
have to trust anybody. So, if I'm like, oh, do I have an accurate copy of the blockchain? I can
compare it to many of my peers and say, okay, yes, it agrees with all of them. So, we're all good.
We're all fine. But it's much more inefficient. It's much more inconvenient. Because if you want
the most trustless option, you have to download the whole blockchain onto your computer.
Now, you have to take up your hard drive space, your bandwidth, your everything to be absolutely
trustless. And when we start thinking about the fact that this is always going to be the tradeoff,
the more we decentralize, the more inefficient it is, the more inconvenient it is,
we need to start asking the question, what things need to be trustless? This is the core question.
What things need to be trustless? Because those are the things that definitely we can
start to decentralize. Knowing that there is going to be a cost. But not everything does
need to be trustless. When we start thinking, okay, to give a very stupid example, and sorry
to call people out here, there was a coin back in the 2017 boom called CampusCoin. And the idea
was that CampusCoin was a cryptocurrency that students would be able to use to pay tuition
or their meal plans or all these things, right? So we're trying to make paying our tuition and
paying for our meal plans for our students a trustless endeavor. And I just didn't see or
hear the outcry of students that were like, these universities are cheating me. I paid the money,
and they said they never received it. I don't trust them. We need to make sure that they're
in check. We need to make sure that this is a whole trustless kind of thing. I didn't see this,
right? This is something that does not need to be trustless. Same with Dentacoin, paying dentists,
right? How many people have been defrauded by their dentist? I mean, maybe there's some of you,
and you might raise your hand and say, I see a need for Dentacoin. But I think for the majority
of us, we're able to pay them and they're like, thank you for your money. Here's your teeth
cleaning, and we're good. So kind of in this whole blockchain area, it makes sense in some
spaces. For example, distributed ledger technology. There is strong motivation for people to try to
mess with money information. Why? Because I can become a millionaire or a billionaire overnight
if I have the capability to mess with the money situation. Or I can be malicious against somebody
else, and they have $100, and now they have $0. There's strong incentive. Money is such a strong
incentive. So kind of this ledger-based information, there is a lot of reason why we
would want to trust people less. So we may be able and willing to decentralize, become more inefficient
so that way we can trust people less. But I mean, we just can't do that for everything.
But here's the big thing about decentralization in terms of blockchain.
It's, as I alluded to earlier, it's not just one thing. When somebody says, we are a decentralized
project, which you hear pretty much everywhere, what does this mean? We're going to start kind
of poking into many of the ways that a project can be decentralized. One way, who writes the code?
Okay. So all of these wallets, all of this protocol, all this stuff needs to
be written by some entity. Usually, like pretty much always by people, right? Who is actually
writing this code? Is it one person? Is it 10 people? Maybe they have 20 CEOs and one engineer
that's actually writing the code. Is this decentralized? They have 20 CEOs, so 20 people
looking over making sure the project is okay, or 20 people in high CT positions, CEO, CTO,
these kinds of things, right? But they don't know the code well enough to be able to check
on this guy or whatever. So functionally, there's one person who is writing this code.
And he has no oversight, which means that this person can put back doors inside. They can,
you know, they can do whatever they want. And nobody's going to be around to stop them.
Who's writing the code? Finances. Who has the money? Who's distributing the money?
If there's one guy or just a very small team of people who is handling the money of a project,
then they can choose... it gets a little subtle here, and it gets a little not quite so on the
nose, but they can hire engineers that maybe they're less than scrupulous, or maybe they're
hiring engineers that are incompetent, or maybe they're hiring engineers that adhere to a certain
design philosophy and not to another one, so they don't have to have as much infighting to get
the... to get the project to go in the direction that they want it to go, rather than another
direction. This kind of stuff is important, because when we're talking about decentralization,
trusting less, are we just like, you know, the development one is the one that's most well known.
Who's writing the code? How many eyes are on the code? But the same thing with money.
How many eyes are on the money? How many eyes are on the money? Who's paying to either hire
somebody or fire somebody or to get this research done? What if, you know, there's two
papers, two research papers that are done for two different, totally different directions,
that we can take a coin in terms of evolving the protocol? Well, the person with the finances
decides to fund an audit for one of those papers, but not the other. Once that audit is complete,
they can say, look, this way has been audited, and they said it was good. This other one is
unaudited, so I really don't think we should do that. Well, just like that, you have influenced
a community because of money decisions. So it's also, decentralization also needs to take place
kind of on a financial level. If there's only one group or a small group of people that has
all of the money, that's something you also need to watch out for.
Research. Who's doing the research? Because, you know, just as I alluded to in the audit thing,
perhaps a researcher comes across two or three different ways that things can go,
you know, evolve from a cryptographic perspective, and they decide to really
pour their all into one of them that they think is the best and not put as much effort into the
other ones. They're going to have a lot more. Do we trust this researcher? And many times,
many times, yes. Many times we think they're not malicious, right? But we can never be
absolutely positively sure. When we're talking about decentralization, when we're talking about
making things trustless, it's not just on the code. It's not just on the money. It's also on
the science. Who's doing the science? And practical stuff. Practical stuff. Who's running
the day-to-day? Who's running the community management stuff? All of these are ways. All
of these are ways that decentralization needs to be looked at. So, when a project says,
we are a decentralized project, maybe they mean in one way, like we have several different
engineers or we're open source. Usually what they mean is we're open source, which means
there are several eyeballs on the project. But when you start digging deeper, you're like, oh,
but we do have a small group of scientists that's pretty impenetrable. Like, as in, it's hard for
you to do your own research. And they're the ones that do the majority of the research. Oh,
who handles the money? Oh, well, see, we've got this little group that's responsible for that.
What kind of oversight do they have? Oh, we just we make sure they do good stuff. So,
they start hand-waving a lot of the centralized areas. They start hand-waving away a lot of the
centralized the ways that they are centralized. And they're like, and we want you to only focus
on the fact that our development is decentralized. Here's what I'm not gonna take a lot of time to
get into, but mining, centralization, decentralization. This is something that,
you know, Monero tried to do with RandomX. Asics tend to lead to centralization of mining over
a period of time to particular people that are able to create new Asics and stuff. And
we don't have time to go over in that. Daniel Kim has kind of discussed this in his talk.
So, I would give that a look. But basically, the goal is to explain to you guys that there
are many ways that something is decentralized. And particularly when a project is centralized
in many areas, I would even say in most areas, they do their best to divert attention to the
areas, very few, that do exist where they are decentralized. And that's usually the protocol
level. Because blockchain by design is pretty much decentralized as long as there's miners and
stuff. Like, it's just the way that it is. And they'll point all eyeballs to that and say,
see, we're decentralized when really everything else is centralized. So, we've gone through the
what. What is it? We gave some definitions. We've gone through the where, right? It can be mining.
It could be development. We've gone through the why. Why would we want to decentralize? Because
we want to reduce trust. Now, we need to talk about the how. And so, to discuss the how,
I'm going to say that I start a new project. This new project, SaveTheMarmots.com. It's not a real
website. I mean, maybe it is. You can go there and check. But it's not something I've created.
But we're gonna say I've started a new project. SaveTheMarmots.com. And you come in and you say,
okay, I want to judge whether this project is worth my time. I want to judge whether this
project is worth my attention, my money, whatever the case. Whatever this cause is for. I want to
make sure that I'm not gonna be wasting any of my resources on this. So, SaveTheMarmots. And I say,
it's very decentralized. It's very decentralized. And so, you're like, ooh, decentralized,
that sounds good. So, that's a couple points in its favor. I may put some money in there.
But we gotta ask the big questions. We gotta ask the big question.
First of all, does this need to be decentralized? This is always, always, always, always the first
big question that you should ask yourself. When they say, we are revolutionizing the real estate
industry, the automobile industry, the trash picking up industry, the dentist industry.
We're decentralizing all of these industries. We need to start thinking. Critical thinking.
Does this need to be decentralized? Is this something that would benefit from a reduction
of trust? Is this something where all the parties don't trust each other? Is this something where
we would be willing to tolerate the inefficiencies? What are the inefficiencies? They may not be as
neat on the surface. So, for blockchain, the inefficiency is it requires proof of work,
which requires electrical consumption. It requires power to secure the blockchain.
It requires the database to be on everybody's computer. That's a lot of space.
So, for blockchain, we might know what those are. But if we start looking at decentralizing
other industries, we may not know, it may not be readily apparent to us,
what are the inefficiencies that are going to come from decentralizing something?
We need to start asking the big question. Does this need to be decentralized?
But there are a few other smaller questions that we can ask. Because that's a big question,
and it's probably the most important. And maybe they're able to give you a good reason why they
think that this should be decentralized. They'll typically say, there is a good use case for this.
Usually it's a bunch of bullshit. But that's what they'll say. There's a good use case for this.
But there's a bunch of smaller questions that we can use to tease out whether, A, this thing needs
to be decentralized, and B, whether the project itself is decentralized. For example, who do I
talk to to get information? Who do I need to approach? Who do I need to talk to to get any
information about this product? If it is largely contained within one website, one person,
one department, right? If it's only a small group of people that is able to give you accurate
information about this, then that means that for the most part, the information is going to be
coming from a select group that's going to tell you what you want to hear. Are there dissenting
viewpoints? Are those dissenting viewpoints given a platform? Assuming they're not trying to spread
misinformation, which I really feel sorry for everyone in this day and age of social media,
where there's so much misinformation everywhere, there's so much noise, it's really hard to
distinguish between what is signal and what is noise, but that is another talk in and of itself.
But are these dissenters who have different ideas that are, in theory, in a perfect world,
presented in a respectful manner, and they have good points, and they address people's points
and stuff, are these people given a platform, or are they silenced? If they are silenced,
then that says that this is a pretty centralized thing, because you need a certain amount of power
to silence these people. If we look at Bitcoin and BTC, from the subreddit perspective,
there was a group of people that didn't like the Bitcoin subreddit, so they made their own.
They made BTC. And regardless of what you think about these two subreddits, I don't really care
what you think about these things, but these voices were not silenced, and they couldn't be
silenced. In one area, they were, but they were very easily able to build a following, build a
community of people who disagreed with the first one. Whereas many people, those things get
attacked, those places get shut down. It's really bad if you start getting like some cease and
desist letters that shows, okay, these people really do own pretty much everything, and it's
not decentralized at all. Question two, what is centralized, and why? So this is a bit of a
this is a bit of a funky thing that I'm going to throw in here. It's probably best
no, let me let me take a step back. Maybe not probably best. Some projects may choose to
centralize some things, and we need to understand why they do that, because maybe that's the way to
go. For example, payments. Maybe it's just easier to get people paid on time in terms of researchers
who decide they can devote their time and energy to this, or engineers, or people who are going to
code full-time. They say, you know what, we have decided to centralize the way we do our finances
so we can pay people on time so they don't leave us. If they can give you a reason for their
centralization, now at that point, you can kind of decide for yourself, well, is this kind of worth
the trust? Maybe, maybe not. But if they're like, oh, no, no, no, that's all decentralized, and
you're like, no, no, no, it's clearly not, right, they need to be able to point to parts of their
project that are centralized, and honest people, which are hard to come by, will tell you, yeah,
we're a decentralized protocol, but this part is centralized, this part is centralized, and this
part is semi-centralized, and here are the reasons why. Okay, now at that point, we are armed
with information. You can run it through your personal philosophies and worldviews to see if
that's something that you're okay with, but it's definitely something that is rarely talked about.
It's kind of hand-waved away. Well, no, it's not really centralized. It is, and they need to be
able to explain why. So talk to your projects, anything that you're interested in. And what can
I learn from history? There is a huge amount of examples of not just blockchain projects, but
open-source projects, business projects, everything, all over the world, where they have failed
because of centralization, and some have failed because of decentralization. There was too many
people that you needed to get an approval from to make a decision, and so you missed out on an
opportunity, right? You couldn't get all 10 CEOs to kind of give a check mark, or all 10 powerful
board members to give a check mark, either because you couldn't reach them, or they couldn't reach a
decision. So you missed out on a big opportunity, and they go under, right? So decentralization is
not always blanket good. Sometimes it can keep you from doing things in a quick fashion, but it
reduces trust. So there's a ton of examples. Please, please, please go look through failed businesses,
failed open-source projects, failed blockchain projects. Try to find out why did they fail?
Oftentimes, one of the underlying issues is that it was either centralized in some way or
decentralized in some way that it shouldn't have been. I'm running short on time, so
conclusion summary in closing. Decentralization is this big buzzword. It doesn't...
we really have to understand what it is, what its goal is, what it's trying to do,
because otherwise, when somebody says, oh, this is decentralized, we're just like, oh, yeah, that's
great, when really it may not necessarily be that way. So, yeah. That's my last slide. That's my
last second. It's 430. My time. I'm an hour ahead of Las Vegas. So it's time for me to back on out
of here. If you have any questions, I will be chilling around in the Discord. I'm more than
happy to talk with you guys about decentralization, what I think about various projects and what they
claim about decentralization, blah, blah, blah. So that's it for me. Thanks for coming to my little
talk, my DEF CON talk. I hope you guys enjoy the rest of the conference. Go check out everything
that we have to offer. Monerovillage.org has the rest of the schedule for the rest of today and
tomorrow. Thank you!
Transcribed by https://otter.ai