Transcript
it is critical, right?
So let's start actually with the basics.
What exactly does an XRPL validator do
and how does the network reach consensus
without mining or staking,
which is another, you know, connected topic?
Yeah, an XRPL data on the XRPL really does
only one thing primarily
and that is transaction ordering.
So basically without validators,
the big problem is how are you going to order transactions
in a ledger, right?
And you need sort of to break a tie
which transaction came first.
Because if you have all no, only nodes on the XRP ledger,
the problem is when there is a transaction submitted,
there could be conflicting transactions, right?
For example, someone sending two transactions
for the same XRP, right?
Someone trying to spend the same XRP twice,
the classic double spend problem.
And to solve a double spend problem,
all you really need is transaction ordering
that globally is accepted.
So if we all accept that your transaction came first,
then mine came second, okay?
And the third transaction that I made
also came as a third time here,
then we solved it, right?
We only need to agree which transaction
sort of came first that the network has seen, okay?
And then we have solved the double spend problem,
that's it, the double spend problem,
really it's just basically that you try to spend
the same XRP twice.
And you solve it by saying,
well, this one came before the other,
and that's it, then you solved it.
Because if we know that for the first transaction came
before the other and we agree on that,
then we can just declare the second transaction has invalid, okay?
You are a jerk, you tried to spend the same XRP twice
invalid rejected, but for this we need transaction ordering.
And XRP like evaluators do exactly that.
They order transactions.
And by, and how they exactly do that is basically
as in a collaborative environment with multiple validators
and they vote on a transaction set.
So the interesting thing I just said
like is transaction ordering, but the exact order
is also irrelevant, right?
So whether your transaction came first
or my transaction came first or the two transactions
I tried to double spend,
whether you put one before the other also doesn't matter, right?
We just need to order them,
have a sequential order of these transactions.
So what validators are doing is basically
they vote on transaction sets, okay?
So on candidates.
And whatever, in a few milliseconds,
this will be solved and then you gain a majority
on one transaction set.
This is how the algorithm works.
And then you basically accept this for the next ledger.
We call this sort of we have a lunch towards a transaction set.
So there's like consensus towards like one transaction set
over time the consensus algorithm is made
that you have a lunch to an agreement, okay?
To let's say a specific transaction set over time,
even though people propose multiple ones.
And then there you go, right?
That's it.
Now you have transaction ordering, you agree globally on it.
That's the job of the validator, right?
And from an excerpt, so on,
and where these lists come into place,
these validator lists is that unlike proof of work,
where the person with the biggest mining equipment
and energy, right, is sort of their transaction ordering
is taken, right?
You can order transaction in a block and that's taken.
Proof of stake the same thing, right?
If you order a transaction and you have enough stake,
the probability that your block is being picked is higher,
eventually will be picked.
So you are basically the decider of the next block structure.
And on the excerpt ledger, we don't have proof of stake,
we don't have proof of work, we have proof of association.
So people can associate themselves with anyone that they want.
And how this exactly works is basically
that you put on a list that you have on your own note,
a list of basically a row of validators
that you want to listen to.
So you can say, hey, I want to listen to,
I see will down there, I see copper down there,
I see Carlos down there, so three validators is perfect.
I can say I want to listen to those three validators.
And they do the transaction ordering, as I said, right, together.
And I put them together in my list and there you go, right?
Now I have basically created my own list with three validators
and I'm going to listen to them.
And the more people listen to the same list,
the more people basically join a network, right?
Because they agree on the same transaction set.
And the excerpt ledger foundation is publishing exactly that.
They publish a list of recommended validators.
And that's what a lot of people in the ecosystem are using.
And that's where Flair was added towards to.
So by adding it to the list,
and because a lot of people use the list,
the excerpt ledger fund, the Flair has become one of the notes
that majority of the nodes on the excerpt ledger
are listening to for transaction ordering.
So yeah.
OK, this made me think immediately
about how this MIV if anything works.
But perhaps that's a question that we can leave until the end,
because it goes in a different direction.
But it's super interesting.
I think I've never thought about how would that work.
You are always thinking on the EVM,
or at least with a proof of stake, right?
But anyway, for the people that are just tuning in,
please make sure to follow along.
And if you have questions, I think we have a list.
And actually, if Sarah is listening, which I'm sure she is,
send me on a select, because right now I'm struggling a little bit,
having just one screen to have the questions,
have the space, and the control there.
So it's going to be more difficult for me to scroll down
throughout the questions of the community.
And we want to open up the mic at the end.
That's another thing.
And of course, make sure to reshare this space.
So we get to more people.
As you can hear from Beth's explanation,
this is really an interesting topic.
I'm pretty sure most people are not aware of many of these details.
And let's move forward with the conversation.
So anyone can technically run an XRPL by data
if I understand correctly, right?
So the next question would be why is being included
in a recommended by data release, such an important distinction,
right?
What kind of changes once a validator joins that least?
Because the XRPL ledger works with association.
So other networks prioritize, typically,
either in some shape or form money, either mining equipment
or steak, which is tokens, which you need to buy.
This is money.
On the XRPL ledger, it's association.
You are free to associate yourself with anyone.
And the more people associate themselves with the same people,
they construct a network.
That's basically the same people who
are listening to the same notes.
And on the XRPL ledger, that's done through these lists.
So anyone can create their own list and just put their validators
on.
But of course, you need overlap.
If you are listening to a list that
has completely different validators than I do,
then we cannot be on the same network.
We're going to fork from each other
because you are listening to something completely different.
I listen to something completely different.
And that's fine because we opted for this.
I opt for not listening to the same notes as you.
And you say, I don't want to listen to the same notes as you have.
So making your own list is very important.
And as a convenient way, a lot of people
also can listen to published lists.
So someone can create a list, put it on the internet,
literally on a website.
And your validator can just pull that.
And the foundation is publishing exactly that.
So it's a recommended list.
Say, hey, look, we recommend this list for the network.
You are free to use it or not.
And a lot of people are using it because they
are trusting the XRPL ledger foundation
and their recommended validators that they put on.
And that list is under maintenance always.
Like we're working on it.
And when there is validators who drop off,
they're going to get removed if there are new candidates
that make a lot of sense.
They get added to, like Flair.
And that's how we develop them.
Once you get added to that list,
because I said this list is used by a lot of people.
You automatically are being listened to by a lot of people.
I like to compare this to a friend group.
Imagine someone who is super popular and knows basically
everybody.
Now you are new to the friend group.
Let's say Flair.
You are new.
And this very popular person, in this case,
the XRPL ledger foundation introduces you to everybody.
And now you have suddenly all the friends.
So it makes a huge difference if you are being introduced
by someone who has a lot of friends or someone who knows nobody.
And that is basically the analogy.
And that's why it matters.
If you are being introduced by a very popular person
to a large group of people, a lot of folks
will come to you and say, hey, they're
going to connect with you quite literally on the network.
And if that wouldn't be the case, you can still go to everybody
and connect.
You can go to each individual person,
try to connect.
It's just like harder.
And that's why it matters.
Or is such a difference, whether you
are being introduced by, in this case, metaphorically
by a very popular person versus if you are not.
But you're free to connect.
That's the analogy of running a validly.
You're free to go and connect with everybody.
Nobody's going to stop you.
But it just makes a difference if you are
being introduced by a very popular person.
Yeah, I kind of get it now.
And certainly this is more like on the reputation side
of things, rather than, like you mentioned,
like with proof of state and with mining,
which is money, really.
So it's a really interesting comparison.
Certainly one of the things that makes the XRP layer unique,
right?
And moving into trusted validator lists
and more about decentralization,
I think we should define what exactly is a unique node list.
And just to be a little bit repetitive here,
but just to be super clear, who decides which validators
to trust and what's the difference between the list
that are published by the XRP layer
foundation and all the parties, such as, for instance,
repo.
Sure.
So every node on the XRPLager has a UNL, right?
Every node.
So if you run an XRPLager node to submit transaction,
to trade on the DEX, meme coins, whatever you do, right,
you typically run a node, OK?
If you're not pulling from public info,
but you run a node.
And this node has a list, OK, because you
want to listen to validators that order transactions,
as we said, that's the main task.
So all of them use a UNL.
So this is why, actually, it's inaccurate
when people say the UNL.
There is no the UNL because every node has a UNL.
So which one do you mean?
Do you mean the node number 100 or the node number 500?
Node list.
So there's node the UNL, actually.
But when people refer to the UNL,
they mean the overlap.
So when all of these nodes run their own stuff, right,
their own list, OK?
And you need an overlap to connect, OK?
The minimum has to be 80%.
Research has shown it's actually a bit more, like 85%, 90%.
But it's in that realm where you need
a lot of overlap with people.
So it means if two nodes run their own UNL, which they do,
then you need, at least, when you look at both lists,
around 90% same validators to connect with each other
so that we can communicate because of the transaction ordering.
You need to always keep in mind why is that the case?
It's because of transaction ordering.
If we disagree on transaction ordering in an x-block,
my chain looks different than yours.
And then, of course, we would like fork.
So about 90% of my friend group needs
to have an overlap with your friend group.
So basically, 90% of our friends are the same people, OK?
So that we are basically in a group.
And everybody has their own friend group.
This is very important.
Everybody, every node has their own friend group.
And this is basically the quote-to-quote UNL
where people refer to it.
There is also not really the UNL, but people use that.
What people mean with the UNL is the default UNL.
So the default UNL is what in the XRP ledger software
published a reference implementation by the foundation.
There is a UNL list recommended.
There is already one in.
Because if you have no list, when you start your network,
the node, it does nothing.
It doesn't listen to anybody.
So the XRP ledger foundation puts a recommended list in
so that you can get started.
So there's a list in it with values.
So when you start your node, you immediately
connect it to the network that the foundation is sort of
recommending.
You are free to change it and do whatever you want after.
This is completely up to you if you are disagreeing
or anything like this.
But this default list is what we can recommend.
And in our reference implementation.
And so when there's two options, either you create your own
list as a node or you listen to somebody else's list.
And a lot of.