Okay, well, anyway, we made a lot of jokes, so it's not wasted.
It was a lot of entertaining.
The panel that you are going to listen to and to watch is very atypical.
You don't need me to introduce the other panelists because they are Jameson Lopp and Jimmy Song,
so of course you already know them.
I will just introduce the topic because the topic is coming from a very quick conversation
I had with Rodolfo when thinking about what to discuss that could be of general interest
during this panel.
It would be mostly like a conversation among friends, so we will just pretend to disagree
a little bit and we will just try to step in and discuss the topic, and possibly we
will migrate from the initial topic to other contingent or entirely different topics.
The starting point would be censorship and regulation of minors.
So it started from the news, let's call it that, that a few weeks ago some news broke
that there was some attempt to regulate the transaction that minors could include so as
to start some kind of KYCing of minors first.
So the first point is identification of minors that so far can be just informal organization,
a mining pool, a mining farm, there is not necessarily a legal status to a mining operation,
not everywhere anyway, and there are a lot of anonymous blocks coming out from nowhere.
And then as a second step, the possibility that these legal entities that are allowed
to mine could also be forced to select the kind of transaction that they want to include
and the kind of transactions that they want to exclude, basically creating censorship,
selective censorship in Bitcoin.
So most people that commented these news were kind of scared and a little bit shocked, but
some of us were actually completely, totally expecting this kind of evolution in the battle
between Bitcoin and the state, mostly because it was part of the security model, if not
since the inception, because that's not the case, Satoshi was extremely optimistic about
mining decentralization, but at least since I would say 2015, 2016, for example, this
discussion reminded me of the crypto economics wiki on the LiBitcoin project.
If you go on LiBitcoin on GitHub, that's basically a library, a set of libraries for Bitcoin
maintained by Eric Bosqueville and initiated by Amir Taki, actually.
And Eric maintains also this kind of wiki about theoretical facts about Bitcoin, the
threat model, the economic model, and stuff like that.
And you can find some voices, some entries on this wiki, in which the model that Eric
is proposing and thinking about is explicitly, then eventually the government starts to seize
hashing power and it tries to compete, overcompete the rest of the market, trying to mine blocks
that actually only include sanctioned, permissioned transaction, which eventually, this is also
my thesis, eventually will go to coincide with empty blocks, because the amount of permissioned
transactions that really need Bitcoin is basically zero, because you need Bitcoin for permissionless
transactions.
Otherwise, you just use PayPal, which is cheaper and faster.
So the government tries to mine empty blocks.
And there is a competition between the government that can fund mining operations at loss, which
are censored mining operation, and the black market and the gray market of transaction,
on the other hand, that will pay more to be included.
And so when there is an economic equilibrium between these two forces, you basically have
some rate of effective transaction field blocks, which may be lower than one block every 10
minutes, because we may have a consistent rate of empty blocks, or even a consistent
rate of reorgs of a few blocks when the government goes above the threshold of 50%.
So my initial reaction, and then I will just leave it to James and then Jimmy to comment,
was that this is nothing really new or unexpected.
It is something serious to consider seriously, and maybe to counteract at the technical level
with better mine pool algorithm that are already being considered, and hoping that the technological
improvement in ASICs basically stops hitting some kind of hard bottom of commodification.
Because if that's happened, then the cost of entering the competition of market and
the competition of mining, even in the black market, becomes lower.
So my reaction to this news will be, there's nothing intrinsically new to this.
Maybe it's news to the 2008 optimists, but it's not really news to the prevalent trend
model that we were facing or imagining, let's say 2015, 2016.
I don't know, guys, if you agree.
Yeah, I mean, the model has been pretty much the same ever since the advent of pooled mining,
which I think was back in 2011 or so, maybe even 2010.
And what we're really going to be talking about here, because this is a fairly complex
topic, is we're talking about game theory, we're talking about economics, we're talking
about incentives.
And so in order for Bitcoin as a protocol and a network to be as robust as possible,
we want it to be decentralized.
And we want it to be decentralized in the sense that anyone who is exerting power or
pressure on the network, especially nation states that have monopoly on violence and
can exert power upon any large entity that they can find, that is where we start to worry
about one nation state having too much power over this one particular aspect of Bitcoin.
Historically, people have talked about China.
What if China does something?
And I've written about that even just in the past few months.
And one of the many things that we hope to continue to happen throughout the years is
that Bitcoin mining continues to grow and grow.
Miners continue to be incentivized to find cheaper and cheaper power.
That will result in more and more of the hash rate spreading out throughout the world as
it searches for that cheap, renewable, and often sequestered power that there's just
no demand for.
So there's a lot of different variables we can talk about that will come into play when
we talk about anyone given nation state being able to exert censorship on the network.
Yeah, I would add that from a technical standpoint, this attack on pools, which is essentially
it would be a regulatory attack, one where the government would say something like pool,
if you let a transaction with let these transactions through that are not KYC, then we're going
to punish you in some way.
I'm not sure that's that workable because you have stratum 2.0, which is coming out.
And I expect that to become a much bigger part of the pooling ecosystem.
And part of stratum 2.0 is that the individual miner that's in the pool gets to decide what
transactions go into the block.
So in a sense, the pool, I mean, right now, stratum 1.0, the pool gets to decide what
transactions go in, but that decision is taken out one step further into the miners that
are actually doing the mining, the actual people that own the miners and they can switch
at any time.
So in a sense, if you do have KYC pools, then all those people that contribute hash rate
to those, they're going to leave and go to ones with stratum 2.0, so they get to decide
what transactions to put in because that will make them the most money.
And that in turn means that the regulation essentially is ineffective and doesn't really
do anything.
And that seems to be the case with a lot of different regulatory attack vectors that exist
in Bitcoin is that the government can try, but it's actually quite easy for the users
to get around it, for the miners to get around it, to get pretty much all the players to
get around it.
And the thing is, most governments are very used to controlling single points of failure.
That's their model is that they regulate that single point of failure.
Even in a mining pool, there's this perception that the pool is the single point of failure.
It's not really, it's not a single point of failure.
There's lots of pools and you can switch pools anytime you want.
So I don't think this works.
Yeah, there is a low switching cost, but I think getting a bit deeper into the weeds,
I'm not sure if anyone's really talked about what are the incentives for more pools to
adopt stratum V2, because if I'm not incorrect, I believe slush pool is the only one that
has implemented it so far.
Is there more that needs to be done for the competition to heat up on stratum to adoption?
Well, maybe I would say the profit motive though.
Why is it?
Is that more profitable that just the more well, so if you're, if you're the pool, if
you mine, whatever the pool tells you that might not be the most efficient, right?
Like if, if the pool gives you an empty block to mine, that sucks.
You're not getting any fees and it's fees get become more a part of it.
This is where we're talking about the fee market.
That becomes a much more critical part of the revenue that you're receiving.
So when you're mining, you want to make sure you get the most fees and there might be other
reasons.
You might have some transactions of your own that you want to include and you need them
to go through relatively quickly.
There might be all sorts of different incentives there, but, but generally if you give more
power to the, to the end user or the end miner in this case, then you're going to be able
to be more efficient with it.
Right now though, you know, it's a, the market isn't mature enough to take those things into
account, but at some point that that's going to come into play because you know, it will,
we will reach more of an equilibrium than we have now, right?
Right now because of price increases and whatnot, it's just insanely profitable even such that
people are putting like S9s back online and so on.
But at a certain point it becomes, you know, like all those little edges add up.
I mean, like a lot of miners, for example, don't use liquid cooling because it's not
worth it.
At some point that's that though, that edge is going to be sufficient enough that they're,
they're going to be incentivized to do it.
I see this in the same.
You know, even stepping back a little bit and I'm not sure we may have lost Jimmy there,
but you know, Giacomo said something about anonymous blocks and you know, that actually
got me to thinking, you know, originally pretty much all of the blocks that were mined were
anonymous and it was really the advent of pooled mining where I think it became kind
of like a status symbol to put a string in the Coinbase that basically said, you know,
we mined this block, but that has kind of created an attack vector, right?
Now perhaps it's good for auditing that someone who's mining with a given pool can then easily
look at the blockchain to see how many blocks that pool mine to make sure they're not getting
gypped by the payouts or whatever, but you know, perhaps the incentives here are wrong
as well and there needs to be an improvement or some sort of push to suggest that miners
should go back to, or at least the pool should go back to anonymously creating those blocks.
Actually, it may have been used as an attack vector actually during the Bitcoin main crisis.
There were some accounts of mining pools that were basically, they say, allegedly blackmailed
into signaling for the New York agreement hard fork.
Otherwise they would not have received the new hardware or something from the ASICS producer.
So basically we put everybody in the same, we tossed everybody in the same definition
of miners, but actually we have, Jimmy was clarifying there are the individual hashers
and these guys are basically the guys connecting the ASICS to the power and cooling down the
ASICS and basically they need an internet connection and some temperature difference
to cool down and that's basically all and electricity, of course, cheap electricity.
Then there are the hardware producers that are a very, very strong centralization point
because it's very expensive to build up an ASICS factory.
And then there are the pools that are coordinating.
So the theory of the scenario that Jimmy put forward is that the choice, the barrier to
switch for individual hashers will become lower and lower with new stratum protocols
and stuff like that.
And so basically you end up to a point where regulating the pool is difficult because the
pool is basically an abstract entity.
They don't need a physical setup.
The pool can just be an anonymous store on your service in theory.
You don't need a physical entity to coordinate.
The hashers, they are a physical entity that are actually very easy to spot because they
consume electricity, they produce heat, but they can switch easily and they can possibly
deny that they are mining the KYC blocks or not.
If they stop signaling their name, for example, that would be very, very good as a new trend.
This is an interesting point.
Yeah, it's worth noting.
I think it's kind of, I consider it a bit of a failure of Bitcoin mining, at least,
that what you were talking about, Giacomo, the concept of being able to have sort of
a pool-less mining or really p2pool, this peer-to-peer decentralized pool was what I
consider to be kind of the holy grail of improving and getting rid of a lot of those attack vector
sectors.
Unfortunately, it just never took off due to, I think, the complications around setup
for it and the fact that it added a bunch of latency, which could be harmful to profitability.
Unfortunately, while the sort of game theory security incentives seem to be there for us
to push for an adoption of p2pool, the economic incentives have not really worked out.
In a way, censorship may fix this, meaning that there are a lot of centralization forces
like logistical, electricity, and synchronization latency, but regulation is a decentralization
force because the more centralized you are, the easier the target you become for regulation
and eventually censorship.
Yeah, and one of the things about mining that is very unique in this regard is that they
can move.
It's not that hard to take a mining setup and bring it from some hydroelectric dam in
China and bring it to Mongolia for the winter during dry season, something like that.
And that's one of the features.
And the fact that they can move jurisdictions relatively easily, I mean, it still takes
a few weeks and everything else, but I'm sure they're working on making that process as
quick and smooth as possible.
But once you get more innovation in that area, it's going to be very difficult to regulate.
Like governments are not used to playing whack-a-mole like that.
They're used to just regulating a single entity and saying, you can't do this.
And that's the choke point at which they say you're not allowed to do X, Y, or Z.
That's a whole new ballgame because all of those physical miners are movable.
They can be taken out of the jurisdiction that's being too harsh.
Something like New York bit license would not work because essentially they would just
move to the next state or move to Canada or something like that and many have.
And that is the key here is that governments are not like, I don't think they're capable
of doing it because this is just so far from their normal paradigm of lawmaking and whatever.
And the miners have way more tools to be able to skirt those regulations and say, okay,
well, we can set up here, but we don't really need to.
It's like being a digital nomad or something like that.
You can go to whichever jurisdiction is most favorable to you economically and regulatorially.
And if the regulatory environment is tough, well, then you're going to go somewhere else
where the combined cost is the least.
And those are just simple economic incentives that point towards a future where it's more
and more decentralized.
I'm curious on this regard what you guys think about one of the few comparison that actually
makes GPU mining way more interesting, which is actually a plausible deniability.
So what I mean is that basically ASICs mining is arguably better than GPU mining because
it's closer to the terminonimical limit and there is not such a thing as ASICs resistance
anyway.
So eventually somebody can create an ASIC.
So the easier the ASICs to create, the better in the long run because there is a lower entry
to competition.
So there is no an argument for ASIC resistance.
And also many argue that the ASICs are better because they create a higher investment that
is completely wasted if you stop mining Bitcoin.
So they create the same cost for attacking Bitcoin if you don't protect yourself with
a short or something else.
So ASICs are good, but they are bad in a few ways.
One way that may be eventually solved by commodification is the monopolization of production, which
for now is less of a problem than a few years ago, but it's still pretty much a problem
because you don't have many alternatives to be made still.
But the other problem, which is in relation with what Jimmy said, is that you cannot really
deny that the SHA-256 ASIC is for mining Bitcoin.
Well, with GPUs and CPUs, you can just set up a data farm where you can just say that
you're doing rendering and then you can deny that you're mining Bitcoin and switch.
But if somebody comes to visit your, if somebody opens your container and you're moving jurisdiction
and somebody asks, what are these things for, you don't really have an alternative answer
to, I'm a Bitcoin hasher, right?
Well, you can.
It just depends on how creative you are.
I mean, they've had miners that heat homes, right?
Like you can disguise it as a heating mechanism and just say, I'm just too cold.
Or there are people that are using the mining heat to create, like to make beer, you know,
like whatever.
There's a lot of ways in which you can use that excess heat and, you know, a clever manufacturer
might, you know, to a negative jurisdiction may end up doing something like that.
Here's a beer making kit.
By the way, it's heated by ASICs.
Not that we're saying you should use it, but I mean, this is the sort of thing that happened
during prohibition is that they would sell these like giant cubes of like grapes and
say, you know, if you leave this in your basement for 90 days, it's going to make wine.
Not that we're telling you you should do that, but, you know, we're just saying you, I know
we know you want this giant cluster of grapes.
Just saying that if you leave it for 90 days, it'll make wine, you know, like which would
be illegal.
But, you know, yeah, that's the way it would probably happen.
Yeah.
We're speculating on, you know, plausible deniability versus actual realistic enforcement.
And so, you know, the issue with any of these things is if it's widespread enough, no government
has no matter how authoritarian has the ability to do 100% enforcement on the entire populace.
It's always a matter of what are the incentives and how much enforcement do they really need
to strike enough fear in the hearts of the populace to have, you know, effectively quash
it.
And what we're talking about here is, you know, an activity that is highly profit motivated.
So like any black market activity, the people who are highly profit motivated are going
to innovate, they're going to find ways to make that money regardless of whether or not
it's legal, illegal, whatever.
I also think that even if you take the most pessimistic view of all of this and assume
nation states that have 100% omniscience and omnipotence ability to do a perfect enforcement
on whatever laws they enact, is that then you get into a very interesting game theory
at the nation state level.
If we believe that Bitcoin is robust enough that it will continue to exist, then now you
have to start thinking about, well, okay, let's say that all of the Bitcoin mining operations
get nationalized.
And we've already seen that happen, I believe, in Venezuela where the military took over.
If I'm not misinformed on that.
So you could get into a situation now where even if a lot of these operations are run
by nation states, we already know that a lot of nation states don't get along with each
other and there's going to still be this continued friction where I would consider it highly
unlikely to ever get into a situation where all of the nation states around the world
could agree to collectively work on perfectly censoring the creation of blocks for Bitcoin.
So we're looking at at least three layers of protection against effective censorship
of transaction.
The first one is just a geopolitical game theory.
You need all the jurisdiction agreeing about that.
Sometimes it happens, for example, right now we are seeing that happening about tax arbitrage
because basically all tax havens are basically disappearing except Dubai and in a way the
USA itself.
But the USA managed to basically destroy many tax havens in the last 10 years, I would say.
So sometimes global coordination can happen, but it's very hard to do.
Then if you do that, you still need the physical enforcement against us ushers moving away
this kind of beer machines.
But it's very it's very nice if you have mining provision together with alcohol provision.
So you can just do the moonshining with with beer and Bitcoin together.
And then the third the third level that will see remain is basically the fact that there
will be a market competition between the empty blocks that are losing the revenue of the
legal transactions and the miners which are included in the legal transaction, which are
making more money out of fees.
Now, there is also another one, which is basically the everything about a second layer.
I just don't just mean lighting network closing and openings, but also, for example, today
there was this live experiment by Chris Belcher and other of of coins.
So one transaction that looks completely normal is actually enforcing another transaction
between parties that are not between the same parties, but which is completely different
economically speaking.
So you can have something like a white listed transaction, which seems to move money from
my key was see address to your key was the address.
But actually, it is a coin swap anchor to do something else on some kind of different
layer, which is another very strong layer of protection to avoid the pessimistic scenarios.
Yeah.
And the thing about censorship is that there there's it's really very impractical on Bitcoin.
Like if you if you try to do that, if you try to eliminate all coin joins and things
like that, you do get coin swap transactions and things like that, which which let you
do that.
And we have additional possible future coin join like things with cross input signature
aggregation or something like that, which would happen post taproot.
But you know, that would cause an economic incentive to coin join, in which case, you
know, like it would be, you know, evil or it would be bad for a company not to do it
because they would be not fulfilling their fiduciary duty to their customers and so on.
So there's a lot of possible ways in in which this will be very, very difficult to enforce.
I will point out, though, that like the little bit of global coordination that we've seen
from the United States mostly is because of the dollar hegemony that exists.
And they're able to enforce a lot of that because they have the ability to sanction
pretty much everybody else due to the fact that the dollar the world runs on a dollar
standard.
And if you don't have the ability to transact with dollars, it's very, very difficult for
any company or country.
You know, as we go more towards a Bitcoin standard, you know, those kinds of enforcements
are become much, much more difficult.
So on multiple fronts, Bitcoin makes sort of this global enforcement of KYC or something
like that.
Very, very difficult.
Do you think that if we were in the in the situation in which all these four layers of
protection, which are basically probably already sufficient in the first layer, which is which
is impossibility of coordination, but assuming coordination and assuming effective to some
level physical censorship of ushers and assuming that they can actually look into transaction
and whitelist everything, do you think that it would be conceivable to have some kind
of have you heard the developers seriously discussing of the possibility of having committed
reveals schemes forks in which basically the miners in order to get fees, they are forced
to to commit before the transact or can actually reveal the two extremes of the transaction
is such is the stuff like that even conceivable or discussed or is complete science fiction?
I think they did something like that with Namecoin, if I remember back in 2011.
So the way that system worked was you revealed only the hash of what you wanted to go the
domain that you wanted and then only after some number of blocks do you actually reveal
that so that you don't get into a race where soon as something shows up, everybody else
tries to get the same domain, something like that.
You don't want people front running it like everyone does in the Ethereum mempool.
Yeah, exactly.
So I mean, that's possible.
I think the solution is a lot more, I don't know, banal.
Like essentially, everyone is master of their own notes.
So if I want if I see governments doing nefarious things, I can just say I'm not going to accept
blocks from them.
And if enough people do that, then that's not Bitcoin anymore.
And if there's a nation state that's just mining empty block after empty block after
empty block, I can just soft work and say, I'm not going to accept those.
Now, the game theory ends up that the nation state, in order to get accepted, has to make
it indistinguishable from a normal block that would happen, in which case, they're just
a miner.
So they end up helping the network instead of trying to hurt the network.
So the fact that users have sovereignty is is a major, major production.
And I think it stops there way before it gets to sort of some protocol hash revealing, or
hash locked transactions or something like that.
Yeah, I mean, you definitely don't want to go down the route where you start subjectively
trying to decide what blocks you're accepting, based on some sort of identification.
Because really, as long as people are still adding blocks, they're adding proof of work,
they're securing the network, it I think becomes untenable.
If you get into a situation where there's enough adversarial hash power, that they are
actively reorging blocks with transactions to basically prevent confirmations from happening
on the network, that is what I would consider to get the network into a state where it's
not usable, if no one can get any transactions through.
And so then it turns into a more interesting game theory of, well, what happens if there
is that type of emergency on the network?
Because at that point, there would probably be enough people who would be incentivized
to fall back from our machine consensus and have discussions at the human consensus layer
of, you know, do we need to think about effectively quashing this adversarial hash rate?
And of course, the sort of ultimate nuclear option with that is to change the algorithm
to nullify all of the ASICs, which effectively restarts the whole ASIC machine race all over
again, you'll have to go through the GPU era and back through ASIC era again.
And it would be fairly unprecedented.
So you know, the incentives required for that would be enormous, and people would have to
agree that it was a major emergency.
Yeah, also, because it would be unprecedented, but it will create a very dangerous precedent,
because then nobody will put the same kind of investment into ASICs asification again
with a new algorithm, because there was this precedent that if it happened once, it can
happen again.
And also, it may be a centralization point, I mean, if the community managed to effectively
coordinate these, it may be a sign of centralization and can be misused for other kinds of nasty
things.
So yeah, I agree that it would be basically a serious nuclear option, really, like Cold
War style at this point, it's still a trap that we can use, like an ultimate trap.
It's because it turns into a waiting game, right?
If you're patient, if you don't have to transact, then you don't care if no transactions have
been confirmed for days or weeks.
So that's the sort of gray area.
It's hard to quantify how the ecosystem as a whole would react to a situation where no
transactions were being confirmed for a very lengthy period of time.
Well, basically, you are increasing the confirmation time, which is a cost.
You are increasing the on-chain transaction cost by increasing the number of confirmation,
which are reorg-free, reorg-safe, and by increasing the time you have to wait for the new actual
confirmation.
Yeah, I mean, I don't think it ever gets to that nuclear option.
Like, if you have 150 block reorg, that's going to set up serious alarm bells and say,
you know what, let's just not accept that other chain that just reorg 150 blocks, something
like that.
At which point, it's like, okay, well, they just wasted 150 blocks worth of hashing power
in order to do what?
They didn't do anything.
And yeah, I mean, that's a social consensus right there.
I don't see that, like, if people on the network see 150 block reorg, I think they would be
up in arms and saying, okay, let's just go and reject it.
And here's a patch to go make sure that we're still on the older one that had 149 blocks.
There we go.
And like, it might have been a nation state that did it, but really, they didn't get to
do anything.
All they did was they, you know, they got, you know, I guess the nodes that decided,
everyone's sovereign over their own node.
There's so many things that could happen, right?
Say a rogue state like North Korea did a reorg like that and used it to attack a US exchange
and steal billions of dollars.
I mean, there's some very interesting new nation state attack level scenarios you can
get into there.
But once again, it turns into geopolitics.
Well, I mean, but ultimately, it's the users in charge.
That's my point.
Even if you had North Korea stealing from the US, it's the users that decide is what
is this like reasonable?
Or is this just like a flat out attack?
And there's an immune system there of users that are going to say, well, you know, I don't
think this, we should accept this and, and go from there.
Because I remember, I remember when CZ asked the Twitter, if users want the reorg, I remember
the immune system is being pretty strong opinion about that.
So I think we really eviscerated all the scenarios.
And I think I will not panic to sell or not that they have any Bitcoin, of course.
But if I, if I did have any Bitcoin, I would probably not panic sell after this conversation.
And I will probably think about having a mining pool in the sense of a liquid, a liquid cooling
system heating up my spa pool with a hybrid system for for plausible deniability.
That would be interesting.