Hello, and welcome to Drivechain Twitter Spaces,
where we talk about Drivechain, which is BIP300 and BIP301.
Welcome to those who've joined.
Moon Settler, great to see you in the audience.
Feel free to raise your hand and come up to speak if you'd like.
Today should be a great conversation.
We might have Post Capone and Super Testnet and others join.
Some of those people I've chatted with about the possibility
of having a conversation about what do ordinals and inscriptions
mean for our understanding of Drivechain and its value proposition.
Welcome to the stage, Paul.
Thank you. I'm excited to learn more about Drivechain.
Oh, great. Great.
It's the talk of the town.
People are interested more than ever.
Definitely was in February.
I mean, hopefully we can continue that in March.
But it did seem like February
saw an enormous amount of increase in interest.
What do you think this ordinals and inscriptions episode
in Bitcoin history means?
In terms of our understanding of Drivechain.
Do you feel that it's...
Well, in general, what it means, a couple of different things.
One thing is there's a group that thinks that Bitcoin should not be used
and that it should only be store of value.
Historically, that group is contrasted with the people
who thought it should be a medium of exchange and it should be used.
And I, of course, I just think that both are very important.
It's like having heat in your house and air conditioning.
And I just think the better that each gets, the better.
But the issue is that there's a few dogmatic people
who just think that there's no reason to use Bitcoin at all
and that any use of the blockchain space
other than for transactions is illegitimate.
And those people have now been kind of shown up and frustrated
because the ordinals thing has taken off.
Because the ordinals thing has taken off
without them being able to do anything about it.
And then the other thing is that every cult is naturally an enemy of fun.
So a cult emphasizes conformity.
And of course, the current Bitcoin culture
has unfortunately some cult-like properties.
And so when they see anyone having fun,
then they know that we can't have that.
Everyone has to just be like a weird monk.
Those are some of the things.
Yeah. I think someone wrote on Twitter,
maybe it was you, that when there are two groups
and one of them is expending energy to enforce conformity of ideas,
then that group will underperform another
because of the energy that the one group
is expending towards just policing itself.
The group having fun will outperform.
That's certainly the case. I don't remember saying that,
but I would mostly agree with it.
The worst thing is that it's not even so much the energy used to police,
it is that innovation is the key to success
and conformity is a natural enemy of novelty,
whereas everything innovative must be novel.
So to me, the novelty is a bigger factor,
but that is also the case that you have to have all these people
spending their time on a self-inflicted handicap.
When you described Bitcoin being used versus just held,
it's an interesting distinction
because of course holding in economic theory is a form of using,
saving is an activity.
So the distinction some would draw is between Bitcoin as money
and Bitcoin as a platform
for a multitude of non-monetary applications.
And then the people would cite, perhaps simplistically,
the Oracle problem.
Well, indeed, but you see actually this is, I think, backwards
because money is a use, you're right, as savings is a use,
all that is correct.
But actually that just proves that the whole point
is that it has to be used for something.
So you make a list of uses and you can put at the top,
you could put savings or money,
but that just means that the reality of that list is undeniable.
And if the list contains money,
the reality of that list is undeniable.
And if the list contained nothing, then it would be useless.
So that's actually flipped around.
Bitcoin should be useful and the more useful, the better.
And if savings is one use,
but we should stack the list with everything we can find.
But when you say all that, are you focusing on the distinction
between economies that are on blockchain versus those that aren't?
Like, let's say all of Amazon were somehow more on the blockchain
versus just using Bitcoin as a payment system.
Isn't that the distinction?
There's another different thing, which is that Bitcoin,
there's 21 million coins,
but that has nothing to do with how much block space there is.
Block space is a completely different thing.
That is something that we also want the demand for to be high
because block space determines the security budget
and the 51% attack cost.
So they're related, but they're completely different goods.
They're just as different as oil and Netflix subscriptions.
Yes. Lately, I've been analogizing it to a store,
a store that just sells meat and milk and cereal
versus a store that sells those things plus candy and toys.
The one with candy and toys is being more permissive
in its value judgments about what to sell,
and it'll probably sell more products.
Right, and it can still sell meat, of course.
So you have someone who really wants candy,
and you're with a group, and you say,
well, if we go to the first store, then someone has to go away empty handed.
But if we all just go to the second store, then there's no issue.
So that would be, in terms of who can attract a group and network effects,
that would seem to be a pretty big advantage.
Yes. The discussion on block space is so important
and probably bodes well for drivechain understanding.
Since we've had a lot of discussion about the security budget recently
in the wake of this ordinals popularity and inscriptions popularity,
but many people were saying,
oh, well, the security budget will be solved
by those doing this ordinals and inscriptions.
But we haven't seen the fees rise very significantly.
We haven't seen fees over $2 per transaction.
We haven't seen total transaction fees per day over $1 million a day.
So we haven't seen a spike in fees rivaling prior spikes,
such as in 2017 or 2021.
Yeah, it's important to notice that Michael Saylor
seems to get this question wrong,
because he does not appreciate the difference
between the block space and the coins.
So because he'll make statements like,
well, it'll be with Bitcoin is $100 trillion asset,
then it'll be a 1% fee.
But that is not a legitimate argument,
because there's absolutely no relationship
between the two goods at all.
So there's no sense in which one would be a percentage of the other
any more than when you buy Netflix subscription,
it should be a certain percentage of the price of oil.
Yes. Everyone listening is welcome to raise their hand
and come on up to the stage if you have a question or a comment or a rant.
I think, yeah, this block space discussion is so important,
because it improves understanding and awareness of how Bitcoin works.
And the block space is a very interesting part
of the drivechain value proposition.
How surprised are you, Paul, at the evolution of ideas on drivechain
and how the security budget became a way of understanding
the value of drivechain,
since that wasn't part of the original inspiration?
Would you say that's a tremendously stunning development?
It seems that way to me personally. Do you feel that way?
I think it is interesting that drivechain already,
you know, when it designed it, it already solved
like three or four different huge problems.
So one of those being, of course, this idea,
this meta consensus idea that people disagree
over what the blockchain should contain.
So you can't have large blocks and small blocks.
You have to have one or the other, and whichever one you choose,
the other group will go away empty handed.
So that's one, but then just people wanting to be creative,
you know, the idea of Vitalik starting Ethereum,
but he can't really do it on Bitcoin,
because he wants a different design.
And make no comment on whether or not that's a good idea or not.
It's just the idea that people would want to try new things.
And that's the relentless march of progress.
So in addition to solving like block size, war, governance,
questions of like the blockchain going from being a one size
fits all to this full spectrum of things.
I think it just happens to be interesting
that it also solves the security budget problem
that people weren't even talking about back in 2015.
There was just a really vague question mark in people's heads
about bees replacing the block subsidy.
So I think that's a very interesting development.
I think unfortunately, the security budget question
is still completely misunderstood to the degree that makes it.
It just makes it more obvious to me that the community is in bad shape,
because you still have people who think that,
for example, a block size increase decreases fees
when there's really no evidence,
that all of the evidence points in the other direction.
And then you have people conflating fee rates and total fees,
which is the difference between saying $4 per gallon
is different than saying $4 per gallon
is comparable to like $20,000 or something.
One is per gallon and the other is just an amount of money.
So the fact that people would even compare those
shows how confused they are, unfortunately.
To be sure I heard you right, you said that a block size increase
would, in your view, increase total fees.
Is that correct?
Yeah, we really don't see.
There's a couple cases such as Bitcoin SV
where the fees are ultra low.
But in general, everyone is paying some tiny fee.
And so each transaction is giving a slightly right,
adding slightly to the end.
What we haven't seen is the fees stay up, right?
They sometimes go up when there's a big bubble or something temporarily.
But this is unreliable. They always fall back down.
And they just if you just look at the amounts
and just use a little common sense,
like they don't go up in the way that would be affected
by an increase in the block size.
So it's not as though they are $200 per transaction.
And then when we add and increase the blocks,
they just behave a very different way when the blocks are full
versus when they're even slightly empty.
When they're even slightly empty, then it's just this $30,
$0.30, $0.30 to $0.80 or less, $0.05 to $0.80 range,
which is just like the range that we get.
And then when they're full, they sometimes go up.
But when they go up, people react with their behavior.
They get annoyed and they transact less.
They batch the transactions.
They wait, they hold off on transacting
until the fees go back down, which they do.
Yes, I understand. That makes sense.
Some of the errors people make is they think
that if the fees aren't high enough for the miners,
then the block size could be reduced.
And that would increase total fees. And that's wrong.
And they also think that the miners could impose a minimum fee
that's higher than the current minimum fee.
And that's wrong.
It's wrong to think that either of those proposals
would increase the total fee rate paid to the miners,
which is what matters.
The minimum fee is like anyone can do that, right?
You could open a Burger King and you could say,
well, I'm not making enough money.
And you could say, well, instead of charging $5 per hamburger,
I'm going to charge $7,000 per hamburger.
I'm going to impose a minimum fee.
That's like stupid that idea is. But yeah, people say that.
Welcome to the stage, Super Testnet,
one of my favorite speakers on Twitter Spaces.
Thank you. Glad to be here.
I wonder if we'll get Post Capone today.
He was invited. Hopefully he'll join.
But maybe if either of you have to drop for any reason, no worries.
But I was wondering maybe we could have a combo between you two
about one of the things you've spoken about a lot,
Super Testnet, in your tweets,
which is the censoring of inscriptions or ordinals.
Is that something you'd be interested in talking with Paul about?
Yeah. Yeah, I'd love to.
Okay. That was sort of the intended topic for you and Post Capone
to perhaps debate.
And maybe Paul has a view that is somewhat aligned with Post's view.
He's anti-censoring. And then you have an interesting view
because you're someone who speaks very favorably about drivechain
and has written a great essay on it.
But I suspect Paul doesn't have the pro-censorship view that you have.
Maybe that's something you could summarize for him.
Sure. Sure. Well, it's gotten even more urgent recently.
Or if urgency is the right word for this.
There's even more people dumping ridiculous data on Bitcoin now
because of the new sovereign roll-ups protocol or idea from Bitkip.
Or not Bitkip.
I have to compliment you on that because it really has been disappointing.
As you may have saw me tweet recently,
like a lot of people, what a lot of people seem to think from the title
is that the roll-up thing is like some kind of Ethereum technology
that we Bitcoiners can now take advantage of or something.
When the reality is not.
But you're one of the few people who seem to have actually noticed that.
That is. But anyway, why don't you continue?
Yeah, well, I would say, actually, I would agree with those who say
it is an Ethereum technology that Bitcoiners can take advantage of.
But we've always been able to like run transactions through the EVM
and then and then dump the output to Bitcoin.
Like that's that's really what they're doing.
It's like it's not running the EVM on.
It's not running Bitcoin transactions through the EVM.
It's running these altcoin transactions through the EVM.
That's what I mean.
Dumping the output. Yeah.
Yeah. So because there's a lot more people dumping ridiculous data on Bitcoin,
using this taproot or not taproot, using this inscriptions, envelopes idea.
I think it's even more of a good idea to block them and just say,
yeah, sorry, you can't do that anymore.
Also, also the fact that Ethereum people are using it now
also makes that harder to do because they have a lot more tenacity
than I think the monkey JPEG people do.
And I think they would stick stick with it a lot longer
and try to overcome any attempts of ours to block it.
So that would be upsetting because I don't know who has more tenacity,
the Ethereum people or the Bitcoin people.
Yeah. So the basic idea is I don't think Bitcoin
is a good place to dump to use as a dumping ground.
I think there are much better places to do that.
And if someone starts using Bitcoin as a dumping ground,
like with inscriptions, I think it's a good idea
if it becomes popular to block it and restore Bitcoin to just being,
you know, just being money. That's that's my opinion.
Well, I kind of agree with some of that.
But really, I just think that the goal of the system
is for the miners to maximize their profits,
which includes all the cost minimization they do in the real world,
but revenue maximization with whatever they do
in the when they sell the block space.
So I see that whoever is going to pay the most for the block space
will be the owner.
And there's really nothing anyone's going to be able to do
to get around that reality.
I don't think that they will be able to because of the universality
of computation, anything that you try to,
if you try to block something that's formatted like that,
you can always you can always re-encode a JPEG as a PNG or something.
You know, it's just zeros and ones that are interpreted a certain way.
Isn't the solution that like you two might both agree with
to reduce the block size,
not for any reason related to security budget,
but exclusively for reasons related to making full nodes less expensive?
Well, the whole architecture needs to be fit for the task.
So if you have small layer one block size,
but you have layer two chains where people can do all this stuff
that they're interested in doing cheaply,
then they will do it over there.
They won't do it on layer one. Everyone's happy.
That's the basic situation.
Yeah, I think a much more feasible plan to stop inscriptions
would be to reduce the block size than to send soft fork them out.
Especially now that Ethereum people are doing it
and they wouldn't give up as easily as I think the NFP people would.
But yeah, I think campaigning for a block size reduction
is probably a better use of my time now
than campaigning for blocking inscriptions.
Yeah, I'm actually in favor of the block size reduction.
And it's interesting to figure out who is and is not,
because like for once, you know, I don't want to speak for Peter Todd.
Anyone can find him on Twitter and look it up.
But I remember and I talked to him,
I think I was either March 2017 or 2018.
Can't remember which one of those.
I was talking to him in person in Italy, I think it was 2017.
And…
I asked him, what did he think of the SegWit discount being a block size increase?
And he said, well, he said something like, and this is I'm just paraphrasing it from memory here,
and anyone can talk to him in person and figure out what his views are.
But he said something along the lines of, well, I think we really might need the space.
So he was worried about not having enough block space, which how about that?
So you never know who's going to endorse what position in this industry, you know,
because everyone is so quirky and independent.
So it'd be interesting to see who is still like, who's into what.
And then, of course, we have, I've told this story a bunch of times, but it's a good one,
because John Carvalho is saying he was all, he was on the warpath for SegWit.
And then it wasn't until, I think, 2019 or 2020 that he even admitted that he didn't,
he just learned that day, years later, that it was a block size increase.
So he didn't even know that it was.
Are you indifferent as to whether the block size is reduced before or after Drivechain is adopted?
Well, I think it's just a more, the whole, the system overall works better
if they, does happen around the same time.
But I think that the reasons for lowering the block size are mostly independent.
I mean, one of the reasons why I'm in favor of lowering the block size is because
of my confidence in the view that people can take advantage of sidechains.
So if that view were totally mistaken, then that would be,
I might need to reconsider all of my other views in that light.
Welcome to the stage, Alex and Post Capone.
Alex and Post Capone.
What are your thoughts on lowering the block size and adopting Drivechain and other matters?
Drivechain is a good idea. Lowering the block size is not a good idea.
Why would you say that?
Well, because lowering the block size essentially disables the Lightning Network
and the nodes are cheap enough at this point.
Well, that is another point that, unfortunately, a lot of people do not notice.
I tweeted about this recently, which is that I have a lot of familiarity with the Lightning Network
and I designed a couple of things, 2015, 2016, some clever Lightning-type things.
They require, it's a long story, but I won't get into it.
But the Lightning Network, in order to do lots of really cool things,
it must be possible to write the state, the intermediate state, to the Layer 1 blockchain
or to whatever the host blockchain is.
And so Lightning needs on-chain fees to be low.
And they did the math in this post, I did Lightning limitations.
The amount of marginal bytes, I mean, I should look it up, but it's quite a few.
To add in, you have to write the HTLC and then you have to spend it later.
So the total amount of bytes needed to write this output, if you get interrupted,
if you get interrupted in your Lightning payment and you say,
okay, the person went AWOL, I have to write it to the state,
I have to write this whole if branch, I have to write the HTLC,
I have to write what the hash should be, I have to write the preimage,
you have to select that as an input and then spend it to wherever,
whatever gets you your hamburger or whatever you were going to spend your money on.
That total thing is like, I don't remember exactly what it was,
but it was like 90 bytes or something, which is a lot
because that is basically as much as a normal to any other on-chain output.
So yes, Super Testnet.
So what I'm saying is if that costs a lot, if that costs like $8 per transaction,
you're really kind of up against the wall with respect to any Lightning payment
that's less than $8 because you'd rather not write it to the blockchain.
So you could say maybe we'll do trust or maybe we'll have tiny little,
this is the interledger idea, we'll have it stream pennies at a time
and we'll just hope that if it breaks, I'll find a different route.
But anyway, you have your hand up.
Yeah, I'd like to elaborate on that point.
I think having a small blockchain or a small block size,
I don't think it breaks the Lightning network.
I think it increases the minimum size of an HTLC.
You could have $100,000 fees on the blockchain
and still the Lightning network would work just for payments greater than $100,000.
And for other payments, you could put them on a drivechain.
You could say, all right, we're going to do micropayments on a drivechain
or you can put the Lightning network on a drivechain.
So that's a good way to think about it.
That is a little like payments less than $100,000,
that's kind of like the exception that proves the rule or whatever.
But yeah, I agree that I think the drivechain,
obviously, I think drivechain scaling is underappreciated.
I think the idea of Lightning on drivechain is underappreciated.
The big disadvantage with drivechain or anything merge-mined
is that it still relies on confirmations
and you get at most one confirmation every mainchain block.
So you need even more sidechain confirmations.
There's no block subsidy.
So basically, the confirmations are slower and worth less on the sidechain.
That's the big disadvantage.
But that's also why it would work just fine for like all these casual things,
coffee, whatever, these things where everyone's using Venmo
or using the fiat world or whatever.
So then when you go to a situation where those are slower
and the confirmations are more onerous to obtain,
well, that's also where you get all the benefits of Lightning
because you open the channel before you go to sleep
and then when you wake up, it has however many 50, 60, 70 confirmations,
you're ready to go for the next week or whatever.
And now in Lightning, the stuff is basically instant.
I don't know if we should turn this into the Lightning complain fest,
but I just saw Warren Togami I have a lot of mixed feelings about,
but I think he knows more about Tor than probably anyone I know.
And he had an interesting rant recently about Tor,
Lightning will not work over Tor.
And I don't know if we should try to unpack that,
but I thought it was very interesting
because he's squarely in the precious LN whatever camp,
but he's very experienced with Tor.
And without Tor, Lightning requires you to disclose your IP address
in order that people can find you and pay you,
which I would think is just a knife to the heart for privacy.
And it would tip the scales back into Lightning being less private than on-chain.
Could you post a link to that thread in which he talks about that?
Let's find it. I'm going to try to find it now.
Because I would say that although the Lightning protocol is currently designed
only supports Tor and Onion,
there's nothing stopping us from adding other data transport protocols to...
I think ITP is better with better performance than Tor.
So I don't know if that will ever happen, but...
I don't think it means if you lose Tor,
it doesn't mean you have to dox your IP address,
or at least maybe right now it does, but we could change that.
Now you have to only dox your I2P address or some other networking layer.
Well, I mean, that sounds like a lot of work, doesn't it?
Not only that, Paul, but like you pointed out,
Tor, I mean, the cost of running a node must also include
not just the physical cost of running a node,
but the permission to be online or to stay alive.
So if Tor doesn't work,
then basically the cost of running a node is essentially infinite.
I found this thing and I'm going to reply.
I don't know how to do spaces and stuff and make it show up in the thing,
but I will reply to the thing and maybe it will show up.
I don't know how it works.
Maybe Henry can do it.
It didn't show up in the nest, but I see it here, so that's useful.
In the comment box.
There it is.
I wish there were more people in Bitcoin who knew
as much about Tor as Porn Togami does.
I put it in the nest.
Okay, cool.
Maybe I'll delete it from the comments then.
Because maybe he'll deny it.
It's all right.
They will come in.
Anyway, Tor is one of those things that often I think people use as an excuse.
They don't do their own thinking.
They just think, oh, is it private?
And then they just think Tor.
They hold up a sign and they just like wave a magic wand.
I once spoke to Warren about half a year ago
in our Twitter spaces about Drivechain.
And I recall him saying that
something to the effect of
he wasn't convinced that
the Drivechain security budget marketing
is entirely convincing.
So he's not a
security budget alarmist.
Not security budget aligned.
Yeah, maybe he's optimistic about the security budget
more than most DriveChainers.
See, this is the other thing.
The security budget
argument
leads to a lack of focus.
People
kind of show up for different reasons
into the Drivechain group and then they leave
for different reasons.
Like if you're there because you want
a cool feature
then
you'll eventually just join the altcoin
that has that feature.
After you join Bitcoin, you wait around.
You're waiting for the feature.
Waiting for Godot.
It doesn't show up and then you just say,
join the Namecoin project.
Are you saying that people could show up to
Drivechain
because of a concern about the security budget
and then
join some other coin
because it doesn't seem to have
the same kind of problem?
In that example, they would join Ethereum or something.
Because it's doing inflation
and doesn't seem to have a problem with that?
Ethereum is not necessarily
doing inflation at all.
It just has more higher fees
and in fact
their ETH1000
is about
if they have enough fees
they will destroy coins.
But if they don't have enough fees then they inflate?
Then they will inflate.
Oh, the 1559.
So inflation is part of their
protocol just as a backup plan?
Yes.
The funny thing is
if you go the Peter Todd route
informally it is also part of
the Bitcoin backup plan
just denied.
I don't
agree that I think that
Peter Todd would find it impossible
to actually succeed
in
adding inflation to Bitcoin
and he may be engagement trolling
but
still it's a thing
that people say in Bitcoin.
Well a big question is
when will these concerns materialize?
Right?
We had concerns about selfish mining
that never became a big topic
in reality it was just a theoretical
concern from 10 years ago.
When will these security budget concerns
become a reality?
When will they manifest as a risk?
We haven't yet seen
a hostile or
greedy miner of any
substantial nature.
Correct?
Well
my own thoughts are that
first of all I never thought selfish mining
would be an issue.
I had like a little refutation of it
at the time but
with the
I think the adversaries of these coins
would have to get good at attacking
decentralized coins first.
They're going to warm up on that.
They're going to get really good
at whatever they have to do
to take down the centralized coin.
You know that only has a few people
running it.
What does that mean?
Well out of all the
coins that exist
there's an existing spectrum.
The Bitcoin SV
would be more centralized.
Ethereum would be more centralized
than PTC.
And the adversaries would warm up
on that.
And then once they warm up, once they have
that cold, then they know all we have to do
is attack the security budget and it will go
from being decentralized to centralized.
And then we can do it.
So it would be like a chess game
where they have to move on
a certain territory first.
Super, did you understand
that whole point he just made?
I didn't.
I was replying to someone
who was wrong on the internet so I didn't
actually listen. What was the point?
Paul, were you saying that
when the security budget problem materializes
if it does, it will first
materialize on BSV and BCH?
No.
The security budget
it only works
the only thing it accomplishes
strategically is it converts
a decentralized cryptocurrency
into a centralized one. Because people
immediately, once the attack starts
they'll look for some kind of
person to say what is the state
of the blockchain, like Coinbase or whatever.
Does that make sense?
Yes.
Yes, this is what...
This is the only thing it does.
It doesn't instantly kill the coin.
It will convert it
from decentralized to centralized.
Understood.
Sounds like an easy way to kill Bitcoin then.
Let's do it.
Exactly. But my point is that the adversaries
understand this industry so poorly
that they will, before they
realize what they can do
they will have to
learn about killing the centralized coins.
And there'll have to be a slang term for this
and there'll have to be examples of this.
This will have to enter the Overton
window, you know. This will be like a
mainstream thing first.
That's because you ask in practice. There's a difference
between practice and theory, of course.
That old joke that
in theory there's no difference between theory and practice
but in practice there is.
If I understood correctly, are you saying that
before Bitcoin is destroyed
by the security budget, if that were to happen
it would first happen to
BCH and BSV?
Oh, no, no.
I think you're saying
51% of the time.
Yeah, right.
Have you ever played a video game or something
where you have to
stun the boss first before you move in
with the Master Sword
and they don't glow red?
They don't glow red unless
you hit them at the right time.
I think another way to put it was
before they try it on Bitcoin
they'll practice it on BCH and BSV
and if it works there, then they'll
be like, hey, maybe we could do this to Bitcoin.
Well, right.
There's a whole spectrum.
So some are completely centralized
already.
So that would be like
something where
like,
I don't know, the best example would be
Solana, maybe, something like that.
Where they would just be like
compliant Solana
or whatever the attack is.
Or, you know,
people from the Mafia
find the Solana validators
or they find whoever it is,
a certain key person,
Vitalik or whatever, and they
broadcast messages.
I think I
interpret the
hypothetical optimal strategy
for a
large-scale attack very differently
than you. It's interesting. I'm hearing
what you're saying. Also, by the way, hey, Paul,
I have very much
enjoyed listening to you
rant,
I'll say, over the years.
It's been quite enjoyable
and nice work. I like your work.
I dig it.
So
I don't
think I see it
in actually the opposite
window.
If I were going
to hypothetically
detail
an ideal attack scenario,
I would say
that
attacking in any sort of
material or demonstrable way
a Solana or even an Ethereum
is like
this half step
that only hardens an industry that you'd
rather see go away.
As long as the
largest chain by market cap
exists,
then you don't really achieve
anything meaningful.
While you
may learn something
in practice, you may
learn some technical skill,
you don't achieve
any ends and you ultimately harden
the system.
I would think that
the optimal avenue
for attack would be to get
a chain that is easier
to attack to be the highest
chain by market cap and then proceed.
Yes.
I agree with you.
They're really two sides of the same thing.
You can imagine the decentralization
as being some kind of antibiotic resistance
or something.
The question is, if you poured
antibiotics into
a lake and you're just pouring
antibiotics into the lake,
then eventually the only bacteria that
are still in the lake will be
antibiotic resistant
bacteria.
Then maybe all of the
antibiotic
bacteria, maybe all of them
can be like, I don't know, I'm just making this up,
but they can be eaten by a certain kind of
algae or something.
You put the algae in.
You do all one and then you do all the other.
I totally agree with that.
I was just kind of saying
you have it the other way around
where it's kind of like
what they'll
probably do is they'll probably say
we have this algae that most of the bacteria
in this lake that we don't like.
It's kind of funny because
from the perspective
of certain people,
Bitcoin will be like the small fry
because there'll be so much more fraud
like FTX
and maybe Solana or whatever
and it'll be easier.
Maybe it'll be easier for them to understand
how...
You can do it either way. Either you can say
listen, we're going to make it
so that the bacteria
that dominate this lake
are all vulnerable
to antibiotics and then we'll just come in
and then
we'll shock it with
huge antibiotics and everything will die.
Or you do it the other way around
where you say...
We will make it so that the only bacteria in this lake are the antibiotic resistant ones
and then we'll do the opposite strategy. So the 51% attack the security budget is what
penalizes the antibiotic resistant bacteria. If this is making any sense at all. I mean,
analogies are not very persuasive, I think. Yeah, no, you are making sense. I think maybe
some of the endpoints where I see it becoming dramatically different aren't... Maybe they're
not packaged in the conversation well. So, okay, let's do this. Let's go back to what we were
talking about earlier where we said like, okay, an attack on the chain would result in a
centralization of the coin due to this perception sort of phenomena where people would rely on
a small subset of actors to inform them of the state of the chain, right? We'd all be like,
oh my God, all the nodes are saying different shit. What does Coinbase say the state of the
chain is, right? Is that a decent consumption of your point? Yes. All right. So I don't know
that this actually is the way that it plays out, but let's say that it is. If an attack
takes place, right? Let's go with the scenario of they make the chain larger, they get Ethereum
to flip Bitcoin, and then the treasury says validators can't attest to blocks containing
transactions, and they force Ethereum into a user-activated soft fork. Well, the system as
an aggregate, I mean, a lot of people, the technicals here will escape them as the industry
is made of goldfish. But for a lot of the system, it will become very apparent that this whole
like distributed infrastructure, node operators, network participants over market participants
kind of conversation space did have real meaning. And you would see an increase, I think you would
see a pretty dramatic increase in network participation as a percentage of market
participation. And especially considering the fact that a lot of the market participants
would probably flee, then that metric would probably skyrocket. And if that metric skyrockets,
then there becomes substantially less of a need for somebody to rely on a centralized perspective
because the percentage of network participants has skyrocketed, right? You have more perspectives
to share, right? Does that make sense? No, I'm a little, I guess. Okay. Yeah, I just see,
I see the evolution of it. And it's all, it's dealing in hypotheticals for sure. Like this is
all like game theory of like war room type stuff. So it plays out in a couple of different ways.
But I mean, you don't see, the point is consensus is very difficult. So people will want to know
what are other people going to use for consensus? And so it would naturally be like
Coinbase or like one person. I think the odds of it being a reliance on Coinbase would be
dramatically lower if an attack were to manifest somewhere else first. There's a great tweet from
Nick Szabo from a few years ago, where he wrote something to the effect of what Bitcoin does is
try to minimize the social coordination. And if there is a insufficiency, he doesn't say
this much elaboration, but he says basically that it wouldn't be good to have the plan B.
Well, if there is an attack because of low fees, insufficient economic security, then
we'll just coordinate to solve it. That wouldn't be good because that's not what Bitcoin does well.
That's what traditional finance does is people get together in a boardroom and make a decision.
Bitcoin tries to do the opposite.
Would it be fair to say that by increasing, if you see a dramatic increase in the number of
network participants as a percentage of market participants,
that you would avoid more of the coordination scenario?
Are you talking about network participants defined as full nodes?
Sure. Yep.
So if there's tons of full nodes out there, if everyone in the world who uses Bitcoin,
all 200 million Bitcoiners each has a full node, then if there is a low total fee rate
and low Coinbase, then a 51% hostile miner can still destroy the network by making it impossible
for anyone to send a transaction because they're reorging at every opportunity.
I'm not saying they can't. I'm specifically addressing this notion of becoming reliant
on a single perspective. I'm not trying to augment the nature of a 51% attack.
I'm just discussing the idea that we would become reliant on a central perspective.
Paul, what do you think of that?
I have a lot of real work to do, so I have to leave soon.
OK.
The idea is that the 51% attack makes it so that the nodes do not tell people what the
unique blockchain history is. So they will need to get that information from somewhere else.
I was being charitable. I was saying instead of just the network dying immediately to the 51%
attack, people will try something. They will tweet at Brian Armstrong
and say, my money is at Coinbase. I'm on liquidity or whatever. Whoever has the most liquidity
will just determine what it is until, you know, whatever. Brian Armstrong is,
heaven forbid, assassinated or whatever the heck happens in this weird scenario,
arrested or whatever. Anyway, I have a lot to do, unfortunately.
Thank you for your time here.
Yeah, I appreciate the time, Paul. It was a pleasure talking to you.
Yeah, likewise.
So, Post, what was your thought in response to that and kind of what I was saying, which is,
I think, the same thing, which is that if you have the hostile attacker reorging at every opportunity,
then the heaviest chain, the chain with most proof of work, is a chain with no transactions.
And if they're doing this forever, then things aren't working. Bitcoin would devalue.
Yeah. So, again, I don't want to augment any of the perceptions of a 51% attack. I guess I would
butt heads a little bit with the idea that it's sort of an insta-death thing. If you have a
substantial number of node operators, if there's some reasonable number of node operators, then
the node operators have to make some sort of decision, right? And this may result in some
social coordination that is less than desirable, but the node operators come to some decision where
they say, okay, we have to change the hashing algorithm or something, right? And just burn
all the ASICs to the ground or something to that effect, right? The node operators would have to
make some decision. And without a doubt, this would be an unbelievably traumatic scenario. I
mean, in the conversation about changing the hash rate, you're destroying good miners' chips.
But I feel like if you reread Paul's 2021 blog post on the security budget, the second of his two
lengthy essays on the security budget at truthcoin.info, what you'll see there
is the argument that I agree with, which is that if Bitcoin deviates from proof of work,
as you said, for example, viewing all of the ASICs as a sunk cost and pivoting to a new
hashing algorithm with a new kind of machine, then the problem there is that the security
budget problem once again exists. I don't disagree with that take.
Right. Okay. So we're on the same page fully then. So it's not good if the plan is,
well, if there aren't enough fees and the government can afford to just press this
button that turns off Bitcoin by doing some mining, then if Jason Lowery builds his hashing
force and then Janet Yellen orders him to make it mine empty blocks and reorg forever, then
we have to, the problem is the cost of trying to overcome that is still there. Even if we change
the algorithm, then we have to change again and again and again. And that's what Bitcoin becomes
is a constant cat and mouse game. Yes. I think what honest network participants
ultimately would sort of like be trying to fight for is the preservation of the state
at least up to the point in the attack or to some point at or beyond the start of the attack.
Right. Like that's what they would want to be preserving. And so there would be an enormous
amount of conversation space that would have to exist around how to preserve that state,
like how to move forward. But the more network participants there are, then the less I think
it becomes dependent on a central observer. So again, like I'm not, I'm not trying to say like,
oh, we can get through a 51% attack and it'll all be hunky dory. That's, that's not the point
at all. It is just very, very simply that I think that in any course of events, like any way in
which the narrative or scenario matures, that results in a higher percentage of network participants
as, you know, as compared to market participants, the less reliant you become
on the centralized perspective. That's, that's really the extent of the point.
Super. Please jump in with your thought and anyone else is also welcome to join the stage.
Yeah. One of the, one of the things that occurred to me during that
back and forth was that right now there's about a two year, or it seems to me that there's about,
we go about two years before we come out with new ASICs because it takes a certain amount of time
to design new chips and to get them to even smaller nanometer levels. And then it takes a
certain amount of time to get the contracts to get them built. But yeah, it seems like we have
about a two year turnaround. And so if you, if you disable the current ASICs, then it'll be about,
it'll buy you about two years of time at current levels before they come out with new ones that,
that hash, whatever your new algorithm is. But it's possible that in the future,
this, this timeframe, this time, this window of time will be greatly reduced.
Because one of the reasons it takes a lot of time is because there's not that many designers.
And another reason why it takes a lot of time is that there's, there's a lot of the companies
that manufacture chips for us have contracts with other companies that they prioritize first.
But should Bitcoin continue to grow as an industry, both of those things could easily change
and you could easily pump out, there could be, there could be, there's a possible future where
new ASICs come out every six months instead of every two years. And in that future, you don't,
changing the hash rate doesn't buy you a lot of time before you have the same problem again,
where like one big 51% attacker has all the ASICs again, because he can afford to buy them,
to buy it, to buy up all the contracts. And then it would come down to how either,
so it would come down to how aggressive the competition in the, the fabrication,
like the end point of the fabrication industry is, or it could potentially include other variables,
such as how controlled by the state, the fabrication process is right. So if, if
competition, if there's a low level of control over the fabrication process and the competition
for the end point of the fabrication supply line is high, then it's not a scenario where the attacker
gets all the chips. It's a scenario where a bunch of new miners get the chips, which is like more
along the lines of how we've seen the ASIC industry evolve thus far. Right. Or it's closer to that end
of the spectrum. So it's, it's not so much like, oh, well they'll just make the chips and then
attack you again. It's given these other set of variables, right? So what, you know, and those other
set of variables might be important to understand as you, as you navigate how to preserve the state,
like the conversation about how do we preserve the state may want to take into consideration, like,
you know, the chip fab industry is like, you know, basically strangleholded by the state.
And there's no chance of us changing the hashing algorithm and coming out of this with like
competitive bids on ASICs. So maybe that doesn't work. Right. But it's not, it's not a static
thing. Like there, there are variables at play. Right. Right. Yeah. There are more variables than
just time it takes to manufacture and time it takes to design. Yeah, that's a, that's a good
point. Thank you for bringing that up. So you want to argue again? That was, that was fun when
we did that last week. Yeah. I mean, so I was telling Henry that I think for like you and I
as individuals, like we've, we've spoken pretty extensively. I don't, I don't know how much more
we'd cover, but for the sake of it being a public conversation, I'm certainly down. And if we do
wind up covering new stuff, then great. Maybe one way to clarify the conversation in a slightly new
light would be to talk about, given your agreement, if I understand correctly, you both agree that
drivechain is a great idea. So given that, why does, why, why does it seem to you super to be
good to think about like censorship when the long run goal would be just smaller block size and,
you know, having things naturally move to a sidechain rather than try to force them out of layer
one through censorship? Well, some forks, some soft forks are easier than others. And I don't,
I don't want to remove one from the table just because there's another one that's out there as
well. However, I, since then a couple of things have changed and I'd like to let Post Malone know
that I think he's, he's basically, or circumstances have won me over to his side since our conversation
happened. In the new circumstances, these roll up things came out, the roll kit protocol came out,
which exacerbates the problem of using Bitcoin as a dumping ground even, even further, but introduces
a new set of players. Now it's not just monkey JPEG people who are dumping data on Bitcoin.
It's like these, these blockchain designers who have a vested financial interest in getting someone
else to host their data for them. And I think that while the NFT people didn't have a lot of,
or don't have a lot of, what's the word? I don't think they, they have stick-to-itiveness,
where if we started blocking them and censoring their transactions, I think they would give up
and move elsewhere pretty quickly. I don't think these Ethereum roll-up devs would. I think they
would fight more our attempts to block them and continue dumping data on Bitcoin through
means other than inscriptions. So I think that's a lost battle now. And I'm just going to advocate
for a soft fork to reduce the block size rather than a soft fork to specifically kick out
inscriptions, which is what you wanted, but you didn't want me to advocate for censoring
inscriptions all along. So congratulations. Now I'm more aligned with you. What do you think?
Well, you know, perhaps the answer is great, but I, I, my answer would really be that I,
like, I find it unfortunate that the alignment comes in the form of like erosion like this.
But the, the idea of roll-ups being something that existed within the use of this,
this right story, this right space, right? Like this. Okay. So the way, the way that Ethereum,
like it trajects its future for roll-ups is by leveraging blob space, right? Like they want,
they want to do this blob space thing and they want to write the roll-up state in the blob space.
And then they want to use smart contracts to settle to Ethereum, right? So what the envelope,
the inscription envelope is, is effectively a cheaper version of the blob space that Ethereum
already wants. And the people paying it, if you're, if you're paying really close attention
to the evolution of several different narratives in Bitcoin over the course of the past year and a
half, then the idea of roll-ups on Bitcoin was fairly apparent as far back as a year to a year
and a half ago. I like, I personally have been sort of like chomping at the bit, sweating in
anticipation of these roll-ups on Bitcoin. Now, in the case of this specific demonstration with
Rollkit, it is, it is a substantially less than desirable, like prototypical demonstration of
how you might start to go about the process. I personally, and this actually counts as a
little bit of a criticism for 300301 as well, I think that mechanism design that does not allow
for trustless deposit and withdrawal that is at the sort of like, let's say, let's call it
native speeds of the L1, right? Like it's slower than the L1 is expected to be. Ultimately results
in users being forced into LPs that facilitate these deposits and withdrawals. And those
processes become sort of like on-chain KYC AML injection services. Not to mention that the, the,
the absence of a trustless nature of deposit or withdrawals completely
weakens a lot of the arguments made by the base layer of like the way things ought to be.
So I'm not, I'm not super, I'm very excited about the conversation that Rollkit
creates, like the conversation space around that, I think is awesome. I think that the way that
that evolves, hopefully is into, you know, better executions than Rollkit is, right?
But as a broad generalization, like we all knew, we all knew the roll-ups were coming and we were
all pretty excited about it. And as I did with inscriptions, I would encourage you to try and
see that there, there, there is a lot to be done with it without compromising Bitcoin, like with,
with it without compromising Bitcoin, without dismissing what we intend Bitcoin to be, without
increasing block space, without
augmenting. all that data into witness data.
Well, yeah, using it as a data availability surface is the point though, right? And there,
there's a million amazing ways to use that. And it's not increasing block space. It's usual,
it's using the block space that was already there. And I, I think where you and I just like,
cannot come to merger is that you think that the block space should only be usable by some very
narrow terms. And I, I would say, I don't think it's really possible, like the, the timeframe
that we would have to wait for that block space to be used in like these very, like sort of
turbo primitive mechanisms that you see it as ought to be used for would be very long.
And by the time we get there, right? Like by the time we get to the point where the
block space is being used by that, then all the things that you don't like about this stuff are
going to be priced out anyway. No, like nobody's going to be printing fucking JPEGs for a half a
million dollars. Like it's not going to happen. I don't want to say it's not going to happen.
But I find it less than likely. Well, it could be less prevalent,
a smaller percentage, a smaller issue if it's price. Oh, right. Yeah. Using blob space to
facilitate other like, okay. So I don't know how familiar you are with like John light and
Dugan's research into validity roll-ups, but like that whole thing goes around this sort of like,
you know, you have, you have to write a proof, right. And like, according to Dugan,
if we take Dugan at his word, then with no like material changes to the Bitcoin protocol,
you could roll up a hundred thousand P2PKH transactions into a single block like today.
Right. And those wouldn't require, those wouldn't require like novel liquidity channels,
like a lightning that wouldn't, you know, there'd be a lot of improvements to the scalability of
Bitcoin and people's ability to use Bitcoin affordably for financial transactions. And
the other really great thing in my, in my mind, the thing that I really like about John's research
thus far is that he has been, you know, very forward about advancing the ways in which Bitcoin
would assert rules on the L2, like on the rollup. We're like, okay, you can only do this type of
shit. Right. And I think that that is a very good angle to approach this from, but that requires us
to write to this, like, we have to be able to use this blob space on Bitcoin to do that. And
interestingly, as it turns out, it's actually about one sixth the cost to write to Bitcoin
than it is, than it is to Ethereum. So the people that are developing these tools in other networks
are actually like, you know, deeply attracted to it now. Yeah. I mean, that's only until we have
a whole other conversation. ETH will probably get cheaper and better.
Say it again. Sorry. I say ETH will likely improve and cheapen.
Oh yeah, yeah, yeah. So like once, once ETH has matured, some other development initiatives,
like they're like, they're, they actually are sort of in this interesting dilemma where
like the current regime of ZK rollups may not actually even demonstrate any material
bandwidth improvement or throughput improvement, which is like kind of, it's like kind of an
interesting dilemma to be in. But for the time being, like there is a lot of interest around
the space. And I think that you don't like what I, I personally have no interest in seeing the
tokenize everything agenda migrate its way to Bitcoin. Like as far as I'm, as far as I'm
concerned, that, that mindset, that, that approach to the development process or the financing
process can sort of fuck off. Like it's all just, it's such a, it's such a horrendously stupid way
to do it for like a vast plurality of reasons. But the, there is a lot of engineering skill there.
And if that engineering skill can be motivated to, I'll use my own Twitter bio as a reference here.
If it can be motivated to plug into the second control report, then like, I think there's a lot
of good to be had in that. And I think super is sort of stuck in this world where, you know,
he, he wants, he wants it to just be these, these sort of like primitive expected Bitcoin transactions
that occur. And I think that there is to, to some degree, a failure to recognize that there are
very, very useful end points for this. And we can't get those end points without also having
the permission to write JPEGs, right? Like I, we, like you, Bitcoin is permissionless. If you
can write data, then you can write data. Right? So right now people are writing JPEGs.
Well, if the mentality was to block it, like the overwhelming BTC-ers,
then it, then it wouldn't be blocked. Is that accurate?
Well, that's definitely not, it's not an overwhelming majority of BTC-ers.
I didn't say it was, I said it was.
Yeah, it's just me.
I'm asking you if it's worth it.
So if the, if the social consensus, if the social consensus were that we should block it,
then I like, you know, I, I would probably still try and rationalize with them.
I'm just saying, what I'm saying, let's have, would it get, would it get blocked?
Let's have super, sorry to interject, but just quickly.
It would require a dramatic fork that changes a number of features over an eight year time period,
well, a seven year time period in order to fork it out.
It would be extraordinarily difficult to remove this.
Or, or if it wasn't these tenacious Ethereum people trying to do roll-ups on it,
then, and it was just the monkey JPEG people, I think they would quickly give up.
If without, it wouldn't take eight years, I think it would take like maybe one year
of soft forking out every attempt to take it in descriptions.
And you'd have to unwind, you'd have to unwind seven years of feature ads.
That's what I'm saying.
You, you would, you would have to go back to 2016 to get rid of this, man.
Like it, it would be rough.
I don't think you would though.
I think if you just, if you just banned, you know, the, the inscription envelope,
the op-if op or op false op-if construction.
No, because then I would just do op true op dump off false op-if.
Like I would just like, there's a million different ways you could write this.
Right.
I know, but I don't think they have the patience to keep on rewriting their software every,
every, you know, three months.
We already have 18 different ways we could do it.
You, you would have to undergo a new soft fork every single time to mitigate something
that takes 45 minutes to slap together.
It just won't work, dude.
I'm sorry.
I don't, I don't want to frame it in that way.
Cause it sounds like, it sounds like it's this like malicious attack when you do that.
Like, okay, remediate.
And I'll just tap it again, this other angle.
Like, I don't, I hate framing it that way, but we have to be pragmatic about it.
It would take the envelope authors 45 minutes to work around a soft fork that you organized.
And now, now you have to organize another soft fork and then they do 45 minutes for
the tinkering.
And then you have to organize another soft fork.
And then they do 45 minutes for the tinkering.
The network winds up like the end of this process is like, that's the step that kills
Bitcoin.
So I don't think it would take them just 45 minutes to, to do all this because here's
one thing that would happen that I think would happen is you've already got like 15
different projects on ordinals and each of them has to do this, do this process of changing
the inscriptions and any new developer who wants to make a new project has to support
all of them.
So they literally merge and update to one GitHub repo and then do a notification in
the discord server that's like, Hey, version 0.5.7 is released.
Download it.
Okay.
I'll put it this way.
It would take less effort than the soft fork.
And that's the problem, right?
That's the reason this doesn't work.
I don't, I don't think that's true though, because even although it takes less effort
in some sense, I think the NFT people aren't willing to put in that, that effort.
Like just, I'm imagining if I'm coming on and I want to make the next ord swap and I
look at the ord documentation and it's like, okay, first you have to support up if, then
you have to support up if, up not up if, then you have to put up true, up not up if,
then you have to do up true to up not if, and you have to support all these different
formats because we're in a fight with Bitcoiners.
I think that you guys would be like, how about we just not.
I think you're underestimating how enthusiastic people are about this.
I personally taught literal hundreds of people how to use CLI so that they could become node
operators.
People are ravenously, ravenously excited about this.
I like, you know, I, the only way, the only way for us to really know is to play the cat
and mouse game.
So I guess go and play it.
But I will tell you that I'm extraordinarily confident in the sort of like analysis that
the cat and mouse game would not work.
And that, that even if you were successful in organizing a dozen back-to-back soft forks,
you still would wind up with people writing envelopes.
Yeah, so I, a reminder, I do agree with you now, because I think that the roll kit thing
changes things and that I think those, those people who want to do roll ups have a lot
more stick-to-itiveness than the NFP people do.
And so they would, they would not give up as easily.
They would, they would fight us back-to-back and they wouldn't care about updating their
software frequently because they already do that.
So were you, were you present or tuned into any of the conversations about roll ups on
Bitcoin prior to this roll kit thing?
Because that conversation space has existed for a year and a half now.
Yeah, I mean, the first deep dive I made into roll ups was when John Light published his
paper on them.
So I guess that wasn't a year and a half ago, but it was like six months ago, I think.
So that was when I first started becoming interested in them.
Although, but even before then, when roll ups first came out on Ethereum, I thought
they were interesting and I heard all good things about them, but I didn't read about
them until, until John Light made his paper.
Yeah, the, the, the introduction of ZK roll ups is really what sort of like gravitated
me back to like doing a lot of research in that ecosystem, because I do, I do think that
there's a lot of extraordinarily interesting use for, you know, use cases for cryptography
are interesting in general, but that one in particular offers a lot for like this industry.
So I guess I would be curious, sans roll kit.
All right, like, let's, let's leave, I think roll kit is like this prototypical conversation
starter.
It's not what we should, like, it's, it's not where we should want to be.
But in the larger conversation, that is the idea of roll ups on Bitcoin, where do, where
would you say you stand?
Yeah, I'm in favor.
I look at them as a probably slightly better version of drivechain in that if you, when
I compare them analytically, I think both of them are trying to take a lot of, a bunch
of state on a separate blockchain, compress it into something small with, with what's
with the drivechain, they compress it into a hash with ZK proofs or with roll ups, they
compress it into a proof, a proof of what happened on the blockchain.
They put that proof or that hash on Bitcoin.
And that's all that nodes have to validate is that one small thing.
I think that, from my perspective, that's, that's really good.
Roll ups are probably even a little bit better than, than drivechain.
So that my general stance is, I'm in favor, but let's do them.
Let's put that whatever we can do to get to a peg.
Let's do.
But I think the, to bring it back to roll kit, I think there are particular things that
they don't, they don't do that.
They don't put just a proof of what happens on an off chain system on Bitcoin.
They put the entire state into witness data, like every single block, the entirety of the
block goes into witness data.
And that's just stupid.
It's like, and the only reason they're doing that is because they're like, we don't want
to run a node of this sidechain and no one else wants to run a node of this sidechain.
So we'll just dump all the blocks onto Bitcoin.
And that's that.
So the way that I thought it would, I'll share with you the way that I thought it would have
been more interesting.
And like, maybe I should have diverted my own time to focus on this and I'd be out in front
of the roll kit conversation.
I don't know.
But what I thought would be more interesting is to define a state transition function in
the initial envelope and then have some sort of, like you, you use some, you know, multi
sig lock channel lock, right?
And some number of participants run the state transition function and they re-envelope,
we really shouldn't call it inscription because they really like, they're running this sort
of other thing that uses the same technique, but it's not really an inscription.
So they write the state transition function and then following the state transition function,
they write, you know, compressed state updates or, you know, hash state updates or state
routes or whatever it is, right?
Like there's a lot of different ways you can go about that.
And I think that would have been like a lot more, that still would have been a prototype
and not necessarily like the way it should be done in the end, but it would have been
like perhaps more desirable than what we have.
I see roll kit as a conversation starter and in its capacity as a conversation starter,
I like it.
If a year from now, roll kit is like the way we're doing business, then I would probably
join you in some very deep criticisms.
Cool.
Well, I guess we'll just wait a year then and see.
Yeah.
I hope that it's, that it isn't the way that the Ethereum people start doing roll ups,
but yeah, I guess I think they don't really care in my estimation of them.
They don't really care where the, about using data efficiently.
They just care about them not having to store it.
They don't want to have to store all that data.
So if they can shove it onto Bitcoin, I think that I would not be, I will not be surprised
if they're happy to do that and then continue doing super inefficient things, but now just
shove it all on Bitcoin as the data availability later.
I think they want a product development surface.
I think if it meant them storing it, they would store it.
But I think that the sort of hype narrative that comes baked into this notion of roll
ups on Bitcoin, it could definitely mature into like extraordinarily rushed development.
We already see problems with rush development, right?
I think that the inscriptions market right now is in this sort of limbo period where
the lazy developers who are rushing production to be first to market are developing far less
than desirable products, while the people that want to see a more robust, trustless,
distributed infrastructure emerge are hammering away at keyboards, trying to get to the point
where they can demonstrate meaningfully, these are the lazy shits, right?
And that'll happen.
I'm working in that team.
I'm working on those things every day.
It will happen.
It's just a process.
I think the interpretation there for the roll kit thing is the people who do want to see
it mature better, the John Lights, the Dougans, the Yous, the Me's, you just got to hammer
at the keyboard, man.
You got to make the better shit.
The only way to beat that.
I'm doing my best.
I'm doing my best.
Most of it's in the drivechain space, but I see them as pretty much equivalent.
The only way to beat it is to outship, man.
That's the only way to do it.
Yeah.
Yeah, you're right.
I agree with that.
Outship them.
If you don't build the better thing, they'll build the worst thing, and then that'll get
popular.
Cool.
Well, we agree too much.
I thought we were going to have a lot of fun arguing, and we only got a minute of that.
So anyway, what were you going to ask?
I was going to try to perhaps summarize and reset where it sounds like we might be in
agreement about smaller blocks being good and drivechain being good, and that things
like ordinals and inscriptions would, if we adopt drivechain and perhaps also a smaller
block size on layer one, then there wouldn't be any inspiration for censorship, and instead
the activity of users of all kinds would flow to their natural place on a sidechain that
is designed to facilitate that use.
Yeah, except I would say it's still censorship.
If you add a soft fork that creates small blocks, you're still censoring out something.
You're specifically censoring out anything larger than data size X.
Any block that's larger than X is what you're censoring.
So that's still censorship.
It's just censorship of a different thing, and perhaps one that more Bitcoiners would
be inclined to say is okay to censor.
Well, not every limit is a censor, right?
The word censorship refers to the imposition of a value judgment on a constituent of a
set of activities or data, doesn't it?
So if I say in this room we're talking about drivechain and everyone's free to talk about
BIP-119 in a different room, then that's not censorship.
That's just something else, isn't it?
In my opinion, it is.
I would say you're imposing a value judgment, but the value judgment is not that talk about
BIP-118 is inherently bad, just that it's bad in this context, because here we're trying
to talk about something else.
If Twitter says there can only be three hosts per Twitter space, that's not censorship.
That's a design choice.
So if they increase to four or to two instead of three as the maximum number, that's not
censorship.
That's just a design choice designed to facilitate value creation.
I would say it is.
I really doubt that anyone here is interested in a philosophical analysis of what counts
as censorship or not.
Okay, fair enough.
I just feel like censorship isn't really part of drivechain or smaller block size.
It's about the design of the protocol to enable the most value creation.
So that's the trick, is that the protocol, like what Satoshi chose, it required a somewhat
centralized planning, right?
He chose the issuance schedule.
He chose the one megabyte block size limit.
He chose 21 million coins.
So he had to make certain choices based on his view of reality. Yeah. Post, please chime in.
Yeah. So those things, it's not really feasible to start a network. If we wanted to start our own
network right now, some more centralized set of decision-making would have to occur before a
decentralized set of decision-making could. So the way I like to think about it is, you know,
like every idea starts in one person's head. And then if other people like the idea, they adopt it
and potentially augment it moving forward. So I don't think there's an... Excuse me. Hold on.
God bless you. I think, yeah, I agree with that. And I also think that you can launch
without having a central... It's possible to launch stuff that has no central
point of control, even if, well, it's just an idea. Yeah. You can do whatever you want with it.
Like some person sits down and writes the initial code and then it gets launched.
And so I guess my point to that is like using that as the sort of...
I don't know. Words are failing me. But to using that as like the argument, I don't think... I do
think that being really sensitive to the design choices of that person when they wrote it is like
really important. Like why was this decision made and should this decision be upheld at Infineon?
Those conversations should absolutely exist. But it doesn't necessarily always make sense to me
to be like, this decision was made... So the peer discovery process, right? This would be a really
good example. The initial peer discovery process for Bitcoin was like, it was literal jank. It was
complete jank. It was connected to Satoshi's IRC server. Right, exactly. You read a list of IPs off
of an IRC channel and you connect it up to them. So that had to be evolved, right? Like that was
an initial decision that was made that we absolutely would be like in the dumbest of
positions if we had said, oh, we should never change this because this is the way Satoshi did
it or something to that effect. So there are a lot of things he did that I think should never be
changed, right? Like if we can never change the 21 million supply, we should not, right?
But I don't think... I don't know. I'm not necessarily even accusing Henry of doing this,
but I think that it is sort of unfortunately common that we do things like say, Satoshi
designed it this way and that should never be changed. I think that we owe it to ourselves
to have better conversation about what should or should not change. I think I am sometimes
tempted to do that. And I wouldn't be surprised if Henry is sometimes the siren song of just
saying, well, Satoshi did it this way because everyone respects him. Satoshi agreed with me on
this. Therefore, you should agree with me too. But of course, Henry knows as well as anyone else
that Satoshi was wrong sometimes. So that's not a sufficient argument. The appeal to Satoshi's
authority is not sufficient, but there's nothing wrong with saying Satoshi agrees with this too.
Yeah, from a historical perspective, what Satoshi thought is a valid inquiry. We can,
as a historical matter, discuss scientifically what did he say based on the evidence of the
record of his public statements and messages. So that's interesting. But of course, what he said
isn't necessarily the truth. He failed to invent the two-way peg for sidechains. So he thought
in the BitDNS thread that every blockchain would be a merged mined altcoin of Bitcoin. He didn't
think of a two-way peg for a sidechain, which is a superior economic design because then instead of
merely sharing fees, every blockchain also shares the same unit of account. So the synthetic Bitcoin
on the sidechains trades one for one with the Bitcoin on layer one because of the two-way peg.
He failed to invent that. That's DriveChainBit300. So he also was like incorrect about a few other
things, like his understanding of network security wasn't the same as ours because he didn't conceive
of a hostile perpetual miner. And therefore he said, perhaps in the future, people will be able
to process fees for free, which wasn't quite right, this idea of relying on ideological mining.
And also he said that we wouldn't care how big the nodes get in the future. It wouldn't matter.
So he wasn't cognizant of our, he didn't have our sense of cheap nodes as being important. So that's
why the big blockers, they focus on those few sentences where he said that we, he effectively
said we'll have just huge blocks and it won't matter that people, ordinary people won't run nodes.
Today, we have a different view, at least within Drivechain, that small blocks and perhaps smaller
blocks is ideal and let the big blocks be opt-in, not mandatory.
I also, Sochi also changed his mind sometimes, and that's important to keep, or that's worth
noting. It's like sometimes he'll say one thing and then later he'll change his mind about it.
Or he'll have like a nuanced take where like there's two conflicting things that he,
nevermind. Yeah, I forget. I had something else and I forget what it was.
Another thing that I disagree with Satoshi about was his Rothbardian
notion that fractional reserve banking isn't good.
Oh yeah, I remember what it was. He didn't, he may not have come up with the two APEG
in that one post where he was discussing merged mining, but he did in the white paper, he talks
about escrow, which it seems that he conceived of as essentially a two of three multi-sig.
And in other posts, he talked about a two of two multi-sig as another way of doing escrow without
a third, without a trusted third party. And I suspect you can make a two way peg out of that.
So he may have thought of the tools we need to do a two way peg without inventing the concept.
But that's not a peer to peer two way peg. And I should have used that more extensive description
of what I was talking about. So BIP300 aspires to be a peer to peer two way peg. Whereas
what Satoshi was talking about in his escrow conversations, and he did seem very excited
about them, is he was interested in having like me send you an order for a box of shoes and you
send me the money, but you're not going to want to send me the money until the box of shoes arrives
in the mail. And therefore the plan would be that you send your money, your payment to the escrow
account. And then I see that it's there. So I know you're good for the money. And then I released
the shipment to you. And then when you get the shipment, you release the payment to me.
And that is Satoshi's arrangement for like escrow without a third party. That was the thing
that he was very excited about is that kind of anarchic voluntarist kind of Silk Road style,
you know, payment system where there's there's minimal reliance on trust. It's not perfect,
but it's trust minimized because I don't have to trust that you have the money. You've proven you
have the money. And then you send me the shoes. And then when you get them, you release the money
to me. So there is a worst case scenario where you spike me and you get the shoes and then don't
release the money. But at least you can't get the money back ever. So that was what he was talking
about, not a two way peg between blockchains. Right. But I think just as Liquid uses a
a large, what is it, 12, 11 of 15 multisig to do a two way peg, I think you can also use
a small, you know, two of two peg or two of two escrow to do a two way peg and then not have
the problems of like having of needing to trust the 11 of 13 people, to be honest.
Wait, can you explain how are how aren't the people being trusted who are running the
federated multisig? Those people are. If you get rid of them and do instead of instead of doing
that model of escrow, you do the two of two escrow that Satoshi talked about. Then if you do that
instead, then I don't think you have to trust the other people. I don't understand. I think that
like I gave you the shoes example where someone's buying shoes from someone else.
And then I talked about how that reduces trust because of the existence of an on chain verifiable
publicly visible escrow account where you send your payment as like an intermediate stop before
it gets to the goods seller. I don't see how that's related to what Liquid does.
Do you hear me?
Super, can you hear me or post? Are you there?
Yeah, I am here. I got distracted for a minute. No problem. Other related nonsense going on.
Sorry, maybe we lost super. But what I was saying is basically, oh, he's back. Okay.
Super, you're. Are you there now? Can you hear me?
Oh. Okay, great. Yeah, my app crashed. Yeah, I don't know. But basically, we've we've touched
on this before. But the basic fundamental idea in my mind is that, like in drivechain,
the two way peg is peer to peer. And like, that's how I see it. I think that's how Paul sees it.
I think that's how Adam Back sees it. Like, it's peer to peer because anyone can get involved in
mining. You don't need anybody's permission. And therefore you can it's peer to peer. You don't
need privileges to become a miner and participate in controlling the hash rate escrow. But in
liquid, all the money in the world might not be enough to get you control of that federated
multisig. So it's not peer to peer. It's not there's there's more than an economic barrier.
There's a social barrier to entry to participate in this in the role of a functionary on liquid.
So it's not peer to peer. That's the difference. That's what blockchain is all about. The whole
space is all about peer to peer protocols. That's like the most important narrative of any in my
personal understanding, which is probably wrong. But let me know your thought.
Yeah, I'm saying if you if each individual who pegged into the sidechain did so through a two
of two multisig and not through an 11 or 15, then I think you get you can get rid of
the get rid of the Federation and still have the two way peg.
I just don't know what that means. But I appreciate you sharing it.
It's not fully worked out yet. So I've got to do some more thinking about it.
And maybe it's just a dumb idea. I mean, are you saying that like billions of people would be
each participating in a two of two multisig? And that would be a user experience that's viable?
Yes, except probably not billions. I think sidechains are probably only get millions of
users each. But yeah, that's that's the basic idea. If if if the pegging operation involves
creating a two of two multisig, then I think you could do it without needing to trust a
without needing to trust anyone but yourself.
Interesting. Well, thank you very much for all of the conversation today. This has been another
drivechain Twitter spaces we've rebranded as drivechain talk, or Bitcoin drivechain talk
is the title here. So we tried that out today. I thought it would maybe be a smash hit. We got
a number of people in the audience and people on stage. Thank you super imposed for being here. I
really appreciate it. I really love listening to both of you speak. I hope that you'll always come
back anytime you'd like to share ideas or listen or chat. So we run these spaces every Monday and
Thursday, usually at 930am pacific time, although we're flexible to have different times. If either
of you would like to do like a debate or a talk about anything in the future, please don't hesitate
to let me know. Henry at LayerTwo Labs like I would love to schedule something around your time
or because both of you are people who I think understand drivechain very well and have all
kinds of exciting other ideas to talk about. So you're welcome to come back whenever you want and
as much as you want. So maybe I'll see you on Thursday or any other time. And our goal here is
to promote understanding of drivechain. I think that we have the truth on our side. So with enough
talk of drivechain in the community, we can incite more support for it and maybe get it to be part of
Bitcoin. Thanks for your time and for hosting. I'm going to I'm going to head out so I can go get
some lunch. See you later. Adios. Thank you for being here and post. Do you have any closing
thoughts? No, I was just going to say thanks and have a good one. That's it. Thank you very much.
Have a great day.