Stephen DeLorme: How Does Bitcoin Consensus Work? Part 2 - Atlanta BitDevs - (EVNT007) - ATL BitLab Podcast

Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Stephen DeLorme (00:00):
So you have a new rule block and new rule block,

(00:01):
and then miners start, you know,mining these legacy rule blocks.
Um, and, uh, they, they, theyget into this idea of a bounty.
So the idea is, let's see, aconsensus change, uh, has just
been confirmed and like, youknow, the last block or whatever.
And as users, we're all gung-ho and it'slike, yeah, they just, you know, you

(00:23):
know, activated some like really cool newop code that like allows me to secure my
Bitcoin in some novel and interesting way.
So I'm gonna send it to like, you know,this particular address and lock my
coins up in this newly activated op code.
Okay.
What you've just done there is you'vecreated a bounty on your coins.
You have created a bounty, um, thatincentivizes malicious miners to change

(00:48):
the rules and use the old client,
This podcast episodeis an event recording.
If you're listening to the audioversion, you might be missing some
context from the speaker's visuals.
You can find the videoversion at atlbitlab.
com.
That's A T L B I T L A B dot com.
There might also be audiencequestions or other background

(01:10):
chatter that's not audible.
Look, event recordings are neverperfect, but we're sharing it here
because we think you're going tofind something valuable in it.
Let's talk a little bit about our sponsorsfirst, and then we'll get onto the show.
This episode is sponsored by ATL BitLab.
ATL BitLab is Atlanta'sfreedom tech hacker space.

(01:30):
We have co working desks,conference rooms, event space,
maker tools, and tons of coffee.
There is a very activecommunity here in the lab.
Every Wednesday night isBitcoin night here in Atlanta.
We also have meetups for cybersecurity,artificial intelligence, decentralized
identity, product design, and more.
We offer day passes and nomad passesfor people who need to use the lab only

(01:50):
occasionally, as well as membershipsfor people who plan to use the lab
more regularly, such as myself.
One of the best things abouthaving a BitLab membership isn't
the amenities, it's the people.
Surrounding yourself with thecommunity helps you learn faster
and helps you build better.
Your creativity becomes amplifiedwhen you work in this space.
That's what I think at least.

(02:10):
If you're interested in becominga member or supporting the space,
please visit us at atlbitlab.
com.
That's A T L B I T L A B dot com.
All right, on to our show.
All right.
So we're here to talk about theBitcoin consensus analysis project.
This is part two, and thisis gonna be the last part.

(02:32):
We cover this.
We don't want to beat this thing to death.
Um, Nitesh uh, covered thefirst part, um, of this.
Um, we, we were trying to schedulethis out in advance and we couldn't
quite work the schedule out to wherehe could, you know, present both of
them in a concise period of time.
So I offered to take the second part.
So, um, and it's a, I'm happy to do itbecause it's a very fascinating paper.

(02:56):
So, uh, this is basically, as itsays, analyzing Bitcoin consensus.
It's risks and protocol updates.
So what the authors of this paper triedto do is they're trying to analyze,
uh, how, uh, people in bitcoin.
Reach consensus, both at like a sociallevel and at a technical level, um, to,

(03:17):
uh, make changes and upgrades to Bitcoin.
Um, and you know, there's alot of, you know, discussion.
Uh, somebody was just mentioning me beforethis, that there's apparently a lot of
more soft fork chatter going around onBitcoin, Twitter today or something.
So, you know, this is something thatkeeps coming up again and again.
But I think this is one of the first,um, attempts I've seen to really break

(03:39):
down, not just the, like, how does itwork at the code layer, but how does
it actually work at the social layer.
Um, so I'm gonna do a brief recapof what Nitesh covered last time.
Um, in the first half.
And I highlighted some stuff.
One thing that, uh, he spoke to usabout was soft forks and hard forks.
And the, the, the basic idea withsoft forks and hard forks is this

(04:01):
is, uh, you know, when you are makinga change to Bitcoin's code base,
a soft fork is kind of like a, a,um, a backwards compatible change.
And a hard fork is a notbackwards compatible change.
So, uh, a soft fork is it, ittightens the rules typically.
Um, it, it, it might, uh, you know, makethe rules tighter or more restrictive.

(04:22):
It's not gonna loosen the rules, though.
Um, and, uh, so yeah, the, the ideais, uh, like a hard fork is always
going to, uh, result in a chain split.
Typically it's going to, um, result inthis situation where, uh, you know, uh,
Bitcoin nodes that fail to upgrade theirsoftware will not be able to participate

(04:42):
in, in, uh, that same blockchain.
So they'll have to have their own,you know, split of the blockchain.
Soft works don't necessarilyhave to result in, in a chain
split, but in certain situationsthey, that could happen.
Um, we'll, we'll dig into chain splitsa little bit more as we get deeper into
tonight's, uh, Nitesh also walked usthrough some activation mechanisms.

(05:04):
The earliest one was flag day.
It was like, you know, at thisdate, uh, or at this block height,
the, the change will take effect.
Um, that kind of worked when,you know, Bitcoin was like 10
nodes on the network or whatever.
But then we started introducingthis idea of like miner signaling
and stuff like that, like BIP nine.
So it's this idea that a miner canactually signal that they're ready

(05:25):
for, uh, a hard fork by putting,you know, some metadata in their,
their block header and all of that.
And then we get into this idea ofuser activated soft forks, which is
this idea that, that, that we call'em the users, the people actually
running Bitcoin nodes, not necessarilythe people, uh, doing miners.
They could actually, uh, youknow, run a Bitcoin node that,

(05:46):
that activates a soft work.
They could run a modified Bitcoin clientor an updated Bitcoin core, um, that,
you know, uh, you know, basically acceptsblocks that, uh, have this new change.
So, um, do you, is there aquestion back there that I see.
Can you say the question again?

AUDIENCE (06:04):
I was just wondering like, um, there were, there's like
a lot of soft works in the earlydays of Bitcoin development, right?
Like sometimes Satoshi they say thatSatoshi released a version, wouldn't
even say what it was, what it fixed.

Yeah.
Were there a lot of soft works in theearly days of Bitcoin's development?
That's something I'mactually a little, yes.
Bitcoin did change a lot.
Like Satoshi in the early days was ableto kind of unilaterally make changes,
um, because, you know, there, there washardly anybody on the network, right?
And so, um, you know, Satoshiwas able to make a quick change,

(06:42):
you know, get on Bitcoin talk orwhatever, say, Hey, I fixed this or
whatever, and, uh, it, it would work.
Um, I'm not actually sure if those, uh,qualified as soft forks or maybe some
of those might have been hard forks.
I'm actually not sure on that.
Um, my, my thought is some ofthose might have been hard forks,
but I'm open to being corrected onthat 'cause I'm not entirely sure.

(07:04):
Um, cool.
I'll move on.
The other kind of, you know,things that Nitesh covered
last time was state of mind.
This is like, I'm gonnatry and blow this up.
Can, can everybody like see this text?
Okay.
Yeah.
Is that good?
Okay.
State of mind is kind of likethe, it's they, they abbreviated.
SOM, you know, you can, uh,I think some real engineering
minds worked on this paper.
Um, so som one, passionateadvocate for change.

(07:26):
SOM two, supportive of a change, uh,som three, apathetic or undecided.
Uh, four is unaware.
Five is not supportive, but notto a degree to spend time, money,
resources toward fighting it.
And then six is passionately against achange and willing to expend resources
and exercise power to fight it.
So as we're getting more into tonight'sstuff, it's like you have this idea that

(07:47):
like, one and two are kind of like pushingin the direction of supporting a change.
Uh, five and six are pushing againstthat change and like, like three
and four are kind of in the middle.
You're either unaware or you're apathetic.
So one and two or four, five and six areagainst three and four are in the middle.
Neutral question,

AUDIENCE (08:06):
just to clarify, when you talking about the state of
mind, that's really more relevantto like bed 32 time changes?
Correct.

It's relevant.
It's, it's describing the,the, the actual people.
Mm-hmm.
Um, and it, it would be relevant to BIP

AUDIENCE (08:19):
context and attachment to what happening.

It would be re it would be relevant to the bip.
The bip.
Was it 34?
Was that it?

AUDIENCE (08:26):
I thought it was 39.

39.
I'm sorry.
'cause I get them all confused'cause they're all freaking numbers.
34. Oh 30.
Uh, yes, it would be relevant to the bip.
34 changes that like where aminer actually, or I'm sorry,
I meant to say BIP nine.
I guess it would berelevant to both of 'em.
BIP 34 is, uh, block versioningand BIP nine is miner signaling.
So yes, the state of mind would berelevant there because the person

(08:48):
operating the mining equipment doeshave to determine, am I for this?
Am I against this?
Am I empathetic?
Um, and it would also be relevant foruser activated soft forks as well,
because the person who needs to choosewhether to upgrade or not upgrade their
Bitcoin note has to choose am I for it?
Am I against it?
It's relevant for all of them.
It's basically for any person.

(09:08):
It's relevant.
Yeah.

AUDIENCE (09:10):
Question.
So, um, I remember back in the day,um, you know, there used to be a way to
basically signal in your node, uh, youcould do a user agent comment and just,
you know, put a a basically a messageto the other node operators in there.

Hmm.

AUDIENCE (09:24):
Signaling for or against different things.
I mean, you could putit whenever you want.

Interesting.

AUDIENCE (09:28):
Um, is that still a thing as far as, you know, trying to gauge
consensus from the, my perspectiveabout the user agent comment?

I'm actually not familiar with the user agent comment one.

AUDIENCE (09:40):
Really?

Yeah.
I'm not sure.

AUDIENCE (09:41):
Old school, this is going back a long, long time.

Yeah.
Interesting.

AUDIENCE (09:44):
But yeah.

Huh.

AUDIENCE (09:45):
It's called user agent comment and it's just a, basically
an option in the configuration file

mm-hmm.

AUDIENCE (09:51):
Where you can add that and then add a note, you know, so,

and it'd be like a what, like a, you know,
plain English note basically.
Yeah.
Oh, pretty much.
Very interesting.
I'm not aware of that.
Okay.
Another question right there,

AUDIENCE (10:00):
just to add real quick to that.
Uh, what Nitesh's going last timethough is they primarily look at who's
accepting the updates to understandthe popularity of our for providers.
Right.
That accurate.

Look at Yeah.
Who's accepting it.
Yeah.
But, well, it'll get a little bitcomp, it'll, it'll get a little
complicated as we get into it tonight.

AUDIENCE (10:18):
Gotcha.

The other kind of, I guess final point to cover after state
of mind, just to recap from what Niteshsaid, was these different stakeholders,
economic nodes, these are, to put itsimply, 'cause I don't want to get
too bogged down by the details here.
Think of these as companies likeexchanges that facilitate the
transfer between Bitcoin and othercurrencies, um, uh, investors.

(10:40):
So that's basically anybodywho invest in Bitcoin.
So if you have a cold storage walletwith, uh, you know, Bitcoin under your
self custody, you are an investor.
Um, if you are someone who is holdingBitcoin on an exchange, well, yeah,
it's not the same thing as selfcustody, but you're still an investor.
Or if you're someone who is like managingan ETF, um, of Bitcoin, again, some

(11:04):
people feel diff, you know, certain waysabout this, but under, at least, uh,
in their paper, they classify all ofthose as different types of investors,
and they do differentiate betweenthose different types of investors.
In the pa uh, uh, the paper, obviouslypeople, self custody investors
are a little bit more nimble.
They can act a lot more quickly thansay, someone who's managing an ETF.
But, um, they, uh, they're,they're all kind of in this bucket

(11:27):
of the investor stakeholder.
Uh, then you've got media influencers.
So just think of these like,um, podcasters, like the Stefan
RAs, Natalie Brunes, PeterMcCormack, McCormack of the world.
Um, you've, uh, got your, uh, miners,obviously people, whether you're running
a single mining rig or whether you'reoperating like a pool or whatever.

(11:49):
It's kind of like that'sthe, that's the, the idea.
Uh, it's, it's miners cansignal for certain changes.
They can direct their hashpower in certain places.
Um, then you've got yourprotocol developers.
These are your people, uh, workingdirectly on the protocol itself.
That typically means peoplewriting code for the piece of
software known as Bitcoin core.

(12:11):
It could also be other things likeother alternative implementations like
BTCD and hypothetical future Bitcoinclients that don't even exist yet, but.
Protocol developers are working directlywith the Bitcoin protocol itself, so
just think of those as bitcoin core devs.
And then you have usersand application developers.
That's our final stakeholder category.

(12:31):
And that might sound a little bitweird here because there might be, you
know, you might, uh, think of yourselfas like, well, I'm just a plebe.
I don't know a lot about code.
I just like to huddle my Bitcoin.
And then you have someone like Chad who'slike made a cool SATs, gg, uh, lightning,
you know, streaming app, and he's enteringa top builder, and you're like, well,
there's a difference between me as a Pleband Chad as an application developer.

(12:55):
Well, actually, according to this paper,you're actually in the same category.
Um, and, uh, you're, you're,you're both the same stakeholder.
And the reason behind that iseven application developers.
Um, even if you write the most complicatedweb application in the world that
interacts with lightning and does allthis cool stuff from the Bitcoin protocol
level, you are a user of that protocol.

(13:18):
So it's like if you're just a plebwith a lightning, like a Bitcoin wallet
and your Bitcoin wallet is talking toa Bitcoin node, Chad has built a cool
web application, but at the end of theday, it's talking to a Bitcoin node.
Bitcoin nodes are the only actual trueparticipants in the Bitcoin protocol.
So whether you're just an average Joeuser or whether you're an application

(13:40):
developer, uh, according to, um, the,the, the way they've structured this,
y'all are just users, applicationdevs are just fan, a fancier way
of saying a user, a Bitcoin user.
So on the same bucket.
Okay, I'm gonna get us intotonight's actual content.
So that was kind of the, the, the, thehigh level recap of what Nitesh talked,
um, uh, uh, LA talked about last time.

(14:03):
So now we're gonna get into, now thatwe've kind of set the groundwork,
state of mind, stakeholders,hard work, soft works, activation
mechanisms, how is consensus achieved?
And so, as they say, I'm, I didn'tprepare a formal slide deck because I
felt like the paper did such a good jobof laying it out in plain English that
I didn't feel the need to repeat it.
So they say Bitcoin consensus change isnot achieved through formal governance,

(14:25):
but through complex interplay actions andreactions among stakeholders, each with
their own unique powers and incentives,they say this process is iterative.
So one stakeholder group will do somethingand then the other stakeholders are
kind of waiting to see what they do.
And then another stakeholder group willcome in and do something and then another
stakeholder group will react to that.
So it's this kind ofback and forth process.

(14:47):
And so SegWit was, um.
You know, well, that was kind of acontentious, uh, uh, change in, in
Bitcoin's history that was, uh, around thetime of the Bitcoin cash fork for those,
um, you know, who may, may not be familiarwith the, the technical word SegWit,
um, but they describe it like this.
So, protocol devs implemented theSegWit code merged into Bitcoin core.
They advocated for its benefits.

(15:09):
Um, they also wrote about the risk ofASIC boost, a method to increase the
efficiency and profitability of miningThat would be incompatible segue.
They disclosed bugs anderrors in alternative clients
such as Bitcoin Unlimited.
They played a crucial role in thetechnical discussions in public
communications that shape the broadercommunities' understanding of the upgrade.
And then again, the miners.
So miners initially showed low support forSegWit, um, but as the market demonstrated

(15:32):
weak demand for competing proposals likeBitcoin and limited and SegWit two x, so
miners face pressure to signal for SegWit.
Um, 84% of the hash power in 2017supported the New York agreement,
but it didn't gain enough traction.
Uh, then you get into the economic nodes.
What were they doing?
Well, exchanges like bitfinex listedfutures markets for Bitcoin Unlimited

(15:53):
and SegWit two x. You might rememberthat for those of us who joined us
for the, uh, block size wars book, um,this allowed for price discovery that
highlighted the economic viabilityor lack thereof of these forks.
So by creating a futures markets forthese potentials, forks, they, they,
you know, allow this price discovery.
Um, that's the power thatonly the economic nodes have.

(16:13):
Um, they also decided how to listthese forks overwhelmingly opting
to list them under separate symbols.
So they didn't call these forks, Bitcoinor BTC, that called them something else.
Um, and then major, no, you know,the economic nodes supported the
New York agreement, but then theyget faced with this grassroots
backlash, and so they wind down.
Um, you have investors, investorsexpress their preferences by trading

(16:35):
futures, contracts for different forks.
Um, you know, they, the sell offof futures for seg and bu that
that indicates a market preference.
Um.
And, uh, the, the market preferenceis kind of like saying don't
change the consensus rules.
Um, and you know, when you havethat many investors selling off the
future, that, that puts pressureon the other stakeholders groups.
This is a quick little marketshowing the, um, the futures markets.

(16:59):
Then you got the users, so the users,um, they advocated for it on social media
for on, you know, forums, other places.
Um, then there's also this discussionof user activated soft forks.
This idea that what if we have a, uh,A-A-A-U-A-S-F client and the users can
decide to activate, uh, seg themselves,um, without needing to get the miners

(17:20):
involved or wait for miner signaling.
And, you know, I, I, I think, uh,what, what they talk about a little
bit in the paper here is there's notreally a lot of evidence that the UASF
itself played a crucial role in SegWit.
But it was kind of like, youknow, perhaps one of these almost
like a, a, you know, an kind oflike a, a, a negotiating tactic.
It's like, Hey, we havethe UASF, we can do this.

(17:45):
Ultimately, segway was activated byother means, but because the UASF there
was kind of thrown out there, it'skind of like a negotiating tactic.
Um, and then the influencers,you know, they, they advocated,
you know, for and against Segway.
It depended on the podcastyou were listening to.
Right.
Um, these people end up, uh, youknow, uh, shaping the, the, uh,
how people feel about these things.

(18:06):
And, you know, they say here thatthe, the influencers and media
outlets played a role in, uh, helpingto reject the New York agreement.
So that's just kind of like how allthese stakeholders, um, you know, uh, uh,
you know, exercise their powers, uh, inalignment with their various incentives.
Yeah.
Question,

AUDIENCE (18:23):
uh, it seems like, does Bitcoin Cash still have a lot of value
in in, in, and why, why is that the case

still?
I don't think it has a lot of value.
It, I don't know what the priceis off the top of my head.
It's certainly not a hundred KI checked,

AUDIENCE (18:37):
it's been up, it's down 1% over five years,

down 1% over five years.
Yeah.
Yeah.
I don't know why it has a lot ofvalue or why people see it as,
some people see it as having value.
I mean, I'm, I'm, I'mchoosing my words carefully.
'cause I think it's, personally,I think it's worthless.
Um, I think, you know, there's pro I Iimagine it's gotta be kind of weird if
you did, if you, if you did go all inon Bitcoin cash after that fork, and

(19:06):
just like if you sold your Bitcoin andturned it into Bitcoin cash, there's,
you know, maybe, maybe you have this,like, hope that it's eventually gonna go
back up one day and, and there might bepeople holding on, um, in that regard.
I, I mean, again, I just don't know howthat market has enough liquidity for that.
I think there's also just agood bit of brand confusion too.

(19:28):
Personally, and this is just my opinion,I don't really know, but I, I, I
think there's a situation where peopleget into Bitcoin for the first time.
So they Google search Bitcoin andit's like, hey bitcoin.com and they
have Bitcoin and Bitcoin cash and itmakes them seem like equal things.
Um, kind of, so yeah, I, I don't know.
I, I don't know why it has,but it's certainly compared to
the Bitcoin's market cap andvaluation is nothing at this point.

(19:50):
Question?
Yeah.

AUDIENCE (19:52):
Going back to the, um, the, uh, financial, the, the exchanges,
you said that they literally createdfutures options on the various branches

Yeah.

AUDIENCE (20:02):
To try to gauge

mm-hmm.

AUDIENCE (20:04):
Price, discovery and, and desirability, I guess.

Mm-hmm.

AUDIENCE (20:09):
So it sounds kind of like, that for me sounds kinda like the,
like predicted or like one of theseother types of like social, like
outcome gambling sites that we seenow that it's a lot more prevalent.

Oh, like a prediction market?

AUDIENCE (20:26):
Yeah.

Yeah.
So, I mean, I think that's a good point.
Yes.
I do personally think it's, it.
I think.
Starting a futures market forsomething does feel, feel somewhat
similar to a prediction market.
I think it serves a, a similar,well maybe on a financial level,
it's not exactly the same thing.
It does serve a pretty,very similar social purpose.
Yeah, exactly.

(20:47):
You're gauging the opinion ofpeople with skin in the game.
That's a great point.
Um, so here's kind of liketalking about segue activation.
It's a lot of the, the stuff wejust were looking at there, but
it's put in more visual form.
So it goes to these key dates fromFebruary 28th, 2015, uh, all the
way through November 8th, 2017.
Um, and it shows the differentstakeholders here and it

(21:09):
shows when they use a power.
And so this is what I mentionedearlier, lightning paper published.
This is users and app devs according to.
The, you know, the authors of, ofthis paper that's, uh, that that's
actually, you know, light, uh, somebodyworking on a lightning protocol or
lightning implementation is actuallya user of the Bitcoin network.
Um, so they're exercising a power bysaying, Hey, what if, you know, we did

(21:32):
this like kind of layer two thing instead.
Um, then you have miners and economicnodes, you know, trying to do
this Bitcoin, um, XT client thing.
You have some protocol devs involvedwith that too, because I don't, I can't
remember the names, but some protocoldev must have been involved in that.
Um, you know, you get Hong KongRoundtable, Bitcoin Classic, you
know, um, you know, the, uh, minersare involved in this discussion here.

(21:57):
Uh, Bitcoin Unlimited, you got theminers, you got exchanges, you have people
talking about it on podcasts and stuff.
Um, you know, you got the,the, the New York agreement.
This has got, you know, you got miners,economic nodes, investors, users.
Um, you know, this was kind of like a,a, an agreement that took place before

(22:17):
the Consensus Conference in 2017.
It was kind of a secret industrymeeting where like the bunch of people
from all these different companiesgot together to try and, um, you know,
discuss a Bitcoin fork basically.
I think that that was one of the thingsthat was really, there was no protocol
developers that I'm aware of, uh, involvedin the New York Agreement, which is part

(22:37):
of what made the New York agreement alittle bit, uh, controversial, is that it
was kind of like, you know, um, you know,again, this is always a, a debate that,
you know, like a philosophical debate.
There are some people whoare like, oh, well, the.
You know, Bitcoin protocol developersare like an elite few who get to
dictate what happens with this.
And, um, you know, but then the, thecounterargument about is like, well, no,

(22:59):
the Bitcoin protocol developers are thepeople who put in the proof of work to
demonstrate they actually know how thistechnology work and their opinions matter.
Right.
So, um, but yes, this was kind of, I,I mean, I see it as kind of a result.
Revolt against the protocol devs.
Yeah.

AUDIENCE (23:13):
Um, so this is all well before my time, so I really don't know a lot
about the conversations that happened.
My understanding is that you basicallyhad a bunch of big economic actors,
like some miners, and it's some bigexchanges who thought that best thing for
Bitcoin for their businesses would be toincrease the block size was a hard fork.

Mm-hmm.

AUDIENCE (23:36):
And they lost.
Um, my question is, had theypursued a soft for, and they
recognized that hard forks were nogood, like absolutely no case can.
They were saying some otherconversations coming right now around
like OP_CAT or op, like CTV, um, doyou think that the lesson would be the

(24:00):
same, the result would be the same?

Hmm, that's a good question.
So if they had tried to pursue, if theminers and other, uh, economic node actors
had tried to pursue a soft fork instead ofa hard fork, would the lesson be the same

AUDIENCE (24:13):
without protocol developer consent?
Like, like core, core, like thebig for contributor consent, they
were against it, but they went withanyway they up cap, for example?

That would be interesting.
Uh, I, I think the lessonmight be the same still.
Um, because the thing about it isthat a, a soft fork could still result
in a chain split, uh, either way.
So I think that like hardforks are definitely.
Uh, contentious and Bitcoin, butSoft Forks are contentious as well.

(24:47):
Like hard forks are almost kind of like, Ifeel like they're kind of like, it's kind
of like the Overton window has shifted towhere they're not even discussed anymore.
Whereas with Soft Forks you canstill, like soft Forks are like in
the Overton window, I feel like,but they're still, they're still
like considered risky and dangerous.
Right.

AUDIENCE (25:07):
The risk is, would be, I guess, from the perspective of the user that they
use the soft fork and there isn't actuallya majority of hash power behind it, and
they're funding getting stolen through a

Yeah.
Yeah.

AUDIENCE (25:22):
So I guess the, the risk in that regard is not that you have like
two versions of Bitcoin, but ratherjust the user loses their money.

Yeah.

AUDIENCE (25:31):
Um, and so I, I just find it interesting 'cause like, like there's,
there's this lesson I feel that waslearned and I, and just well before
time, I don't really know what happened.
But it seems like we almost maybe swungin a direction which was maybe too strong
the other way, which is there is no rolefor economic actors in this process.
It's purely an open source thingled by the core developers.

(25:55):
When you could make an art, youcould had it on differently.
Maybe the less would be that there ischecks and balances, but, um, anyone can
contribute to core, anyone can for it and
Sure.
Yeah.
Question
maybe to refre, uh, maybe, uh, torephrase, uh, Joshua's, uh, question,
is it possible for large economic actorsto like force a change in, even with

(26:23):
protocol developers not wanting it?
So,

yeah,

AUDIENCE (26:27):
because I think we could add like.
ETF.
Mm-hmm.
And, and now even potentiallySovereign Wealth Fund kind of actors
into that second column here, thatmight create more momentum for a
potential change that may be counter

Sure.

AUDIENCE (26:44):
To what developers would be wanting.

I think let's just keep going because I think it'll
all become clear as we keep going.
Uh, I think what the authors of thispaper are actually saying is that
like it already does happen that way.
Like, uh, already what they're tryingto say here is economic nodes already
do exert influence, um, in media.
Influencers do exert influence,users do exert influence.

(27:08):
All these groups do exert influence.
And so that's actually the point of thisprocess is that all of these stakeholders
do have power and they do have influence,but it's the way the, the areas in
which they can wield them are limited.
So I think that these are valid questionsand great points and I think that, um.
You, you, you know, I think someonewho's smart enough to work on

(27:30):
Bitcoin core protocol design.
I, I, I don't wanna put words in anybody'smouth, but I think it's kind of like,
I, I think that, uh, that the discussionaround the protocol, you kind of have
to be able to carry your weight, uh,like intellectually to be able to have
really, really serious, meaningfulconversations about the protocol.
And it, it might be, yeah, I mean,it's kind of like you have to put in

(27:52):
the proof of work to show, you canhave a discussion about the protocol.
Um, but having said that,whether economic nodes, all these
stakeholders, they all have power.
And so that's kind of like whatthis whole paper is talking about.
Let's move on.
'cause that, that's basically what we'retalking about is like, you know, every,
everyone's smart in thinking ahead here.
This, this is basicallywhat this paper is about.

(28:12):
So let's, let's look at this stakeholderpowers during a consensus change.
This next section of the paper istalking about how stakeholders have
varying degrees of power duringthe course of a consensus change.
And so this graph here, Ithink lays it out pretty well.
You have, uh, different, likethe, the, uh, amount of power you

(28:35):
have in a given stakeholder class.
They say changes over the courseof, um, a consensus change.
And so I'm not sure if everybodycan see the, the color super well.
Uh, actually I think, uh, becauseI have night mode on, on this
monitor, I can actually see it,uh, the color's better over here.
So we've got blue line for economicnodes, red line for investors,

(28:56):
or I'm sorry, that's a orange.
Orange line for investors, green formedia influencers, red for miners.
Uh, that is a purple for protocoldevelopers and a black for users.
Uh, the purple and the black kind ofblend together, but we will roll with it.
So let's say you're gonna makeit, you want to, uh, you know,
some kind of new feature or somekind of change to the protocol.

(29:17):
We'll start at the beginning.
We'll roughly say thisis protocol ideation.
So at this point, investorshave the least amount of power.
Miners have a little bit of power.
Um, then they say that, what is it?
The green, the media influencershave a lot of power and the protocol
devs have a lot of power, and theusers are kind of in the middle.
And so the idea here is that theprotocol at the protocol stage, this

(29:40):
is where the, the, the protocol devsare at their most powerful because
this is during the protocol ideation.
They're, they're, uh, coming up withideas, whether it's being discussed in
GitHub threads, mailing lists, posts, chatrooms, conferences, wherever it is, this
is where the ideas are flowing freely.
Um, and a lot of, uh, you know, powerfulconversations happening, um, and maybe

(30:05):
codes being written, but it's all, youknow, experimental kind of stuff, right?
It's not like actual deployable code.
Then at this, this time, it, it, itsounds, it's a little counterintuitive
seeming at first, but the mediainfluencers, they say actually have
a lot of power at this time becausethis is the point in time when.
You know, no one, you know, no one isrunning this code in production, right?

(30:26):
It's just ideas that developers aretalking about and maybe experimenting
with on their own laptops.
So, of course, podcasters, um, you know,invite them on to talk about their stuff.
You know, you've probably watchedepisodes of like Stefan Lara or Peter
McCormack, where they're interviewingsome developer and they're talking
about some, oh, you know, this OPcode's gonna fix everything, right?

(30:47):
Um.
That's kind of what that is, is atthis time, the media influencers
actually have the ability to take ideasthat are being discussed on mailing
lists and at conferences, and really,really elevate these ideas and raise
awareness, um, in, in people's minds.
They also have the ability to bury theseideas or to not talk about them or to
talk shit about these ideas, right?

(31:09):
That, that, that these are all onthe table, but that's the power
that the media influencers have.
So they're like themost powerful ones here.
Then you get into the ideaof the bit being implemented.
Implemented is that's, that's wherethey're writing code for it, they're
testing it, all of this kind of stuff.
Um.
And then you get over herewhere it says, merged into core
signaling threshold starts.

(31:30):
So in this example scenario, we're goingto, you know, um, let's say it's SegWit,
um, or taproot or something like that.
We're gonna actually mergethe code into Bitcoin core.
Um, and there's going to be a minersignaling mechanism in place, uh, you
know, by whatever means, and it happens.
Okay?

(31:50):
So at this point, look, the protocoldevs and the media influencers
actually their power starts to decline.
It doesn't go all the way down, butit goes down the middle at this point.
Uh, this is where the miners start togain more power, because I just said we're
talking about a miner signaling mechanism.
So as soon as it's merged into coreand that code can be download, that
Bitcoin version can be downloadedand the miners can start running

(32:13):
it, their power goes way up.
They can actually start signalingor not signaling it depends on
how they wanna wield their power.
Um, also the, uh, the investors,uh, you know, um, uh, go up during
this time that they say, um, uh, Iwanna double check that in a minute.
I think that was just because the ideathat, um, if a, you know, there's the

(32:35):
possibility of a, you know, contentiousfork, um, investors can kind of, uh, you
know, signal, um, what they want to do.
Um, the economic nodes are starting togain, uh, power at this point as well too.
And so this is where you start gettinginto the ideas of like, you know,
futures markets and things like that.

(32:56):
And when the signaling threshold.
Um, you know, hits and, you know, it, it,it's actually, uh, you know, the, the,
the, the consensus change is actuallygoing to happen because the miners
have, um, hit that signaling threshold.
The, the economic nodesare at their most powerful.
So at this point in time, you know, uh,you know, from the signaling all the way

(33:17):
through, it actually hitting, they cando things like launch futures markets.
Like, okay, you know what, if there isa, a chain split or a new coin, well,
we can start a futures market for that.
Um, or investors could be like, oh man,I don't like these changes to Bitcoin.
I'm gonna sell my Bitcoin and, youknow, buy some shit Bitcoin instead.
Like they, they could do that.
And you know, again, wh whether youthink Bitcoin price action, you know, is

(33:41):
affected by this sort of thing or not, Idon't know that's up for discussion, but
it's like the investors have all theseways of signaling their preferences.
Based on, um, you know, whichexchanges they use, whether they buy
their Bitcoin or sell it, whetherthey, you know, sell off futures
on other coins, things like that.

(34:02):
Um, then you have the, this likething where like the majority of
clients, um, uh, start to adopt it.
So again, the power shifts a little bit.
Um, you know, again, the miners,they, they're still, they still
have the ability to kind of choose.
Um, they, they don't havethe, the signaling power.
They can however, choose if they wantto keep mining under the new, um, you

(34:24):
know, consensus changes or if theywanna switch back to the old ones.
Um, you know, the protocol devs, they,they, their power starts to go down.
It doesn't plummet all the way because,uh, you know, the, the, the, the protocol
devs and the media influencers, theycan, uh, start to shape public discourse.
Like certainly if someone is a veryexperienced protocol dev, their opinion.

(34:48):
Might carry a little bit more weight,um, hopefully carries a lot more weight
if they know what they're talking about.
Um, you know, the, the mediainfluencers, um, you know, still
have big fan followings and all that.
Um, but yeah, uh, the users are,are kind of in the middle here.
They, they don't, they never really getto the, the upper echelons of power here.

(35:09):
Um, but, uh, you know, what's, what'skind of interesting though is that the
investors, the investors get a, a, youknow, the most amount of power at the top.
And I'd say a lot of us, you know,a lot of people fall into the
category, both users and investors.
This is where investors have a ton ofpower, because again, if there is a, uh,
a chain split or something like that, inthe case of like Bitcoin and Bitcoin cash,

(35:32):
well, you can sell your Bitcoin cash.
For Bitcoin.
So you're creating like a, if yousold sell your Bitcoin cash, um, you
know, you're, you're hurting the priceof Bitcoin cash and then you're also
creating more like demand pressureinside of, on the Bitcoin side of things.
So, um, you know, the investorsend up wielding, uh, you

(35:53):
know, a, a ton of power there.
And at this point, I think they showthe, the economic node power going down.
It's because once, once, you know,the majority of clients have adopted
it, it's really in the hands ofthe investors at that point for
the market to kind of decide, um,if it likes this change or not.
Um, so yeah, they all have powers.
It just, it kind of depends.

(36:14):
That's what the authorsof the paper are saying.
Um, any questions, thoughts about,about, about all of that kind of stuff?

AUDIENCE (36:23):
I guess, uh, just a thought I had about, you know, the
economic notes and the investors.
Greg, I. So, I mean, I guessthat's more of a reason for people
to hold their own keys, right?
Because they can vote with their money.
Their money being Bitcoin.

Yeah.

AUDIENCE (36:37):
They believe to be Bitcoin.
Right.
So I mean, just looking at it, it,you know, when you think investors
nowadays, people think the ETF right?
But almost all of 'em, except for one,doesn't even prove they have the keys.

Yeah.

AUDIENCE (36:49):
They just say that they, they don't prove it.

Yeah.
And they do argue, uh, earlierin the paper that yes, if you do
have self custody, it enables youas an investor to move quicker.
Right?
I think ultimately it mattershow much Bitcoin you have.
Um, because like the more Bitcoinyou have, the more kind of, like,
if you think of it as a vote.
Um, it's kind of like the more Bitcoinyou have, the more votes you have.
You know, if you, if if, if, if, if Idon't like something and I sell half a

(37:13):
Bitcoin cash, that's not as much of a,um, a profound statement as if somebody
sells 10 Bitcoin cash for Bitcoin.
But certainly, um, if I haveself custody, I can move quicker.
And so with enough people with selfcustody Yeah, you, you, you have the
ability to move quicker as an investor.

AUDIENCE (37:30):
Yeah.
'cause I mean there's what,19.85 million already mine.

Yeah.

AUDIENCE (37:34):
And there's supposedly only a couple million on the exchanges, right?

Yeah.

AUDIENCE (37:37):
So while they may be looked at as the economic nodes today, the
reality is those who hold the keys arethe real economic actors in the protocol.

Yeah.

AUDIENCE (37:45):
So I think that's just kind of a chart tick.

Mm-hmm.

AUDIENCE (37:48):
Advocate for self custody, in my opinion.

Mm-hmm.
Yeah.
So did I, oh, sorry, go ahead.

AUDIENCE (37:53):
I don't know if, uh, this fits into what we were just talking
about, but was it ibit that changedtheir, their thing such that they
can allow for in kind redemption.
There was a recent changeto one of the ETFs.

What do you mean when you say in kind redemption?

AUDIENCE (38:08):
So in kind redemption, everything that was approved last year.
Oh yeah.
Had, if you, um, if you had like abunch of it Then there's an option and
there's limits and, and you have to bea certain type of investor and stuff.
But there was an option youcould, you could sell for cash.
Well it's just like selling it.

(38:28):
Right.
But for a lot of these ETFs they havewhat they call inkind redemption.
So if it's like a gold ETF, thenyou can take your ETF shares and
convert them to gold and theywill have to give you the gold.
One of the ETFs, I think it was Ibitrecently changed such that if you're a big
holder of that ETF, you could actually.

(38:51):
Say, give me the Bitcoin thatrepresents this block of of ET TF share.
And there would have to be a mechanismwhich basically they would've to
transfer a certain amount fromtheir holdings into your possession.
Again, certain qualifications andstuff, but it kind of, instead of

(39:12):
a trust me bro sort of model thatthe ETFs are under, it's like I
could kind of make a run on the ETF.
Like you would make a run onthe bank if enough of the major
investors wanted to pull out actualBitcoin, they could show that.

Mm-hmm.

AUDIENCE (39:29):
The F actually had it.
It's kind of like lookingis the golden, the

Yeah.

AUDIENCE (39:33):
Knox kind of like thing.

Yeah.
That's cool.

AUDIENCE (39:36):
So my question in this context is like, does that mean that their
economic power is potentially elevated?
If you can move the market or if youcan pull, like, I guess, I guess it's
really equates the amount of Bitcoin,but equates to the voting power.
So in that case, it's kind of slightlyelevated based on that, the nature

(39:59):
of that ETO, uh, that's thoughts.

Yeah, I'm not sure.
Uh, I mean, maybe just by having alot of Bitcoin, uh, I don't know.
so

AUDIENCE (40:08):
I guess the question that I, that comes is like, could a big
hedge fund like rock BlackRock, like.
Yeah.
Do something in the marketthat substantively changes
any of this consensus.
They could, they somehowexert power in a way.

Yeah.

AUDIENCE (40:28):
That is inordinate.
Like

if they're an investor in Bitcoin, they can
do it according to this paper.
Again, that's, that's the thingis that the investors have power.
So I'm gonna move us on here, uh,because we got, uh, we got some
more, more shit to get through.
Um, one point I just wanna highlight here.
The fact that the relative powersof the stakeholders fluctuate
throughout the consensus changeprocess suggests that stakeholders

(40:50):
should have a countercyclical behaviorwith respect to their state of mind.
That was a maybe a, I don't know,very engineery way of putting it.
Um, but the plain English way of, ofsaying it is, um, like you don't, like
when you need to exercise your power.
When, when the time comes,when you are as a stakeholder,

(41:14):
you are at your most powerful.
You don't want to be stuck inthe middle at three or four.
You don't want to be unaware or apatheticof a change when you're the most powerful.
So obviously as a protocoldev, you don't want to be.
Um, you know, apathetic atthe beginning of this process.
Um, similarly as an economicnode, you don't want to be

(41:37):
apathetic or unaware going intothis critical part of the change.
You want to follow this and know whatit is so that when the time comes
where your stakeholder group is themost powerful, you're prepared to act.
so yeah, I mean, again, this isjust, this section here is just
talking about consensus and Bitcoin.
How do you measure it?
How do you actually figureout where everybody's at?

(41:58):
Um, that's kind of, I think maybe whatyou might be, might be getting to, uh,
there with your comment is that, um,how do you actually, you know, before
you start implementing it, determinehow people feel and it's messy.
This is the social layer of Bitcoin.
It's not, um, it, there's not an easy wayof doing it when it comes to exchanges,
economic nodes, you can see theirpress releases and stuff like that.

(42:20):
You can't really see what client,uh, uh, version they're running.
But you can see their press release.
You can look at their productdevelopment announcements.
With investors, you can see how they'refeeling by the price of Bitcoin.
You can see how competingforks of Bitcoin are doing.
Um.
You can see, press, press releasesfrom like major investment firms

(42:42):
and ETFs and things like that.
See what their views are, see howthey're investing their money.
Um, you know, uh, youknow, engagement and views.
That's kind of like how you see, uh,the media influencers, uh, take on it.
What, what topics are they choosingto promote and engage with?
Um, of course they say, youknow, it's sometimes difficult.
You know, you, you know, particularlylike with like Twitter, API, if you really

(43:05):
wanna scale the Twitter API, um, or if youwant to, uh, you know, you know, peruse
the Twitter API at scale, I think yougotta play, pay a decent chunk of change.
But, um, assuming you have, uh, have the,the money to pay, you could in theory
survey social media for social sentiment.
Um, miner signaling, obviously that'sthe version, bits and the metadata

(43:26):
on the blocks you can see press,uh, press releases from major pools.
Um, of course.
A miner could lie, they couldactually, um, you know, do a version
bit signal without actually updatingto a particular client version.
So there's not always a way of knowing.
Oh, with Protocol devs, this isobviously the, the largest one
they can write and implement about.

(43:47):
Uh, they, they can write about stuff,they can implement code, um, they can
advocate for proposals, you know, dopress releases, social media posts,
talk at conferences, write code, writetests, um, Bitcoin Inquisition, it's
like a CNET for testing new op codes.
You can see what people are merging intoInquisition, what people are actually
using on Bitcoin, Inquisit Inquisition.

(44:10):
Um, but, you know, you know,again, as I say, developer's
influence is often indirect.
Um, it, it just kind of depends onif their proposals get accepted,
uh, you know, accepted or rejected.
And it's also, you know, subject,you know, the, uh, a proposal might
be, um, a very strictly defined, um.
Uh, you know, piece of code within the,the, the mind of the protocol engineer,

(44:32):
but it's kind of gonna be interpretedby the community and the way that,
you know, people want to interpret it.
And it's not always, uh, you know, maybeit's accurate, maybe it's not, but it is
gonna be interpreted in a different way.
And then users and application devs,um, you know, again, press releases,
product developments, all that.
Again, application devs andusers are in the same bucket.
Um, but again, it's hard.

(44:54):
Um, they also say, you know, likeyou can actually measure how people
feel about stuff by the absence ofpress releases or the absence of, you
know, you know, commentary online.
If you see a company or an individualnot talking about something, um,
maybe that's a signal that they'renot into it, or maybe they are

(45:14):
into it and, uh, their company isworking on something with this new.
A consensus change and theycan't talk about, 'cause it's
considered a, a, a trade secret.
So, um, you know, they, they suggestthat, um, maybe, um, as a stakeholder,
whatever kind of stakeholder you are,if you see, uh, another stakeholder
who's kind of being silent on an issue,maybe you wanna kind of talk to them,

(45:38):
have a one-on-one chat, just kind of,you know, uh, gauge the temperature,
how they feel about a certain proposal.
That way you can figure outwhat state of mind they're in.
Um, and, and, and all of that.
But measuring actual, like sociallayer, human consensus is messy.
And it's, it's, it's not somethingthat we can do accurately.

(45:58):
They do suggest some ideas for, um, howwe maybe could improve this process.
Um, you know, if we had more transparencyfrom economic nodes, that's basically
like exchanges telling you what versionof bitcoin core they're running and like
publishing it very publicly and openly.
Um, you could have, uh, predictionmarkets for, uh, Bitcoin protocol changes.
So, um, you know, you got poly market.

(46:21):
It's a super fascinating, uh, youknow, it's, it's a, it's a. You know,
alt coin protocol sort of thing.
So, you know, I don't love that.
But the, the product itself is fascinating'cause you can see people bidding on
all kinds of stuff, like from, youknow, who's gonna win the election to,
who's gonna win the Super Bowl to like,you know, is this celebrity gonna say
this word in the next 12 months on tv?

(46:42):
Um, and, uh, maybe you, uh, could makesome kind of Bitcoin based, uh, prediction
market, um, for certain protocol changes.
Um, prediction markets, um, with enoughscale and enough market liquidity seem
to do a good job of predicting things.
Um, but you know, that'smaybe dinner conversation.
Um, maybe there's an implementationof an anonymous cryptographic

(47:04):
at a station mechanism.
So it could allow stakeholders like a,you know, a large Bitcoin holder or a
frequent transactor to, um, you know, signa message saying they agree with a change
without, uh, revealing their identity.
Um, that's a little kind.
Hand wavy and woo woo to me, uh,that's some real, you know, moon
math cryptographic black magic.

(47:25):
I'm not entirely sure how that,that that would, um, work, but I
mean, if you just kind of squint atthe idea, it sounds kind of cool.
Um, and then aggregation websitesthat summarize support, um, or lack
of support for different protocols.
So like, you know, for anyone that'sbeen following my user experience work
over the couple of years, I've launchedwebsites like bitcoin qr.dev, and, uh,

(47:48):
I've helped with the Bolt twelve.orgredesign and also win taproot.org.
These are simple websites that show like,well, which company and Wallet supports?
Taproot addresses whichcompany and wallet supports.
Uh, bolt 12, which supports, uh,BIP 21, you know, uh, QR codes
with lightning parameters embedded.
These are like simpleapplication layer things.

(48:08):
What if you took that idea but youscaled it to something much bigger
industry-wide and made it for protocolchanges instead of for like the little
dinky application there stuff I do.
Um.
So, yeah.
Um, we're gonna get into some like,uh, you know, you know, interesting
weird territory here and, um, I, I, Ikind of don't wanna take questions on

(48:31):
that previous one 'cause I wanna makesure we get into the, like, the really
crazy, weird chain split stuff that Ithink everyone will find fascinating.
Now, when I was reading this, I wasinitially a little bit conflicted with
the order in which they laid out theinformation in this paper and I thought
about rearranging it for everybody andputting chain splits at the beginning.

(48:53):
But then after getting through it,I realized I don't really know the
best way to organize this content.
It's Bitcoin.
It's a lot to wrap your head around.
Um, so at this part of the people,they start talking about how, like,
okay, most of the consensus changes,or really all of 'em so far have
been in Bitcoin core in the future.

(49:13):
Maybe that's not the case.
Maybe people start making.
Uh, Bitcoin written inother programming languages.
And, um, it was, you know, and, andmaybe these, uh, uh, new Bitcoin
clients actually, um, actually, youknow, uh, gain adoption or maybe people
actually wanna run these clients.
Um, so they call these alternativeconsensus clients, um, and,

(49:37):
you know, they, they highlight,it's a major departure.
It could involve risks,but it, it is an option.
Um, so, you know, there couldbe different, um, you know,
motivations for that sort of thing.
Um.
You know, you know, economic nodes,maybe they has something to do
with more trading or more fees.
Um, with the media influencers,maybe it's, you know, just a, a

(49:58):
chance to be in the spotlight.
You know, potential to gain attentionand credibility, be seen as promoting
innovation, all of this kind of stuff.
Um, you know, maybe minerswant to get more transaction
fees out of the whole deal.
Um, maybe protocol devs want to addnew features, all this kind of stuff.
But if you get to a point to where wecan't do that in Bitcoin core, then

(50:18):
people, you know, could start pursuingthese alternative consensus clients.
So they lay out this kind ofnice little flow chart here.
It's like your upgrade pass forBitcoin are right now we're here.
You get your upgrade merged intoBitcoin core, and it basically is gonna
happen probably like historical networkupgrades, like Segway and Taproot.

(50:39):
It could instead be upgrade, couldbe merged into an alternative client.
And if it doesn't reach miner,uh, hash rate threshold, then
you know, the alternative clientfails to activate the upgrade.
And, you know, people probably,excuse me, probably stop using it.
Um, it could be that it actually reachesa miner hash rate, you know, threshold.

(50:59):
Um, and if there is low or medium economicnode adoption, meaning if exchanges the
economic node actors, uh, fail to adoptthe alternative, uh, consensus client,
or they don't, you know, not enoughof 'em do, then you're gonna have a
chain split, which we're gonna get to.
If a high economic node adoptionhappens, then you could have a world

(51:20):
with alternative consensus clients.
And maybe that's the future.
But what could happen though is that,that other path, that chain split risk.
So what is a chain split?
So a chain split's the scenario wherethe blockchain forks into two chains
and you basically have two coinsand you know, this has happened.

(51:40):
Um.
You know, a a a couple times inlike Bitcoin and blockchain history?
Um, I wanna say some, I, I mightbe wrong in this, but I, I want to
say it happened in the very, veryearly days of Bitcoin when there was
like, hardly anybody working on it.
Um, you know, this must like 2010or something, 2011, and the protocol
devs very quickly, like hashedit out and got it fixed, right?

(52:02):
So there was a chain split in theearly days in Bitcoin, I believe.
Um, it happened in Ethereum, I think inlike, uh, 2021, um, due to a bug between
like, uh, uh, between two different,um, implementations of Ethereum.
Um, it also happened, uh, youknow, with us, with, uh, Bitcoin
and Bitcoin cash, you know, weended up with two different coins.

(52:23):
So there was Ethereum, the chainsplit in a 2016 after the DAO hack.
Um, you know, basically, uh,these hackers stole a bunch of
money from a smart contract.
I'm kind of paraphrasing.
And, um, you know, some people inthe Ethereum community wanted to
roll that back and take the moneyback from the attackers, give it
back to, uh, the rightful owners.
And, you know, some people didn't.

(52:44):
And so it forked intoEthereum and Ethereum Classic.
Um, some people did not upgrade tothe Ethereum client that, um, uh,
rolled back those transactions.
And so you ended up with a chain split.
Um, so whenever there's a significantdisagreement like this over protocol
consensus rules, this is gonnacreate, um, uh, a chain split.

AUDIENCE (53:05):
Just a quick note of the recent, uh, $1.5 billion Ethereum heist, uh, with
that occurred, I don't think there's anyrolling it back, but there was discussion.
It seemed about doing a rollback.

Oh boy.

AUDIENCE (53:23):
Applications might be.
Uh, I, I think now we've gone way too farto like even consider it, but there was,
it seemed like a window of time when theEthereum community was considering rolling
it back, which would've probably tankedtheir whole project, in my opinion, but

Wow.

AUDIENCE (53:40):
That's like, uh, it, these things do happen and
these, these issues come up.

Yeah.
That's wild times.
And that, that would've been crazyif there had been another split.
but for anyone, so for anyone who'snot kind of, uh, away, you know, has
never been in Bitcoin for a chainsplit, you know, for those of us who
are here during the Bitcoin cash fork,you end up with money on both chains.
That's the weird thing about it,and that's, that, that's the kind of
hard thing to think about, but it'sbecause the blockchain has diverged.

(54:08):
It's like you have this ledger with allthese pages in the ledger that you're
stacking up, and then suddenly it forks.
And it's like both chains are the sameup until a point, but then it forks
off over here and over here, and you,you, they, they then have different
histories after that point in time.
Um, this is, you know, it's likeimagine, you know, uh, uh, I don't know.

(54:30):
Imagine the time travel movie wheresomeone, you know, fails to, uh, you
know, uh, goes back in time and preventstheir friend from getting killed, and
then you end up with two timelines.
One where their friend is aliveand one where their friend is dead.
Same basic idea.
Um, you know, conceptually, scientificallyscience fiction, metaphorically speaking.
You end up with coins on both chains.

(54:50):
So this is where the economic nodes,uh, you know, come in, in into play.
It's this like market warfare oflike, which coin do people wanna hold?
Are you apathetic?
Are you in the middle?
Are you, state of mind, three and four,and you're just gonna hold onto coins
on both chains and see what happens?
Are you for the chain andyou're gonna sell the original
chain and hold onto the split?

(55:12):
Or are you against the change andyou're gonna, you know, uh, you
know, sell the old and buy the new?
Like, what are you gonna do?
Um, but you know, the,the, the economic nodes.
So investors have a, a role to playin that, and the economic nodes have a
role to play in that because they canassign ticker symbols and they can, you
know, choose which kinds of markets theywanna honor and all that kind of stuff.

(55:32):
Um, so you also have theidea of hash rate allocation.
So miners have a role toplay during a chain split.
Um, miners can decide whichchain, uh, that they want to,
uh, direct their hash power at.
Um, and if a bunch of, you know,miners move to one chain, the
other chain could be 51% attacked.

(55:53):
Like right now, we, you know, wetalk about, you know, what, what
if Bitcoin could be 51% attacked?
What if all the largestpools were one, one player?
Well, imagine if all the hash powerin Bitcoin or like a whole bunch of it
suddenly started mining something else.
Well, then I. You would have, youcould probably have a 51% attack.

(56:14):
Um, you also have this idea of replayattacks, which is an interesting one.
You might remember thisfrom the block size wars.
Replay attacks are basically like, let'ssay I have, um, uh, a transaction on
Bitcoin and I broadcast a transactionto send somebody some Bitcoin.
But somewhere along the waythis, um, you know, uh, Bitcoin

(56:35):
transaction I broadcasted makes itsway over to a Bitcoin cash node.
And, uh, my Bitcoin cash, uh, accidentallygets spent as well, or vice versa, right?
So you have this situation whereyou only intended to spend your
coins on one chain, but you know, itaccidentally got spent on both chains.
Um, so these are replay attacks andI believe with, uh, I believe there

(56:56):
were replay attacks, um, replay,attack, uh, mechanisms, like prevention
mechanisms put in place, um, uh,uh, before segment was activated.
But, uh, I might needto be corrected on that.
so you know, the long-term viability,it's like the success of each
change depends on its ability togarner sufficient economic activity.
You need people using thechain for economic activity.

(57:17):
You need developer support.
You need, you need miner participation.
Um, there's also like this debate kindof, that can happen in a chain split.
Like which one is the real Bitcoin?
And I think at this point, youknow, we all feel like, hey,
Bitcoin's the real Bitcoin.
It's the one called BTC,it's Bitcoin core, right?
And um, but this is likea ideological difference.
If you, you're on the other sideof that, maybe you think the

(57:38):
other one's, you know, Bitcoin.
Um, so yeah, chain splits are,uh, uh, pretty, um, you know,
uh, pretty, uh, scary situation.
Um, oh yeah, this is what I reallywanted to get into is like all
these, these things right here.
Okay.
So there's this idea that after aconsensus chain change comes into

(58:02):
effect, um, like let's say the minersignal for it and, you know, the
consensus change updates, minerscould still change their tune, right?
So like, let's say you havethese blue blocks, which are
the newly proposed rules.
These are like, call thesethe soft fork blocks.
These are like the blocks thathave some new Bitcoin feature

(58:22):
that was just activated.
So this consensus changes activated wheremining blue blocks that contain all of our
new rules, and then somebody broadcasts alegacy transaction that's like, you know,
um, not viable under the new tightenedrule set, but they broadcast it, but it
has a ton of fees associated with it.
It's like, yeah, it, you're a miner andyou see this transaction come in and it's

(58:46):
like, okay, these aren't the new rules.
These are the old rules.
But there's a ton of fees here.
I kind of want to include thesetransactions in a block, even though
it breaks, it's the old rules andnot the new rules called a bribe.
Called a bribe.
Banjo says, yeah, this is,that's exactly what can happen.
So in this scenario righthere, the pool mines it.
So you have a new rule block and newrule block, and then miners start, you

(59:09):
know, mining these legacy rule blocks.
Um, and, uh, they, they, theyget into this idea of a bounty.
So the idea is, let's see, aconsensus change, uh, has just
been confirmed and like, youknow, the last block or whatever.
And as users, we're all gung-ho and it'slike, yeah, they just, you know, you
know, activated some like really cool newop code that like allows me to secure my

(59:33):
Bitcoin in some novel and interesting way.
So I'm gonna send it to like, you know,this particular address and lock my
coins up in this newly activated op code.
Okay.
What you've just done there is you'vecreated a bounty on your coins.
You have created a bounty, um, thatincentivizes malicious miners to change

(59:54):
the rules and use the old client,

Stephen D's video recording (59:57):
uh, because let's say I'm like, yeah, you
know, let's say I'm like super rich andI've got like 10 Bitcoin or something.
Like, I love this new op code.
I'm gonna send, I'm gonna lock up10 Bitcoin with this new op code.
Well now I've just created a10 Bitcoin incentive for the
miners to change the rules on me.
So, you know, I. I, I newlyproposed rules transaction.

(01:00:19):
I, I put my, my, my Bitcoin in thisnew op code, it gets mined in a block.
I put another, uh, you know, one in there.
It gets mined in a block.
Something is weird aboutthe font, uh, color on this.
I don't know why it's like showing up solight, but it says newly proposed rules.
Block Bounty, 200 Bitcoin.
I've sent 200 Bitcoin to anaddress locked up in a novel new

(01:00:40):
op code that was just activated.
So then somebody, uh, sees thisand they're like, well, under the
old rule set, uh, you know, my,if I use an older Bitcoin client,
it doesn't recognize the new, thenewly tightened rule you're using.
So I could actuallyspend that transaction.
And then a pool sees it, a miningpool sees it, and they mine it.

(01:01:04):
This creates a chain split, and so there'sone, uh, there's one block over here, uh,
that, that still has my Bitcoin in it.
It still has my 200 Bitcoin in it.
And then in the legacy rules,fork miners are mining blocks that
have basically stolen by Bitcoin.
So there's a chain split there.
And you know, I mean, it's,it's an interesting thing.

(01:01:26):
Then it's like, well, which chain wins?
And obviously my Bitcoin is on,you know, I want the blue chain to
win because it has my 200 Bitcoinlocked up in this novel new op code.
But what if the orange chain wins?
What?
Sorry?

AUDIENCE (01:01:40):
Only one miner's gonna get that 200 Bitcoin, right?
So why would all the other miners on top?

Stephen D's video recording (01:01:45):
Well, well think about it like this.
So what if, um, what if I.Um, what if there, what if?
And instead of just thinking aboutit as me, uh, locking up my, say, 200
Bitcoin in this, uh, uh, you know, newop code, let's say a lot of people do it.
So let's say all around the world, abunch of people start, you know, locking

(01:02:06):
up their Bitcoin in this new op code.
So a bunch of people create transactionsspending the coins from the new op
code, and they pay a huge transactionfee to the miners because you're
stealing somebody's Bitcoin, right?
So it's like, it was a hundred Bitcoin.
As a miner, I could justpay you 10 Bitcoin and you'd
probably be good with it, right?

(01:02:26):
Like, that would probablybe enough for you.
And let's say I broadcasted a ton ofthese transactions, a lot of miners would
probably wanna scoop up that money, right?
So, but I, I mean, it just dependson how much, if there's enough of
these bounty transactions beingclaimed, maybe enough evil miners
start mining these blocks to where.

(01:02:48):
You know, to gain the transaction feesto where they don't wanna roll it back.

So it's almost like the bounty has to reach like
a critical mass to incentivizeminers to cause a change split.

I think that's a good way of putting it.
That, yeah, I, I, I, that wouldbe the way I would imagine it is
that the bounty would need to hita critical mass to incentivize
enough miners to want to split.
Yeah.

So the inference here is do not use a new feature for some
period of time after it starts.

I think that is a correct inference to not use
a new feature for a little bit of time.
Like maybe I. If a, if a new consensuschange comes into effect, maybe don't
send out of an abundance of caution.
Perhaps don't send your coins to any kindof, uh, address or script that, um, you

(01:03:37):
know, utilizes some novel new featureuntil the network is kind of stabilized
and we're certain that there's not achain split and, and, and all of that.
Um, so yeah.
Um, I, so I think that's about it.
Um, that, you know, we, we've kindof beaten the bounty thing to death.
Um, and I, I think I'm gonna go aheadand just bring us to a conclusion here.

(01:04:02):
Um, yeah, that's probably it.
There's a lot of stuffwe could talk about here.
Um, oh, I think this might be a good placeto just kind of wrap it up right here.
So this is kind of the, thefinal questions that the,
the author has put forth.
I'm gonna, I. It's not super long list.
I'm not gonna like readevery line in this.
I'm just gonna coverthe key bullet points.

(01:04:23):
Um, so assessing whether a proposed changeis achieved consensus can be challenging
due to bitcoin's decentralized nature.
Consensus is not formalized through votes,but it's instead gauged by the absence
of strong sustained opposition and theoverall sentiment of the community.
So you can look to severalindicators and resources.
So I say one mailing list discussions two.
GitHub activity three, the BIP process.

(01:04:43):
We talk about those a lot at bit devs.
People writing these documentswith suggestive improvements.
Uh, technical conferences, workshops,podcasts, uh, miner signaling, node
adoption, economic node statements.
In other words, press releasesfrom exchanges, tweets from
exchanges, things like that.
Um, community sentiment.
This is a little squishier andsoft, but how do people talk

(01:05:05):
about it at your local bit?
Devs meetup.
How do, um, you know, uh, you know,people talk about it on Twitter,
just this kind of, what's thevibe when talking about something?
Technical analysis and reviews, likelooking for people that have really
reviewed the code for new changes.
What do they say about it?
Um, have people run this on Testnetand cnet, and if so, are people

(01:05:27):
using it on Testnet and cnet?
And is it working?
Is it working as expected?
Are there bugs?
That sort of stuff.
Um, market sentiment, youknow, derivatives, futures,
the price of Bitcoin itself.
Um, and then user activation.
This is the extreme case of peopleactually using UASF clients.
So these are all things that, whateverkind of stakeholder you are, and I think

(01:05:49):
a lot of us in the room, we probably fallinto multiple stakeholder categories here.
Um, and only you can know whichof those categories you fall into.
Um, but you know, we, we all, I thinkbelong to a, a different subsets of
different stakeholder categories.
That's, that according to all those paperthat's, that's, uh, from what they've
analyzed, that's how this process works.

(01:06:09):
So now you know your powers, you know,when you're the most powerful, you,
your, your, your role is to go out intothe world now and be involved in these
discussions and figure out how to, uh,best exercise your power, um, so that
you get the Bitcoin that you want.
So that's, that's it.
That's tonight's conversation.

Stephen DeLorme (01:06:36):
Hey, thanks for listening.
I hope you enjoyed this episode.
If you want to learn more aboutanything that we discussed, you can
look for links in the show notesthat should be in your podcast
player, or you can go to atlbitlab.
com slash podcast on a final note.
If you found this information usefuland you want to help support us, you
can always send us a tip in Bitcoin.

(01:06:57):
Your support really helps us that wecan keep bringing you content like this.
All right.
Catch you later.

All Episodes

Stephen DeLorme: How Does Bitcoin Consensus Work? Part 2 - Atlanta BitDevs - (EVNT007)

Episode Transcript

Popular Podcasts

On Purpose with Jay Shetty

Dateline NBC

Cold Case Files: Miami

.css-15opob5{left:0;position:absolute;top:0.8rem;} All Episodes

.css-14f5ked{margin:0;word-break:break-word;display:-webkit-box;-webkit-box-orient:vertical;box-orient:vertical;-webkit-line-clamp:2;overflow:hidden;}Stephen DeLorme: How Does Bitcoin Consensus Work? Part 2 - Atlanta BitDevs - (EVNT007)