WEBVTT

1
00:00:00.160 --> 00:00:03.040
<v Speaker 1>Welcome to the deep dive. So you sent over a

2
00:00:03.080 --> 00:00:06.240
<v Speaker 1>pretty hefty stack of sources this week, a really technical

3
00:00:06.280 --> 00:00:08.880
<v Speaker 1>look at the foundations of blockchain.

4
00:00:09.000 --> 00:00:11.560
<v Speaker 2>Yeah. Dense stuff, yeah, but important if you want to

5
00:00:11.599 --> 00:00:12.679
<v Speaker 2>get past the surface level.

6
00:00:12.839 --> 00:00:15.759
<v Speaker 1>Absolutely, we're really trying to sidestep all the investment talk,

7
00:00:15.880 --> 00:00:18.800
<v Speaker 1>the hype around prices. Our goal here is to get

8
00:00:18.800 --> 00:00:23.239
<v Speaker 1>a clear blueprint, like what is the underlying math and crucially,

9
00:00:23.559 --> 00:00:26.399
<v Speaker 1>how does it enforce this idea of an unchangeable record,

10
00:00:26.679 --> 00:00:30.160
<v Speaker 1>especially comparing you know, bitcoin and ethereum they work quite differently.

11
00:00:30.199 --> 00:00:32.240
<v Speaker 2>That's the core of it. To really see where this

12
00:00:32.320 --> 00:00:34.840
<v Speaker 2>tech might go, you have to understand how it uses

13
00:00:34.880 --> 00:00:39.679
<v Speaker 2>cryptography and interestingly, game theory to basically build a system

14
00:00:40.200 --> 00:00:43.759
<v Speaker 2>that doesn't need you to trust any person or institution. Okay,

15
00:00:43.920 --> 00:00:46.079
<v Speaker 2>we're digging into the engineering choices that make it all

16
00:00:46.119 --> 00:00:46.679
<v Speaker 2>hang together.

17
00:00:46.799 --> 00:00:49.119
<v Speaker 1>Right, So let's start at square one. Why even invent this?

18
00:00:49.240 --> 00:00:51.640
<v Speaker 1>I mean it comes down to something ancient, right, centralized

19
00:00:51.640 --> 00:00:56.200
<v Speaker 1>trust For well forever, any real transaction land stocks just

20
00:00:56.200 --> 00:00:59.159
<v Speaker 1>sending him money. You needed someone in the middle a bank,

21
00:00:59.560 --> 00:01:03.359
<v Speaker 1>a government in office, an escro agent, someone to sign

22
00:01:03.439 --> 00:01:03.880
<v Speaker 1>off on it.

23
00:01:04.159 --> 00:01:09.680
<v Speaker 2>And those middlemen they create bottlenecks. Your sources really hammered

24
00:01:09.719 --> 00:01:12.799
<v Speaker 2>this home. The pain points are obvious. International payments can

25
00:01:12.799 --> 00:01:16.079
<v Speaker 2>take days. Cost of fortune. Yeah, and you never have

26
00:01:16.159 --> 00:01:20.560
<v Speaker 2>absolute finality. A central party can always step in reverse things,

27
00:01:21.000 --> 00:01:23.640
<v Speaker 2>maybe due to regulations or even pressure.

28
00:01:24.079 --> 00:01:27.239
<v Speaker 1>And Satoshi Nakamoto, whoever they are, they saw this right.

29
00:01:27.680 --> 00:01:31.400
<v Speaker 1>That the Internet changed how we share information, but money

30
00:01:31.519 --> 00:01:34.040
<v Speaker 1>it was still basically stuck in the eighties model of

31
00:01:34.079 --> 00:01:36.079
<v Speaker 1>digital entries in a bank's private book.

32
00:01:36.000 --> 00:01:38.120
<v Speaker 2>Still needed that trusted third party. You still had to.

33
00:01:38.079 --> 00:01:40.120
<v Speaker 1>Trust the bank wouldn't mess with the number exactly.

34
00:01:40.439 --> 00:01:43.319
<v Speaker 2>And that's the gap blockchain aims to fill. It's often called,

35
00:01:43.400 --> 00:01:46.159
<v Speaker 2>you know, the missing piece of the Internet puzzle. It's

36
00:01:46.159 --> 00:01:47.799
<v Speaker 2>set up as a peer to peer way to keep

37
00:01:47.840 --> 00:01:51.920
<v Speaker 2>records specifically for transacting value. And that value doesn't just

38
00:01:51.959 --> 00:01:54.480
<v Speaker 2>mean money, it could be anything. But it does it

39
00:01:54.480 --> 00:02:00.280
<v Speaker 2>in a way that's cryptographically secure, verifiable, without needing that intermediary.

40
00:02:00.120 --> 00:02:04.959
<v Speaker 1>Swapping out trusted people for guaranteed rules basically rules baked

41
00:02:04.959 --> 00:02:06.280
<v Speaker 1>into the system itself.

42
00:02:06.079 --> 00:02:08.800
<v Speaker 2>Precisely, rules enforced by math and incentives.

43
00:02:09.000 --> 00:02:11.560
<v Speaker 1>Okay, so if you take out the middleman, you need

44
00:02:11.599 --> 00:02:15.759
<v Speaker 1>something incredibly strong to make sure everyone plays fair. Your

45
00:02:15.840 --> 00:02:19.400
<v Speaker 1>sources point to two main pillars holding this whole trustless

46
00:02:19.439 --> 00:02:22.639
<v Speaker 1>idea up. Let's start with the one ensuring the data

47
00:02:22.680 --> 00:02:25.159
<v Speaker 1>itself isn't messed with cryptography.

48
00:02:25.400 --> 00:02:28.240
<v Speaker 2>Right. The first piece of that guarantee is the hash function.

49
00:02:28.639 --> 00:02:30.280
<v Speaker 2>You could sort of picture it like a one way

50
00:02:30.280 --> 00:02:32.960
<v Speaker 2>blender for data. Can put in any amount of data,

51
00:02:32.960 --> 00:02:36.719
<v Speaker 2>transaction details, that document, whatever, and it turns out a unique,

52
00:02:37.039 --> 00:02:40.759
<v Speaker 2>fixed length string of characters. Yea, like a digital fingerprint.

53
00:02:41.120 --> 00:02:44.479
<v Speaker 2>Bitcoin uses SAHA two fifty six, which gives you a

54
00:02:44.479 --> 00:02:46.159
<v Speaker 2>two hundred and fifty six bit output.

55
00:02:46.319 --> 00:02:48.560
<v Speaker 1>And the one way part is critical. Why can't you

56
00:02:48.639 --> 00:02:51.280
<v Speaker 1>just reverse engineer the input from that fingerprint?

57
00:02:51.360 --> 00:02:54.000
<v Speaker 2>That's the magic property. It's called pre image resistance or

58
00:02:54.039 --> 00:02:57.080
<v Speaker 2>sometimes just hiding. Hiding, Okay, it means given that output hash,

59
00:02:57.280 --> 00:03:01.400
<v Speaker 2>trying to find the original input data. It's computationally infeasible. Yeah,

60
00:03:01.439 --> 00:03:04.400
<v Speaker 2>it would take current computers thousands, maybe millions of years.

61
00:03:04.680 --> 00:03:08.159
<v Speaker 2>And this lets you do something called commitment. You can

62
00:03:08.280 --> 00:03:12.199
<v Speaker 2>prove you know something or decided something just by sharing

63
00:03:12.240 --> 00:03:15.439
<v Speaker 2>its hash publicly. You commit to it without revealing the

64
00:03:15.479 --> 00:03:19.039
<v Speaker 2>actual data until later, like sealing a bid in an auction.

65
00:03:19.240 --> 00:03:21.879
<v Speaker 1>Got it? So, hashing locks down the data. But what

66
00:03:22.000 --> 00:03:24.360
<v Speaker 1>about proving who sent it, who owns what.

67
00:03:24.759 --> 00:03:28.400
<v Speaker 2>That's where the second crypto tool comes in, asymmetric key cryptography.

68
00:03:28.759 --> 00:03:30.319
<v Speaker 2>You know public and private.

69
00:03:30.080 --> 00:03:33.840
<v Speaker 1>Keys, right, Your private key is the secret sauce exactly.

70
00:03:33.960 --> 00:03:36.639
<v Speaker 2>You keep it absolutely secret. You use it to create

71
00:03:36.719 --> 00:03:42.439
<v Speaker 2>a unique mathematical signature on a specifical transaction, a digital signature.

72
00:03:42.000 --> 00:03:43.919
<v Speaker 1>And the public key that's the one you share.

73
00:03:43.840 --> 00:03:47.120
<v Speaker 2>Yep, broadcast it far and wide. Anyone can use your

74
00:03:47.120 --> 00:03:50.439
<v Speaker 2>public key to check that a specific digital signature could

75
00:03:50.479 --> 00:03:53.240
<v Speaker 2>only have been created by your corresponding private key.

76
00:03:53.319 --> 00:03:55.479
<v Speaker 1>Ah, so that gives you authentication proof it was.

77
00:03:55.439 --> 00:03:58.719
<v Speaker 2>You and non repudiation. You can't later claim you didn't

78
00:03:58.719 --> 00:04:01.360
<v Speaker 2>authorize that transaction. The math proves you did.

79
00:04:01.520 --> 00:04:05.560
<v Speaker 1>Okay, that seems solid data integrity via hashing owner identity

80
00:04:05.639 --> 00:04:09.439
<v Speaker 1>via signatures. But this all assumes people want to play

81
00:04:09.439 --> 00:04:12.879
<v Speaker 1>by the rules. What stops a bunch of users nodes

82
00:04:12.919 --> 00:04:16.040
<v Speaker 1>from getting together and just agreeing to write fake history

83
00:04:16.279 --> 00:04:17.399
<v Speaker 1>to benefit themselves.

84
00:04:17.560 --> 00:04:20.879
<v Speaker 2>Ah Yeah, that's the other side of the coin, and

85
00:04:20.920 --> 00:04:24.199
<v Speaker 2>that brings us to the second pillar game theory. This

86
00:04:24.399 --> 00:04:27.160
<v Speaker 2>is what ensures the system as a whole remains stable

87
00:04:27.600 --> 00:04:28.879
<v Speaker 2>even with bad actors.

88
00:04:29.319 --> 00:04:32.920
<v Speaker 1>This addresses that old computer science problem, right, the Byzantine

89
00:04:32.920 --> 00:04:33.639
<v Speaker 1>general's problem.

90
00:04:33.720 --> 00:04:35.240
<v Speaker 2>That's the one. How do you get a bunch of

91
00:04:35.319 --> 00:04:38.680
<v Speaker 2>distributed actors who can't fully trust each other to agree

92
00:04:38.680 --> 00:04:41.399
<v Speaker 2>on a single truth when some might be actively trying

93
00:04:41.399 --> 00:04:42.879
<v Speaker 2>to sabotage the consensus.

94
00:04:43.040 --> 00:04:45.639
<v Speaker 1>So the system's design almost assumes people are going to

95
00:04:45.639 --> 00:04:48.680
<v Speaker 1>be self interested, maybe even greedy. It's not built on

96
00:04:48.800 --> 00:04:50.319
<v Speaker 1>hoping everyone's nice, that's.

97
00:04:50.240 --> 00:04:53.120
<v Speaker 2>A great way to put it. It essentially ignores morality.

98
00:04:53.639 --> 00:04:58.120
<v Speaker 2>It relies on pure rational self interest. Game theory helps

99
00:04:58.160 --> 00:05:01.360
<v Speaker 2>design the system so it reaches what's called a Nash equilibrium,

100
00:05:01.560 --> 00:05:05.519
<v Speaker 2>meaning that each individual participant, acting solely to maximize their

101
00:05:05.519 --> 00:05:09.000
<v Speaker 2>own financial gain, finds that their best strategy is actually

102
00:05:09.000 --> 00:05:12.519
<v Speaker 2>to follow the rules and maintain the system's integrity.

103
00:05:12.279 --> 00:05:16.480
<v Speaker 1>Because trying to cheat is just too expensive or not

104
00:05:16.560 --> 00:05:17.120
<v Speaker 1>worth the risk.

105
00:05:17.399 --> 00:05:21.920
<v Speaker 2>Exactly the cost of successfully attacking the network, say, trying

106
00:05:21.920 --> 00:05:26.199
<v Speaker 2>to double spend coins by rewriting history, involves immense computational

107
00:05:26.279 --> 00:05:30.120
<v Speaker 2>power and therefore cost. This cost has to be designed

108
00:05:30.160 --> 00:05:32.680
<v Speaker 2>to always outweigh the potential profit from.

109
00:05:32.560 --> 00:05:35.240
<v Speaker 1>The fraud, and that's where things like mining rewards and

110
00:05:35.279 --> 00:05:37.639
<v Speaker 1>transaction fees come in. They're not just payments.

111
00:05:37.720 --> 00:05:41.759
<v Speaker 2>They're the incentives. They're the carefully calculated game theory mechanics

112
00:05:42.040 --> 00:05:45.199
<v Speaker 2>that make honesty the most profitable strategy for the majority

113
00:05:45.199 --> 00:05:45.920
<v Speaker 2>of participants.

114
00:05:45.920 --> 00:05:48.079
<v Speaker 1>Okay, I think I'm getting the framework. We've got crypto

115
00:05:48.120 --> 00:05:51.160
<v Speaker 1>locking down the data and identity and game theory making

116
00:05:51.160 --> 00:05:54.040
<v Speaker 1>sure people are incentivized to keep the system honest. So

117
00:05:54.079 --> 00:05:57.959
<v Speaker 1>how does this translate into the actual chain, solid record?

118
00:05:58.360 --> 00:06:00.879
<v Speaker 2>Right? This is where the structure comes in. The core

119
00:06:01.000 --> 00:06:04.839
<v Speaker 2>idea is immutability by design, It's built into the data

120
00:06:04.879 --> 00:06:10.360
<v Speaker 2>structure itself. Blocks which are just bundles of validated transaction. Yeah,

121
00:06:10.439 --> 00:06:13.800
<v Speaker 2>they're linked together one after the other sequentchly. But the

122
00:06:13.879 --> 00:06:17.879
<v Speaker 2>link isn't just saying block five follows block four. The

123
00:06:17.959 --> 00:06:21.879
<v Speaker 2>link is a cryptographic hash pointer, meaning meaning the header

124
00:06:21.959 --> 00:06:26.120
<v Speaker 2>data of block five contains the unique cryptographic hash, that

125
00:06:26.199 --> 00:06:28.800
<v Speaker 2>fingerprint we talked about of the entirety of block four.

126
00:06:29.000 --> 00:06:31.639
<v Speaker 1>Ah. Okay, So that's the enforcement.

127
00:06:31.279 --> 00:06:35.759
<v Speaker 2>That's the linchpin. It creates this incredibly strong self enforcing chain.

128
00:06:36.600 --> 00:06:38.839
<v Speaker 2>If you go back and try to tamper with anything

129
00:06:38.879 --> 00:06:42.120
<v Speaker 2>in block four, change the transaction amount by even a

130
00:06:42.160 --> 00:06:42.920
<v Speaker 2>tiny bit.

131
00:06:42.839 --> 00:06:45.319
<v Speaker 1>The hash of block four changes instantly.

132
00:06:45.000 --> 00:06:48.240
<v Speaker 2>Completely changes which means the pointer store in block five

133
00:06:48.319 --> 00:06:50.759
<v Speaker 2>is now wrong. It points to a version of block

134
00:06:50.800 --> 00:06:52.600
<v Speaker 2>four that doesn't exist anymore.

135
00:06:52.240 --> 00:06:54.160
<v Speaker 1>And that breaks the link, breaks the link to block.

136
00:06:54.000 --> 00:06:56.160
<v Speaker 2>Five, which breaks the link to block six, and so on. Yeah,

137
00:06:56.240 --> 00:06:59.160
<v Speaker 2>the entire chain becomes mathematically invalid from that point of

138
00:06:59.199 --> 00:07:02.199
<v Speaker 2>tampering forward. Anyone checking can see it immediately.

139
00:07:02.319 --> 00:07:05.680
<v Speaker 1>Hold on, that's actually pretty elegant. The integrity isn't down

140
00:07:05.680 --> 00:07:08.519
<v Speaker 1>to a security guard or a firewall policy. It's just

141
00:07:08.560 --> 00:07:12.680
<v Speaker 1>a mathematical consequence. Tampering breaks the math, so the system

142
00:07:12.720 --> 00:07:14.000
<v Speaker 1>inherently rejects.

143
00:07:13.639 --> 00:07:18.319
<v Speaker 2>It exactly right now. That leads to another challenge. Efficiency.

144
00:07:18.560 --> 00:07:22.360
<v Speaker 2>If this chain gets massive and every node validating the

145
00:07:22.399 --> 00:07:26.360
<v Speaker 2>network needs to confirm nothing's been tampered with? Yeah, how

146
00:07:26.360 --> 00:07:29.199
<v Speaker 2>does that even work? Does it grind to a halt? Yeah?

147
00:07:29.240 --> 00:07:32.160
<v Speaker 1>That sounds like a huge overhead. Does every single person

148
00:07:32.240 --> 00:07:35.800
<v Speaker 1>running a node have to download and check gigabytes terabytes

149
00:07:35.839 --> 00:07:38.560
<v Speaker 1>of history just to verify one recent transaction?

150
00:07:38.839 --> 00:07:42.000
<v Speaker 2>Thankfully, No, that would never scale. The solution here is

151
00:07:42.120 --> 00:07:46.319
<v Speaker 2>another clever data structure used inside each block. The Merkele tree.

152
00:07:46.439 --> 00:07:47.800
<v Speaker 1>Okay, Merkle tree, what's that too?

153
00:07:47.959 --> 00:07:49.920
<v Speaker 2>Think of it as a tree structure but built out

154
00:07:49.920 --> 00:07:52.680
<v Speaker 2>of hashes. At the bottom of the leaves are the

155
00:07:52.680 --> 00:07:56.639
<v Speaker 2>hashes of individual transactions within the block. Then pairs of

156
00:07:56.680 --> 00:07:59.000
<v Speaker 2>these hashes are hashed together to form the layer above,

157
00:07:59.480 --> 00:08:01.519
<v Speaker 2>and pairs of those hashes are hash and so on

158
00:08:01.879 --> 00:08:03.480
<v Speaker 2>all the way up to a single root hash at

159
00:08:03.519 --> 00:08:03.839
<v Speaker 2>the top.

160
00:08:04.079 --> 00:08:07.639
<v Speaker 1>And that single root hash somehow summarizes all the transactions in.

161
00:08:07.600 --> 00:08:12.240
<v Speaker 2>The block precisely. It's a compact cryptographic summary of everything inside.

162
00:08:12.399 --> 00:08:14.839
<v Speaker 1>So how does that help Joe user check their payment?

163
00:08:15.279 --> 00:08:18.079
<v Speaker 2>It enables us called proof without possession. You don't need

164
00:08:18.079 --> 00:08:21.800
<v Speaker 2>the whole blocks transaction list to prove your transaction is included.

165
00:08:21.839 --> 00:08:25.079
<v Speaker 2>You only need your transaction hash and the few sibling

166
00:08:25.160 --> 00:08:27.759
<v Speaker 2>hashes along the specific path up the tree to the

167
00:08:27.800 --> 00:08:31.000
<v Speaker 2>root hash. Uh like a shortcut, a very efficient shortcut.

168
00:08:31.160 --> 00:08:34.799
<v Speaker 2>Instead of checking say, oh, thousand transactions one by one,

169
00:08:34.960 --> 00:08:38.360
<v Speaker 2>which is linear time, you only need maybe ten hashes,

170
00:08:38.559 --> 00:08:40.840
<v Speaker 2>which is loggerithmic time left lit make time.

171
00:08:40.879 --> 00:08:41.120
<v Speaker 1>Okay.

172
00:08:41.159 --> 00:08:44.200
<v Speaker 3>Put that in perspective, it means checking gets only slightly

173
00:08:44.279 --> 00:08:48.440
<v Speaker 3>harder even if the number of transactions grows enormously, Like

174
00:08:48.519 --> 00:08:51.159
<v Speaker 3>checking one transaction in a block with a million entries,

175
00:08:51.399 --> 00:08:52.240
<v Speaker 3>isn't that much harder?

176
00:08:52.240 --> 00:08:54.240
<v Speaker 2>Than checking one in a block with a thousand, It

177
00:08:54.360 --> 00:08:56.799
<v Speaker 2>scales incredibly well compared to checking every single one.

178
00:08:56.840 --> 00:09:01.679
<v Speaker 1>Okay, foundations laid crypto game fear, the hash linked chain,

179
00:09:02.360 --> 00:09:05.919
<v Speaker 1>Merkele trees for efficiency. Now let's compare the big players.

180
00:09:06.559 --> 00:09:10.279
<v Speaker 1>Bitcoin and Ethereum dominate, but your sources stress they are

181
00:09:10.320 --> 00:09:13.840
<v Speaker 1>fundamentally different beasts, built for different things.

182
00:09:13.799 --> 00:09:17.879
<v Speaker 2>Totally different goals, leading to different architectures. Bitcoin was conceived

183
00:09:17.919 --> 00:09:21.679
<v Speaker 2>purely as electronic cash, a peer to peer value transfer system,

184
00:09:22.200 --> 00:09:25.000
<v Speaker 2>and its core design reflects that it uses the UTXO

185
00:09:25.120 --> 00:09:29.000
<v Speaker 2>model that stands for unstent transaction output. This is probably

186
00:09:29.000 --> 00:09:33.000
<v Speaker 2>the biggest conceptual hurdle for people used to traditional banking.

187
00:09:33.159 --> 00:09:35.559
<v Speaker 1>Because bitcoin doesn't really have accounts or balances and the

188
00:09:35.600 --> 00:09:36.639
<v Speaker 1>way we normally.

189
00:09:36.279 --> 00:09:39.000
<v Speaker 2>Think of them exactly. There's no central ledger saying Alice

190
00:09:39.000 --> 00:09:42.440
<v Speaker 2>has five btc. Instead, the state of the Bitcoin network

191
00:09:42.480 --> 00:09:46.240
<v Speaker 2>is simply the entire collection of all existing utxos. These

192
00:09:46.240 --> 00:09:49.080
<v Speaker 2>are like specific IOUs or digital coins that haven't been

193
00:09:49.080 --> 00:09:49.639
<v Speaker 2>spent yet.

194
00:09:49.720 --> 00:09:51.879
<v Speaker 1>So if I have five btc, what I actually have

195
00:09:52.240 --> 00:09:56.320
<v Speaker 1>is maybe one utxo worth two btc from a previous transaction,

196
00:09:56.440 --> 00:09:59.519
<v Speaker 1>and another one worth three btc from a different one.

197
00:09:59.360 --> 00:10:02.159
<v Speaker 2>Precisely, and when you want to spend say four BTC,

198
00:10:02.559 --> 00:10:05.799
<v Speaker 2>your transaction doesn't just debit an account. It consumes your

199
00:10:05.840 --> 00:10:10.000
<v Speaker 2>two btc utxo and your three btc utxo.

200
00:10:09.200 --> 00:10:10.480
<v Speaker 3>Completely destroys them.

201
00:10:10.559 --> 00:10:13.720
<v Speaker 2>Functionally, yes, and it creates new utxos, one worth four

202
00:10:13.759 --> 00:10:15.080
<v Speaker 2>BTC for the recipient and.

203
00:10:15.039 --> 00:10:16.799
<v Speaker 1>One worth one BTC that comes back to.

204
00:10:16.759 --> 00:10:20.080
<v Speaker 2>Me has change exactly. It's very much like using physical cash.

205
00:10:20.399 --> 00:10:22.919
<v Speaker 2>You hand over specific bills, you get different bills back

206
00:10:22.960 --> 00:10:26.679
<v Speaker 2>as change. Every node knows all the available utxos, so

207
00:10:26.720 --> 00:10:28.600
<v Speaker 2>they know you can't spend the same bill twice.

208
00:10:28.919 --> 00:10:32.639
<v Speaker 1>Makes sense, simple, focused on the transfer, and Bitcoin's scripting

209
00:10:32.720 --> 00:10:34.080
<v Speaker 1>language reflects that too, doesn't it.

210
00:10:34.080 --> 00:10:39.320
<v Speaker 2>It's quite limited deliberately, so it's simple. It's not turing complete,

211
00:10:39.360 --> 00:10:44.080
<v Speaker 2>meaning it can't perform arbitrary complex calculations. This enhances security

212
00:10:44.360 --> 00:10:47.960
<v Speaker 2>and keeps the focus squarely on its core function, moving value.

213
00:10:48.080 --> 00:10:50.960
<v Speaker 1>Okay, then there's a theeum designed from the ground up

214
00:10:51.000 --> 00:10:54.159
<v Speaker 1>to be something much broader, a general purpose platform, a

215
00:10:54.200 --> 00:10:55.039
<v Speaker 1>world computer.

216
00:10:55.320 --> 00:10:57.960
<v Speaker 2>Right, and it takes a completely different approach the account

217
00:10:58.000 --> 00:11:01.039
<v Speaker 2>balance model, much more like traditional computing or banking.

218
00:11:01.360 --> 00:11:04.720
<v Speaker 1>So Ethereum does track accounts with balances.

219
00:11:04.840 --> 00:11:07.799
<v Speaker 2>Yes, the state of Ethereum is the state of all accounts.

220
00:11:08.000 --> 00:11:11.039
<v Speaker 2>Each account has a balance in ether, a nonce like

221
00:11:11.080 --> 00:11:15.960
<v Speaker 2>a transaction counter. Importantly, potentially code and storage associated with it, and.

222
00:11:16.000 --> 00:11:18.440
<v Speaker 1>Your source is mentioned. Two types of accounts, they're the

223
00:11:18.440 --> 00:11:19.840
<v Speaker 1>ones people control.

224
00:11:19.639 --> 00:11:24.080
<v Speaker 2>Externally owned accounts eoas. These are controlled by private keys,

225
00:11:24.480 --> 00:11:27.519
<v Speaker 2>just like Bitcoin addresses. You use your private key to

226
00:11:27.600 --> 00:11:29.480
<v Speaker 2>sign transactions sent from your EOA.

227
00:11:29.879 --> 00:11:32.159
<v Speaker 1>And then there are contract accounts, and this seems to

228
00:11:32.200 --> 00:11:33.440
<v Speaker 1>be the huge innovation.

229
00:11:33.799 --> 00:11:37.919
<v Speaker 2>Absolutely, contract accounts don't have private keys. They're controlled only

230
00:11:37.960 --> 00:11:40.480
<v Speaker 2>by their internal code. That code is what we call

231
00:11:40.519 --> 00:11:44.559
<v Speaker 2>a smart contract. Okay, these smart contracts can be incredibly complex.

232
00:11:44.639 --> 00:11:48.960
<v Speaker 2>Because Ethereum's virtual machine is turing complete. They can execute

233
00:11:48.960 --> 00:11:53.600
<v Speaker 2>basically any logic you can program, run decentralized organizations, manage

234
00:11:53.639 --> 00:11:57.799
<v Speaker 2>financial instruments, power games, anything. They activate when an EOA

235
00:11:58.240 --> 00:12:00.480
<v Speaker 2>or another contract send them a transaction.

236
00:12:00.600 --> 00:12:04.080
<v Speaker 1>But wait, turn complete. That sounds risky on a shared network.

237
00:12:04.399 --> 00:12:06.720
<v Speaker 1>If you can run any code, what stops someone writing

238
00:12:06.759 --> 00:12:09.639
<v Speaker 1>code that just runs forever in an infinite loop, or

239
00:12:09.679 --> 00:12:13.159
<v Speaker 1>does some crazy complex calculation that bogs down every node

240
00:12:13.200 --> 00:12:16.279
<v Speaker 1>trying to validate it. A denial of service attack baked

241
00:12:16.320 --> 00:12:17.240
<v Speaker 1>into a transaction.

242
00:12:17.480 --> 00:12:20.960
<v Speaker 2>That is the fundamental danger, and ethereum solution brings us

243
00:12:21.039 --> 00:12:22.320
<v Speaker 2>right back to game theory again.

244
00:12:22.960 --> 00:12:25.919
<v Speaker 1>The gas system, Ah, gas, I've heard about this. It's

245
00:12:25.960 --> 00:12:28.159
<v Speaker 1>like fuel for transactions exactly.

246
00:12:28.200 --> 00:12:32.440
<v Speaker 2>It's an economic mechanism to prevent computational abuse. Every single

247
00:12:32.480 --> 00:12:37.159
<v Speaker 2>elementary operation a contract performs adding two numbers, storing data

248
00:12:37.600 --> 00:12:40.759
<v Speaker 2>using network bandwidth has a pre defined cost and gas.

249
00:12:41.120 --> 00:12:43.519
<v Speaker 1>So complex operations cost more.

250
00:12:43.399 --> 00:12:46.879
<v Speaker 2>Gas much more, and storage is particularly expensive. When you

251
00:12:46.919 --> 00:12:49.720
<v Speaker 2>send a transaction. To interact with a contract, you have

252
00:12:49.759 --> 00:12:53.360
<v Speaker 2>to specify two things. A gas limit the maximum gas

253
00:12:53.360 --> 00:12:56.320
<v Speaker 2>you're willing to burn, and a gas price how ether

254
00:12:56.440 --> 00:12:58.159
<v Speaker 2>you'll pay per unit of gas.

255
00:12:58.320 --> 00:13:01.480
<v Speaker 1>So the total fee is gas u used times gas price.

256
00:13:01.600 --> 00:13:05.240
<v Speaker 2>Correct miners prioritize transactions offering a higher gas price.

257
00:13:05.519 --> 00:13:08.039
<v Speaker 1>And the crucial part, what if the calculation is too

258
00:13:08.039 --> 00:13:10.200
<v Speaker 1>complex and hits the gas on that you set, Then.

259
00:13:10.120 --> 00:13:14.200
<v Speaker 2>The transaction execution halts immediately. Any changes it tried to

260
00:13:14.200 --> 00:13:16.759
<v Speaker 2>make to the blockchain state are instantly reverted as if

261
00:13:16.759 --> 00:13:17.440
<v Speaker 2>it never happens.

262
00:13:17.480 --> 00:13:18.000
<v Speaker 1>It fails.

263
00:13:18.120 --> 00:13:21.159
<v Speaker 2>But and this is key, you the sender still pay

264
00:13:21.200 --> 00:13:23.399
<v Speaker 2>the fee for all the gas that was consumed up

265
00:13:23.440 --> 00:13:24.480
<v Speaker 2>until the point it ran out.

266
00:13:24.639 --> 00:13:27.840
<v Speaker 1>Ah, so there's a direct financial penalty for sending an

267
00:13:27.840 --> 00:13:30.480
<v Speaker 1>efficient code or trying to spam the network. You lose

268
00:13:30.519 --> 00:13:32.399
<v Speaker 1>your fee even if the transaction fails.

269
00:13:32.720 --> 00:13:37.799
<v Speaker 2>Precisely, it forces users to be rational about the computational

270
00:13:37.840 --> 00:13:41.639
<v Speaker 2>resources they ask the network to expend. It aligns incentives

271
00:13:41.720 --> 00:13:45.759
<v Speaker 2>beautifully protecting the network while rewarding miners for the actual

272
00:13:45.840 --> 00:13:49.559
<v Speaker 2>work done. It's a really clever piece of economic.

273
00:13:49.159 --> 00:13:51.960
<v Speaker 1>Engineering that really clarifies things. Okay, so wrapping up, we've

274
00:13:51.960 --> 00:13:56.159
<v Speaker 1>got the cryptographic building blocks, hashing, and digital signatures providing

275
00:13:56.200 --> 00:14:00.240
<v Speaker 1>that core integrity. We have game theory providing stability through

276
00:14:00.240 --> 00:14:05.639
<v Speaker 1>economic incentives, whether it's Bitcoin's mining rewards or Ethereum's gas system.

277
00:14:05.679 --> 00:14:09.919
<v Speaker 1>And we see these reflected and fundamentally different architectures. Bitcoin's

278
00:14:10.039 --> 00:14:14.879
<v Speaker 1>UTXO model like digital cash, simple and focused versus Ethereum's

279
00:14:14.879 --> 00:14:19.200
<v Speaker 1>account based model enabling complex turing, complete smart contracts. This

280
00:14:19.320 --> 00:14:20.919
<v Speaker 1>world computer idea.

281
00:14:20.679 --> 00:14:23.279
<v Speaker 2>Big difference in philosophy and capability, but we've also touched

282
00:14:23.279 --> 00:14:24.159
<v Speaker 2>on this core tension.

283
00:14:24.720 --> 00:14:27.960
<v Speaker 1>For these systems to be truly decentralized and trustless, every

284
00:14:28.000 --> 00:14:31.200
<v Speaker 1>node generally has to process and validate everything Yeah.

285
00:14:31.000 --> 00:14:33.480
<v Speaker 2>That redundancy is the source of security, but it's also

286
00:14:33.519 --> 00:14:34.840
<v Speaker 2>the bottleneck.

287
00:14:34.480 --> 00:14:38.440
<v Speaker 1>Which leads directly to arguably the biggest hurdle for this technology,

288
00:14:38.480 --> 00:14:39.559
<v Speaker 1>right scalability.

289
00:14:39.879 --> 00:14:43.279
<v Speaker 2>Absolutely. If every node does all the work, the whole

290
00:14:43.320 --> 00:14:46.879
<v Speaker 2>network can only process as many transactions as its slowest

291
00:14:46.879 --> 00:14:50.600
<v Speaker 2>participants can handle. As you add more users and transactions,

292
00:14:50.879 --> 00:14:54.679
<v Speaker 2>things slow down. That trade off between decentralization and raw

293
00:14:54.759 --> 00:14:58.840
<v Speaker 2>speed is still the major challenge for widespread global adoption.

294
00:14:59.039 --> 00:15:02.039
<v Speaker 2>Yeah it works, yes, but can it work at visa scale?

295
00:15:02.399 --> 00:15:02.919
<v Speaker 2>Not yet?

296
00:15:03.240 --> 00:15:04.919
<v Speaker 1>So for you listening, If you want to dig deeper,

297
00:15:05.000 --> 00:15:07.519
<v Speaker 1>the logical next step is to look into how developers

298
00:15:07.559 --> 00:15:10.879
<v Speaker 1>are trying to break that bottleneck. Your source materials mentioned

299
00:15:10.879 --> 00:15:12.000
<v Speaker 1>a couple of main approaches.

300
00:15:12.200 --> 00:15:15.600
<v Speaker 2>Yeah, two big ones to explore. First is sharding. This

301
00:15:15.720 --> 00:15:19.480
<v Speaker 2>tries to partition the blockchain's data and processing load, so

302
00:15:19.519 --> 00:15:22.360
<v Speaker 2>instead of every node checking everything, nodes might only be

303
00:15:22.399 --> 00:15:25.879
<v Speaker 2>responsible for a specific chard or section of the network. Activity.

304
00:15:26.000 --> 00:15:26.879
<v Speaker 2>Makes it much more.

305
00:15:26.720 --> 00:15:29.120
<v Speaker 1>Parallel, okay, divide and conquer kind of.

306
00:15:29.200 --> 00:15:34.000
<v Speaker 2>Yeah. The second major area is off chain computation. Think

307
00:15:34.080 --> 00:15:38.240
<v Speaker 2>systems like Bitcoin's Lightning network or Ethereum's Layer two. Solutions

308
00:15:38.279 --> 00:15:41.200
<v Speaker 2>like raiden or roll ups. The idea here is to

309
00:15:41.200 --> 00:15:45.039
<v Speaker 2>handle lots of small, frequent transactions off the main blockchain

310
00:15:45.679 --> 00:15:49.080
<v Speaker 2>in separate channels rivately yeah between the participants, but still

311
00:15:49.080 --> 00:15:53.360
<v Speaker 2>cryptographically secured. Then only the final net settlement or a

312
00:15:53.639 --> 00:15:56.879
<v Speaker 2>summary of those off chain transactions gets recorded back onto

313
00:15:56.919 --> 00:16:00.279
<v Speaker 2>the main slower, more expensive blockchain. It could potentially allow

314
00:16:00.320 --> 00:16:01.799
<v Speaker 2>millions of transactions per second.

315
00:16:01.919 --> 00:16:05.799
<v Speaker 1>Fascinating, so moving computation off the main stage to speed

316
00:16:05.840 --> 00:16:08.480
<v Speaker 1>things up definitely areas worth exploring further.

317
00:16:09.000 --> 00:16:10.799
<v Speaker 2>For sure, the scaling race is where a lot of

318
00:16:10.799 --> 00:16:11.960
<v Speaker 2>the innovations is happening now.

319
00:16:12.039 --> 00:16:14.480
<v Speaker 1>Well, thank you. This was incredibly helpful in mapping out

320
00:16:14.519 --> 00:16:18.360
<v Speaker 1>the actual mechanics underneath it all. Understanding that fundamental architecture

321
00:16:18.360 --> 00:16:22.279
<v Speaker 1>really is key to evaluating where this technology might realistically

322
00:16:22.320 --> 00:16:23.320
<v Speaker 1>go my pleasure.

323
00:16:23.679 --> 00:16:27.080
<v Speaker 2>It's complex, but getting the basics right is crucial