1
00:00:00,000 --> 00:00:08,000
All right, guys, I'm glad to introduce Paul Sztorc. Is that the proper way to
pronounce it? Paul Sztorc?

2
00:00:08,000 --> 00:00:10,000
No, but I'm used to it.

3
00:00:10,000 --> 00:00:13,000
So let's hear how the proper way to pronounce it.

4
00:00:13,000 --> 00:00:19,000
It's Sztorc. In Polish, C is a different letter. It's pronounced T-S. Paul Sztorc.

5
00:00:19,000 --> 00:00:33,000
Okay, yeah. Okay, so Paul Sztorc is joining us. He's been active in the Bitcoin
project since late 2010, 2011-ish.

6
00:00:33,000 --> 00:00:40,000
He's helped produce great technology for Bitcoin, such as the Truthcoin protocol.

7
00:00:40,000 --> 00:00:49,000
Truthcoin is not an altcoin. It's a prediction market protocol, which has been then
called Bitcoin Hivemind as the implementation.

8
00:00:49,000 --> 00:00:57,000
And his latest project and his current focus is on Drivechains. Drivechains is not
actually new.

9
00:00:57,000 --> 00:01:04,000
It's something he proposed a couple of years ago, and it's sort of this version of
sidechains,

10
00:01:04,000 --> 00:01:10,000
which represents some of the earlier efforts of what sidechains were meant to be.

11
00:01:10,000 --> 00:01:18,000
And now they've got some testnet Drivechain things going on, and so there's a lot of
interesting development going on there.

12
00:01:18,000 --> 00:01:24,000
So welcome, Paul. Thanks for joining us to discuss your work.

13
00:01:24,000 --> 00:01:26,000
Thanks for having me.

14
00:01:26,000 --> 00:01:32,000
Okay, so a lot of people don't necessarily know who you are, so let's provide a bit
of background to them.

15
00:01:32,000 --> 00:01:42,000
So tell us then when you got into Bitcoin and why, and what motivates you to stay on
the project in 2018?

16
00:01:42,000 --> 00:01:50,000
Well, really, it's funny because I have a kind of a vulgar origin, because it's
the...

17
00:01:50,000 --> 00:01:54,000
I'm sorry, how do I shut off this user-joined-your-channel thing here?

18
00:01:54,000 --> 00:02:00,000
Go to the sound pack in self. If you go to the self menu, sound pack, and turn off.

19
00:02:00,000 --> 00:02:05,000
So either sound's deactivated, and that should make it so that...

20
00:02:05,000 --> 00:02:08,000
Oh, okay, there we go. Sound's deactivated.

21
00:02:08,000 --> 00:02:11,000
Yeah.

22
00:02:11,000 --> 00:02:14,000
Oh, someone says I have to restart TeamSpeak. Is that true?

23
00:02:14,000 --> 00:02:19,000
If it pops up again, I guess maybe you can do that, but let's see if it works
otherwise.

24
00:02:19,000 --> 00:02:27,000
All right, so far so good. All right, classic whatever, AV issues.

25
00:02:27,000 --> 00:02:33,000
Okay, so we're good? We're still doing this? Are we doing it all?

26
00:02:33,000 --> 00:02:35,000
No, no, just keep going.

27
00:02:35,000 --> 00:02:38,000
All right, we're going. Great, I love it. We're doing it live.

28
00:02:38,000 --> 00:02:48,000
Yes, so the vulgar origin story is just the Wired Silk Road article, and I read this
article and I just thought...

29
00:02:49,000 --> 00:02:54,000
Well, that article is, of course, about the anonymous drug market where you can buy
anything.

30
00:02:54,000 --> 00:03:00,000
I think that was the subtitle, and the article was they are addicted to drugs or
something.

31
00:03:00,000 --> 00:03:05,000
They would figure out a way to cheat it if they could cheat it.

32
00:03:05,000 --> 00:03:09,000
And I was like, so it must be pretty solid actually. It's living in the real world.

33
00:03:09,000 --> 00:03:12,000
And then I was like, this seems really complicated to use.

34
00:03:12,000 --> 00:03:15,000
Why don't they use other things like credit cards?

35
00:03:15,000 --> 00:03:22,000
And I learned about why, because your credit card is attached to your identity,
which is something that had never even occurred to me until then.

36
00:03:22,000 --> 00:03:35,000
And so, yeah, I kind of just read this article, and I just thought, wow, that's
really interesting that commerce has kind of sprouted up despite this immense
transaction cost, this immense friction.

37
00:03:35,000 --> 00:03:45,000
And then I just did a little bit more research, and Satoshi had solved the problem
of how to distribute the initial coins, which is something that I kind of thought
would be impossible,

38
00:03:45,000 --> 00:03:52,000
like how are you going to get people to switch from their money to your money in a
gradual sense, because not everyone's going to hear about it at the same time.

39
00:03:52,000 --> 00:03:56,000
And then that would be a headache, but the mining solved that issue.

40
00:03:56,000 --> 00:04:09,000
And then I realized that it could – there was a plan for it to actually be cheaper
and at a grander scale than like all the rest of the payment systems of the world.

41
00:04:09,000 --> 00:04:13,000
And I just thought, well, this thing will work in the black markets, and it will
work in the regular markets.

42
00:04:13,000 --> 00:04:17,000
And I just thought, well, then it's only a matter – that's kind of it.

43
00:04:17,000 --> 00:04:20,000
And then the more reading you do, the more into it you get.

44
00:04:20,000 --> 00:04:22,000
I got some coins from Gavin's faucet.

45
00:04:22,000 --> 00:04:27,000
I bought some coins using BitInstant, and then you have to use the stupid –

46
00:04:29,000 --> 00:04:34,000
Thought it was a scam and everything, and then suddenly when Wikileaks accepted it,
I was like, holy shit, that's the value, right?

47
00:04:34,000 --> 00:04:40,000
And similar to Silk Road being able to – it's not that you enjoy that it's being
used for these kind of purposes necessarily.

48
00:04:40,000 --> 00:04:48,000
It's just that it's a solid proof of concept, a demonstration of that feature of
non-censurability and such.

49
00:04:48,000 --> 00:04:49,000
Exactly.

50
00:04:49,000 --> 00:04:55,000
Okay, so let's hear about – so right now Bitcoin is quite tribal, right?

51
00:04:55,000 --> 00:04:58,000
There's the big blockers. There's the core supporters and stuff.

52
00:04:58,000 --> 00:05:01,000
So give us an idea of where your head is on this.

53
00:05:01,000 --> 00:05:03,000
Are you a big blocker?

54
00:05:03,000 --> 00:05:07,000
How are you with all these politics these days of Bitcoin?

55
00:05:08,000 --> 00:05:16,000
Well, I think politics is often something you do when you aren't clever enough to
figure out how to have it both ways.

56
00:05:16,000 --> 00:05:23,000
There are tradeoffs. I'm a diehard economist and I have a lot of training and
background in economics.

57
00:05:23,000 --> 00:05:30,000
So we do live in a world that has tradeoffs, but we also live in a world that has
technological progress

58
00:05:30,000 --> 00:05:37,000
and in a world that has a growth, an unbounded growth of knowledge over time.

59
00:05:37,000 --> 00:05:44,000
So we get better and better and better at – sometimes there's a tradeoff between X
and Y,

60
00:05:44,000 --> 00:05:48,000
but then someone will invent something and the tradeoff is meaningless.

61
00:05:48,000 --> 00:05:52,000
If you have – sometimes we have to work together to solve problems,

62
00:05:52,000 --> 00:05:56,000
but someone will invent something that makes it so it doesn't matter as much.

63
00:05:56,000 --> 00:06:00,000
So like on an airplane, you can get noise-canceling headphones and a sleeping mask,

64
00:06:00,000 --> 00:06:03,000
and then what you do doesn't affect other people.

65
00:06:03,000 --> 00:06:07,000
So you can have this kind of – these tradeoffs and this interconnectedness,

66
00:06:07,000 --> 00:06:14,000
but I really subscribe to the philosophy that you should be able to find something
that works for everyone eventually.

67
00:06:14,000 --> 00:06:23,000
And in that, can you talk a bit then about how that – so for example, when Bitcoin
Cash split off,

68
00:06:23,000 --> 00:06:32,000
there was a lot of discussion in the years before about extension blocks and
different solutions of, like you say, being able to have it both ways.

69
00:06:32,000 --> 00:06:34,000
Could you talk a bit about this?

70
00:06:34,000 --> 00:06:40,000
Yes. There was – I think – I don't remember. I think it was Johnson Lau.

71
00:06:40,000 --> 00:06:51,000
I don't exactly remember who got it going in 2013, but Adam Back kind of explained
this in detail in Bitcoin Dev mailing list

72
00:06:51,000 --> 00:06:59,000
that you could have kind of a soft fork where some people got larger blocks and
other people stayed with the smaller blocks,

73
00:06:59,000 --> 00:07:05,000
and that is an example of what I'm talking about where everyone can get what they
want and there's no conflict.

74
00:07:05,000 --> 00:07:09,000
And so I thought that was great. That was a pretty good idea.

75
00:07:09,000 --> 00:07:18,000
I mean it had issues that were later discovered, and I think also there was immense
peer pressure that Dr. Back fell under,

76
00:07:18,000 --> 00:07:25,000
and I think he kind of withdrew the idea and also the philosophy of optimistic
technological progress.

77
00:07:25,000 --> 00:07:31,000
I think that was also withdrawn, that it was – from then on out, it was going to be
us versus them.

78
00:07:31,000 --> 00:07:39,000
Right, and we'll get more into that as we focus more specifically on the technical
side of sidechains,

79
00:07:39,000 --> 00:07:45,000
but let's talk a bit more then about – so you mentioned you're – I trained as an
economist,

80
00:07:45,000 --> 00:07:50,000
so a lot of guys in crypto, like it's this great intersection of computer science
and finance,

81
00:07:50,000 --> 00:07:55,000
and so you get a lot of guys who are sort of finance guys that are attracted to the
project.

82
00:07:55,000 --> 00:07:59,000
You get purely crypto guys who are just in it for the cypherpunk aspects.

83
00:07:59,000 --> 00:08:10,000
So you're actually – even though you say you're a trained economist and you're from
Yale and such doing statistical economics type work,

84
00:08:10,000 --> 00:08:16,000
you're also a C++ developer, so you're actually like – as well as designing
protocols,

85
00:08:16,000 --> 00:08:19,000
you've also even written the code to implement some of this.

86
00:08:19,000 --> 00:08:26,000
So did you self-teach C and C++, or did you sort of like – were you one of these
guys who already was great at coding,

87
00:08:26,000 --> 00:08:32,000
so you wanted to have your formal education in something non-computer science so you
wouldn't waste your time?

88
00:08:32,000 --> 00:08:39,000
Yeah, I have no formal training in programming, and I don't think I'm that good at
it.

89
00:08:39,000 --> 00:08:45,000
My favorite languages would be like Python and R. R is like a research language,

90
00:08:45,000 --> 00:08:53,000
but yeah, what I will do is I'll usually write something in one of those and then
translate it over to C++,

91
00:08:53,000 --> 00:08:58,000
and I'm not very confident in my C++ at all, but it gets done.

92
00:08:58,000 --> 00:09:05,000
Right. Okay, so in terms of Bitcoin development in general, a lot of people –

93
00:09:05,000 --> 00:09:12,000
Yeah, I should mention there's a great – Cryptex is an associate of mine.

94
00:09:12,000 --> 00:09:15,000
He does all the really good C++ for Drivechain.

95
00:09:15,000 --> 00:09:22,000
He's done all that, so he's done all the real coding work, so he's an awesome guy.

96
00:09:22,000 --> 00:09:25,000
Yeah, I was actually running the – and it's very easy.

97
00:09:25,000 --> 00:09:30,000
For everyone who's curious, go to drivechain.info, and there's a very simple guide.

98
00:09:30,000 --> 00:09:36,000
If you're running Linux, just download the binaries, and you can get the sidechain
one running.

99
00:09:36,000 --> 00:09:43,000
You can get the DriveNet stuff running, and you can really see the things in action
on a little testnet right now.

100
00:09:43,000 --> 00:09:49,000
It's very, very simple, and so I presume most of this work is from this Cryptex
colleague

101
00:09:49,000 --> 00:09:53,000
who's actually been around the Bitcoin for a very long time, and also yourself.

102
00:09:53,000 --> 00:09:57,000
So between you two, you're doing most of the work currently, yeah?

103
00:09:57,000 --> 00:10:05,000
Yeah, on the actual Drivechain project, I think actually Patrick has probably
written every line of code that is there now,

104
00:10:05,000 --> 00:10:08,000
and no one else has, I think.

105
00:10:08,000 --> 00:10:15,000
But people will contribute in various other ways, review documentation, et cetera,
trying stuff out.

106
00:10:15,000 --> 00:10:18,000
We have people who have made add-ons already.

107
00:10:18,000 --> 00:10:24,000
So in the design, the design calls for a number of things to happen in code,

108
00:10:24,000 --> 00:10:27,000
but it also calls for one thing to happen in the world of business,

109
00:10:27,000 --> 00:10:34,000
which is for people to compete on offering services to quickly move coins from the
sidechain back to the mainchain.

110
00:10:34,000 --> 00:10:39,000
We have someone who, Andreas Brecken, the guy who did that Lightning Network review,

111
00:10:39,000 --> 00:10:44,000
the guy who runs ShiftCoin.com, and he did the Lightning Network review where he
connected to,

112
00:10:44,000 --> 00:10:50,000
he controlled like more than half of the Lightning Network with like a 50 Bitcoin
Lightning Network node or something.

113
00:10:50,000 --> 00:10:52,000
That guy has made a cool website.

114
00:10:52,000 --> 00:10:57,000
It's something like, I think it's something like HerokuApp.Sideshift.io or
something.

115
00:10:57,000 --> 00:11:01,000
We'll have to get the link for everyone in particular, but if you go to
drivechain.info,

116
00:11:01,000 --> 00:11:04,000
I'd link to it in the first couple of sentences.

117
00:11:04,000 --> 00:11:07,000
So we have other things like that. It's very interesting stuff.

118
00:11:07,000 --> 00:11:14,000
Yeah. Andreas has actually been on the TeamSpeak since 2014, 2015.

119
00:11:14,000 --> 00:11:17,000
So we know him from a long time ago when he was working at Drivechain.

120
00:11:17,000 --> 00:11:19,000
Yeah, very funny guy. Very clever.

121
00:11:19,000 --> 00:11:22,000
Yeah, he's great. I thought the review of Lightning was actually brilliant.

122
00:11:22,000 --> 00:11:24,000
And a lot of people got, for whatever reason, triggered by it,

123
00:11:24,000 --> 00:11:26,000
but I thought it was very thorough, technically focused.

124
00:11:26,000 --> 00:11:29,000
I thought it was a great approach.

125
00:11:29,000 --> 00:11:33,000
I actually used his faucet to get started when I was playing around with the
Drivechain.

126
00:11:33,000 --> 00:11:35,000
Oh, yes. He also made a faucet. It's great.

127
00:11:35,000 --> 00:11:39,000
Yeah, unfortunately, we live in a very polarized time, which makes it very
difficult.

128
00:11:39,000 --> 00:11:43,000
Anyone who's interested in like discourse, it's very hard to say anything

129
00:11:43,000 --> 00:11:48,000
and just have it be a piece of knowledge or information.

130
00:11:48,000 --> 00:11:54,000
It has to immediately be polarized into a kind of emblem of support for one side or
another.

131
00:11:54,000 --> 00:11:59,000
And actually discussing ideas is very difficult.

132
00:11:59,000 --> 00:12:03,000
Yeah, and I find that whenever you're discussing like a single topic with someone

133
00:12:03,000 --> 00:12:06,000
and you might be representing some kind of opinion,

134
00:12:06,000 --> 00:12:10,000
then they try to associate you with the tribe that is most commonly expressing this
opinion.

135
00:12:10,000 --> 00:12:14,000
So then they'll suddenly start attributing all kinds of other opinions on other
topics to you,

136
00:12:14,000 --> 00:12:17,000
even though it's really some tribe that they're really referring to.

137
00:12:18,000 --> 00:12:23,000
Yes, those people are victims as well as not only are they making a mistake

138
00:12:23,000 --> 00:12:28,000
and misinterpreting what's being said and misrepresenting someone's views, which is
wrong,

139
00:12:28,000 --> 00:12:36,000
but also they are sort of victims of this polarization because they feel a deep
insecurity,

140
00:12:36,000 --> 00:12:44,000
which is such that whenever they encounter anything that is deviant or threatening
to their camp,

141
00:12:44,000 --> 00:12:50,000
they feel the need to advertise very loudly to everyone around that they don't
associate with it.

142
00:12:50,000 --> 00:13:01,000
So it's really their own kind of fear that people will think that they're
associating with you.

143
00:13:03,000 --> 00:13:07,000
So that sounds probably like a little psychological babble or something,

144
00:13:07,000 --> 00:13:12,000
but really these people are kind of – you have to feel sorry for them, I think, more
than anything else.

145
00:13:12,000 --> 00:13:15,000
Yes, it's like their virtue signaling to their tribe members that,

146
00:13:15,000 --> 00:13:18,000
hey, don't worry, I'm not related to this guy.

147
00:13:18,000 --> 00:13:20,000
That's just a virtue signaling exercise.

148
00:13:20,000 --> 00:13:24,000
So it has nothing to do with you or your ideas, which is, again,

149
00:13:24,000 --> 00:13:28,000
it's just disappointing for anyone who's actually interested in real discourse
because it's very difficult.

150
00:13:29,000 --> 00:13:30,000
Exactly, yes.

151
00:13:30,000 --> 00:13:37,000
So last month in Riga, at the same time that this conference,

152
00:13:37,000 --> 00:13:40,000
this sort of Bitcoin-focused development conference in Riga by Honeybadger,

153
00:13:40,000 --> 00:13:43,000
there was this big bug that was discovered.

154
00:13:43,000 --> 00:13:47,000
And not many people have really discussed it,

155
00:13:47,000 --> 00:13:51,000
and I was kind of vocal in the conference of trying to get that on the agenda

156
00:13:51,000 --> 00:13:53,000
and discussing the implications and all these things.

157
00:13:53,000 --> 00:13:57,000
Can you sort of tell us what you think about the bug in general,

158
00:13:57,000 --> 00:14:02,000
how it was disclosed, whether it was an appropriate process that it was done in?

159
00:14:02,000 --> 00:14:09,000
And if you think that, in general, the community has reacted properly to the
significance of this.

160
00:14:10,000 --> 00:14:13,000
You know, actually, I don't think I know a lot.

161
00:14:13,000 --> 00:14:20,000
I'd rather talk about something where I'm much more of an expert relative to other
people.

162
00:14:20,000 --> 00:14:26,000
I mean, all I know is that it's very rare a bug appeared that was sort of where it
was

163
00:14:26,000 --> 00:14:31,000
kind of a very high level of catastrophe because it required people to update.

164
00:14:31,000 --> 00:14:37,000
What's more interesting to me is the fact that many people just ignored the request

165
00:14:37,000 --> 00:14:40,000
from senior developers to upgrade their software.

166
00:14:40,000 --> 00:14:45,000
So this idea that this thing came out, it's really important that everyone upgrade
immediately,

167
00:14:45,000 --> 00:14:49,000
big problem, but actually no problems really happened,

168
00:14:49,000 --> 00:14:55,000
and also like very few people upgraded, which I find to be more interesting.

169
00:14:55,000 --> 00:14:59,000
I've always wondered, the SegWit2x project failed,

170
00:14:59,000 --> 00:15:04,000
and you can either look at that as just a rejection of change

171
00:15:04,000 --> 00:15:08,000
and say, like, we just like the status quo and we don't want anyone messing with
this,

172
00:15:08,000 --> 00:15:16,000
or you can kind of look at it as an endorsement of the small block view and the kind
of pro dev view.

173
00:15:16,000 --> 00:15:18,000
And I think a lot of people look at it as an endorsement.

174
00:15:18,000 --> 00:15:23,000
They say, like, our side won, talking about the small blocker side.

175
00:15:23,000 --> 00:15:26,000
But, you know, it's also possible that just no one cares about anything.

176
00:15:26,000 --> 00:15:33,000
And so no one won and that no one, you know, there's no—there's just apathy is the
big winner.

177
00:15:33,000 --> 00:15:38,000
So that I think—I think that is kind of an interesting data point along that.

178
00:15:38,000 --> 00:15:44,000
But one of the things that the big blockers at least will say is that, well,

179
00:15:44,000 --> 00:15:48,000
this code that led to the bug was not properly—

180
00:15:48,000 --> 00:15:52,000
there wasn't enough checks and balances in terms of the review process

181
00:15:52,000 --> 00:15:56,000
because all these Blockstream employees are involved and they all love each other

182
00:15:56,000 --> 00:15:59,000
and just sort of give the thumbs up on everything.

183
00:15:59,000 --> 00:16:01,000
How much validity do you think there is to this argument?

184
00:16:01,000 --> 00:16:06,000
And do you feel like Bitcoin's development process is open enough currently?

185
00:16:06,000 --> 00:16:11,000
Well, I've heard actually a lot of people complain that the quality of the code has
been declining

186
00:16:11,000 --> 00:16:15,000
and that a lot of changes are just kind of being made for frivolous reasons

187
00:16:15,000 --> 00:16:23,000
and that there's a kind of—that no one kind of really knows even what's going on
anymore.

188
00:16:23,000 --> 00:16:26,000
Like originally Satoshi wrote the code in a big block,

189
00:16:26,000 --> 00:16:30,000
which anyone who's familiar with software development knows that that is kind of
like—

190
00:16:30,000 --> 00:16:33,000
that is kind of like an easy way to just get something to work.

191
00:16:33,000 --> 00:16:37,000
And then when you work on a big team, of course, they split it out.

192
00:16:37,000 --> 00:16:42,000
Or when the project is very large, you're supposed to split it out into like little
modular pieces.

193
00:16:43,000 --> 00:16:46,000
And I don't know, like it's hard to say.

194
00:16:46,000 --> 00:16:49,000
I think it's not even really relevant.

195
00:16:49,000 --> 00:16:52,000
The important thing is just can we recover from mistakes in the code.

196
00:16:52,000 --> 00:16:55,000
And so when people point out like, oh, there was a bug,

197
00:16:55,000 --> 00:17:00,000
like I didn't really think it was that interesting when it happened in BCH

198
00:17:00,000 --> 00:17:03,000
and I don't think it's necessarily that interesting when it happens in BTC.

199
00:17:03,000 --> 00:17:05,000
Although, of course, it was interesting for different reasons.

200
00:17:05,000 --> 00:17:07,000
There's kind of an exception that proves the rule.

201
00:17:07,000 --> 00:17:13,000
When the bugs started appearing in BCH, it was important because it may have
indicated something else,

202
00:17:13,000 --> 00:17:18,000
which is that these people are kind of very reckless and they have no idea what
they're doing.

203
00:17:18,000 --> 00:17:25,000
And these are just a few bugs of a giant bug storm that's coming down the road in
the BCH side.

204
00:17:25,000 --> 00:17:28,000
So that was why that was kind of relevant.

205
00:17:28,000 --> 00:17:34,000
And then in BTC, I think it is relevant because this is really like the first like
big screw up.

206
00:17:34,000 --> 00:17:38,000
And then it's kind of relevant because it may indicate, well, why weren't there
screw ups before?

207
00:17:38,000 --> 00:17:46,000
Of course, there were screw ups before, but this is the first screw up that was
like, you know, it was like not first.

208
00:17:46,000 --> 00:17:49,000
First of all, it's the first screw up that was found from outside.

209
00:17:49,000 --> 00:17:52,000
So it was the first like potentially exploitable screw up.

210
00:17:52,000 --> 00:17:57,000
Everything else before was someone inside noticed something and it was fixed long
before.

211
00:17:57,000 --> 00:18:00,000
You know, there was kind of never any risk of damage.

212
00:18:01,000 --> 00:18:08,000
So they are kind of interesting because, but in general, I don't think like a bug is
only important if like no one could ever fix it.

213
00:18:08,000 --> 00:18:11,000
But something where the bug is very fixable is fine.

214
00:18:11,000 --> 00:18:13,000
Like, well, the DAO hacker is an example.

215
00:18:13,000 --> 00:18:19,000
That's the bug was sort of an unfixable one because it involved the guy withdrawing
a bunch of money.

216
00:18:19,000 --> 00:18:21,000
And then now it's like that's his money.

217
00:18:21,000 --> 00:18:23,000
So we can't roll back the chain.

218
00:18:23,000 --> 00:18:24,000
Of course, that's exactly what they did.

219
00:18:24,000 --> 00:18:25,000
So they did fix it.

220
00:18:25,000 --> 00:18:28,000
So it's ironically, they kind of adopted my framework there.

221
00:18:28,000 --> 00:18:30,000
But they can't.

222
00:18:30,000 --> 00:18:39,000
You really want, we live in a world where, you know, humans are imperfect and
fallible and our knowledge is incomplete.

223
00:18:39,000 --> 00:18:42,000
And, you know, we live in a world where everything is imperfect.

224
00:18:42,000 --> 00:18:51,000
And the only thing we can do is try to correct mistakes when we discover them and
try to continue correcting and get better and better and better each time.

225
00:18:51,000 --> 00:18:55,000
That's just the way everything works in all fields of human progress and knowledge.

226
00:18:55,000 --> 00:18:57,000
And it's just like everywhere.

227
00:18:57,000 --> 00:19:00,000
And so there's no exception in Bitcoin development.

228
00:19:00,000 --> 00:19:05,000
You just want to make sure that the errors don't do a lot of damage and that they're
fixed quickly.

229
00:19:05,000 --> 00:19:16,000
And the institution of the soft fork protects us in many ways because if a version
is, it has to be a bug in like basically all the versions in order to significantly
affect users.

230
00:19:16,000 --> 00:19:21,000
So the constant tightening of the rules hardens the protocol further and further.

231
00:19:21,000 --> 00:19:29,000
But yeah, and just in general, to answer your question, I think that the protocol
probably should stop changing in general.

232
00:19:29,000 --> 00:19:34,000
Like it should just completely harden at the consensus management layer.

233
00:19:34,000 --> 00:19:39,000
And I think sidechains are really the big opportunity to do that.

234
00:19:39,000 --> 00:19:48,000
You harden up this one layer and then you have complete freedom to experiment with
new, completely new software applications.

235
00:19:48,000 --> 00:19:50,000
And I think that's just the ideal.

236
00:19:50,000 --> 00:19:59,000
That's almost like a sign from fate that this is around the time to kind of wrap it
up, I think.

237
00:19:59,000 --> 00:20:12,000
Yeah, and I think that social consensus aspect, like the fact that we have a strong
community that can respond to bugs and get it fixed within a matter of hours and get
it properly implemented and stuff.

238
00:20:12,000 --> 00:20:17,000
I think that's probably the strongest feature of Bitcoin is that we do have this
strong community.

239
00:20:17,000 --> 00:20:20,000
But that's a great segue into the meat of what we wanted to get to.

240
00:20:20,000 --> 00:20:23,000
So we've got some nice sort of context of where you're coming from.

241
00:20:23,000 --> 00:20:25,000
And so sidechains.

242
00:20:25,000 --> 00:20:33,000
So I've been a fan of sidechains since back in like 2014, 15, when there was a lot
of hype around them.

243
00:20:33,000 --> 00:20:45,000
And the Bitcoin maximalists back then used to be essentially, I mean, I've been
essentially a maximalist since the start and always thought shit coins were shit and
all that.

244
00:20:46,000 --> 00:20:48,000
Yeah, I think I have as well.

245
00:20:48,000 --> 00:20:54,000
But the whole thing back then was, listen, if you want to have all of these
experiments, then do it on a sidechain.

246
00:20:54,000 --> 00:20:56,000
So we're going to facilitate.

247
00:20:56,000 --> 00:20:59,000
Bitcoin is the king and I'm all about Bitcoin.

248
00:20:59,000 --> 00:21:02,000
But obviously, we should have the flexibility to experiment.

249
00:21:02,000 --> 00:21:05,000
And obviously, it should be with Bitcoin in the form of sidechains.

250
00:21:05,000 --> 00:21:10,000
It was never just like Bitcoin is the best and everything that it does, it covers
everything.

251
00:21:10,000 --> 00:21:13,000
And we don't need any of these alternative technologies.

252
00:21:13,000 --> 00:21:17,000
But somehow it's changed the past couple of years, I've noticed.

253
00:21:17,000 --> 00:21:25,000
Like this, I got less and less about doing trustless sidechains and became more
about, you know, Rootstock initially was saying they were going to do things in all
this great way.

254
00:21:25,000 --> 00:21:27,000
And then they got more federated.

255
00:21:27,000 --> 00:21:32,000
And then Blockstream's proposition on Liquid became more and more about just a
trusted federation.

256
00:21:32,000 --> 00:21:38,000
So maybe you can let's start very, very basic and talk about what really defines a
sidechain.

257
00:21:39,000 --> 00:21:47,000
And how, for example, does a federated trusted system fit into that definition?

258
00:21:47,000 --> 00:21:49,000
Oh, wow, that was a lot of things.

259
00:21:49,000 --> 00:21:50,000
Okay, yeah, that was cool.

260
00:21:50,000 --> 00:22:04,000
First, I want to comment on something, which is that I agree with you that it's kind
of ironic that these altcoins started to pop up and they were originally just so
boring that they could be easily dismissed.

261
00:22:04,000 --> 00:22:08,000
But then in 2015, there was this kind of like, well, you can just put them on a
sidechain.

262
00:22:08,000 --> 00:22:14,000
But the sidechains, the altcoin technology was still very boring and kind of
ridiculous.

263
00:22:14,000 --> 00:22:21,000
And now we've had like a complete inversion of that where now actually the altcoin
technology is sort of getting somewhat respectable.

264
00:22:21,000 --> 00:22:28,000
Where you have something like Monero or you have like Sciacoin, you have these other
projects.

265
00:22:29,000 --> 00:22:37,000
Namecoin has been around for a while, but now it's kind of accepted that actually
there's a big demand for ERC-20 tokens at the very least.

266
00:22:37,000 --> 00:22:47,000
There's a very loud and seeming demand for cheaper transaction fee like large block
BCH type things.

267
00:22:48,000 --> 00:22:52,000
So people argue about whether or not this demand is illusory or not.

268
00:22:52,000 --> 00:23:02,000
But I just think it's ironic that the technology in the modern, in the last few
years is now the altcoin technology is much more serious than it was.

269
00:23:02,000 --> 00:23:13,000
And yet the kind of sidechain neglect has also like the opinion or the kind of
reliance on sidechains has kind of fallen kind of at the same time.

270
00:23:13,000 --> 00:23:17,000
So it's kind of gotten in a very bizarre kind of inversion there.

271
00:23:17,000 --> 00:23:18,000
But yeah, you asked it.

272
00:23:18,000 --> 00:23:25,000
I mean, the definition of a sidechain I think is there are, you know, it's like how
do you define like a stop sign?

273
00:23:25,000 --> 00:23:29,000
There's a lot of, you know, it depends on those certain list of attributes.

274
00:23:29,000 --> 00:23:34,000
And then there's like a context and then there's like infinite levels of description
you can reach.

275
00:23:34,000 --> 00:23:42,000
But definitely the first aspect is that it is a blockchain where there is a token
that can move to other blockchains and back.

276
00:23:42,000 --> 00:23:51,000
So that is the – that's the major feature is that you have something that is a
blockchain and then the tokens can travel to different pieces of software.

277
00:23:51,000 --> 00:23:54,000
And then hopefully maybe change owners and then come back.

278
00:23:54,000 --> 00:24:01,000
So there's a kind of accounting dependence but otherwise an independence.

279
00:24:02,000 --> 00:24:04,000
What about the trust dimension though?

280
00:24:04,000 --> 00:24:09,000
How important is it that these sidechains are along the trustlessness spectrum?

281
00:24:09,000 --> 00:24:14,000
How important is it that they're actually trustless and not federated, trusted in
some form?

282
00:24:14,000 --> 00:24:18,000
Well, right. That's what – you can keep describing more and more detail.

283
00:24:18,000 --> 00:24:20,000
So it's like what is a blockchain?

284
00:24:20,000 --> 00:24:29,000
Like if someone could just say, well, just give all of your Bitcoin to me and my
friends and we will take care of it and we'll tell you what we're doing.

285
00:24:29,000 --> 00:24:33,000
And then when you ask for the Bitcoin, we'll give it back to you.

286
00:24:33,000 --> 00:24:36,000
And then you just put labels on things.

287
00:24:36,000 --> 00:24:41,000
You say that, OK, that is a blockchain and then that – therefore that is a
sidechain.

288
00:24:41,000 --> 00:24:53,000
To me, I really think that if someone is going to say that they have invented
blockchain technology, what they should really have invented to be slightly more
precise but kind of needlessly verbose about it.

289
00:24:53,000 --> 00:24:56,000
They're really – you're really inventing like a kind of a sidechain factory.

290
00:24:56,000 --> 00:24:58,000
You're really saying here's this technology.

291
00:24:58,000 --> 00:25:00,000
You can use it to spin up.

292
00:25:00,514 --> 00:25:04,514
You can fork it or you can click a button inside of it or something.

293
00:25:04,514 --> 00:25:06,514
You can spin up your own sidechain whenever you want.

294
00:25:06,514 --> 00:25:14,514
And so I find the federated model to fail on a number of criteria.

295
00:25:14,514 --> 00:25:21,514
One is that like you can't reuse it other than in this trivial sense of creating
another –

296
00:25:21,514 --> 00:25:28,514
you have to pick another set of people that you want to trust to hold onto the
money, another set of keys.

297
00:25:28,514 --> 00:25:35,514
But I kind of also feel that it is not – well, there's a lot to say about it.

298
00:25:35,514 --> 00:25:38,514
So I don't know which direction we should go first.

299
00:25:38,514 --> 00:25:44,514
But I think it's like – well, let me say one other thing though is that the
multisignature account,

300
00:25:44,514 --> 00:25:50,514
that's kind of like an infinite degrees of freedom situation where you can just kind
of say anything you want goes on in there.

301
00:25:51,514 --> 00:25:56,514
And whether or not something goes on in there or doesn't, it kind of doesn't matter
because at the end of the day,

302
00:25:56,514 --> 00:25:59,514
the only thing that matters is if the keys sign off on something.

303
00:25:59,514 --> 00:26:01,514
So it's kind of an infinite regress.

304
00:26:01,514 --> 00:26:06,514
It says we'll solve this trust problem, but it doesn't say –

305
00:26:06,514 --> 00:26:11,514
there's no relationship between the trust problem that you're solving and the
solution.

306
00:26:11,514 --> 00:26:14,514
So it's kind of the explanation that's not an explanation.

307
00:26:14,514 --> 00:26:19,514
It just says, well, how are we going to secure this other blockchain?

308
00:26:19,514 --> 00:26:21,514
And it's like, well, these other guys will do it.

309
00:26:21,514 --> 00:26:23,514
But it's like, how are those guys going to do it?

310
00:26:23,514 --> 00:26:27,514
It's kind of just an empty explanation in my opinion.

311
00:26:27,514 --> 00:26:29,514
Right, yeah.

312
00:26:29,514 --> 00:26:36,514
So let's talk about sidechains as if Drivechain is the ultimate sidechain just to
make things easier.

313
00:26:36,514 --> 00:26:40,514
Well, that would be easier for me, of course, because I can talk about my own
project,

314
00:26:40,514 --> 00:26:45,514
and I don't want to speak for other people any more than I have to.

315
00:26:45,514 --> 00:26:46,514
Right, exactly.

316
00:26:46,514 --> 00:26:53,514
So let's focus, just so everyone listening also understands that now the context is
going to be sidechains

317
00:26:53,514 --> 00:26:56,514
as defined by Drivechain's approach.

318
00:26:56,514 --> 00:26:59,514
So let's talk about what makes a sidechain.

319
00:26:59,514 --> 00:27:02,514
So let's say I want to – maybe it's helpful for people,

320
00:27:02,514 --> 00:27:06,514
because a lot of the traders here aren't necessarily super technical on the
development side.

321
00:27:06,514 --> 00:27:12,514
So let's talk about what it looks like if we were to do Ethereum in a Drivechain.

322
00:27:12,514 --> 00:27:16,514
Because there's a lot of guys that love the scripting aspects of Ethereum and having
that kind of flexibility.

323
00:27:16,514 --> 00:27:21,514
So how does that look in a Drivechain context where people can have all the
experiments and the fun,

324
00:27:21,514 --> 00:27:22,514
but what are they really doing?

325
00:27:22,514 --> 00:27:27,514
Like talk about the Bitcoin UTXOs that are required to go into the sidechain

326
00:27:27,514 --> 00:27:33,514
and how that SPV proofing just breaks this down for people who aren't necessarily
super technical.

327
00:27:34,514 --> 00:27:35,514
Yeah, okay.

328
00:27:35,514 --> 00:27:42,514
Well, Drivechain is a method for transferring Bitcoin back and forth among different
blockchains.

329
00:27:42,514 --> 00:27:45,514
So that was kind of what we talked about before.

330
00:27:45,514 --> 00:27:51,514
So the primary goal of Drivechain is to have fully ignorable sidechains.

331
00:27:51,514 --> 00:27:54,514
So you want the parent chain, the main chain.

332
00:27:54,514 --> 00:27:58,514
It doesn't need to know what its sort of children are doing specifically.

333
00:27:59,514 --> 00:28:06,514
So in that sense, you would not be on the hook for the Ethereum rules or any bugs in
Ethereum or any weird stuff like that.

334
00:28:06,514 --> 00:28:08,514
So that's kind of the primary goal.

335
00:28:08,514 --> 00:28:11,514
The secondary goal is, of course, that you want to have some kind of redeemability.

336
00:28:11,514 --> 00:28:16,514
You want to be able to guarantee that each sidechain Bitcoin, when you own a Bitcoin
on the sidechain,

337
00:28:16,514 --> 00:28:21,514
you can redeem it for one Bitcoin back on the regular main chain.

338
00:28:21,514 --> 00:28:24,514
Or if there are sidechains, have sidechains then.

339
00:28:25,514 --> 00:28:31,514
If you've got a Bitcoin on a child, it can go to the parent chain and then it can go
to its grandparent chain or whatever.

340
00:28:33,514 --> 00:28:34,514
It's difficult to describe.

341
00:28:34,514 --> 00:28:39,514
It's difficult to describe this technology because I think that those two goals are
actually mutually exclusive.

342
00:28:39,514 --> 00:28:42,514
You cannot have both at once.

343
00:28:42,514 --> 00:28:49,514
But you try with some clever rearranging to try and get like 100% of the first
ignorable sidechains.

344
00:28:49,514 --> 00:28:54,514
And you try to get like whatever, like 90, 99% of the second.

345
00:28:54,514 --> 00:28:56,514
But you actually cannot get them both at the same time.

346
00:28:56,514 --> 00:29:03,514
And so the challenge when explaining sidechain technology is that you bring up
something along the first axis

347
00:29:03,514 --> 00:29:05,514
and then people will criticize it along the second axis.

348
00:29:05,514 --> 00:29:09,514
And then if you bring something up on the second axis, they will criticize it on the
first axis.

349
00:29:10,514 --> 00:29:13,514
And that is not really the game.

350
00:29:13,514 --> 00:29:16,514
The game is not to get perfection on both axes.

351
00:29:17,514 --> 00:29:22,514
The goal is to get perfection on the first axis, which is ignorable sidechains.

352
00:29:23,514 --> 00:29:27,514
And just to do as well as you can on redeemability.

353
00:29:28,514 --> 00:29:35,514
So that's something that I think people should keep in mind if they want to study
the criticisms of sidechains.

354
00:29:36,514 --> 00:29:42,514
In particular, the fact that you have...

355
00:29:45,514 --> 00:29:47,514
Well, so you're asking how an Ethereum sidechain works.

356
00:29:47,514 --> 00:29:49,514
So what you want about...

357
00:29:49,514 --> 00:29:55,514
The reason that fully ignorable sidechains are important is because that is what
frees you up to just allow the sidechain developers

358
00:29:55,514 --> 00:29:58,514
to just do whatever crazy, terrible ideas they want.

359
00:29:58,514 --> 00:30:02,514
So I think that Turing completeness is not a great idea.

360
00:30:03,514 --> 00:30:05,514
But it shouldn't really be up to me, right?

361
00:30:05,514 --> 00:30:10,514
When Snapchat came out, I thought it was a terrible idea too, but now I use it like
every day.

362
00:30:11,514 --> 00:30:13,514
I thought Twitter was a dumb idea when it came out.

363
00:30:13,514 --> 00:30:17,514
So we don't really want it to be up to any person to judge.

364
00:30:17,514 --> 00:30:21,514
It should be up to the users to decide what's a good idea and what's a bad idea.

365
00:30:21,514 --> 00:30:25,514
So you need this ignorability criterion so that someone can do their weird Ethereum
idea.

366
00:30:25,514 --> 00:30:29,514
And then what happens is that you would send money from Bitcoin.

367
00:30:30,514 --> 00:30:32,514
And this is a little bit easier if you actually just do it.

368
00:30:32,514 --> 00:30:36,514
So as you mentioned, you can go to drivechain.info and download our test software.

369
00:30:37,514 --> 00:30:40,514
And you can just see there's like a little window.

370
00:30:40,514 --> 00:30:44,514
You just sort of click a button and you send money to a sidechain address.

371
00:30:44,514 --> 00:30:50,514
And it's deposited. It goes away in your one window, but then it shows up in the
other window.

372
00:30:50,514 --> 00:30:53,514
So you're just sending Bitcoin to a different software application.

373
00:30:53,514 --> 00:30:58,514
You can send it to Ethereum, and then once it lands in Ethereum, it can do any crazy
thing that it wants.

374
00:30:58,514 --> 00:31:00,514
Now, of course, the trick is getting it back.

375
00:31:00,514 --> 00:31:04,514
Yeah, but actually, let's stop there then and explore that more.

376
00:31:04,514 --> 00:31:06,514
So let's pretend I'm Vitalik.

377
00:31:06,514 --> 00:31:09,514
And let's rewind back to 2014-15.

378
00:31:09,514 --> 00:31:13,514
And I'm thinking, oh man, I want to get all these changes into Bitcoin, but they
won't allow me.

379
00:31:13,514 --> 00:31:17,514
And I'm talking to you now, Paul, and I'm saying, Paul, I'm going to go make my own
shitcoin.

380
00:31:17,514 --> 00:31:19,514
I'm going to make a foundation. We're going to do all this shit.

381
00:31:19,514 --> 00:31:21,514
We're going to sell tokens to the public. We're going to raise this money.

382
00:31:21,514 --> 00:31:24,514
We're going to maybe break securities laws. Who knows? We're just going to do it.

383
00:31:24,514 --> 00:31:26,514
Yeah, right. Exactly.

384
00:31:26,514 --> 00:31:31,514
And you say, Vitalik, stop. Don't do that. I've got this great idea.

385
00:31:31,514 --> 00:31:34,514
I've got Drivechains. Put your projects in the Drivechain.

386
00:31:34,514 --> 00:31:38,514
And then I'm Vitalik, and I say, okay, can I still do a token sale?

387
00:31:38,514 --> 00:31:39,514
Like, how does this look?

388
00:31:39,514 --> 00:31:46,514
Will I still be able to create X number of Ethereum tokens, and they're backed by
the UTXOs that I've put into the sidechain?

389
00:31:46,514 --> 00:31:51,514
How would Vitalik then go to the next step and say, all right, Paul, you convinced
me.

390
00:31:51,514 --> 00:31:53,514
I'm going to do Ethereum in the sidechain.

391
00:31:54,514 --> 00:32:05,514
Well, I think it's not so much about – I think there's a misconception that this is
about convincing the altcoin developers to switch over.

392
00:32:05,514 --> 00:32:10,514
It's really kind of more – the relationship is, I think, much more adversarial.

393
00:32:10,514 --> 00:32:22,514
It basically says, Vitalik, make it a sidechain, or after you do all of the hard
work and build it and test it and do all this stuff,

394
00:32:22,514 --> 00:32:28,514
building out the technology and proving that it works and that some people find it
useful,

395
00:32:28,514 --> 00:32:36,514
as soon as you do that, we will just steal all of your source code and we will make
it a sidechain with very, very low effort to ourselves.

396
00:32:37,514 --> 00:32:46,514
Once it's reached its ultimate form, we'll just take it, and then all the Bitcoin
users will have all the benefits of the Ethereum technology.

397
00:32:46,514 --> 00:32:52,514
So why would you buy one Bitcoin and one ETH when you can instead just buy one
Bitcoin that does the exact same thing?

398
00:32:52,514 --> 00:32:55,514
So it's kind of much more of a threat.

399
00:32:55,514 --> 00:32:58,514
There's no need – there are a lot of misconceptions.

400
00:32:58,514 --> 00:33:06,514
I will throw out these little lines about, well, sidechains will destroy altcoins or
something, but it doesn't really need to do that.

401
00:33:06,514 --> 00:33:17,514
It's just like a standing kind of option to do that and a standing kind of threat
that if someone wants to build work on new blockchain technology,

402
00:33:17,514 --> 00:33:19,514
that we will just take it.

403
00:33:19,514 --> 00:33:29,514
I think there is an element of persuasion in it, though, which is that if you use a
sidechain, it kind of proves that you're not out to just make a quick buck,

404
00:33:29,514 --> 00:33:40,514
whereas if you – in a world where sidechains exist, what's the marginal decision on
should I make an altcoin versus should I make a sidechain?

405
00:33:40,514 --> 00:33:43,514
People will just say, well, why are you making an altcoin?

406
00:33:43,514 --> 00:33:50,514
And they'll say, well, because I love the technology, and they'll be like, well,
okay, but you could do that in a different way without making an altcoin.

407
00:33:50,514 --> 00:33:59,514
So it completely removes the pretext of a need for altcoin, and I think almost all
these markets are very highly pretext-driven.

408
00:33:59,514 --> 00:34:12,514
They're all just like fake excuses, and it's just like they're hoping that most of
the traders and investors involved don't figure out how empty it all is until
everyone else has caught and run.

409
00:34:12,514 --> 00:34:21,514
But aren't Drivechains flexible enough that someone would still be able to – because
like you said, the chain has its own token on it, right, as well?

410
00:34:21,514 --> 00:34:23,514
Oh, well, you can – so this is the thing.

411
00:34:23,514 --> 00:34:28,514
If you put Bitcoin into the sidechain, then you can only get as many Bitcoin out of
it that you put in.

412
00:34:28,514 --> 00:34:30,514
But on the sidechain, you can do that every one.

413
00:34:30,514 --> 00:34:35,514
You can make two – you can make extra types of tokens, and they can be over there.

414
00:34:36,514 --> 00:34:46,514
I would imagine – I've long talked about like a counterparty-like sidechain because
the integration between counterparty and Bitcoin is a little awkward technically and
somewhat byte inefficient.

415
00:34:46,514 --> 00:34:50,514
Excuse me? It works just fine for Rare Pepe, as well.

416
00:34:51,514 --> 00:35:02,514
Yeah, so I think it certainly works, but there's this question of – there's always
this tension between like is – like a Luke Jr. type will say that it doesn't belong
there.

417
00:35:02,514 --> 00:35:13,514
And so why not put all the asset stuff where it's in one place, and then anyone
who's interested in assets can just download like some supplemental data, and they
can play over there.

418
00:35:13,514 --> 00:35:23,514
And then stuff would be cheaper over there because, of course, Bitcoin transaction
fees go up for people who really want to use these unique Bitcoin properties of a
small block Bitcoin chain.

419
00:35:23,514 --> 00:35:29,514
They really want it to be super secure, super durable money that's very
censorship-resistant.

420
00:35:29,514 --> 00:35:36,514
Those people may be willing to pay a premium for that, but that crowds out the Rare
Pepe counterparty transactions.

421
00:35:36,514 --> 00:35:43,514
And so it's better if you have a different chain that doesn't take itself as
seriously, that has lower fees, that has all the assets on it.

422
00:35:43,514 --> 00:35:46,514
And I have specced this out as well as a BitAssets sidechain.

423
00:35:46,514 --> 00:35:52,514
I have kind of written down how I think that would work in an efficient way.

424
00:35:52,514 --> 00:36:03,514
So that is – you can have other assets, but you probably wouldn't have Ether
specifically because Ether is like – Ether is the competitor to BTC in the Ethereum
world.

425
00:36:03,514 --> 00:36:14,514
So you would move Bitcoin over to the Ethereum sidechain, and you would use BTC over
there to pay gas costs and pay whatever, whatever wacky Ethereum costs of the day.

426
00:36:14,514 --> 00:36:25,514
You'd pay those in BTC, and those BTC that you pay, they would go to the Ethereum
sidechain miners, so to speak, and then – but in that world, you could do many other
things.

427
00:36:25,514 --> 00:36:28,514
You could have ERC tokens inside that world though.

428
00:36:28,514 --> 00:36:33,514
Right, and also wouldn't they – would they still have like – let's say would they
have their own proof-of-work algo and stuff?

429
00:36:33,514 --> 00:36:42,514
Like wouldn't they be able to have their own little economy within that sidechain
even so they could still have Ether gas and all that stuff, but then underneath it
is these Bitcoin UTXOs that are backing it?

430
00:36:43,514 --> 00:37:02,514
Yeah, they could. You could have – you could say that what we're going to do is
we're going to make a sidechain where you can send BTC, but actually what's going on
is that BTC are just going to be poked around by Ethereum smart contracts and then
withdrawn.

431
00:37:02,514 --> 00:37:23,514
So you could do that and you could have Ether be in there. They kind of already did
that with BTC relay sort of. Not quite because it doesn't settle back to the Bitcoin
chain because the Bitcoin chain doesn't know what is going on in Ethereum at all in
any sense, even in a limited sidechain sense.

432
00:37:24,514 --> 00:37:31,514
But yeah, you could actually do that if you wanted. So you could have a side – you
have basically infinite degrees of freedom on the sidechain. You can write anything
you want over there.

433
00:37:32,514 --> 00:37:47,514
I don't think – although you could have your own mining, I have long believed that
in practice no one ultimately will because you have a free option to use merge
mining or as I have sort of tweaked it into something called Blind Merged Mining.

434
00:37:47,514 --> 00:37:57,514
That's just – it's just a free option and you get basically all Bitcoin's hash rate,
and I just think that's just such an enticing offer that no one will really want to
pass it up.

435
00:37:57,514 --> 00:38:09,514
In particular, the sidechain has to pay transaction fees just like every blockchain,
but it has no block subsidy. There's no 12.5 Bitcoin being created every 10 minutes.

436
00:38:10,514 --> 00:38:20,514
So the problem is if you ever get into a lull where there's just few transaction
fees, then there's nothing. There's absolutely no incentive to mine at all.

437
00:38:21,514 --> 00:38:28,514
So you're really going to have that demand way up and stable, and that would –
that's very hard to guarantee, especially if the sidechain is new.

438
00:38:28,514 --> 00:38:38,514
So I just think it would like starve out and die very early on if it didn't use this
option. So it's just – I think in practice everyone will do Blind Merged Mining.

439
00:38:39,514 --> 00:38:44,514
But yeah, you are free to do other weird things, proof of stake, whatever. You could
do any weird thing you want.

440
00:38:45,514 --> 00:38:53,514
You can even just do something that says your signature is what – this is the
authority on what happens in the – you could even do a Corda thing.

441
00:38:53,514 --> 00:39:01,514
You could do like a really weird – but I don't think in practice anyone will do any
of these things because I think first of all, most of them have no value.

442
00:39:02,514 --> 00:39:08,514
These are just things that – this is a pretext that I mentioned before that just
people want to do something different because there's a lot of money for funding
blockchain.

443
00:39:09,514 --> 00:39:14,514
So people can – so people could in principle run ICO scams in a sidechain.

444
00:39:15,514 --> 00:39:21,514
Oh yeah, definitely. Well, that's why I think you would make the BitAsset sidechain
and then from there you just say here's our token.

445
00:39:22,514 --> 00:39:31,514
And then once you have those ingredients, then I'm sure a bunch of disreputable
people will just say – they'll just staple a bunch of scam onto that and go with it.

446
00:39:32,514 --> 00:39:38,514
But I don't – who knows? You never know how much this stuff depends on context,
right? Like maybe 2017 was just a magical year for that. I don't know.

447
00:39:39,514 --> 00:39:51,514
So we have 21 million coins that will ever be created in our project. That means
there's a limited number of UTXOs that could be utilized for various things like
sidechains.

448
00:39:52,514 --> 00:39:54,514
So what are the constraints for –

449
00:39:55,514 --> 00:40:04,514
Right. Well, you're talking about what is the upgrade. It's like there's an upgrade
to support SegWit and there is an upgrade to support this technology.

450
00:40:05,514 --> 00:40:07,514
That's what you're talking about, right?

451
00:40:08,514 --> 00:40:20,514
Well, yeah. So the mainchain has to track – it has to track which sidechains exist
but only – they're like a ledger entry. It's like a row in Microsoft Excel.

452
00:40:21,514 --> 00:40:24,514
They need to track which sidechains exist and how much Bitcoin is in them.

453
00:40:25,514 --> 00:40:29,514
But they have to kind of do that anyway for any user and any key pair that's
generated.

454
00:40:30,514 --> 00:40:40,514
So they have to just track a couple other things too like when someone tries to
initiate a withdrawal from the sidechain, they have to track when the withdrawal has
first been proposed.

455
00:40:41,514 --> 00:40:45,514
And then they have to keep track of all the withdrawals of all the sidechains for a
while.

456
00:40:46,514 --> 00:40:50,514
But all that stuff, all that information is published inside the blockchain already.

457
00:40:51,514 --> 00:41:00,514
So it respects all of the one megabyte block limit or what we now have is a weirder
thing, one megabyte, four megabyte block size, weight limit.

458
00:41:01,514 --> 00:41:10,514
But all that stuff is respected because the Drivechain is just processing data that
is – that has to be placed into the blockchain anyway.

459
00:41:11,514 --> 00:41:14,514
So the main bottleneck is in that direction.

460
00:41:15,514 --> 00:41:27,514
It's very quite light I think and there are a lot of – I don't know about – people
compare like lines of code changed and there's obviously a lot of debate over
whether or not the usefulness of that metric.

461
00:41:28,514 --> 00:41:37,514
I don't – this is a very – this is an optional thing just like SegWit is optional,
just like all soft forks are optional.

462
00:41:38,514 --> 00:41:39,514
You don't actually need to upgrade.

463
00:41:39,514 --> 00:41:40,514
But there's a lot of debate about that.

464
00:41:41,514 --> 00:41:46,514
Some people say that software updates are never optional and you should always be
running the latest software.

465
00:41:47,514 --> 00:41:48,514
That's not what I think at all.

466
00:41:49,514 --> 00:41:53,514
But these things are all debated by lots of people in different directions.

467
00:41:54,514 --> 00:41:57,514
Yeah, I mean paid-to-script hash was a soft fork, right?

468
00:41:58,514 --> 00:41:59,514
So that's like –

469
00:42:00,514 --> 00:42:05,514
Yes, the way all these work, as people probably know now, it's kind of more common
knowledge now I think than it was.

470
00:42:05,514 --> 00:42:15,514
But the way these things work is you take a path where the money would sort of be
immediately spendable by anyone and that path already exists.

471
00:42:16,514 --> 00:42:28,514
A bunch of these paths already exist in Bitcoin and then you attach – you kind of
like attach a filter to that that catches everything that you don't want and then
what's left is what you want.

472
00:42:28,514 --> 00:42:39,514
So all the invalid Drivechain withdrawals would hit the filter and then the block
would be invalid and that's how these soft forks work in practice most of the time.

473
00:42:40,514 --> 00:42:43,514
And so let's talk about the security of these sidechains.

474
00:42:44,514 --> 00:42:51,514
So one of the common criticisms is that miners are in control with these sidechains.

475
00:42:52,514 --> 00:43:03,514
They can sneak in and do a withdrawal from the sidechain and unless users are
actively monitoring for this, then they won't be able to stop it.

476
00:43:04,514 --> 00:43:13,514
Can you explain what role miners are playing in the sidechain and how does the
sidechain security compare to sort of the miner incentives on the mainchain?

477
00:43:14,514 --> 00:43:16,514
Yes, there's a lot to say there.

478
00:43:16,514 --> 00:43:20,514
So first I'm going to bring up that this is this issue of the two axes.

479
00:43:21,514 --> 00:43:25,514
So you want the sidechain to be ignorable, but we also want it to be secure.

480
00:43:26,514 --> 00:43:27,514
We also want the redeemability idea.

481
00:43:28,514 --> 00:43:37,514
So this is like you try to respond on the redeemability side and then people will
switch axes and they will complain about how it's less ignorable or something as a
result.

482
00:43:38,514 --> 00:43:51,514
But yeah, I think the way that – I think what people have to keep in mind is that
the – well, I don't know like how to explain it like in what order.

483
00:43:52,514 --> 00:44:00,514
But what really happens is that the – I don't know like what were you actually
asking about?

484
00:44:01,514 --> 00:44:07,514
So a lot of people perceive sidechains as being significantly less secure.

485
00:44:08,514 --> 00:44:12,514
So people are concerned that miners – people are concerned that miners are –

486
00:44:13,514 --> 00:44:14,514
Yes, I understand.

487
00:44:15,514 --> 00:44:16,514
Yeah, so I just want to say –

488
00:44:17,514 --> 00:44:18,514
Sorry, it's just such a broad thing.

489
00:44:19,514 --> 00:44:26,514
So the first point is to bring up is that it shouldn't really be up to random other
people what you do with your own money.

490
00:44:26,514 --> 00:44:35,514
So the second axis is the less important of the two axes intentionally because if –
the first axis affects everyone else.

491
00:44:36,514 --> 00:44:42,514
So if what you do – if the sidechain isn't ignorable, then the whole network is
along for the ride.

492
00:44:43,514 --> 00:44:51,514
But if you lose your own money by putting it in a dumb sidechain, then that's kind
of like your fault and you should be – the sidechain is going to have some benefits.

493
00:44:52,514 --> 00:44:53,514
It's going to be doing something new.

494
00:44:53,514 --> 00:44:57,514
So it's going to have some benefits or at the very least some perceived benefits.

495
00:44:58,514 --> 00:45:00,514
And so each user should be able to decide.

496
00:45:01,514 --> 00:45:05,514
They should have the option to say, well, these benefits are sort of worth the cost.

497
00:45:06,514 --> 00:45:13,514
That also – you see, this is why the question is kind of bizarre to answer because
it has many kind of layers.

498
00:45:14,514 --> 00:45:15,514
So the first layer is these two axes.

499
00:45:15,514 --> 00:45:31,514
The second layer is that the second axis is almost – it's almost a 100% free pass on
my part because no matter how irredeemable the coins are, it's still possible that
some user would decide that it's in their best interest to hazard the risk.

500
00:45:32,514 --> 00:45:36,514
So users should be free to choose, like freedom and all of that stuff.

501
00:45:37,514 --> 00:45:50,514
If the user can send their money to Bitcoin Eater, don't send, and they can destroy
their Bitcoin or if they can deposit to a really dumb website or if they can invest
it in some dumb scam, they should be able to try anything they want to try.

502
00:45:51,514 --> 00:45:55,514
And the only way we know that the idea is bad is after the fact for sure.

503
00:45:56,514 --> 00:45:57,514
But before the fact, we don't know.

504
00:45:58,514 --> 00:46:00,514
There's always a possibility that the idea actually will be really, really great.

505
00:46:01,514 --> 00:46:02,514
So that's kind of the second layer.

506
00:46:03,514 --> 00:46:07,514
First layer is these two axes, which are mutually exclusive.

507
00:46:08,514 --> 00:46:12,514
But the second layer is that the second axis, there's almost no grounds for
criticizing it.

508
00:46:13,514 --> 00:46:20,514
Other than to say it's so insecure that it's going to be completely useless and that
– so whether or not we add it is kind of superfluous, so why are we wasting our time
talking about it?

509
00:46:21,514 --> 00:46:23,514
But other than that, you get kind of a free pass on the second axis.

510
00:46:24,514 --> 00:46:45,514
But the third answer is – the third layer is to explain exactly why I think the
withdrawals that are broadcast from the sidechain will match the withdrawals that
are processed by the mainchain, even though there is no connection between the two
pieces of software because it's the first criterion that the sidechain be ignorable.

511
00:46:46,514 --> 00:46:48,514
So why do I think that they will match?

512
00:46:49,514 --> 00:46:56,514
And the answer is that per sidechain, withdrawals are forced to be very, very robust
case scenario.

513
00:46:57,514 --> 00:46:59,514
There's one withdrawal per sidechain.

514
00:47:00,514 --> 00:47:01,514
This is all per sidechain.

515
00:47:02,514 --> 00:47:05,514
One withdrawal every three months, every 13,150 blocks.

516
00:47:06,514 --> 00:47:15,514
So the withdrawals are very, very infrequent and the withdrawals are proposed first
in a phase where – well, it's a continuous operation.

517
00:47:16,514 --> 00:47:20,514
But when the withdrawals are proposed, only one of them can be advanced at a time.

518
00:47:21,514 --> 00:47:33,514
So it's kind of as though there's a bunch of trains leaving New York City and
they're headed to Los Angeles or I don't know, a bunch of airplanes leaving New York
City traveling to – I don't know.

519
00:47:34,514 --> 00:47:38,514
I don't know what would be a better one for airplanes, but I hate to make it too
US-centric.

520
00:47:39,514 --> 00:47:44,514
But New York City is in the northeast and Los Angeles is in the southwest of the
United States.

521
00:47:45,514 --> 00:47:46,514
So these trains are leaving New York City.

522
00:47:47,514 --> 00:47:58,514
They're traveling southwest and it's about 3,000 miles or so, which is what is that
like – I don't know, 5,000 kilometers or something, whatever it is.

523
00:47:59,514 --> 00:48:04,514
And the trains are leaving, but the way it's set up is that the trains are proposed.

524
00:48:05,514 --> 00:48:09,514
We're going to say we're going to take all this money and give it – that's some of
the money that's in the sidechain.

525
00:48:10,514 --> 00:48:11,514
We're going to give it to the owners.

526
00:48:12,514 --> 00:48:21,514
It's new owners on the main chain and only a few of those – or excuse me, only one
of those trains can advance at a time and reach its destination in Los Angeles.

527
00:48:21,514 --> 00:48:26,514
So the total number of successful withdrawals, possible withdrawals is only one.

528
00:48:27,514 --> 00:48:35,514
It's a very, very slow rate of one per three months, and you get to see it coming
from a mile away, and you see it coming in a very fixed state.

529
00:48:36,514 --> 00:48:49,514
So you're sitting there in Los Angeles and you say, well, the sidechain has 200
bitcoin in it, and they want to take 13 of those and give it to X and one of those
and give it to Y and half of one and give it to Z.

530
00:48:49,514 --> 00:48:55,514
And there's a train that has that information, and it's slowly, slowly making its
way across the country.

531
00:48:56,514 --> 00:48:59,514
And so you have – it's very, very easy to see.

532
00:49:00,514 --> 00:49:02,514
You mentioned earlier the responsiveness of the community to bugs and things.

533
00:49:03,514 --> 00:49:10,514
This is like – people will know immediately when the invalid withdrawal is proposed,
all the sidechain users and really the entire internet.

534
00:49:11,514 --> 00:49:12,514
All humans will know.

535
00:49:13,514 --> 00:49:14,514
How do they check?

536
00:49:15,514 --> 00:49:16,514
How would they check it though?

537
00:49:16,514 --> 00:49:19,514
Oh, of course. In the sidechain, it's literally going to be in the headers.

538
00:49:20,514 --> 00:49:26,514
So it's going to be in every header, whichever – there will be only like one or two
hashes.

539
00:49:27,514 --> 00:49:33,514
It will be a hash of the actual transaction that you want to put into the mainchain
bitcoin.

540
00:49:34,514 --> 00:49:36,514
So it's not computationally intensive. It's just a simple check of the hash.

541
00:49:37,514 --> 00:49:42,514
Yes. So when it's introduced on the mainchain, the hash is placed into a coinbase
transaction so that everyone can see it.

542
00:49:43,514 --> 00:49:45,514
So it's just a 256-bit hash.

543
00:49:46,514 --> 00:49:54,514
This is 32 – it's this thing that you're used to seeing where it's just a kind of
relatively short string of alphanumeric characters.

544
00:49:56,073 --> 00:49:59,073
From a user perspective, it will show up in the mainchain.

545
00:49:59,073 --> 00:50:05,073
It's not the mainchain full node guys that have to do those checks. It's the
sidechain guys, right?

546
00:50:05,073 --> 00:50:17,073
Yes. Well, the mainchain, it will be valid no matter what it is. So no matter which
train reaches, if the train reaches LA and it says pay all the money out of the
sidechain to this one address,

547
00:50:17,073 --> 00:50:21,073
then that will be what is considered valid in the Drivechain rules.

548
00:50:21,073 --> 00:50:29,073
So another issue in explaining Drivechain – there are many – but another kind of
hangup is that there are two – the word validation is kind of like double used.

549
00:50:29,073 --> 00:50:35,073
So there's the rules of Drivechain, which are this train thing that I'm trying to
describe.

550
00:50:35,073 --> 00:50:45,073
But then there's also is the train that arrived in Los Angeles – is that actually
the correct one that all the sidechain nodes know to be the true blockchain history?

551
00:50:45,073 --> 00:50:51,073
And the mainchain can never know that while the sidechain is ignorable, which is the
first criterion.

552
00:50:51,073 --> 00:50:59,073
But what if a miner says, okay, all these users are saying this is a fake
withdrawal, but fuck you, I'm going to include it anyway and put it on the
mainchain. What happens next?

553
00:50:59,073 --> 00:51:11,073
Yeah, well, there is – so there are several lines of defense that the user can
attempt to make along the second axis if they so choose to try and make the
redeemability better.

554
00:51:11,073 --> 00:51:25,073
So one is that they can just – it's very easy to just kind of – it's basically going
to be like a right-click and then left-click operation to just kind of flag an
individual train when it's going through like Kansas City or Arizona or something.

555
00:51:25,073 --> 00:51:33,073
Those are cities in the middle of the United States. Again, sorry for anyone who is
living in Europe and doesn't – not familiar with U.S. geography or something.

556
00:51:33,073 --> 00:51:51,073
But the train has been moving across. After a few months, you realize that this
train really is trying to make it to LA, even though it probably – even though it's
wrong and there's a lot of barriers in its path that someone is trying to steal
money from the sidechain.

557
00:51:51,073 --> 00:52:10,073
The wrong train is trying to make it to LA, and you can just say, well, that train –
we UASF so that the train can never actually dock in LA. It just has to wait outside
the city until it expires, and they expire after six months, after 26,000 blocks.

558
00:52:10,073 --> 00:52:26,073
But what does that look like for – so let's pretend this is happening. Jihan is
sabotaging some kind of SegWit sidechain because he's an evil motherfucker and he
generates some fake withdrawal. Someone is trying to scam the sidechain and pull
money out, so he includes it anyway.

559
00:52:26,073 --> 00:52:35,073
So how many of us need to – UASF – how many of us need to actually be running a
specific altered node implementation that censors essentially his fake withdrawal?

560
00:52:35,073 --> 00:52:48,073
Well, it depends on innumerable factors, of course, because it's quite a complicated
situation overall. But the important thing to mention is what I was talking about
before, which is that what we really want is we just want ways to recover from
mistakes.

561
00:52:49,073 --> 00:53:00,073
And we want low-cost kind of solutions that don't harm users if they fail. So it
should be obvious. Actually, I don't think we really explained.

562
00:53:00,073 --> 00:53:09,073
So what miners can do on the mainchain is they can move the train one space forward
each block, and that is the SPV proof in Drivechain.

563
00:53:09,073 --> 00:53:19,073
In other setups, you're just literally dumping all the sidechain's headers in. That
was Blockstream's original idea from Appendix B.

564
00:53:19,073 --> 00:53:30,073
You're literally putting a bunch of sidechain headers in the block to try to
literally prove the work. But in Drivechain, you just allow the mainchain miners to
take some train, any train they want.

565
00:53:30,073 --> 00:53:44,073
They could do it without looking at any trains or none of the trains. They can move
it forward one space and then all the other trains of that sidechain type, any other
withdrawals that have been attempted, automatically move back a space.

566
00:53:44,073 --> 00:53:51,073
Or the miners can just abstain and do nothing, or they can actually hit a panic
button, which moves all the trains back by one space.

567
00:53:51,073 --> 00:54:03,073
So the train is not even going to be able to reach LA in time unless 51% of the hash
rate is determined to make the withdrawal happen.

568
00:54:03,073 --> 00:54:07,073
So that's the first line of defense is that it's just kind of difficult to withdraw
from the sidechain.

569
00:54:07,073 --> 00:54:12,073
Sidechain withdrawals are just infrequent and slow, and they can probably fail some
of the time.

570
00:54:12,073 --> 00:54:29,073
So this is kind of a long slog. You're kind of running down a very, very long
hallway, and you have to really hope you get to the end of the hallway before
someone locks the door on the other side or before the timer runs out and you have
to start over.

571
00:54:29,073 --> 00:54:35,073
But how significant of an issue does that pose for the UX in terms of you need
liquidity to move on and off the sidechain, right?

572
00:54:35,073 --> 00:54:42,073
Yeah, so I'll get to that. So these issues are all related. I'll get to that in a
second. Didn't we just bring up Andreas, though? He made that site.

573
00:54:42,073 --> 00:54:43,073
Yeah, yeah.

574
00:54:43,073 --> 00:54:57,073
So yeah, the design calls for a lot of things to happen in code, but it calls for
one thing to happen in the world of business, which is that it calls for someone to
create a kind of website and be a kind of investment banker, like broker type
person.

575
00:54:58,073 --> 00:55:12,073
So you have a bunch of sidechain coins, and if you want them out immediately, you
just kind of do basically – Andreas has named it very well. He's naming it
SideShift, so it's kind of a pun on ShapeShift.

576
00:55:12,073 --> 00:55:13,073
But it has to be profitable, right?

577
00:55:13,073 --> 00:55:26,073
Right. So the guy's going to charge kind of like a 1% cut or whatever it is. But
since anyone can do it, anyone who has mainchain funds can do it, it should be very
competitive, and it should just be time value of money for three months, which is…

578
00:55:27,073 --> 00:55:33,073
But there's other risks associated that the investment banker type person has to
account for in terms of the risk that they are taking.

579
00:55:33,073 --> 00:55:34,073
Correct.

580
00:55:34,073 --> 00:55:39,073
You just said that some withdrawals may fail. Does that include legitimate
withdrawals?

581
00:55:39,073 --> 00:55:40,073
Yes, it does.

582
00:55:40,073 --> 00:55:51,073
So maybe they're trying to close an ARB. Yeah, exactly. So if they're trying to
close an ARB, and then oh shit, the withdrawal failed, but it was totally legit,
then they're fucked in terms of the funds that they already expended for the
middleman, and suddenly they're wiped out.

583
00:55:52,073 --> 00:56:00,073
No, no, no. Let me explain. I don't think I maybe explained it right. So what you do
is you swap your sidechain. You have 100 sidechain coins.

584
00:56:01,073 --> 00:56:14,073
You swap them, not using the slow Drivechain process at all, using regular Bitcoin
transfers. You swap 100 sidechain coins for 99 mainchain coins.

585
00:56:14,073 --> 00:56:31,073
So 100 sidechain coins go from you to Andreas, and then 99 of his separately in the
mainchain world, but possibly connected by cryptography, the hash time lock
contracts, these other things, or possibly just connected to the website and its
brand or whatever.

586
00:56:31,073 --> 00:56:46,073
That's a simpler first version. But 100 of your sidechain coins go to Andreas, 99 of
Andreas's mainchain coins go to you. Those are two separate things and two separate
networks. They may be connected through cryptography or through the brand or the
website or these other things.

587
00:56:46,073 --> 00:57:00,073
And so then Andreas is going to end up with 100 sidechain coins, and then he's going
to collect these from all these different people. He's going to pack them all into
one UTXO on the side. This is all happening on the sidechain. This has nothing to do
with Drivechain. This could happen with altcoins right now so far.

588
00:57:00,073 --> 00:57:27,073
He's going to pack all of these up into one package that has like 30,000 sidechain
BTC, and then he's going to put that on the next train. So then whatever, like maybe
three and a half months, if the next train's leaving in a half a month, he's going
to put that stuff on the train, and then he's going to pick it up three months after
that in Los Angeles when it shows up.

589
00:57:27,073 --> 00:57:46,073
So does that make sense? What he's risking is the possibility that it will never
ever succeed or that it will succeed so late, like years down the road, that his
time value of money has wiped out his capital, his return of capital employed.

590
00:57:46,073 --> 00:58:00,073
Or he's also risking that there will be some kind of problem with the sidechain that
it will just completely fail to operate for some reason, and so he can never prove
that the funds are his, even though that's also pretty unlikely and recoverable.

591
00:58:17,073 --> 00:58:33,073
In that one scenario, I would really not want to be Andreas, because if someone can
figure out how to do inflation, they crack that open, and then they send Andreas all
this money that they've inflated, and Andreas gives his mainchain money, and then
Andreas starts to withdraw.

592
00:58:33,073 --> 00:58:51,073
But then it's like, oh no, now we've got a big problem because it looks like lots of
money is trying to get on this train, but there's only so much can exit on the other
side, and then it's realized that something has gone wrong and that it's going to
get pennies on the dollar or nothing.

593
00:58:51,073 --> 00:59:00,073
So that's the one thing where I would really not want to be the guy operating the
kind of investment banker liquidity thing.

594
00:59:00,073 --> 00:59:08,073
See, this is what inevitably happens in this conversation, is that I start talking
about security, and then it immediately goes to won't that be too slow to be useful.

595
00:59:08,073 --> 00:59:16,073
And so I don't know if we want to return – sometimes we forget to return to the
security dimension, or we can just go wherever the conversation wants to go.

596
00:59:16,073 --> 00:59:26,073
Well, so yeah, I mean a lot of – at least I personally, one of my big ideas, use
cases with sidechains was privacy aspects.

597
00:59:26,073 --> 00:59:40,073
So ZK, Starks, sort of now they've done it in Zcash, but – and also like ring
signatures, sort of Monero-type sidechain, and Mimblewimble, so like a Grin
sidechain type thing.

598
00:59:40,073 --> 00:59:48,073
Oh, absolutely. There's no problem with like privacy in general, but there's certain
types of privacy that risk the inflation thing.

599
00:59:48,073 --> 00:59:57,073
As you may have – this is discussed recently, and if that is ever possible, then I
would be very uneasy.

600
00:59:57,073 --> 01:00:13,073
But as long as something like ring signatures and Monero, that is an example of
enhanced privacy, but where I would think I would still be very comfortable in
Andreas' role, I would be like – because I really think that this is done correctly,
and they have these proofs too.

601
01:00:13,073 --> 01:00:16,073
Sidechain just doesn't work if there's inflation.

602
01:00:17,073 --> 01:00:24,073
Yeah, and so I think there's a concept here that some people might be now skeptical
and all this, but I think people misunderstand.

603
01:00:24,073 --> 01:00:28,073
In crypto in general, everything is a gamble in a way, right?

604
01:00:28,073 --> 01:00:38,073
So you have to always – that's why you require a certain number of confirmations,
and we essentially rely upon – smart people are validating their own transactions.

605
01:00:38,073 --> 01:00:40,073
They run in their own node, but there's also miners out there.

606
01:00:41,073 --> 01:00:42,073
A lot of people are just trusting things.

607
01:00:42,073 --> 01:00:50,073
They're running SPV implementations, and they're just trusting that the miners are
behaving and their node that they're connected to is behaving and such.

608
01:00:50,073 --> 01:01:03,073
And so a lot of people are concerned that some of the trustlessness or the security
is lost because there's this SPV proofing.

609
01:01:04,073 --> 01:01:09,073
Can you kind of discuss what – like –

610
01:01:09,073 --> 01:01:10,073
Yes.

611
01:01:10,073 --> 01:01:11,073
No, I can.

612
01:01:11,073 --> 01:01:13,073
I can very briefly because it's these two axes again.

613
01:01:13,073 --> 01:01:24,073
So the only way you can have it be perfectly secure is to have Bitcoin enforce all
of the rules, but then you've just de facto evil forked or hard forked Bitcoin.

614
01:01:24,073 --> 01:01:26,073
It depends on your definition for these words.

615
01:01:26,073 --> 01:01:32,073
Unfortunately, many different people use many different definitions for hard fork,
soft fork, et cetera.

616
01:01:32,073 --> 01:01:45,073
But it's generally agreed that an evil fork is something where the user is kind of
more or less forced to upgrade against their will in something that is more
burdensome computationally for their computer.

617
01:01:45,073 --> 01:01:48,073
So that's kind of one of the more stable definitions.

618
01:01:48,073 --> 01:01:53,073
Unfortunately, the definitions are really messed up, but that's what you don't want
sidechains to be.

619
01:01:54,073 --> 01:02:04,073
You want sidechains to be like optional plug-ins, like a Firefox plug-in or
something, and it doesn't matter to you if your neighbor downloads some stupid
plug-in and blows up his computer.

620
01:02:04,073 --> 01:02:07,073
Then that doesn't really matter for you.

621
01:02:07,073 --> 01:02:18,073
So that's the thing is that it is worse, but all the criterions in that direction –
again, as I say, this is that sort of second layer I was talking about, the first
layer being those two axes.

622
01:02:19,073 --> 01:02:23,073
The second layer is that you kind of get a free pass on all of that because you have
consumer sovereignty.

623
01:02:23,073 --> 01:02:26,073
A consumer gets to decide, you know what? I don't know.

624
01:02:26,073 --> 01:02:30,073
This sidechain looks like not a lot of people use it, and the developers look shady.

625
01:02:30,073 --> 01:02:32,073
Software is buggy.

626
01:02:32,073 --> 01:02:37,073
So okay, it's kind of like you swim at your own risk, no lifeguard on duty or
whatever.

627
01:02:37,073 --> 01:02:44,073
But then on the other hand, if you have a sidechain that is very, very, very
popular, tons and tons of people are using it.

628
01:02:44,073 --> 01:02:47,073
And this is very similar if you try to choose what other software you run.

629
01:02:47,073 --> 01:02:49,073
What kind of operating system do you want to run?

630
01:02:49,073 --> 01:02:54,073
You really want to run one that lots of people are looking at and lots of people are
working on.

631
01:02:54,073 --> 01:02:56,073
This is big network effects and security.

632
01:02:56,073 --> 01:02:58,073
Or maybe you don't.

633
01:02:58,073 --> 01:03:06,073
People used to argue that Apple would never get viruses, macOS, because it was not
as popular.

634
01:03:06,073 --> 01:03:10,073
So you have all these different things, and so the security is really a kind of –

635
01:03:11,073 --> 01:03:12,073
Well, let's try to –

636
01:03:12,073 --> 01:03:13,073
Yeah, okay.

637
01:03:13,073 --> 01:03:20,073
So I wanted to sort of poke you into the direction of – because you've used this
concept of betting on the profitable mining strategy.

638
01:03:20,073 --> 01:03:25,073
And how this plays into the overall tradeoffs that are going on in the crypto space.

639
01:03:25,073 --> 01:03:31,073
Like in general, miners can do all kinds of horrible things, but you're betting that
they're going to be greedy and they're going to act in their own interest.

640
01:03:31,073 --> 01:03:33,073
And that's kind of what holds the whole system together.

641
01:03:33,073 --> 01:03:38,073
And as you go further along that spectrum, maybe you can talk about it in that
context.

642
01:03:38,073 --> 01:03:39,073
Yes.

643
01:03:39,073 --> 01:03:48,073
Well, I think this is the security context that we had to depart momentarily to talk
about why it wouldn't be totally useless to only allow one withdrawal every three
months.

644
01:03:48,073 --> 01:03:57,073
So the reason that – so we just discussed that actually from the user's point of
view, it would be more or less instant.

645
01:03:57,073 --> 01:04:02,073
And only specialists will be walking these coins back slowly over the three months.

646
01:04:02,073 --> 01:04:17,073
So now the question is, do you really think that miners will do the wrong thing and
have the real hash as reported by the sidechain's headers, real hash won't match the
hash that is placed –

647
01:04:17,073 --> 01:04:22,073
that is materialized in New York City and that it is marching towards Los Angeles?

648
01:04:23,073 --> 01:04:42,073
And the answer to that is just that if miners are really willing to do kind of
flagrant things just to get some more bitcoin for themselves today at the expense of
kind of interfering with the network, they have always had many other opportunities
to do that.

649
01:04:42,073 --> 01:04:46,073
And throughout bitcoin's history, they can – the most obvious is just to reorg the
chain.

650
01:04:46,073 --> 01:04:56,073
So miner could separate of his 100-block maturity period for the coins that he
generates, miner could just buy a bunch of coins from Coinbase.

651
01:04:56,073 --> 01:04:58,073
Jian Wu could take – it's funny.

652
01:04:58,073 --> 01:05:01,073
It's better with names, right, because it's a little bit easier to understand.

653
01:05:01,073 --> 01:05:02,073
It's more fun.

654
01:05:02,073 --> 01:05:05,073
So Jian Wu shows up with like $100 million.

655
01:05:05,073 --> 01:05:07,073
He goes to Coinbase.

656
01:05:07,073 --> 01:05:09,073
He gives $100 million to Brian Armstrong.

657
01:05:09,073 --> 01:05:11,073
Brian Armstrong gives him whatever, bitcoin.

658
01:05:11,073 --> 01:05:13,073
I don't know what it is.

659
01:05:13,073 --> 01:05:15,073
I don't even know, like 5,000 bitcoin, whatever.

660
01:05:15,073 --> 01:05:20,073
And then Jian Wu just – he deposits this bitcoin to his computer.

661
01:05:20,073 --> 01:05:27,073
Then he sends it back to Coinbase and sells them for – he deposits them to Coinbase
and waits six confirmations.

662
01:05:27,073 --> 01:05:32,073
And then Brian Armstrong credits him with the thing and then he re-withdraws the
$100 million.

663
01:05:33,073 --> 01:05:37,073
And once that's set up, you'd think we're back where we were before.

664
01:05:37,073 --> 01:05:47,073
But of course we're not because Jian Wu now calls up whatever, his conciliary or
whatever it is, running all the mining equipment.

665
01:05:47,073 --> 01:05:51,073
And he says, roll back the chain seven blocks ago.

666
01:05:51,073 --> 01:06:01,073
He rolls back the chain seven blocks ago and bang, the deposit to Coinbase now isn't
in the longest chain or the heaviest chain.

667
01:06:01,073 --> 01:06:03,073
He uses different terminology.

668
01:06:03,073 --> 01:06:05,073
Again, terminology is ruined in bitcoin.

669
01:06:05,073 --> 01:06:07,073
But the point is he can just do that.

670
01:06:07,073 --> 01:06:11,073
If you've got more than 50% of the hash rate, you can do that.

671
01:06:11,073 --> 01:06:19,073
So SegWit2x showed that you could have a reasonably organized 83% of the hash rate.

672
01:06:19,073 --> 01:06:25,073
And of course it becomes much easier the more hash rate you have and the fewer
blocks that you need to re-org.

673
01:06:25,073 --> 01:06:31,073
And so yeah, if you can re-org seven blocks, then you can beat the system and just
re-withdraw that money.

674
01:06:31,073 --> 01:06:36,073
Then Jian Wu's got $100 million and he's got his 5,000 bitcoin back.

675
01:06:36,073 --> 01:06:43,073
So he just double-spent basically against Brian Armstrong and emptied Coinbase's
bank account.

676
01:06:43,073 --> 01:06:45,073
So that's something that's already possible.

677
01:06:45,073 --> 01:06:48,073
And so this is just to try to expand the scale.

678
01:06:48,073 --> 01:06:53,073
You mentioned earlier that the community is very responsive to bugs, which is true.

679
01:06:53,073 --> 01:06:56,073
And I even have that on drivechain.info and there's little memes there.

680
01:06:56,073 --> 01:07:02,073
I talk about how we had the inflation bug a long time ago that was fixed in a few
hours.

681
01:07:02,073 --> 01:07:13,073
We had the March 2013 chain fork, which had very few – that was a very brief time
before that resolved itself kind of more or less automatically.

682
01:07:14,073 --> 01:07:20,073
So we had the July 2015 thing that resolved itself completely automatically.

683
01:07:20,073 --> 01:07:30,073
So we've seen that there is some resource here where people will kind of – there's a
kind of – people will monitor the network.

684
01:07:30,073 --> 01:07:36,073
You can't exploit this resource in its raw form because it will just be exhausted at
scale.

685
01:07:36,073 --> 01:07:41,073
But that's what this technology tries to do is just flattens it so it takes at least
three.

686
01:07:41,073 --> 01:07:47,073
And if you only have a minority, if you only have 51 percent – excuse me, not a
minority.

687
01:07:47,073 --> 01:07:50,073
I mean a very bare, slim majority.

688
01:07:50,073 --> 01:07:59,073
If you have only 51 percent, then it's really going to take six months actually
because you can only move the train forward once every – it will be finding so many
blocks.

689
01:07:59,073 --> 01:08:03,073
This train will be moving forward twice as slowly as it otherwise would.

690
01:08:03,073 --> 01:08:05,073
So it will just barely make it to LA in time.

691
01:08:05,073 --> 01:08:07,073
So it will take even longer if you have less hash rate.

692
01:08:08,073 --> 01:08:19,073
But yeah, we have this – so this notion of that the community can fix problems
within ten blocks very, very reliably, fix severe problems with no warning.

693
01:08:19,073 --> 01:08:29,073
But then we just set this up so that there's plenty of warning, a huge amount of
warning, and there's also no disagreement over what should be done, if anything.

694
01:08:29,073 --> 01:08:36,073
There might be disagreement over what to do, but it would be like do we right click
on this and UASF to block it or not.

695
01:08:36,073 --> 01:08:38,073
So that would be very, very, very simple.

696
01:08:38,073 --> 01:08:46,073
So it's an entirely different character in my view because we get – it can only
happen – it can't happen like over and over again.

697
01:08:46,073 --> 01:08:51,073
It can only happen once at a time per sidechain I suppose, but it can only happen
once at a time.

698
01:08:51,073 --> 01:08:58,073
It can only happen once per three to six months, and you're warned about it in
advance, like three to six months in advance.

699
01:08:58,073 --> 01:09:00,073
And you know exactly what to do about it.

700
01:09:01,073 --> 01:09:05,073
or to have a debate over what solution to the problem is.

701
01:09:05,073 --> 01:09:14,073
But it gets even better than that, which is that if you have 50% and you try to do
this – you try to ram this thing down.

702
01:09:14,073 --> 01:09:18,073
This is funny because I didn't actually answer your – I have to talk about your
point about profitability still.

703
01:09:18,073 --> 01:09:29,073
But if you try to ram this invalid transaction through and the people UASF to block
it and they fail, not enough people – you asked this question earlier.

704
01:09:29,073 --> 01:09:32,073
Not enough people UASF, so it's a very weak UASF.

705
01:09:32,073 --> 01:09:34,073
But then nothing happens to those people.

706
01:09:34,073 --> 01:09:40,073
Those people just – they live on a blockchain that's kind of paused and they don't
see any new blocks coming in,

707
01:09:40,073 --> 01:09:43,073
or they only see an invalid block and they see other invalid blocks.

708
01:09:43,073 --> 01:09:46,073
So all those people have to do is kind of wait.

709
01:09:46,073 --> 01:09:50,073
They don't have to do this collectively or together, but they just have to wait and
try to stick it out.

710
01:09:50,073 --> 01:09:56,073
And then one by one, they'll just decide, oh, I guess the UASF failed, and they
right-click and they undo the UASF.

711
01:09:56,073 --> 01:10:02,073
And then they're back on the history where the evil train docked in Los Angeles.

712
01:10:02,073 --> 01:10:05,073
So you kind of lose nothing by attempting a UASF really.

713
01:10:05,073 --> 01:10:07,073
If you fail, you just have to give up.

714
01:10:07,073 --> 01:10:12,073
But then it's not like your money is – it's not like you have any –

715
01:10:12,073 --> 01:10:17,073
But what if you're the one who is getting the money stolen in the same –

716
01:10:17,073 --> 01:10:24,073
Yeah, well, then you're hoping the UASF succeeds, but you – then obviously if it
doesn't succeed, then your money is stolen.

717
01:10:24,073 --> 01:10:27,073
But it can, of course, be reversed, like what if legitimate sidechain users –

718
01:10:27,073 --> 01:10:32,073
poor Andreas is trying to just get his money back that he bought for $0.99 on the
dollar.

719
01:10:32,073 --> 01:10:37,073
He doesn't want frivolous UASFs to block his trains that are real.

720
01:10:37,073 --> 01:10:44,073
But yeah, so that's a nice thing, but I didn't actually answer your question, which
is that I assume there's –

721
01:10:44,073 --> 01:10:48,073
it's a very reliable assumption that – see, as I explained earlier,

722
01:10:48,073 --> 01:10:53,073
I think almost all of these sidechains in practice will have to be merged amounts,

723
01:10:53,073 --> 01:10:59,073
and they may fall unpredictably or may even fall as a result of a self-fulfilling
prophecy that could cause each other.

724
01:10:59,073 --> 01:11:01,073
People might be like, oh, this sidechain is going to die.

725
01:11:01,073 --> 01:11:04,073
I'm not going to transact on it because the hash rate is too low.

726
01:11:04,073 --> 01:11:07,073
That could cause the hash rate to fall in the absence of merge mining.

727
01:11:07,073 --> 01:11:11,073
So I just think the merge mining deal is too good, too sweet to pass up.

728
01:11:11,073 --> 01:11:14,073
Everyone is going to take it because they get amazing hash rate all the time,

729
01:11:14,073 --> 01:11:18,073
and it's just easier to just leave it on, set it, and forget it.

730
01:11:18,073 --> 01:11:24,073
The chances that transaction fee volumes fall towards zero probably still have very
high hash rate.

731
01:11:24,073 --> 01:11:30,073
So I think everyone will take that deal, which means that the sidechain miners and
the mainchain miners will all be the same people.

732
01:11:30,073 --> 01:11:36,073
And you just have to ask yourself – the purchasing power of each Bitcoin,

733
01:11:36,073 --> 01:11:42,073
which is to say the price, the $5,000, what are we, $6,500 today or these days or
whatever?

734
01:11:43,073 --> 01:11:53,073
That is the most significant driver of miner profits because they buy all their
stuff in US dollars and then they're paid in Bitcoin.

735
01:11:53,073 --> 01:12:04,073
So there's a big multiple reason that they would at least not overtly in such a
flagrant way blow up the possibility of sidechains being useful on the Bitcoin
network

736
01:12:04,073 --> 01:12:10,073
by just kind of stealing them all in this very flagrant and slow, kind of
unforgivable,

737
01:12:10,073 --> 01:12:16,073
and kind of blowing up the whole infrastructure of sidechains.

738
01:12:16,073 --> 01:12:19,073
Because if you steal from one, it kind of makes it calls into question like,

739
01:12:19,073 --> 01:12:23,073
is anything in this infrastructure, in the sidechains infrastructure,

740
01:12:23,073 --> 01:12:29,073
value their existence and enforce the integrity of them in the sense that we need
to,

741
01:12:30,073 --> 01:12:36,073
many of us need to then be on the UASF train if a miner like Jihan was trying to
pull some kind of shit like that.

742
01:12:36,073 --> 01:12:39,073
Because then if one sidechain is fucked, nobody's going to trust the other ones.

743
01:12:39,073 --> 01:12:49,073
But I wanted to highlight because Stop and Decrypt, our Bitcoin moderator, is making
a point on the point regarding the mining.

744
01:12:49,073 --> 01:12:53,073
So he's in the chat. I don't know if you're able to see the chat there.

745
01:12:53,073 --> 01:12:59,073
But he was mentioning on this point about comparing the incentive to cheat for a
miner on the mainchain versus on a sidechain.

746
01:12:59,073 --> 01:13:06,073
He's making the point here that the more people that are using the sidechain, the
more profitable it is to attack.

747
01:13:06,073 --> 01:13:11,073
So the risk is not the same as the mainchain, because as the liquidity on the
sidechain grows,

748
01:13:11,073 --> 01:13:16,073
the incentive might exceed the analog for doing this kind of attack on the
mainchain.

749
01:13:16,073 --> 01:13:21,073
Can you comment a bit about more like how this risk is compared versus sidechain?

750
01:13:22,073 --> 01:13:26,073
Yes. Well, that's what you want. You want the benefits and the costs to scale
together.

751
01:13:26,073 --> 01:13:29,073
You don't want them to fall out of whack.

752
01:13:29,073 --> 01:13:33,073
So as the sidechain becomes more popular, there's more Bitcoin deposited to it,

753
01:13:33,073 --> 01:13:37,073
which means that there's more Bitcoin that you can steal.

754
01:13:37,073 --> 01:13:41,073
But it also means that the transaction fees on that sidechain will be going up.

755
01:13:41,073 --> 01:13:51,073
And it also means that a theft from that sidechain is much more likely to indicate
an end of like the sidechain experiment on Bitcoin.

756
01:13:51,073 --> 01:13:56,073
And I really think that in this world, I don't know exactly how everything will
shake out,

757
01:13:56,073 --> 01:14:03,073
but in a world where, like, for example, the Bitcoin dominance index is down into
like the 50s,

758
01:14:03,073 --> 01:14:08,073
when it used to be like 95, you know, all the time, and you wouldn't even talk about
it because they're so high.

759
01:14:09,073 --> 01:14:16,073
I think that, you know, we have a lot of potential innovation going on,

760
01:14:16,073 --> 01:14:22,073
especially the nature of the consensus software, which is that it mandates total
agreement down to the last byte.

761
01:14:22,073 --> 01:14:26,073
So what I'm trying to say is basically that sidechain tech is valuable,

762
01:14:26,073 --> 01:14:35,073
and that a sidechain enabled token is going to be more valuable than a token that
cannot travel among different chains.

763
01:14:35,073 --> 01:14:39,073
And so I really think that as the sidechain becomes more popular,

764
01:14:39,073 --> 01:14:45,073
attacking it would be seen as much more of an indicator that the whole sidechains
experiment is dead.

765
01:14:45,073 --> 01:14:52,073
I would actually be much more worried about the unpopular sidechains and trying to
convince everyone that, you know,

766
01:14:52,796 --> 01:15:00,796
to care about your thing, which you probably would not, and then your redeemability
would fall apart. But you still have that first access, the unpopular sidechain.

767
01:15:00,796 --> 01:15:07,796
I think the other thing I would mention is that as the incentive to attack grows, so
too does the incentive for users to hold that accountable, right?

768
01:15:07,796 --> 01:15:08,796
Yes, exactly.

769
01:15:08,796 --> 01:15:27,796
So if the pot is so valuable that Jihan's like, holy shit, I'm going to put my head
in there and take some money, then there's some guys who have a lot at stake as well
on the other side who are playing against him who will say, all right, I need to run
UASF, I need to inform people about this and get people in the community to care so
that these funds are protected.

770
01:15:28,796 --> 01:15:43,796
Yes, there is another analogy to what I was talking about before where instead of
Jihan reorging seven blocks, he can make even more money. It's just mathematical
certainty by reorging even more transactions. He can go back 70 blocks or 700
blocks.

771
01:15:44,796 --> 01:16:00,796
At a certain point, he can – so this is a little complicated to explain, but even if
he doesn't control the money, he can double spend – a 51% minor coalition can double
spend money that they don't have.

772
01:16:00,796 --> 01:16:13,796
So all they have to do is reorg the chain and then not allow the old transactions in
unless there is – unless they receive a follow-on transaction that gives most of the
money to them, 99.9% or whatever.

773
01:16:14,796 --> 01:16:21,796
They can say either you're giving it to me – well, I don't want to get it too bogged
down, but I presented on this in Lisbon.

774
01:16:22,796 --> 01:16:33,796
Long story short is any Bitcoin transaction that the miners can put into the
blockchain, they could redirect to themselves if they actually have the muscle to
fork it out.

775
01:16:33,796 --> 01:16:43,796
So why not just take all the Bitcoin transactions for seven blocks? Why not just
make it 70 blocks or 700 blocks? You can steal more and more Bitcoin that way.

776
01:16:44,796 --> 01:16:52,796
But of course, the more blocks you're reorging, the more you're just kind of saying
this whole network is just useless.

777
01:16:53,796 --> 01:17:03,796
Yeah, yeah. Okay, so I want to give you a quote and then you can respond to it
however you see fit. It's from Thamos. It was also from r.slicebitcoin. He's not
here in the chat.

778
01:17:04,796 --> 01:17:05,796
Yes, I've heard of him.

779
01:17:06,796 --> 01:17:17,796
He's not here in the chat, but I sort of got a response from him on this topic of
Drivechains and comparing them to Lightning Network and comparing them to federated
chains.

780
01:17:18,796 --> 01:17:26,796
He said, Drivechain is like Bitcoin SPV and it's worse than federated sidechains. So
what's your comment?

781
01:17:27,796 --> 01:17:30,796
Oh, well now we have to talk about federated sidechains.

782
01:17:30,796 --> 01:17:39,796
Yeah, it's a broader segue. So we're outside of Drivechain being the definition of
sidechains. Let's step out of that for the purpose of being able to discuss.

783
01:17:40,796 --> 01:17:46,796
Because I'm going to segue into Liquid and want to discuss Liquid's implementation.
So I want to give this quote sort of as a segue.

784
01:17:47,796 --> 01:17:52,796
So for everyone's sake, mentally detach yourself from Drivechain being the
definition for sidechains.

785
01:17:53,796 --> 01:18:02,796
So let's start now. Take it out of that and say, so Thamos has said, Drivechain is
like Bitcoin SPV, but it's worse than federated sidechains.

786
01:18:03,796 --> 01:18:10,796
Yes. Well, first of all, I'm going to claim the privilege of that second layer,
which is that it shouldn't be up to any of us to decide.

787
01:18:10,796 --> 01:18:15,796
The user should be able to decide what they want to use and they can decide whatever
they like.

788
01:18:16,796 --> 01:18:23,796
Like maybe they like sending their money to one website and just leaving it in the
control of the website admin and they can do anything they want.

789
01:18:24,796 --> 01:18:28,796
Or they could use Drivechain or they could use a federated sidechain. They can use
whatever they want.

790
01:18:29,796 --> 01:18:37,796
So the criticisms of projects are not a grounds for kind of not trying it, which I
think people often kind of conflate this to.

791
01:18:37,796 --> 01:18:43,796
But that being said, I really don't understand the federated model at all.

792
01:18:44,796 --> 01:18:53,796
I think I brought up earlier that I find it to be like an empty explanation because
you try to say that you're solving this trust problem by trusting a group of people.

793
01:18:54,796 --> 01:19:02,796
Or you try to say, well, this blockchain is going to be managed by 23 other people,
but you don't explain how they are going to manage it.

794
01:19:02,796 --> 01:19:14,796
They sort of do because they say, they try to say that, well, they make a bunch of
claims that I find to be kind of, not only are they baffling to me, but I think
they're almost like insulting to the intelligence of the audience.

795
01:19:15,796 --> 01:19:28,796
Like some of the claims are that all these keys are going to be inside of these
thermite machines that will like explode if anyone tampers with them and just like
potentially destroy an office building.

796
01:19:28,796 --> 01:19:34,796
Or I don't know exactly how they work, but it just seems like kind of, just seems
like not very cypherpunk.

797
01:19:35,796 --> 01:19:38,796
Like if this kind of stuff would work, then we would have done it with Bitcoin,
right?

798
01:19:39,796 --> 01:19:47,796
We would have just had Bitcoin set up with, oh, that was the secret all along to
just put Liberty Reserve on trusted hardware around the world.

799
01:19:48,796 --> 01:19:55,796
And the more problematic is this claim that the keys are actually on this hardware.

800
01:19:55,796 --> 01:20:14,796
I mean, it's kind of philosophically, it's almost as though part of the pitch for
the federated sidechain is a closed source in a kind of philosophical sense, which
is that all this stuff is said about the federation softwares on these trusted
hardware security things.

801
01:20:15,796 --> 01:20:20,796
And the hardware security things are geographically and jurisdictionally
distributed.

802
01:20:20,796 --> 01:20:27,796
But that's kind of like, isn't that also kind of like, I think that's kind of, you
know, of course they wouldn't put them all in one room somewhere, right?

803
01:20:28,796 --> 01:20:30,796
Like obviously you could like, you'd spread them out.

804
01:20:31,796 --> 01:20:35,796
These are just kind of like weird, fake kind of explanations for things.

805
01:20:36,796 --> 01:20:45,796
And it's just, this doesn't really, the whole point of Bitcoin is we have this error
recovery, even if a meteor strikes all the mining equipment and blows it up.

806
01:20:46,796 --> 01:20:50,796
People would turn on old mining equipment or there'd be some, Bitcoin would be able
to survive.

807
01:20:51,796 --> 01:20:57,796
Someone would hard fork the thing to a new proof of work or different ASIC hardware
or something.

808
01:20:58,796 --> 01:21:04,796
So there's all this ability to recover in a kind of organic way from mining
failures.

809
01:21:05,796 --> 01:21:10,796
But this is like, it's just a static 23 people and you just need 15.

810
01:21:11,796 --> 01:21:16,796
And this involves them, I mean, we need to have their identity or we don't.

811
01:21:17,796 --> 01:21:18,796
Either way, it's problematic.

812
01:21:19,796 --> 01:21:23,796
If we don't have their identity, then we don't know that they aren't all secretly
the same person.

813
01:21:24,796 --> 01:21:37,796
We don't, if we do not have their identity, then they can be arrested by some kind
of, obviously some kind of banking treaty or something signed with the United
States.

814
01:21:38,796 --> 01:21:44,796
Allows all kinds of broad discretion in terms of just like gish galop and just dump
a bunch of random complaints.

815
01:21:45,796 --> 01:21:51,796
So I will focus on one, which is that it's really just the same problem that you
have with an exchange.

816
01:21:52,796 --> 01:21:55,796
We're talking about the federated, I'm talking about the federated peg in general.

817
01:21:56,796 --> 01:22:02,796
The same problem you have with an exchange, which is a problem of scale at security
that I'll explain in a second.

818
01:22:03,796 --> 01:22:05,796
But it's just kind of on a larger scale.

819
01:22:06,796 --> 01:22:11,796
So for Liquid specifically, it's just on a larger scale because it's a bunch of
exchanges grouped together into like a big exchange.

820
01:22:12,796 --> 01:22:16,796
So the problem with the exchange is that they get big and then there's a bunch of
money there.

821
01:22:17,796 --> 01:22:18,796
And then everyone wants to hack the exchange.

822
01:22:18,796 --> 01:22:24,796
And so at a certain point, it just becomes unmanageable.

823
01:22:25,796 --> 01:22:36,796
Like there are, let's just see in this scheme, there's 15, you need 15 out of 23
signatures, or you need to steal 15 out of 23 boxes,

824
01:22:37,796 --> 01:22:43,796
or you need to go to the people in charge of feeding information into those boxes
and compromise 15 of them.

825
01:22:43,796 --> 01:22:50,796
Whatever you need to do, you need to like capture 15 people's families and start
cutting off their limbs or something.

826
01:22:51,796 --> 01:22:53,796
Whatever you need to do, you need to compromise 15 out of 23.

827
01:22:54,796 --> 01:23:03,796
Now, if there's something like, if there's only $15,000 on this sidechain, then
there's only $1,000 budget per person.

828
01:23:04,796 --> 01:23:11,796
But if there's $15 million on the sidechain, then it's a budget of a million per
compromise that you need to hit.

829
01:23:11,796 --> 01:23:16,796
And then of course, like, you know, right now, Bitcoin has what, like hundreds of
billions of dollars.

830
01:23:17,796 --> 01:23:20,796
So if you have, eventually the scale just becomes so high.

831
01:23:21,796 --> 01:23:23,796
This is the problem with any centralized thing is the single point of failure.

832
01:23:24,796 --> 01:23:27,796
It just becomes the cost benefit on compromising that single point.

833
01:23:28,796 --> 01:23:31,796
It just starts to just become, I think, completely unmanageable.

834
01:23:32,796 --> 01:23:37,796
And to complicate that even further, you have no way of, when you first enter the
system,

835
01:23:37,796 --> 01:23:41,796
you're like the first guy to deposit into the federated sidechain.

836
01:23:42,796 --> 01:23:44,796
You don't know how big it will be like while you're asleep.

837
01:23:45,796 --> 01:23:47,796
Like maybe you go to sleep and then people deposit a trillion dollars.

838
01:23:48,796 --> 01:23:51,796
And then everyone decides, okay, time to compromise the sidechain.

839
01:23:52,796 --> 01:23:55,796
And when they compromise it, they're not going to just steal, they're going to steal
all of the money.

840
01:23:56,796 --> 01:23:57,796
So that includes yours.

841
01:23:58,796 --> 01:24:01,796
I just don't think, and so I think it's the same problem with exchanges.

842
01:24:02,796 --> 01:24:06,796
And that's why people don't leave their money on exchanges, which is basically the
reason why they have less, much less liquidity.

843
01:24:07,796 --> 01:24:08,796
In the first place.

844
01:24:09,796 --> 01:24:19,796
And I think in turn, we'll see the exact same effect play out where I don't know if
people will feel comfortable leaving money on a federated sidechain.

845
01:24:20,796 --> 01:24:31,796
Because they won't know exactly at what point the scale will sort of tip and it will
become so large that either bribing or threatening or whatever.

846
01:24:32,796 --> 01:24:35,796
These people will just eventually kind of tip over that line.

847
01:24:35,796 --> 01:24:44,796
But mostly I object to the claim that a multi-signature kind of output plus a bunch
of stuff that we don't see.

848
01:24:45,796 --> 01:24:48,796
I object to the claim that that is like an innovation at all.

849
01:24:49,796 --> 01:24:50,796
I think it's not really anything.

850
01:24:51,796 --> 01:24:58,796
Okay, and so that's a segue into this week to add some specifics to Liquid.

851
01:24:59,796 --> 01:25:10,796
So Blockstream's main sort of blockbuster product, the specific sidechain that got a
lot of attention in 2014-15.

852
01:25:11,796 --> 01:25:16,796
Because the concept of sidechain is one thing, but then there was this specific
thing called Liquid, which was supposed to be like this great thing.

853
01:25:16,796 --> 01:25:23,796
All traders were going to use them to ARB because they'd be able to easily move coin
around via the sidechain.

854
01:25:24,796 --> 01:25:30,796
And then exchanges would participate and be able to easily facilitate arbitrage,
which would add efficiencies to the market and all these other great things.

855
01:25:31,796 --> 01:25:33,796
So they finally actually released it.

856
01:25:34,796 --> 01:25:37,796
It's production live now this week and there's been a lot of press around it.

857
01:25:38,796 --> 01:25:44,796
And you generated a bit of controversy on Twitter by tweeting that it's not a real
sidechain.

858
01:25:45,796 --> 01:25:47,796
Oh yeah, that is kind of funny.

859
01:25:48,796 --> 01:25:55,796
I mean, it's very funny because the person I checked and someone asked me a question
and that person has like 35 followers or something.

860
01:25:56,796 --> 01:26:00,796
And I just like replied to them and then it turned out this big thing.

861
01:26:01,796 --> 01:26:09,796
So I thought that was actually very funny that I kind of went back later and checked
out some one guy with like 35 followers kind of actually started this giant thing.

862
01:26:10,796 --> 01:26:11,796
I know, it's funny, isn't it?

863
01:26:11,796 --> 01:26:16,796
And so Samson Mao, I don't forget what his title is, something at Blockstream.

864
01:26:17,796 --> 01:26:18,796
Chief Joke Assembler.

865
01:26:19,796 --> 01:26:20,796
Chief Meme Officer.

866
01:26:21,796 --> 01:26:24,796
No, I'm actually quite fond of Samson, I think he's a good guy.

867
01:26:25,796 --> 01:26:26,796
He's a great comedian.

868
01:26:27,796 --> 01:26:30,796
So he took offense to this, right?

869
01:26:31,796 --> 01:26:34,796
And responded to you and said, I don't have the exact text here, but he said
something.

870
01:26:35,796 --> 01:26:39,796
It was like chest sort of disciplining you as if you were a child.

871
01:26:40,796 --> 01:26:41,796
Like, come on, enough of this nonsense.

872
01:26:42,796 --> 01:26:44,796
It is a sidechain, stop calling it not a sidechain.

873
01:26:45,796 --> 01:26:47,796
So what is the deal here?

874
01:26:48,796 --> 01:26:49,796
Is it a sidechain or not?

875
01:26:50,796 --> 01:26:56,796
Oh, no, I actually kind of object to the, it's not Liquid specifically, I object to
the federated model as being a sidechain.

876
01:26:57,796 --> 01:26:59,796
Of course, as I explained, it can be anything.

877
01:27:00,796 --> 01:27:06,796
If you have a blockchain and you have a token that moves among blockchains, then
that's kind of like the first pass definition of a sidechain.

878
01:27:07,796 --> 01:27:11,796
But that kind of just, that punts that question away into like, okay, what exactly
is a blockchain?

879
01:27:12,796 --> 01:27:18,796
Is a blockchain something if it's just a bunch of people and their friends say that
they'll give you your money back?

880
01:27:19,796 --> 01:27:20,796
Is that a blockchain?

881
01:27:21,796 --> 01:27:23,796
It kind of punts a lot of that stuff around.

882
01:27:23,796 --> 01:27:26,796
Like R3, the R3 private chain, is that a real blockchain?

883
01:27:27,796 --> 01:27:28,796
We don't know, right, no one knows.

884
01:27:29,796 --> 01:27:30,796
Bullshit.

885
01:27:31,796 --> 01:27:39,796
So it, but yeah, I think it's very clear, though, that what, yeah, I was amazed that
he went for that.

886
01:27:40,796 --> 01:27:49,796
And even later on in the thread, Adam Back shows up and he sort of, I don't know
exactly what word you use, he either confesses or laments or something like that.

887
01:27:49,796 --> 01:27:55,796
He says that, yes, unfortunately, now the word sidechain does apply to two
completely different things.

888
01:27:56,796 --> 01:27:59,796
So I don't know if he was agreeing with me or Samson or both of us or just trying to
be helpful.

889
01:28:00,796 --> 01:28:11,796
But there, I think anyone who knows the context, and that's what I replied to
Samson, I said, go back to 2015 and look at what was really going on.

890
01:28:11,796 --> 01:28:17,796
You see very clearly there was this time period where a lot of altcoins were coming
out.

891
01:28:18,796 --> 01:28:21,796
That was the era when, so let's try to set the stage here.

892
01:28:22,796 --> 01:28:26,796
The original sidechain's white paper, that was published in October of 2014.

893
01:28:27,796 --> 01:28:39,796
And 2014 was the year that all this weird stuff was really starting to come out,
like BitShares and NXT, Counterparty, all this weird stuff was coming out.

894
01:28:39,796 --> 01:28:45,796
So before then, it had just been Bitcoin, which was a serious project, Namecoin,
which was also a serious project.

895
01:28:46,796 --> 01:28:53,796
And that was not as sort of popular, but also like kind of like it had a real
justification for existing, like technologically.

896
01:28:54,796 --> 01:29:00,796
And then there was some stuff that was kind of just like a copy and paste thing that
was very shallow.

897
01:29:01,796 --> 01:29:03,796
So like Litecoin, etc.

898
01:29:03,796 --> 01:29:06,796
But in 2014, things started to get very different.

899
01:29:07,796 --> 01:29:09,796
And Ethereum was coming out, all this stuff was coming out.

900
01:29:10,796 --> 01:29:20,796
So in October 2014, this idea was published that we can just take all of those
features from sidechains and simulate them in Bitcoin.

901
01:29:21,796 --> 01:29:25,796
And you can capture all of the features of the altcoins and keep it within the
Bitcoin network.

902
01:29:26,796 --> 01:29:35,796
And so as I say, when someone says that they have come out with a sidechain's
technology, when someone like Blockstream, who is very highly technically regarded,

903
01:29:36,796 --> 01:29:44,796
that put out this paper in the past and that has made all these statements, you
know, in a context of late 2014,

904
01:29:45,796 --> 01:29:52,796
when they say that they have new sidechain's technology, what I really kind of hear
is that,

905
01:29:52,796 --> 01:30:00,796
which is maybe, you know, it might be on me, but what I hear is that they have a
kind of new technology that people can use to create sidechains.

906
01:30:01,796 --> 01:30:03,796
But they really have not done that.

907
01:30:04,796 --> 01:30:05,796
That's not really what they have.

908
01:30:06,796 --> 01:30:12,796
They don't have what was originally promised, which was that the mainchain would do
some kind of thing to do an SPV proof

909
01:30:13,796 --> 01:30:18,796
and kind of check that the sidechain had burnt its money, the money that is coming
back by a train.

910
01:30:19,796 --> 01:30:21,796
That was like part of the design.

911
01:30:22,796 --> 01:30:28,796
And also, as I said, I kind of more generally object to just putting a bunch of
stuff,

912
01:30:29,796 --> 01:30:31,796
saying that a bunch of stuff is behind a multi-signature address.

913
01:30:32,796 --> 01:30:36,796
I object to that claim that that is anything.

914
01:30:37,796 --> 01:30:41,796
I think if a different organization did that, that wasn't Blockstream, they would be
laughed at.

915
01:30:42,796 --> 01:30:49,796
And in fact, I have to point out that the thermite boxes idea, that was actually
literally laughed at for most of 2015.

916
01:30:50,796 --> 01:30:56,796
If anyone brought up sidechain's technology, people would post a GIF of someone's
house exploding or something,

917
01:30:57,796 --> 01:30:58,796
and everyone would just laugh at how weird it was.

918
01:30:59,796 --> 01:31:09,796
It took Chris Zeros and Josh Unseth months of questioning to get anyone to even
admit that this weird thermite box idea was even going around.

919
01:31:10,796 --> 01:31:13,796
And we have no way of knowing what is on the thermite.

920
01:31:14,796 --> 01:31:19,796
Someone in the audience has no way of knowing what's in the, just as someone in the
audience has no way of knowing what's going on,

921
01:31:20,796 --> 01:31:27,796
has no way of knowing what is going on at the Federal Reserve or in the inner
workings of the US Congress or something.

922
01:31:28,796 --> 01:31:29,796
You don't know of any of this.

923
01:31:30,796 --> 01:31:33,796
Where's the key that's in this box?

924
01:31:34,796 --> 01:31:35,796
Is it actually in the box or is something else going on?

925
01:31:36,796 --> 01:31:38,796
Was the key before they put it in the box?

926
01:31:39,796 --> 01:31:41,796
Did they publish it to the New York Times?

927
01:31:42,796 --> 01:31:44,796
No one knows anything about really what the setup is.

928
01:31:45,796 --> 01:31:51,796
So it's kind of just like a bunch of shady misdirection in my opinion.

929
01:31:52,796 --> 01:31:59,796
Yeah, and I get this feeling that Adam Peck on some level knows better,

930
01:32:00,796 --> 01:32:07,796
just because I remember how excited I was about Liquid and how excited I was about
Psychange back in early 2015-ish.

931
01:32:08,796 --> 01:32:11,796
Yeah, it's a very common comment that people say that.

932
01:32:12,796 --> 01:32:13,796
And Bitfinex was going to implement it.

933
01:32:14,796 --> 01:32:19,796
I remember talking to Phil about it and we were like, oh, it's going to be so easy
to R, it's going to be so great, it's going to be trustless.

934
01:32:20,796 --> 01:32:24,796
And I feel like Adam Peck is a brilliant cryptographer.

935
01:32:25,796 --> 01:32:33,796
Even in the current implementation of Liquid, the way that they've managed to handle
the peg,

936
01:32:34,796 --> 01:32:40,796
the way they've got IA, issued assets and confidential transactions, that crypto is
great.

937
01:32:41,796 --> 01:32:42,796
The tech is great there.

938
01:32:43,796 --> 01:32:44,796
Yeah, I think we should.

939
01:32:45,796 --> 01:32:48,796
I'm sorry to interrupt, but I do want to say I think we should do it.

940
01:32:49,796 --> 01:32:50,796
We should do it.

941
01:32:51,796 --> 01:32:52,796
I think it's a good idea and people should try it.

942
01:32:52,796 --> 01:32:53,796
And I think it's great.

943
01:32:54,796 --> 01:33:00,796
It's very difficult to ship software, so everyone should congratulate them on the
software release.

944
01:33:01,796 --> 01:33:07,796
That's different from saying that it is Psychange technology, though, because this
just might be on me.

945
01:33:08,796 --> 01:33:10,796
But I really think that if you say that you are coming out with Psychange
technology,

946
01:33:11,796 --> 01:33:15,796
what I would have in mind would be something that's open source that someone could
use.

947
01:33:16,796 --> 01:33:22,796
The example you used was make a spin up an Ethereum sidechain or something like
that.

948
01:33:23,796 --> 01:33:28,796
I designed Drivechain in order to bring my prediction market software that I think
we were talking about before.

949
01:33:29,796 --> 01:33:34,796
I don't know if we mentioned it on the air, but I designed Drivechain in order to
bring the prediction market software to Bitcoin

950
01:33:35,796 --> 01:33:37,796
to achieve a purpose.

951
01:33:38,796 --> 01:33:46,796
So it's kind of like if I can't achieve that purpose and if you can't decide a coin
developer can achieve similar purpose

952
01:33:47,796 --> 01:33:53,796
and if you can't get Roger Ver to put Bitcoin Cash large block sidechain on Bitcoin,
then using this technology.

953
01:33:54,796 --> 01:33:59,796
I mean, the way it would work if you had to use this technology would be to find
your own functionaries,

954
01:34:00,796 --> 01:34:01,796
find your own group of people to trust.

955
01:34:02,796 --> 01:34:04,796
And I think that is where it starts to really break down.

956
01:34:05,796 --> 01:34:08,796
And you see this for kind of what a shallow contribution it is.

957
01:34:09,796 --> 01:34:11,796
I don't know what to make of it.

958
01:34:12,796 --> 01:34:14,796
Yeah, that was the kicker to what I was saying.

959
01:34:15,796 --> 01:34:17,796
Acknowledging all the great tech that they put together.

960
01:34:18,796 --> 01:34:25,796
And then what I was going to finish saying was, and it feels sort of spoiled by the
fact that it's in this fucking shitty federated setup.

961
01:34:26,796 --> 01:34:30,796
And Blockstream, I mean, they've got some of the most brilliant cartographers in the
entire world.

962
01:34:31,796 --> 01:34:35,796
Some of the best collection of talent, better than all of the shitcoin shit out
there.

963
01:34:36,796 --> 01:34:37,796
They do.

964
01:34:38,796 --> 01:34:42,796
And they decided this was going to be like all that great tech packed into a
federated context.

965
01:34:43,796 --> 01:34:48,796
I feel like it's like so close to the finish line, like so close, but so far at the
same time.

966
01:34:49,796 --> 01:34:52,796
I mean, I hope that they end up doing some more interesting stuff beyond that.

967
01:34:53,796 --> 01:34:57,796
But let's talk about what's the worst that can happen in this context.

968
01:34:58,796 --> 01:35:00,796
So let's say the most evil federation is completely fucked.

969
01:35:01,796 --> 01:35:07,796
And let's say now, let's imagine that 50% of the volume on spot exchanges now is
being done using Liquid, these LBTCs.

970
01:35:08,796 --> 01:35:14,796
So the peg is enforced cryptographically and exchanges in the federation are
confirming transactions on their little sidechain.

971
01:35:15,796 --> 01:35:21,796
And people are moving around LBTC and they're redeeming it every now and then into
mainchain BTC and all this stuff.

972
01:35:22,796 --> 01:35:26,796
Let's now imagine that it's the evil and corrupted in the worst way.

973
01:35:27,796 --> 01:35:28,796
What can happen to people?

974
01:35:28,796 --> 01:35:33,796
Yes, well, this is an excellent question, because I think we should actually, I
would like, we should sort of ask them.

975
01:35:34,796 --> 01:35:41,796
But I, you know, I have some thoughts, which are that it's claimed that there's all
this stuff, again, in these like tamper proof machines.

976
01:35:42,796 --> 01:35:46,796
And these machines supposedly, they have like the only copy of the private key.

977
01:35:47,796 --> 01:35:54,796
And they supposedly are running this software that also tracks the block creation.

978
01:35:55,796 --> 01:36:02,796
Although actually, I think actually the latest version, they split that up so that
there's a different group of 15 out of 32.

979
01:36:03,796 --> 01:36:06,796
So the way this way things work in Bitcoin is I think there's a lot of rapid
technological progress.

980
01:36:07,796 --> 01:36:09,796
So someone will publish a paper or something.

981
01:36:10,796 --> 01:36:12,796
And then you finally get around to reading it two weeks later.

982
01:36:13,796 --> 01:36:16,796
And then as far as you know, the author of the paper has like figured out a bunch
more stuff and the design has changed.

983
01:36:17,796 --> 01:36:21,796
So I don't want to speak as really an expert on Liquid because I'm not.

984
01:36:21,796 --> 01:36:28,796
But what I do know for a fact is that by definition, they say that it's not, it
doesn't require any modifications to Bitcoin.

985
01:36:29,796 --> 01:36:39,796
Which means the rub there is that ultimately the Bitcoin network is just checking to
see if enough keys have signed on withdrawing the money.

986
01:36:40,796 --> 01:36:51,796
So that is the thing is that all this stuff is said about the keys only exist in
these certain places that are generated by this weird hardware that, you know, I
don't know.

987
01:36:52,796 --> 01:36:58,796
So there are certain kind of like laws of physics and laws of like information
processing that you kind of can't break.

988
01:36:59,796 --> 01:37:00,796
So I don't know the specific details.

989
01:37:01,796 --> 01:37:06,796
But what I do know is if you have those keys, you can just withdraw the money from
the sidechain, just assign it to yourself.

990
01:37:06,796 --> 01:37:16,796
And if you, it depends on like what other weird compromises are made, like these
boxes might all be in secure locations.

991
01:37:17,796 --> 01:37:20,796
But someone has got to feed them like an internet connection, right?

992
01:37:21,796 --> 01:37:28,796
They got to feed them an internet connection to let them know what's happening on
the exchanges so that they know which legitimate withdrawals to process.

993
01:37:29,796 --> 01:37:33,796
So there's all these things that must be possible.

994
01:37:33,796 --> 01:37:34,796
You know what I mean?

995
01:37:34,796 --> 01:37:42,796
Like if it's possible for this to be useful, there must be possible to just trick
this to get on a different network and withdraw all the money.

996
01:37:43,796 --> 01:37:57,796
So the worst case scenario is, of course, that you compromise enough of the 23 keys
or you compromise enough of the whatever it is, you know, whatever dimension they
chose to put the weakest link on, you would get 15 of 23 of those.

997
01:37:58,796 --> 01:38:01,796
You trick them onto a different internet or whatever you would do.

998
01:38:02,796 --> 01:38:07,796
Or you threaten their families or the people feeding, whatever it is, you compromise
that and then you get all the money.

999
01:38:08,796 --> 01:38:12,796
This is a special case, of course, because it's the biggest and because of other
reasons.

1000
01:38:13,796 --> 01:38:14,796
In this case, they're hostile.

1001
01:38:15,796 --> 01:38:16,796
Of course, right? That's kind of what I, yeah.

1002
01:38:17,796 --> 01:38:22,796
But, you know, I don't know how much, is this really high on Kraken's priority list?

1003
01:38:23,796 --> 01:38:28,796
Kraken doesn't even allow you to withdraw to base 32 formatted addresses.

1004
01:38:29,796 --> 01:38:30,796
Are they really integrating this?

1005
01:38:31,796 --> 01:38:32,796
If I were an exchange, I wouldn't touch.

1006
01:38:33,796 --> 01:38:34,796
I would make as few changes.

1007
01:38:35,796 --> 01:38:36,796
I would be so paranoid.

1008
01:38:37,796 --> 01:38:38,796
I would never make changes.

1009
01:38:39,796 --> 01:38:40,796
I would have the changes printed out.

1010
01:38:41,796 --> 01:38:43,796
I would be handwriting them at home to make sure that they were right.

1011
01:38:44,796 --> 01:38:46,796
And I would be super paranoid about making any changes.

1012
01:38:46,796 --> 01:38:55,796
The other kind of point of evidence against Liquid specifically, this is Liquid
specifically versus federated sidechains more generally,

1013
01:38:56,796 --> 01:39:02,796
which is that if this were really an issue, like I did mention that I asked this at
a Blockstream party once,

1014
01:39:03,796 --> 01:39:06,796
like how it would compete with the Lightning Network, which is that if it were like
a huge,

1015
01:39:07,796 --> 01:39:11,796
if this were a really big issue, I think the exchanges would have been able to
figure something out,

1016
01:39:11,796 --> 01:39:17,796
either using regular kind of contracts signed by lawyers, dumb contracts,

1017
01:39:18,796 --> 01:39:21,796
but also they could have used something like payment channels,

1018
01:39:22,796 --> 01:39:26,796
and I thought there was actually a lot of talk of this, like using like a Tier Nolan
channel or whatever it was,

1019
01:39:27,796 --> 01:39:31,796
and there was talk of, I thought BitGo, right, was doing something at some point?

1020
01:39:32,796 --> 01:39:33,796
Yeah, well, BitGo had –

1021
01:39:34,796 --> 01:39:35,796
Of course, and then it exploded.

1022
01:39:36,796 --> 01:39:39,796
And then everyone planned to figure out everyone else and we don't really know
exactly how that –

1023
01:39:39,796 --> 01:39:42,796
I don't know how that shook out. Probably a lot of people do.

1024
01:39:43,796 --> 01:39:47,796
Yeah, we have a pretty good idea about how it shook out, but there's a lot of
confidentiality and shit around it.

1025
01:39:48,796 --> 01:39:54,796
But the issue is that if this were a really big deal, why don't the exchanges just
set up payment channels with each other,

1026
01:39:55,796 --> 01:39:57,796
or as soon as SegWit was fixed,

1027
01:39:58,398 --> 01:40:05,398
why didn't they just connect to a big lightning hub, which would have – if all the
exchanges connect to a lightning hub,

1028
01:40:05,398 --> 01:40:11,398
I think that has the exact same effect. I think it's actually better. It's basically
the same.

1029
01:40:11,398 --> 01:40:17,398
It's all the benefits of liquid, but I think even stronger security because if
anyone screws you,

1030
01:40:17,398 --> 01:40:20,398
you steal all of their money, which that might even be just too harsh.

1031
01:40:20,398 --> 01:40:26,398
They might need to do a different version of lightning that's less harsh. I don't
know actually about that.

1032
01:40:26,398 --> 01:40:33,398
That's kind of an offhand comment, so don't read too much into that. But everyone in
one hub, everyone in one lightning –

1033
01:40:33,398 --> 01:40:40,398
everyone has a lightning node. As long as they have a hub that they deposit, all of
the Bitcoin that they would have deposited

1034
01:40:40,398 --> 01:40:47,398
into the liquid network seems that it would be strictly an improvement because
they'd be able to do all that stuff

1035
01:40:47,398 --> 01:40:52,398
without paying any transaction fees and without having to worry about the
functionaries at all,

1036
01:40:52,398 --> 01:40:57,398
and it would also be a great excuse. It would be even more private because the
confidential transactions

1037
01:40:57,398 --> 01:41:01,398
wouldn't even be an issue because they wouldn't appear on any blockchain at all.

1038
01:41:01,398 --> 01:41:06,398
So not only would the amounts be hidden, but everything would be hidden because
there would just not be anything.

1039
01:41:06,398 --> 01:41:14,398
So I think the settlement would be faster. You'd get to try out the real lightning
technology

1040
01:41:14,398 --> 01:41:26,398
versus this kind of new experimental technology that has never been – to have this
new kind of round-robin Paxos thing.

1041
01:41:26,398 --> 01:41:31,398
I didn't really look into it myself. It's not my area of expertise, but you don't
want to roll your own crypto.

1042
01:41:31,398 --> 01:41:36,398
That's a basic principle. So they have this new thing. Why take a chance on the new
thing

1043
01:41:36,398 --> 01:41:40,398
versus the lightning network that everyone's going to be cheerleading for?

1044
01:41:40,398 --> 01:41:45,398
So that's a criticism of just Liquid specifically, that if this were a big need,

1045
01:41:45,398 --> 01:41:50,398
then maybe the exchanges would have set up payment channels with each other.

1046
01:41:50,398 --> 01:41:54,398
I don't know. Maybe there's some reason why they didn't do that or didn't want to do
that.

1047
01:41:54,398 --> 01:42:02,398
I don't know. I just don't see why the barriers stopping them from using lightning
network.

1048
01:42:02,398 --> 01:42:05,398
I don't know. Don't those apply also to Liquid? I don't know.

1049
01:42:05,398 --> 01:42:07,398
Yeah, well, exactly. Yeah, yeah. That's the point, right?

1050
01:42:07,398 --> 01:42:13,398
Because I think people would be astonished if they understood just how low the
knowledge is of crypto developers

1051
01:42:13,398 --> 01:42:15,398
and exchanges who are implementing these things.

1052
01:42:15,398 --> 01:42:20,398
They're really not geniuses that are totally on the ball with all of the new tech
and stuff like that.

1053
01:42:20,398 --> 01:42:23,398
They're just developers that are implementing what they're told, right?

1054
01:42:23,398 --> 01:42:31,398
Yes, well, a good indicator of that is that Coinbase never went for – they never
went for the extension block, the easy win.

1055
01:42:31,398 --> 01:42:34,398
And they went for SegWit2x instead and just lost big.

1056
01:42:34,398 --> 01:42:39,398
But they could have so easily, if they had 83% hash rate, they could have soft
forked the extension block in.

1057
01:42:39,398 --> 01:42:42,398
It's always baffled me why they didn't take the easy win.

1058
01:42:42,398 --> 01:42:46,398
They insisted on trying and failing big.

1059
01:42:46,398 --> 01:42:50,398
I don't even know if there was some kind of four-dimensional chess going on

1060
01:42:50,398 --> 01:42:52,398
and they wanted to lose the whole time. I thought it was so baffling.

1061
01:42:52,398 --> 01:42:54,398
And Adam Banks supported it.

1062
01:42:54,398 --> 01:42:57,398
I mean, everyone was – that would have been – that was the path that I thought was
going to happen,

1063
01:42:57,398 --> 01:42:59,398
like at least back in 2015-ish.

1064
01:42:59,398 --> 01:43:01,398
Yes, 2014, 2015, right?

1065
01:43:01,398 --> 01:43:04,398
That was very old and then it kind of got swept under the rug.

1066
01:43:04,398 --> 01:43:07,398
Yeah, yeah. And then the whole thing became, okay, fork off.

1067
01:43:07,398 --> 01:43:09,398
You really didn't see it.

1068
01:43:09,398 --> 01:43:16,398
In the scaling conferences in late 2015, the extension block idea had kind of sort
of been –

1069
01:43:16,398 --> 01:43:19,398
sort of fizzled out or sort of wasn't considered interesting.

1070
01:43:19,398 --> 01:43:24,398
And then right after Hong Kong, it was all about SegWit Lightning.

1071
01:43:25,398 --> 01:43:29,398
There was all just a big push to just do that as quick as possible.

1072
01:43:29,398 --> 01:43:34,398
Yeah, and I think the narrative from Liquid and Blockstream now is like,

1073
01:43:34,398 --> 01:43:38,398
well, use for micropayments Lightning and for high-value payments Liquid,

1074
01:43:38,398 --> 01:43:42,398
which to me is a bit strange because, like you said, it's a bit more experimental on
Liquid.

1075
01:43:42,398 --> 01:43:45,398
So maybe I don't actually want to trust high-value payments just yet.

1076
01:43:45,398 --> 01:43:49,398
Maybe I want to see how it goes for a bit, make sure that things are okay.

1077
01:43:49,398 --> 01:43:54,398
So I'm not entirely sure either exactly how the use case is going to evolve

1078
01:43:54,398 --> 01:43:56,398
and how many exchanges are actually going to use it.

1079
01:43:56,398 --> 01:43:59,398
And just to sort of echo what – do this a little bit differently.

1080
01:43:59,398 --> 01:44:03,398
So it could be that – and I know that have they made because it's not just,

1081
01:44:03,398 --> 01:44:08,398
oh, here's this optimization in terms of faster blocks, faster confirmations.

1082
01:44:08,398 --> 01:44:10,398
But there's no risk. Exchanges use Liquid.

1083
01:44:10,398 --> 01:44:14,398
We don't have to worry about hacks anymore because the whole federation has to get
hacked

1084
01:44:14,398 --> 01:44:17,398
instead of this one exchange, which is total nonsense because –

1085
01:44:17,398 --> 01:44:20,398
The stuff exploded or something weird.

1086
01:44:20,398 --> 01:44:22,398
I don't want to make light of that, but it is very funny.

1087
01:44:22,398 --> 01:44:25,398
I'm just remembering all those GIFs that people used to post.

1088
01:44:25,398 --> 01:44:31,398
But yeah, I think I would like to know more about like what happens if they don't
like one of them.

1089
01:44:31,398 --> 01:44:35,398
Do they can like replace one with someone else or double copies of yourself

1090
01:44:35,398 --> 01:44:39,398
and then you add up to 15, you plus 14 clones.

1091
01:44:39,398 --> 01:44:41,398
And so I don't know. Is it fixed?

1092
01:44:41,398 --> 01:44:45,398
If so, that has its own set of 23 great guys.

1093
01:44:45,398 --> 01:44:49,398
And even if eight of them screw up, you only need 15.

1094
01:44:49,398 --> 01:44:52,398
So I don't know. I'd like to learn.

1095
01:44:52,398 --> 01:44:57,398
Sort of doubt about whether this was compromised and whether in the future

1096
01:44:57,398 --> 01:44:59,398
it can then be used to unravel the whole thing.

1097
01:44:59,398 --> 01:45:01,398
It's the same issue, right?

1098
01:45:01,398 --> 01:45:05,398
It was funny you mentioned that. It's kind of – I think it is similar.

1099
01:45:05,398 --> 01:45:07,398
I've never really been that worried about it because it actually –

1100
01:45:07,398 --> 01:45:09,398
it flows in the other direction, right?

1101
01:45:09,398 --> 01:45:12,398
The more people over time, the more people who just forget.

1102
01:45:12,398 --> 01:45:15,398
Again, I don't – this is a little macabre.

1103
01:45:15,398 --> 01:45:20,398
But over time, people like get hit by a bus or something and they're dead or their
apartment burned.

1104
01:45:20,398 --> 01:45:25,398
So you're right. It is similar that whenever you're interested with the Drivechain
model

1105
01:45:25,398 --> 01:45:31,398
or when you switch from thinking about Bitcoin to thinking about Drivechain, it's
exactly the same.

1106
01:45:31,398 --> 01:45:35,398
Yeah, it's just a calculated risk that we're all taking every day

1107
01:45:35,398 --> 01:45:37,398
when we're dealing with any crypto stuff.

1108
01:45:37,398 --> 01:45:38,398
There's tradeoffs.

1109
01:45:38,398 --> 01:45:42,398
And in the end, I think the way that you put it was pretty succinct.

1110
01:45:42,398 --> 01:45:47,398
I don't know if you said it yet or if I had read you saying it before,

1111
01:45:47,398 --> 01:45:52,398
but that it's like you're taking a bet on a particular mining profitability
strategy.

1112
01:45:52,398 --> 01:45:55,398
And I think that's really insightful that in all of these cases,

1113
01:45:55,398 --> 01:45:58,398
it's just different gradients of the same bet.

1114
01:46:00,398 --> 01:46:06,398
Yeah, whenever you interact with other people, you have this thing that's like a
physical –

1115
01:46:06,398 --> 01:46:10,398
it's as if it were a physical object in your brain because it is knowledge

1116
01:46:10,398 --> 01:46:12,398
that's encoded in your brain in some ways.

1117
01:46:12,398 --> 01:46:16,398
You have like a real thing. It's not like a fake subjective thing.

1118
01:46:16,398 --> 01:46:20,398
You objectively have real beliefs about what other people do

1119
01:46:20,398 --> 01:46:25,398
and they are in the physical reality and they have consequences.

1120
01:46:25,398 --> 01:46:29,398
Like if you watch people drive on the freeway, they mostly don't hit each other

1121
01:46:29,398 --> 01:46:34,398
even though there's just these little lines of white paint separating all the roads.

1122
01:46:34,398 --> 01:46:36,398
It's as if they were something much stronger.

1123
01:46:36,398 --> 01:46:39,398
There's a real like law of physics kind of operating there.

1124
01:46:39,398 --> 01:46:48,398
So it's not a thing to just be dismissed because it holds everything together and it
holds –

1125
01:46:48,398 --> 01:46:53,398
Right, and I think just like you say where you need to –

1126
01:46:57,398 --> 01:47:00,398
let me think the best way to put this.

1127
01:47:01,398 --> 01:47:05,398
Actually, I was going to make a point there but I think I'm not going to bother
because I think it'll –

1128
01:47:05,398 --> 01:47:07,398
Ah, yes, that's okay.

1129
01:47:07,398 --> 01:47:10,398
I think it'll maybe derail things so I don't want to go down that road.

1130
01:47:10,398 --> 01:47:16,398
But so let's – I think let's stop talking about Liquid because we need to –

1131
01:47:16,398 --> 01:47:21,398
I want to invite some Blockstream guys on so we can discuss exactly what their
implementation is currently

1132
01:47:21,398 --> 01:47:26,398
and then we can just sort of go based off what they're actually saying

1133
01:47:26,398 --> 01:47:28,398
and not what we are assuming that they are doing.

1134
01:47:28,398 --> 01:47:31,398
So, yes, I think –

1135
01:47:31,398 --> 01:47:35,398
Oh, yes, no, anything like audience questions or whatever, let's do anything.

1136
01:47:35,398 --> 01:47:36,398
Yes, perfect.

1137
01:47:36,398 --> 01:47:45,398
So, okay, one thing actually I wanted to ask before that I hadn't quite gotten to in
the miscellaneous part

1138
01:47:45,398 --> 01:47:46,398
was this Bee Foundation.

1139
01:47:46,398 --> 01:47:48,398
Have you done anything?

1140
01:47:48,398 --> 01:47:49,398
I know that –

1141
01:47:49,398 --> 01:47:50,398
Started a different club.

1142
01:47:50,398 --> 01:47:52,398
They just started like Satoshi Roundtable but with a different name.

1143
01:47:52,398 --> 01:47:56,398
Then it wouldn't have meant anything.

1144
01:47:56,398 --> 01:47:59,398
But I think maybe there is actually something to that though,

1145
01:47:59,398 --> 01:48:03,398
to the specifically the kind of the audacity of using the word like the foundation

1146
01:48:03,398 --> 01:48:05,398
or phrasing it that way.

1147
01:48:05,398 --> 01:48:08,398
There might actually be something to that.

1148
01:48:08,398 --> 01:48:14,398
Just as a kind of guess, I might say that it might be a good sign.

1149
01:48:14,398 --> 01:48:18,398
It might be a good indicator that stuff is going to go really wrong

1150
01:48:18,398 --> 01:48:22,398
for all the people who are in that group over the next 6 to 12 months

1151
01:48:22,398 --> 01:48:25,398
because that's like the hubris moment.

1152
01:48:25,398 --> 01:48:27,398
I don't really have a lot of thoughts about it.

1153
01:48:27,398 --> 01:48:32,398
I mean, obviously, if it mattered, then that would be a weird attack vector for
Bitcoin.

1154
01:48:32,398 --> 01:48:34,398
It would just indicate that Bitcoin is –

1155
01:48:34,398 --> 01:48:35,398
Really weak, right?

1156
01:48:35,398 --> 01:48:36,398
Is really weak, right?

1157
01:48:36,398 --> 01:48:37,398
Yes.

1158
01:48:37,398 --> 01:48:40,398
What about –

1159
01:48:40,398 --> 01:48:45,398
So, I mean, what do you think about the incentives overall in Bitcoin development?

1160
01:48:45,398 --> 01:48:46,398
Do you think that –

1161
01:48:46,398 --> 01:48:47,398
Actually, let me ask a different question.

1162
01:48:47,398 --> 01:48:49,398
What about code of conduct in open source projects?

1163
01:48:49,398 --> 01:48:59,398
So Linux has recently generated some conflict over basically resigning sort of from
the Linux project

1164
01:48:59,398 --> 01:49:02,398
because there's this code of conduct that has been generating things.

1165
01:49:02,398 --> 01:49:04,398
And I know Grin has a code of conduct,

1166
01:49:04,398 --> 01:49:07,398
and there's discussions maybe about Bitcoin doing something.

1167
01:49:07,398 --> 01:49:12,398
What do you feel about this and how this would impact the project?

1168
01:49:12,398 --> 01:49:14,398
The issue is not –

1169
01:49:14,398 --> 01:49:16,398
As I said earlier, the issue is not –

1170
01:49:16,398 --> 01:49:21,398
People will make mistakes when they write software

1171
01:49:21,398 --> 01:49:25,398
and when they create processes that write software,

1172
01:49:25,398 --> 01:49:27,398
and everything we do, there will be mistakes.

1173
01:49:27,398 --> 01:49:30,398
So the issue is just to make sure that we will survive the mistakes.

1174
01:49:30,398 --> 01:49:34,398
Like Karl Popper and his current Bitcoin developers are wrong about something.

1175
01:49:34,398 --> 01:49:36,398
You just say, well, as long as that doesn't matter too much,

1176
01:49:36,398 --> 01:49:40,398
as long as there's some pathway to correction, then it doesn't matter.

1177
01:49:40,398 --> 01:49:44,398
But if there is no way to correct the mistake,

1178
01:49:44,398 --> 01:49:47,398
then whether or not there actually currently is a mistake kind of doesn't really
matter.

1179
01:49:47,398 --> 01:49:51,398
It just means that it's only a matter of time before some mistake shows up

1180
01:49:51,398 --> 01:49:53,398
and it just blows up everything.

1181
01:49:53,398 --> 01:49:58,398
So the key is to have improvement, error correction over time.

1182
01:49:58,398 --> 01:50:03,398
So what you really need in all human endeavors

1183
01:50:03,398 --> 01:50:07,398
is to make sure that all of your mistakes will be caught quickly,

1184
01:50:08,398 --> 01:50:14,398
which is to start a hard fork or start a spinoff altcoin,

1185
01:50:14,398 --> 01:50:20,398
or Bitcoin is often unfortunately a kind of a source of distress

1186
01:50:20,398 --> 01:50:22,398
or psychological hangups for people.

1187
01:50:22,398 --> 01:50:25,398
But you only need the token to work, and you get paid if you own it.

1188
01:50:25,398 --> 01:50:27,398
So you get paid in Bitcoin.

1189
01:50:27,398 --> 01:50:30,398
So with Bitcoin, you can trade in these markets.

1190
01:50:30,398 --> 01:50:32,398
You can create a market.

1191
01:50:32,398 --> 01:50:34,398
You can buy and sell in the market.

1192
01:50:34,398 --> 01:50:38,398
And that's also how the people who resolve the outcomes get paid.

1193
01:50:38,398 --> 01:50:42,398
So you can do basically everything with Bitcoin if you just want to be a mere user.

1194
01:50:42,398 --> 01:50:46,398
But if you want to be an employee, you have to kind of stake up some money.

1195
01:50:46,398 --> 01:50:51,398
But it's not quite a stake because this asset has a free-floating exchange rate.

1196
01:50:51,398 --> 01:50:55,398
So this corporation will rise in value as the network becomes more valuable,

1197
01:50:55,398 --> 01:50:58,398
and it will fall in value as the network becomes less valuable.

1198
01:50:58,398 --> 01:51:02,398
And that is a kind of disincentive to do like an exit scam.

1199
01:51:02,398 --> 01:51:07,398
That's kind of a similar problem to the exchange problem, except much worse.

1200
01:51:07,398 --> 01:51:10,398
The exchange gets big, and then people want to hack the exchange.

1201
01:51:10,398 --> 01:51:13,398
But then you have these things like what was it called?

1202
01:51:13,398 --> 01:51:14,398
Sheep market or something.

1203
01:51:14,398 --> 01:51:15,398
I don't remember.

1204
01:51:15,398 --> 01:51:20,398
The other dark market where they get really big, and then the owners just decide,

1205
01:51:20,398 --> 01:51:23,398
well, operating this is a lot of work, and it's risky,

1206
01:51:23,398 --> 01:51:27,398
and instead I could just cut and run with all this Bitcoin at a certain scale

1207
01:51:27,398 --> 01:51:28,398
that becomes an issue.

1208
01:51:28,398 --> 01:51:30,398
So with this, it shouldn't become an issue

1209
01:51:30,398 --> 01:51:32,398
because if you do a great job and grow the company,

1210
01:51:32,398 --> 01:51:35,398
you can just cash out the same way you could with a real company.

1211
01:51:35,398 --> 01:51:42,398
You could sell some of your vote coin or what would be called rep in mine.

1212
01:51:42,398 --> 01:51:44,398
And so you have that aspect of it.

1213
01:51:44,398 --> 01:51:49,398
And then I ask all of the individual people to report on all the outcomes.

1214
01:51:49,398 --> 01:51:53,398
So it's actually similar to Drivechain and similar to Bitcoin itself

1215
01:51:53,398 --> 01:51:58,398
where a lot of stuff that is happening across a long period of time

1216
01:51:58,398 --> 01:52:02,398
in many dimensions, that's crunched up into a few discrete events.

1217
01:52:02,398 --> 01:52:06,398
So just like in Bitcoin, you have all these transactions that are crunched up

1218
01:52:06,398 --> 01:52:09,398
into little blocks that happen once every 10 minutes,

1219
01:52:09,398 --> 01:52:12,398
and Satoshi put the proof of work on the blocks,

1220
01:52:12,398 --> 01:52:17,398
on the discrete blocks that just arrive in chumps, big, big lumps at a time,

1221
01:52:17,398 --> 01:52:20,398
big drips instead of a continuous stream.

1222
01:52:20,398 --> 01:52:24,398
And similarly in Drivechain, the withdrawals,

1223
01:52:24,398 --> 01:52:27,398
there's a lot of stuff happening on the sidechain, weird rules,

1224
01:52:27,398 --> 01:52:29,398
that you don't see.

1225
01:52:29,398 --> 01:52:31,398
Are the rules being followed? You don't know.

1226
01:52:31,398 --> 01:52:34,398
Are the blocks like super big over there? You don't know.

1227
01:52:34,398 --> 01:52:36,398
But it's compressed into one.

1228
01:52:36,398 --> 01:52:39,398
Someone like Andreas can compress all the stuff over there

1229
01:52:39,398 --> 01:52:43,398
and then just with one little UTXO request every three months,

1230
01:52:43,398 --> 01:52:45,398
he can sync it all across.

1231
01:52:45,398 --> 01:52:52,398
So similarly, I guess that's a kind of common theme in design for me, I guess.

1232
01:52:53,398 --> 01:52:58,398
In Truthcoin Hivemind, I have lots of different events

1233
01:52:58,398 --> 01:53:01,398
that are all resolved at the same time.

1234
01:53:01,398 --> 01:53:05,398
And all the people who own this corporation, they submit events.

1235
01:53:05,398 --> 01:53:08,398
They report on each event.

1236
01:53:08,398 --> 01:53:11,398
So you have a big matrix of everyone's reports.

1237
01:53:11,398 --> 01:53:15,398
They throw in a ballot for each thing, or in practice, they do it all at once.

1238
01:53:15,398 --> 01:53:19,398
But you have this big matrix, and then I do something kind of interesting

1239
01:53:19,398 --> 01:53:25,398
where if your results agree with most other people's results,

1240
01:53:25,398 --> 01:53:29,398
well, first of all, if everyone is in complete unanimous agreement, nothing happens.

1241
01:53:29,398 --> 01:53:34,398
But if there's any disagreement, the people who disagree the most are penalized.

1242
01:53:34,398 --> 01:53:38,398
So there's this intense conformity that is established.

1243
01:53:38,398 --> 01:53:43,398
And it's even kind of cleverly done where if people disagree on something

1244
01:53:43,398 --> 01:53:46,398
that most other people agreed on, they're penalized more.

1245
01:53:46,398 --> 01:53:50,398
And so there's a very big kind of, there's kind of this big multidimensional
surface.

1246
01:53:50,398 --> 01:53:54,398
This is like a statistical trick that you can read about in the white paper.

1247
01:53:54,398 --> 01:53:58,398
And so there's like a big, it's like a big hill, but in many dimensions.

1248
01:53:58,398 --> 01:54:03,398
And if you are too deviant, you lose some of your shares of this hypothetical
corporation,

1249
01:54:03,398 --> 01:54:06,398
and they go to other people.

1250
01:54:06,398 --> 01:54:10,398
And then like in Drivechain, there's a kind of, there's a long,

1251
01:54:10,398 --> 01:54:14,398
there's just these long delays, these long periods where the thing,

1252
01:54:14,398 --> 01:54:19,398
the network can just be on pause instead of the resolution process can be on pause,

1253
01:54:19,398 --> 01:54:21,398
not the network itself.

1254
01:54:21,398 --> 01:54:22,398
So you have all these people trading.

1255
01:54:22,398 --> 01:54:26,398
There's a cool little graphic in the white paper where there's all these people
trading,

1256
01:54:26,398 --> 01:54:31,398
and then that part is allowed to happen very quickly.

1257
01:54:31,398 --> 01:54:36,398
But then as the stuff becomes more and more important,

1258
01:54:36,398 --> 01:54:41,398
more integral to the resolution process, it happens more rarely,

1259
01:54:41,398 --> 01:54:45,398
and there's much more kind of checking, and there's much more inconvenience.

1260
01:54:45,398 --> 01:54:47,398
There's much more kind of clamping down on it.

1261
01:54:47,398 --> 01:54:50,398
So there's supposed to be a kind of mix up between those things.

1262
01:54:50,398 --> 01:54:51,398
So I don't know.

1263
01:54:51,398 --> 01:54:54,398
I don't know if that's going to count as overengineered bullshit,

1264
01:54:54,398 --> 01:54:58,398
but I tried to make it as simple as I could while still providing some guarantees.

1265
01:54:58,398 --> 01:55:03,398
I gave a talk at QCon that people can find about the Oracle problem,

1266
01:55:03,398 --> 01:55:10,398
and the talk is about this issue of how do you get stuff that we know in the real
world,

1267
01:55:10,398 --> 01:55:13,398
we know that Donald Trump was elected president in the United States,

1268
01:55:13,398 --> 01:55:16,398
how do we get that represented in the blockchain?

1269
01:55:16,398 --> 01:55:20,398
And that is what the talk is about, and there are many, many challenges to that.

1270
01:55:20,398 --> 01:55:23,398
And it's an interesting talk if you're interested in that kind of thing.

1271
01:55:23,398 --> 01:55:27,398
I do think that, again, the medium is the message sort of,

1272
01:55:27,398 --> 01:55:33,398
so probably, I don't know, you should do more reading on like a PDF

1273
01:55:33,398 --> 01:55:37,398
than listening on a podcast, but maybe that's enough to get people interested.

1274
01:55:37,398 --> 01:55:41,398
Yeah, I mean, this is an inherently super complex topic,

1275
01:55:41,398 --> 01:55:44,398
and so I don't want to necessarily go too much deeper,

1276
01:55:44,398 --> 01:55:50,398
but maybe you can tell people who's the bookie in Hivemind?

1277
01:55:50,398 --> 01:55:55,398
Are you doing a paramutual bidding structure where you need to have a guy who's on
the other end,

1278
01:55:55,398 --> 01:56:00,398
and then everything just sort of balances out because you're just going from one
person's pocket to someone else's?

1279
01:56:00,398 --> 01:56:06,398
Or is there some kind of like bookie incentives going on from a third party or what?

1280
01:56:07,398 --> 01:56:11,398
No, there is no. The blockchain is the bookie, and this is quite –

1281
01:56:11,398 --> 01:56:15,398
I don't think people appreciate this. This is a complete convenience thing.

1282
01:56:15,398 --> 01:56:21,398
This just sort of coincidentally happened, and it just happened to be very lucky,

1283
01:56:21,398 --> 01:56:27,398
but there's something invented to address a problem of liquidity in prediction
markets

1284
01:56:27,398 --> 01:56:35,398
that also happens to double as very, very useful in a UTXO-based blockchain world.

1285
01:56:35,398 --> 01:56:37,398
So it's just really good luck.

1286
01:56:37,398 --> 01:56:42,398
But what's going on is that the first person, as I said, you buy a slot if you have
a question,

1287
01:56:42,398 --> 01:56:46,398
and then you also have to front some money if you want to make a market.

1288
01:56:46,398 --> 01:56:50,398
But you can – once you have these questions like who is going to be –

1289
01:56:50,398 --> 01:56:54,398
will Hillary Clinton be elected president in the 2016 election or something?

1290
01:56:54,398 --> 01:56:58,398
Or I guess we should now say – I have to update all my lines.

1291
01:56:58,398 --> 01:57:03,398
So I should say will Donald Trump be reelected president in the 2020 election in the
United States?

1292
01:57:03,398 --> 01:57:08,398
And that would be resolved in like December 2020 after the election happens in
November.

1293
01:57:08,398 --> 01:57:14,398
But you buy a slot, and then people could reuse this question in many different
markets

1294
01:57:14,398 --> 01:57:17,398
because it's possible to have really cool markets that interact with each other,

1295
01:57:17,398 --> 01:57:23,398
which unfortunately is – we probably can't explain right now, but I assure you it's
really, really cool,

1296
01:57:23,398 --> 01:57:30,398
and it's the key to like fixing a lot of big things that are broken in society today
very easily.

1297
01:57:30,398 --> 01:57:34,398
So you got to kind of start small. We can't talk about that right now.

1298
01:57:34,398 --> 01:57:39,398
But let's just say you make the slot, and then you make one market that you're just
trying to bet on

1299
01:57:39,398 --> 01:57:42,398
whether or not Donald Trump will or will not be elected,

1300
01:57:42,398 --> 01:57:45,398
and you don't care about these other beautiful features that are very, very awesome.

1301
01:57:45,398 --> 01:57:47,398
You have to take my word for it.

1302
01:57:47,398 --> 01:57:56,398
Then when you create the market, you front some money in order to create the market
in the first place.

1303
01:57:56,398 --> 01:58:00,398
So these people who buy the slot and who create the market, they get a cut of the –

1304
01:58:00,398 --> 01:58:03,398
they are compensated later, and they get a cut of the trading fees,

1305
01:58:03,398 --> 01:58:06,398
and the people who resolve the market get a cut of the trading fees.

1306
01:58:06,398 --> 01:58:09,398
So that's 50%, 25%, 25%.

1307
01:58:09,398 --> 01:58:13,398
And I only bring this up because I don't want to say like money is coming for free
or something.

1308
01:58:13,398 --> 01:58:18,398
So I know this explanation kind of sucks, but when the guy fronts the – yeah.

1309
01:58:18,398 --> 01:58:22,398
The big problem is that the market making, you have to price it.

1310
01:58:22,398 --> 01:58:25,398
It's just like you're market making any commodity, whether you're market making
futures.

1311
01:58:25,398 --> 01:58:27,398
This isn't that problem. This doesn't have any problems.

1312
01:58:27,398 --> 01:58:29,398
This is the greatest, luckiest thing to happen.

1313
01:58:29,398 --> 01:58:32,398
For anyone who wanted to put prediction markets on a blockchain,

1314
01:58:32,398 --> 01:58:36,398
this is like beautiful luck raining down from the sky solving problems magically.

1315
01:58:36,398 --> 01:58:39,398
Okay, good. Well, that's a big – holy shit.

1316
01:58:39,398 --> 01:58:41,398
I mean, go on. That's a lot to say.

1317
01:58:41,398 --> 01:58:45,398
But so let's say I want to bet $1,000 that Trump –

1318
01:58:45,398 --> 01:58:48,398
hopefully to God that he gets reelected.

1319
01:58:48,398 --> 01:58:52,398
Then how do I get paid? What's my multiple on this payment?

1320
01:58:52,398 --> 01:58:55,398
What's the price that I'm getting filled at? Like how does that even work?

1321
01:58:55,398 --> 01:58:57,398
Yeah, I don't know how you feel about Microsoft Excel,

1322
01:58:57,398 --> 01:59:00,398
but I made a little Excel spreadsheet with like many examples,

1323
01:59:00,398 --> 01:59:02,398
and you can just try it out yourself.

1324
01:59:02,398 --> 01:59:05,398
I should probably get some kind of Heroku app for this actually.

1325
01:59:05,398 --> 01:59:08,398
That's probably a good idea. I should put that on. I should write that down or
something.

1326
01:59:08,398 --> 01:59:14,398
So I know Excel is not the best explanation because it's so difficult to see.

1327
01:59:14,398 --> 01:59:17,398
Yeah, and I think there's actually a lot of layers.

1328
01:59:17,398 --> 01:59:20,398
That's also why I wanted to kind of avoid in general the Hivemind discussion

1329
01:59:20,398 --> 01:59:23,398
because I think we could go on another three hours basically.

1330
01:59:23,398 --> 01:59:25,398
Yes, we probably could.

1331
01:59:25,398 --> 01:59:29,398
But what I want to say is this first guy kind of makes a sacrifice trade,

1332
01:59:29,398 --> 01:59:31,398
the guy who makes the market.

1333
01:59:31,398 --> 01:59:34,398
He puts a bunch of money and that he's never getting back.

1334
01:59:34,398 --> 01:59:36,398
So he's got to hope that he makes it up on the trading fees.

1335
01:59:36,398 --> 01:59:40,398
So he wants to make a market. It's a very excellent incentive for him.

1336
01:59:40,398 --> 01:59:43,398
He wants to make a market that other people will use.

1337
01:59:43,398 --> 01:59:46,398
Now, he also – he fronts his money up.

1338
01:59:46,398 --> 01:59:48,398
It depends on how much money he puts up.

1339
01:59:49,398 --> 01:59:51,398
He just has to put up some positive amount.

1340
01:59:51,398 --> 01:59:54,398
Otherwise, we get division by zero and nothing works.

1341
01:59:54,398 --> 01:59:59,398
But if you – let's say the guy puts up $1,000.

1342
01:59:59,398 --> 02:00:03,398
This $1,000 is kind of wagered on all sides at once.

1343
02:00:03,398 --> 02:00:09,398
And so what you do is when you bet your $1,000, this is a kind of calculus thing.

1344
02:00:09,398 --> 02:00:13,398
If people understand calculus, it's probably a lot easier to understand.

1345
02:00:13,398 --> 02:00:20,398
Where it's kind of like this guy has started off – the market starts off completely
balanced.

1346
02:00:20,398 --> 02:00:27,398
So if it just has these two states, you have 50% for yes and 50% for no

1347
02:00:27,398 --> 02:00:30,398
because Donald Trump is either going to be reelected or he's not.

1348
02:00:30,398 --> 02:00:34,398
Maybe he won't be reelected because he was hit by an asteroid or something.

1349
02:00:34,398 --> 02:00:38,398
But whatever it is, no is going to happen unless yes happens.

1350
02:00:38,398 --> 02:00:40,398
So if he's elected, it's yes.

1351
02:00:40,398 --> 02:00:47,398
If anything else happens, if anything where you would normally say all bets are off
happens, they're not off.

1352
02:00:47,398 --> 02:00:49,398
It's no.

1353
02:00:49,398 --> 02:00:51,398
So no happened.

1354
02:00:51,398 --> 02:00:52,398
So either yes or no is happening.

1355
02:00:52,398 --> 02:00:57,398
So when it starts off, these would each be available for $0.50.

1356
02:00:57,398 --> 02:01:02,398
Let's just say instead of – you just have to imagine that you have to replace –

1357
02:01:02,398 --> 02:01:07,398
when I say $0.50, you just have to replace it with like $0.50 or $0.50 or something.

1358
02:01:07,398 --> 02:01:10,398
That's not an integral part of the explanation.

1359
02:01:10,398 --> 02:01:18,398
So they start off at 0.5 units and when you first start to buy, you want to bet that
Donald Trump is elected.

1360
02:01:18,398 --> 02:01:27,398
Your first couple of pennies for a while, they will be trickling out and you will
buy one share or whatever it is.

1361
02:01:27,398 --> 02:01:30,398
You'll buy 150th of a share because it's $0.50.

1362
02:01:30,398 --> 02:01:34,398
We're going penny by penny here in a calculus sense.

1363
02:01:34,398 --> 02:01:39,398
So you buy 150th of a share at a time at $0.50, the price.

1364
02:01:39,398 --> 02:01:45,398
But as you spend more and more pennies, the scales will tilt and you will eventually
be betting against yourself.

1365
02:01:45,398 --> 02:01:51,398
So if you bet like a million dollars, if the first guy fronts $1,000 – and again, I
apologize.

1366
02:01:51,398 --> 02:01:55,398
I'm sure this is a really painful explanation for anyone who doesn't know calculus.

1367
02:01:55,398 --> 02:01:57,398
Probably even if you do, it's probably just horrible.

1368
02:01:57,398 --> 02:02:03,398
But let's imagine Aaron shows up, Alice shows up on day one.

1369
02:02:03,398 --> 02:02:06,398
Alice creates the market and spends $1,000.

1370
02:02:06,398 --> 02:02:11,398
Bob shows up and Bob bets $10 million that Donald Trump will win.

1371
02:02:11,398 --> 02:02:21,398
What Bob is going to end up with is he's going to end up with having spent $10
million mostly at $1 per share.

1372
02:02:21,398 --> 02:02:29,398
So he will have bet a lot and slid the scales against himself and then once he's
exhausted the $1,000 that he was betting against,

1373
02:02:29,398 --> 02:02:36,398
he will be buying one share at a cost of $1 for $1 a share for all of his 100
percentage points.

1374
02:02:36,398 --> 02:02:43,398
So basically he'll acquire a bunch of shares that he can at best sell for exactly
what he bought them for.

1375
02:02:43,398 --> 02:02:45,398
So I'm sure this explanation is terrible.

1376
02:02:45,398 --> 02:02:48,398
But what happens is there's just this rule.

1377
02:02:48,398 --> 02:02:55,398
There's this formula and when you update the formula, you pay a certain amount and
if you pay a certain amount,

1378
02:02:55,398 --> 02:02:59,398
you get a certain amount of shares and you just update this over and over and over
again.

1379
02:02:59,398 --> 02:03:06,398
So it's like one UTXO that just keeps multiple people keep interacting with it and
it just moves in a big line.

1380
02:03:06,398 --> 02:03:08,398
So you just keep updating it.

1381
02:03:08,398 --> 02:03:15,398
So let's say Trump gets caught having sex with the monkey and so now his chances are
so much lower.

1382
02:03:15,398 --> 02:03:17,398
So there has to be a reaction in the market, right?

1383
02:03:17,398 --> 02:03:27,398
The chances are not lower until people either sell their yes shares or they buy the
no shares.

1384
02:03:27,398 --> 02:03:33,398
And so when they start selling the yes shares, they will immediately get good prices
because the default state –

1385
02:03:33,398 --> 02:03:37,398
it's kind of like a seesaw in this sense and in a multivariate sense, it's like
balancing a tray on a pin.

1386
02:03:37,398 --> 02:03:41,398
And yeah, I agree with whoever said that.

1387
02:03:41,398 --> 02:03:43,398
Yeah, he would be super popular honestly.

1388
02:03:44,398 --> 02:03:45,398
People would love it.

1389
02:03:45,398 --> 02:03:53,398
But it's like a seesaw where if a huge pile of money is on one side, you get less
favorable pricing.

1390
02:03:53,398 --> 02:03:56,398
It's kind of like a seesaw except if you turn the ruler upside down unfortunately.

1391
02:03:56,398 --> 02:04:02,398
So this is like the price you're buying would be – if you dumped a ton of money on
one side,

1392
02:04:02,398 --> 02:04:08,398
then you would be – if you wanted to buy even further, you would be paying a dollar
for something

1393
02:04:08,398 --> 02:04:11,398
and you'd get a share back that would only be worth a dollar if you were right.

1394
02:04:11,398 --> 02:04:16,398
So at a certain point, it becomes very pointless because already so much money is on
one side

1395
02:04:16,398 --> 02:04:22,398
that the seesaw is giving you this thing that's worth a dollar if Donald Trump wins
and zero if he doesn't win

1396
02:04:22,398 --> 02:04:24,398
and it's selling you that for one dollar.

1397
02:04:24,398 --> 02:04:26,398
So it's actually terrible.

1398
02:04:26,398 --> 02:04:31,398
So the first couple of people who take their money out, they will be getting it at
great prices.

1399
02:04:31,398 --> 02:04:33,398
So you really want to be the first guy to sell.

1400
02:04:33,398 --> 02:04:35,398
And then eventually – or buy, exactly.

1401
02:04:35,398 --> 02:04:43,398
So the market will react and try to mirror the real world as quickly as it possibly
can.

1402
02:04:43,398 --> 02:04:46,398
I mean this is how market makers operate in normal markets anyway, right?

1403
02:04:46,398 --> 02:04:47,398
So they're going to –

1404
02:04:47,398 --> 02:04:48,398
Yeah, so this is great.

1405
02:04:48,398 --> 02:04:49,398
So there's always liquidity.

1406
02:04:49,398 --> 02:04:52,398
You can always place a trade for some amount.

1407
02:04:52,398 --> 02:04:56,099
It may be a small amount, but you can always place a trade.

1408
02:04:56,099 --> 02:05:09,099
No matter what. So you can always bet against someone. It's just funny that because
of the way the calculus works, you will eventually be betting against yourself at
terrible odds. So the magnitude is the hang up now.

1409
02:05:09,099 --> 02:05:27,099
But the cool thing is the guy who creates the market, he can choose. He can say, I
really want to attract a lot of people to this market. He can front a ton of money
when he sets it up and this makes it permanently much more liquid throughout its
entire lifetime because there's just a bunch of money like in the middle at the
fulcrum, in the middle of the seesaw.

1410
02:05:27,099 --> 02:05:45,099
It's kind of like weighing it down that he kind of sacrifices. The other very cool
thing is that if the market has independent social value, so for example, if it's a
market about whether or not Donald Trump will, if elected, keep unemployment down.

1411
02:05:45,099 --> 02:06:02,099
So if Donald Trump would be good for the economy, you could have, or if good for any
other thing, you could have people donate at any time. At any time, they can put
more money on the fulcrum of the seesaw and make it more and more liquid at higher
and higher amounts.

1412
02:06:02,099 --> 02:06:21,099
So they can be permanently more liquid. So just you can have a sort of a – what I'm
imagining is that kind of like we have these organizations in the United States that
are like Annenberg Political Fact Check or stuff that's like Get Out the Vote or
something where it's like people who are trying to make democracy work better.

1413
02:06:22,099 --> 02:06:39,099
But instead, we could have our own thing where we have like a Frederick Hayek
version where we raise a bunch of money and we just donate it to the market. You can
donate to a market and the blockchain will enforce all the rules and make sure that
it just goes split among all the traders all the time.

1414
02:06:40,099 --> 02:06:52,099
But again, the way it works is the traders don't get this money until the trading is
over, and in the meanwhile, it makes the seesaw much heavier. So a given dollar on
one side won't move it as much. So it's actually very, very cool.

1415
02:06:53,099 --> 02:06:56,099
Yeah, and it would be a lot better than this scammy fact check.

1416
02:06:56,099 --> 02:07:08,099
It's very, very cool, and it's also really – yeah, I know because it's – the problem
with any other thing, anything that's not a prediction market basically, is that you
have the who watches the watchman problem.

1417
02:07:08,099 --> 02:07:09,099
Exactly.

1418
02:07:09,099 --> 02:07:10,099
So it's going to regress.

1419
02:07:11,099 --> 02:07:24,099
These scammy fact check fucking people, they just really have an agenda anyway. So
they'll say something they disagree with is like partially true or only partially
false, and then there's some other thing that's bullshit.

1420
02:07:24,099 --> 02:07:29,099
They'll attach some other bias to it. So these fact checkers are just liars anyway.

1421
02:07:30,099 --> 02:07:48,099
It's unsustainable. So if someone did very, very well as a fact checker, that would
only plant the seeds of the future destruction because it's only a matter of time
before – no matter how great it is before the people who made it great quit or die
or become bribed or get drunk with power so to speak.

1422
02:07:48,099 --> 02:07:54,099
So it's only – the better it is, it just means that it will – just that much earlier
before it's ruined.

1423
02:07:54,099 --> 02:08:04,099
Exactly, because we need to enforce anarchy. There needs to be no rulers. There
needs to always be just a self-correcting sort of sustainable system that's just
operating with us, the users, interacting with it.

1424
02:08:04,099 --> 02:08:08,099
There shouldn't be authorities that are ruling things. That's the wrong way to go.

1425
02:08:08,099 --> 02:08:14,099
But yeah, I like the way – so I get high level of what you're talking about. So this
is how a lot of market makers operate, right?

1426
02:08:14,099 --> 02:08:24,099
So you have a starting price and if you're making markets in futures, for example,
you have to put your bids and offers in and if people – if you're trading against
the users who are going mega long, then you'll adjust the price as they are getting
mega long.

1427
02:08:24,099 --> 02:08:33,099
You're going to be going – moving up your bid-offer spread and stuff. So this is
very similar to some basic market making algos that you'll see in the markets out
there.

1428
02:08:33,099 --> 02:08:41,099
So I think it's intriguing. I think we should have a separate talk probably in the
future on only this because I think a lot of – we're traders here.

1429
02:08:41,099 --> 02:08:54,099
So a lot of us are interested in prediction markets and how you've been tackling
this and I think many of us probably want to help you out in testing the development
and any other sort of ways that we could help.

1430
02:08:55,099 --> 02:09:06,099
But does anyone out there have sort of Drivechain related questions? So I've reduced
the talk power in this channel now so that many of you, if not most of you, should
be able to just talk.

1431
02:09:06,099 --> 02:09:19,099
So if any of you are suspicious or skeptical of some of the things that Paul has
said and you want to challenge him, now is your time to sort of challenge anyone
that wants to – or has just comments or feedback or anything.

1432
02:09:19,099 --> 02:09:22,099
Now is a great time to get that.

1433
02:09:25,099 --> 02:09:29,099
Flipper, maybe you or Sean, anyone, Sheffield, anyone out there.

1434
02:09:36,099 --> 02:09:43,099
Well, I guess we could just say Drivechains versus merge mining. Is there a
difference?

1435
02:09:44,099 --> 02:09:56,099
I think yes. Drivechain, again, I think is the – I would describe as the system for
moving – for walking the funds back from the sidechain to the main chain.

1436
02:09:56,099 --> 02:10:04,099
And so what the sidechain does to produce its own blocks is its own business. So I
think they would be different, yes.

1437
02:10:05,099 --> 02:10:12,099
Although, again, I say in practice I expect – I think it would be crazy. I don't
know why anyone would pass up.

1438
02:10:12,099 --> 02:10:19,099
It's like your friend lives in a beautiful mansion and he lets you stay there for
free for merge mining.

1439
02:10:19,099 --> 02:10:30,099
And it's like – or you have no money and you could just live on the street and do
your own thing, but you have no resources because all you have is the – whatever
transaction fees you think you might be able to get.

1440
02:10:31,099 --> 02:10:35,099
So you're like a beggar in the street or you're a guy who's unemployed, like trying
to look for a job.

1441
02:10:35,099 --> 02:10:42,099
But so it's like that's your alternative to just staying in your super rich friend's
Bitcoin house that has like a ton of hash rate.

1442
02:10:42,099 --> 02:10:45,099
I just – I don't know why people would not take that.

1443
02:10:45,099 --> 02:10:51,099
Well, what if they want performance? What if they want performance, right? They
don't want to suffer under the Bitcoin confirmation time.

1444
02:10:52,099 --> 02:10:57,099
Well, it's like – you're talking about the interblock time of 10 minutes?

1445
02:10:57,099 --> 02:10:58,099
Yes.

1446
02:10:58,099 --> 02:11:04,099
I don't know. Like I think if people want to do those things, I think –

1447
02:11:04,099 --> 02:11:15,099
I'm imagining implementing say EOS on a Drivechain to Bitcoin instead of an atomic
swap. What would be the difference there?

1448
02:11:16,099 --> 02:11:27,099
I think – well, first of all, I think it is possible to do a kind of merged mining
that has another layer like in between where you have a second proof of work inside
of that.

1449
02:11:27,099 --> 02:11:31,099
So it's like building a shed on your person's mansion property or something.

1450
02:11:31,099 --> 02:11:42,099
You have like its own tinier thing and you would have to set it up so that
ultimately all the fees went to the main chain miners because otherwise it wouldn't
be incentive compatible.

1451
02:11:43,099 --> 02:11:54,099
But if you did, you could have like interblocks. You could have like blocks that
were in between the blocks and then you could increase the interblock time or you
could rather – you could decrease.

1452
02:11:54,099 --> 02:11:59,099
You could increase the frequency of blocks and you could make it so that there were
more blocks in 10 minutes.

1453
02:11:59,099 --> 02:12:09,099
But I think most of the people who – no offense to you, sir, but most of the people
who try to alter that parameter, I think they are just like toying with powerful
forces.

1454
02:12:10,099 --> 02:12:17,099
There's not really a lot of upside in my view and a lot of potential craziness.

1455
02:12:17,099 --> 02:12:23,099
But that is again exactly why we have the sidechains so that people can try out
their crazy ideas.

1456
02:12:23,099 --> 02:12:25,099
But I don't see –

1457
02:12:25,099 --> 02:12:28,099
You're referring to your criticism of proof of stake for example?

1458
02:12:28,099 --> 02:12:29,099
Oh, well, yeah.

1459
02:12:29,099 --> 02:12:30,099
I read your article.

1460
02:12:30,099 --> 02:12:57,099
I do think proof of stake just is proof of work but like with weird paint on it that
says stake and that it ultimately is like kind of – also smuggles in many potential
technical issues that are not even relevant to the economic problems that I brought
up, which is that it's kind of – basically achieves the same result with similar
resource drain on the economy.

1461
02:12:58,099 --> 02:13:02,099
But yeah, I don't know why people would want faster blocks.

1462
02:13:02,099 --> 02:13:06,099
Usually, you want – you could set something up in a block.

1463
02:13:06,099 --> 02:13:15,099
You could do something that's very lightning-like or you can set something up in a
block and then you can do things like inside that new layer.

1464
02:13:15,099 --> 02:13:20,099
So this would be like layer 2 of a layer 2 if a sidechain is layer 2.

1465
02:13:20,099 --> 02:13:22,099
So it would be like a layer 3 I suppose.

1466
02:13:22,099 --> 02:13:26,099
And I think that is much more stable way of setting all this stuff up.

1467
02:13:26,099 --> 02:13:31,099
But again, I think it would be possible to – I'm not sure.

1468
02:13:31,099 --> 02:13:37,099
Again, I can't guarantee the behavior because it's a really strange idea in a couple
of ways.

1469
02:13:37,099 --> 02:13:43,099
But you should be able to make faster blocks even with merge mining and with blind
merge mining.

1470
02:13:43,099 --> 02:13:44,099
But yeah, I don't know why.

1471
02:13:44,099 --> 02:13:47,099
I really don't recommend that at all.

1472
02:13:47,099 --> 02:13:57,099
But like I said, how would you compare say the atomic swap version of transacting
between say Bitcoin and EOS versus the Drivechain version?

1473
02:13:57,099 --> 02:14:01,099
I don't think I really understood what you mean.

1474
02:14:01,099 --> 02:14:06,099
Like atomic swaps are something that I would want in Drivechain.

1475
02:14:06,099 --> 02:14:11,099
They would be like helping the Andreas part so that Andreas and his customers didn't
have to trust each other.

1476
02:14:11,099 --> 02:14:15,099
So they would be a nice thing to have.

1477
02:14:15,099 --> 02:14:19,099
Are you referring to just someone atomic swapping with an altcoin?

1478
02:14:19,099 --> 02:14:20,099
Is that what you're referring to?

1479
02:14:20,099 --> 02:14:21,099
I don't know.

1480
02:14:21,099 --> 02:14:26,099
No, it was a bit sideways because I forgot about your Andreas example.

1481
02:14:26,099 --> 02:14:37,099
And I guess the Drivechain ecosystem would have to have atomic swaps as part of it
to solve that problem for somebody who wants to be the arbiter in the middle.

1482
02:14:37,099 --> 02:14:41,099
Yes, that three-month delay would be like a real killer otherwise.

1483
02:14:42,099 --> 02:14:49,099
But since you can specialize and have one person, that person would be like –
Andreas would be like a patience specialist or something.

1484
02:14:49,099 --> 02:14:52,099
It would just be his job to be more patient than other people.

1485
02:14:52,099 --> 02:14:55,099
And since you can just do that, then there's no reason why not.

1486
02:14:55,099 --> 02:14:58,099
And when I originally wrote the post, I had it at like two weeks or something.

1487
02:14:58,099 --> 02:15:09,099
But then over time, we just made it easier and easier for the Andreas character and
I just made it longer and longer because you get basically free security out of that
in my opinion.

1488
02:15:09,099 --> 02:15:15,099
So hypothetically, somebody wins the bet on Trump's reelection in Hivemind.

1489
02:15:15,099 --> 02:15:19,099
They get paid out in Truthcoin.

1490
02:15:19,099 --> 02:15:26,099
OK, I'll explain because I think this is kind of complicated and there's just a lot
of moving pieces even though they are simple I think.

1491
02:15:26,099 --> 02:15:34,099
The end point of your explanation, take us from winning the bet in Hivemind to
getting the BTC back on your normal mainnet wallet.

1492
02:15:34,099 --> 02:15:35,099
OK, no problem.

1493
02:15:35,099 --> 02:15:36,099
I'll start even earlier.

1494
02:15:36,099 --> 02:15:43,099
So let's say you show up in Bitcoin and you have – let's just say you mine a block
and you have 12.5 Bitcoin.

1495
02:15:43,099 --> 02:15:46,099
So on mainchain world, and then you would deposit it into the sidechain.

1496
02:15:46,099 --> 02:15:49,099
You pay a transaction fee on mainchain Bitcoin.

1497
02:15:49,099 --> 02:15:52,099
So now you have 12.49, but it shows up on the sidechain.

1498
02:15:52,099 --> 02:15:55,099
It shows up in the Bitcoin Hivemind blockchain.

1499
02:15:55,099 --> 02:16:00,099
Then you buy a bunch of shares of Donald Trump's reelection and I don't know what
they are now.

1500
02:16:00,099 --> 02:16:02,099
I think they're like – he's like 60%.

1501
02:16:02,099 --> 02:16:03,099
I don't know.

1502
02:16:03,099 --> 02:16:07,099
Actually, someone should check electionbettingodds.com and see what it is.

1503
02:16:07,099 --> 02:16:09,099
I think you have to adjust for the other Republican candidates.

1504
02:16:09,099 --> 02:16:11,099
It's like divide or something.

1505
02:16:11,099 --> 02:16:16,099
So I don't know what it is, but let's just say it's – you buy him for 60%.

1506
02:16:16,099 --> 02:16:27,099
So you get whatever that would be, that 12.49 divided by 0.6 number of shares.

1507
02:16:28,099 --> 02:16:30,099
Let's say he actually wins.

1508
02:16:30,099 --> 02:16:32,099
So you don't actually wait until the bet is resolved.

1509
02:16:32,099 --> 02:16:34,099
I wouldn't.

1510
02:16:34,099 --> 02:16:38,099
Day after election day or even during at like 8.30 p.m.

1511
02:16:38,099 --> 02:16:45,099
when all the major news networks announce that he has won and the challenger calls
him to resign.

1512
02:16:45,099 --> 02:16:53,099
The price will be – the price of the seesaw, everyone will have been fleeing the
loser end of the seesaw.

1513
02:16:53,099 --> 02:16:57,099
Or excuse me, the seesaw edge that is up that has less money on it.

1514
02:16:57,099 --> 02:17:00,099
Everyone will be withdrawing because those people can still get money for a while.

1515
02:17:00,099 --> 02:17:06,099
But eventually that will go to the top and it will hit zero and then those less
people will get nothing.

1516
02:17:06,099 --> 02:17:09,099
And at that point, you will be able to sell.

1517
02:17:09,099 --> 02:17:17,099
As soon as that happens, you will be able to sell your shares that you purchased for
60 cents for – or 60, I would say 0.6.

1518
02:17:17,099 --> 02:17:21,099
The shares that you purchased for 60 percentage points, you'd be able to sell them
for 100 percentage points.

1519
02:17:21,099 --> 02:17:23,099
So then you'll have more Bitcoin.

1520
02:17:23,099 --> 02:17:32,099
You will have – in fact, you have exactly whatever it is, 12.49 divided by 0.6 minus
any transaction fees you paid.

1521
02:17:32,099 --> 02:17:34,099
So you'll have that many shares.

1522
02:17:38,099 --> 02:17:40,099
You got cut off there it seems.

1523
02:17:40,099 --> 02:17:42,099
Yeah, sorry. I just wanted to divide it.

1524
02:17:42,099 --> 02:17:46,099
So you'd have 20.83 roughly.

1525
02:17:46,099 --> 02:17:49,099
You have like 20.8 Bitcoin at that point.

1526
02:17:49,099 --> 02:17:56,099
And then you would go to Andreas' service, which would hopefully be trustless and
have all kinds of cool atomic swaps in it.

1527
02:17:56,099 --> 02:17:59,099
Or you'd go to Eric Voorhees or whatever maybe.

1528
02:17:59,099 --> 02:18:02,099
Probably Eric Voorhees would have a thing for this.

1529
02:18:02,099 --> 02:18:05,099
I don't know about modern Eric Voorhees, right?

1530
02:18:05,099 --> 02:18:09,099
I think we all like – what's the fate of ShapeShift?

1531
02:18:09,099 --> 02:18:11,099
It's a little unclear at this point.

1532
02:18:11,099 --> 02:18:15,099
But whoever it is, you send your 20.83 to them.

1533
02:18:15,099 --> 02:18:17,099
They charge you 1 percent.

1534
02:18:17,099 --> 02:18:21,099
So then you get 20.6 out on the other end.

1535
02:18:21,099 --> 02:18:23,099
And then you've got it back on mainchain Bitcoin.

1536
02:18:23,099 --> 02:18:24,099
So you get 20.6.

1537
02:18:24,099 --> 02:18:25,099
So that's it.

1538
02:18:25,099 --> 02:18:27,099
You went from 12.5 to 20.6.

1539
02:18:27,099 --> 02:18:30,099
That's the voyage.

1540
02:18:32,099 --> 02:18:34,099
All right.

1541
02:18:38,099 --> 02:18:40,099
Well, Swap, are we done?

1542
02:18:41,099 --> 02:18:42,099
I mean, yeah.

1543
02:18:42,099 --> 02:18:52,099
If anyone has any other questions either on DragTrain or Hivemind stuff, I mean,
we've been going at it now like four hours or so.

1544
02:18:52,099 --> 02:18:54,099
I'm excited.

1545
02:18:54,099 --> 02:18:57,099
I agree with Roger on it.

1546
02:18:57,099 --> 02:18:58,099
Yeah, no.

1547
02:18:58,099 --> 02:18:59,099
I mean, it's great stuff.

1548
02:18:59,099 --> 02:19:03,099
I mean, Augur butchered the whole fucking idea.

1549
02:19:03,099 --> 02:19:10,099
And that's why I'm glad I dumped all my fucking 20x to my investment and just got
the fuck out of that shit.

1550
02:19:10,099 --> 02:19:13,099
And we'll see what the backholders there – what their fate is.

1551
02:19:13,099 --> 02:19:20,099
But I'd much rather, yeah, see how this Truthcoin – I still, in my head, think of it
Truthcoin.

1552
02:19:20,099 --> 02:19:21,099
I got to get used to Hivemind.

1553
02:19:21,099 --> 02:19:22,099
That's the sort of name.

1554
02:19:22,099 --> 02:19:24,099
But I know it.

1555
02:19:24,099 --> 02:19:27,099
I know it from very going back to as a Truthcoin.

1556
02:19:27,099 --> 02:19:29,099
And so this is in my head how I still reference it.

1557
02:19:29,099 --> 02:19:36,099
But I think it sounds very interesting, the sort of progress you've made since then
in terms of how to solve this market-making and stuff.

1558
02:19:36,099 --> 02:19:39,099
So I'm going to definitely investigate more.

1559
02:19:39,099 --> 02:19:46,099
I think a lot of people here are, in particular, as traders, this is something of
very high interest for us.

1560
02:19:46,099 --> 02:19:48,099
As well, I got one random question.

1561
02:19:48,099 --> 02:19:49,099
Sure, sure.

1562
02:19:49,099 --> 02:19:54,099
Paul, do you have any opinion on the new crop of stablecoins coming out?

1563
02:19:54,099 --> 02:19:59,099
And if you do have one, how could they interact with Drivechains?

1564
02:19:59,099 --> 02:20:07,099
That's an interesting question because stablecoins are – I actually think – I
actually wrote about this.

1565
02:20:07,099 --> 02:20:13,099
So let me actually find the link and I'll post it in, and then I will talk a little
bit.

1566
02:20:13,099 --> 02:20:14,099
But let me find it for one second.

1567
02:20:14,099 --> 02:20:19,099
Telegram is still down. This is so annoying.

1568
02:20:19,099 --> 02:20:20,099
It's back.

1569
02:20:20,099 --> 02:20:21,099
Is it?

1570
02:20:21,099 --> 02:20:22,099
Why is it still showing –

1571
02:20:22,099 --> 02:20:23,099
Fake news!

1572
02:20:23,099 --> 02:20:28,099
It's still showing like clocks and shit for me where it's not loading.

1573
02:20:33,099 --> 02:20:38,099
Okay, so basically the funny thing is I don't know about – I wrote this a long time
ago.

1574
02:20:38,099 --> 02:20:40,099
It's from January 2015.

1575
02:20:40,099 --> 02:20:47,099
So I don't know how well it's held up given the popularity of Tether, but I suppose
I could be vindicated at any moment if Tether explodes.

1576
02:20:47,099 --> 02:20:53,099
But the thing is the BitUSD, it tries to be the best of all the worlds.

1577
02:20:53,099 --> 02:21:01,099
It tries to say it's the stability of the money you already have plus all the cool
blockchain features.

1578
02:21:01,099 --> 02:21:05,099
But in my opinion, it kind of ends up being the worst of both worlds instead.

1579
02:21:05,099 --> 02:21:11,099
It says there's no upside, so you can't make a ton of money by investing in this.

1580
02:21:11,099 --> 02:21:19,099
And it has all these weird technical and trust and new technology kind of hangups
and issues.

1581
02:21:19,099 --> 02:21:28,099
So I don't really think that it is – but it's funny because the one context that I
think it does work is that it works inside of prediction markets very well.

1582
02:21:28,099 --> 02:21:39,099
And the other kind of crazy thing is that prediction markets create – they can
create stable coins all by themselves that are like fully collateralized and
enforced by the blockchain.

1583
02:21:39,099 --> 02:21:46,099
So I think they're a major upgrade because what you do is you just have – you
deposit money in and then you have someone bet.

1584
02:21:46,099 --> 02:21:52,099
You bet that the Bitcoin exchange rate will go down inside of a prediction market.

1585
02:21:52,099 --> 02:21:54,099
So will the exchange rate go up or down?

1586
02:21:54,099 --> 02:22:01,099
And you bet on down and then other arbitrageurs will bet on up and kind of it will
cancel out in the back end when they do arbitrage.

1587
02:22:01,099 --> 02:22:11,099
But basically what's happening is the people who bet that the exchange rate will go
down, if they're right, they get more Bitcoin.

1588
02:22:11,099 --> 02:22:15,099
And they get exactly as much Bitcoin as they lost in US dollar value terms.

1589
02:22:16,099 --> 02:22:18,099
So you get more – so it ends up being the same thing.

1590
02:22:18,099 --> 02:22:24,099
You get more Bitcoin but you still have – you start with $20 worth of Bitcoin, which
is only some amount.

1591
02:22:24,099 --> 02:22:27,099
The price of Bitcoin plummets by 70%.

1592
02:22:27,099 --> 02:22:30,099
You get much more Bitcoin though when you sell.

1593
02:22:30,099 --> 02:22:33,099
So you still end up with $20 worth of Bitcoin.

1594
02:22:33,099 --> 02:22:41,099
So it actually kind of ends up doing the same thing and this is the thing I skipped
over explaining before, which is that you can have these markets that take on many
different questions at once.

1595
02:22:42,099 --> 02:22:51,099
So what you do is something really crazy where you bet that Donald Trump would win
and the Bitcoin exchange rate would go down and that the US employment rate would go
up or something.

1596
02:22:51,099 --> 02:23:05,099
And none of these are – they all – not only do they not – there's no liquidity
problem but they actually vastly amplify each other's liquidity because it's –
again, it's like a – instead of a seesaw, it's like a dish and some of the money is
like spilling into the center.

1597
02:23:05,099 --> 02:23:08,099
And so it actually ends up being really, really cool.

1598
02:23:08,099 --> 02:23:16,099
I think people are going to really like it when they understand it but it's going to
be very hard to explain unless you want to look at the Excel sheet and even then it
will probably be very hard to explain.

1599
02:23:16,099 --> 02:23:24,099
But when people see it – the fun thing about this type of thing is that when
people's money is at stake, their ability to learn things just like goes through the
roof.

1600
02:23:24,099 --> 02:23:29,099
They suddenly become – there's like expert students.

1601
02:23:29,099 --> 02:23:36,099
So I think once this is hammered out in the real world, everyone will learn it all
very quickly.

1602
02:23:37,099 --> 02:23:43,099
But you can in fact construct a stable coin inside of a prediction market world and
you can use that to bet.

1603
02:23:43,099 --> 02:23:52,099
But you can do even cooler things like if you want to bet on something that's going
to happen two or three years from now, you might think something like – what am I
going to do?

1604
02:23:52,099 --> 02:23:55,099
Because the price of Bitcoin might go way up.

1605
02:23:55,099 --> 02:24:00,099
So why should I invest in shares of whatever, something that is going to happen?

1606
02:24:00,099 --> 02:24:03,099
Why should I invest in Donald Trump's reelection?

1607
02:24:03,099 --> 02:24:06,099
Because sure, I might be able to double my money.

1608
02:24:06,099 --> 02:24:13,099
But if I just keep it in crypto, I might multiply it by 10X or who knows what might
happen to it.

1609
02:24:13,099 --> 02:24:17,099
So maybe you think that's a good idea or maybe you think that's a bad idea.

1610
02:24:17,099 --> 02:24:20,099
You want the money in crypto or do you want it in US dollars?

1611
02:24:20,099 --> 02:24:23,099
It's like a very confusing knot to untie.

1612
02:24:24,099 --> 02:24:33,099
So what you can do is you can bet that the US stock market will go down, like the
S&P 500.

1613
02:24:33,099 --> 02:24:37,099
You can bet against that while you bet against whatever your prediction market thing
is.

1614
02:24:37,099 --> 02:24:47,099
And then it's basically as if you transformed your crypto into shares of the
industrial average, Dow Jones Industrial Average or the S&P 500, and you wagered
those.

1615
02:24:47,099 --> 02:24:52,099
You said I'll wager you one Dow Jones Industrial Average share that Donald Trump
wins.

1616
02:24:52,099 --> 02:24:55,099
And if he wins, you owe me two Dow Jones Industrial Average shares.

1617
02:24:55,099 --> 02:24:58,099
And if he loses, then you can take mine.

1618
02:24:58,099 --> 02:25:10,099
And in that way, you can bet on things that will happen like five or six years from
now in a kind of sane way because otherwise it would be crazy because you'd have to
either bet in crypto or you'd have to bet even in stablecoin.

1619
02:25:10,099 --> 02:25:12,099
That would also be kind of crazy, right?

1620
02:25:12,099 --> 02:25:14,099
Because what if the price of Bitcoin goes really well?

1621
02:25:14,099 --> 02:25:21,099
Then you'll really regret having bet in stablecoin because then you'll be getting it
back in stablecoin.

1622
02:25:21,099 --> 02:25:25,099
And you could have had it in crypto, that 100X or something.

1623
02:25:25,099 --> 02:25:31,099
Have you ever considered hiring Swapman to program in some leverage?

1624
02:25:31,099 --> 02:25:38,099
It's funny you mentioned that because you can do – you can't do – it's known.

1625
02:25:39,099 --> 02:25:40,099
I don't know.

1626
02:25:40,099 --> 02:25:53,099
You guys are traders, but I'm an academic kind of from – and I took some financial
engineering classes and a call option is equal to just having a lot of leverage in
the underlying.

1627
02:25:53,099 --> 02:26:04,099
So if you buy a call option to buy something, I don't know, a share of – you buy a
share of GE with leverage, that's the same as buying a call option in GE.

1628
02:26:05,099 --> 02:26:07,099
It depends on the specific numbers, exactly how much of each you have.

1629
02:26:07,099 --> 02:26:10,099
It's not quite, but I understand what you're trying to say.

1630
02:26:10,099 --> 02:26:14,099
I trade options and futures, so I know the domains very well.

1631
02:26:14,099 --> 02:26:15,099
Yeah, exactly.

1632
02:26:15,099 --> 02:26:23,099
So all you need to do – so in this, someone creates a question and you can create
the market, but you can also create funny versions of the question.

1633
02:26:23,099 --> 02:26:25,099
So you can say here's the underlying question.

1634
02:26:25,099 --> 02:26:32,099
It's going to resolve somewhere between zero and one based on the US stock market's
performance within some range, a big min-max range.

1635
02:26:32,099 --> 02:26:39,099
You could say maybe it will be that number squared or it will be that number
logarithm or some other weird other thing.

1636
02:26:39,099 --> 02:26:47,099
You could use that for all kinds of fancy things, but one thing is you can just say,
okay, we're going to do a call option transform on that.

1637
02:26:47,099 --> 02:26:54,099
So if it expires above six, then we recast it where 6.6 is zero and one is one.

1638
02:26:54,099 --> 02:27:04,099
But then if it's below six, it's just fixed at zero, and then it has basically
become a call option, and then it has basically become leverage.

1639
02:27:04,099 --> 02:27:08,099
You would do that on the question side?

1640
02:27:08,099 --> 02:27:10,099
No, you would not do it on the question side.

1641
02:27:11,099 --> 02:27:19,099
The question is just what is the Dow Jones Industrial Average on this exact date or
what was rather because the people who are answering it will be in the future.

1642
02:27:19,099 --> 02:27:25,099
So today you would say what will the Dow Jones Industrial Average be on January 5th,
2020?

1643
02:27:25,099 --> 02:27:28,099
They would answer that question in like March of 2020.

1644
02:27:28,099 --> 02:27:36,099
They'd have to go back and check, but then you would bet on it today, and it would
be as if you had owned it because the price should track the whole time.

1645
02:27:37,099 --> 02:27:40,099
Because if it doesn't, then you just go – you buy low and sell high.

1646
02:27:40,099 --> 02:27:41,099
You do arbitrage.

1647
02:27:41,099 --> 02:27:46,099
So it's as if you have the Dow Jones Industrial Average inside of this system the
whole time.

1648
02:27:48,099 --> 02:28:02,099
Have you ever imagined doing leverage the other way where, say, if it's currently
50-50, I get to 10X a 50 bet, but then I'm busted if the price only moves 10% down.

1649
02:28:02,099 --> 02:28:05,099
So let's say to 45.

1650
02:28:06,099 --> 02:28:10,099
What you would do is you would not use that particular market.

1651
02:28:10,099 --> 02:28:24,099
I mean there's actually other ways of doing it because these things, when you buy
the shares, the shares are like themselves tokens, and you can trade those in the
real world using any kind of non-blockchain world, but that would be lame.

1652
02:28:24,099 --> 02:28:25,099
So you wouldn't want to do that.

1653
02:28:25,099 --> 02:28:35,099
But what you would want to do is you would want to hope that someone would make a
different type of market where – so the questions show up, and the question is what
is the Dow Jones Industrial Average?

1654
02:28:35,099 --> 02:28:43,099
And once that's up, someone might make 30 or 40 markets that all have different
amounts of leverage and different features, and some might plug in.

1655
02:28:43,099 --> 02:28:44,099
Someone would make one.

1656
02:28:44,099 --> 02:28:48,099
Would the Dow Jones Industrial Average be higher if we elected Donald Trump?

1657
02:28:48,099 --> 02:28:53,099
That would be Donald Trump election cross Dow Jones Industrial Average price.

1658
02:28:53,099 --> 02:29:18,099
But other people might just make one that's just Dow Jones Industrial Average with a
lot of liquidity in it, and one person might make one that says Dow Jones Industrial
Average with a range trimmed so that, as you say, it has much more price volatility
within certain real world value ranges of the Dow Jones Industrial Average.

1659
02:29:19,099 --> 02:29:24,099
But otherwise, it's just everyone loses their money here or something like that.

1660
02:29:24,099 --> 02:29:26,099
So you can kind of try to configure that.

1661
02:29:26,099 --> 02:29:29,099
The issue though is that, of course, you always need counterparties.

1662
02:29:29,099 --> 02:29:31,099
Otherwise, the liquidity will suck.

1663
02:29:31,099 --> 02:29:41,099
It will never go to zero because of this market scoring rule thing that we discussed
earlier, but it can still just suck where you can only trade $5 at a time, and
that's because you need counterparty.

1664
02:29:41,099 --> 02:29:46,099
If you have people that you're going to bet against, you can go all day back and
forth.

1665
02:29:48,099 --> 02:29:50,099
So it's a challenge.

1666
02:29:50,789 --> 02:29:53,789
I think you can go back and forth all day, can't you, Paul?

1667
02:29:55,789 --> 02:30:02,789
No, this has been fantastic. Your passion is so clear on this, and you're doing
everything for the right reasons.

1668
02:30:03,789 --> 02:30:11,789
So the question is, I guess, what's the funding model of both, whether it's
Drivechain or Truthcoin? What's the situation there?

1669
02:30:12,789 --> 02:30:25,789
I think it doesn't really matter because the – I mean, I would like to, since Roger
did actually help out way early on, I would like to cut him in on some of those,
what I call vote coins, what would be called rep.

1670
02:30:26,789 --> 02:30:33,789
But again, see, actually none of that actually makes any difference because in order
to be a user, you don't need any of that. That's only for the employees.

1671
02:30:33,789 --> 02:30:44,789
And you're obligated to report on the outcomes, so you're giving someone a task as
well. Those things may be valuable, but the true value is in enabling the service,
which is all BTC.

1672
02:30:45,789 --> 02:30:55,789
But the other thing is that the price of Bitcoin, ever since this stuff has been
discussed, everyone has gotten so wealthy that I'm not exactly sure that it's even
matters.

1673
02:30:56,789 --> 02:31:00,789
That's the point. I mean, Drivechain, do you guys need help with getting developers?

1674
02:31:00,789 --> 02:31:13,789
No, we don't need any funding, but I think obviously the more – if you have
intrinsically motivated, technically minded people who understand Bitcoin, that is –
those people, that's worth more than money.

1675
02:31:14,789 --> 02:31:26,789
Money is only worth what it can buy, right? You cannot just go out and say, hey, I'd
like to hire someone who knows about Bitcoin and who understands the scaling debate
and who understands various people and their concerns about centralization.

1676
02:31:26,789 --> 02:31:48,789
That's very difficult. So those people, if you are interested and you want to help,
or if you know someone who is interested and want to help, or if you want to hire
someone who is qualified and can help, then obviously that would be great if you
could go to drivechain.info, download the test DriveNet software, and then play with
it and find bugs.

1677
02:31:48,789 --> 02:31:56,789
Ideally, fix the bugs also, but just finding the bugs is also helpful and just that
kind of thing is very helpful.

1678
02:31:57,789 --> 02:32:06,789
I think especially in this space, the best product comes out of people who are just
doing it from their heart and people that are driven by passion.

1679
02:32:06,789 --> 02:32:34,789
You don't work 80 hours a week because, oh, my salary is really nice and high and
I'm super motivated. You work 80 hours a week because you are just super passionate
and driven about something, and that's what makes you spend all night not even
thinking you're spending all night to work on it.

1680
02:32:34,789 --> 02:32:51,789
And that's where you come up with something really great. So I was just sort of
curious if there was any gaps in the project that there's – whether there was some
kind of need to – like for example, over at Grin, they're doing a security audit
fundraiser kind of thing, and Monero does periodic fundraisers.

1681
02:32:51,789 --> 02:32:56,789
So I was just curious if you had any sort of coherent funding models there.

1682
02:32:56,789 --> 02:33:12,789
Well, we could. I mean I don't know. That type of thing – I mean the skill of being
a software developer and the skill of being like a cool blockchain like software
architect, those are two different skills.

1683
02:33:12,789 --> 02:33:32,789
And then the skill of like managing a team, that's a completely different skill,
like a team of managing a programming team or something. These are all very
different skills, and I really don't think – I've only got like a few – I've only
got a couple of little skills, and I really don't want to – those things, I have no
idea if they work or not.

1684
02:33:32,789 --> 02:33:43,789
It's not my area of expertise. I do worry a lot when money gets involved, then
people spend a lot of time squabbling over who gets the money and how, and it ends
up being a distraction.

1685
02:33:44,789 --> 02:34:03,789
And especially for stuff like security audit, does that really work or do we just
have to just take the bugs as they come and just try to fix them? I don't know. Has
a security audit ever really moved the needle from insecure to secure?

1686
02:34:03,789 --> 02:34:17,789
I don't know about that. I don't know if it's moved the needle in that sense, but I
think – wasn't it Augur or one of these others that they had an audit done and they
had to change the entire code language that they used or something from one thing to
solidity or something?

1687
02:34:18,789 --> 02:34:32,789
Yes, but you see in – that's also – either the needle was secure and then they just
did this for appeasement and then they moved it from secure to secure, which is to
say not at all, or it was broken and then they found a bunch of things.

1688
02:34:32,789 --> 02:34:46,789
But where is the final destination of this needle? It's probably just going to –
sure, they rewrite the whole thing as they were instructed, but probably given how
they reached that situation in the first place, probably the final destination of
that needle is just going to land on insecure again.

1689
02:34:46,789 --> 02:34:58,789
That's just the way I kind of – again, I don't really know. Maybe these things work
really well and I have no idea what's the best way to get a piece of software
produced.

1690
02:34:58,789 --> 02:35:10,789
I don't know that at all really. I just try to do the best I can to think about what
I think would be really cool, blockchain technology, and work on it like myself and
talk to other people who are working on it.

1691
02:35:11,789 --> 02:35:25,789
That's all you kind of get from me in that regard. I don't know the best – if we
raised a bunch of money, what would I do? Give it to the people who are already
working on the software just kind of arbitrarily or it's like a one-time bonus or
would I try to hire people?

1692
02:35:25,789 --> 02:35:38,789
What criteria would I use? It would just take more time away from me working to
interview people. I don't know. I have no idea how to do that. I just – I don't know
at all.

1693
02:35:38,789 --> 02:35:52,789
Well, I think that when it comes to any of these things, whenever there's a big idea
and people want to get involved and they just – people want to – the immediate thing
– I mean you have this GoFundMe scam economy that has emerged where people do dumb
shit.

1694
02:35:52,789 --> 02:35:58,789
They get attention for doing dumb shit and being stupid and they use it to raise
money. Like help me. I'm an idiot. Give me money for it.

1695
02:35:59,789 --> 02:36:10,789
So people react to this and they want to – people show their care for something by
sending money to it, whether it's a good thing or a bad thing. And so immediately
people think like, oh, I want to get involved.

1696
02:36:10,789 --> 02:36:20,789
I want to – I'm not necessarily – someone thinks maybe they're – I'm not necessarily
a coder. I'm not necessarily able to help in these other ways. So how do I – people
would want to show it by giving money, right?

1697
02:36:20,789 --> 02:36:23,789
How about if you took money and then used it for liquidity?

1698
02:36:23,789 --> 02:36:25,789
Right, yeah. I mean some –

1699
02:36:25,789 --> 02:36:37,789
Yeah, that would be really cool. That would actually be neat. We could do – but see
that – it would almost be better to do that as an ongoing basis though because you
could say – I could say – I completely agree with you.

1700
02:36:37,789 --> 02:36:39,789
No, I totally agree 100 percent.

1701
02:36:39,789 --> 02:36:43,789
Because remember you had that model about the people that could donate to the
contracts.

1702
02:36:43,789 --> 02:36:44,789
Exactly, yes.

1703
02:36:44,789 --> 02:36:49,789
Yeah, you could just make it – especially if you can economically incentivize people
to do that.

1704
02:36:50,789 --> 02:36:55,789
That would be an ongoing thing though because that's independent of producing this
piece of software.

1705
02:36:55,789 --> 02:37:10,789
But yeah, I completely agree. I would want to say something like – I would go out
there with a microphone at the charity gala or something and I would be like, well,
all these other people right now, they're raising all this money.

1706
02:37:11,789 --> 02:37:30,789
The RNC, the Republican National Committee, and the Democratic National Committee,
they're raising all this money to manipulate public opinion and they're spending all
this money on political advertising and they are going to – if they're elected,
they're going to get access to a huge budget and direct all these resources around
and they're going to give kickbacks to their friends.

1707
02:37:31,789 --> 02:37:49,789
And like I'm going to say, this huge amount of money is set against actual human
kind of prosperity and like honest elections, for lack of a better word, which
unfortunately, they're really – there's not a great word for that, what I'm trying
to describe.

1708
02:37:49,789 --> 02:38:07,789
But I'd go out there and say, look, there's all that. So if you want to fight back
against that, I can guarantee that 100% of all Bitcoin donated will be used exactly
for this purpose and this is exactly why it will help fix all those problems.

1709
02:38:07,789 --> 02:38:19,789
And yeah, I think if you got – you would be able to get – and again, I don't think
people – because I haven't explained it. You'd be able to make a market with many,
many dimensions at once. It would be huge.

1710
02:38:19,789 --> 02:38:34,789
You'd have like Donald Trump's election, the congressional election, unemployment
rate, life expectancy, all these things. You can put them all in one big giant super
seesaw and then you dump all the money in that and it would work on all of them at
once. It works fantastic.

1711
02:38:34,789 --> 02:38:46,789
So I would be able to kind of optimize the use, I think, and I'd say, look, this is
what you're buying. It all goes exactly to this purpose. And yeah, I think that
would be a really cool fundraising thing.

1712
02:38:46,789 --> 02:39:03,789
I do believe also that that is an essential part of the sort of futarky part of the
project because it is an issue often. These prediction markets, they do often need –
they need like a little bit of money to kind of get them going.

1713
02:39:03,789 --> 02:39:23,789
Once they get going, they do well, but there is a kind of vision of prediction
markets that they'll just fix everything kind of for free and as someone who has
been intensely involved in the prediction market world basically since I was like 12
or 13 years old, it never works that way.

1714
02:39:23,789 --> 02:39:37,789
You really do need some money to kickstart the engine, so to speak, and so I
definitely would be doing – I think that is a part of the future, the funding for
the market liquidity. I think that absolutely will happen.

1715
02:39:37,789 --> 02:39:56,789
Yeah, because as you say, there's like a surplus of capital in the system now.
There's a lot of us that have been around for a bit and –

1716
02:39:56,789 --> 02:40:21,789
Yeah, stop. I mean most of us have been plowing it into scam ICOs and multiplying
our money even more.

1717
02:40:21,789 --> 02:40:38,789
That game is over now, but it's nice to see quality open source projects and why not
give people a chance to help out and maybe sort of like make amends for all of the
scams that have been fueled over the years because of us.

1718
02:40:38,789 --> 02:40:52,789
You have the – and the Truthcoin model is for the employees of the corporation,
right? Would there be any way to tie that in to those who wanted to be say liquidity
providers and maybe get a cut of the trading fee?

1719
02:40:52,789 --> 02:41:11,789
Nothing prevents you from doing all of them at once. You could buy 1% of the total
rep, the total of Voltcoin, and you could create the question. You would buy a
question slot and you could create the market and add a bunch of liquidity to it.

1720
02:41:11,789 --> 02:41:27,789
At that point, you would be looking at at least 50% because you get 25% from –
everyone who uses the question gets the 25%, and everyone who uses the market – the
market creator gets 25%, the question creator gets 25% of all the trading fees.

1721
02:41:27,789 --> 02:41:44,789
So when people trade, there's a tiny – you got to take your cut to keep everything
moving. So there's a – not only is there a transaction fee for the miners, but
there's a trading fee, and that is what goes to the people who make the questions
and who make the markets.

1722
02:41:44,789 --> 02:42:05,789
Yeah, but what if somebody was neutral? Like what if somebody was neutral on the bet
but wanted to pile into the liquidity provision that the question creator is in?

1723
02:42:06,789 --> 02:42:30,789
I would imagine that some people would specialize in – they would just be like, oh,
all I do is I wait for the slots to become available, and I buy the slots, and then
I write the best questions, and they're the clearest questions in the world, and I'm
like an ex-lawyer, and I love English language, and I write super clear, and I put
all the hyperlinks in there or something or blah, blah, blah, whatever.

1724
02:42:30,789 --> 02:42:44,789
I know exactly what it is that people want, what they're going to disagree about
over the next three months, the next year or something. So I buy the slots, and I
write all the questions about what people will disagree about, and I write them very
clearly so that they work.

1725
02:42:44,789 --> 02:42:58,789
And then you buy the slots, and then in return you get 25% if lots of people use
them in markets and those markets have a lot of trading in them. So someone could
just do that or someone could just do – they could just create the markets.

1726
02:42:58,789 --> 02:43:08,789
So you could have a bunch of great questions already there, and you could say, you
know what? I bet people really want to bet on these two things at once or this other
thing is not being here.

1727
02:43:08,789 --> 02:43:18,789
So let's say here's a market. This is a really good market over there, market number
302. It's a perfect market, but it just doesn't have enough liquidity in it.

1728
02:43:18,789 --> 02:43:26,789
So I'm just going to copy that perfectly. I'm just going to copy market 302, and I'm
going to make it the new market, market 720.

1729
02:43:26,789 --> 02:43:38,789
720 is going to have the exact same questions and the exact same dimensions. So it
will be exactly the same, but I'm going to just put a ton of money in there as
liquidity, and you'll lose all that money up front, remember?

1730
02:43:38,789 --> 02:43:54,789
But you get 25% of any trading in this market. So if you're right about the market,
the only thing being wrong with it being that it needed a little bit more liquidity,
then you could potentially just on the 25% alone remake the money that you invested.

1731
02:43:54,789 --> 02:44:16,789
So it still depends on the trading. You never know. There's a lot of trading
sometimes. Like a foreign exchange market has like $5 trillion worth of trading
every 24-hour period. That's a ridiculous amount of money if you're taking your
tiny, tiny, tiny peanuts, 10,000th of a percent of that and you're splitting it
many, many ways.

1732
02:44:16,789 --> 02:44:37,789
What if you could just pile on into the one that already exists and get a stake in
the amount of trading fees? So it becomes a scale where if you're the original
creator, you have 100 of them, 100%. Then let's say, okay, I'm neutral on the
question, but I want to pile into that liquidity pool.