1
00:00:01,000 --> 00:00:06,000
This is epicenter episode 318 
with guest David vork. 

2
00:00:20,500 --> 00:00:24,300
Hi, welcome to epicenter. 
My name is Sebastian cuchillo, I

3
00:00:24,300 --> 00:00:29,000
hope everyone's having a good 
pre year-end holiday week to all

4
00:00:29,000 --> 00:00:31,600
of you who will be traveling 
over the next few days. 

5
00:00:32,000 --> 00:00:36,800
I wish you the best of luck and 
patience, I find that traveling 

6
00:00:36,800 --> 00:00:40,100
at this time of year can be 
particularly exhausting, but 

7
00:00:40,300 --> 00:00:44,500
Payoff of being with family and 
doing nothing for about a week 

8
00:00:44,800 --> 00:00:48,000
is always well worth it. 
So I arrived in Canada over the 

9
00:00:48,008 --> 00:00:50,700
weekend will be here for the 
next 10 days or so. 

10
00:00:50,700 --> 00:00:54,400
Before heading back to Europe, 
for the new year today, our 

11
00:00:54,400 --> 00:00:59,500
guest is David florrick, David 
is the founder and CEO of Saia 

12
00:01:00,100 --> 00:01:02,700
size an interesting project for 
a number of reasons. 

13
00:01:03,400 --> 00:01:05,800
Now, before doing this episode, 
I have to admit. 

14
00:01:06,500 --> 00:01:10,100
I was not extremely familiar 
with SIA but I'm really happy 

15
00:01:10,200 --> 00:01:12,000
That I was able to learn about 
it. 

16
00:01:12,700 --> 00:01:15,300
It's a decentralized file 
storage platform which has been 

17
00:01:15,300 --> 00:01:19,500
around since 2014. 
Now, needless to say, it's 

18
00:01:19,500 --> 00:01:22,000
definitely less talked about 
than other projects in this 

19
00:01:22,000 --> 00:01:27,100
category, I'm thinking of ipfs 
and storage for instance but I 

20
00:01:27,100 --> 00:01:30,300
think it deserves a lot more 
attention than it does. 

21
00:01:31,600 --> 00:01:35,900
Sia is a true decentralized 
Dropbox and its potential is 

22
00:01:35,900 --> 00:01:38,500
huge. 
The technology seems quite 

23
00:01:38,500 --> 00:01:43,200
mature and the cost of Files on 
Sia is surprisingly affordable 

24
00:01:44,100 --> 00:01:48,200
and as someone who runs his own 
at home Cloud Server, I can 

25
00:01:48,200 --> 00:01:50,900
definitely see sayaka 
complimenting or straight up 

26
00:01:50,900 --> 00:01:53,900
replacing that set up. 
Sonny. 

27
00:01:53,900 --> 00:01:57,300
And I talked to David about the 
history of the project, how it 

28
00:01:57,308 --> 00:02:00,600
works under the hood and some of
the really interesting features 

29
00:02:00,600 --> 00:02:04,500
of has like seed based file 
recovery which allows you to 

30
00:02:04,500 --> 00:02:09,300
recover your entire file 
Library, just to seat which is 

31
00:02:09,300 --> 00:02:10,800
one of my favorite things. 
About this product. 

32
00:02:10,800 --> 00:02:14,300
We also got into David's views 
on proof of work. 

33
00:02:14,600 --> 00:02:18,700
David's, a strong proponent of 
proof of work and even started a

34
00:02:18,700 --> 00:02:21,700
company, which builds minors for
Sia. 

35
00:02:22,000 --> 00:02:27,100
Now a few years ago, there was 
some drama surrounding a hard 

36
00:02:27,100 --> 00:02:31,600
Fork of the Sia protocol which 
was somewhat contentious going 

37
00:02:31,600 --> 00:02:34,100
into the interview. 
I actually had a little context 

38
00:02:34,100 --> 00:02:36,300
for what happened during that 
hard fork. 

39
00:02:36,300 --> 00:02:40,200
And so my reaction to it was as 
genuine as it could have been. 

40
00:02:40,200 --> 00:02:44,400
And anyway, I'm curious if you 
all knew if this and what you 

41
00:02:44,400 --> 00:02:47,800
think about what happened, so 
it's quite possible that we have

42
00:02:47,900 --> 00:02:51,000
David back on the show in the 
near future. 

43
00:02:51,600 --> 00:02:55,200
He's quite outspoken on the 
topic of trusted setups in zkp 

44
00:02:55,200 --> 00:02:58,100
systems. 
In fact, I first encountered 

45
00:02:58,100 --> 00:03:02,200
David, when he gave a talk on 
trusted setups at Stark 

46
00:03:02,200 --> 00:03:05,300
recessions and Tel Aviv. 
But unfortunately, we didn't 

47
00:03:05,300 --> 00:03:08,800
have time to get into this as it
deserves an entire episode on 

48
00:03:08,800 --> 00:03:11,500
its own. 
Before you go to the interview, 

49
00:03:11,500 --> 00:03:13,100
I'd like to tell you about our 
sponsors. 

50
00:03:13,100 --> 00:03:17,100
For today's episode Papo is a 
community of creators sharing 

51
00:03:17,100 --> 00:03:20,100
short video content on crypto 
and Entrepreneurship. 

52
00:03:20,300 --> 00:03:23,600
And on Peppo, you show your love
for Content creators with Peppa 

53
00:03:23,600 --> 00:03:25,600
coins. 
When you like something, you 

54
00:03:25,600 --> 00:03:27,800
send Peppa coins. 
When you want to take part in a 

55
00:03:27,800 --> 00:03:30,600
conversation, you show your 
interest by putting up coins as 

56
00:03:30,600 --> 00:03:32,900
well. 
So if you want to try it out, 

57
00:03:32,900 --> 00:03:35,700
now is the perfect time because 
the Peppa team has launched the 

58
00:03:35,700 --> 00:03:39,100
home for the holidays challenge.
So as we're all spending time 

59
00:03:39,100 --> 00:03:42,100
with family over them, Few 
weeks, it's inevitable that 

60
00:03:42,100 --> 00:03:45,900
someone at some point will ask 
you to explain what is Krypto 

61
00:03:46,300 --> 00:03:48,900
the challenge is to share how 
you explain crypto to your 

62
00:03:48,900 --> 00:03:52,400
family in a 30-second video. 
And those are two thousand 

63
00:03:52,400 --> 00:03:56,100
dollar prize pool and pepper 
coins for the top three videos 

64
00:03:56,100 --> 00:03:59,700
as voted by the community. 
And to be honest, your odds of 

65
00:03:59,700 --> 00:04:02,000
winning, this are actually 
pretty good because as I'm 

66
00:04:02,000 --> 00:04:04,400
recording this, there's only 
about 15 entries. 

67
00:04:05,100 --> 00:04:07,700
So to participate in the 
challenge, download the Peppo 

68
00:04:07,700 --> 00:04:10,000
app on your Android or iOS 
device. 

69
00:04:10,200 --> 00:04:13,200
And log in with your Twitter 
account, then open your mobile 

70
00:04:13,200 --> 00:04:17,500
browser and go to epicenter, 
rocks / Peppo as pep. 

71
00:04:17,500 --> 00:04:20,899
Oh, and that'll take you right 
to the contest thread and then 

72
00:04:20,899 --> 00:04:24,300
reply with your unique. 
What is crypto explanation by 

73
00:04:24,300 --> 00:04:28,200
December 20th at 12 p.m. 
New York time to enter the 

74
00:04:28,200 --> 00:04:30,000
contest. 
And while you're there, you can 

75
00:04:30,000 --> 00:04:32,800
also check out my entry and send
me some peppercorns if you like.

76
00:04:33,400 --> 00:04:35,500
We like to thank Peppo for the 
support of the podcast. 

77
00:04:36,700 --> 00:04:38,500
We're also brought to you by 
etoro. 

78
00:04:39,000 --> 00:04:40,700
Now if you're interested in 
getting into the financial 

79
00:04:40,700 --> 00:04:42,700
markets, you don't know where to
start. 

80
00:04:42,900 --> 00:04:47,600
Itõs a great place to begin 
building your portfolio, utero 

81
00:04:47,600 --> 00:04:51,000
makes trading easy because you 
get access to stocks bonds, 

82
00:04:51,000 --> 00:04:54,900
ETFs, indices, Commodities, 
currencies and crypto. 

83
00:04:55,000 --> 00:04:59,900
In one single easy-to-use 
platform utero is unlike any 

84
00:04:59,900 --> 00:05:01,700
trading platform you've ever 
experienced. 

85
00:05:02,000 --> 00:05:04,500
If you're used to using your 
Banks trading and investment 

86
00:05:04,500 --> 00:05:07,500
dashboard your interest 
something Ting if you're used to

87
00:05:07,500 --> 00:05:11,100
using your Banks trading or 
investing dashboard you're in 

88
00:05:11,100 --> 00:05:12,700
for something vastly different 
here. 

89
00:05:13,100 --> 00:05:15,700
And that's because each row is a
social trading platform. 

90
00:05:16,000 --> 00:05:19,000
When you join etoro, you're 
joining a community of 12 

91
00:05:19,000 --> 00:05:20,800
million Traders from around the 
world. 

92
00:05:21,100 --> 00:05:23,800
They're talking about trading 
sharing charts and they're 

93
00:05:23,800 --> 00:05:27,500
talking about crypto to the best
thing about etoro is the copy 

94
00:05:27,500 --> 00:05:30,000
Trader feature. 
You can automatically copy the 

95
00:05:30,000 --> 00:05:34,000
trades of the top crypto Traders
at the exact price in real time.

96
00:05:34,300 --> 00:05:36,900
Now these are people who spend a
lot of I'm researching the 

97
00:05:36,900 --> 00:05:39,300
market developing their 
strategies and building their 

98
00:05:39,300 --> 00:05:42,600
portfolio and you can copy their
trade so that whatever gains 

99
00:05:42,600 --> 00:05:45,600
they make, you'll make the same 
profits proportional to your 

100
00:05:45,600 --> 00:05:48,400
investment. 
And one of my favorite things 

101
00:05:48,400 --> 00:05:52,100
about copy Trader is that I get 
to see the risk profile of each 

102
00:05:52,100 --> 00:05:54,800
of these Traders. 
So I can build my own strategies

103
00:05:54,800 --> 00:05:58,500
investing certain amounts with 
low medium or high risk Traders,

104
00:05:58,500 --> 00:06:02,500
and learning from their traits. 
So to get started, go to 

105
00:06:02,500 --> 00:06:07,400
etoro.com, that's ET Oro. 
Your account and start trading 

106
00:06:07,500 --> 00:06:10,600
and copying trades today. 
Now, of course, this is not 

107
00:06:10,600 --> 00:06:13,600
investment advice. 
This is my personal opinion and 

108
00:06:13,800 --> 00:06:16,800
you should always consider the 
risks and investing, especially,

109
00:06:16,800 --> 00:06:19,200
with crypto currencies. 
As they are highly volatile. 

110
00:06:19,800 --> 00:06:22,300
We'd like to thank each row for 
those support of the podcast. 

111
00:06:23,100 --> 00:06:26,100
And with that, here's our 
interview with David work. 

112
00:06:28,600 --> 00:06:31,300
We're here with David vorik. 
David is the co-founder of the 

113
00:06:31,300 --> 00:06:33,300
sign at work. 
David, thanks for joining us. 

114
00:06:34,000 --> 00:06:37,000
Thanks glad to glad to be here 
to tell us. 

115
00:06:37,000 --> 00:06:39,200
How did you get involved in 
crypto? 

116
00:06:39,200 --> 00:06:40,900
And what's your background? 
What were you doing before? 

117
00:06:41,100 --> 00:06:42,400
Getting into the grip of the 
space. 

118
00:06:43,400 --> 00:06:49,800
Yeah, so I got involved freshman
year of college, this is in 2011

119
00:06:49,800 --> 00:06:54,600
and basically what happened was 
a someone from my dorm, just pop

120
00:06:54,600 --> 00:06:57,600
their head, through my doorway 
and said, you should check out 

121
00:06:57,600 --> 00:06:59,400
Bitcoin, and then they 
disappeared. 

122
00:06:59,600 --> 00:07:02,300
So I was like, one sentence that
changed my life. 

123
00:07:02,400 --> 00:07:04,400
It was your own little Satoshi. 
Yeah. 

124
00:07:05,600 --> 00:07:07,500
So, he didn't even care that 
much. 

125
00:07:07,500 --> 00:07:10,400
But Bitcoin wasn't wasn't really
his thing, but he knew it would 

126
00:07:10,400 --> 00:07:12,700
be my thing and he was 
definitely right. 

127
00:07:12,800 --> 00:07:17,800
So, I I just basically from the 
age of 18 dived, really deep 

128
00:07:17,800 --> 00:07:21,800
into Bitcoin started, studying 
it, looking at it, experimenting

129
00:07:21,800 --> 00:07:23,900
with it, trying to find ways to 
make it better. 

130
00:07:24,400 --> 00:07:28,300
And by the time I graduated I 
decided to start a company 

131
00:07:28,800 --> 00:07:31,200
around blockchain and 
cryptocurrency that ended up 

132
00:07:31,200 --> 00:07:35,800
being the Sian at work. 
So yeah, sigh of platform was 

133
00:07:35,800 --> 00:07:39,600
started by me and my co-founder 
we both did it straight out of 

134
00:07:39,608 --> 00:07:43,500
college so we don't have any 
anything but Tween. 

135
00:07:43,600 --> 00:07:46,300
So I guess I've always 
professionally been a Bitcoin 

136
00:07:46,300 --> 00:07:49,500
guy. 
What was it about you that you 

137
00:07:49,500 --> 00:07:52,200
know your friend knew that you 
would be super attracted to like

138
00:07:52,200 --> 00:07:55,500
working on bitcoin stuff? 
That's a great question. 

139
00:07:55,500 --> 00:07:58,900
I think he knew that I was 
pretty libertarian. 

140
00:07:59,000 --> 00:08:02,200
I was pretty pretty excited 
about Technologies like tour and

141
00:08:02,200 --> 00:08:05,900
BitTorrent and Bitcoin kind of 
falls under the same like 

142
00:08:06,300 --> 00:08:11,900
internet and you know, fight the
man gain independence, sort of 

143
00:08:12,200 --> 00:08:14,900
philosophy. 
So I think I think he knew that 

144
00:08:14,900 --> 00:08:21,700
there would be a match there. 
So you started Sia in 2014. 

145
00:08:22,000 --> 00:08:27,300
Yep. 
And so, I mean, I only recently 

146
00:08:27,300 --> 00:08:31,300
learned about Sia, when did you 
guys go live or what are the 

147
00:08:31,300 --> 00:08:33,700
were the different sort of, 
like, iterations of that? 

148
00:08:33,700 --> 00:08:36,500
Because it seems that sighs, you
know, gotten much more attention

149
00:08:36,500 --> 00:08:40,299
recently than it did like, you 
know, 2015 or 2016. 

150
00:08:41,400 --> 00:08:46,900
Yeah, so we launched the first 
version of the network, in 2015,

151
00:08:46,900 --> 00:08:52,300
June June 2015, which seems kind
of insane to say that from from 

152
00:08:52,300 --> 00:08:56,900
here because at that point we 
were just like panicked about, 

153
00:08:57,000 --> 00:08:59,600
you know, is storage going to 
come to Market first as file 

154
00:08:59,600 --> 00:09:00,900
coin. 
Going to come to Market first, 

155
00:09:00,900 --> 00:09:04,900
like, we really wanted to be the
first platform that worked when 

156
00:09:04,900 --> 00:09:09,100
we were and then it was like, it
took all the way until 2019. 

157
00:09:09,100 --> 00:09:12,100
When are we've launched? 
For there to really be a 

158
00:09:12,100 --> 00:09:15,200
decentralized like competitor. 
But yeah. 

159
00:09:15,200 --> 00:09:19,400
So we first launched this, I 
Network way back in 2015 and 

160
00:09:19,400 --> 00:09:21,600
then we've just been working on 
it, building it. 

161
00:09:21,600 --> 00:09:24,200
I think we're a bunch of 
Builders more than, like, 

162
00:09:24,200 --> 00:09:27,800
marketers are businessmen. 
And so, as a team, especially 

163
00:09:27,800 --> 00:09:30,300
early on before we had, you 
know, grown and expanded, and 

164
00:09:30,300 --> 00:09:33,000
kind of filled out the roles 
that we were missing, we were 

165
00:09:33,000 --> 00:09:35,100
just focus on building up the 
technology. 

166
00:09:35,500 --> 00:09:39,100
And now today, we actually have 
people on the team who are 

167
00:09:39,100 --> 00:09:42,500
responsible for growing. 
And product strategy and like 

168
00:09:42,500 --> 00:09:45,000
customer fit. 
And so I think that that's 

169
00:09:45,000 --> 00:09:48,000
really where more of the 
attention is coming from is we 

170
00:09:48,000 --> 00:09:51,600
have we have non Builders or 
people who do more than just 

171
00:09:51,700 --> 00:09:56,300
write code helping sigh out. 
Yeah, it was our marking that 

172
00:09:56,400 --> 00:09:58,500
you guys have a pretty 
impressive blog presence. 

173
00:09:58,500 --> 00:10:02,400
Like your blog is pretty 
frequently, you know, seeing new

174
00:10:02,400 --> 00:10:04,700
articles and things and like 
Community updates quite 

175
00:10:04,700 --> 00:10:06,400
regularly. 
So I think that's quite good. 

176
00:10:07,300 --> 00:10:12,400
So you mentioned storage and 
ipfs, explain how you differ 

177
00:10:12,400 --> 00:10:16,900
from those two products. 
So I think in our audience a lot

178
00:10:16,900 --> 00:10:18,500
of people will be familiar with 
ipfs. 

179
00:10:19,000 --> 00:10:22,500
We also did an episode with 
storage like years ago, but 

180
00:10:22,500 --> 00:10:27,400
yeah, how is he a different Yes,
I think Chief way we distinguish

181
00:10:27,400 --> 00:10:30,600
from our competitors that we're 
fully decentralized. 

182
00:10:30,700 --> 00:10:36,800
So at no point in the standard 
like Sia flow or architecture, 

183
00:10:37,300 --> 00:10:39,700
do you have to deal with the 
centralized party? 

184
00:10:40,000 --> 00:10:43,000
And so, like storage is pretty 
transparent about the fact that 

185
00:10:43,000 --> 00:10:45,200
their architecture has 
centralized elements to it. 

186
00:10:45,200 --> 00:10:47,400
They have, you know, satellites 
and coordinators, and they, you 

187
00:10:47,400 --> 00:10:51,400
know, they need if storage. 
The company disappears, the 

188
00:10:51,400 --> 00:10:55,400
network stops working ipfs. 
It also has like a similar 

189
00:10:55,400 --> 00:10:58,600
weakness in that. 
Like when you put data on ipfs, 

190
00:10:58,600 --> 00:11:01,300
it's not guaranteed to stick 
around, unless you contract, a 

191
00:11:01,308 --> 00:11:04,100
centralized like a pending 
service like in fear or 

192
00:11:04,100 --> 00:11:05,900
something. 
And so, on the side Network, 

193
00:11:06,300 --> 00:11:08,900
none of these limitations exist.
Who I, you know, today, when you

194
00:11:08,900 --> 00:11:13,200
put data on the sign Network, it
goes up in a permanent way. 

195
00:11:13,200 --> 00:11:16,100
And and because you're paying 
for it, you don't have to worry 

196
00:11:16,100 --> 00:11:16,800
about pinning. 
It. 

197
00:11:16,800 --> 00:11:19,700
You don't have to find a 
centralized party if our company

198
00:11:19,700 --> 00:11:22,400
disappears tomorrow. 
You know the website would go 

199
00:11:22,400 --> 00:11:26,700
down but the people So using 
this, I app would not notice. 

200
00:11:26,800 --> 00:11:29,100
There would be no decrease in 
speed. 

201
00:11:29,100 --> 00:11:31,500
You would, you know, you'd still
be able to upload and download 

202
00:11:31,500 --> 00:11:35,400
and form new contracts and so in
that sense size like a fully 

203
00:11:35,400 --> 00:11:39,500
fully independent Network Okay. 
So do you think that people are 

204
00:11:39,500 --> 00:11:43,900
aware of the fact that, you 
know, ipfs for instance, has 

205
00:11:44,300 --> 00:11:47,600
these centralized elements and 
why does it matter? 

206
00:11:48,800 --> 00:11:52,300
So, I think most developers that
are building applications on top

207
00:11:52,300 --> 00:11:55,700
of ipfs are aware because they 
either, either they run into the

208
00:11:55,708 --> 00:11:58,500
problem that their files 
disappear or they are actively 

209
00:11:58,500 --> 00:12:00,600
Contracting. 
These centralized services. 

210
00:12:01,100 --> 00:12:04,100
And I think it's pretty 
transparent from that side of 

211
00:12:04,100 --> 00:12:06,200
things. 
I don't know if users people who

212
00:12:06,200 --> 00:12:10,500
use apps that use The fs are 
aware that these centralized 

213
00:12:10,500 --> 00:12:15,000
elements are in play, but I 
think it is important because 

214
00:12:15,700 --> 00:12:18,900
what it means is that some 
third-party has the ability to 

215
00:12:18,900 --> 00:12:21,600
shut down the app. 
If in fira decides that they're 

216
00:12:21,600 --> 00:12:25,500
not serving your apps files 
anymore, you know, with one 

217
00:12:25,500 --> 00:12:28,500
executive decision, someone, you
know, some leader at the 

218
00:12:28,500 --> 00:12:32,500
inferior company can decide to 
Cripple your entire application 

219
00:12:33,300 --> 00:12:36,600
and that really like 
decentralization is about 

220
00:12:36,600 --> 00:12:39,000
building things. 
Things that other people don't 

221
00:12:39,000 --> 00:12:42,200
control. 
And so we see, we see that as a 

222
00:12:42,200 --> 00:12:46,900
huge weakness and really we want
to build a web and like an 

223
00:12:46,900 --> 00:12:50,600
application system where there 
is no counter party, who can 

224
00:12:50,600 --> 00:12:55,000
decide to turn off your access 
to an application or an app. 

225
00:12:56,400 --> 00:12:59,600
But isn't it true that you could
use ipfs in? 

226
00:12:59,600 --> 00:13:01,300
I mean like so you're 
Contracting out these 

227
00:13:01,300 --> 00:13:05,500
centralized services but it is 
possible as well to utilize like

228
00:13:05,500 --> 00:13:10,600
layers on top of ipfs to build a
decentralized network of storage

229
00:13:10,600 --> 00:13:11,600
providers. 
Is it Mom? 

230
00:13:13,100 --> 00:13:16,500
I think that remains to be seen 
the ipfs team would say that it 

231
00:13:16,500 --> 00:13:18,800
is possible. 
However, no one's done such a 

232
00:13:18,808 --> 00:13:22,000
thing and I think that there 
would be substantial challenges,

233
00:13:22,400 --> 00:13:24,200
another. 
Another thing I'd point out 

234
00:13:24,200 --> 00:13:28,300
about ipfs and I think this is 
also something developers on top

235
00:13:28,300 --> 00:13:31,400
of ipfs are well aware of is 
that the architecture is just 

236
00:13:31,400 --> 00:13:34,600
fundamentally very very slow. 
So if you want to open up a file

237
00:13:34,600 --> 00:13:37,700
especially one that doesn't get 
access very often, it's going to

238
00:13:37,700 --> 00:13:39,800
take you many seconds to get 
that file open. 

239
00:13:40,300 --> 00:13:44,000
And the way, the Sigh network is
built is fundamentally Much much

240
00:13:44,000 --> 00:13:46,500
faster. 
Not only is it is it much easier

241
00:13:46,500 --> 00:13:51,000
and like the storage element of 
SIA is we started with storage 

242
00:13:51,000 --> 00:13:53,000
first? 
And then we're moving on to 

243
00:13:53,200 --> 00:13:56,100
these other elements second 
whereas whereas ipfs, the 

244
00:13:56,100 --> 00:14:00,100
storage has to be like bolted on
later and that really I think 

245
00:14:00,100 --> 00:14:02,400
has a big impact to the user 
experience. 

246
00:14:03,300 --> 00:14:06,200
I'd be first since I are not 
exactly, you know, comparable 

247
00:14:06,200 --> 00:14:08,700
because one is more of a Content
delivery Network and the other 

248
00:14:08,700 --> 00:14:11,600
is sort of a decentralized cloud
service. 

249
00:14:11,600 --> 00:14:15,900
Very different Use cases, right?
And even that like you know, 

250
00:14:15,900 --> 00:14:19,700
maybe a closer comparison would 
be to like file coin but even 

251
00:14:19,700 --> 00:14:22,100
that I think there's some quite 
a few differences. 

252
00:14:22,100 --> 00:14:25,300
And so maybe like before we 
start to like delve more into 

253
00:14:25,300 --> 00:14:28,900
these comparison, maybe we 
should get a good base ground of

254
00:14:28,900 --> 00:14:33,500
what Sia is trying to do, what 
is the product here. 

255
00:14:33,900 --> 00:14:38,000
Sia is a progression. 
So it's a technology that we are

256
00:14:38,000 --> 00:14:40,500
iterating rapidly on the very 
first thing. 

257
00:14:40,500 --> 00:14:42,800
That's, I could really do is 
archival storage. 

258
00:14:42,800 --> 00:14:47,200
So if you have data that you 
want to backup or protect this, 

259
00:14:47,200 --> 00:14:51,400
I now work is a very fast, very 
efficient, very cost-effective 

260
00:14:51,400 --> 00:14:55,600
solution, robust, you know, 
decentralized place, where you 

261
00:14:55,600 --> 00:14:58,900
can store your data. 
So, first and foremost, like, 

262
00:14:58,900 --> 00:15:02,600
Sia is a in archival platform or
a backup solution. 

263
00:15:02,600 --> 00:15:06,600
But the Reason that that's what 
it is first is because that's 

264
00:15:06,800 --> 00:15:11,200
what we felt as you look at more
things you can do with 

265
00:15:11,200 --> 00:15:14,200
distributed data or 
decentralized data that was the 

266
00:15:14,200 --> 00:15:16,900
easiest thing to build. 
So that's where we put all of 

267
00:15:16,900 --> 00:15:18,600
our attention because above 
everything else. 

268
00:15:18,600 --> 00:15:23,000
We wanted a practical 
decentralized storage platform 

269
00:15:23,100 --> 00:15:28,500
as fast as possible so that V1 
would be sort of like a personal

270
00:15:28,600 --> 00:15:30,300
cloud. 
Let's say I had a bunch of files

271
00:15:30,300 --> 00:15:33,100
on my Dropbox instead of putting
them on Dropbox, I would go 

272
00:15:33,100 --> 00:15:36,000
ahead and put them. 
And the scien at work, but I 

273
00:15:36,000 --> 00:15:39,600
wouldn't be serving a website, 
for example, from this V1. 

274
00:15:40,500 --> 00:15:43,700
Yeah, so V1 doesn't do do 
website, serving, or content 

275
00:15:43,700 --> 00:15:47,200
delivery. 
However, the two does do website

276
00:15:47,200 --> 00:15:52,200
serving and content delivery and
so as we're finishing up, you 

277
00:15:52,200 --> 00:15:57,100
know, the final edges on this 
archival storage or like object 

278
00:15:57,100 --> 00:16:00,400
storage platform. 
Really, where that moves on. 

279
00:16:00,400 --> 00:16:03,800
If you think about the pieces, 
we have hundreds. 

280
00:16:03,900 --> 00:16:08,300
Of hosts all over the world. 
You know, near many major cities

281
00:16:08,600 --> 00:16:11,600
if you want to download a file 
and it's on the sign Network, 

282
00:16:11,600 --> 00:16:13,800
the sign network is going to 
have the you know the 

283
00:16:13,800 --> 00:16:16,700
distribution in the locality in 
the ability to serve that file 

284
00:16:16,700 --> 00:16:19,500
very quickly. 
And so we've been building, you 

285
00:16:19,500 --> 00:16:23,300
know, from day one starting in 
2014 with the idea of eventually

286
00:16:23,300 --> 00:16:28,000
turning into known ipfs style, 
content delivery Network or you 

287
00:16:28,000 --> 00:16:30,200
know competing with like saying 
Akamai. 

288
00:16:30,700 --> 00:16:33,500
That's just we've known that 
that's going to be step two 

289
00:16:33,500 --> 00:16:34,300
knots. 
One. 

290
00:16:34,400 --> 00:16:37,700
So that's kind of where we're 
we're pointed for the future. 

291
00:16:38,900 --> 00:16:41,400
So is this a V2? 
What's live today? 

292
00:16:41,400 --> 00:16:43,800
Or is it V1 that's live right 
now. 

293
00:16:44,600 --> 00:16:48,500
V1 is live right now and we are 
actively building both to make 

294
00:16:48,500 --> 00:16:53,500
the one better and to bring V-22
production, I see. 

295
00:16:53,700 --> 00:16:56,700
So let's talk a little bit of 
the start off by talking about 

296
00:16:56,700 --> 00:17:00,000
V1 and that and like some of the
economics and design around that

297
00:17:00,000 --> 00:17:02,200
and then we can shift towards 
V2. 

298
00:17:02,500 --> 00:17:05,900
So, in this V1, could you tell 
us a little bit about? 

299
00:17:05,900 --> 00:17:08,500
Okay, I have a, you know, on my 
hard drive. 

300
00:17:08,700 --> 00:17:13,500
Currently I have a movie 
collection of 100 gigabytes and 

301
00:17:13,500 --> 00:17:16,300
I want to get rid of my hard 
drive and put it on the side 

302
00:17:16,300 --> 00:17:19,800
Network and you walk me through 
what is the process of doing 

303
00:17:19,800 --> 00:17:21,700
that? 
Both what I'm doing as a user 

304
00:17:21,700 --> 00:17:24,099
and then what's happening 
technically behind the scenes. 

305
00:17:25,400 --> 00:17:27,900
Yeah. 
So as a user, what you do is you

306
00:17:27,900 --> 00:17:31,200
go to our website, you download 
Sia or if you want to get it 

307
00:17:31,200 --> 00:17:35,600
some other way, you could clone 
the GitHub repo or get lab clone

308
00:17:35,600 --> 00:17:37,800
the get lab repo and build it 
yourself. 

309
00:17:37,900 --> 00:17:41,800
Either way, you're gonna have to
get get the sigh of software and

310
00:17:41,800 --> 00:17:45,400
run it, then it's going to have 
to sink the chain and it's going

311
00:17:45,400 --> 00:17:48,100
to need funding. 
So the science Network 

312
00:17:48,100 --> 00:17:50,900
philosophy is that everything 
runs on money when you upload 

313
00:17:50,900 --> 00:17:53,100
the cost money when you store 
data that cost money, when you 

314
00:17:53,100 --> 00:17:56,500
download that cost money and 
it's All on a big Open 

315
00:17:56,500 --> 00:17:58,600
Marketplace. 
So, things are generally 

316
00:17:59,000 --> 00:18:01,700
extremely cost competitive 
because every single host 

317
00:18:01,700 --> 00:18:04,400
honest, I network is competing 
with every other host on this, I

318
00:18:04,400 --> 00:18:08,400
Network and price is a big 
factor in yeah, which which 

319
00:18:08,400 --> 00:18:11,200
hosts get selected. 
So that prices are constantly 

320
00:18:11,200 --> 00:18:13,600
being pushed down on this, I 
Network. 

321
00:18:14,000 --> 00:18:17,400
So you're going to have to give 
this I client Society coins. 

322
00:18:17,400 --> 00:18:19,700
And then you have to set up an 
allowance, which is basically 

323
00:18:19,700 --> 00:18:23,100
just like a control mechanism so
it knows how much money it's 

324
00:18:23,100 --> 00:18:25,000
allowed to spend that way. 
It doesn't. 

325
00:18:25,100 --> 00:18:27,200
It Go and spend, you know, a 
thousand dollars a month or 

326
00:18:27,200 --> 00:18:29,800
something. 
But really, that's just a, like 

327
00:18:29,800 --> 00:18:33,500
a safety feature, more than more
than it like, a critical 

328
00:18:33,500 --> 00:18:35,600
element. 
Once you're there, once you're 

329
00:18:35,600 --> 00:18:38,800
set up, you'll be able to just 
upload your files to the sign 

330
00:18:38,800 --> 00:18:41,600
Network. 
So you will either, you know, 

331
00:18:41,600 --> 00:18:45,100
drag-and-drop or, you know, open
the folder, uploader order a 

332
00:18:45,108 --> 00:18:49,600
couple couple ways to put files 
onto the side client, but once 

333
00:18:49,600 --> 00:18:51,100
you start uploading them, it'll 
go ahead. 

334
00:18:51,500 --> 00:18:53,700
It will saturate Your Home 
Connection. 

335
00:18:54,100 --> 00:18:57,500
So, PSI networking Do about 400 
megabits per second. 

336
00:18:57,800 --> 00:19:00,000
Most people's home connections 
are not that fast. 

337
00:19:00,200 --> 00:19:04,000
So for 100 gigabytes it might 
take you, you know, one day a 

338
00:19:04,000 --> 00:19:07,000
full day of uploading to get 
that onto the Sigh Network at 

339
00:19:07,000 --> 00:19:09,800
which point your data is on the 
sign Network and you can 

340
00:19:09,800 --> 00:19:13,400
download it anytime Saya today 
supports video streaming. 

341
00:19:13,400 --> 00:19:17,000
So in your case, you gave the 
movies as an example. 

342
00:19:17,300 --> 00:19:20,500
If you wanna watch your movies, 
this is actually something you 

343
00:19:20,500 --> 00:19:21,900
can do directly from the sign 
out work. 

344
00:19:21,900 --> 00:19:24,200
So you don't need, you don't 
even have to download them and 

345
00:19:24,200 --> 00:19:26,300
then watch them. 
You can just stream them 

346
00:19:26,300 --> 00:19:29,900
directly from the sign at work 
and then the other thing that 

347
00:19:29,900 --> 00:19:34,200
the cyan Network supports which 
I think is really critical to 

348
00:19:34,200 --> 00:19:36,400
making it a practical. 
Backup solution. 

349
00:19:36,500 --> 00:19:40,400
Is seat base file recovery. 
So once all your data is 

350
00:19:40,400 --> 00:19:44,400
uploaded and it's in good health
and the you, I will show you you

351
00:19:44,400 --> 00:19:48,100
know the health of your files 
you can create a snapshot. 

352
00:19:48,200 --> 00:19:50,800
So there's a there's a tool to 
walk through the says create a 

353
00:19:50,800 --> 00:19:54,600
snapshot it'll make a snapshot, 
it will upload the snapshot, 

354
00:19:54,600 --> 00:20:00,300
it'll take Maybe an hour to get 
everything in place and then a 

355
00:20:00,300 --> 00:20:03,300
little report that the Snapchat 
was successful and has 

356
00:20:03,300 --> 00:20:05,200
completed. 
And what that means is that your

357
00:20:05,200 --> 00:20:08,800
wallet seed that you use to fund
your account. 

358
00:20:09,200 --> 00:20:11,900
Also now has access to your 
data. 

359
00:20:12,300 --> 00:20:15,800
So if you lose your computer and
you have to start over, you get 

360
00:20:15,800 --> 00:20:19,400
a new computer you install Sia, 
and you, you know, you open up 

361
00:20:19,400 --> 00:20:23,800
the type in your seed screen, 
when you put your seat in, not 

362
00:20:23,800 --> 00:20:27,300
only will you get Your money 
back, like a traditional crypto 

363
00:20:27,300 --> 00:20:31,400
currency wallet, but you will 
also get that snapshot back and 

364
00:20:31,400 --> 00:20:34,600
you'll have all your data. 
What exactly is going into that 

365
00:20:34,600 --> 00:20:40,300
snapshot Beyond just my crypto 
private key, for example, Yeah. 

366
00:20:40,400 --> 00:20:43,900
So when you upload data to the 
sign Network, while you're doing

367
00:20:43,900 --> 00:20:46,800
is you're storing it on a bunch 
of machines around the world. 

368
00:20:47,100 --> 00:20:52,900
Each piece of data ends up on 30
different machines and to make 

369
00:20:52,900 --> 00:20:57,300
things fast you have to, we 
don't have like a DHT or any 

370
00:20:57,300 --> 00:20:59,500
sort of look at mechanism. 
So what you need to do is you 

371
00:20:59,500 --> 00:21:03,300
have to remember where each 
piece is, you're going to have 

372
00:21:03,300 --> 00:21:05,800
encryption key for that piece, 
you're going to have that IP 

373
00:21:05,800 --> 00:21:08,300
address of the host, and you're 
going to have the content ID 

374
00:21:08,700 --> 00:21:12,700
that you Give to the host to get
the piece back and so for 100 

375
00:21:12,700 --> 00:21:16,000
gigabytes might have, you know, 
50,000 pieces on it. 

376
00:21:16,300 --> 00:21:18,000
So you're going to have to 
remember all the metadata for 

377
00:21:18,000 --> 00:21:20,900
those 50,000 pieces. 
So, what a, what a snapshot is 

378
00:21:21,400 --> 00:21:26,500
is, essentially, it bundles up 
all that metadata and stores it 

379
00:21:26,500 --> 00:21:30,300
on the sign Network in a guess, 
kind of complicated here, but in

380
00:21:30,300 --> 00:21:33,800
a way that we can recover it 
using only your file seat, 

381
00:21:34,100 --> 00:21:37,000
basically, you give the you 
distribute the metadata to the 

382
00:21:37,008 --> 00:21:39,500
hosts in a very specific. 
A turn. 

383
00:21:39,900 --> 00:21:42,500
When you're doing a recovery 
later, you can ask posts for a 

384
00:21:42,508 --> 00:21:46,700
specific location of data rather
than a specific content address 

385
00:21:47,300 --> 00:21:50,900
that location will unpack to all
of your metadata. 

386
00:21:52,400 --> 00:21:56,300
That's really cool. 
So essentially you could store 

387
00:21:56,300 --> 00:22:01,800
files in Sia and this opens up 
some interesting use cases. 

388
00:22:01,800 --> 00:22:05,500
So someone could do like an 
interesting kind of inheritance 

389
00:22:05,500 --> 00:22:10,200
planning scheme where, you know,
they upload something to Sia. 

390
00:22:10,400 --> 00:22:14,000
It could even be like, encrypted
blockchain private keys, for 

391
00:22:14,000 --> 00:22:18,900
instance, and then have a single
key that's stored somewhere. 

392
00:22:18,900 --> 00:22:23,000
And then as part of a will, for 
example, that key gets Given to 

393
00:22:23,000 --> 00:22:26,200
the benefit of the beneficiary 
and then that beneficiary uses 

394
00:22:26,200 --> 00:22:28,500
that key to get files. 
It could be pictures, it could 

395
00:22:28,500 --> 00:22:33,200
be letters or whatever, but 
you'd be sort of like a lock box

396
00:22:33,200 --> 00:22:36,700
where you keep files until like 
a later date, it would be 

397
00:22:36,700 --> 00:22:38,800
interesting. 
Also, if you know, if you guys 

398
00:22:38,800 --> 00:22:42,600
were to implement some sort of 
time lock mechanism, where like 

399
00:22:42,600 --> 00:22:45,900
a seed could only unlock files. 
I'm just like thinking out loud 

400
00:22:45,900 --> 00:22:47,000
here. 
But like, we're a see could only

401
00:22:47,000 --> 00:22:49,900
unlock files after a single, 
like, after some time has passed

402
00:22:49,900 --> 00:22:52,400
or something like that. 
Yeah, off the top of my Head. 

403
00:22:52,400 --> 00:22:55,800
I don't know if there's an easy 
way to do that but definitely 

404
00:22:56,000 --> 00:22:59,500
aside from the time control 
thing like creating a lockbox on

405
00:22:59,500 --> 00:23:02,700
size something that you could do
today, I believe. 

406
00:23:03,600 --> 00:23:05,900
Yeah, that's cool. 
So let's talk about the the 

407
00:23:05,900 --> 00:23:10,000
architecture a little bit 
because Sia is a little 

408
00:23:10,000 --> 00:23:13,200
different from what we already 
talked about how different it is

409
00:23:13,200 --> 00:23:16,200
from the other file storage 
systems that exist. 

410
00:23:16,600 --> 00:23:20,400
But it leverages a lot of 
different technologies that we 

411
00:23:20,400 --> 00:23:22,300
all know, and understand and the
crypto A. 

412
00:23:22,300 --> 00:23:25,700
So it has some sort of a 
blockchain mechanism, but it 

413
00:23:25,700 --> 00:23:29,200
also has file storage and also 
uses payment channels and and 

414
00:23:29,200 --> 00:23:31,000
encryption. 
So talk about the technical 

415
00:23:31,000 --> 00:23:34,200
architecture and how all those 
those building blocks fit 

416
00:23:34,200 --> 00:23:38,100
together. 
Yeah, so philosophically, when 

417
00:23:38,100 --> 00:23:43,700
we were building style, we use 
to like bedrocks, just two 

418
00:23:43,700 --> 00:23:46,200
requirements. 
The first requirement is that it

419
00:23:46,200 --> 00:23:51,200
has to be fully decentralized. 
So that decentralization always 

420
00:23:51,200 --> 00:23:54,800
Trump's every Other decision, if
it's not decentralized, then 

421
00:23:54,800 --> 00:23:56,700
that we don't make. 
You know, we don't go down that 

422
00:23:56,700 --> 00:24:01,900
route but then the second thing 
that we always focus on it and 

423
00:24:01,900 --> 00:24:03,600
we ask ourselves every time we 
build. 

424
00:24:03,600 --> 00:24:07,500
Something is how fast is? 
This, is it as fast as possible?

425
00:24:07,700 --> 00:24:11,000
Like is there theoretically, a 
faster way to do something and 

426
00:24:11,000 --> 00:24:13,500
if the answer is yes, there's 
theoretically faster ways to do 

427
00:24:13,500 --> 00:24:17,100
it. 
We make sure that we build so 

428
00:24:17,100 --> 00:24:20,300
that performance is ahead of 
everything else. 

429
00:24:20,300 --> 00:24:24,700
And so what that means on the 
sign, Today is basically and 

430
00:24:24,700 --> 00:24:28,300
moving forward, that the whole 
thing is set up to be completely

431
00:24:28,300 --> 00:24:31,300
point to point. 
So, it was something that a lot 

432
00:24:31,300 --> 00:24:34,400
of distributed systems do, is 
they have some sort of look up 

433
00:24:34,400 --> 00:24:38,300
routing or content, hashing or 
the they'll have these 

434
00:24:38,300 --> 00:24:41,900
distributed algorithms to find 
data, and that means that you 

435
00:24:41,900 --> 00:24:44,200
have to ask someone who has to 
ask, someone who has to ask 

436
00:24:44,200 --> 00:24:47,100
someone and that, like, 
multi-hop step takes a lot of 

437
00:24:47,108 --> 00:24:50,600
time on the sign Network. 
Every single request is point to

438
00:24:50,608 --> 00:24:52,600
point as soon as you want to. 
File. 

439
00:24:52,800 --> 00:24:55,600
You immediately know exactly 
who's storing the file. 

440
00:24:55,900 --> 00:24:58,800
You find that out in, you know, 
in under a microsecond. 

441
00:24:59,100 --> 00:25:02,900
And then after that, then you 
can do the the network request 

442
00:25:03,100 --> 00:25:05,500
to fetch the data. 
So I think that's something 

443
00:25:05,500 --> 00:25:07,900
that's really critical to us. 
So another thing is like 

444
00:25:07,900 --> 00:25:11,000
scalability, blockchains don't 
scale very well. 

445
00:25:11,000 --> 00:25:13,300
They only get a couple 
transactions per second on the 

446
00:25:13,300 --> 00:25:16,000
side Network. 
You know, it's I think during, 

447
00:25:16,000 --> 00:25:18,600
you know, peak times, the sine 
of work is probably seen more 

448
00:25:18,600 --> 00:25:21,100
than a million transactions in a
single hour. 

449
00:25:21,400 --> 00:25:23,100
And of course, that's Happening 
on chain. 

450
00:25:23,100 --> 00:25:26,100
That's all going over payment 
channels. 

451
00:25:26,100 --> 00:25:31,000
And so we use something similar 
to the inner Ledger protocol. 

452
00:25:31,000 --> 00:25:34,100
Basically every time you 
download data, you download a 

453
00:25:34,108 --> 00:25:37,100
small piece of data. 
So you pay for it first then you

454
00:25:37,100 --> 00:25:39,500
trust, the host is going to send
it to you. 

455
00:25:39,500 --> 00:25:42,400
So the host could could steal 
that small payment but it's very

456
00:25:42,400 --> 00:25:45,400
tiny and if they do you know 
they're dishonest, you know, not

457
00:25:45,400 --> 00:25:47,200
to use them anymore, you know, 
to penalize them. 

458
00:25:47,200 --> 00:25:51,900
And so there's very tiny amount 
of trust that you extend to the 

459
00:25:52,100 --> 00:25:53,800
Stand. 
They give you data then you know

460
00:25:53,800 --> 00:25:56,100
extend another tiny amount of 
trust then they give you the 

461
00:25:56,100 --> 00:25:59,900
data and this allows us to scale
very well you know we don't need

462
00:25:59,900 --> 00:26:02,600
a payment Channel system that's 
as complicated as the lightning 

463
00:26:02,600 --> 00:26:05,200
Network we do data overstate 
channels. 

464
00:26:05,200 --> 00:26:08,300
So you know we're not just 
sending money or simultaneously 

465
00:26:08,300 --> 00:26:12,100
updating the file contract for 
what the host is required to 

466
00:26:12,100 --> 00:26:15,900
store. 
So this is specifically for when

467
00:26:15,900 --> 00:26:19,000
I'm retrieving data, right? 
So what, when I'm, for example, 

468
00:26:19,000 --> 00:26:20,800
couldn't a new contract with 
someone. 

469
00:26:21,100 --> 00:26:26,000
Then I You know, we'll make a 
new what were they called file 

470
00:26:26,000 --> 00:26:28,600
contracts or something about 
contracts, have five contract 

471
00:26:28,900 --> 00:26:31,600
and then if I'm constantly 
receive requesting data, let's 

472
00:26:31,600 --> 00:26:33,600
say I'm streaming it like a 
movie or something. 

473
00:26:33,600 --> 00:26:35,300
Right. 
Then I use the payment Channel 

474
00:26:35,800 --> 00:26:39,600
but what if I don't want to be 
like retrieving data, very 

475
00:26:39,600 --> 00:26:41,200
often. 
I'm just using it primarily for 

476
00:26:41,200 --> 00:26:44,300
archival purposes. 
Do I still use payment channels?

477
00:26:44,300 --> 00:26:50,200
And that case Yep. 
So what we do is when you create

478
00:26:50,200 --> 00:26:54,100
your allowance, what happens is,
the side note is going to go and

479
00:26:54,100 --> 00:26:58,500
form a contract with basically 
the 50 best host on the network.

480
00:26:58,700 --> 00:27:02,600
So you're going to end up with 
50 state channels that you use 

481
00:27:02,600 --> 00:27:04,100
for both uploading and 
downloading. 

482
00:27:04,100 --> 00:27:07,500
And so those State channels have
a lifetime of several months and

483
00:27:07,500 --> 00:27:11,400
then any time you upload or 
download and update these State 

484
00:27:11,400 --> 00:27:15,100
channels and so you only need to
make on chain transactions on 

485
00:27:15,100 --> 00:27:18,600
the sign at work. 
About once a month and then 

486
00:27:18,600 --> 00:27:22,600
otherwise all activity, 
uploading downloading or even 

487
00:27:22,600 --> 00:27:24,400
just, you know, storing and 
sitting there. 

488
00:27:24,600 --> 00:27:27,200
It's going to happen through 
these payment Channels State 

489
00:27:27,200 --> 00:27:32,000
channels. 
And so, when I give this data to

490
00:27:32,000 --> 00:27:35,600
50 best host, right? 
I'm assuming there's some sort 

491
00:27:35,600 --> 00:27:38,600
of reed-solomon typing coding 
going on. 

492
00:27:39,200 --> 00:27:42,600
What is generally the like best 
practices here to make sure that

493
00:27:42,600 --> 00:27:45,500
your data is well redundant 
fide. 

494
00:27:46,900 --> 00:27:49,900
So you hit the nail on the head 
with Reed Solomon by default we 

495
00:27:49,900 --> 00:27:54,100
use a 10 of 30 scheme. 
So every piece of data is going 

496
00:27:54,100 --> 00:27:57,600
to go to 30 out of those 50 
hosts and of the 30 who receive 

497
00:27:57,600 --> 00:28:02,500
it, any 10 is sufficient to 
recover the original data. 

498
00:28:03,100 --> 00:28:05,300
And so basically what this I 
networks going to do because 

499
00:28:05,300 --> 00:28:08,700
hosts on the side Network. 
They're not unreliable right 

500
00:28:08,700 --> 00:28:10,300
there. 
They there tend to be pretty 

501
00:28:10,300 --> 00:28:13,600
reliable, but they're not Amazon
reliable, and we don't expect 

502
00:28:13,600 --> 00:28:16,500
them to be Amazon reliable, 
because that's very expensive. 

503
00:28:16,700 --> 00:28:18,500
We want, we want to be cost 
competitive. 

504
00:28:19,100 --> 00:28:23,200
So instead, we just Monitor and 
we see have host gone offline 

505
00:28:23,700 --> 00:28:24,900
are, there are, there are 
issues. 

506
00:28:24,900 --> 00:28:27,500
And so we have this constant 
monitoring service that's 

507
00:28:27,500 --> 00:28:29,500
checking in on the health of 
your files. 

508
00:28:30,600 --> 00:28:34,300
And this information, the health
information that is being 

509
00:28:34,300 --> 00:28:37,000
monitored, I presume is being 
monitored by the individual 

510
00:28:37,000 --> 00:28:38,400
nodes. 
Yep. 

511
00:28:38,400 --> 00:28:42,000
So the person who uploaded the 
data is the person who's doing 

512
00:28:42,000 --> 00:28:46,600
the health monitoring this 
information somehow, get stored 

513
00:28:46,600 --> 00:28:49,200
somewhere where the entirety of 
the network has access to it 

514
00:28:49,200 --> 00:28:51,900
like, is, is that where the 
blockchain comes in or like 

515
00:28:51,900 --> 00:28:54,300
explain where also the Block 
Chain falls into this 

516
00:28:54,300 --> 00:28:56,900
architecture? 
We'll come back to the 

517
00:28:56,900 --> 00:29:00,000
blockchain thing. 
So the answer is no, the health.

518
00:29:00,000 --> 00:29:03,300
The health is not monitored by 
anyone except the person who 

519
00:29:03,300 --> 00:29:07,200
uploaded the data, and I think 
this is one key departure from 

520
00:29:07,700 --> 00:29:10,900
file coins Ambitions. 
So it's a very difficult, 

521
00:29:10,900 --> 00:29:16,400
theoretical problem to do, 
decentralized data repair. 

522
00:29:16,800 --> 00:29:21,300
So what file coin wants to do is
that you upload data to the file

523
00:29:21,300 --> 00:29:24,300
core Network and then you 
disappear and the file coin 

524
00:29:24,300 --> 00:29:25,800
network will. 
Matically. 

525
00:29:25,800 --> 00:29:28,300
Restore the health of your file 
is host, go offline. 

526
00:29:28,800 --> 00:29:32,300
But the challenge here is that 
file coin can't restore the 

527
00:29:32,300 --> 00:29:36,100
health of your file, unless it 
knows how to erase your code 

528
00:29:36,100 --> 00:29:39,000
things, if it knows how to race 
your code things host can 

529
00:29:39,000 --> 00:29:42,200
collaborate with each other to D
duplicate the data and create 

530
00:29:42,200 --> 00:29:45,000
this sort of false redundancy 
make things look highly 

531
00:29:45,000 --> 00:29:47,600
redundant when they're not or 
make things look decentralized 

532
00:29:47,600 --> 00:29:51,100
when they're actually, you know,
it's six copies of the same file

533
00:29:51,100 --> 00:29:55,800
on one hard drive and the answer
to, like, how do You prevent 

534
00:29:55,800 --> 00:29:59,400
host from doing that is not easy
and I would, I would say, is a 

535
00:29:59,400 --> 00:30:01,900
strictly unsolved problem. 
I don't believe file coin to 

536
00:30:01,900 --> 00:30:04,700
solve the problem. 
Certainly it's been one of the 

537
00:30:04,700 --> 00:30:06,300
challenges for them and 
launching. 

538
00:30:06,900 --> 00:30:11,000
So we sidestep the issue by just
saying the network will not 

539
00:30:11,000 --> 00:30:14,600
repair your files for you. 
You have to repair the files 

540
00:30:14,600 --> 00:30:18,400
yourself and so that goes back 
to your question, Sonny, what do

541
00:30:18,400 --> 00:30:20,400
you have to do? 
Is someone storing data on the 

542
00:30:20,400 --> 00:30:23,300
sign Network? 
You have to run your node every 

543
00:30:23,300 --> 00:30:24,400
once in awhile. 
You know. 

544
00:30:24,400 --> 00:30:28,600
It doesn't need to be Beyond 
every day but maybe you know if 

545
00:30:28,600 --> 00:30:31,600
you leave it overnight once a 
week or you know, leave it on 

546
00:30:31,600 --> 00:30:34,700
all weekend once a month, 
that'll give it a chance to 

547
00:30:34,700 --> 00:30:37,200
check in with the files. 
See what the health is. 

548
00:30:37,200 --> 00:30:40,800
If any files are kind of low on 
health it can it can restore 

549
00:30:40,800 --> 00:30:44,700
them to full health. 
How does that still help with 

550
00:30:44,700 --> 00:30:48,400
solving like civil attack on the
host side? 

551
00:30:49,000 --> 00:30:52,100
How do I know I'm not getting 
false redundancy on PSI 0? 

552
00:30:53,700 --> 00:30:58,200
Yeah, so the important thing 
here actually so simple. 

553
00:30:58,200 --> 00:31:00,100
Several attacks and false 
redundancy are. 

554
00:31:00,100 --> 00:31:03,400
I think two different questions.
False redundancy we have a super

555
00:31:03,400 --> 00:31:07,900
simple solution which is that we
encrypt the data after we erase 

556
00:31:07,900 --> 00:31:10,500
your coded. 
So first you make these 30 

557
00:31:10,500 --> 00:31:13,200
redundant pieces of. 
And of course every piece is 

558
00:31:13,200 --> 00:31:16,300
different and then you encrypt 
each piece with a different key.

559
00:31:16,900 --> 00:31:20,500
So we have super high confidence
that at least three copies of 

560
00:31:20,500 --> 00:31:23,500
the data like definitely exists 
because we No, no one, no one 

561
00:31:23,500 --> 00:31:27,000
has the cryptographic ability to
deduplicate, that was his term. 

562
00:31:27,000 --> 00:31:30,500
Use Erasure code. 
Yeah, Erasure coding Erasure, 

563
00:31:30,500 --> 00:31:35,200
coded data means I give you a 
big set of data and then if some

564
00:31:35,200 --> 00:31:38,800
attacker or some, you know, 
processed in the middle cuts out

565
00:31:38,800 --> 00:31:41,000
pieces of it, erases parts of 
it. 

566
00:31:41,600 --> 00:31:45,100
The racial coding means that up 
to some threshold of erase data.

567
00:31:45,500 --> 00:31:49,300
You can restore the things that 
got deleted and so, it doesn't 

568
00:31:49,300 --> 00:31:52,100
matter that things got lost. 
Okay. 

569
00:31:52,100 --> 00:31:54,600
Is this similar to how? 
Like a raid would work? 

570
00:31:54,600 --> 00:31:58,100
Some like that? 
Yeah, raid is is a very simple 

571
00:31:58,100 --> 00:32:02,400
form of racial coding. 
So I guess one of the reasons it

572
00:32:02,400 --> 00:32:04,700
makes sentence idea but it's a 
little bit different in file 

573
00:32:04,700 --> 00:32:09,400
coin or at least entire V1 is 
because here it's my personal 

574
00:32:09,400 --> 00:32:12,700
data that I'm trying to 
distribute a bunch of them, a 

575
00:32:12,708 --> 00:32:15,700
number of hosts. 
And so what I can do is I can 

576
00:32:15,700 --> 00:32:19,200
encrypt to each of the host I 
can give a different encrypted 

577
00:32:19,200 --> 00:32:22,900
copy, and I know that they can't
convert, they can't be colluding

578
00:32:22,900 --> 00:32:26,600
because they don't know how to 
decrypt, one person's data and 

579
00:32:26,600 --> 00:32:28,300
turn it into another person's 
data. 

580
00:32:28,600 --> 00:32:32,600
But with file coin, what they're
Do is more for public data 

581
00:32:32,600 --> 00:32:35,200
rather than private data. 
And so they're assuming, let's 

582
00:32:35,200 --> 00:32:39,500
say it's Wikipedia, right? 
And so there's no private 

583
00:32:39,500 --> 00:32:42,800
decryption key, you know, that 
that decryption key has to be 

584
00:32:42,800 --> 00:32:46,300
public to everyone in the world.
So how do you solve this in? 

585
00:32:46,300 --> 00:32:52,100
For example, Sy R V2, So the 
solution that we are favoring 

586
00:32:52,100 --> 00:32:55,400
right now, is that you would 
actually publish, you'd have two

587
00:32:55,400 --> 00:32:58,300
versions of the file. 
So let's say for a public thing,

588
00:32:58,300 --> 00:33:03,000
instead of doing a 10 of 30, you
might do a 10 of 40 or 10 and 50

589
00:33:03,500 --> 00:33:07,100
and then you would publish 20 
out of those 50 pieces. 

590
00:33:07,300 --> 00:33:10,800
Which means that if hosts are 
colluding, if those 20 pieces 

591
00:33:10,800 --> 00:33:14,400
disappear, you still have these 
30 pieces that you know, you 

592
00:33:14,400 --> 00:33:16,500
know, can't be colluded over 
because you never release the 

593
00:33:16,500 --> 00:33:20,100
keys to those and it's also kind
of gives you a nice Canary 

594
00:33:20,400 --> 00:33:24,000
because it lets you see if I was
circulating if suddenly a file 

595
00:33:24,000 --> 00:33:27,000
unexpectedly disappears. 
You know that hosts are 

596
00:33:27,000 --> 00:33:31,300
colluding over over the public 
data but doesn't matter because 

597
00:33:31,300 --> 00:33:33,200
because you kept some of the 
redundancy back. 

598
00:33:33,200 --> 00:33:36,700
So I think I think that's the 
strategy will be using in V2 for

599
00:33:36,700 --> 00:33:39,500
the that public data. 
You just won't won't publish all

600
00:33:39,500 --> 00:33:43,100
of it at once. 
So you mentioned like the 

601
00:33:43,100 --> 00:33:46,400
decentral, you know, you really 
focus on the decentralisation as

602
00:33:46,400 --> 00:33:50,900
a primary focus, obviously the 
decentralization helps us with 

603
00:33:50,900 --> 00:33:55,500
the censorship resistance and, 
you know, all that fun stuff but

604
00:33:55,500 --> 00:33:58,300
does it also help with the cost 
of these things. 

605
00:33:58,300 --> 00:34:02,200
So you mentioned that cyan cost 
of storage is like connects 

606
00:34:02,200 --> 00:34:04,600
lower than AWS. 
Is it something about the 

607
00:34:04,600 --> 00:34:08,199
decentralization that allows 
this to happen or could I have 

608
00:34:08,199 --> 00:34:12,699
created a centralized Airbnb for
Storage and I could have gone to

609
00:34:12,699 --> 00:34:16,800
the same cost savings. 
Yeah, that's a great question. 

610
00:34:16,800 --> 00:34:20,900
So I think in the long term, the
decentralization is absolutely 

611
00:34:20,900 --> 00:34:23,400
critical to how the pricing 
model Works. 

612
00:34:23,800 --> 00:34:27,400
What we see with centralized 
systems and services, is that 

613
00:34:27,400 --> 00:34:30,699
even if they're competitive for 
a while, once they've obtained 

614
00:34:30,699 --> 00:34:33,500
the brand, once they've obtained
the trust once they've obtained,

615
00:34:33,500 --> 00:34:37,699
the market share, they always, I
mean, Google, Google is a famous

616
00:34:37,699 --> 00:34:40,600
recent example, they always 
switch gears from being, you 

617
00:34:40,600 --> 00:34:43,900
know, super consumer-friendly, 
super cost oriented super 

618
00:34:44,100 --> 00:34:46,900
everything that people. 
Want to being this more value, 

619
00:34:46,900 --> 00:34:50,400
extracting prices go up terms of
service, get nasty. 

620
00:34:50,900 --> 00:34:53,600
Even if a centralized service 
can provide good prices in the 

621
00:34:53,600 --> 00:34:58,600
short term you're really 
depending on the Goodwill of 

622
00:34:58,600 --> 00:35:02,500
that entity to maintain that, 
good pricing, once you depend on

623
00:35:02,500 --> 00:35:05,700
it and and I think we've seen 
consistently in the business 

624
00:35:05,700 --> 00:35:09,600
world you know once a business 
has established a moat it Cashes

625
00:35:09,600 --> 00:35:14,000
in on that mode by increasing 
prices and just making making 

626
00:35:14,000 --> 00:35:15,600
things better for the company. 
Open e and worse for the 

627
00:35:15,607 --> 00:35:18,000
consumer. 
And so I think in the long term,

628
00:35:18,200 --> 00:35:21,100
this decentralization is really 
important for Consumer 

629
00:35:21,100 --> 00:35:23,300
Protection. 
It's the only way that consumers

630
00:35:23,300 --> 00:35:26,700
can guarantee that their service
will continue to work as it did 

631
00:35:26,700 --> 00:35:30,300
when they joined. 
So, let's come back to the 

632
00:35:30,300 --> 00:35:32,400
technical components just 
briefly. 

633
00:35:32,600 --> 00:35:34,100
What is the blockchain fit into 
this? 

634
00:35:34,100 --> 00:35:35,800
Because I know there's a 
blockchain in here somewhere. 

635
00:35:35,900 --> 00:35:40,900
Yes, great question. 
So basically the chief challenge

636
00:35:40,900 --> 00:35:44,800
that we faced with a 
decentralized payment Network, 

637
00:35:45,000 --> 00:35:47,200
Right. 
So we're changing value for 

638
00:35:47,200 --> 00:35:50,900
storage. 
And we're also requiring hosts 

639
00:35:50,900 --> 00:35:55,500
to fulfill certain promises and 
the big promises like in the 

640
00:35:55,500 --> 00:35:59,100
file contract that the file 
contract says, you have to hold 

641
00:35:59,100 --> 00:36:00,800
this data for this amount of 
time. 

642
00:36:01,300 --> 00:36:04,700
And so, what we need is some 
sort of system that can check in

643
00:36:04,700 --> 00:36:07,000
on the hosts and then can 
decide. 

644
00:36:07,200 --> 00:36:10,300
Okay, you you kept the data, 
here's a payout or oh, no, you 

645
00:36:10,300 --> 00:36:12,500
didn't keep the data. 
We're going to, you know, we're 

646
00:36:12,500 --> 00:36:14,200
going to penalize. 
You were going to slash you. 

647
00:36:14,900 --> 00:36:19,300
And in a decentralized setting, 
I think the only way to do that 

648
00:36:19,300 --> 00:36:21,800
is to have this, this Block 
Chain where everyone can 

649
00:36:21,800 --> 00:36:24,900
validate, did the host, fulfill 
the promise or not. 

650
00:36:24,900 --> 00:36:27,800
And then and then based on that 
update, this decentralized 

651
00:36:27,800 --> 00:36:30,600
ledger to say whether or not the
host gets paid. 

652
00:36:30,800 --> 00:36:33,700
So the where the blockchain 
comes in is that stores, the 

653
00:36:33,700 --> 00:36:38,500
file contracts and specifically 
it audits, whether or not the 

654
00:36:38,500 --> 00:36:42,400
host has correctly stored, the 
data that the host was supposed 

655
00:36:42,400 --> 00:36:44,300
to sort and then and then from 
there we can. 

656
00:36:45,100 --> 00:36:48,500
Nope, analyzer pay the host of. 
So that's why we need our own 

657
00:36:48,500 --> 00:36:52,400
blockchain, then it's also just 
helpful in general. 

658
00:36:52,600 --> 00:36:55,300
Everything runs on money. 
If you want a decentralized 

659
00:36:55,300 --> 00:36:58,000
money I think you need a 
blockchain somewhere in there 

660
00:36:58,100 --> 00:37:00,600
but for that you could use, you 
know, Bitcoin or whatever. 

661
00:37:00,900 --> 00:37:02,600
So we need Sia for file 
contracts. 

662
00:37:04,500 --> 00:37:08,300
So the settlement layer is, is 
the blockchain. 

663
00:37:08,600 --> 00:37:16,300
This also stores, the contracts 
between users and hosts and the 

664
00:37:16,300 --> 00:37:21,400
payments for storing but also 
downloading and uploading files 

665
00:37:21,400 --> 00:37:24,900
happen over the payment channel.
So there's a layer one block 

666
00:37:24,900 --> 00:37:29,500
chain and LR2 payment Channel 
solution, or state channel built

667
00:37:29,500 --> 00:37:33,300
into into Sia with all within 
this last egg. 

668
00:37:34,000 --> 00:37:35,200
Yep. 
That's correct. 

669
00:37:35,500 --> 00:37:37,300
Okay. 
And which blockchain are you 

670
00:37:37,300 --> 00:37:39,300
using? 
Or did you build your own or are

671
00:37:39,300 --> 00:37:41,000
you leveraging an existing 
framework? 

672
00:37:42,300 --> 00:37:46,000
Yeah, so we built our own, it's 
highly inspired by the Bitcoin 

673
00:37:46,000 --> 00:37:48,700
blockchain but it is from the 
ground up. 

674
00:37:48,700 --> 00:37:52,600
So we wrote, we wrote all the 
code that's in our in our 

675
00:37:52,600 --> 00:37:58,000
software but it's very similar 
in design, philosophy to the 

676
00:37:58,000 --> 00:38:02,700
Bitcoin blockchain. 
What is the functionality of 

677
00:38:02,800 --> 00:38:05,700
Saia? 
That was of the base layer, the 

678
00:38:05,700 --> 00:38:08,500
L1 layer that wasn't available 
on bitcoin. 

679
00:38:08,500 --> 00:38:12,500
Like could we have done this as 
an L2 solution using Bitcoin, 

680
00:38:12,500 --> 00:38:14,400
like payment channels where 
their op codes that were 

681
00:38:14,400 --> 00:38:16,400
missing? 
That were needed. 

682
00:38:17,300 --> 00:38:22,100
Yeah, so you have to remember 
that we launched Sia in 2015, 

683
00:38:22,200 --> 00:38:25,700
June 2015. 
That's pretty etherium believe. 

684
00:38:25,700 --> 00:38:29,400
It that definitely pretty tender
moment and so we were we were 

685
00:38:29,400 --> 00:38:33,400
working with Much, much smaller,
technology base Bitcoin. 

686
00:38:33,400 --> 00:38:35,800
Today still can't do the Sci-Fi 
contracts. 

687
00:38:36,400 --> 00:38:38,900
We're missing two really 
critical things. 

688
00:38:38,900 --> 00:38:42,900
I think for this, I Network the 
first is that Bitcoin doesn't 

689
00:38:42,900 --> 00:38:46,000
have updatable payment channels.
So the lightning, our groups 

690
00:38:46,000 --> 00:38:49,700
based on HTML CSS, and these 
like crazy Draconian penalty 

691
00:38:49,700 --> 00:38:51,700
schemes. 
That really just don't don't 

692
00:38:51,700 --> 00:38:57,600
work with size complexity, but 
L2 is an example of a network 

693
00:38:57,900 --> 00:39:00,900
being built or an extension to 
bitcoin and I think if we had 

694
00:39:00,900 --> 00:39:04,900
L2, we might be able to do 
everything we need on the 

695
00:39:04,900 --> 00:39:07,100
Bitcoin Network directly. 
The other thing that this I 

696
00:39:07,100 --> 00:39:09,800
Network really needs that the 
Bitcoin network doesn't offer. 

697
00:39:09,800 --> 00:39:12,200
Is we need decentralized 
entropy. 

698
00:39:12,800 --> 00:39:14,900
And so we get this by looking at
the Block cache. 

699
00:39:15,400 --> 00:39:19,700
So, the block hash is, you know,
that the tail B are highly 

700
00:39:19,700 --> 00:39:24,900
random and we can depend on 
these two essentially have host 

701
00:39:24,900 --> 00:39:28,100
construct, secure proofs that 
they're actually storing data. 

702
00:39:28,100 --> 00:39:29,900
And that's, that's a bit of a 
more complicated. 

703
00:39:30,100 --> 00:39:34,500
A topic there, you know, because
black ashes are not perfectly 

704
00:39:34,500 --> 00:39:37,600
random, you have to be careful 
with how you use them, but 

705
00:39:37,700 --> 00:39:40,100
deciding how work is careful 
with how it uses black ashes and

706
00:39:40,100 --> 00:39:42,900
make sure that despite the fact 
that they're not perfectly 

707
00:39:42,900 --> 00:39:46,100
random every everything still 
checks out economically. 

708
00:39:47,100 --> 00:39:49,800
I mean, obviously the Bitcoin 
network has block hashes but I 

709
00:39:49,808 --> 00:39:52,700
guess the issue is that you 
can't access that data from 

710
00:39:52,700 --> 00:39:55,700
within a you T XO contract. 
That's correct. 

711
00:39:55,900 --> 00:39:59,100
So that's the challenge. 
And so if we got that and we 

712
00:39:59,100 --> 00:40:04,300
also got l to then I think you 
could probably build something 

713
00:40:04,300 --> 00:40:07,300
that's functional equivalent to 
size file contracts on paper 

714
00:40:07,300 --> 00:40:10,700
coin. 
So something that maybe you know

715
00:40:10,700 --> 00:40:15,000
maybe a lot of lessons don't 
know is that you are sort of in 

716
00:40:15,000 --> 00:40:19,200
part inspired me to go work on 
Cosmos because I met you the 

717
00:40:19,200 --> 00:40:23,400
first time at coined as 
consensus in 2017 and at the 

718
00:40:23,400 --> 00:40:27,100
time I was interning at a 
consensus, the why? 

719
00:40:27,500 --> 00:40:31,600
And roughly around the same time
as when storge or storage or 

720
00:40:31,600 --> 00:40:34,200
however you say it just 
transition to shifting onto 

721
00:40:34,200 --> 00:40:37,200
aetherium. 
And so I was kind of asking you 

722
00:40:37,200 --> 00:40:39,400
like why don't? 
You guys shift onto a theorem as

723
00:40:39,400 --> 00:40:40,800
well. 
It seems like that's that's the 

724
00:40:40,800 --> 00:40:44,000
hot thing these days and then 
you kind of went and pitched me 

725
00:40:44,000 --> 00:40:47,300
on explain to me why 
application-specific blockchains

726
00:40:47,300 --> 00:40:49,400
makes sense which is kind of 
what led me down to Cosmos. 

727
00:40:49,700 --> 00:40:53,000
So could you explain a little 
bit now for the listeners of why

728
00:40:53,000 --> 00:40:55,100
not shift onto aetherium or 
something like that? 

729
00:40:56,500 --> 00:40:58,400
Yeah, absolutely this is a good 
point. 

730
00:40:58,500 --> 00:41:02,000
Even even if Bitcoin had all the
Primitives that we needed to 

731
00:41:02,000 --> 00:41:06,000
build Saya on top of Bitcoin, we
probably still would have our 

732
00:41:06,000 --> 00:41:09,100
own blockchain. 
And and the reason for this 

733
00:41:09,500 --> 00:41:13,700
comes down to governance. 
So on the sign Network every 

734
00:41:13,700 --> 00:41:17,100
single user of the side 
blockchain is storage oriented. 

735
00:41:17,100 --> 00:41:18,800
They're thinking about how do I 
store data? 

736
00:41:18,800 --> 00:41:20,700
How do I retrieve data? 
How do I make money from 

737
00:41:20,700 --> 00:41:23,900
hosting? 
They're all aligned around this 

738
00:41:23,900 --> 00:41:25,700
common goal of data storage. 
Courage. 

739
00:41:25,700 --> 00:41:28,600
And so that means, if something 
comes up, if we need to make 

740
00:41:28,600 --> 00:41:31,200
some sort of blockchain 
governance decision a hard for 

741
00:41:31,200 --> 00:41:34,200
core, or we need to extend the 
protocol in some way or 

742
00:41:34,200 --> 00:41:38,100
something like that, everybody's
aligned around the same goal. 

743
00:41:38,100 --> 00:41:41,900
And so it's going to be easy to 
support the common use case on 

744
00:41:41,900 --> 00:41:46,100
the blockchain like etherium, 
you have this competing interest

745
00:41:46,100 --> 00:41:47,600
problem. 
This, this political thing where

746
00:41:47,600 --> 00:41:50,000
you have some people who really 
care about stable coins and 

747
00:41:50,000 --> 00:41:51,500
defy. 
Some people who really care 

748
00:41:51,500 --> 00:41:53,700
about like crypto, kitties some 
people who really care about 

749
00:41:53,700 --> 00:41:56,100
icos. 
And then also, People who care 

750
00:41:56,100 --> 00:41:58,800
about storage. 
And so, if the storage Network, 

751
00:41:59,100 --> 00:42:02,500
you know, says all we need to do
XYZ to make storage successful 

752
00:42:02,700 --> 00:42:07,200
and the D5 people are like, wait
XYZ is going to harm the defy, 

753
00:42:07,200 --> 00:42:11,600
use case you have this internal 
battle and it's it's much more 

754
00:42:11,600 --> 00:42:14,800
difficult for the storage people
to push through storage stuff. 

755
00:42:14,800 --> 00:42:17,600
Or if like if defies the big 
thing and defies like yeah we 

756
00:42:17,600 --> 00:42:20,000
need x y z in the storage 
communities like wait that's 

757
00:42:20,000 --> 00:42:23,100
going to break us? 
Well too bad like you're a small

758
00:42:23,100 --> 00:42:27,300
player and so I really Like the 
application, specific blockchain

759
00:42:27,300 --> 00:42:31,100
because it means that we're 
never every governance decision.

760
00:42:31,100 --> 00:42:34,500
Every Community decision is 
oriented around a common goal, 

761
00:42:34,800 --> 00:42:36,500
and that just makes the storage 
platform. 

762
00:42:36,500 --> 00:42:39,500
A lot more powerful and a lot 
more agile. 

763
00:42:40,900 --> 00:42:44,700
When you build see initially 
proof of stake was hadn't yet 

764
00:42:44,700 --> 00:42:47,300
been experimented with at least 
not to the scale at which is 

765
00:42:47,300 --> 00:42:50,200
today. 
The Cs still used proof of work.

766
00:42:51,900 --> 00:42:55,000
Yeah, so PSI is a proof-of-work 
blockchain, I'm a huge 

767
00:42:55,000 --> 00:42:58,000
proof-of-work, proponent I'm 
really not a fan of proof of 

768
00:42:58,000 --> 00:43:02,900
stake, although I know lots of 
people are happy to debate me on

769
00:43:02,900 --> 00:43:05,800
that. 
I think right now where we are 

770
00:43:05,800 --> 00:43:12,700
with the side Block Chain is 
actually 95 to 98% of our time 

771
00:43:13,300 --> 00:43:16,200
is focused on the application 
layer. 

772
00:43:16,200 --> 00:43:19,600
So we assume, you know, a 
decentralized consensus 

773
00:43:19,600 --> 00:43:22,700
mechanism, we assume a file 
contract, Summa payment Channel 

774
00:43:22,700 --> 00:43:25,600
system. 
And then once we assume all 

775
00:43:25,600 --> 00:43:31,200
those things, 98% of our Dev 
time is focused on using those 

776
00:43:31,200 --> 00:43:34,000
Primitives to build a 
decentralized storage Network. 

777
00:43:34,000 --> 00:43:37,600
And so, you know if three years 
from now, it turns out that that

778
00:43:37,600 --> 00:43:40,200
proof of stake is is absolutely 
the way to go. 

779
00:43:40,200 --> 00:43:42,800
Or if three years from now, it 
turns out that we need to switch

780
00:43:43,400 --> 00:43:45,300
our, underlying consensus 
mechanism. 

781
00:43:45,800 --> 00:43:48,300
That's something we're not 
really worried about today, 

782
00:43:48,300 --> 00:43:51,500
because doing so we should be 
able to carry over. 

783
00:43:51,700 --> 00:43:55,500
Over 98% of our effort. 
So I think right now we don't 

784
00:43:55,500 --> 00:43:58,400
even think of the consensus 
element just just because it's 

785
00:43:58,400 --> 00:44:01,800
so, it's so influx. 
In terms of the leading 

786
00:44:01,800 --> 00:44:04,900
research, we're really happy to 
be proof of work and we think we

787
00:44:04,900 --> 00:44:08,300
have a secure system as is. 
And then, if we can get more by 

788
00:44:08,300 --> 00:44:10,600
switching, we're happy to wait 
it out. 

789
00:44:10,600 --> 00:44:13,300
And we don't we don't think 
there will be a big switching 

790
00:44:13,300 --> 00:44:14,800
cost. 
Most of the things we've built. 

791
00:44:14,800 --> 00:44:18,100
We'll just we'll just be able to
transplant over if we ever do 

792
00:44:18,100 --> 00:44:21,600
need to switch. 
Have you ever thought about 

793
00:44:21,600 --> 00:44:24,400
maybe using some sort of proof 
of storage style system? 

794
00:44:24,400 --> 00:44:28,500
So similar to like Chia kind of 
stuff where let's say providers 

795
00:44:28,500 --> 00:44:31,800
don't have contracts that 
they're fulfilling, but they 

796
00:44:31,800 --> 00:44:36,400
have extra space, maybe they can
use that for consensus and then 

797
00:44:36,400 --> 00:44:40,800
as contracts come in, they can, 
you know, shift that around or 

798
00:44:40,800 --> 00:44:43,400
even if you take it even a step 
further file coin seems to want 

799
00:44:43,400 --> 00:44:45,200
to see. 
They think they can even do 

800
00:44:45,200 --> 00:44:50,000
proof of useful storage. 
Not useless storage like Jia you

801
00:44:50,000 --> 00:44:52,100
think that's even a feasible 
thing? 

802
00:44:53,200 --> 00:44:56,000
Yeah. 
So I don't think proof of useful

803
00:44:56,000 --> 00:45:00,000
storage is ever going to be a 
desirable consensus mechanism. 

804
00:45:00,200 --> 00:45:03,400
She of course has has 
sidestepped the issue by by 

805
00:45:03,400 --> 00:45:06,400
having proof of useless storage.
I think they have their own set 

806
00:45:06,400 --> 00:45:08,500
of challenges primarily the time
element. 

807
00:45:08,500 --> 00:45:10,900
How do you prove the data has 
been stored for a specific 

808
00:45:10,900 --> 00:45:13,100
amount of time? 
Is our, is there ways to cheat 

809
00:45:13,100 --> 00:45:17,000
that? 
So fun fact, back in in 2014 

810
00:45:17,000 --> 00:45:19,500
when we started building this, I
Network it was actually a His 

811
00:45:19,500 --> 00:45:23,700
storage base system, I was bft 
based, it looked very similar to

812
00:45:23,700 --> 00:45:25,200
a lot of proof of stake networks
today. 

813
00:45:25,200 --> 00:45:31,100
It had a lot of 51% assumptions,
more, 67% assumptions and then 

814
00:45:31,100 --> 00:45:34,400
it was all based on showing 
storage and proving storage. 

815
00:45:34,800 --> 00:45:37,400
And so I actually took this and 
presented it to the Bitcoin core

816
00:45:37,400 --> 00:45:41,400
developers and specifically, 
Greg Maxwell kind of broke it 

817
00:45:41,400 --> 00:45:45,600
down and he pointed out all 
these issues with what we had 

818
00:45:45,600 --> 00:45:48,300
built. 
And he convinced me 

819
00:45:48,300 --> 00:45:52,600
substantially That I was not 
ready to build my own consensus 

820
00:45:52,600 --> 00:45:54,300
system and that I should just 
use proof of work. 

821
00:45:54,300 --> 00:45:58,800
So we kind of abandon that, I 
think many of the issues that he

822
00:45:58,800 --> 00:46:01,900
pointed out then still apply 
today both tsuchiya and to file 

823
00:46:01,900 --> 00:46:04,000
coin. 
I don't think they've solved a 

824
00:46:04,000 --> 00:46:07,900
lot of the fundamental issues, 
but also like I'm happy to see 

825
00:46:07,900 --> 00:46:09,200
the research. 
Happy to see the 

826
00:46:09,200 --> 00:46:11,000
experimentation. 
Maybe they'll have some 

827
00:46:11,000 --> 00:46:14,500
incredible, you know, Discovery 
or breakthrough and in the 

828
00:46:14,500 --> 00:46:18,400
meantime as like someone who's 
trying to build something that's

829
00:46:18,400 --> 00:46:20,300
going to be. 
As soon as possible. 

830
00:46:20,300 --> 00:46:23,300
I think, proof-of-work does but 
our use case and so since it's 

831
00:46:24,000 --> 00:46:27,000
since it's what we know best, 
we're happy to use that for now.

832
00:46:27,000 --> 00:46:29,700
And like I said, you know, if 
there's some big consensus 

833
00:46:29,700 --> 00:46:32,800
breakthrough, I think it'll be 
relatively easy for us to move 

834
00:46:32,800 --> 00:46:37,100
over down the road. 
So given that we're talking 

835
00:46:37,100 --> 00:46:39,800
about proof-of-work, you know, 
the other thing that you're 

836
00:46:39,800 --> 00:46:43,100
quite well known for in the 
crypto Community is also a lot 

837
00:46:43,100 --> 00:46:46,500
of your research on Asics and 
whatnot. 

838
00:46:46,600 --> 00:46:49,200
That day that we met in 2017. 
You convinced me of two things 

839
00:46:49,200 --> 00:46:52,300
one which was applications of 
block chains and to Asics are 

840
00:46:52,300 --> 00:46:55,600
good which heavily inspired a 
lot of my design for how atoms 

841
00:46:55,600 --> 00:47:00,100
work in this Cosmos system. 
But so can tell us a little bit 

842
00:47:00,100 --> 00:47:04,000
about why you really believe in 
a six and then why you even went

843
00:47:04,000 --> 00:47:09,100
ahead and built an Sick company.
It's very interesting question. 

844
00:47:09,100 --> 00:47:13,700
So the reason I believe in A6 is
because it creates alignment 

845
00:47:13,700 --> 00:47:17,000
with your consensus Builders and
the long-term health of the 

846
00:47:17,000 --> 00:47:19,700
network. 
Well, and the long-term value of

847
00:47:19,700 --> 00:47:23,100
the coins that they mind 
basically an Asic is a giant 

848
00:47:23,100 --> 00:47:27,100
upfront payment for Hardware 
that has exactly one purpose. 

849
00:47:27,100 --> 00:47:31,700
The only way you can see Roi on 
that Hardware is to mine coins 

850
00:47:31,700 --> 00:47:33,100
and those coins have to have 
value. 

851
00:47:33,100 --> 00:47:34,800
So if you're mining coins, would
have no value. 

852
00:47:34,800 --> 00:47:38,200
You're never going to see Roi to
the best of our Ability. 

853
00:47:38,200 --> 00:47:41,200
We want to build proof-of-work, 
blockchain such that. 

854
00:47:41,500 --> 00:47:44,400
If the Network's not healthy, 
the coin price, is also not 

855
00:47:44,400 --> 00:47:46,300
healthy in the miners aren't 
getting paid that way. 

856
00:47:46,300 --> 00:47:49,600
The miners are strongly. 
Motivated to push the network 

857
00:47:49,600 --> 00:47:52,200
forward and and maintain the 
health of the network. 

858
00:47:52,200 --> 00:47:54,300
Make sure it's not getting 
attacked etcetera. 

859
00:47:54,700 --> 00:47:59,600
Whereas with like GPU mining, if
one network fails, you just jump

860
00:47:59,600 --> 00:48:02,200
to the next one. 
You know, if you don't even know

861
00:48:02,200 --> 00:48:04,900
it's GPU mind these days. 
Well, I'll say grin because 

862
00:48:04,900 --> 00:48:07,500
there are no granny six yet if 
grin fails. 

863
00:48:07,600 --> 00:48:10,800
GPU minor can just jump back to 
aetherium and they'll see maybe,

864
00:48:11,000 --> 00:48:14,800
you know, one to two percent 
reduction in their revenue but 

865
00:48:14,800 --> 00:48:18,100
it's not going to be a big deal.
They're not they're not married 

866
00:48:18,100 --> 00:48:21,800
to grin the same way that like 
in Asic minor is going to be 

867
00:48:21,800 --> 00:48:25,800
married to its cryptocurrency. 
So we think that's very valuable

868
00:48:25,800 --> 00:48:28,800
from a security perspective. 
Yeah. 

869
00:48:28,800 --> 00:48:33,300
And so then the follow-up to 
that was sort of why did you go 

870
00:48:33,300 --> 00:48:39,700
ahead and start an Asic company?
And 2017 nebulous created a 

871
00:48:39,700 --> 00:48:44,200
subsidiary called Obelisk. 
Obelisk purpose was to become a 

872
00:48:44,207 --> 00:48:46,000
mining and nebulous is your 
company. 

873
00:48:46,700 --> 00:48:50,600
Yes, nebulous is the parent 
company of Saia nebulous has two

874
00:48:50,600 --> 00:48:53,800
major projects. 
One is Saya and the other is 

875
00:48:53,800 --> 00:48:57,200
obelisk. 
So between 2014 and 2017, Sia 

876
00:48:57,200 --> 00:49:00,800
was the only focus for nebulous 
starting in 2017. 

877
00:49:00,800 --> 00:49:05,500
It also had this mining focus. 
And in 2017, this is when a lot 

878
00:49:05,500 --> 00:49:08,000
of The drama with B main was 
really heating up, but main was 

879
00:49:08,000 --> 00:49:12,300
making a lot of political power 
plays against Bitcoin, but that 

880
00:49:12,300 --> 00:49:15,900
main was controlling a lot of 
the alt, and we were and, and 

881
00:49:15,900 --> 00:49:18,300
also just the general sentiment 
at the time. 

882
00:49:18,300 --> 00:49:21,000
And, you know, my personal 
belief at the time was that 

883
00:49:21,000 --> 00:49:25,800
having distributed hash rate was
really important and so we 

884
00:49:25,800 --> 00:49:29,500
didn't want any single party to 
own more than 51% of the hash 

885
00:49:29,500 --> 00:49:32,900
rate of the sign at work and we 
felt bit main was a big risk. 

886
00:49:32,900 --> 00:49:37,600
So Obelisk was actually an 
attempt to You give users an 

887
00:49:37,600 --> 00:49:40,800
alternative to bit main for 
mining on the side Network. 

888
00:49:41,400 --> 00:49:44,600
And so we we did we created an 
Asic company. 

889
00:49:45,000 --> 00:49:48,500
We designed a chip, we 
manufactured the chip we 

890
00:49:48,500 --> 00:49:51,300
designed a rig and manufactured 
the rig and we shipped something

891
00:49:51,300 --> 00:49:55,800
like I think fifteen thousand 
machines total, the purchasers 

892
00:49:55,800 --> 00:49:58,600
on the sign Network. 
And so for a while we had some 

893
00:49:58,600 --> 00:50:02,100
of the best distributed hash 
rate of any a sick Mind Block 

894
00:50:02,100 --> 00:50:04,500
Chain. 
As it turns out, I think that 

895
00:50:04,600 --> 00:50:07,600
the more we studied Bitcoin. 
And if you look at the people 

896
00:50:07,600 --> 00:50:12,800
who really dive deep and proof 
of work in 2019, and I think 

897
00:50:12,800 --> 00:50:16,700
this is really just an idea that
that started to take root in mid

898
00:50:16,700 --> 00:50:23,600
to late 2019, is that it doesn't
actually matter if a majority of

899
00:50:23,600 --> 00:50:27,400
the hash rate is controlled by 
one entity because that entity 

900
00:50:27,400 --> 00:50:31,300
is still financially heavily 
aligned that entity still has to

901
00:50:31,300 --> 00:50:36,200
worry about hard Forks. 
So really all you want From a 

902
00:50:36,200 --> 00:50:40,200
security perspective is one 
entity, that's heavily 

903
00:50:40,200 --> 00:50:43,700
financially aligned with your 
network and then to an entity 

904
00:50:43,700 --> 00:50:45,600
that that is capable of being 
threatened. 

905
00:50:45,600 --> 00:50:47,900
So if it's a perfect Monopoly 
and they know no one's ever 

906
00:50:47,900 --> 00:50:50,000
going to compete with them, 
that's a bad situation. 

907
00:50:50,600 --> 00:50:53,800
But if they have like, 80% of 
the hash rate and there's some, 

908
00:50:53,800 --> 00:50:55,700
you know, there are some other 
companies or some other 

909
00:50:55,700 --> 00:50:58,700
opportunity for people to step 
in, if they get lazy or slow or 

910
00:50:58,700 --> 00:51:02,100
stop innovating, There's an 
opportunity for that 80% to 

911
00:51:02,100 --> 00:51:05,800
switch hands from one company to
another, that I think is, As you

912
00:51:05,808 --> 00:51:10,200
know, it's sufficient for a 
secure network and this is quite

913
00:51:10,200 --> 00:51:13,300
a controversial opinion, I do 
think over the next two or three

914
00:51:13,300 --> 00:51:15,400
years, it will become more 
broadly accepted. 

915
00:51:15,400 --> 00:51:18,400
I think it's very sound from a 
research perspective, but is 

916
00:51:18,400 --> 00:51:22,000
certainly very different from 
how most people were thinking 

917
00:51:22,000 --> 00:51:27,400
about proof-of-work, blockchains
between 2010 and 2009 teen. 

918
00:51:29,000 --> 00:51:34,300
And so what was like different 
about what Obelisk was doing? 

919
00:51:34,300 --> 00:51:37,100
Is you know, I know one of the 
things is there's a there's a 

920
00:51:37,107 --> 00:51:41,400
heavy emphasis on Open Source 
Hardware what made Obelisk 

921
00:51:41,500 --> 00:51:44,500
different than b main or no 
silicon and stuff. 

922
00:51:45,600 --> 00:51:48,400
Yeah, so we had a couple of 
really big things. 

923
00:51:48,400 --> 00:51:50,300
We cared about 11 was open 
source. 

924
00:51:50,300 --> 00:51:51,800
We wanted the firmware to be 
open source. 

925
00:51:51,800 --> 00:51:54,200
One the hardware to be open 
source when the gate level to be

926
00:51:54,200 --> 00:51:56,700
open source. 
We wanted competitive hashing 

927
00:51:56,700 --> 00:51:59,400
Network, we wanted it to be such
that, you know, a bit main 

928
00:51:59,400 --> 00:52:03,500
couldn't lock, everyone out. 
So we wanted to pave the steps 

929
00:52:04,100 --> 00:52:06,400
so that if someone wanted to 
come in and compete with us, 

930
00:52:06,900 --> 00:52:10,400
they would have is low barrier 
to entry as possible to do that.

931
00:52:11,100 --> 00:52:14,400
Another thing that we really 
cared about is transparency and 

932
00:52:14,400 --> 00:52:20,400
this actually Important to the 
person buying the Asic, it's 

933
00:52:20,400 --> 00:52:22,700
less important to the Network 
Health, more important to the 

934
00:52:22,700 --> 00:52:25,600
person buying Asic. 
What we saw a bit Main and, you 

935
00:52:25,600 --> 00:52:30,300
know, silicon do and continue to
do even today as they would sell

936
00:52:30,300 --> 00:52:32,300
a bunch of hash rate. 
They wouldn't tell you how much 

937
00:52:32,300 --> 00:52:36,500
it manufactured and so miners 
would go in and they make these 

938
00:52:36,500 --> 00:52:39,600
projections like okay. 
Well as long as bit main made 

939
00:52:39,600 --> 00:52:44,100
less than 20,000 machines, then 
I'm going to Roi and I'm going 

940
00:52:44,100 --> 00:52:46,500
to make a ton of money and then 
A bit main did was they made, 

941
00:52:46,500 --> 00:52:51,000
you know, 150,000 machines and 
they sold all 150,000 and so 

942
00:52:51,200 --> 00:52:55,500
they have 150,000 customers who 
all Roi only, if less than 20 

943
00:52:55,500 --> 00:52:58,700
thousand machines were sold. 
And the result is that if you 

944
00:52:58,700 --> 00:53:01,800
know the block reward for two 
years is 40 million dollars but 

945
00:53:01,800 --> 00:53:05,800
made a profit of 100 million 
dollars but main made more money

946
00:53:06,100 --> 00:53:10,300
than was possible to even mine 
selling you know a 64, a 

947
00:53:10,300 --> 00:53:12,600
cryptocurrency. 
And of course this money comes 

948
00:53:12,600 --> 00:53:14,600
directly out of the pockets of 
Asic buyers. 

949
00:53:14,900 --> 00:53:18,100
And so Almost really wanted to 
fight a lot of the mining 

950
00:53:18,100 --> 00:53:21,900
spaces. 
Absolutely saturated with these 

951
00:53:22,200 --> 00:53:26,700
crazy dark patterns patterns 
designed to abuse the customer 

952
00:53:26,900 --> 00:53:30,000
and abused the ecosystem for the
profit of the manufacturer. 

953
00:53:30,000 --> 00:53:33,000
And I both really wanted to take
a stand against that, you know, 

954
00:53:33,000 --> 00:53:36,300
we just kind of realized that b 
if B Maine has sales channels 

955
00:53:36,900 --> 00:53:40,500
and they're going to be able to 
sell, it's very exhausting for 

956
00:53:40,500 --> 00:53:43,400
obelus to try and like shut down
those dark patterns it. 

957
00:53:43,400 --> 00:53:45,400
But Maine has a source of people
that they can talk. 

958
00:53:45,600 --> 00:53:47,600
To and sell to that obelus, 
doesn't, you know? 

959
00:53:47,600 --> 00:53:49,600
It doesn't have communication 
Channel with. 

960
00:53:49,900 --> 00:53:52,800
There's nothing we can do to 
like Stop Those sales. 

961
00:53:54,800 --> 00:53:57,200
So, could you tell us a little 
bit about this, you know, 

962
00:53:57,300 --> 00:54:01,500
Infamous hard fork in see? 
Where I remember, I was like 

963
00:54:01,500 --> 00:54:04,400
peripherally following it, but 
there was a lot of drama around 

964
00:54:04,400 --> 00:54:08,100
essentially, it is hard Fork 
design through brick a lot of 

965
00:54:08,100 --> 00:54:14,300
the pit Main and in our silicon 
a 6 but not the Obelisk ones 

966
00:54:14,600 --> 00:54:15,800
one. 
How does that even work? 

967
00:54:15,800 --> 00:54:18,800
Technically and to why was that 
done? 

968
00:54:18,800 --> 00:54:20,900
And is that like you know, 
potentially a conflict of 

969
00:54:20,900 --> 00:54:25,600
interest Yeah. 
So I will say that is the most 

970
00:54:25,600 --> 00:54:29,000
stressful and difficult decision
that I've ever had to make. 

971
00:54:29,300 --> 00:54:33,000
So practically speaking, there 
was no conflict of interest 

972
00:54:33,100 --> 00:54:36,100
Obelisk could already sold all 
of its Hardware obelus doesn't 

973
00:54:36,100 --> 00:54:39,400
mind itself, never did. 
So we've never mind a 

974
00:54:39,400 --> 00:54:41,100
substantial amount on the sign 
Network. 

975
00:54:41,100 --> 00:54:43,300
Like, yeah, sure, we had 
machines that we had to test way

976
00:54:43,308 --> 00:54:45,700
to make sure they worked and 
that that was happening on the 

977
00:54:45,700 --> 00:54:48,400
side network. 
But we never controlled like say

978
00:54:48,400 --> 00:54:52,700
even 1% of the Sia hash rate as 
Obelisk. 

979
00:54:52,700 --> 00:54:56,700
And and so the actual conflict 
of interest was a lot smaller 

980
00:54:56,700 --> 00:54:58,300
than the apparent conflict of 
interest. 

981
00:54:58,300 --> 00:55:01,600
There's still a conflict of 
interest there just because you 

982
00:55:01,600 --> 00:55:05,300
know we did control Obelisk more
sales for Obelisk meant you no 

983
00:55:05,300 --> 00:55:09,000
more revenue for us more profit 
for us, but a lot of people 

984
00:55:09,000 --> 00:55:12,000
assume that Obelisk was mining 
Obelisk had claimed at one point

985
00:55:12,000 --> 00:55:16,400
and it says, Obelisk on nebulous
is primary revenue at the time. 

986
00:55:16,400 --> 00:55:21,200
Yes, it was the primary cash 
flow for nebulous and you know, 

987
00:55:21,200 --> 00:55:23,500
Obelisk could stay. 
That it would not mind more than

988
00:55:23,500 --> 00:55:25,700
25% of the hash rate of the 
network. 

989
00:55:26,100 --> 00:55:30,000
Which of course, most people 
assumed that we were mining 25% 

990
00:55:30,000 --> 00:55:32,800
of the hash rate of the network,
we had kind of I think 20% one 

991
00:55:32,800 --> 00:55:36,100
of those either way we had 
carved out the ability for us to

992
00:55:36,100 --> 00:55:40,100
- oh substantial fraction a 
network that we never did didn't

993
00:55:40,100 --> 00:55:42,700
help with the apparent conflict 
of interest. 

994
00:55:42,700 --> 00:55:46,000
So you had all these like 
conflict of interest issues that

995
00:55:46,000 --> 00:55:49,900
both made it difficult to think,
clearly made it made it 

996
00:55:49,900 --> 00:55:53,400
difficult to establish as a 
leader that That you are just 

997
00:55:53,400 --> 00:55:56,400
internally to convince yourself 
that you're making an impartial 

998
00:55:56,400 --> 00:55:59,500
correct decision. 
But then it made it basically 

999
00:55:59,500 --> 00:56:02,800
like almost impossible to 
present to the rest of the world

1000
00:56:02,800 --> 00:56:05,700
that the decision was impartial.
So I think that, you know, that 

1001
00:56:05,700 --> 00:56:10,000
was a very difficult situation, 
but I also think that we made 

1002
00:56:10,000 --> 00:56:14,400
the right decision overall and 
I'll dive into that a little bit

1003
00:56:14,400 --> 00:56:16,200
more. 
So the background of the 

1004
00:56:16,200 --> 00:56:21,200
situation is like Obelisk, had 
announced this open transparent 

1005
00:56:21,500 --> 00:56:24,900
effort To decentralize the 
mining of SIA. 

1006
00:56:25,600 --> 00:56:29,300
So one critical mistake we made 
is we had never established like

1007
00:56:29,300 --> 00:56:32,300
a policy as you know, going into
manufacturing. 

1008
00:56:32,300 --> 00:56:37,700
We never established a policy on
secret Asics as you know just a 

1009
00:56:37,700 --> 00:56:41,000
just a governance tools but we 
had mentioned several times that

1010
00:56:41,000 --> 00:56:46,200
we felt that if someone had made
a 6 in secret, this would be 

1011
00:56:46,200 --> 00:56:51,200
considered a hostile action 
because it comes back to the you

1012
00:56:51,200 --> 00:56:55,800
know, if bit Main cells 90,000 
machines to a market that can 

1013
00:56:55,800 --> 00:56:59,000
only reasonably sustain the 
production of 10,000 machines, 

1014
00:56:59,200 --> 00:57:01,100
everybody gets screwed, accept 
it name. 

1015
00:57:01,700 --> 00:57:04,300
And so, secret Asics come into 
the exact same thing. 

1016
00:57:04,300 --> 00:57:06,900
We felt at the time that the 
story of secret A6. 

1017
00:57:06,900 --> 00:57:09,100
You mean like a manufacturer 
secretly? 

1018
00:57:09,100 --> 00:57:12,300
Making a six and then dumping 
them on to the market. 

1019
00:57:13,200 --> 00:57:15,100
Yeah, exactly. 
So, the only way to make an 

1020
00:57:15,100 --> 00:57:18,400
informed decision as a consumer.
When you're buying a six, is to 

1021
00:57:18,400 --> 00:57:21,700
know how many Asics are on the 
market because it's a zero-sum 

1022
00:57:21,700 --> 00:57:22,600
game. 
More Asics. 

1023
00:57:22,600 --> 00:57:24,100
There are the less money you 
make. 

1024
00:57:25,200 --> 00:57:28,900
We held the position internally 
and we had spoken about it. 

1025
00:57:29,000 --> 00:57:32,100
Nonchalantly we'd never 
establish a policy that secret a

1026
00:57:32,100 --> 00:57:37,200
6 would be considered an attack 
on the network because it would 

1027
00:57:37,200 --> 00:57:40,400
substantially damaged. 
The revenue of people who had 

1028
00:57:40,400 --> 00:57:42,200
attempted to make informed 
decisions. 

1029
00:57:43,200 --> 00:57:46,300
Of course, what happened was bit
main announced with one week 

1030
00:57:46,300 --> 00:57:49,000
notice. 
Hey, we have PSI a six that were

1031
00:57:49,000 --> 00:57:52,400
shipping and then we discovered 
through some back-channeling 

1032
00:57:52,400 --> 00:57:54,100
that they made ninety thousand 
machines. 

1033
00:57:54,100 --> 00:57:57,500
And again, the sign Network 
could sustain maybe ten, maybe 

1034
00:57:57,500 --> 00:58:00,900
twenty thousand machines a bit 
main had manufactured 90,000. 

1035
00:58:01,400 --> 00:58:02,700
That's what our information 
said. 

1036
00:58:02,700 --> 00:58:05,000
So, it was this like transparent
attack. 

1037
00:58:05,500 --> 00:58:09,800
Then I received a call from BTC 
Drac of all people who said, 

1038
00:58:09,800 --> 00:58:14,100
like, hey, we've been working on
our own a We've been working 

1039
00:58:14,100 --> 00:58:18,400
with in our silicon and so we 
also have a secret, a 60, what 

1040
00:58:18,400 --> 00:58:21,900
you can do is you can break bit 
Main and just use our Asic 

1041
00:58:21,900 --> 00:58:25,600
instead, which of course, like 
completely misses the point that

1042
00:58:25,600 --> 00:58:28,800
the problem is the secret Asic. 
It's not B main specifically, 

1043
00:58:28,800 --> 00:58:32,100
but at the action of creating a 
secret Asic, so it would 

1044
00:58:32,100 --> 00:58:35,400
definitely every purchaser of 
Obelisk Hardware perceive this 

1045
00:58:35,400 --> 00:58:40,500
as an attack on them. 
And then because it was a huge 

1046
00:58:40,500 --> 00:58:42,900
fraction of the committed Side 
Community had purchased. 

1047
00:58:43,000 --> 00:58:46,600
Jobless card where it was kind 
of an attack on the sign Network

1048
00:58:46,600 --> 00:58:48,200
as a whole. 
And so that this was like, the 

1049
00:58:48,200 --> 00:58:50,600
big dramatic challenge. 
You had all these people 

1050
00:58:51,000 --> 00:58:53,200
suddenly accounts that had never
been seen before. 

1051
00:58:53,200 --> 00:58:56,500
We're very active on Discord 
with a talking exclusively about

1052
00:58:56,500 --> 00:58:59,200
how bit Mains a good actor and 
how you can't break bit Main and

1053
00:58:59,200 --> 00:59:02,800
like, just this like, hold 
social engineering mess. 

1054
00:59:03,800 --> 00:59:07,200
Ultimately, we decided and it 
really, really came down to 

1055
00:59:07,200 --> 00:59:10,500
enter silicon was actively 
mining themselves as a 

1056
00:59:10,500 --> 00:59:12,700
manufacturer more than 51% of 
the network. 

1057
00:59:12,700 --> 00:59:15,400
So, not only did they dumped a 
ton of machines on the network 

1058
00:59:15,400 --> 00:59:17,800
then? 
Also overproduce, they also kept

1059
00:59:17,800 --> 00:59:20,800
most of the machines for 
themselves and so, you know, 

1060
00:59:20,800 --> 00:59:24,200
silicon and a party that had 
already attacked the network and

1061
00:59:24,200 --> 00:59:28,500
shown to be acting in bad. 
Faith also had the ability to, 

1062
00:59:28,500 --> 00:59:32,700
you know, 51% the network That 
was kind of what pushed me all 

1063
00:59:32,700 --> 00:59:33,800
the way over and said like, 
okay. 

1064
00:59:33,800 --> 00:59:37,400
Now we have enough, enough 
reason to Fork, even with all 

1065
00:59:37,400 --> 00:59:40,200
this like conflict of interest 
mess, we're going to go ahead 

1066
00:59:40,200 --> 00:59:44,500
and Fork but like, one common 
theme that I felt in every 

1067
00:59:44,500 --> 00:59:48,200
interaction that I had with the 
other mining manufacturers. 

1068
00:59:48,500 --> 00:59:50,800
You know, specifically bit Main 
and you know silicon but we were

1069
00:59:50,800 --> 00:59:52,600
talking to other manufacturers 
as well. 

1070
00:59:52,600 --> 00:59:55,000
Was the sort of like Manifest 
Destiny. 

1071
00:59:55,000 --> 00:59:58,000
They're like Sia is a 
decentralized network. 

1072
00:59:59,100 --> 01:00:02,600
If there's a way for us to make 
a ton of money by crapping all 

1073
01:00:02,600 --> 01:00:05,800
over the network, then it's all 
right to do so and it's our 

1074
01:00:05,800 --> 01:00:09,700
right to make a giant mess of 
SIA and then make a, my a ton of

1075
01:00:09,700 --> 01:00:12,700
money off of that. 
And so that I just felt like had

1076
01:00:12,700 --> 01:00:16,300
to be shut down. 
So a big part of the fork was 

1077
01:00:16,300 --> 01:00:20,500
also this element is like no you
have to behave like we as a 

1078
01:00:20,600 --> 01:00:24,300
community can make decisions 
that directly impact your 

1079
01:00:24,300 --> 01:00:26,500
financial status and your 
financial gain. 

1080
01:00:26,500 --> 01:00:29,700
And if you're going to be like a
bad bodyguard, we're going to 

1081
01:00:29,700 --> 01:00:31,300
fire you. 
And so that's what we did. 

1082
01:00:31,300 --> 01:00:34,400
We forked and something that I 
definitely notice following the 

1083
01:00:34,400 --> 01:00:37,400
fork is that mining 
manufacturers for other 

1084
01:00:37,600 --> 01:00:40,700
proof-of-work, cryptocurrencies 
suddenly became a lot more. 

1085
01:00:40,700 --> 01:00:44,400
Attentive to the developers 
desires developers are suddenly 

1086
01:00:44,600 --> 01:00:48,200
Out of the discussion when 
miners are being manufactured, 

1087
01:00:48,200 --> 01:00:50,800
because they realized that there
can be Forks. 

1088
01:00:51,900 --> 01:00:55,200
So those who have bought your 
own mining machines as well at 

1089
01:00:55,200 --> 01:00:58,500
all below Square also then 
basically forked out of the 

1090
01:00:58,500 --> 01:01:03,200
network. 
Now there's this strategy that 

1091
01:01:03,300 --> 01:01:05,700
most if not all Asic 
manufacturers. 

1092
01:01:05,700 --> 01:01:10,400
Do I again think we need to 
credit Greg Maxwell with coming 

1093
01:01:10,400 --> 01:01:13,400
up with the idea of a. 
Basically what you do is you 

1094
01:01:13,500 --> 01:01:17,200
make your Hardware capable of 
mining the main algorithm. 

1095
01:01:17,500 --> 01:01:20,700
Then you also add, you know, 
just a little Branch somewhere 

1096
01:01:20,700 --> 01:01:24,200
in the chips, I'm wearing the 
gates that allows it to mine. 

1097
01:01:24,200 --> 01:01:27,900
This alternate algorithm so it's
just a slightly modified 

1098
01:01:27,900 --> 01:01:31,900
algorithm there are you A 
million ways to make a slightly 

1099
01:01:31,900 --> 01:01:34,900
modified algorithm every 
manufacturer when they pick at 

1100
01:01:34,900 --> 01:01:36,800
week is going to pick a 
different tweak. 

1101
01:01:37,300 --> 01:01:42,300
And what that allows us to do is
hard Fork to tweak the mining 

1102
01:01:42,300 --> 01:01:44,000
algorithm. 
So that instead of doing, you 

1103
01:01:44,008 --> 01:01:49,000
know, mining algorithm a, you do
mining algorithm a prime, or if 

1104
01:01:49,000 --> 01:01:51,700
there's one bad actor and for 
good actors, you tweak it so 

1105
01:01:51,700 --> 01:01:54,300
that it's a prime or a prime 
prime. 

1106
01:01:55,500 --> 01:01:58,900
Isn't this also building some 
kind of a secret Asic? 

1107
01:01:59,300 --> 01:02:02,600
I can't help but to see the 
irony in all of this because on 

1108
01:02:02,600 --> 01:02:05,900
one hand, you've been saying 
since the beginning that you're 

1109
01:02:05,900 --> 01:02:11,000
40 centralization and I really 
think you are but by doing this,

1110
01:02:11,000 --> 01:02:16,500
by acting as a privileged actor,
who had access to the community 

1111
01:02:16,500 --> 01:02:21,800
and the code of Sion network, 
but also making an Asic that you

1112
01:02:21,800 --> 01:02:25,800
could change the algorithm on a 
whim How do you see yourself as 

1113
01:02:25,800 --> 01:02:31,300
different from the free market 
actors that are just producing a

1114
01:02:31,300 --> 01:02:33,700
6 in a free market? 
And you know, playing by the 

1115
01:02:33,700 --> 01:02:37,000
rules of the market that are 
allowed to make a 6 and sell 

1116
01:02:37,000 --> 01:02:39,800
them to customers if they want 
to and it's up to them on 

1117
01:02:39,800 --> 01:02:44,100
whether they want to inform the 
market of their desire to do so.

1118
01:02:45,300 --> 01:02:49,100
so it's important to clarify 
like you know, silicon also had 

1119
01:02:49,400 --> 01:02:53,000
this switch in it and we don't 
know for certain but I would be 

1120
01:02:53,400 --> 01:02:57,700
very surprised a bit main didn't
also have support for some 

1121
01:02:57,700 --> 01:03:01,600
alternate algorithm What's the 
point of them putting a switch 

1122
01:03:01,600 --> 01:03:03,000
in it? 
Because how could they choose, 

1123
01:03:03,000 --> 01:03:05,900
what the next hashing algorithm 
would be unless they're also the

1124
01:03:05,900 --> 01:03:09,800
Developers. 
So it's about governance, what 

1125
01:03:09,900 --> 01:03:13,600
bit main is doing what Obelisk 
was doing and before we 

1126
01:03:13,600 --> 01:03:16,000
conclude, we should definitely 
get into the actual mechanics of

1127
01:03:16,000 --> 01:03:17,500
the fork. 
Because I think it's very 

1128
01:03:17,500 --> 01:03:19,900
interesting what we did. 
And, and very important, the way

1129
01:03:19,900 --> 01:03:24,200
we executed the fort, but before
that as a manufacturer, when you

1130
01:03:24,200 --> 01:03:28,500
put this alternate algorithm 
just tweaked algorithm into your

1131
01:03:28,500 --> 01:03:30,600
Asic. 
What you're doing is you're 

1132
01:03:30,600 --> 01:03:36,400
offering the community, the 
option to switch to exclude some

1133
01:03:36,400 --> 01:03:38,100
A6, and not others. 
So you Hang them. 

1134
01:03:38,100 --> 01:03:42,100
The flexibility to make 
governance decision to exclude 

1135
01:03:42,100 --> 01:03:46,900
other manufacturers but retain 
you If manufacturers aren't 

1136
01:03:46,900 --> 01:03:50,900
doing this, the only tool that 
the community has is like a nuke

1137
01:03:50,900 --> 01:03:55,200
is to break all a six at once. 
If every manufacturer has this 

1138
01:03:55,500 --> 01:03:58,500
tweak that they either disclosed
to the developers or just close 

1139
01:03:58,500 --> 01:04:02,000
the community, you know, at 
their discretion then instead of

1140
01:04:02,000 --> 01:04:05,200
dropping a nuke that kills 
everything, you can drop a 

1141
01:04:05,207 --> 01:04:09,000
targeted nuke that leaves 
certain parties alive. 

1142
01:04:10,100 --> 01:04:12,900
Manufacturers do this because it
gives the community a choice. 

1143
01:04:12,900 --> 01:04:15,100
There's very little downside to 
the manufacturing. 

1144
01:04:15,100 --> 01:04:19,500
For example, a possible 
scenario, where we would have or

1145
01:04:19,500 --> 01:04:22,400
where, the side and network 
would have selected. 

1146
01:04:22,400 --> 01:04:25,200
Bit means a six over in a 
silicon is like, you know, 

1147
01:04:25,200 --> 01:04:28,400
silicon really was like the bad 
actor in the space. 

1148
01:04:28,400 --> 01:04:32,200
So if we assume that obelus 
didn't exist, it made it come to

1149
01:04:32,200 --> 01:04:34,900
Market than in a silicon came to
Market with much Superior 

1150
01:04:34,900 --> 01:04:37,500
Hardware. 
You know, silicon has 51 percent

1151
01:04:37,500 --> 01:04:40,700
of the hash rate, this included.
You know, silicon Hardware was 

1152
01:04:40,700 --> 01:04:44,600
Superior to B mains. 
If you know, silicon had 51 

1153
01:04:44,600 --> 01:04:47,900
percent and the network is now 
like being attacked or 

1154
01:04:47,900 --> 01:04:49,900
something. 
But man could have come forward 

1155
01:04:49,900 --> 01:04:53,700
and said instead of breaking all
a six by clearly the Bad actors.

1156
01:04:53,700 --> 01:04:56,600
In our silicon we have this 
tweak that we know in our 

1157
01:04:56,600 --> 01:04:59,500
silicon doesn't have or we 
strongly suspect in a silicon 

1158
01:04:59,500 --> 01:05:04,600
doesn't have Who's to decide 
that a bat and a silicon or bit 

1159
01:05:04,600 --> 01:05:05,700
mate. 
I mean, I don't have any opinion

1160
01:05:05,700 --> 01:05:09,600
on this, but who's to decide who
are The Bad actors and do you 

1161
01:05:09,600 --> 01:05:13,600
not feel that as a privileged 
actor in this whole ecosystem, 

1162
01:05:14,100 --> 01:05:17,900
by making the decision, to 
tweak, the algorithm to favor, 

1163
01:05:18,300 --> 01:05:23,500
that of with your miners, were 
ready for that is not a huge 

1164
01:05:23,500 --> 01:05:27,800
conflict of interest. 
This really gets into the deep 

1165
01:05:27,800 --> 01:05:32,400
nitty-gritty of decentralized 
governance, it's difficult for 

1166
01:05:32,400 --> 01:05:34,400
people to wrap their heads 
around. 

1167
01:05:34,800 --> 01:05:37,700
How leadership Works in a 
decentralized way and actually 

1168
01:05:37,700 --> 01:05:41,200
to the detriment of 
decentralization in a 

1169
01:05:41,207 --> 01:05:43,200
decentralized network. 
If there's one liter and 

1170
01:05:43,200 --> 01:05:46,600
everyone assumes that the leader
has the unilateral ability to 

1171
01:05:46,600 --> 01:05:50,000
make a decision that leader. 
In fact does have the unilateral

1172
01:05:50,000 --> 01:05:52,600
ability to make a decision. 
We see this I think very 

1173
01:05:52,600 --> 01:05:57,500
strongly in aetherium It also is
strong in the sign at work. 

1174
01:05:57,500 --> 01:06:01,900
However, we are doing everything
we can to show people that we 

1175
01:06:01,900 --> 01:06:05,700
don't have the unilateral 
ability to make a decision. 

1176
01:06:05,700 --> 01:06:09,500
And the way we structured, our 
hard Fork, I think underscores 

1177
01:06:09,500 --> 01:06:11,600
very well. 
Like, I think it's the right way

1178
01:06:11,600 --> 01:06:16,300
to do it, and we made sure that 
any dissenting minority could 

1179
01:06:16,500 --> 01:06:18,900
opt out, or the hard Fork was 
opt-in. 

1180
01:06:18,900 --> 01:06:21,700
So you could choose not to opt 
into the hard Fork. 

1181
01:06:21,700 --> 01:06:25,300
You could neglect opt into the 
hard fork and Remain on the old 

1182
01:06:25,300 --> 01:06:28,900
Network. 
The important thing about 

1183
01:06:28,900 --> 01:06:34,300
decentralization is that every 
political decision each person 

1184
01:06:34,300 --> 01:06:39,300
can make on their own, and they 
can go any way they want and be 

1185
01:06:39,300 --> 01:06:42,900
successful. 
In activating this hard fork in 

1186
01:06:42,900 --> 01:06:46,500
activating the switch, the 
important thing was to make sure

1187
01:06:46,500 --> 01:06:50,300
that anybody who didn't like the
hard Fork, anybody who wanted to

1188
01:06:50,300 --> 01:06:54,200
resist the hard Fork, had easy 
trivial means to do so or any 

1189
01:06:54,200 --> 01:06:57,500
group that wanted to opt out and
not follow. 

1190
01:06:57,500 --> 01:07:01,400
The hard Fork would not have to 
work very hard, would have to do

1191
01:07:01,400 --> 01:07:05,500
very little work in order to 
maintain the old Network. 

1192
01:07:05,500 --> 01:07:10,500
So that was very important to us
and we did it came down to about

1193
01:07:10,900 --> 01:07:14,900
Things to make sure that this 
hard Fork was not our decision, 

1194
01:07:14,900 --> 01:07:17,600
but like the collectives 
decision each person on their 

1195
01:07:17,600 --> 01:07:20,500
own, got to decide. 
So the first thing is that, PSY 

1196
01:07:20,500 --> 01:07:22,500
network does not have mandatory 
upgrades. 

1197
01:07:22,800 --> 01:07:26,800
So every time you upgrade your 
sign owed, you have to do it 

1198
01:07:26,900 --> 01:07:28,400
yourself. 
And that means that the 

1199
01:07:28,400 --> 01:07:30,900
developers cannot push an update
out onto the network. 

1200
01:07:30,900 --> 01:07:34,200
So, we couldn't just sign a 
binary, send a bunch of messages

1201
01:07:34,200 --> 01:07:38,100
around and then have the default
Network be this hard for. 

1202
01:07:38,100 --> 01:07:41,100
So if we wanted to instigate a 
hard, Fork. 

1203
01:07:41,100 --> 01:07:44,600
We'd have to convince every 
member individually to upgrade. 

1204
01:07:45,600 --> 01:07:48,800
The second thing we did was, we 
did replay protection and wipe 

1205
01:07:48,800 --> 01:07:52,200
out protection and these 
conversations that came up a lot

1206
01:07:52,200 --> 01:07:54,300
in Bitcoin with USF and and 
whatnot. 

1207
01:07:54,300 --> 01:07:58,900
But what it means is that the 
transaction format on the hard 

1208
01:07:58,900 --> 01:08:01,900
Fork Network on the new network,
where we change, the proof work 

1209
01:08:01,900 --> 01:08:05,600
algorithm was slightly tweaked. 
So that transactions that you 

1210
01:08:05,600 --> 01:08:08,600
signed on the noon hour, we're 
completely invalid on the old 

1211
01:08:08,600 --> 01:08:11,400
Network and then vice versa, 
transaction signed on the old 

1212
01:08:11,400 --> 01:08:14,300
Network were completely invalid 
on the new network so you had 

1213
01:08:14,300 --> 01:08:17,899
no. 
No fear of, you know, trying to 

1214
01:08:17,899 --> 01:08:21,399
send someone saw a classic coins
and then receiving your side 

1215
01:08:21,399 --> 01:08:24,899
coins as well and no fear of, 
you know, vice versa happening. 

1216
01:08:25,200 --> 01:08:28,200
And then wipe out protection, 
similarly means that the 

1217
01:08:28,200 --> 01:08:31,600
proof-of-work algorithms are 
immediately different from each 

1218
01:08:31,600 --> 01:08:34,399
other on both networks. 
So if one network, you know, 

1219
01:08:34,399 --> 01:08:37,899
gets more work than the other 
notes will not switch from one 

1220
01:08:37,899 --> 01:08:40,899
network to the other that you 
don't have to worry about this, 

1221
01:08:40,899 --> 01:08:44,600
massive, this other network 51%,
attacking you, or wiping out a 

1222
01:08:44,600 --> 01:08:48,700
ton of History. 
So, the next thing we did was 

1223
01:08:48,899 --> 01:08:52,700
developer maintenance, we wanted
to make sure the people on the 

1224
01:08:52,700 --> 01:08:56,500
old Network would be trivially 
able to continue merging our 

1225
01:08:56,500 --> 01:08:58,600
updates. 
I'm going to continue, merging 

1226
01:08:58,600 --> 01:09:00,600
the code that were writing for 
the new network. 

1227
01:09:01,500 --> 01:09:05,500
If we had misread the situation 
and the true Side Community, the

1228
01:09:05,500 --> 01:09:08,399
users of the side Community. 
Really did not believe in the 

1229
01:09:08,399 --> 01:09:11,100
hard Fork. 
The Sigh Network would continue 

1230
01:09:11,100 --> 01:09:13,000
to benefit from all of the 
develop work. 

1231
01:09:13,000 --> 01:09:14,600
We were doing. 
We would see that we had made a 

1232
01:09:14,600 --> 01:09:16,200
mistake. 
We could switch back to the old 

1233
01:09:17,100 --> 01:09:22,600
So, the line of code difference 
between the hard fork and the 

1234
01:09:22,608 --> 01:09:25,300
not hard fork with the replay 
protection with the Wipeout 

1235
01:09:25,300 --> 01:09:26,600
protection, with the proof of 
work. 

1236
01:09:26,600 --> 01:09:29,800
Change was three lines of code 
and they're actually just 

1237
01:09:29,800 --> 01:09:31,800
configuration switches and a 
header somewhere. 

1238
01:09:32,800 --> 01:09:37,600
You do not need any developer 
ability to keep the two networks

1239
01:09:38,000 --> 01:09:40,100
synced with each other. 
In terms of code, there are no 

1240
01:09:40,100 --> 01:09:43,000
merge conflicts. 
There's no challenge in keeping 

1241
01:09:43,000 --> 01:09:45,399
the two networks updated, which 
we felt was important. 

1242
01:09:45,399 --> 01:09:49,100
You don't need an experienced 
developer to maintain the Sia, 

1243
01:09:49,100 --> 01:09:50,899
classic chain. 
You just need someone who knows 

1244
01:09:50,899 --> 01:09:53,200
how to click the merge button on
gitlab. 

1245
01:09:54,200 --> 01:09:59,100
Finally, we made it so that the 
storage network files would not 

1246
01:09:59,100 --> 01:10:01,400
be impacted. 
So whether you chose to go on to

1247
01:10:01,400 --> 01:10:03,800
the new sign at work or stay on 
the old sign at work. 

1248
01:10:04,100 --> 01:10:07,600
And whether your hosts chose to 
go onto the new sign at work or 

1249
01:10:07,600 --> 01:10:11,600
not, your files would stay 
intact, which I think was 

1250
01:10:11,600 --> 01:10:14,800
probably the neatest most 
interesting trick that we 

1251
01:10:14,800 --> 01:10:18,700
pulled. 
But basically, if you personally

1252
01:10:18,900 --> 01:10:21,900
did not like the hard Fork, you 
wouldn't have to go along with 

1253
01:10:21,900 --> 01:10:23,100
it, you wouldn't lose your 
files. 

1254
01:10:23,100 --> 01:10:25,900
Your He wouldn't be at risk, 
it'd be easy to maintain. 

1255
01:10:25,900 --> 01:10:28,700
You'd continue to receive, you 
know, updates as we made the 

1256
01:10:28,700 --> 01:10:30,900
network faster and more scalable
Etc. 

1257
01:10:31,900 --> 01:10:36,400
It's very easy for the community
to reject the leadership's 

1258
01:10:36,400 --> 01:10:40,700
decision to do this hard fork. 
And I think that's really what 

1259
01:10:40,700 --> 01:10:44,200
separates us from like a Google 
just passing down to users. 

1260
01:10:44,200 --> 01:10:48,000
Oh now your Android phones. 
Have DRM now your Block Chain 

1261
01:10:48,200 --> 01:10:50,500
uses a different proof of work 
algorithm and we breakdown 0. 

1262
01:10:50,500 --> 01:10:53,100
Silicon is very easy for the 
community. 

1263
01:10:53,500 --> 01:10:55,800
I know that's not what we want 
and we're not going to go along 

1264
01:10:55,800 --> 01:10:58,700
with it. 
Final thing is if you, you know,

1265
01:10:59,000 --> 01:11:01,400
live under a rock or whatever, 
you don't pay attention to any 

1266
01:11:01,400 --> 01:11:04,000
of the news because we can 
update the network. 

1267
01:11:04,000 --> 01:11:06,400
If you didn't update your 
software, if you didn't know 

1268
01:11:06,400 --> 01:11:09,500
that, this hard Fork thing was 
happening, your default was to 

1269
01:11:09,500 --> 01:11:13,100
end up on the old Network. 
So no one no one ended up on the

1270
01:11:13,100 --> 01:11:16,400
hard Fork on accident they chose
to upgrade to the hard Fork 

1271
01:11:16,400 --> 01:11:20,000
Network. 
That kind of illustrates the 

1272
01:11:20,000 --> 01:11:21,800
lengths. 
We went to to make sure that 

1273
01:11:21,800 --> 01:11:25,200
this was a decision not handed 
down from the top. 

1274
01:11:25,200 --> 01:11:28,600
But one made by the community, 
in a in a decentralized way. 

1275
01:11:29,800 --> 01:11:32,900
Going back to the for Content 
itself. 

1276
01:11:33,200 --> 01:11:35,700
Would you primarily sort of 
hurt? 

1277
01:11:35,700 --> 01:11:39,400
It bit me and had already sold 
all of their a sex at this 

1278
01:11:39,400 --> 01:11:41,100
point. 
Weren't you no longer hurting B 

1279
01:11:41,100 --> 01:11:43,100
made, but now you're kind of 
punishing. 

1280
01:11:43,100 --> 01:11:45,600
Everyone who bought a 64-bit 
Maine. 

1281
01:11:46,700 --> 01:11:51,100
Yes, which is a good point at in
practice. 

1282
01:11:51,100 --> 01:11:54,500
Actually, the only people we 
were hurting were in a silicon 

1283
01:11:55,200 --> 01:11:59,700
and silicon was mining more than
51% of the hash rate themselves.

1284
01:11:59,700 --> 01:12:02,900
So it wasn't customers of a no 
silicon, who was mining. 

1285
01:12:03,300 --> 01:12:06,700
There's in a silicon themselves 
who is mining customers have 

1286
01:12:06,700 --> 01:12:09,900
been 0 silicon had already lost 
90 plus percent of their money, 

1287
01:12:09,900 --> 01:12:13,100
just giving it in our silicon at
these ridiculous. 

1288
01:12:13,100 --> 01:12:16,900
Margins with these assumptions 
that in our silicon new at, At 

1289
01:12:16,900 --> 01:12:20,400
time of sale were incorrect, but
it was more than happy to sell 

1290
01:12:20,400 --> 01:12:24,100
an overpriced device to a 
customer that main was in a 

1291
01:12:24,108 --> 01:12:26,800
similar situation. 
All their customers that already

1292
01:12:26,800 --> 01:12:30,900
lost all their money because one
bit main oversold to the money 

1293
01:12:30,900 --> 01:12:33,800
was all gone anyway into in a 
silicon came in and completely 

1294
01:12:33,800 --> 01:12:36,800
destroyed bit mate. 
So, all the money was gone. 

1295
01:12:37,100 --> 01:12:41,000
Anyway, so in practice, we were 
only hurting in a silicon but 

1296
01:12:41,000 --> 01:12:45,400
philosophically I think if 
you're going to buy Hardware to 

1297
01:12:45,400 --> 01:12:49,800
mine, On a network. 
It's kind of your imperative to 

1298
01:12:49,800 --> 01:12:52,100
do research and make sure that 
your purchase is not 

1299
01:12:52,100 --> 01:12:55,900
contributing to the downfall of 
that Network or, you know, 

1300
01:12:55,900 --> 01:12:58,400
contributing to drama or 
contributing, to instability 

1301
01:12:58,400 --> 01:13:00,500
Etc. 
And so even though we're not 

1302
01:13:00,500 --> 01:13:03,700
hurting bit main directly, I do 
think there is some moral 

1303
01:13:03,700 --> 01:13:06,800
responsibility on the people who
gave bit made money and 

1304
01:13:06,800 --> 01:13:09,500
contributed to the disaster that
happened. 

1305
01:13:09,800 --> 01:13:12,300
But as it were, we didn't hurt 
those people anyway. 

1306
01:13:13,400 --> 01:13:16,000
I don't know how you can say 
that mean, like the people that 

1307
01:13:16,000 --> 01:13:19,000
are buying the harder from bit 
manner consumers, the people 

1308
01:13:19,000 --> 01:13:21,300
that are buying a hardware 
formula silicon as consumers, 

1309
01:13:21,300 --> 01:13:25,700
they're looking for, you know, 
the best minor. 

1310
01:13:26,100 --> 01:13:29,800
And they might not even have 
that much skin in the game when 

1311
01:13:29,800 --> 01:13:34,300
it comes to Sia. 
To put that on the consumer as 

1312
01:13:34,300 --> 01:13:38,400
their responsibility to choose 
the right minor and have to 

1313
01:13:38,400 --> 01:13:42,000
understand all these political 
implications, and all this 

1314
01:13:42,000 --> 01:13:45,500
drama, you know, happening 
underneath all of this, it's 

1315
01:13:45,500 --> 01:13:48,600
nonsensical. 
I mean maybe this goes down to 

1316
01:13:48,600 --> 01:13:51,400
like the application specific 
chain and look even better. 

1317
01:13:51,400 --> 01:13:54,700
That's why he believes in A6. 
Were he expects miners to be 

1318
01:13:54,700 --> 01:13:57,400
active participant in the 
network? 

1319
01:13:58,400 --> 01:14:01,000
You can be if you want but you 
know, have to be an active 

1320
01:14:01,000 --> 01:14:03,500
participant Network. 
You can see the simply as an 

1321
01:14:03,500 --> 01:14:05,100
investment opportunity or 
whatever. 

1322
01:14:06,000 --> 01:14:10,100
I do want to come back to Sia 
and the evolution of 

1323
01:14:10,100 --> 01:14:13,200
decentralized storage and get 
your thoughts and word. 

1324
01:14:13,200 --> 01:14:20,000
You see decentralized storage 
and 35 and even maybe 10 years 

1325
01:14:20,000 --> 01:14:23,100
ahead. 
And the next one year, we're 

1326
01:14:23,100 --> 01:14:27,200
going to see a lot of power 
users starting to put their 

1327
01:14:27,200 --> 01:14:30,800
storage onto decentralized 
mediums. 

1328
01:14:30,800 --> 01:14:33,400
I believe so, probably as a 
secondary backup. 

1329
01:14:33,400 --> 01:14:35,800
People are increasingly getting 
nervous about the control that 

1330
01:14:35,800 --> 01:14:39,300
AWS has about the control that 
Google cloud has size a very 

1331
01:14:39,300 --> 01:14:41,700
cheap alternative. 
So if you're worried about 

1332
01:14:41,700 --> 01:14:45,100
Google handing down this, like 
nasty terms of service decision,

1333
01:14:45,100 --> 01:14:47,300
L we terminated your account. 
You don't get your data back, 

1334
01:14:47,300 --> 01:14:51,100
you know you're done sigh of for
just a little bit of her head 

1335
01:14:51,100 --> 01:14:54,100
like a ten percent overhead of 
what you pay Google, you can 

1336
01:14:54,100 --> 01:14:57,200
ensure that your data is immune 
to this sort of centralized 

1337
01:14:57,200 --> 01:14:59,500
decision. 
So I think over the next year 

1338
01:14:59,500 --> 01:15:01,600
you know we're going to see 
increasingly people moving that 

1339
01:15:01,600 --> 01:15:04,700
direction in the two to three 
year Horizon. 

1340
01:15:04,700 --> 01:15:07,900
I think we're also going to see 
a big shift in how contents 

1341
01:15:07,900 --> 01:15:11,600
distributed on the internet. 
So you know when you go to Vimeo

1342
01:15:11,600 --> 01:15:14,100
and you watch a video when you 
go to Netflix and watch a video,

1343
01:15:14,400 --> 01:15:16,700
I think it's very likely that 
that data is going to be coming 

1344
01:15:16,700 --> 01:15:20,700
from a decentralized network as 
opposed to the current like 

1345
01:15:20,900 --> 01:15:22,500
Centralized content 
distribution. 

1346
01:15:23,700 --> 01:15:27,000
When you do live video 
streaming, how are we making it 

1347
01:15:27,000 --> 01:15:30,700
efficient to like get data from 
so many different sources? 

1348
01:15:30,700 --> 01:15:35,500
Decrypt it and like reverse 
Erasure coded and to provide a 

1349
01:15:35,508 --> 01:15:39,700
seamless streaming experience. 
It basically comes down to a lot

1350
01:15:39,700 --> 01:15:41,900
of parallelism. 
The data needs to go through a 

1351
01:15:41,900 --> 01:15:45,700
fan-out process because it, you 
know, that the uploader only has

1352
01:15:45,700 --> 01:15:47,400
so much bandwidth. 
So they're going to upload to 

1353
01:15:47,400 --> 01:15:50,200
someone who has a lot more 
bandwidth and that person's 

1354
01:15:50,200 --> 01:15:52,800
probably not going to distribute
directly to users, they're 

1355
01:15:52,800 --> 01:15:55,100
probably going to upload 
themselves to people with even 

1356
01:15:55,100 --> 01:15:57,400
more bandwidth. 
So you get this big fan out that

1357
01:15:57,400 --> 01:16:02,100
takes maybe one second or so' 
fan out, you know, three or four

1358
01:16:02,100 --> 01:16:05,300
layers, and then once your 
fanned out four layers now, now 

1359
01:16:05,300 --> 01:16:08,000
you have, I mean, four layers is
enough to have the entire 

1360
01:16:08,200 --> 01:16:11,000
Bandwidth of the sign Network to
distribute content. 

1361
01:16:11,000 --> 01:16:15,300
So the answer is a lot of 
scalability, engineering 

1362
01:16:15,300 --> 01:16:18,900
essentially, these can all be 
done using classical techniques 

1363
01:16:18,900 --> 01:16:21,400
and then you just make sure that
the The Primitives are 

1364
01:16:21,400 --> 01:16:25,800
decentralized I'd like to get 
your thoughts on this and I'm 

1365
01:16:25,800 --> 01:16:28,300
very much looking forward to 
having like more decentralized 

1366
01:16:28,300 --> 01:16:32,000
storage solutions available and 
people using decentralized 

1367
01:16:32,000 --> 01:16:34,400
Origins. 
I think like we agree on on the 

1368
01:16:34,400 --> 01:16:37,100
underlying premise that one 
would be better off. 

1369
01:16:37,200 --> 01:16:39,400
Storing their files in a 
personal files on be centralized

1370
01:16:39,400 --> 01:16:42,400
storage, that's encrypted, 
that's censorship resistant 

1371
01:16:42,400 --> 01:16:46,200
rather than on like Google or 
Google Cloud, some like that I 

1372
01:16:46,208 --> 01:16:51,400
see a problem with the space 
getting big enough to sustain 

1373
01:16:51,700 --> 01:16:55,000
storage needs, one of the blog 
posts on your blog. 

1374
01:16:55,000 --> 01:16:58,800
I forget which one it says that 
you know by 2025 will were 

1375
01:16:58,800 --> 01:17:04,300
expected to create like over 400
petabytes of data per day and to

1376
01:17:04,300 --> 01:17:09,000
expect that decentralized 
storage networks which rely you 

1377
01:17:09,000 --> 01:17:11,500
know if we want to keep it to 
centralize primarily on storage 

1378
01:17:11,500 --> 01:17:15,400
provided by individuals to 
expect that storage needs. 

1379
01:17:15,500 --> 01:17:20,300
Need to be met by that. 
I think is we're unlikely to see

1380
01:17:20,300 --> 01:17:22,600
that. 
And the reason why I think 

1381
01:17:22,600 --> 01:17:27,800
that's the case is that there is
a trend where perhaps 10 or 15 

1382
01:17:27,800 --> 01:17:32,200
years ago, a lot of people had 
storage in their house, right? 

1383
01:17:32,200 --> 01:17:34,700
Like there were probably more 
people that I mean, this is my 

1384
01:17:34,700 --> 01:17:38,700
assumption, but probably more 
people had NOS hard drives or 

1385
01:17:38,708 --> 01:17:43,200
USB hard drives because cloud 
storage wasn't so prevalent and 

1386
01:17:43,200 --> 01:17:47,500
cheap and even like free, but in
Namely, it's becoming Niche 

1387
01:17:47,500 --> 01:17:50,900
thing, you know, I think I know 
three people that have like an 

1388
01:17:50,900 --> 01:17:54,900
ass or like a, an external hard 
drive that's constantly 

1389
01:17:54,900 --> 01:17:57,200
connected to the internet. 
And I have the same issue with 

1390
01:17:57,200 --> 01:17:59,400
like something like space mesh, 
for example, which built on the 

1391
01:17:59,400 --> 01:18:03,200
same premise that there will be 
an abundance of decentralized. 

1392
01:18:03,400 --> 01:18:07,700
Highly available storage 
available out there, Is this 

1393
01:18:07,700 --> 01:18:09,200
something you guys have thought 
about? 

1394
01:18:09,200 --> 01:18:12,300
And how do you counter this? 
You know what? 

1395
01:18:12,300 --> 01:18:15,200
If we go into a trend where 
people just don't have a whole 

1396
01:18:15,200 --> 01:18:18,600
lot of storage available to put 
up on the network? 

1397
01:18:18,900 --> 01:18:22,200
This is something we've thought 
about a lot, basically, from day

1398
01:18:22,200 --> 01:18:25,200
one, and you'll see in a second 
that we answer this problem, 

1399
01:18:25,200 --> 01:18:29,400
very effectively to justify your
concern when we looked at other 

1400
01:18:29,400 --> 01:18:33,800
peer-to-peer storage systems 
that had popped up in the past, 

1401
01:18:33,800 --> 01:18:38,400
whether it's BitTorrent Space, 
Monkey, send form, all of them, 

1402
01:18:38,400 --> 01:18:40,700
kind of were built with this 
assumption that the people who 

1403
01:18:40,700 --> 01:18:44,800
are storing data and the people 
who have storage are the same 

1404
01:18:44,800 --> 01:18:47,200
people. 
When this really like cause 

1405
01:18:47,200 --> 01:18:50,300
struggles on the network, they 
As had these imbalances where 

1406
01:18:50,300 --> 01:18:53,900
either they had, you know, too 
much storage or they had too 

1407
01:18:53,900 --> 01:18:57,700
much data and it was very 
difficult to convince people who

1408
01:18:57,700 --> 01:19:01,600
wanted to use you know say 10 
terabytes of cloud to also have 

1409
01:19:01,600 --> 01:19:04,300
20 terabytes at home serving up 
to the cloud. 

1410
01:19:04,300 --> 01:19:08,600
So in Scio we intentionally very
sharply divorce, these two 

1411
01:19:08,600 --> 01:19:13,600
situations, that's why we pay 
people So we actually envisioned

1412
01:19:13,600 --> 01:19:16,900
for the sign network is just 
like mining Farms. 

1413
01:19:16,900 --> 01:19:19,000
You know, we have mining Farms 
all over the world that do 

1414
01:19:19,000 --> 01:19:23,100
proof-of-work Mining and they do
it because they make money, not 

1415
01:19:23,100 --> 01:19:26,000
because they have some, you 
know, ideological 

1416
01:19:26,200 --> 01:19:29,200
decentralization imperative, but
because there's profit to be 

1417
01:19:29,200 --> 01:19:31,000
made. 
So we see the same thing for 

1418
01:19:31,000 --> 01:19:34,900
Sia, we pay people to store data
on an Open Marketplace. 

1419
01:19:35,300 --> 01:19:39,600
If there's a lot of demand for 
storage, people will set up data

1420
01:19:39,600 --> 01:19:43,000
centers that are designed And 
just to service this I Network 

1421
01:19:43,300 --> 01:19:46,500
and so we will be able to meet 
an unlimited amount of demand. 

1422
01:19:46,500 --> 01:19:50,600
So long as that that demand is 
unlimited at your, at a price 

1423
01:19:50,600 --> 01:19:52,400
point. 
That makes sense for people to 

1424
01:19:52,400 --> 01:19:54,300
open up data centers. 
And so, so we have this 

1425
01:19:54,300 --> 01:19:57,400
Marketplace and we don't 
actually expect storage to come 

1426
01:19:57,400 --> 01:20:00,000
from individuals. 
We expect storage to come from, 

1427
01:20:00,300 --> 01:20:04,400
professional data centers that 
are tuned to, you know, be 

1428
01:20:04,400 --> 01:20:08,900
profitable on the sign Network. 
You're expecting a sort of 

1429
01:20:08,900 --> 01:20:15,900
parallel Data Center Market to 
emerge as a parallel, to the AWS

1430
01:20:15,900 --> 01:20:20,000
is the Azure, is the Google's of
the world that are serving 

1431
01:20:20,100 --> 01:20:24,600
purely serving the sign at work 
and providing storage for that 

1432
01:20:24,600 --> 01:20:26,500
decentralized Network. 
And perhaps even other 

1433
01:20:26,500 --> 01:20:30,000
decentralized storage networks. 
Yep, that's correct. 

1434
01:20:30,800 --> 01:20:35,900
And what about leveraging 
existing storage like partnering

1435
01:20:35,900 --> 01:20:40,300
with With manufacturers of 
embedded devices and using that,

1436
01:20:40,300 --> 01:20:44,300
as a way, to incentivize people 
to basically pay off devices 

1437
01:20:44,300 --> 01:20:46,500
that they would buy, right? 
So if you have like a Roku or 

1438
01:20:46,500 --> 01:20:49,900
something, that has a hard drive
in it, well then as a user of 

1439
01:20:49,900 --> 01:20:52,000
that hard drive, you're already 
getting your automatically 

1440
01:20:52,000 --> 01:20:54,000
getting back. 
Sort of like you're paying for 

1441
01:20:54,000 --> 01:20:57,700
it with these these Sia rewards 
that you're gaining from just 

1442
01:20:57,700 --> 01:21:01,000
having it plugged in. 
We've looked at that a number of

1443
01:21:01,000 --> 01:21:03,000
times, and I think the 
conclusion is actually always 

1444
01:21:03,000 --> 01:21:04,600
been the same. 
Is that the economics? 

1445
01:21:04,600 --> 01:21:08,500
Don't, don't make sense. 
Once you have these kind of like

1446
01:21:08,500 --> 01:21:12,900
mining, it be the same sort of 
idea, as you know just having 

1447
01:21:12,900 --> 01:21:15,700
your phone mining all the time. 
The truth is that your phone is 

1448
01:21:15,700 --> 01:21:18,100
not a device that's optimized 
around Mining. 

1449
01:21:18,100 --> 01:21:21,700
And therefore it's one, two, 
three, four, five orders of 

1450
01:21:21,700 --> 01:21:25,900
magnitude less efficient than a 
6 in a Data Center. 

1451
01:21:26,200 --> 01:21:28,600
And I think storage is going to 
end up being much the same way, 

1452
01:21:28,600 --> 01:21:31,000
you know. 
Oh if you took if you increase 

1453
01:21:31,000 --> 01:21:34,700
the cost of like say a smart TV 
by a hundred dollars to put a 

1454
01:21:34,708 --> 01:21:37,300
bunch of storage in it, that 
smart TV is going to make the 

1455
01:21:37,300 --> 01:21:41,600
company, may be one dollar a 
month in storage, Revenue by 

1456
01:21:41,600 --> 01:21:44,800
serving on the sign Network. 
Where is like an optimize data 

1457
01:21:44,800 --> 01:21:47,800
center can get that same $1 a 
month of revenue for maybe an 

1458
01:21:47,800 --> 01:21:53,300
additional $25 a build-out cost.
Competitively speaking, I think 

1459
01:21:53,700 --> 01:21:58,400
even if the smart TV can lean in
on things like free space. 

1460
01:21:58,700 --> 01:22:01,600
Yeah, like free physical space 
free electricity Etc. 

1461
01:22:01,600 --> 01:22:04,000
It's still just doesn't make 
economic sense. 

1462
01:22:04,000 --> 01:22:08,000
So I don't see that happening in
the future just purely because 

1463
01:22:08,500 --> 01:22:12,000
economies of scale are really, 
really sharp when you're 

1464
01:22:12,000 --> 01:22:14,800
specializing. 
Very much looking forward to 

1465
01:22:14,800 --> 01:22:20,100
seeing how this plays out and I 
know I'll probably be using Sia,

1466
01:22:20,400 --> 01:22:24,400
providing spaces a host of my 
Synology Nas will be glad to 

1467
01:22:24,400 --> 01:22:26,200
have you. 
All right, thanks a lot for 

1468
01:22:26,200 --> 01:22:27,800
coming on the show. 
David thank you. 

1469
01:22:28,000 --> 01:22:31,900
Glad to be here. 
Thank you for joining us on this

1470
01:22:31,900 --> 01:22:34,300
week's episode. 
We release new episodes every 

1471
01:22:34,300 --> 01:22:36,200
week. 
You can find And subscribe to 

1472
01:22:36,208 --> 01:22:40,100
the show on iTunes Spotify, 
YouTube SoundCloud or wherever 

1473
01:22:40,100 --> 01:22:42,400
you listen to podcast. 
And if you have a Google home or

1474
01:22:42,400 --> 01:22:45,300
Alexa device, you can tell it to
listen to the latest episode of 

1475
01:22:45,300 --> 01:22:49,200
the epicenter podcast, go to 
epicenter, .t V /, subscribe for

1476
01:22:49,200 --> 01:22:51,800
a full list of places where you 
can watch and listen, while 

1477
01:22:51,800 --> 01:22:54,100
you're there, be sure to sign up
for the newsletter so you get 

1478
01:22:54,100 --> 01:22:57,400
new episodes in your inbox as 
they're released if you want to 

1479
01:22:57,407 --> 01:23:00,400
interact with Us guest earlier. 
A podcast listeners you can 

1480
01:23:00,400 --> 01:23:03,100
follow us on Twitter and please 
leave us a review on iTunes 

1481
01:23:03,300 --> 01:23:05,700
helps people find the show and 
we're always happy to read them.

1482
01:23:06,500 --> 01:23:09,100
Well thanks so much and we look 
forward to being back next week.