1
00:00:00,000 --> 00:00:04,160
Welcome to the deep dive our mission here is well, it's pretty simple

2
00:00:04,160 --> 00:00:08,880
We take these really complex ground breaking projects the kind of stuff

3
00:00:08,880 --> 00:00:13,100
You'd normally need a PhD to understand and we try to just you know, pull out the

4
00:00:13,100 --> 00:00:13,940
core insights for you

5
00:00:13,940 --> 00:00:18,440
And today we are tackling a huge one. We're not just talking about a better app or

6
00:00:18,440 --> 00:00:18,980
an upgrade

7
00:00:18,980 --> 00:00:21,680
No, we're talking about framework that is designed to

8
00:00:21,680 --> 00:00:26,820
To replace the entire foundation of the internet. It's all about security and

9
00:00:26,820 --> 00:00:28,440
privacy by design

10
00:00:28,440 --> 00:00:32,860
We are diving into gene in it exactly and we're gonna explore why the network we

11
00:00:32,860 --> 00:00:34,300
all use every day is well

12
00:00:34,300 --> 00:00:39,050
Fundamentally broken and how this peer-to-peer approach aims to fix it. It's a big

13
00:00:39,050 --> 00:00:39,800
topic it is

14
00:00:39,800 --> 00:00:43,260
So before we peel back all those layers, we have to give a huge and necessary

15
00:00:43,260 --> 00:00:46,200
Thank you to the supporter of this deep dive safe server

16
00:00:46,200 --> 00:00:50,100
Mm-hmm safe server focuses on hosting exactly this type of software and they

17
00:00:50,100 --> 00:00:52,440
support you in your digital transformation efforts

18
00:00:52,440 --> 00:00:54,480
You can find out a lot more at

19
00:00:54,720 --> 00:01:00,250
WWW safe server dot DE. Okay. So the whole premise here is that the internet we

20
00:01:00,250 --> 00:01:00,640
rely on

21
00:01:00,640 --> 00:01:04,960
It was never actually built to protect us, right? It has these deep

22
00:01:04,960 --> 00:01:10,220
Architectural flaws that you know, they allow for mass surveillance for centralization

23
00:01:10,220 --> 00:01:12,540
of power and gnet is the answer to that

24
00:01:12,540 --> 00:01:14,120
It's a decades-long

25
00:01:14,120 --> 00:01:19,800
Academic effort to go back and fix those underlying flaws to build a network where

26
00:01:19,800 --> 00:01:22,200
privacy isn't just an add-on. It's

27
00:01:23,160 --> 00:01:27,120
It's guaranteed. Okay, so let's unpack this the source material kicks off with this

28
00:01:27,120 --> 00:01:30,840
really powerful analogy to help us get our heads around just how

29
00:01:30,840 --> 00:01:34,480
Damaged the current system is yeah, I love this one

30
00:01:34,480 --> 00:01:37,880
They describe the conventional internet as a vast system of roads

31
00:01:37,880 --> 00:01:43,550
But these roads are just filled with these deep unavoidable potholes and worse they're

32
00:01:43,550 --> 00:01:46,000
crawling with highwaymen a perfect metaphor

33
00:01:46,000 --> 00:01:48,000
Really? It shows a systemic failure

34
00:01:48,000 --> 00:01:52,400
Imagine you're driving your car and your car is your data or your device down this

35
00:01:52,400 --> 00:01:52,760
road

36
00:01:52,760 --> 00:01:56,510
Okay, even if you lock the doors the car itself could be hijacked the long arms of

37
00:01:56,510 --> 00:01:56,880
these

38
00:01:56,880 --> 00:02:01,950
Unaccountable actors can just reach in and steal your stuff and they use that data

39
00:02:01,950 --> 00:02:03,680
against you or they sell it, right?

40
00:02:03,680 --> 00:02:07,040
And the scariest part of that which I think really captures the modern problem

41
00:02:07,040 --> 00:02:10,240
Is that you the driver you often can't even tell it's happening

42
00:02:10,240 --> 00:02:15,310
You don't even notice the theft and you certainly can't hold the scroungers as they

43
00:02:15,310 --> 00:02:17,680
call them accountable in any real way

44
00:02:17,840 --> 00:02:22,060
So where does that weakness come from? Well, it traces right back to the 70s. The

45
00:02:22,060 --> 00:02:23,560
sources are really clear on this

46
00:02:23,560 --> 00:02:28,140
The original internet protocols were just not developed with security in mind at

47
00:02:28,140 --> 00:02:29,200
all not at all

48
00:02:29,200 --> 00:02:34,130
So try to stop someone from say faking their network address or preventing a

49
00:02:34,130 --> 00:02:36,440
central router from learning your metadata

50
00:02:36,440 --> 00:02:41,210
It's incredibly hard. It's non-trivial and sometimes it's just plain impossible

51
00:02:41,210 --> 00:02:42,480
with the current architecture

52
00:02:42,720 --> 00:02:47,610
Now most of us think we've got this covered right we use VPNs HTTPS end-to-end

53
00:02:47,610 --> 00:02:48,560
encryption

54
00:02:48,560 --> 00:02:53,040
But the sources really stress that focusing just on the content of the message

55
00:02:53,040 --> 00:02:58,250
Misses the biggest threat the metadata metadata. Exactly. Absolutely because your

56
00:02:58,250 --> 00:02:58,880
metadata

57
00:02:58,880 --> 00:03:03,040
That's who you talk to when you talk to them how often how much data you exchange

58
00:03:03,040 --> 00:03:08,580
That is profoundly revealing more so than the message itself sometimes often

59
00:03:08,580 --> 00:03:13,660
Yeah, so the sources make this massive claim that exposing this metadata actually

60
00:03:13,660 --> 00:03:15,400
threatens democracy itself

61
00:03:15,400 --> 00:03:15,800
Okay

62
00:03:15,800 --> 00:03:20,260
how that seems like a huge leap it all connects to the chilling effect if a

63
00:03:20,260 --> 00:03:22,800
powerful entity be it a corporation or a

64
00:03:22,800 --> 00:03:27,620
Government can map out who is talking to whom they can identify journalists sources

65
00:03:27,620 --> 00:03:29,400
or dissidents precisely

66
00:03:29,400 --> 00:03:33,040
They do it just by analyzing the timing and frequency of communication

67
00:03:33,040 --> 00:03:38,260
So what happens people's ability to exercise free speech just erodes you start to

68
00:03:38,260 --> 00:03:39,060
self censor

69
00:03:39,060 --> 00:03:44,220
I see so even with your messages encrypted this metadata trail allows for a total

70
00:03:44,220 --> 00:03:44,480
identification

71
00:03:44,480 --> 00:03:49,370
Centralization of power and ultimately political control that paints a very clear

72
00:03:49,370 --> 00:03:50,720
picture of the problem

73
00:03:50,720 --> 00:03:55,800
So if the old internet is that broken crime-ridden road then Geonet is the project

74
00:03:55,800 --> 00:03:56,940
trying to build a brand new

75
00:03:56,940 --> 00:04:00,450
Secure freeway underneath it all okay. Let's pivot to the solution though for a

76
00:04:00,450 --> 00:04:02,240
beginner someone just hearing about this

77
00:04:02,240 --> 00:04:05,710
How would you define gene it you can think of it as a whole new set of rules for

78
00:04:05,710 --> 00:04:06,600
the internet a?

79
00:04:06,600 --> 00:04:12,400
Network protocol stack a stack okay, and its whole goal is to be a direct technical

80
00:04:12,400 --> 00:04:14,140
replacement for the old

81
00:04:14,140 --> 00:04:19,970
Insecure stack we use now and it achieves that privacy by design idea. Yes, not

82
00:04:19,970 --> 00:04:22,400
through patches or add-ons, but through

83
00:04:22,400 --> 00:04:25,900
Fundamental improvements to how things like

84
00:04:26,240 --> 00:04:30,490
Addressing routing and naming actually work. It sounds like they're building a

85
00:04:30,490 --> 00:04:33,200
really complex foundation not just one simple app

86
00:04:33,200 --> 00:04:37,060
Why is that framework approach so necessary for privacy? Why not just build a

87
00:04:37,060 --> 00:04:39,040
better app? That is a crucial question

88
00:04:39,040 --> 00:04:42,990
I mean if you look at the bigger picture in the peer-to-peer space you just see

89
00:04:42,990 --> 00:04:44,440
massive inefficiency

90
00:04:44,440 --> 00:04:48,480
How so every single PDP project is constantly reinventing the wheel

91
00:04:48,480 --> 00:04:52,960
They all spend huge amounts of time and money trying to build reliable encrypted

92
00:04:52,960 --> 00:04:55,160
channels or tools for resource discovery

93
00:04:55,160 --> 00:04:59,510
And if they get it even slightly wrong, the users metadata is instantly leaked all

94
00:04:59,510 --> 00:05:00,560
that work is for nothing

95
00:05:00,560 --> 00:05:05,060
So G minute is providing those common pre-built components that are already robust

96
00:05:05,060 --> 00:05:07,200
and tested to be metadata preserving

97
00:05:07,200 --> 00:05:11,850
Exactly. It offers that solid foundation. It's rooted in you know, almost two

98
00:05:11,850 --> 00:05:14,120
decades of continuous academic research

99
00:05:14,120 --> 00:05:18,290
It's designed to withstand intense scrutiny. So developers can just focus on

100
00:05:18,290 --> 00:05:19,240
building their apps

101
00:05:19,680 --> 00:05:23,080
Let's move beyond the tech for a second and look at the philosophy behind it

102
00:05:23,080 --> 00:05:25,480
because that seems to inform all the design choices

103
00:05:25,480 --> 00:05:29,760
You said the goal is to support a free and open society. That's right

104
00:05:29,760 --> 00:05:35,500
I mean today our freedoms are restricted by constant monitoring by centralized

105
00:05:35,500 --> 00:05:36,280
network hubs

106
00:05:36,280 --> 00:05:41,440
By software we can't even inspect and things like traffic shapers and firewalls

107
00:05:41,440 --> 00:05:42,160
that just block

108
00:05:42,160 --> 00:05:46,720
Communication all of it and here's where it gets for me really interesting the

109
00:05:46,720 --> 00:05:49,080
software itself embodies that philosophy

110
00:05:49,160 --> 00:05:53,600
The project is explicitly free software as in freedom, right?

111
00:05:53,600 --> 00:05:57,860
Which gives the user four specific guarantees the freedom to run the network to

112
00:05:57,860 --> 00:06:00,140
study the code to share information and to

113
00:06:00,140 --> 00:06:02,800
Modify it or build new things on top of it

114
00:06:02,800 --> 00:06:04,680
It's all about putting you back in control of your data

115
00:06:04,680 --> 00:06:08,570
You determine what you share and with who without that external pressure to

116
00:06:08,570 --> 00:06:12,680
compromise your privacy just to participate now speaking of philosophy

117
00:06:12,680 --> 00:06:16,200
The sources take a very strong I'd say almost

118
00:06:17,120 --> 00:06:19,940
controversial stance against some current tech trends

119
00:06:19,940 --> 00:06:22,560
specifically

120
00:06:22,560 --> 00:06:24,060
blockchains and

121
00:06:24,060 --> 00:06:28,100
Distributed ledgers. They absolutely do the sources state that the hype around all

122
00:06:28,100 --> 00:06:30,200
this is and I'm quoting here

123
00:06:30,200 --> 00:06:34,200
Detrimental to the health of our planet Wow, that's a serious claim

124
00:06:34,200 --> 00:06:38,760
But wait, isn't a decentralized ledger like the gold standard right now for trust

125
00:06:38,760 --> 00:06:40,440
without a central authority

126
00:06:40,440 --> 00:06:45,410
How can Gina do that without it and that's the key difference? Yes ledgers achieve

127
00:06:45,410 --> 00:06:45,680
trust

128
00:06:45,680 --> 00:06:49,860
But they do it through these incredibly wasteful consensus mechanisms. You mean

129
00:06:49,860 --> 00:06:51,760
like proof-of-work mining exactly

130
00:06:51,760 --> 00:06:55,480
It requires massive continuous energy use just to keep a public record that

131
00:06:55,480 --> 00:06:56,400
everyone agrees on

132
00:06:56,400 --> 00:07:00,820
The sources argue these systems just don't scale and are frankly

133
00:07:00,820 --> 00:07:05,230
Ecologically unsustainable so Junionet's approach is fundamentally different

134
00:07:05,230 --> 00:07:08,660
completely. They use decentralization only where it provides the most value

135
00:07:08,660 --> 00:07:12,550
Like for naming or key management, they don't require a public globally validated

136
00:07:12,550 --> 00:07:14,560
ledger for every little interaction

137
00:07:14,560 --> 00:07:18,440
So it's more target. It's smart targeted decentralization. Not this broad

138
00:07:18,440 --> 00:07:20,180
computationally expensive approach

139
00:07:20,180 --> 00:07:24,550
It provides a path that is actually efficient and sustainable. This all sounds

140
00:07:24,550 --> 00:07:26,100
incredibly ambitious

141
00:07:26,100 --> 00:07:30,300
Maybe a bit high-level. So what does it all mean for the person listening at home?

142
00:07:30,300 --> 00:07:33,340
What can you actually do with Junionet today?

143
00:07:33,340 --> 00:07:36,590
Well, a number of applications have already been built on the framework that show

144
00:07:36,590 --> 00:07:37,740
what it can do. Like what?

145
00:07:37,740 --> 00:07:41,200
We're looking at things like anonymous and robust file sharing

146
00:07:42,020 --> 00:07:45,020
Decentralized and confidential telephony and you mentioned

147
00:07:45,020 --> 00:07:50,810
Decentralized naming and identity which sounds critical. It is it's called GNS the

148
00:07:50,810 --> 00:07:52,020
GNU name system

149
00:07:52,020 --> 00:07:55,380
Think of it as a replacement for DNS the domain name system

150
00:07:55,380 --> 00:07:59,740
We use now which is a huge point of centralization and control a massive one

151
00:07:59,740 --> 00:08:03,570
Yeah, GNS lets users manage their own identities and name their own services

152
00:08:03,570 --> 00:08:06,780
without relying on those centrally controlled authorities

153
00:08:06,780 --> 00:08:11,400
It's a huge step towards real network autonomy and beyond GNS and telephony you

154
00:08:11,400 --> 00:08:14,640
said file sharing. Yes anonymous file sharing

155
00:08:14,640 --> 00:08:19,920
It's sort of the original promise of P2P tech, but actually secured and anonymized

156
00:08:19,920 --> 00:08:21,500
at the protocol level

157
00:08:21,500 --> 00:08:26,700
They've also developed a way to tunnel normal IP traffic over GNE app

158
00:08:26,700 --> 00:08:31,180
So you could use it as a kind of anonymizing layer for your regular web browsing

159
00:08:31,180 --> 00:08:31,820
potentially

160
00:08:31,820 --> 00:08:35,950
Yes. Now it's really important for our listeners to know where the project stands.

161
00:08:35,950 --> 00:08:38,140
This isn't finished polished software

162
00:08:38,140 --> 00:08:42,540
Is it no not at all. The sources are very clear. It is an LBHA release

163
00:08:42,540 --> 00:08:47,490
That means there are known significant bugs. There are missing features. This is

164
00:08:47,490 --> 00:08:48,580
the research frontier

165
00:08:48,580 --> 00:08:51,060
But it is usable for someone who's curious

166
00:08:51,060 --> 00:08:55,520
Okay, so for those in our audience who are comfortable messing around with their

167
00:08:55,520 --> 00:08:58,180
operating system, how do they get started?

168
00:08:58,180 --> 00:09:01,320
We don't need the exact commands. But what's the entry point like? Well, the great

169
00:09:01,320 --> 00:09:05,460
thing is that it integrates pretty smoothly into modern operating systems

170
00:09:05,460 --> 00:09:08,140
It's available for many Linux distributions

171
00:09:08,140 --> 00:09:12,970
Debian Ubuntu Arch Linux and you can usually just install it with your system's

172
00:09:12,970 --> 00:09:15,100
package manager and it runs as a local node

173
00:09:15,100 --> 00:09:19,250
On your machine. That's right. And once that node is running you interact with the

174
00:09:19,250 --> 00:09:21,020
network through some simple tools

175
00:09:21,020 --> 00:09:24,660
So a command line interface there's one for people who prefer the console

176
00:09:24,660 --> 00:09:28,880
Yeah, simple commands to search the network download files or publish your own

177
00:09:28,880 --> 00:09:29,400
content

178
00:09:29,400 --> 00:09:32,020
But for users who want a graphical interface

179
00:09:32,020 --> 00:09:36,430
There are separate GUI tools for setting it up and for using things like the file

180
00:09:36,430 --> 00:09:37,380
sharing system

181
00:09:37,380 --> 00:09:41,600
Those are options. The focus is on functionality and user agency. This has been a

182
00:09:41,600 --> 00:09:43,220
really fascinating deep dive

183
00:09:43,220 --> 00:09:49,800
Geonet is I mean, it's a crucial attempt to build a robust private foundation for

184
00:09:49,800 --> 00:09:51,860
the next generation of the Internet

185
00:09:51,860 --> 00:09:55,380
It's directly addressing these deep security flaws this

186
00:09:55,380 --> 00:09:59,660
Rampant exposure of metadata. It's an enormous undertaking one

187
00:09:59,660 --> 00:10:02,940
That's required nearly 20 years of sustained work and that kind of raises an

188
00:10:02,940 --> 00:10:05,320
important final question for you the listener

189
00:10:05,320 --> 00:10:09,530
What challenge does the inherent complexity of a truly privacy preserving

190
00:10:09,530 --> 00:10:11,600
architecture like G unit present?

191
00:10:11,600 --> 00:10:16,490
When you compare it to the simple convenience of the centralized broken systems we

192
00:10:16,490 --> 00:10:18,260
use every day. That's the real question

193
00:10:18,500 --> 00:10:23,380
What trade-offs will people accept and what will it ultimately take for essential

194
00:10:23,380 --> 00:10:25,540
projects like this to move from the research frontier?

195
00:10:25,540 --> 00:10:31,300
To you know widespread mainstream adoption something to think about definitely

196
00:10:31,300 --> 00:10:34,660
Thank you for joining us for this deep dive and once again a huge

197
00:10:34,660 --> 00:10:38,780
Thank you to our supporter safe server for helping us bring you these insights. We

198
00:10:38,780 --> 00:10:40,060
will catch you next time