1
00:00:00,000 --> 00:00:14,990
*34c3 intro*

2
00:00:14,990 --> 00:00:22,770
Herald: And now please join me in welcoming
Caleb, for his talk BGP and the Rule of Custom

3
00:00:22,770 --> 00:00:26,120
*Applause*

4
00:00:26,120 --> 00:00:30,050
Caleb James DeLisle: Thank you.

5
00:00:30,050 --> 00:00:32,790
Thank you and thanks for

6
00:00:32,790 --> 00:00:38,870
coming. Tonight I'm going to speak to you
about the BGP protocol but it's not going

7
00:00:38,870 --> 00:00:44,469
to be that technical. Specifically I'm
going to concentrate on the way that BGP

8
00:00:44,469 --> 00:00:53,969
molds human interactions. So there I'm
aiming at making this a reasonably

9
00:00:53,969 --> 00:00:58,530
accessible talk so if you catch me using
slightly incorrect terms for something

10
00:00:58,530 --> 00:01:04,159
consider that I'm optimizing not only for
the clarity of the correctness but also

11
00:01:04,159 --> 00:01:10,860
for the widest possible audience. So
protocols that are at their root systems

12
00:01:10,860 --> 00:01:16,750
of governance. So I'm gonna propose a
framework to think about governance and

13
00:01:16,750 --> 00:01:22,820
then as I describe BGP we can we will be
able to analyze it through the lens of

14
00:01:22,820 --> 00:01:31,240
that framework and so as my slide shows
here I find it useful to dichotomize

15
00:01:31,240 --> 00:01:36,750
governance systems into either
institutional like democracy or network

16
00:01:36,750 --> 00:01:42,890
like the family or like the CCC, and when
you think about institutional governance

17
00:01:42,890 --> 00:01:46,930
think about democracy, think about law,
think about rights, equality, these are

18
00:01:46,930 --> 00:01:52,579
all concepts which are protected for us by
an institution in this case it's the state

19
00:01:52,579 --> 00:01:55,690
and when you think about networks think
more about the family, think about

20
00:01:55,690 --> 00:02:00,479
reputation, honor, mutual respect. These
are representations of the network

21
00:02:00,479 --> 00:02:04,219
governance model, and the network
governance model turns out to be very

22
00:02:04,219 --> 00:02:09,419
important to BGP. Now I'm gonna argue that
we need both institutional and networks

23
00:02:09,419 --> 00:02:19,080
governance and each one should be used for
what it's best for. So but first I'm going

24
00:02:19,080 --> 00:02:25,209
to talk about where BGP came from. The
year was 1989 and there were two people

25
00:02:25,209 --> 00:02:30,930
named Kirk Lougheed and Yaakov Rekhter and
they were having lunch and at the time the

26
00:02:30,930 --> 00:02:37,390
Internet of the day, then known as NSFNET
was facing impending collapse let's say.

27
00:02:37,390 --> 00:02:42,620
The NSFNET was at the time, experiencing
explosive growth and the EGP

28
00:02:42,620 --> 00:02:47,680
routing protocol was reaching the point
where it just could no longer work. So

29
00:02:47,680 --> 00:02:52,290
during that lunch they defined a new
protocol, which they jokingly referred to

30
00:02:52,290 --> 00:02:55,510
as a two napkin protocol, because they had
drawn their

31
00:02:55,510 --> 00:03:02,479
diagrams out on those napkins. And we have
here some photographs of those napkins.

32
00:03:02,479 --> 00:03:08,600
Something to understand about BGP at the
time, and it was that at this time the so-

33
00:03:08,600 --> 00:03:13,750
called Internet was considered kind of
this weird anarchist experiment. You see

34
00:03:13,750 --> 00:03:20,269
real networks used grown-up protocols such
as X.25 which among other things made sure

35
00:03:20,269 --> 00:03:25,189
at the protocol level that when you sent a
piece of data it would actually reach its

36
00:03:25,189 --> 00:03:31,750
destination. By contrast, Internet Protocol
was what we call a best-effort protocol

37
00:03:31,750 --> 00:03:35,930
meaning that sometimes a packet might
arrive mangled or maybe it wouldn't arrive

38
00:03:35,930 --> 00:03:42,470
at all. Computers on the Internet usually
just use the TCP protocol to put the

39
00:03:42,470 --> 00:03:47,790
pieces back together and to resend the
broken ones but TCP is not really part of

40
00:03:47,790 --> 00:03:51,720
the Internet itself. It's more an
application that runs over it and it's

41
00:03:51,720 --> 00:03:57,439
telling that in this era the Internet was
referred to as TCP / IP because at the

42
00:03:57,439 --> 00:04:04,610
time it was expected that any network must
be providing reliable transport. Now the

43
00:04:04,610 --> 00:04:09,409
effect of IP's simplicity as compared to
other networking protocols cannot be

44
00:04:09,409 --> 00:04:17,108
overstated and to consider a different
protocol even the one used in telephone

45
00:04:17,108 --> 00:04:22,380
networks even to this day it's so mind-
numbingly complex that it actually

46
00:04:22,380 --> 00:04:27,280
encourages the telephone companies to
monopolize in order to manage the protocol

47
00:04:27,280 --> 00:04:32,750
that they use. In fact the SONET protocol
that's used in the American telephone

48
00:04:32,750 --> 00:04:36,590
companies - there's an equivalent one here
in Europe - it actually requires

49
00:04:36,590 --> 00:04:41,190
synchronized atomic clocks in the routers
in order to schedule the messages so that

50
00:04:41,190 --> 00:04:45,580
there's an empty moment in each line
that's just for the message to pass

51
00:04:45,580 --> 00:04:52,190
through. And in contrast the simplicity of
IP and the political design of BGP have

52
00:04:52,190 --> 00:05:03,150
allowed for just about anybody to become a
network operator. So what is BGP? BGP is

53
00:05:03,150 --> 00:05:09,090
a protocol which every Internet router
uses to talk to other routers when they're

54
00:05:09,090 --> 00:05:15,210
talking across an organizational boundary.
See inside an organization you control all

55
00:05:15,210 --> 00:05:18,860
the computers so you can have them talk
any way they want

56
00:05:18,860 --> 00:05:25,110
but when you talk across boundaries
the lingua franca of

57
00:05:25,110 --> 00:05:33,100
routing protocols is BGP. And BGP involves
two types of identifiers which

58
00:05:33,100 --> 00:05:39,000
organizations need to apply for. These are
IP addresses which most people know about

59
00:05:39,000 --> 00:05:44,050
and there are the lesser-known autonomous
system numbers and each network provider

60
00:05:44,050 --> 00:05:49,780
has an AS number. Both IP addresses and
AS numbers are issued by ICANN through its

61
00:05:49,780 --> 00:05:56,320
regional registries which manage this an
issuance kind of in accordance to need.

62
00:05:56,320 --> 00:06:02,130
And an autonomous system is a network
operator such as an ISP or hosting

63
00:06:02,130 --> 00:06:07,919
provider and having the AS number
means that they're able to speak as equals

64
00:06:07,919 --> 00:06:13,819
with all the other network operators. It
also means that when they interconnect

65
00:06:13,819 --> 00:06:18,080
with another network operator the fact
that they're interconnected is public and

66
00:06:18,080 --> 00:06:26,029
that's a very important little piece of
BGP. Now BGP is unlike the EGP routing

67
00:06:26,029 --> 00:06:31,190
protocol before it, a mesh protocol and
this fact has proven highly significant

68
00:06:31,190 --> 00:06:36,840
because it's created what I call the
imperative to peer. And to understand the

69
00:06:36,840 --> 00:06:42,030
imperative to peer, I'll give you a scenario.
Let's say that you and I are each

70
00:06:42,030 --> 00:06:47,489
a medium-sized network operator. So your
customers want to talk to my customers, my

71
00:06:47,489 --> 00:06:52,259
customers want to talk to your customers
but as medium-sized network operators we

72
00:06:52,259 --> 00:06:56,270
both need to buy Internet from somebody
else that's bigger than us. In the

73
00:06:56,270 --> 00:07:01,650
industry we call this the upstream
provider and if my customers are asking

74
00:07:01,650 --> 00:07:06,509
for web sites hosted by your customers
normally what's going to happen is I'm

75
00:07:06,509 --> 00:07:10,139
going to get those requests and I'm gonna
have to send them to my upstream provider,

76
00:07:10,139 --> 00:07:12,590
they're going to give them to your
upstream provider who is going to give

77
00:07:12,590 --> 00:07:18,379
them to you. But since we're both paying
for those links to our upstream providers

78
00:07:18,379 --> 00:07:26,060
there's, we're sort of paying for that
traffic in a way, and now if you and I

79
00:07:26,060 --> 00:07:30,419
happen to have routers in the same
datacenter then we could just run a wire

80
00:07:30,419 --> 00:07:35,949
across the room and then peer with one
another and BGP will bypass our upstream

81
00:07:35,949 --> 00:07:38,330
providers and route the traffic between me
and you and

82
00:07:38,330 --> 00:07:42,580
you and me. And it's much more
efficient. Now it's important to

83
00:07:42,580 --> 00:07:48,370
understand that peering means you and your
customers will talk directly to me and my

84
00:07:48,370 --> 00:07:54,520
customers. What it doesn't mean is that you
can use me to reach my other peers, or

85
00:07:54,520 --> 00:07:59,289
my peers peers, or my upstream, or someone
else. To have to do that, you'd have to be

86
00:07:59,289 --> 00:08:05,080
my customer because peering has this
limitation it's usually mutually

87
00:08:05,080 --> 00:08:10,830
beneficial for two ISPs to just peer and
usually it happens with no money changing

88
00:08:10,830 --> 00:08:16,550
hands and in fact it almost always happens
there's not even any paper contract

89
00:08:16,550 --> 00:08:23,430
that they just say that makes sense let's
run a wire. Now in this scenario you and I

90
00:08:23,430 --> 00:08:28,199
needed to have the upstream provider and
you might wonder well who doesn't need an

91
00:08:28,199 --> 00:08:33,260
upstream provider there must be someone at
the top where is the core. Well the core

92
00:08:33,260 --> 00:08:46,710
is a series of what we call Tier 1
providers. And in this picture we have

93
00:08:46,710 --> 00:08:51,890
white lines which represent the peering
agreements and the red lines represent the

94
00:08:51,890 --> 00:08:57,170
customer agreements. The Tier 1's are the
people up at the top and they can reach

95
00:08:57,170 --> 00:09:03,410
anywhere on the Internet using a peering
agreement or a customer. They don't need

96
00:09:03,410 --> 00:09:10,230
to buy Internet from anyone. However they
do need to pay the upkeep on their massive

97
00:09:10,230 --> 00:09:14,510
fiber-optic networks which give them the
global reach to get these peering

98
00:09:14,510 --> 00:09:17,890
agreements and to get these customers
which makes them a Tier 1 in the first

99
00:09:17,890 --> 00:09:23,640
place. And you can also see some
interesting things in the case you can

100
00:09:23,640 --> 00:09:29,509
have you can be a customer and also a peer
and you can also have multiple upstream

101
00:09:29,509 --> 00:09:38,060
providers. But with Tier 1 there's a bit
of politics, see Tier 1's are hesitant to

102
00:09:38,060 --> 00:09:43,000
peer with smaller operators if there is a
chance that if they refuse to peer the

103
00:09:43,000 --> 00:09:48,210
smaller one might alternatively become a
customer and so it's a bit like marriages

104
00:09:48,210 --> 00:09:53,959
between wealthy families because peering
between Tier 1's it's a complex process and

105
00:09:53,959 --> 00:09:59,480
they're each striving to ensure reciprocity
of value. Now small networks

106
00:09:59,480 --> 00:10:03,030
on the other hand they're not so concerned
about this and so they're ready

107
00:10:03,030 --> 00:10:07,721
to peer with each other quite liberally
and this has created a situation known in

108
00:10:07,721 --> 00:10:11,940
the industry as doughnut peering where in
the Tier 1's are actually increasingly

109
00:10:11,940 --> 00:10:18,199
being routed around and it's worthy of
reflection the fact that while SONET with

110
00:10:18,199 --> 00:10:24,460
a synchronized atomic clocks has made an
incentive to monopoly. BGP with it's

111
00:10:24,460 --> 00:10:31,520
imperative to peer has created a situation
where monopoly is discouraged. However

112
00:10:31,520 --> 00:10:37,730
this system also has means of preventing
bad behavior on the Internet which it

113
00:10:37,730 --> 00:10:42,900
works astonishingly well while at the same
time preserving almost absolute free

114
00:10:42,900 --> 00:10:51,760
expression. So this is a chart of the
percentage of all email that is spam and

115
00:10:51,760 --> 00:10:57,480
to understand why this chart is remarkable
consider what a bad protocol email

116
00:10:57,480 --> 00:11:03,660
is. Email is basically a push protocol
with an unlimited free speech. So all you

117
00:11:03,660 --> 00:11:07,980
got to do is get on the Internet, fire up a
mail server and you can just send spam to

118
00:11:07,980 --> 00:11:20,750
anyone. It's magic. So really this
percentage should be like over 99% but

119
00:11:20,750 --> 00:11:29,059
it's not. Well email is an old and heavily
used protocol and the network operators

120
00:11:29,059 --> 00:11:33,959
have made it kind of a special case. Email
is actually one of the few protocols where

121
00:11:33,959 --> 00:11:38,640
messing with it can get you kicked off the
whole Internet even if you have your own

122
00:11:38,640 --> 00:11:43,189
network AS number even if you are a
network operator with peers and everything

123
00:11:43,189 --> 00:11:51,799
you can still get chucked off the Internet
if you mess around with mail. So I'm gonna

124
00:11:51,799 --> 00:11:56,809
try to explain how this works and I'm
going to do it by trying to think of

125
00:11:56,809 --> 00:12:07,809
different types of actors that are, that
exist in the Internet service sphere. So

126
00:12:07,809 --> 00:12:14,280
there is a lot of actors of course but
I've made these four main categories which

127
00:12:14,280 --> 00:12:19,230
I think helps to explain the situation.
And they're the customer, the provider,

128
00:12:19,230 --> 00:12:25,130
the network operator and the civil society
organization. So the customer is someone

129
00:12:25,130 --> 00:12:30,290
like me. I take an IP address on loan from
my cable provider at home and I take one

130
00:12:30,290 --> 00:12:36,371
from my web hosting provider. It's not my
IP address and it's not assigned to me. It

131
00:12:36,371 --> 00:12:39,070
doesn't have my name on it and they can
take

132
00:12:39,070 --> 00:12:45,490
it back when I stop working with them.
But that means I have relative anonymity

133
00:12:45,490 --> 00:12:49,939
because I'm not out there looking for
peers I have my provider they know who I

134
00:12:49,939 --> 00:12:55,780
am, they know me, I know them, nobody else
needs to know who that IP address is

135
00:12:55,780 --> 00:13:01,040
associated with. However my provider can
trivially turn off my access to the

136
00:13:01,040 --> 00:13:05,950
Internet but by the same token I can
usually choose which provider to patronize

137
00:13:05,950 --> 00:13:11,980
as well modulo the monopolies in the cable
companies. Providers on the other hand

138
00:13:11,980 --> 00:13:15,450
they are not anonymous they have to
maintain relationships with network

139
00:13:15,450 --> 00:13:21,510
operators, they have to seek customers.
They're out in the public. Providers also

140
00:13:21,510 --> 00:13:27,079
have an incentive to keep customers so
they can't be obviously scammy and they

141
00:13:27,079 --> 00:13:31,750
shouldn't, they can't provide bad service.
They probably shouldn't be scummy

142
00:13:31,750 --> 00:13:36,010
and they can't provide bad service. They
also have an imperative to stay friendly

143
00:13:36,010 --> 00:13:40,540
with at least one network operator. If all
the network operators just hate them then

144
00:13:40,540 --> 00:13:46,830
they can't find Internet. Providers have
their own IP addresses they're assigned to

145
00:13:46,830 --> 00:13:51,550
them from the Internet registries and so
they can connect with multiple network

146
00:13:51,550 --> 00:13:55,930
operators including even having peers
although in practice they often keep their

147
00:13:55,930 --> 00:14:03,030
networks fairly simple. But because of the
way BGP exchanges information, their peers

148
00:14:03,030 --> 00:14:08,240
and their providers are publicly known,
whereas me the little customer borrowing

149
00:14:08,240 --> 00:14:14,540
an IP address I'm fairly opaque. Now
network operators they're somewhat like

150
00:14:14,540 --> 00:14:19,920
providers, they provide a service but in
general they make their business around

151
00:14:19,920 --> 00:14:23,800
providing raw Internet access to smaller
providers and that's why I've

152
00:14:23,800 --> 00:14:29,160
differentiated them. Network operators are
very much not anonymous they have to have

153
00:14:29,160 --> 00:14:34,799
large numbers of peering agreements and
customers which again all of these

154
00:14:34,799 --> 00:14:40,330
connections are transparent, because of the
way BGP works and they're strongly

155
00:14:40,330 --> 00:14:46,270
pressured by the imperative to peer. If
they're not able to find peers then all

156
00:14:46,270 --> 00:14:49,219
the network traffic will have to be paid
for. It can even squeeze them out of the

157
00:14:49,219 --> 00:14:53,140
market. They do have som
powers though they can

158
00:14:53,140 --> 00:14:57,169
refuse to peer with a network operator and
they can even disconnect a customer which

159
00:14:57,169 --> 00:15:01,929
they might do to maintain their
reputation. Last group here is civil

160
00:15:01,929 --> 00:15:07,059
society these are organizations like Team
Cymru and Spamhaus. They dedicate their

161
00:15:07,059 --> 00:15:10,829
time to shedding light on the bad guys of
the Internet. They're not anonymous as

162
00:15:10,829 --> 00:15:15,580
organisations although their members can
be hidden and they also maintain lists of

163
00:15:15,580 --> 00:15:20,059
IP addresses and AS numbers which are
either known to be operated by spam

164
00:15:20,059 --> 00:15:27,020
organizations or which are just unused and
shouldn't be existing. Internet civil

165
00:15:27,020 --> 00:15:31,280
society doesn't have any direct power but
their power comes from their reputation

166
00:15:31,280 --> 00:15:36,540
for providing valid and useful data.
They're also able to do their job because

167
00:15:36,540 --> 00:15:42,540
of the transparency of providers and
network operators which is built into BGP.

168
00:15:42,540 --> 00:15:46,380
I'm going to tell you a story about two
providers. One is called McColo and the

169
00:15:46,380 --> 00:15:51,120
other is called PRQ, both of these
organizations were founded in 2004 and

170
00:15:51,120 --> 00:15:55,410
both of them have been subject to certain
controversy but tellingly one of them

171
00:15:55,410 --> 00:16:03,280
still remains with us and the other one is
long gone. So this is a splash page of

172
00:16:03,280 --> 00:16:08,309
McColo back in 2008 before it went dark.
McColo was founded by a 19 year old

173
00:16:08,309 --> 00:16:13,380
student named Nicola McColo and it thrived
for four years before being taken down.

174
00:16:13,380 --> 00:16:16,890
McColo provided what's known as
bulletproof hosting. That means hosting

175
00:16:16,890 --> 00:16:21,650
where the provider will keep your server
online no matter what you do with it.

176
00:16:21,650 --> 00:16:25,949
Bulletproof hosting providers choose not
to cooperate with civil society or even

177
00:16:25,949 --> 00:16:32,240
law enforcement unless they're forced to.
And in November of 2008 the Washington

178
00:16:32,240 --> 00:16:37,410
Post gathered some damning evidence that
McColo was a hosting provider mostly

179
00:16:37,410 --> 00:16:42,829
interested in providing service for
spammers. What's interesting is that

180
00:16:42,829 --> 00:16:46,779
rather than send this evidence to the
police they sent it to McColo's network

181
00:16:46,779 --> 00:16:51,300
operators. Like many providers McColo
bought Internet access from two major

182
00:16:51,300 --> 00:16:54,449
network operators. In this case it was a
Hurricane Electric and Global

183
00:16:54,449 --> 00:16:58,630
Crossing. When Hurricane and Global
Crossing were given this information from

184
00:16:58,630 --> 00:17:03,050
The Washington Post they voluntarily chose
to abruptly ceased doing business with

185
00:17:03,050 --> 00:17:08,190
McColo and the provider was caught off
guard and it and all of its customers went

186
00:17:08,190 --> 00:17:18,679
offline. Here is the global volume of spam
which dropped to that day by as much as

187
00:17:18,679 --> 00:17:24,040
75%. Needless to say no other network
provider was beginning or ready to begin

188
00:17:24,040 --> 00:17:32,640
selling service to McColo and their
business crumbled. Now PRQ is in some ways

189
00:17:32,640 --> 00:17:40,110
similar to McColo *heeey* and in some ways
they're quite different. PRQ was founded

190
00:17:40,110 --> 00:17:44,780
by two Swedish guys known in BitTorrent
circles as anakata and Tiamo and they

191
00:17:44,780 --> 00:17:49,280
provide what I call last resort hosting.
They've hosted highly controversial

192
00:17:49,280 --> 00:17:53,850
websites such as WikiLeaks but they're
probably best known for hosting The Pirate

193
00:17:53,850 --> 00:17:58,750
Bay. The Pirate Bay stands out is probably
one of the most famous websites to

194
00:17:58,750 --> 00:18:06,180
publicly flaunt copyright. Going to the
extent of actually publishing abuse

195
00:18:06,180 --> 00:18:17,780
complaints along with their sarcastic and
humiliating responses. This is something

196
00:18:17,780 --> 00:18:22,140
that the copyright industry had never seen
before. High-power lawyers are just not

197
00:18:22,140 --> 00:18:28,530
accustomed to getting replies signed go
fuck yourself. So all four of the founders

198
00:18:28,530 --> 00:18:31,910
have spent some time in prison and the
site's data centers have been raided

199
00:18:31,910 --> 00:18:36,520
multiple times and supposedly Hollywood
even use threats of trade sanctions

200
00:18:36,520 --> 00:18:43,030
against Sweden to force them to shut this
thing down. But we find that as McColo has

201
00:18:43,030 --> 00:18:48,440
drifted into historical obscurity, The
Pirate Bay is still alive and has even

202
00:18:48,440 --> 00:18:54,440
become something of a cultural
institution. And we also find that unlike

203
00:18:54,440 --> 00:18:59,340
McColo, PRQ and The Pirate Bay have never
had any problem with their network

204
00:18:59,340 --> 00:19:05,290
operators. So one of the fundamental
tenets of the Internet is that network

205
00:19:05,290 --> 00:19:10,670
operators are morally but not legally
responsible for the activities of their

206
00:19:10,670 --> 00:19:15,560
customers and peers. So they may choose
who they do business with and they will

207
00:19:15,560 --> 00:19:24,560
not be held to legal account for these
decisions. What are the lessons that we

208
00:19:24,560 --> 00:19:28,220
can take away from this? 
Recently there's been a lot of

209
00:19:28,220 --> 00:19:33,470
work done on federated social networking
protocols. I'm sure some of you will take

210
00:19:33,470 --> 00:19:38,860
part in this development but the vast
majority of you will be evaluating them to

211
00:19:38,860 --> 00:19:43,900
make decisions about which technology to
adopt and I urge you to give some thought

212
00:19:43,900 --> 00:19:48,880
toward the political identities of the
protocols which you make or which you

213
00:19:48,880 --> 00:19:55,850
choose to make your own. Like many things
systems of communication are defined

214
00:19:55,850 --> 00:20:02,310
largely by what they reject, whether that
be packets larger than 1500 bytes or Nazi

215
00:20:02,310 --> 00:20:08,610
propaganda. And in a protocol I identify
three main ways that these rules can be

216
00:20:08,610 --> 00:20:13,981
defined. The first is what we hard code
into the software source code. Some

217
00:20:13,981 --> 00:20:18,260
examples are message formats and
permission systems but hardcoded rules

218
00:20:18,260 --> 00:20:23,950
can be extended further with cryptography
and especially with block chains. Hard

219
00:20:23,950 --> 00:20:29,060
coded rules are a perfect example of
institutional governance they can be very

220
00:20:29,060 --> 00:20:34,330
fair. They are very fair because code
applies the rules equally to everyone and

221
00:20:34,330 --> 00:20:38,820
for things which we consider a basic human
right such as private communications and

222
00:20:38,820 --> 00:20:43,150
things which are easily quantifiable in
software, hardcoding can be the best

223
00:20:43,150 --> 00:20:49,890
solution. However hard rules do have a
downside, Ethereum a cryptocurrency based

224
00:20:49,890 --> 00:20:54,120
heavily on the libertarian philosophy of
freedom of contract found itself in a bit

225
00:20:54,120 --> 00:20:59,050
of a quandary when a bug was discovered in
one of the very significant contract

226
00:20:59,050 --> 00:21:03,330
allowing for all the money to be stolen
out of that contract. For those who don't

227
00:21:03,330 --> 00:21:07,950
follow the topic Ethereum was hard forked
in order to stop the errant contract and

228
00:21:07,950 --> 00:21:14,380
everybody had to update and while the fork
itself protected the participants in the

229
00:21:14,380 --> 00:21:21,230
contract it struck a serious blow to the
fundamental philosophy of Ethereum and it

230
00:21:21,230 --> 00:21:25,620
serves as a warning that we have a
downside to hardcoding. The second source

231
00:21:25,620 --> 00:21:30,231
is of course the central authority. The
Internet uses ICANN as a central authority

232
00:21:30,231 --> 00:21:35,650
to manage allocation of domain names, IP
addresses and autonomous system numbers.

233
00:21:35,650 --> 00:21:38,530
The central authority
is kind of a poster child of institutional

234
00:21:38,530 --> 00:21:43,240
governance. Like hardcoded rules, rule by
central authority also tends to be

235
00:21:43,240 --> 00:21:49,750
egalitarian. Moreover a central authority
is actually capable of equity because it

236
00:21:49,750 --> 00:21:54,061
can comprehend people's different
situations and adapt to them. Something

237
00:21:54,061 --> 00:22:00,220
that a hardcoded rule cannot do but a
central authority like a hardcoded rule

238
00:22:00,220 --> 00:22:06,010
is prone to coldness and bureaucracy and
moreover it's quite difficult to create

239
00:22:06,010 --> 00:22:11,170
central authorities which do not give
certain individuals unaccountable power

240
00:22:11,170 --> 00:22:16,730
over others. The third source of rules or
in this case customs where this talk gets

241
00:22:16,730 --> 00:22:22,800
its name is from the network and this is
how the network operators and how BGP

242
00:22:22,800 --> 00:22:26,890
mostly managed to keep email spam and
other bad things off the Internet. The

243
00:22:26,890 --> 00:22:31,580
network is by far the most humane form of
governance. Social norms are passed along

244
00:22:31,580 --> 00:22:36,340
from friend to friend rather than rules
being forced down upon people by central

245
00:22:36,340 --> 00:22:41,350
authority or source code. And we see
network-like systems in families ancient

246
00:22:41,350 --> 00:22:46,670
tribal societies as well as royalty and
elite in kingdoms and of course online.

247
00:22:46,670 --> 00:22:53,990
However in electronic networks we tend to
associate it with, we get to choose who

248
00:22:53,990 --> 00:23:00,060
we associate with from any connected
person in the world and but networks also

249
00:23:00,060 --> 00:23:04,830
have a downside they're not egalitarian in
any way. Those central in the network are

250
00:23:04,830 --> 00:23:10,140
simply more powerful than those on the
edges. In the feudal system law was often

251
00:23:10,140 --> 00:23:14,410
applied differently to a person based on
how they were dressed or what family they

252
00:23:14,410 --> 00:23:20,610
were from. And the origin of the term rule
of law it was initially described as a

253
00:23:20,610 --> 00:23:27,340
better alternative to rule of the king or
rule of man. We have the law we

254
00:23:27,340 --> 00:23:33,380
have the state because people demanded
them as in this case the inequity of the

255
00:23:33,380 --> 00:23:38,260
network rule proved unacceptable. But
there's an important difference between

256
00:23:38,260 --> 00:23:45,580
BGP's rule of custom and the patently
unjust feudal system. BGP is transparent.

257
00:23:45,580 --> 00:23:49,630
We know which network operators are
interconnected with whom and we know who

258
00:23:49,630 --> 00:23:55,080
is protecting the bad actors
and in every case we find dystopia

259
00:23:55,080 --> 00:23:59,550
whenever there's power without
transparency either in opaque proprietary

260
00:23:59,550 --> 00:24:05,240
code, unaccountable central authorities or
in networks which form mafias, secret

261
00:24:05,240 --> 00:24:11,350
societies in the feudal system. And I
think the key message from BGP has to be

262
00:24:11,350 --> 00:24:17,930
that whether in central authorities or in
networks. Power and privacy do not mix.

263
00:24:17,930 --> 00:24:23,820
With any measure of power there must be
equal transparency and accountability.

264
00:24:23,820 --> 00:24:29,960
*Applause*

265
00:24:29,960 --> 00:24:36,150
So in closing. I hope you go out there and
make protocols and make systems which far

266
00:24:36,150 --> 00:24:43,860
surpass what we have today and to do that
I suggest looking at the past and what

267
00:24:43,860 --> 00:24:49,270
protocols have been highly successful and
try to identify why they worked. Also look

268
00:24:49,270 --> 00:24:56,620
at the ones that didn't work. Look at the
things that died in a hell storm of spam

269
00:24:56,620 --> 00:25:01,640
and try to keep it simple. I developed
some software before and something I

270
00:25:01,640 --> 00:25:06,750
learned the hard way is that complexity
becomes the enemy of adoption so maybe

271
00:25:06,750 --> 00:25:11,409
it's a good idea we all write our
protocols on napkins. Thank you.

272
00:25:11,409 --> 00:25:21,920
*Applause*

273
00:25:21,920 --> 00:25:24,260
Herald: Thank you Caleb. 
We now have five minutes

274
00:25:24,260 --> 00:25:28,330
for questions you know the drill. Please
line up at the microphones there's four

275
00:25:28,330 --> 00:25:32,430
microphones two in the middle, one there,
one there. please don't leave the room

276
00:25:32,430 --> 00:25:36,220
until the talk is over. If you stood up
you can leave but the rest please

277
00:25:36,220 --> 00:25:39,360
stay seated you can spare five minutes
it's a lot of noise that people keep

278
00:25:39,360 --> 00:25:43,490
leaving during the Q&A. 
Microphone number one please.

279
00:25:43,490 --> 00:25:49,940
Question: Hello I'm from Sweden. I
really like to talk to have a overview of

280
00:25:49,940 --> 00:25:56,570
this I think it was mostly correct. I
would say that I'm working with both the

281
00:25:56,570 --> 00:26:05,870
BGP and DNS and those association with
ICANN but I think you put them all on the

282
00:26:05,870 --> 00:26:10,260
you give them a bit too much power they
are not that powerful unless you describe

283
00:26:10,260 --> 00:26:16,180
them. They don't decide everything about
DNS and everything. They're running the

284
00:26:16,180 --> 00:26:21,080
IANA contract for these domain names and
numbers and stuff like that. On behalf of

285
00:26:21,080 --> 00:26:27,240
the community like the IETF and other
multi-stakeholder organizations. So they

286
00:26:27,240 --> 00:26:33,040
only have the power over the new top-level
domains they don't have the power over the

287
00:26:33,040 --> 00:26:41,100
common domain or .se domain in Sweden or
.ch in Switzerland so if .se or .ch would

288
00:26:41,100 --> 00:26:46,100
like to have The Pirate Bay, ICANN has
nothing to do with that.

289
00:26:46,100 --> 00:26:47,730
Herald: Could you get to the question please?

290
00:26:47,730 --> 00:26:50,590
Question: No it's not a question. 
More to show that ICANN is

291
00:26:50,590 --> 00:26:55,680
not the king.
Herald: There are other people queueing please.

292
00:26:55,680 --> 00:27:00,200
Herald: Microphone number two please.
Question 2: Thank you. Great great

293
00:27:00,200 --> 00:27:09,380
lecture. So what actually is the reason
why some illegal activities are tolerated

294
00:27:09,380 --> 00:27:16,841
by ISPs and some like child abuse is not?
Speaker: That's a great question. I

295
00:27:16,841 --> 00:27:27,050
think that what it comes down to is what
is socially acceptable. So when you have

296
00:27:27,050 --> 00:27:31,400
why does some activity, why does some
illegal activity like smoking a joint

297
00:27:31,400 --> 00:27:38,080
outside, not cause you to have the police and
other activity like murdering somebody

298
00:27:38,080 --> 00:27:42,540
would. It's all about what is socially
acceptable to the people around you and

299
00:27:42,540 --> 00:27:46,990
what is socially acceptable to the people
around them and in the case of ISPs

300
00:27:46,990 --> 00:27:50,600
somebody is peering with the the people
that serve The Pirate Bay somebody else

301
00:27:50,600 --> 00:27:55,040
looks at that and says it's not, I don't
feel that that's a problem somebody else

302
00:27:55,040 --> 00:27:58,920
looks at them and says I don't feel but
people who think that is a problem and so

303
00:27:58,920 --> 00:28:02,120
it's okay. I hope this answers your
question.

304
00:28:02,120 --> 00:28:07,180
Herald: Do we have an Internet question?
No. Okay microphone number one please.

305
00:28:07,180 --> 00:28:13,590
Question 3: Why do you think network
operators being morally but not literally

306
00:28:13,590 --> 00:28:18,500
responsible for what content they accept
on the network worked so well for McColo

307
00:28:18,500 --> 00:28:22,690
and PRQ and doesn't work at all for social
media?

308
00:28:22,690 --> 00:28:27,370
Speaker: Oh great great question. Why it
doesn't work. If I understand you

309
00:28:27,370 --> 00:28:35,340
properly? Why Twitter is still a crap
hole. It is. Well here's the thing Twitter

310
00:28:35,340 --> 00:28:42,130
has it's - I am probably getting sued
- it's a, it's basically an

311
00:28:42,130 --> 00:28:46,230
institutional governance system. They said
like we're in charge everybody is flat on

312
00:28:46,230 --> 00:28:53,050
here so on top of a network governance
system they built in a institutional

313
00:28:53,050 --> 00:28:57,210
governance system and the institutional
governance system it's like a high school

314
00:28:57,210 --> 00:29:04,940
it's just it's terrible. So I should use
this opportunity to plug Mastodon because

315
00:29:04,940 --> 00:29:10,480
Mastodon is an example of a system which
is federated and looks well it doesn't

316
00:29:10,480 --> 00:29:15,779
look exactly like BGP but I think it will
in five years.

317
00:29:15,779 --> 00:29:22,519
*Applause*

318
00:29:22,519 --> 00:29:24,710
Herald: One hopefully brief question from
two please.

319
00:29:24,710 --> 00:29:29,330
Question 4: Yes. Hi. Thank you very much
for talking about BGP. As someone who

320
00:29:29,330 --> 00:29:33,100
knows a lot about BGP and I was kind of
wondering if you could help answer a

321
00:29:33,100 --> 00:29:38,251
question and that's that I kind of view
BGP is kind of like a static protocol and

322
00:29:38,251 --> 00:29:42,190
it was stuck it was written on two napkins
and it's been a little bit expanded beyond

323
00:29:42,190 --> 00:29:48,420
that. My concern is like with, with rogue
countries or someone else harnessing IP

324
00:29:48,420 --> 00:29:53,880
blocks and like sync holding traffic for
google.com or facebook.com which has

325
00:29:53,880 --> 00:29:57,580
happened several times in the last couple
years can you think of a way where you can

326
00:29:57,580 --> 00:30:03,929
either get BGP to conquer that problem or
with the new decentralized protocol to

327
00:30:03,929 --> 00:30:08,160
conquer that problem?
Speaker: Filters, filters, filters.

328
00:30:08,160 --> 00:30:14,110
Basically. I don't know that much about
BGP but I know that some people are in the

329
00:30:14,110 --> 00:30:18,150
business of making their filters from the
data that's in the WHOIS database that

330
00:30:18,150 --> 00:30:22,120
just say this is what you're allowed to
announce cool this is we'll build a filter

331
00:30:22,120 --> 00:30:27,480
that's what your your box can send to us.
The problem is that ISPs are lazy and we

332
00:30:27,480 --> 00:30:33,350
don't have we don't have standardized
stuff for making these filters so we end

333
00:30:33,350 --> 00:30:37,180
up with a lot of people just sit putting
no filter and saying yeah announce

334
00:30:37,180 --> 00:30:41,100
whatever you want and then you know China
announces Google and all the traffic goes

335
00:30:41,100 --> 00:30:45,030
like this.
Herald: And that's all for today.

336
00:30:45,030 --> 00:30:47,513
Thank you Caleb.

337
00:30:47,513 --> 00:30:50,064
*Applause*

338
00:30:50,064 --> 00:31:11,061
*34c3 outro*