[Music]
the politics of algorithms what does
they have to do with one another an
algorithm is a mathematical description
of a computational process for mapping
some inputs to some outputs it is very
platonic object what does it have to do
with the vagaries and discretionary
aspects of human politics well it turns
out in the concept of cryptocurrencies
and Bitcoin I’m mostly going to be
talking specifically about Bitcoin a lot
and in order to see why I want you to
think about a Bitcoin and to talk about
Bitcoin as this mass based
cryptocurrency a crypto currency backed
by mathematics and there’s algorithms
and I would like to clear the
misconception that this is only
mathematics and it is only algorithms
there is a human element
these are human systems with human
politics and they are of primary
importance in those networks but before
I get into that the Keynesian kiddie
contests so a nice words is actually
going to be relevant I’m not just trying
to show you cat pictures we’re gonna
play a little game and in that game I’m
gonna show you nine fluffy pictures and
and and you have to pick the cutest one
now this is a positive sum game so we
can all be winners tonight there’s one
issue that I don’t know which one is a
cutest one so what we’re going to do is
we’re going to pick whichever one most
people select and that’s going to be the
cutest one and so in order for you to
win tonight you have to pick the same
picture that you expect most people are
going to pick this is how you win a
commission contest so without further
ado here’s the kittens and we have a we
have a special entrance tonight this is
going to date me but number five is
Alfie the character from an NBC’s 80s
sitcom and he really likes cats so you
have nine entrants and you have to pick
a number and once again if you want to
win you have to pick the number you
expect most people are going to select
so I’m just going to give you a few
seconds to think about that
all right so just by show of hands do we
have anyone who selected kiddin number
six so cute kitten with the flowers okay
I see a couple of hands here all right
it’s good and I’ll cut to the chase
has anyone selected alpha number five
all right okay so the reason that Alf
can be a winner even though he’s not
even a kitten and even though he’s not
particularly that cute is simply that he
is a shelling point so shelling point
were introduced by economist Thomas
Schelling in the 1960s and the idea is
that if you have a game where humans
need to coordinate them something and
they have a hard time communicating or
maybe communication is expensive they
are going to find a solution and that
solution is going to be something that
seems reasonable to them or natural and
here the natural answer is a middle one
it’s it’s obvious even though he’s not
nationally cute and he’s not necessarily
a kitten so how does that relate to
Bitcoin so you know we’ve talked about
Bitcoin it’s this crypto currency one of
the most impressive aspect of Bitcoin is
it’s gross in 2010
it took 10,000 bitcoins for purchasing
two pizzas today this would be worse
around 50 million dollars so crazy price
appreciation but what I find fascinating
is a talk around Bitcoin between I would
say 2009 and 2013 14 where it was really
in the context of the US financial
crisis and everyone is contrasting it
with human politics so you have all of
these discretionary actions from the US
Federal Reserve from the US Treasury and
in contrast of that bitcoin is supposed
to be this island of pure mathematics
where you know nothing gets in the way
of the algorithm it’s completely
independent from human affairs and you
see it in the memes at the time that
kind of reflects the sight gas so here
Bitcoin your money is secured by the
laws of the universe so you know here it
makes the case that it would take the
entire energy of a star and some type of
sci-fi technology if we ever wanted to
to break Bitcoin and you know one that’s
probably not true and two it’s you
it’s really making the point that this
is really a lot of physics which are
backing this but you know what’s
stronger than the laws of physics the
laws of mathematics so you know bitcoin
is mass versus governments the currency
built with mathematics if you go and be
going on org it will tell you that
bitcoin is backed by mathematics which
normally would mean that you should be
able to exchange one Bitcoin from one
mathematics which doesn’t make a whole
lot of sense now one particular view of
Bitcoin which is rather well-known now
is that Bitcoin and I can have at most
21 million coin it’s not 22 it’s not 20
it’s it’s this number and if you look at
this number it’s clearly a human chosen
number right they are special number in
mathematics and like a hundred xx
decision they’re special numbers in
physics 21 million is human chosen and
it could have been 42 it could have been
210 and that wouldn’t make a big
difference the only difference it make
would be the same thing as if you’re
counting in dollars or cents so that
doesn’t make a big difference but
somehow the network enforces that now
it’s not the laws of mathematics which
are gonna be enforcing this human chosen
number it’s not the laws of physics the
reason that between can have 21 million
coin in not 22 is that if I create my
own version of Bitcoin and I come to you
and I say hey I have this new version of
Bitcoin it just so happens that it has
different rules well you’re not gonna
accept it you’re gonna say that’s a
Bitcoin that’s not that’s not in a Roth
set that we all agreed upon and so the
value of a Bitcoin to you is predicated
on the propensity of other people to
actually accept it and so the mechanism
that enforces this type of properties on
the network is the same mechanism that
means that everyone is going to pay a
certain picture in a contest it’s based
on a shilling point and it’s based on a
human and very political decision now it
can be stable it can work like Bitcoin
is working Beast Khan has been doing
fantastically but it’s a misconception
to assume that this is enforced by
mathematics mathematics is a backbone
but in the end it’s a human system so
the French this is Michel Foucault he’s
a French post structuralist and
post-structuralist would tell you that
everything is a social construct
mathematics is a social construct physic
is a social construct and I don’t think
they’re right or if the
are there writing some trivial sense but
for Bitcoin and for money and
cryptocurrencies in general I think they
would be right in an interesting sense
yes
Bitcoin has algorithms and yes it has
mathematics but Nene and what you really
makes the network is a social construct
around it it’s the shared vision that a
lot of people have using this network of
saying yes we are going to ascribe value
to this and yes we’re going to go buy
this set of rules and not by any other
set of rules this is what actually makes
Bitcoin so why does any of this matter
why do you know why do we care if it’s
working you know why why make this
distinction it is just pedantry on my
part well not quite
so it gets interesting when you look at
innovation so Bitcoin the beacon paper
came out of 2008 Network launched in
2009 there’s been a lot of innovation
since then because obviously academics
saw this great innovation and started
working at it and there were a lot of
proposals so here’s just three I like
one is called Bitcoin and gee it’s an
academic proposal for increasing the
throughput that’s the amount of
transactions per second for Bitcoin
about 10 X to 100x Oh near no one’s
talking about it zero cash was an
academic paper it gave rise to a
cryptocurrency called e cash now which
preserves privacy it turns out that if
you make a transaction in Bitcoin the
entire world can see it so that’s you
know that’s an interesting development
or a cerium which implemented smart
contracts which are these electronic
agreements that can be enforced in the
same way that the network enforces
transactions if you go to a Archive
which is the largest repository of
scientific paper now you’ll see this you
know great increase of papers mentioning
Bitcoin so there’s a lot of active
research why does it matter well unless
Bitcoin was perfect from inception you
might want to change the rule sets you
might want to make the rule sets better
because you have new ideas but how do
you do this let’s say that we were all
on focused on number 5 and all of the
sudden we actually you know what we
would be better off having chosen number
6 right how do you actually move that
shilling point well there’s two way we
can do it one way is I’m on stage I
could tell everyone hey everyone you
know everyone pick number six
we’re gonna win by picking number six
and because I have a microphone I have a
loud voice made me I’m able to do that
or we can try to have these informal
discussions but it’s it’s a messy
process and what typically happens when
you try to have innovation in the system
it’s something called a fork so the idea
of a fork is that you start with one
system and then someone say hey you know
what we should move from Alpha and we
should you know we should move to kiddin
number six because that’s you know
that’s clearly a better ID and we better
served and some people are gonna say yes
you know what six is a good idea six is
legitimate this is what we should all be
focusing on but other people will say no
actually I prefer the other version and
so the value that was in your network is
now going to be split into two one part
of that value is going to go to the
people who are looking at one network as
as legitimate and the other to the other
network and it’s not going to be exactly
zero one because there’s always
uncertainty things can happen so
typically when you see Forks it’s tended
to be something like 10% 90%
historically but it’s still what is
going to capture most of the value now
what do they capture the value best
based on is it’s the best Fork who is
going to win is it the fork that’s the
most obvious I would argue that at least
right now the main criterion for picking
a fork is legitimacy and as a use case
there’s been a debate reaching in a
Bitcoin community for about a year and a
half now on the size of blocks so as you
may know a Bitcoin is based on a
blockchain which is instead a structure
that produces a block every about 10
minutes and blocks have traditionally
been this 1 megabyte files that contain
transaction now there’s been one faction
in a between community who’s been saying
hey if we owned if only we had bigger
blocks then we could have more
transactions and if this would be great
because we need to have more
transactions well the other side has
been saying well you know that may be
true but if we do that our network is
going to be less decentralized some
parties are gonna have more influence
than others and when I find fascinating
is that in a lot of these debates both
sides were quoting Satoshi Nakamoto the
pseudonymous
founder of Bitcoin to each other
now Satoshi has not been seen online for
the past few years but they were saying
hey clearly in the vision of Satoshi we
need to have big blocks and the other
one who say no no clearly look at this
clearly we should have small blocks and
so they were engaging in this whole
hermeneutics for weird reasons so one
possible reason might be that it’s
simply argument from Authority they’re
saying well clearly said she knew what
they were doing and so they would have a
good opinion on the matter but there’s
also a lot of very smart people looking
at because there’s a lot of very smart
opinion so I don’t think this is what’s
happening what’s happening is a play for
legitimacy it’s two sides saying I’m the
legitimate branch and so I want to take
the economic value with me whereas the
other side said oh no no no no we’re the
legitimate branch we should have a value
now we’ve moved completely away from
algorithm we’ve moved away from
mathematics but what’s going to
determine the rules of this ledger is
the interpretation of the right of the
writing of a paper reading 9 years ago
this this is completely the realm of
politics and the human element is
nothing to do with algorithm it’s
nothing to do with math the dynamic at
play between those groups he sadly
captured in a game known as the game of
chicken so again this is gonna date me
this is James Dean rebel without a cause
and so in a game of chicken you have two
cars and they’re racing toward each
other and you win the game by not
swerving because the other side who is
slightly less crazy than you is gonna
swerve before and the problem with a
game like this is that if both sides are
you know equally and willing to swerve
the majors crash towards one another so
you have the same dynamic at playing a
fork because you may not want to fork
you don’t want to divide the the network
which you primarily want is you want to
steer it but you’re gonna try to
convince the other side that no matter
what you’re gonna try to you know you’re
gonna go your own way no matter what
you’re gonna keep going in a straight
line and both sides do that and it’s
really not an effective way of doing
governance so what I’ve been working on
is formalized governance for blockchain
and cryptocurrencies
and so just a word about governance so
one way to think about it is imagine in
your own an apartment building not only
apartment built a partner
in an apartment building so you own your
apartment everything you do in your
apartment is your own business I don’t
care the neighbors hopefully don’t care
but what happens when you need to decide
when the heat should be turned on for
the building or should the fast choose
the facade of the building be repainted
you need a way of making these decisions
which are going to engage the common
parts of the building and in order to do
that you need to have governance now we
have several options you can do nothing
you can say we don’t like governance we
don’t want to have governance we own our
apartment and that’s enough and that may
be fine but then the paint is going to
chip on your facade and you’re not gonna
have heating your building in the winter
that’s not that great you could try not
having any process and just letting
whoever yell the loudest make those
decisions but again it’s still gonna
engage your resources that’s not
necessarily what you want so what can we
do well instead of having all these
informal processes we can formalize we
can create rules for saying we have come
in parts and there are common parts in
block chains one you know the most
important of which being innovation so
we all agree that we need to innovate we
need to introduce some time changes in a
platform but if we’re gonna do that we
need to have a process because if we
don’t have a process it’s going to be
determined by completely random means
that have to do with shelling points
instead of having to do with merits so
one simple way of doing this is simply
voting you can have you know block shins
or great records keepers they’re a great
way of coordinating action and so you
can have a vote you can say to people
okay we’re gonna have a new proposal for
innovation do you actually want to do
this and it’s very important for people
to be able to express their intention
without having to try to guess what
other people are doing you want people
to tell them this is what I prefer and
not this is what I expect everyone else
is going to prefer now it’s a very it’s
a subtle distinction but it makes a
whole world of difference so if I just
want to leave one ID one important idea
tonight is that yes there are algorithms
at play there are mathematics but in the
end we’re dealing with human systems
right cryptocurrencies block chains they
are social construct the algorithm is
just a scaffolding upon which you flesh
out all of these human activities so
when designing
systems think about the human element
don’t think of them purely as technology
think of them as enablers of social
structure and only Suzette will be able
to really leverage the coordinating
abilities of blockchains and related
technologies
[Applause]
the politics of algorithms what does
they have to do with one another an
algorithm is a mathematical description
of a computational process for mapping
some inputs to some outputs it is very
platonic object what does it have to do
with the vagaries and discretionary
aspects of human politics well it turns
out in the concept of cryptocurrencies
and Bitcoin I’m mostly going to be
talking specifically about Bitcoin a lot
and in order to see why I want you to
think about a Bitcoin and to talk about
Bitcoin as this mass based
cryptocurrency a crypto currency backed
by mathematics and there’s algorithms
and I would like to clear the
misconception that this is only
mathematics and it is only algorithms
there is a human element
these are human systems with human
politics and they are of primary
importance in those networks but before
I get into that the Keynesian kiddie
contests so a nice words is actually
going to be relevant I’m not just trying
to show you cat pictures we’re gonna
play a little game and in that game I’m
gonna show you nine fluffy pictures and
and and you have to pick the cutest one
now this is a positive sum game so we
can all be winners tonight there’s one
issue that I don’t know which one is a
cutest one so what we’re going to do is
we’re going to pick whichever one most
people select and that’s going to be the
cutest one and so in order for you to
win tonight you have to pick the same
picture that you expect most people are
going to pick this is how you win a
commission contest so without further
ado here’s the kittens and we have a we
have a special entrance tonight this is
going to date me but number five is
Alfie the character from an NBC’s 80s
sitcom and he really likes cats so you
have nine entrants and you have to pick
a number and once again if you want to
win you have to pick the number you
expect most people are going to select
so I’m just going to give you a few
seconds to think about that
all right so just by show of hands do we
have anyone who selected kiddin number
six so cute kitten with the flowers okay
I see a couple of hands here all right
it’s good and I’ll cut to the chase
has anyone selected alpha number five
all right okay so the reason that Alf
can be a winner even though he’s not
even a kitten and even though he’s not
particularly that cute is simply that he
is a shelling point so shelling point
were introduced by economist Thomas
Schelling in the 1960s and the idea is
that if you have a game where humans
need to coordinate them something and
they have a hard time communicating or
maybe communication is expensive they
are going to find a solution and that
solution is going to be something that
seems reasonable to them or natural and
here the natural answer is a middle one
it’s it’s obvious even though he’s not
nationally cute and he’s not necessarily
a kitten so how does that relate to
Bitcoin so you know we’ve talked about
Bitcoin it’s this crypto currency one of
the most impressive aspect of Bitcoin is
it’s gross in 2010
it took 10,000 bitcoins for purchasing
two pizzas today this would be worse
around 50 million dollars so crazy price
appreciation but what I find fascinating
is a talk around Bitcoin between I would
say 2009 and 2013 14 where it was really
in the context of the US financial
crisis and everyone is contrasting it
with human politics so you have all of
these discretionary actions from the US
Federal Reserve from the US Treasury and
in contrast of that bitcoin is supposed
to be this island of pure mathematics
where you know nothing gets in the way
of the algorithm it’s completely
independent from human affairs and you
see it in the memes at the time that
kind of reflects the sight gas so here
Bitcoin your money is secured by the
laws of the universe so you know here it
makes the case that it would take the
entire energy of a star and some type of
sci-fi technology if we ever wanted to
to break Bitcoin and you know one that’s
probably not true and two it’s you
it’s really making the point that this
is really a lot of physics which are
backing this but you know what’s
stronger than the laws of physics the
laws of mathematics so you know bitcoin
is mass versus governments the currency
built with mathematics if you go and be
going on org it will tell you that
bitcoin is backed by mathematics which
normally would mean that you should be
able to exchange one Bitcoin from one
mathematics which doesn’t make a whole
lot of sense now one particular view of
Bitcoin which is rather well-known now
is that Bitcoin and I can have at most
21 million coin it’s not 22 it’s not 20
it’s it’s this number and if you look at
this number it’s clearly a human chosen
number right they are special number in
mathematics and like a hundred xx
decision they’re special numbers in
physics 21 million is human chosen and
it could have been 42 it could have been
210 and that wouldn’t make a big
difference the only difference it make
would be the same thing as if you’re
counting in dollars or cents so that
doesn’t make a big difference but
somehow the network enforces that now
it’s not the laws of mathematics which
are gonna be enforcing this human chosen
number it’s not the laws of physics the
reason that between can have 21 million
coin in not 22 is that if I create my
own version of Bitcoin and I come to you
and I say hey I have this new version of
Bitcoin it just so happens that it has
different rules well you’re not gonna
accept it you’re gonna say that’s a
Bitcoin that’s not that’s not in a Roth
set that we all agreed upon and so the
value of a Bitcoin to you is predicated
on the propensity of other people to
actually accept it and so the mechanism
that enforces this type of properties on
the network is the same mechanism that
means that everyone is going to pay a
certain picture in a contest it’s based
on a shilling point and it’s based on a
human and very political decision now it
can be stable it can work like Bitcoin
is working Beast Khan has been doing
fantastically but it’s a misconception
to assume that this is enforced by
mathematics mathematics is a backbone
but in the end it’s a human system so
the French this is Michel Foucault he’s
a French post structuralist and
post-structuralist would tell you that
everything is a social construct
mathematics is a social construct physic
is a social construct and I don’t think
they’re right or if the
are there writing some trivial sense but
for Bitcoin and for money and
cryptocurrencies in general I think they
would be right in an interesting sense
yes
Bitcoin has algorithms and yes it has
mathematics but Nene and what you really
makes the network is a social construct
around it it’s the shared vision that a
lot of people have using this network of
saying yes we are going to ascribe value
to this and yes we’re going to go buy
this set of rules and not by any other
set of rules this is what actually makes
Bitcoin so why does any of this matter
why do you know why do we care if it’s
working you know why why make this
distinction it is just pedantry on my
part well not quite
so it gets interesting when you look at
innovation so Bitcoin the beacon paper
came out of 2008 Network launched in
2009 there’s been a lot of innovation
since then because obviously academics
saw this great innovation and started
working at it and there were a lot of
proposals so here’s just three I like
one is called Bitcoin and gee it’s an
academic proposal for increasing the
throughput that’s the amount of
transactions per second for Bitcoin
about 10 X to 100x Oh near no one’s
talking about it zero cash was an
academic paper it gave rise to a
cryptocurrency called e cash now which
preserves privacy it turns out that if
you make a transaction in Bitcoin the
entire world can see it so that’s you
know that’s an interesting development
or a cerium which implemented smart
contracts which are these electronic
agreements that can be enforced in the
same way that the network enforces
transactions if you go to a Archive
which is the largest repository of
scientific paper now you’ll see this you
know great increase of papers mentioning
Bitcoin so there’s a lot of active
research why does it matter well unless
Bitcoin was perfect from inception you
might want to change the rule sets you
might want to make the rule sets better
because you have new ideas but how do
you do this let’s say that we were all
on focused on number 5 and all of the
sudden we actually you know what we
would be better off having chosen number
6 right how do you actually move that
shilling point well there’s two way we
can do it one way is I’m on stage I
could tell everyone hey everyone you
know everyone pick number six
we’re gonna win by picking number six
and because I have a microphone I have a
loud voice made me I’m able to do that
or we can try to have these informal
discussions but it’s it’s a messy
process and what typically happens when
you try to have innovation in the system
it’s something called a fork so the idea
of a fork is that you start with one
system and then someone say hey you know
what we should move from Alpha and we
should you know we should move to kiddin
number six because that’s you know
that’s clearly a better ID and we better
served and some people are gonna say yes
you know what six is a good idea six is
legitimate this is what we should all be
focusing on but other people will say no
actually I prefer the other version and
so the value that was in your network is
now going to be split into two one part
of that value is going to go to the
people who are looking at one network as
as legitimate and the other to the other
network and it’s not going to be exactly
zero one because there’s always
uncertainty things can happen so
typically when you see Forks it’s tended
to be something like 10% 90%
historically but it’s still what is
going to capture most of the value now
what do they capture the value best
based on is it’s the best Fork who is
going to win is it the fork that’s the
most obvious I would argue that at least
right now the main criterion for picking
a fork is legitimacy and as a use case
there’s been a debate reaching in a
Bitcoin community for about a year and a
half now on the size of blocks so as you
may know a Bitcoin is based on a
blockchain which is instead a structure
that produces a block every about 10
minutes and blocks have traditionally
been this 1 megabyte files that contain
transaction now there’s been one faction
in a between community who’s been saying
hey if we owned if only we had bigger
blocks then we could have more
transactions and if this would be great
because we need to have more
transactions well the other side has
been saying well you know that may be
true but if we do that our network is
going to be less decentralized some
parties are gonna have more influence
than others and when I find fascinating
is that in a lot of these debates both
sides were quoting Satoshi Nakamoto the
pseudonymous
founder of Bitcoin to each other
now Satoshi has not been seen online for
the past few years but they were saying
hey clearly in the vision of Satoshi we
need to have big blocks and the other
one who say no no clearly look at this
clearly we should have small blocks and
so they were engaging in this whole
hermeneutics for weird reasons so one
possible reason might be that it’s
simply argument from Authority they’re
saying well clearly said she knew what
they were doing and so they would have a
good opinion on the matter but there’s
also a lot of very smart people looking
at because there’s a lot of very smart
opinion so I don’t think this is what’s
happening what’s happening is a play for
legitimacy it’s two sides saying I’m the
legitimate branch and so I want to take
the economic value with me whereas the
other side said oh no no no no we’re the
legitimate branch we should have a value
now we’ve moved completely away from
algorithm we’ve moved away from
mathematics but what’s going to
determine the rules of this ledger is
the interpretation of the right of the
writing of a paper reading 9 years ago
this this is completely the realm of
politics and the human element is
nothing to do with algorithm it’s
nothing to do with math the dynamic at
play between those groups he sadly
captured in a game known as the game of
chicken so again this is gonna date me
this is James Dean rebel without a cause
and so in a game of chicken you have two
cars and they’re racing toward each
other and you win the game by not
swerving because the other side who is
slightly less crazy than you is gonna
swerve before and the problem with a
game like this is that if both sides are
you know equally and willing to swerve
the majors crash towards one another so
you have the same dynamic at playing a
fork because you may not want to fork
you don’t want to divide the the network
which you primarily want is you want to
steer it but you’re gonna try to
convince the other side that no matter
what you’re gonna try to you know you’re
gonna go your own way no matter what
you’re gonna keep going in a straight
line and both sides do that and it’s
really not an effective way of doing
governance so what I’ve been working on
is formalized governance for blockchain
and cryptocurrencies
and so just a word about governance so
one way to think about it is imagine in
your own an apartment building not only
apartment built a partner
in an apartment building so you own your
apartment everything you do in your
apartment is your own business I don’t
care the neighbors hopefully don’t care
but what happens when you need to decide
when the heat should be turned on for
the building or should the fast choose
the facade of the building be repainted
you need a way of making these decisions
which are going to engage the common
parts of the building and in order to do
that you need to have governance now we
have several options you can do nothing
you can say we don’t like governance we
don’t want to have governance we own our
apartment and that’s enough and that may
be fine but then the paint is going to
chip on your facade and you’re not gonna
have heating your building in the winter
that’s not that great you could try not
having any process and just letting
whoever yell the loudest make those
decisions but again it’s still gonna
engage your resources that’s not
necessarily what you want so what can we
do well instead of having all these
informal processes we can formalize we
can create rules for saying we have come
in parts and there are common parts in
block chains one you know the most
important of which being innovation so
we all agree that we need to innovate we
need to introduce some time changes in a
platform but if we’re gonna do that we
need to have a process because if we
don’t have a process it’s going to be
determined by completely random means
that have to do with shelling points
instead of having to do with merits so
one simple way of doing this is simply
voting you can have you know block shins
or great records keepers they’re a great
way of coordinating action and so you
can have a vote you can say to people
okay we’re gonna have a new proposal for
innovation do you actually want to do
this and it’s very important for people
to be able to express their intention
without having to try to guess what
other people are doing you want people
to tell them this is what I prefer and
not this is what I expect everyone else
is going to prefer now it’s a very it’s
a subtle distinction but it makes a
whole world of difference so if I just
want to leave one ID one important idea
tonight is that yes there are algorithms
at play there are mathematics but in the
end we’re dealing with human systems
right cryptocurrencies block chains they
are social construct the algorithm is
just a scaffolding upon which you flesh
out all of these human activities so
when designing
systems think about the human element
don’t think of them purely as technology
think of them as enablers of social
structure and only Suzette will be able
to really leverage the coordinating
abilities of blockchains and related
technologies
[Applause]
Please follow and like us: