►
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Hey
everyone
thanks
for
having
me
great
to
be
here.
My
name
is
tom
trobridge,
I'm
speaking
about
fluence,
but
really
about
first
about
the
need
for
web3
storage
payments
and
compute.
But
first
I
want
to
thank
the
ipfs
community,
the
filecoin
community,
juan
particular,
but
everybody
for
having
us
and
partnering
with
us.
It's
a
terrific
ecosystem
to
be
a
part
of.
A
We
think
that
there's
nothing
more
important
than
storage
as
one
of
the
foundational
kind
of
critical
components
of
web3,
and
so
I'm
going
to
talk
about
two
other
components,
obviously
particularly
compute,
but
without
storage.
None
of
this
happens.
None
of
this
works.
So
I'm
thrilled
to
see
this
group
here
and
for
us
to
be
a
part
of
this
community.
So
my
name
is
tom
trobridge,
I'm
co-founder
of
fluence
labs.
A
So,
first,
in
terms
of
censorship,
you
know:
big
tech
is
really
becoming
an
arm
of
governments
and
that
shouldn't
be
super
surprising,
because
every
single
industry,
the
larger
it
gets
governments
have
to
interact
with
it
and
those
businesses
need
to
interact
with
governments,
and
so
you
can
see
governments
coming
down
really
hard,
particularly
in
you
know,
there's
lots
of
examples
in
russia.
Censoring
google
censoring!
You
should
youtube
that
you
sort
of
expect
from
totalitarian
governments
right
that
has
happened
in
china.
This
is
not
new.
A
India
was
one
of
the
first
examples
where
this
was
really
clear,
where
the
indian
government
forced
whatsapp
to
de-platform,
effectively
opposition
politicians
and
actually
stop
a
whole
variety
of
messaging
around
a
variety
of
protests
that
farmers
were
doing
and
that's
an
obvious,
explicit,
democratic
government
exerting
its
influence
through
single
points
of
contact
with
with
huge
user
bases.
They
were
very
easy
to
do
because
they're
web
2
and
can
be
can
be
pressured
given
shareholders
given
in
jail
potential
for
employees
etc.
A
A
more
recent
example
is,
you
know,
twitter
blocking
the
post
hunter
biden
story,
which
we
now
find
out
was
actually
real
and
they
didn't
allow
any
discussion
of
that.
There's
a
whole
number
of
covert
related
elements
too.
That
now
seem
true
that
twitter
didn't
allow
discussion
of
as
well,
so
whether
that
was
explicit,
government
discussion
or
just
censorship
from
implied
censorship
is
a
little
bit
unclear.
A
But
the
point
is:
you've
got
these
big
platforms
that
people
rely
on
that
have
vulnerabilities
for
censorship,
but
besides
that,
you
also
have
the
hosting
element
that
happens:
the
hosting
component
of
big
tech
and
when
you
have
hosting
dominated
by
a
handful
of
large
companies,
you
lead
to.
That
leads
to
a
whole
number
of
potential
issues
regarding
not
just
censorship,
but
also
pricing
and
also
innovation,
stifling
and
so
the
top
three
hosting
companies
control
about
60
percent
of
the
cloud
and
they're
growing
at
50
a
year
and
they're,
incredibly
large,
powerful
businesses.
A
And
so
why
are
there
they're
driving
the
returns
of
these
companies?
And
why
is
that
the
case?
It's
because
they're
hard
to
leave
and
they
perform
a
very
good
service
when
you're
small.
They
give
you
a
lot
of
technology
to
actually
accelerate
the
development
of
your
applications
and
your
businesses
early
on,
but
then
you're
unable
to
leave
very
easily
afterwards.
A
A
A
You
know
die
on
parlor's
kind
of
social
good
neces
right,
but
at
the
same
time
you
got
to
realize
that
that
is
that
provides
real
warning
to
other
groups
on
that
platform
about
how
quickly
they
can
be
taken
off
with
the
with
effectively
arbitrarily
and
that's
what
happens
when
you
private
companies
in
control
of
this,
they
can
pull
the
switch
on
you
and
even
if
you
haven't
violated
tus
by
the
time,
you've
fought
that
out.
Your
business
is
seriously
jeopardized
if
it's
still
alive
and
resilience
and
you
make
those
trade-offs.
A
If
you
have
incredible
resilience,
you
that's,
maybe
some
some
reasonable.
That
might
be
a
reasonable
trade-off
to
some
extent,
but
the
larger
these
businesses
get
the
more
complex
they
are
and
the
more
the
resilience
has
been
threatened.
And
so
we've
seen
amazon
cloud
go
down.
We've
seen
facebook
go
down,
we've
seen
what's
app,
go
down,
all
of
these
businesses
have
suffered
serious
outages
and
they
have
gotten.
They
haven't
gotten
less
as
they've
gotten.
This
more
infrequent
as
the
businesses
have
scaled
they've
gotten
more
frequent,
and
we
don't
expect
this
to
stop.
A
These
are
hard
to
run
they're
complex
businesses.
There's
a
lot
of
there's
incredible,
different
single
points
of
failure.
Multiple
single
points
of
failure
given
geographies
and
data
centers
involved
and
human
error
comes
into
play
regularly
with
almost
all
of
these
outages,
and
so
you,
besides
having
censorship,
you
have
resilience
issues
which
were
supposed
to
be
going
in
the
other
direction
and
being
more
resilient
as
technology
advances
and
the
question
I
think
everyone
here
knows
the
answer
to
but
I'll
ask
it
rhetorically.
A
Is
this
the
internet
we
signed
up
for-
and
I
don't
think
so
and
tim
berners-lee
did
not
have
this
in
mind
when
he
created
you
know,
basically,
the
the
hyperlink
and
the
easy
access
information
that
we
all
take
for
granted
now
on
the
world
wide
web.
A
But
the
issue
is
that
internet
economics
rewards
scale,
and
that
was
something
not
easily
foreseen
by
everybody
initially,
but
the
scale
that
a
global
platform
can
catalyze
leads
to
ever
increasing
oligopolist
potential,
where
the
larger
you
are,
the
more
you
can
spend
on
advertising
the
more
you
can
spend
infrastructure,
the
more
you
can
attract
people
and
that
then
allows
you
to
out
compete.
A
The
second
biggest
and
that's
kind
of
flywheel
keeps
going
and
that's
why
you
have
and
why
you've
gotten
to
the
place
of
the
amazons
and
the
googles
and
the
facebooks,
and
you
know
you
na
you
name,
you
know
the
the
the
providers,
the
microsoft's
right
and
so
that
continues
and
they
because
they're
so
large.
They
also,
as
I
mentioned
earlier,
have
to
have
relationships
with
governments.
There's
no
large
industry
in
the
world
that
doesn't
have
a
very
close
relationship
with
governments
both
need
to,
and
so
that
also
is
not
going
to
go
away.
A
It's
only
going
to
continue
and
the
question
is
only
to
how
much,
and
so
that
is
the
real
question.
So
how
did
we
get
here
and
how
do
we
evolve
beyond
this?
Because
governments
are
just
sort
of
dipping
their
toes
in
this,
their
extreme
potentially
in
places
like
russia
and
china,
but
the
u.s.
It's
only
going
to
increase
and
we
probably
don't
even
know
the
level
of
actual
involvement
and
relationship
between
these
companies
right
now,
and
it
probably
is
only
going
to
continue
so
we
started
off
with
mainframes
and
desktops
right.
A
That
was
what
web
1.0
and
then
kicking
and
screaming
people
were
pulled
into
cloud
computing,
and
I
remember
very
well
back.
You
know
I
was
looking
at
and
was
even
investing
in
some
of
these
cloud
providers
back
in
the
90s
and
people
were
like
yeah
I'd
rather
host
it
internally.
I'd
rather
have
more
control
over
my
data.
It's
cheaper!
It's
more
resilient,
well
fast
forward.
A
Now
all
those
reasons
that
people
want
to
host
internally
are
why
they're
at
the
cloud,
but
those
exact
reasons
are
why
we
think
people
are
all
going
to
move
to
peer-to-peer
platforms
because
peer-to-peer
will
be
more
resilient,
he'll
be
cheaper
and
will
be
less
it'll,
be
censorship,
resistant,
and
so
that's
why
we
think
open
peer-to-peer
networks
are
the
future.
There
are
no
centralized
bottlenecks.
A
There
are
no
single
points
of
failure
and
there's
much
more
censorship
resistant,
and
that
is
really,
I
think,
we'll
go
through
each
one
of
these
at
a
time
to
show
why
we
think
each
one
of
these
is
compelling,
but
first
peer-to-peer
can
scale
decentralized,
open
architecture.
You
know
it
allows
the
addition
of
limitless
resources
effectively
on
demand,
and
so
it's
sort
of
when
pricing
when
there's
a
shortage.
A
And
so,
when
you
have
an
open
protocol
on
a
peer-to-peer
network
on
which
an
open,
decentralized
application
or
a
dow
is
running,
you
really
have
a
massive
censorship
or
massive
benefit
of
censorship.
Resistance.
Given
the
distributed
nature
of
both
the
code,
the
hardware
and
the
actual
human
talent
behind
it-
and
you
need
all
three
of
those
things
in
order
for
things
to
work
in
a
in
a
way
that
will
actually
resist
some
of
the
obvious
interference
that
we've
seen
and
then
in
terms
of
network
resilience.
A
You
have
applications
that
live
in
the
network
and
they
can
be
hosted
in
different
geographies
by
different
providers
by
different
clouds.
They
can
move
very
easily
and
if
any
host
would
like
to
host
an
application,
they
can
do
so
right
now.
This
is
this
is
there's
great
benefit
to
this,
there's
also
risk
to
it
right.
There
are
applications,
I'm
sure
none
of
us
would
like
to
see
that
could
exist
more
easily
in
this
type
of
a
world,
and
so
there
is.
A
But
that
is
something
I
just
have
to
call
out
that
it's
not
purely
positive
to
have
an
open,
permissionless
peer-to-peer
system,
because
you
can
have
things
that
we
don't
want
right,
and
so
this
is
a
trade-off
that
we're
all
going
to
have
to
face,
and
unfortunately
I
don't
have
the
answers
for
it,
but
I
think
that
we're
shifting
right
now
towards
so
much
centralization
and
so
much
censorship
that
we
probably
air
a
little
bit
on
the
other.
A
It's
going
to
go
the
other
way,
hopefully
for
a
while
and
we'll
figure
out
kind
of
where,
where
that
middle
middle
place
should
be
so,
but
peer-to-peer
platforms
right
now
lack
the
compute
layer
and
so
peer-to-peer
requires
three
things
to
be
a
fully
functioning
ecosystem.
A
You
need
payments,
you
need
storage
and
you
need
compute
right,
and
so
we
know
that
two
of
those
exist
and
are
working
very
well,
but
the
compute
layer
allows
the
off-chain
scalable
computation,
without
which
you're
very
limited
in
what
you
can
do.
And
so,
let's
talk
about
open
peer-to-peer
traction.
Everybody
knows
the
traction
of
crypto
right.
Crypto
started
all
this
payments
and
whether
it's
bitcoin,
whether
it's
ethereum,
whether
it's
you
know,
name
your
your,
your
your
coin
of
choice
that
is
working,
it's
functioning,
it's
global
that
has
been
solved.
A
Largely
storage,
thanks
to
ipfs
and
filecoin,
is
also
seems
to
be
solved
and
only
getting
more
and
more
solved
based
on
the
presentations
that
I
just
heard
today,
which
is
fabulous
right,
you
have
recall,
is
much
faster.
Storage
is
only
increasing
dramatically
huge
amount
of
storage
on
it
and
I
think
that's
only
going
to
increase,
but
compute
has
been
a
whole
so
far
and
you
need
compute
to
have
fully
decentralized
applications.
A
A
And
then
you
have
decentralized
hardware,
marketplaces
and
that's
fine,
but
those
also
are
generally
managed
on
chain,
and
that
is
about
sort
of
more
renting
hardware
and
it's
not
actually
about
offering
the
computation
services.
So
it's
sort
of
a
layer
below
the
actual,
clear
compute
that
applications
need.
So
what's
missing,
you
want
to
eliminate
entrance
bearers
entrance
barriers
and
you
want
sort
of
function,
addressable,
storage
and
the
ability
to
run
full
featured
back
ends,
and
basically
also
if
you
want
be
pluggable
to
block
chains.
A
If
you
want
to
have
consensus
as
part
of
your
compute,
and
so
you
basically
need
to
have
general
purpose,
peer-to-peer,
compute
and
that's
what
fluence
is
working
on
and
it's
any
computation
it
should
be
is
supported
and
it's
function,
addressable
permissionless
with
a
universal
runtime,
and
so
that's
what
we're
building
and
whether
it's
us
that
are
successful
with
it
or
others.
We
think
that
is
a
critical
component
of
making
web3
a
fully
functional
and
featured
solution.
A
You
know
we
talk
about
where
else
this
can
be
used.
You
know.
One
element
here
is
de-platforming
and
I
kind
of
referenced
it
earlier,
but
you
need
to
have
this
ability
of
computation
distributed
in
order
to
avoid
d-platforming
and
we've
seen
that
happen
in
the
web.
2
space
right
you've
seen
twitter,
kill
off
a
bunch
of
features,
you've
seen
facebook,
the
platform
I'm
dating
myself,
obviously,
but
farmville
way
back
when
and
so
you've
seen
d
platforming
happen,
both
from
an
application
level.
You've
seen
it
a
parlor
on
a
on
a
company
sort
of
level.
A
You've
seen
d
platforming
happen
across
the
space
and
the
idea
is
to
create
an
architecture
that
prevents
de-platforming,
and
that
is
what
we're
focusing
on
the
area
people
haven't
focused
on.
So
much
is
deep.
Platforming
still
is
a
risk
even
in
web
3.,
and
that's
because
there's
still
centralized
developer
apis
with
a
lot
of
the
web.
3
businesses
and
web
3
applications
that
we
know
very
well
and
no.
This
hasn't
happened
yet
we've
run
some
risks
of
it,
but
this
is
still
an
issue.
A
You
also
have
centralized
backends
for
very
well-known
and
important
web
3
businesses
right,
and
so
this
is
not
a
critique
of
these
businesses.
There
aren't
real
alternatives
right
now.
That's
what
we
are
trying
to
provide
others
are
trying
to
provide
because
you
to
be
fully
functional
web3,
you
can't
rely
on
centralized
developer
apis.
A
You
can't
have
centralized
backends
just
to
serve
a
decentralized
user
interface
right,
so
you
have
to
get
a
step
deeper
and
have
this
compute
layer,
fully
functional
and
and
and
and
live,
and
so
how
do
we
build
these
best
peer-to-peer
applications
and
this
in
this
community?
I
think
people
will
appreciate
it,
but
it
doesn't
have
to
be
blockchain
based
all
right.
Every
people
here
understand
that
storage
is
not
all
on
chain
right.
A
The
incentives
are
on
chain
and
the
verifications
are
on
chain,
but
the
storage
doesn't
have
to
be,
and
so
compute
also
doesn't
have
to
be
on
chain,
and
if
it
is
on
chain,
you
run
into
a
variety
of
constraints,
and
so
also,
let's
not
forget
the
current
hugely
successful
cloud
businesses.
Their
compute
is
not
on
chain
and
they're,
doing
just
fine
with
a
huge
number
and
of
very
important
significant
customers
who
pay
them
a
lot
of
money,
and
so
I
like
we
like
to
look
at
it.
A
This
way
where
you
have
centralized
computing
and
on-chain
computing
and
in
the
middle
is
where,
where
fluence
exists
and
where
we
think
peer-to-peer
computing
exists,
where
you
have
no
middlemen,
you
don't
have
the
redundancy
you
do
of
of
blockchain-based
computing,
but
it
is
verifiable.
A
You
can
plug
in
consensus
and
you
can
anchor
on
chain
it's
far
cheaper
and
far
faster
than
on-chain
computing,
and
it's
not
deterministic,
and
so
I
think
in
this
group
I
probably
have
to
spend
less
time
on
that,
given
your
familiarity
with
ipfs
and
filecoin,
and
how
that
storage
and
verification
works.
But
it
is
important
that
people
sometimes
equate
peer-to-peer
with
on-chain-
and
you
know
all
you
do
is
go
back
to
the
napster
days
to
realize
right.
That's
not
that
doesn't
have
to
be
the
case
and
and
started
off,
frankly,
not
being
the
case.
A
So
what
is
fluence?
It's
a
peer-to-peer
application
platform,
and
so
it's
a
computing
network
of
nodes.
It
is
a
programming
language
that
is
the
first
programming
language
for
native
peer-to-peer
application
composition
and
then
that
allows
basically
a
code
creator
economy
I'll
talk
about,
because
I
think
that
that
kind
of
it
gets
to
some
of
the
resilience
I
was
referencing
earlier.
A
But
what
this
application
language
does,
which
is
called
aqua.
It
basically
allows
functions
to
exist
into
and
to
to
run
without
central
coordination,
and
so
you
can
create
a
function
that
looks
up.
The
next
basically
looks
up
the
location
for
the
computation
to
be
performed,
completes
it
and
then
looks
up
the
next
next
pier
to
perform
the
computation
and
then
continues
and
it's
very
different
than
the
client
server
or
you
know
the
the
the
client
server
type
of
of
computation,
and
that
is
a
terrific
innovation
and
the
basically
the
application.
A
Construction
in
aqua
allows
you
to
create
these
app.
The
the
the
applications
inc,
much
much
faster
than
if
you
were
to
try
to,
for
example,
use
rust,
which
we
think
is
kind
of
the
closest
the
closest
peer
to
it.
It's
based
on
pi,
calculus
and
basically
views
applications
as
network
protocols
and
I'm
going
to
skip
through
some
of
this
because
we're
running
a
little
bit
a
lot
of
time,
but
you
can
see
what's
available
now,
which
is
pure
discovery.
A
So
if
you
think
about
how
software
and
code
has
been
developed
so
far,
you
have
look
at
this.
It's
version,
1
version,
1.1
version
1.2
and
the
author
or
company,
creates
version
1.0
it
discontinues
it.
You
move
to
version
1.2
it
discontinues
it
and
then
you
can
see
all
the
hosts
are
hosting
version,
1.2
we're
kind
of
comfortable
and
that's
what
normally
happens,
but
if
you're
running
a
peer-to-peer
network
and
on
the
flu
and
the
fluence
architecture
allows
this
you.
A
Basically,
when
an
author
puts
version
one
out
there,
it's
out
there
and
as
long
as
a
host
has
a
customer
that
wants
to
pay
for
that
version.
One
code:
it's
there.
If
the
author
wants
to
change
it,
he
can't
it's
already
out
there
right.
The
author
can
post
version.
1.1
host
can
host
that
author
can
create
version.
1.2
host
can
host
that,
but
there
is
no
record
there.
A
It
also
prevents
disgruntled
developers
from
inserting
malicious
code
into
stuff
into
into
existing
kind
of
applications,
which
we've
seen
happen
as
well,
and
so
what
that
basically
and
just
to
go
back
on
this,
the
way
we've
architected
this
is
code
is-
is
hosted.
The
host
shares
a
small
amount
of
that
revenue
with
the
author,
and
so
some
developer
using
the
code
will
pay
a
hosting
fee.
Just
like
you
do
on
any
other.
A
This
is
the
code
creator,
economy
concept
that
actually
will
allow
open
source
developers
to
issue
code,
have
it
work
and
actually
have
some
revenue
associated
for
that
for
as
long
as
that
code
continues
to
be
served,
and
so
they
kind
of
mention
this
that
you
know
nodes
are
basically
you're
sort
of
no
longer
out
of
you're
out
of
that
upgrade
cycle
as
a
user
and
as
a
developer,
you
benefit
from
anyone
using
it,
and
so
where
are
we
right?
Now?
We've
launched
the
aqua
and
the
peer-to-peer
network.
It's
live.
You
can
look
at
you.
A
Can
you
can
look
at
it
currently
we're
working
on
launching
a
dow
that
will
happen
this
year
to
govern
our
network
and
then
the
compute
economy,
which
is
effectively
that
smart
contract-based
relationship
between
the
code
that
is
created
and
the
host
that
hosts
it?
That
is
the
element
that
we're
working
on
next
and
should
be
out
later
this
year
as
well.
A
So
you
know
find
us
here
at
our
developerhub
dash.fluence.dev,
and
you
know
I
guess
what
I
and
I'd
also
say
that
here's
some
other
elements,
the
fluencefluence.network
our
discord
as
well.
A
I
don't
think
we
want
our
content,
monitored
by
nation
states
right
and
so
right
now,
they're
really
high
barriers
to
building
independently,
not
on
these
three
companies
and
fluence
is
trying
to
build
a
world
and
a
network,
an
environment
that
makes
it
easier
and
simpler
and
we
hope,
even
better,
to
to
to
create
and
build
outside
of
all
of
these
ecosystems
and
in
an
ecosystem
governed
by
and
and
that
rewards
the
whole
developer
community
globally,
and
that
is
that
is
our
vision.
A
And
so
I
just
say
here
to
support
web3
host
contribute,
follow,
and
you
know
we're
super
excited
to
be
working
as
closely
with
ipfs
and
filecoin,
as
we
can
to
help
enable
that
world
that
I
think
we
all
believe
in,
and
so
with
that.
I
thank
you
for
your
time
and
attention
and
happy
to
take
a
few
questions.
A
So
it's
that
we're
working
on
that
right
now,
so
we
have
nodes
up
currently
and
so
they're.
They
phrase
it
this
way
it's
gonna
be
more
about
reliability
than
power
and
right.
So
it's
not
like
bitcoin.
It's
not
like
mining,
it's
more
about
the
computation
level
you
want
to
provide
and
then
your
cost
of
providing
that
versus
the
cost
of
other
people
providing
it.
So
I
think
it's
going
to
be
more
about
what
people
will
value
might
think
is
going
to
be
more
about
reliability
than
raw
computation
power.
A
We
don't
think
of
fluence
as
being
ideal
for
like
sequencing,
a
genome
or
something
like
that.
It's
not
that
raw
power,
but
it
is
about.
We
think
people
will
care
more
about
reliability,
so
bandwidth
up
time.
You
know,
network
up
time,
etc,
and
I
think
that's
going
to
be
more
of
a
focus
than
than
raw
power.
A
I
mean
listen
and
the
answer
is
no,
but
machine
learning
is
like
things
that
we're
like
yeah
there's
so
many
different
kinds.
So
I
don't
know-
and
the
other
point
I
should
mention
is
also
we've
seen-
people
want
to
run
fluence
nodes
internally
on
their
own
systems
as
well,
so
they
we
right
now
what
people
are
doing
influence
is
actually
using
their
own
hardware
to
run
applications
and
to
develop
on
top
of
it,
and
so
I
think
that
we
could
see
that
happen
internally.
A
I
don't
know
whether
you
will
use
it
to
have
so
from
that
side.
Yes,
I
don't
think
it's
you
will
see.
People
say
let
me
get
six
different
hosts
around
the
world
and
together
kind
of
run,
a
machine
learning
application.
I
doubt
that,
but
will
someone
either
do
it
in
house?
Maybe,
or
will
someone
find
one
host
that
has
sufficient
kind
of
both
bandwidth
and
computation
to
do
it?
Possibly?
A
But
it's
not
a.
Let's
find
you
know
spare
laptop
capacity
around
the
world
and
pull
it
all
together.
We
don't,
we
don't
think
that's
the
realistic
use
and
we've
got
breakfast
tomorrow
at
nine.
If
anybody
wants
to
meet
more
of
the
team
and
our
decentralized
summit
tomorrow
at
the
vox
hotel,
so
we're
on
the
main-
oh,
oh
sorry,
metropolitan
my
bad,
but
we
should
just
look
at
the
the
the
general
schedule
and
we're
on
it
all
right
for
devconnect
thanks
everybody
appreciate
it.