►
Description
In 2010, the Africa Internet technical community embraced the 80/20 Initiative - an ambitious vision to have 80% of all Internet traffic locally accessible with only 20% being sourced internationally by 2020. In this webinar we try to understand what has changed across the continent during this period and what lessons can we draw from this initiative going forward using the findings from our latest study, “Moving Toward an Interconnected Africa: The 80/20 Initiative”.
SPEAKERS
Dawit Bekele, Regional Vice President - Africa, Internet Society
Jane Coffin, Senior Vice President, Internet Growth, Internet Society
Andrew Baskett, Edge Network Strategy Manager, Facebook
Michael Kende, Senior Fellow and Visiting Lecturer, Graduate Institute, Geneva
MODERATOR
Michuki Mwangi, Senior Director, Internet Technology and Development, Internet Society
READ THE REPORT
https://www.internetsociety.org/resources/doc/2021/moving-toward-an-interconnected-africa-the-80-20-initiative/
MORE ABOUT THE EVENT
https://www.internetsociety.org/events/moving-towards-an-interconnected-africa-report-launch-webinar/
EXPLAINER - WHAT IS EIGHTY FOR AFRICA
https://www.internetsociety.org/resources/doc/2020/explainer-what-is-eighty-for-africa/
#80forAfrica
A
Good
morning,
good
afternoon,
good
evening,
depending
on
where
you
are
welcome
to
this
really
special
moment
for
us
all,
where
we
are
going
to
witness
the
launch
of
a
very
important
study
that
we've
been
working
on
for
nearly
the
last
10
months
or
more,
but
actually
highlights
work.
That
has
been
ongoing
for
the
last
decade
and
it
really
helps
us
to
take
a
step
back
and
look
at
what
the
work
that
we've
been
doing
for
the
last
decade.
With
respect
to
interconnection,
peering
interconnection
and
traffic
exchange
in
africa
looks
like
my
name
is
amichukimongi.
A
I'm
going
to
be
your
moderator
this
afternoon
and
I'd
like
to
just
welcome
you
to
join
us
today.
We
have
a
great
set
of
speakers
and
I
hope
that
you
will
be
able
to
take
away
something
that's
useful
from
this
session.
So,
let's
start
with
the
objective
of
this
session.
So,
as
you
all
know,
we've
had
you
know
the
work
of
the
peering
and
interconnection
going
on
for
quite
a
number
of
years.
A
The
work,
especially
here
at
the
internet
society
and
so
we're
looking
to
present
the
report,
which
sort
of
looks
at
the
80
20
initiative,
which
is
the
goal
that
the
community
took
on
after
you
know,
it
was
shared
across
by
the
internet
society
of
having
80
percent
of
the
traffic
locally
accessible
and
20
percent
coming
from
outside
of
the
region,
and
so
we
like
to
see
based
on
the
study
that
has
been
done.
What
were
the
findings?
A
What
are
the
things
that
we've
been
will
be
able
to
learn
from
the
study
itself
or
the
report
itself
the
lessons
and
what
are
the
next
steps
going
forward?
So
we
hope
that
michael
is
going
to
be
able
to
share
with
us
what
are
the
things
that
he
learned
through
it
and
the
one
thing
that
we
are
quite
happy
about.
Having
michael
do
this
is
because
there's
the
papers
that
are
actually
preceding
this,
that
sort
of
set
the
stage
for
this
report
of
the
kenya
and
nigeria
studies
were
done
by
michael.
A
So
he
has
good
background
with
respect
to
the
study
itself,
we're
ready
to
get
started
and
without
further
ado,
I'd
like
to
invite
our
colleague
from
facebook
that's
andrew
basket.
Andrew
baskett
is
the
edge
strategy
manager
at
facebook
and
he's
also
been
a
very
good
partner
and
collaborator
with
us
here
at
the
internet
society
having
initiated
our
partnership
project
with
facebook,
going
back
to
2019
2018
actually,
and
that's
looking
at
the
ixp
development
work
which
started
in
africa
now.
A
The
project
has
been
expanded,
covering
that
in
america
and
the
asia
pacific
and
prior
to
joining
facebook,
akama
of
andrew
watts
at
akamai,
and
that's
where
we
really
got
a
chance
to
work
with
him
along
the
way
that
we've
done
in
africa
and
we're
really
excited
that
he
and
through
him
and
facebook
we've
been
able
to
collaborate
on
this
study.
So
andrew.
Just
please
welcome
and
give
us
a
few
remarks.
B
Thank
you
so
much
mitchuki,
it's
it's
so
exciting
to
represent
facebook
and
and
our
support
for
the
internet
society
and
and
be
part
of
this
session.
You
know
for
for
those
that
might
not
know
at
facebook.
Our
mission
is
to
connect
everyone
and
we
invest
in
foundational
and
transformative
technology
and
initiatives
to
support
that
in
order
to
bring
more
people
online
to
the
faster
open,
more
reliable
and
efficient
internet.
So
in
that
sense,
facebook's
interest
and
the
internet
society
and
our
community
are
very,
very
closely
aligned.
B
I
had
a
chance
to
review
the
report
and
work
work
with
the
team
on
it
and
I'm
just
so
impressed
by
it
by
it
and
what
the
industry
has
accomplished
so
far,
and
it's
so
great
to
have
michael
work
on
a
lot
of
these
reports
and
and
have
that
continuity
that
you
mentioned
rajuki,
so
I'm
just
so
happy
and
looking
forward
to
to
everyone's
feedback
on
how
this
went.
So
thinking
back
to
my
first
african
peering
and
interconnection
forum
in
2012,
the
interconnection
ecosystem
was
significantly
different.
B
Facebook
had
very
little
traffic
served
locally
in
the
continent
and
no
physical
infrastructure
fast
forward
to
today.
Facebook
is
present
in
over
20
internet
exchanges
in
africa
alone
and
support
more
than
90
of
our
traffic
from
within
africa.
It
will
be
impossible
to
do
this
and
achieve
these
results
without
these
efforts
and
the
growth
of
ixps
across
the
region.
So
everyone
involved
in
the
community
had
a
had
a
role
in
that
which
is
great
to
just
think
about,
as
we
look
forward
to
even
more
challenges
in
the
future.
B
So
I'm
really
excited
to
to
see
what
people
think
about
the
report,
which
I
believe
will
anchor
many
of
the
conversations
we
need
to
have
with
the
community
industry
and
policy
collaborators,
and
it's
even
more
important
that
we
work
closely
together
to
improve
the
resiliency
of
the
african
internet
as
we
look
toward
the
next
set
of
challenges
that
we'll
have
so
with
that.
It's
a
pleasure
to
hand
it
over
to
the
back
to
you
and
chucky
and
jane
endowed.
A
A
At
the
very
start
of
the
work
going
back,
the
interconnection
and
traffic
exchange
work
going
back
to
2010,
though
it
was
pretty
much
the
person
in
charge
of
the
development
work
in
africa,
and
we
worked
very
well
with
him
in
a
lot
of
the
work
we
did
on
developing
afb
the
community
and
the
ixp
work,
and
I
would
like
to
just
give
him
that
opportunity
to
sort
of
share
his
perspectives
as
well,
based
on
where
we
are
and
how
he
has
seen
that
evolution
over
to
you.
Darwit.
C
As
much
has
said,
it's
been
over
ten
years
that
we
have
been
working
on
the
interconnection
and
traffic
exchange
program
and,
as
you
can
imagine,
when
you
are
on
such
a
long
journey,
sometimes
you
wonder
whether
you
are
having
impact,
whether
you
are
in
the
right
direction
and
so
on,
and
I
have
to
say
that
this
study
comes
at
a
very
right
time.
In
fact,
it
shows
us
a
little
bit
of
it.
It
works
as
a
compass.
C
I
don't
remember
exactly
the
the
percentage,
but
it
was
maybe
10
or
20
percent
of
the
traffic
exchange
locally,
so
that
goal
seemed
unrealistic
and
frankly,
at
the
very
best
it
was
audacious,
and
I
have
to
say
that
the
study
has
shown
us
that
we
might
have
not
reached
the
goal
everywhere,
but
we
did
reach
or
approached
it,
at
least
in
three
countries-
south
africa,
kenya
and
nigeria,
and
well,
if
we
have
not
reached
other
countries
in
other
countries.
At
least.
A
A
He
came
and
said
why
don't
we
have
a
very
bold
statement
like
objective
that
we
work
towards
and
if
you
remember
around
the
time
there
was
the
whole,
what
was
it
called
the
u.n
millennium
development
goals
for
2015,
and
now
it's
thought
was
like.
We
should
have
something
as
big
and
as
audacious
as
that.
So
when
we
said
it,
he
said
no,
that's
too
ambitious,
but
it's
good
that
we
try
and
aim.
A
But
yes,
so
I
just
want
to
recognize
that
now
it's
trigger-
and
you
know,
challenge
to
us
as
a
team
to
come
up
with
something
very
bold
and
ambitious
that
we
can
go
out
to
the
community
with
led
to
sort
of
us
being
here
where
we
are
today.
Thank
you,
david.
Next,
I
would
like
to
give
the
opportunity
to
jane
coffin.
A
Jen
coughing
is
our
senior
vice
president
for
internet
growth,
and
most
of
you
probably
have
met
jen
in
one
way
or
the
other,
but
she's,
probably
a
go-getter
person
who
is
responsible
for
the
internet,
society's
growth
project
teams
and
under
the
growth
project
teams.
We
focus
on
community
networks,
interconnection
and
traffic,
ixps,
peering
community
development
and
also
a
very
interesting
new
project,
called
internet
measurement,
which
is
looking
at
the
health
of
the
internet.
A
She's
also
been
involved
in
this
work.
She
joined
the
internet
society
in
2012
and
took
over
from
where
we
were
working
with
our
colleague
karen
rose,
who
some
of
you
may
know,
and
so
she's
sort
of
taken
on
that
mantle
and
carried
it
forward
and
made
sure
that
all
the
goals
and
visions
that
we
had
sort
of
laid
out
at
the
very
beginning.
D
Thank
you
michiki,
and
thank
you
everyone
for
being
here
this.
This
webinar
really
helps
encapsulate
and
show
the
world
the
importance
of
the
continuity
we've
established
at
the
internet
society
with
our
partners
and
a
big
thank
you
to
all
our
partners,
but
the
continuity
of
working
with
michael
and
working
with
all
of
you
to
show
the
importance
of
internet
exchange,
points,
interconnection
and
the
communities
to
support
them.
D
We
have
had,
over
the
years
many
debates
about
how
to
move
forward
and
improve
what
we
do
as
we
go
endowed
in
machuke
led
the
way
with
the
innovation
of
80
20
by
2020,
and
by
setting
those
goals
we
had
benchmarks.
We
were
looking
at
findings
and
data
to
improve
what
we
do
and,
as
many
know
out
there,
data
is
critical
to
proving
your
point
to
proving
that
building
infrastructure.
D
So
this
is
a
partnership
to
build
the
skills
and
the
infrastructure
to
improve
lives,
to
improve
technology,
to
improve
traffic
flows.
So
it's
not
just
about
the
data,
it's
about
the
local
circumstances
as
well
to
build
a
bigger
regional
and
global
community
of
interconnected
networks,
for
better
data
flows
and
for
better,
as
I
said
before,
at
gdp.
D
So
that's
why
daoid
and
machuchi
set
those
goals
they
created
the
benchmarks.
So
we've
got
the
findings
in
this
report,
which
michael
will
speak
about
just
as
a
snapshot
for
you.
There
are
47
ixps
according
to
fyx
right
now
on
the
in
africa,
with
in
43
cities
in
33
countries.
When
I
first
started
at
the
internet
society,
I
believe
that
number
was
17..
D
The
bottom
line
is,
is
that
there
has
been
change
and
that's
due
to
the
work
that
many
have
done.
Daoit
mentioned
a
lot
of
those
organizations,
it's
very
important
to
realize
that
it's
an
interconnected
network
of
partners
as
well.
That's
part
of
that
foundational,
along
with
foundational
technology.
It's
foundational
partners
in
foundational
support,
so
I
would
just
close
by
saying
that
you're
going
to
hear
about
the
enabling
factors
that
are
so
important
to
develop
the
infrastructure
and
again
I
would
be
remiss
if
I
didn't
say
thank
you
to
the
community.
D
We
can't
do
this
alone
and
the
internet
society
often
says
that
we
do
not
want
to
be
a
single
point
of
failure.
As
a
partner
like
in
a
technical
network,
you
don't
need
single
points
of
failure.
You
certainly
don't
want
that
as
a
partner.
So
thank
you
to
everyone
for
joining
us
today
and
thank
you
to
all
of
the
great
people
who
worked
on
this
report
and
to
everyone
in
the
community.
D
A
Thank
you
so
then
the
moment
is
here.
So
I
like
to
take
this
opportunity
to
invite
michael,
michael
well,
there's
a
lot
to
say
about
michael,
but
essentially
he's
our
go
to
experts
on
mata's,
ixp
I'll
do
a
plug-in
for
him.
He
recently
had
a
book.
I
don't
know
whether
it
would
be
visible
with
my
background
on.
Let
me
see
choose
background
to
be
none
and
he
had
this.
The
flip
side
of
free,
which
is
by
michael
kendeigh,
understanding
the
economics
of
the
internet.
I
think
it's
a
really
good
book.
A
I
started
reading
it.
Thank
you,
michael
and
he's
been
a
great
partner
for
all
the
studies
we've
done
and
I
believe
with
you
know,
through
that
collaboration
and
partnership
we've
had
with
him
we've
gotten
to
this
points.
So
without
taking
too
much
more
of
this
of
the
time
we
have
michael
over
to
you
to
take
us
through
what
the
report
summary
and
findings
are.
E
E
So
I
won't
spend
too
much
time
on
this
slide.
It's
been
talked
about,
but
we
have
this
80
20
rule
that
80
of
internet
traffic
would
be
locally
accessible
or
localized.
That
doesn't
mean
it's
all
created
in
africa,
but
it
it
originates
and
terminates
within
a
country
or
region
of
africa,
which
is
very
important
for
the
development
of
the
ecosystem.
As
we'll
see
with
one
of
the
next
slides.
E
The
internet
exchange
points
are
key
to
meeting
this,
and
the
ite
program
has
really
helped
to
realize
this
goal.
As
as
as
machukery
and
others
mentioned,
there
have
been
the
two
studies,
kind
of
2012
and
2020,
looking
at
the
benefits
of
the
exchange
points
in
kenya
and
nigeria.
I'll
summarize,
those
for
results
in
a
moment
and
what
this
study
does
is
really
looks
at
a
number
of
different
levels.
E
E
E
So
it
was
a
really
a
win-win-win
for
the
consumers,
the
isps
and
the
content
providers
that
were
providing
content,
and
you
can
see
on
the
right
some
of
the
benefits
and
how
they
changed
over
the
years
from
2012
to
2020.
In
particular,
the
cost
savings
were
enormous
because
of
the
amount
of
traffic
that
was
able
to
be
exchanged
locally
next
slide,
please,
and
so
in
the
2020
study.
E
E
E
E
The
reason
being
in
some
of
the
data
centers,
where
there's
nodes
of
ixps
there's
a
lot
of
private
traffic
exchange
so
that
wouldn't
be
measured
by
the
peak
traffic
measure,
which
is
a
common
measure
of
the
value
or
effect
of
an
ixp.
So
privately
exchange
traffic
won't
be
measured,
but
all
those
networks
for
some
of
their
traffic
at
least
will
connect
to
the
exchange.
So
it's
still
picked
up
in
that
important
number
of
how
many
networks
are
connected
next
slide.
Please.
E
So
the
number
of
connected
networks
is
determined
by
a
number
of
factors
or
enablers.
The
first
is
in
in
many
countries
the
isps
and
mmos
access
networks
together,
typically
the
foundation.
They
want
to
save
money
by
not
sending
all
of
their
traffic
abroad
just
to
bring
it
back,
exchange
it
at
a
foreign
ixp
and
bring
it
back.
So
typically,
a
group
of
access
networks
will
get
together
form
an
ixp
and
the
more
isps
there
are
in
the
country.
Because
it's
been
liberalized
and
there's
a
lot
of
competition,
the
more
access
networks
there
will
be.
E
So
that's
the
kind
of
the
first
step
is
is
a
liberalized
market,
creates
a
lot
of
access
networks
that
could
connect,
but
that's
really
the
lower
bound
of
the
number
of
members
looking
at
in
kenya,
the
ixp,
the
when
we
counted
had
four
m
o's
and
nine
isps,
and
there
was
a
little
overlap
there
out
of
more
than
50
connected
networks.
So
it
wasn't
just
access
networks,
it's
content
providers,
international
carriers,
governments,
enterprises,
institutions
can
all
connect,
and
so
what
you
need
is
beyond
just
a
lot
of
you
know.
E
Liberalized
telecom
regulations
is
a
content
environment
that
allows
that
helps
to
attract
international
content
providers,
content,
delivery
networks
to
put
their
caches
in
a
country
to
deliver
content
and
the
generation
of
local
content,
users,
governments
and
others,
and
that
should
be
hosted
locally.
That
means
you
need
a
local
data
center,
that's
affordable
and
reliable,
so
that
people
can
put
their
content
locally
and
exchange
it
through
the
ixp
and
often
that's
the
same
place.
E
E
E
E
The
cost
of
the
ixp
is
of,
of
course,
important,
but
if
you
look
at
the
graph
at
the
right,
it's
not
it's
not
as
important
as
you
might
think.
So,
vertically.
You
have
the
annual
connection
cost
of
an
ixp
over
it's
the
annual
cost,
but
over
three
years
to
take
into
account
the
joining
fee.
The
annual
fee
and
the
monthly
port
fee
and
to
the
horizontally
is
the
number
of
connected
networks.
E
Now
you
see
some
like
in
in
south
africa
that
are
free
that
have
a
lot
of
connected
members
which
you
would
expect,
but
there's
a
number
of
them
that
have
very
low
cost,
that
don't
have
that
many
members
on
the
left,
lower
left
and
the
ones
that
are
over
50
other
than
south
africa
and
kenya
and
nigeria
have
annual
fees
that
are
roughly
in
the
middle.
So
there
are
some
fees,
people
are
still
connecting,
and
the
idea
is
that,
no
matter
what
the
cost
is
and
it
should
be
low,
it
should
be
affordable.
E
It's
always
you're
always
going
to
save
money.
In
all
the
cases
we
looked
at,
it's
cheaper
to
connect
to
the
ixp
than
to
send
your
traffic
abroad
to
be
exchanged,
and
the
final
piece
is
to
to
really
focus
on
awareness
and
capacity
building.
That
was
the
focus
of
a
lot
of
the
survey,
questions
and
and
work.
We
did
in
the
deep
dive
countries,
particularly
now
with
lockdowns,
and
it's
not
possible
to
travel
and
bring
people
together.
E
E
So
this
next
slide,
I
think,
is,
is
really
a
highlight
and
something
to
be
remembered
from
this
report,
and
it's
saying
the
following
that,
of
course,
every
new
network
will
add
the
total
add
to
the
total
amount
of
traffic,
because
they'll
bring
traffic
to
the
ixp.
That's
exchanged,
but
what's
interesting
here
is
that
it's
not
just
every
new
network
on
average
increases
the
total
amount
of
traffic
per
per
network.
E
E
The
total
amount
of
traffic
increases
so
much
that
the
average
per
network
increases,
and
that
really
suggests
there's
going
to
be
a
very
strong
correlation
between
the
number
of
connected
networks
in
a
country
or
in
an
ixp
and
the
achievement
of
the
localization
80
20
goal
and
in
fact
the
one
country
that's
at
80
percent
or
more
south
africa
has
well
over
400
connected
networks
at
one
of
its
ixps
and
the
two
other
countries
that
are
at
70
percent.
Nigeria
and
kenya
both
have
over
50
connected
networks.
E
E
E
E
I
won't
write
run
through
all
the
results
on
the
right,
except
for
the
bottom,
one,
that
local
content
is
still
on
the
right
hand,
side
that
still
could
be
much
more
local
content
generated
and
hosted
in
the
country
and
that'll
really
increase
the
number
of
networks
joining
the
ixp
and
the
amount
of
traffic
exchange
there,
as
we
just
saw
next
slide,
please
so.
Finally,
we
did
these
deep
dives
on
on
six
case
study
countries
next
slide.
E
E
So
that
can
be
lowered
with
more
competition,
a
lot
of
discussion
about
increasing
awareness
to
bring
in
more
members.
One
of
the
things
that
we
really
highlighted
was
the
value
of
a
good
website
for
the
ixp.
It
really
gave
a
lot
of
information
and
a
lot
of
links,
particularly
in
these
times
where
meetings
aren't
possible.
Some
of
the
countries
were
very
successful,
bringing
together
members
and
potential
members
for
awareness
and
capacity
building
when
that's
not
possible.
E
The
website
has
to
stand
in
to
some
degree
and
that
can
that
can
be
highlighted
and
improved
in
a
number
of
the
countries
to
really
emphasize
the
value
of
the
ixp
next
slide.
Please
a
note
quickly
on
the
content,
delivery
networks,
they're
very
important
source
of
traffic,
some
of
them
providing
their
own
content
such
as
facebook
and
google
and
others
like
akamai,
are
carrying
third-party
content.
E
The
total
amount
of
traffic
from
that
cdn
into
the
country.
Whether
local
or
international
also
went
up
highlighting
the
value
of
putting
making
the
content
available
locally
lowers
the
latency
increases
the
usage
and
increases
the
total
amount
of
traffic,
so
that
was
quite
a
quite
a
good
finding
and
really
highlighted
the
value
of
localizing
traffic.
Localizing
content.
E
D
E
E
Have
the
connectivity
international
within
the
country
to
bring
the
content
in
and
to
allow
providers,
regional
and
local
providers
to
connect
to
the
ixp
content
regulations
to
attract
the
international
cdns,
generate
more
local
content
and
have
it
available
to
connect
through
the
ixp,
then.
Finally,
the
ixp
policies
that
that
are
liberalized
to
allow
all
of
these
different
members
to
join
the
ixp
and
that's
what
generates
the
members,
the
member
number,
which
generates
the
local
traffic.
E
So
with
that,
I
will
conclude
and
again
thank
you
to
everyone
who
was
involved,
who
sponsored
and
helped
to
to
create
this.
This
report,
which
hopefully,
will
be
very
useful
step
in
looking
forward
at
what
the
next
realistic
goal
is
for
the
traffic
exchange
community
and
how
to
achieve
it.
So
with
that
I'll
turn
it
over
to
you.
Matuki
thanks.
A
A
It
suddenly
gives
a
very
good
picture
of
where
we
are
as
a
continent
and
certainly
also
the
recommendations
on
what
we
really
need
to
do,
and
I
think
I'll
sort
of
just
like
to
pick
up
from
where
you've
the
last
point
which
you've
raised
as
we
go
into
the
q
a
session
and
ask
you
from
an
expert
point
of
view
and,
of
course,
happy
to
hear
what
andrew's
thoughts
are
and
maybe
jane
and
and
how
it
is.
You
know
what
should
the
community
be
thinking
from
the
next
goal?
A
D
I
would
agree
with
michael
more
data
data
data
data
and,
if
you
can't
measure
it,
you
can't
improve
it
as
michael
and
others
have
said,
I'd
say
a
roadmap,
maybe
I'm
not
sure
what
we
could
call
it,
but
another
set
of
goals
and
something
to
aim
at
it's
always
good.
For
coming
back
to
people
and
saying
we
had
this
target,
we
met
these
objectives.
D
B
I
think
that
that
was
all
well
stated.
Of
course,
different
regions
and
countries
will
have
will
be
in
different
stages,
and
one
thing
we've
been
thinking
about
the
past
year
since
cobot
hit
is
the
is
even
more
so
on
the
reliability
aspects
of
of
the
infrastructure.
So
not
only
our
own
infrastructure,
which
of
course
we
can
have
additional
paths,
and
we
can
do
what
we
can
from
an
investment
standpoint
to
support
that.
B
B
So
one
of
our
team's
major
goals
is
to
improve
the
reliability
of
our
services
and
when
thinking
about
the
data,
we
have
data
on
our
own
reliability
of
our
network
infrastructure.
But
it's
hard
to
tell
if
we're
sharing
the
same
fallback
paths
as
the
wider
industry.
There's
not
really
a
data
source
for
that.
We
know
that
from
relationships
and
talking
to
people
and-
and
we
think
of
that,
as
as
a
big
risk,
not
only
from
just
a
pure,
how
many
paths
are
there,
you
know
perspective
who's
controlling
those.
B
If
this
is
if
it's
just
one
or
two
players,
is
that
an
issue
or
not,
but
also,
I
think,
about
the
interconnection
of
the
continent,
and
that
will
help
us
so
much
in
terms
of
thinking
about
africa
as
a
as
a
as
an
internet
that
stands
on
its
own
does
not
rely
on
europe
or
other
regions,
and
I
think
that's
the
next
step
when
thinking
about
okay
do
we
need
these
international
links?
Well,
we
shouldn't
the
internet
is
already
already
here.
B
A
A
A
A
C
C
A
As
a
result,
it
means
that
users
are
not
benefiting
from
the
local
content
or
the
traffic
that's
actually
locally
available,
because
with
vpns
they're
now
appearing
to
be
in
other
locations.
So
how
do
we
make
interventions
to
block
this
from
undoing
all
the
hard
work
that's
been
put
in
place
to
achieve
the
80.
A
B
I
think
also
there
there's
a
lot
in
what
ben
said
here.
I
think
there's
some
political
questions
there
in
terms
of
what
is
what
is
your
political
risk
strategy
for
my
network
infrastructure
perspective,
and
that
means,
if,
if
all,
if
all
of
your
infrastructure
is
in
one
particular
market,
that
does
represent
a
potential
risk.
B
If
that,
if
that
market
decides
that
you
know
they're
going
to
control
or
adjust,
you
know
what
they're
doing
from
a
dpi
perspective,
but
also,
if
we
think
about
what's
happened
in
other
markets
from
a
vpn
perspective,
vpn
providers
also
have
infrastructure
within
various
locations.
So
there's
no
reason
why
these
vpn
providers
can't
build
out
infrastructure
into
africa
in
the
future.
Maybe
today,
that's
too
cost
prohibitive
or
different
markets.
B
A
E
But
I
think
politicians,
world
over
are
never
really
that
good
at
second
order
effects,
and
the
second
order
effect
in
this
case
is
that
the
data
localization
probably
isn't
going
to
meet
any
of
the
goals
that
you
want.
It'll
lower
the
amount
of
traffic-
that's
in
the
country,
because
it
requires
a
significant
investment
in
every
country
that
has
those
requirements,
and
so
some
companies
may
just
choose
not
to
invest
in
the
in
the
country.
E
If
the
market
doesn't
look
like
it's
going
to
get
big
enough
because
they
don't
want
to
reinvest
the
same
everywhere,
the
value
of
data,
the
value
a
lot
of
it
is
to
bring
it
together
to
be
able
to
analyze
it
and
then
learn
from
it
and
get
economies
of
scale.
So
those
aren't
going
to
be
met.
It
may
not
be
more
secure
than
having
it
in
a
cloud.
You
know
very
secure
cloud
network
because
everyone
knows
where
all
the
traffic
the
data
is
if
it
has
to
be
localized.
E
But
again
it's
it's
a
hard
message,
because
you
know
everyone
likes
sovereignty,
that's
what
nations
are
and
and
when
they're
told
the
value
of
data
and
and
local
traffic
exchange.
It
seems
like
a
good
solution.
So
that's
a
great
question
and
it's
it's
a
challenging
one,
but
I
think
one
has
to
just
go
back
after
every
point
of
what
the
justification
is
and
show
that
it's
not
going
to
likely
meet
any
of
them.
D
Yeah-
and
I
would
say
we
have
to
separate
the
issues
of
infrastructure,
build
out
from
those
issues
of
what
happens
once
you
have
that
content.
I
appreciate
the
digital
sovereignty
issue
greatly,
as
others
have
said,
but
we
know
that
more
infrastructure
and
localized
infrastructure
leads
to
better
socioeconomic
development
and
to
andrew's
point
it
means
a
lot
of
different
things
from
the
policy
regulatory
side
too.
But
if
we
can
separate
the
let's
build
infrastructure
from
what
do
we
do
once
it's
built?
We
really
do
need
more
infrastructure
built.
A
Okay,
thank
you.
So
we
are
suddenly
running
out
of
time,
and
I
see
we
still
have
a
few
more
questions
to
answer
and
so
in
the
interest
of
time.
What
we'll
do
is
we'll
take
those
questions
and
we'll
try
and
respond
to
them
directly
and
at
least
get
the
expert
to
respond
to
you.
So
next
up,
we
suddenly
just
want
to
make
sure
that
everybody
gets
an
opportunity
to
live
here
with
the
report.
A
A
I
think
there's
a
great
question
that
came
in
from
ben
roberts
with
respect
to
are
there
people
making
decisions
that
could
be
undoing
all
the
efforts
that
are
happening,
and
so
we
like
to
invite
all
of
you
to
join
us
at
the
next
conversation
platforms,
where
we
can
have
discussions
of
follow-on
conversations,
the
central
asia,
peering
forum,
virtual
pairing
series.
Sorry
that's
taking
place
on
the
13th
of
july.
A
A
We
are
also
looking
for
program
committee
members
to
help
join
the
afp
program
committee
and
again
that
will
certainly
help
boost
the
conversations
with
the
expertise
from
the
community,
so
last
and
not
least,
is
to
give
you
the
opportunity
to
take
a
snapshot
of
that
qr
code.
That
is
where
we
have
the
report
and
as
you're
doing
so,
I
would
like
to
take
this
opportunity
to
thank
all
the
people
who
have
really
played
their
role
in
supporting
the
the
reproduction
of
this
report.
A
Suddenly
start
with
thanking
michael
for
all
his
tireless
effort
and
working
with
us
and
agreeing
to
work
with
us
again,
one
more
time
to
produce
this
report
to
write
this
report,
and
then
our
partners,
facebook,
for
certainly
having
the
faith
in
us
that
we
can
be
able
to
pull
this
through.
We
got
a
lot
of
support
from
fix
and
the
internet
exchange
points
in
africa
and
all
those
individuals
who
actually
contributed
and
took
their
time
to
participate
in
grueling
interviews
in
this
time
of
the
pandemic.
A
A
I
also
like
to
thank
the
isoc
leadership,
jane
darwitz,
the
ceo
for
really
approving
that
this
can
actually
get
done.
And,
of
course,
we
have
a
strong
team
at
the
internet
society
led
by
a
comms
person,
israel
and
gila
massema
ivana,
a
couple
of
others
from
the
comms
team,
who
really
played
an
important
role
to
make
sure
that
the
report
got
to
this
stage.
My
heart's
off
to
you.
It
was
not
an
easy
task
to
pull
through.
A
So
thank
you
for
all
your
teamwork
collaboration
and
really
to
all
our
partners
out
there.
Thank
you
again.
We
always
say
that
the
work
of
ixp
development
and
interconnection
is
about
partners,
and
we
are
certainly
here
because
of
all
the
collaboration
and
the
partnership.
We've
had
I'll,
also
like
to
mention
that
we
have
the
french
version
coming
out.
So
the
french,
the
interpreted
version
thanks
to
our
colleagues,
ma
sema
and
gila,
for
making
sure
that
this
is
done,
and
so
we
should
be
actually
having
that
ready
before
the
end
of
this
week.
A
So
the
english
is
ready.
The
french
will
be
coming
out
shortly.
So
please
pay
attention
to
this
space
and
you'll
be
able
to
go
to
the
website
and
download
it,
and
we're
also
inviting
you
to
join
us
in
those
future
forums
where
we
can
have
that
further
discussion
and
more
detailed
conversations
around
the
the
findings
of
this
report
and
what
we
should
be
thinking
about
as
the
next
future
goal
for
the
80
20
goal
initiative.
So
thank
you
very
much
for
all
your
time.
A
We're
six
minutes
over
time,
and
we
do
appreciate
you
taking
time
to
join
us
today
and
look
forward
to
the
conversations
going
online
in
the
social
media
platform
and
all
other
spaces
available.
Thank
you.
Do
have
a
good
rest
of
your
day
wherever
you
are
and
thank
you
again
for
all
the
support
and
collaboration
you
next
time.
Thank.