►
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
Hello,
my
name
is
mahmoud,
I'm
from
fleek,
I'm
here
with
my
team
today,
and
we
wanted
to
tell
you
the
state
of
fleek
and
what's
next
for
us,
so
I
I
know
many
of
you
have.
You
know
used
fleek
before.
Hopefully
you
all
love
it
like
we
do
and
like
we
provide
sort
of
like
a
few
things
that
are
pretty
useful
in
web3,
which
are
like
web
3
hosting
for
websites
like
jam
stack
websites.
A
I
guess
in
2021,
so
basically
like
approximately
it's
like
90
is
like
webs
sorry
gaming
as
well
as
storage,
and
then
the
10
involves
websites
and
everything
else,
so
yeah
so
like
what's
next
for
us
like
once
once
we
saw
this
data,
we
wanted
to
do
something
about
it,
so
yeah
brace
yourself.
We
have
a
lot
in
the
pipeline.
Basically,
and
we
want
to
go
on
to
this.
A
Endeavor
to
you
know,
improve
our
ipfs
content,
retrieval
our
our
hosting
services
and,
like
obviously,
we
want
to
add
all
the
cool
features
that
you
would
get
with.
Ipfs,
hosting
and
storage.
Like
you
know,
image
optimization,
you
know,
persistence
with
file
coin.
All
that
cool
stuff
like
serverless
functions,
but
beside
that
we
wanted
to
focus
on
like
two
main
things,
which
is
like
upload
speed
of
content
as
well
as
retrieval,
as
well
as
increase
availability
of
our
services
globally
and
sort
of
yeah
on
a
global
scale.
A
Yeah,
that's
that's
pretty
much
like
what
motivated
us
is.
We
saw
that
stat.
We
wanted
to
do
something
about
it
and
now
we're
here
and
the
product
we
want
to
build
now
is
like
to
unify
the
entire
suite
of
things
to
kind
of
push
us
towards
a
more
interoperable
and
unified
system
that
can
integrate
and
use
other
platforms
like
filecoin.
We
use
ipfs
dns
everything
yeah.
So
let
me
see
so
starting
this
year.
A
A
Sorry
between
web
2
and
web
3
is
dwarf
and
event.
It's
like
a
temporary
layer.
That's
going
to
help
us
improve.
You
know.
Latency
is
going
to
improve
upload
speeds
for
our
content.
It's
going
to
help
us
for
retrievals
yeah
and
the
the
main
idea
in
dorb.
Is
it
when
we
want
it
to
be
non-custodial
so
you'd
like
we
want
to
control
less
of
the
platform,
so
we
want
to
give
back
freedom
to
our
customers
and
the
users
end
users,
which
obviously
means
less
problems
for
us
less
take
down
notices.
A
So
we
don't
have
to
worry
about
that
and
yeah
and,
like
I
said
it
is
a
temporary
layer
but
we're
working
on
something
like
I
said,
it's
our
shortcut
to
web3
and
it's
only
half
of
our
plan,
the
next
plan
or
the
next
half
of
this
plan
is
obviously
like
this
new
product.
It's
called
fleeq
network,
aka,
orsa,
basically
and
yeah.
We
want
to
create
a
like
a
content,
delivery
network,
a
decentralized
content,
delivery
network
and
basically
it's.
A
We
want
to
use
everything
that
we've
learned
over
the
past
year
and
everybody
on
the
team's
experience
with
lip
b2b,
with
decentralized
hosting,
with
running
our
own
gateway,
to
create
sort
of
a
decentralized
content,
delivery
network,
that's
decentralized,
eventually,
meaning
that
anybody
can
run
a
node.
Everybody
can
participate,
you
can
scale,
meaning
like
you,
can
just
run
end
nodes
and
increase
capacity
or
cache
capacity
of
the
network
yeah
and
like
web3
content,
meaning
that
everything
is
represented
as
an
ipld.
A
A
So
if
one
node
goes
down
and
you're
streaming
data
from
them,
you
can
we
can
switch
the
streaming
from
one
node
to
another
and
yeah
and
and
it's
b2b,
so
anybody
can
run
it
and
yeah.
The
architecture
is
entirely
based
on
limp
b2b
like
right
here.
So
we
have
we've
set
some
variables
that
we're
not
too
sure
about
yet
so
we're
still
researching
and
we're
still
obviously
running
a
lot
of
tests
to
make
sure
that
you
know
we
can.
A
We
can
propagate
messages
in
the
network
pretty
quickly
and
yeah,
so
it's
completely
based
on
lippy,
tippy
and
yeah.
So
the
first
thing
in
here
is
the
governance
layer
which,
like
I
said,
we're
working
on
this
like
warping,
meaning
that
we
want
to
go
from
web
2
to
web3
and
to
get
there.
We
need
to
use
some
centralized
pieces
initially,
and
this
is
kind
of
like
this
line,
that
in
the
middle
we
have
some
edge
nodes
which
you
can
call
them
cache.
A
And
then
you
have
this
like
governance
layer
right
now,
which
we
want
to
replace
it
with
a
blockchain,
eventually
like
a
bfd
or
a
hot
stuff,
governance
or
consensus,
and
all
it
does
right
now
is
as
a
node.
You
register
tls
dns
and
you
register
with
the
governance
layer
right
now
and
you
we,
you
know
we
have
a
list
of
all
the
nodes
and
then
we
can
cherry
pick,
which
is,
is,
you
know,
closest
to
the
end
user
and
we
can
serve
them
requests
and
yeah.
A
That's
kind
of
the
first
I'm
going
to
give
like
a
brief
talk
over
flick
network.
We're
going
to
announce
student
starts.
I
just
want
to
give
we're
going
to
give
more
information
later
and
then
we
have
a
gateway
which
is
like
you
can
think
of
it
as
an
ipf
as
gateway,
and
this
is
the
v0
of
the
ipfs
gateway
is
not
obvious
for
fleek
network
gateway,
it's
similar
to
ipfs
gateway
with
a
load
balancer,
and
then
it
connects
to
the
to
the
node
right.
So
just
a
simple
gateway.
A
Actually
let
me
talk
about
that.
A
little
bit
is
like
for.
After
v0,
we
want
to
improve
the
gateway
as
in
instead
of
going
to
the
gateway
you
can
go
to
directly
to
the
node
and
you
can
stream
data
and
make
sure
it's.
You
know
you
can
verify
incrementally
using
graph
sync
or
similar
using
a
cloud
for
cloud
figure,
worker
or
dwarf
yeah.
So
in
content
routing,
we
decided
to
use
a
dhd
slash
gossip
approach.
A
A
We've
gone
from
like
sort
of
a
dht
approach
to
a
gossip,
dht
and
and
then
we
have
indexers,
which
this
week
has
been
amazing,
because
everybody's
talking
about
indexers-
and
I
just
put
this
slide
in
this
week-
we
were
talking
about
indexes,
but
I
wasn't
sure
which
part
of
our
network
we
should
introduce
this
into
because
lotus
has
it
in
their
nodes
and
yeah.
It's
basically,
you
can
think
of
clients
requesting
a
great
ways,
and
then
this
big
circle
right
here
is
a
node
that
has
an
nginx
proxy
sitting
in
front
of
it.
A
A
So
if
the
indexing
doesn't
work,
we
have
to
go
to
our
content
routing
and
if
the
cache
node
doesn't
have
the
dent
node,
then
we
have
to
go
to
origin
which
is
filecoin
or
similarly,
then
we
have
to
get
the
data
store,
it
locally,
cache
it
and
on
a
repeated
request,
you
should
have
the
data
on
the
node
and
that's
how
the
indexing
work
and
we
should
announce
technically
we're,
not
storage
providers.
So
we
shouldn't
announce
to
an
indexer,
but
at
the
same
time
maybe
we
can
announce
and
it'll
improve
that
availability
and
then
pre-discovery.
A
A
A
But
with
all
of
that,
like
I'm
giving
a
brief,
we
have
an
entire
paper
that
we
wrote
it's
only
an
introductory
paper,
because
it
does
involve
a
consensus
on
a
blockchain.
We
wanted
to
write
a
paper,
I'm
I'm
gonna
link
it.
I
I
didn't
actually
have
it
here,
but
I'm
going
to
link
that
paper
and
it
talks
more
into
detail
about
everything
and
yeah.
So
like
it's,
not
all
rainbows
like
there's
a
lot
of
trade-offs,
meaning
that
you
know
doing
something
or
having
the
ability
to
do
something.
A
It
might
does
not
mean
it's
it's
practical
to
do
and
you
can
have
a
lot
of
attacks
like
sybil
or
eclipse
attacks.
We've
even
like
looked
at
some
papers
when
it
comes
to
like
boot,
boot
attacks,
meaning
with
ethereum.
For
example,
some
nodes
can
alter
the
the
view
of
the
nodes
connecting
to
the
network.
A
Initially,
and
they
can
they
can
eclipse
you
to
prevent
you
from
finding
honest
nodes.
So
like
it's
a
big
thing
for
us
too,
and
for
us
we're
trying
to
solve
it
using
a
consensus
layer
stake,
maybe
for
sybil
it's
more
about,
I
mean
for
eclipse,
it's
like
also
ip
rate
limiting
and
things
like
that
yeah
for
sybil,
it's
like
yeah
proof
of
stake.
We
need
to
stick
like
a
number
of
tokens
or
something.
A
Similarly,
if
you
have
any
feedback,
please
let
me
know
we're
open
to
feedback
and
we're
available,
and
so
yeah
we've
we're
actually
almost
done
with
mission
number
one
which
is
really
cool.
Recently
we,
like
I
said
we
started
this
three
three
four
months
ago.
We've
done
about
a
bunch
of
research.
We
want
to
choose
which
tools
are
pretty
cool.
A
We
found
lip
b2b
and
we
knew
about
lip2b
and
we're
almost
done
with
sort
of
mission
number
one
which
is
like
we're,
calling
it
first
exploration,
which
is
outlining
the
the
v0
of
the
spec
doing
some
cool
stuff
and
hopefully
we're
going
to
run
some
nodes
globally
and
have
a
test
net.
Once
we
announce
and
yeah,
we
have
other
missions,
mission,
number
n,
which
are
like
governance
based
falcoin
and
like
hot
stuff.
A
If
we're
going
to
introduce
some
consensus
layer,
we
also
want
to
expand
a
lot
like
scale
out
massively
and,
like
I
said,
intro
like
integrate
with
other
retrieval
markets
or
retrieve
clients,
and
we
want
to
see
how
we
want
to
do
that
and
that's
why
we're
here.
We
want
to
talk
to
everyone.
So
if
you
want
to
talk
to
me
or
the
team,
please
do
I
want
to
talk
to
you
client
verification
workers.
Eventually,
we
want
to
also
like
run
again
client
clients.
A
In
I
mean
we
want
clients
to
just
directly
trustlessly
connect
to
the
cache
node,
and
you
know
stream
data
back
incrementally
verify
in
graph
sync:
it's
not
user
experience
friendly,
but
we
we
think
that
sas
services
are
going
to
improve
that,
and
then
we
have
optimization
and
fleet
web
3,
which
is
like
this
entire.
We
want
to
revamp
our
entire
fleet
co,
which
is
different
than
fleek
network
fleet.co
is
going
to
be
revamped.
A
It's
going
to
have
so
many
cool
things
now
you
know
with
team
management,
image,
optimization
and
all
that
cool
stuff,
and
then
the
last
one
is
kind
of
like
my
and
the
teams,
like
the
rest
of
the
team's
interest
now,
which
is
trying
to
do
something
with
zero
knowledge,
proofs
and
compute.
So
we've
heard
some
talks
this
week
about
compute
over
ips
rpfs,
but
we
want
to
do
like
serverless
compute
over
ipfs
and
yeah.
A
Like
I
have
some
open
questions,
you
can
ask
me
questions
or
ask
the
team
questions
and
yeah
there's
a
lot
more
stuff.
You
want
to
do
like.
We
can
have
a
lot
of
data
ingested
into
our
network
and
we're
going
to
provide
a
lot
of
data
using
the
cdn.
So
we
want
to
even
explore
using
an
fvm
for
content,
analysis
plus
some
ml,
and
can
we
run
a
cairo
vm
on
top
of
fvm?
I
guess
you
have
to
ask
other
or
you
guys
and
then
there's
an
open
question
about
lurk.
A
It's
like
a
recent
thing
that
was
like
provided
in
april
again
and
yeah.
I
appreciate
you
thanks
for
listening.
We
plan
to
announce
more
information
in
the
next
couple
months,
we're
thrilled
to
be
here
and
yeah.
We
can
all
build
a
better
retrieval
market
at
open
source
feedback
appreciate
it.
Thank
you.