►
From YouTube: Core Dev Meeting #5 - October 1st, 2018
Description
A
All
right,
it
should
be
live
on
YouTube,
welcome
everyone
to
quarter
meetings
this
time,
we're
doing
like
it
was
later
agenda
based
on
sort
of
voting
with
the
voting
gap,
just
in
order
to
get
three
to
five
copies
that
we
can
talk
about
more
in
depth,
I
suppose,
having
ten
things
that
we
talk
about
so
with
that
said,
I
guess
we
can
get
started
with
the
first
topic,
which
is
rethinking
mail
server
version
Adam.
Do
you
want
to
take
away.
B
Yeah
sure
so
just
just
bring
back
what
we.
What
we're
talking
in.
On
the
last
a
call.
There
was
discussion
about
log
based
communication
and
moving
off
I'm
ascending
to
Pierre,
to
ipfs
or
swarm
in
order
to
decentralize
the
storage
and
in
order
to
get
rid
of
my
services
as
they
are
now
and
yeah.
So
the
other
version
of
that
that
I've
been
thinking
of
is
to
kind
of
reward.
B
B
B
B
B
Yeah
I
just
wanted
to
share
this
this
vision-
maybe
it's
not
necessarily
any
concrete
example
or
lakes
for
specific,
like
an
any
specific
specification,
and
that
describes
is
in
details,
but
just
like
a
vision,
so
so
the
other.
The
other
important
thing
here
is
that
this
history,
even
if
we
would
create
this
whisper
history,
that
would
be
like
four
camels.
B
So,
for
example,
if
from
one
point
of
time
someone
would
like
to
be
independent
from
us
why
we
would
run
the
servers
and
someone
would
like
to
also
like
forked,
pretty
much
so
run
their
offline
history
and
use
their
services
to
access
this
somehow,
because
they
any
reason
they
could
do
it
and
they
could
point
to
to
our
whisperer
history.
So
they
would
not
start
from
scratch.
They
wouldn't
they
did,
would
still
be
able
to
access
the
history
before
the
Iran
service.
So
that's
also
the
important
part
of
this
idea,
but
yeah
so
yeah.
B
That's
pretty
much
the
description
of
that
as
I
said.
Hopefully,
this
week
I
can
start,
and
maybe
someone
else
would
like
to
help
me
to
write
a
read
more
details
about
it
and
get
into
details
of
of
IP
fsm
swarm
of
the
features
that
they
have.
That
could
help
with
that.
With
this
solution,
yeah,
that's
pretty
much.
If
you
have
any
questions,
let
me
know,
but
I
I'll
try
to
be
more
specific.
This
week.
B
I
think
that
this
is
this
is
really
connected
with
our
cluster,
so
the
centralizing
our
cluster
and
removing
our
cluster
as
being
like
a
single
point
of
failure,
because
right
now
my
servers
pretty
much.
This
is
the
this
is
the
they
were
even
worse
than
whisper
notes,
because
whisper
notes,
it's
very
easy
right
now
to
run
them
even
when
we
had
a
call
on
Friday
with
guys
from
top
note,
and
they
showed
us
how
this
solution
works.
B
You
know
if
the
note
does
not
need
to
have
an
in
disk
space
and
then
any
stuff
like
that
and
and
and
whisper
notes,
can
also
use
this
discovery.
Peers
discovery,
protocol
easily
that's
fairly
easy
to
decentralize
and
actually
get
rid
of
whisper
notes
from
our
from
our
server.
Even
we
could
use
subtle
desktop
for
that,
but
with
my
server
is
much
harder
because
you
need
to
have
these
services.
This
nodes,
like
like,
like
a
chief,
pretty
high
uptime,
so
I,
believe
that
this
is.
This.
Is
the
solution
for
for
that
for
yeah.
C
B
I
think
that
that's
that
still
will
be
possible,
but
you
just
need
to
make
a
little
more
requests,
so
it
wouldn't
be
enough
to
just
make
a
single
request
for
IP.
If
I
just
wanted
and
say,
I
can
get
me
any
messages
from
the
last
day.
You
would
need
to
get
like
kind
of
index
tree
figured
out
which
chunks
you
need
to
load
and
then
make
a
couple
of
requests
to
get
the
chance
and
that
you
need
to
to
get
offline
messages
for
the
chart
public
chance.
That
was
private
chance
that
you
are
part
of.
B
C
I've
got
a
question
and
so
I'm
interested,
and
this
is
a
slightly
off-topic.
You
know
if
it
is,
but
will
this
type
of
architecture
facilitate
a
UX
where
I
mean?
Essentially,
someone
could
log
in
to
a
different
client
but
a
client
that
works
on
top
of
the
status
network
and
log
in
with
their
a
contact
record,
or
this
would
be
their
a
key
and
be
able
to
retrieve
all
the
contacts
and
chats
but
on
our
completely
different
lines.
I.
B
Mean
the
ipfs
or
swarm
can
definitely
be
used
for
that
I
guess
we
already.
There
was
already
at
such
an
idea
where
we
were
talking
about
storing
the
contacts
and
other
things
on
IPSS
or
swarm,
and
so
that
you
could
easily
retrieve
it
on
the
new
device.
Yeah
I
mean
definitely
I.
Think
that
whisper
I
think
that
swarm
or
IP
is
ipfs
are
technologies
that
can
be
used
for
that
and
I
don't
see
any
other
alternatives
right
now,
but
yet
something
some
centralized
search
can
definitely
be
an
answer
here.
It.
C
A
A
C
A
B
A
How
do
you
guys
think
at
this
place
into
I
guess
maybe
a
federal
your
point
of
view
in
terms
of
the
push
notification,
work
and
sort
of
having
either
as
a
relay
when
one
idea
was
that
it
should
be
the
same
node
as
mail
service?
But
if
we
remove
this
special
node,
yeah
I,
guess
how
do
people
look
at
that
I
curious
to
see
Paris
here?
That
is
point
of
view.
E
A
I
guess
I
mean
from
from
the
white
paper,
there's
like
a
specific
use
case
that
def
is
they
the
reason?
The
way
we
implement
the
mail
server?
Is
that
it's
for
offline
in
boxing
and
push
notifications,
but
right
now
the
way
position
if
he
gives
his
work
is
this
tightly
coupled
to
a
specific
provider,
and
this
is
one
to
one
person
so
person
a
has
direct
address
to
send
a
trigger
push
the
variation
at
first
and
be.
A
But
that
means
that
there's
no
sort
of
spam
protection-
and
you
have
other
causes
address
and
you
can't
change
personification
provider
and
it
doesn't
support
sort
of
public
chat
notification
and
so
on
and
one
way
of
solving
that
and
also
creating
some
kind
of
economics
around
it,
as
described
in
a
white
paper.
It's
to
have
a
level
of
indirection
where
the
level
of
indirection
is
to
delegate
to
specific
trusted
note
which
will
be
the
same
as
the
mail
server
node.
And
then
you
have
some
registration
logic,
so
you
register.
E
Yeah
the
the
way
that
where
the
work
was
left
with
push
notifications
v2,
there
was
still
the
needs
to
communicate.
You
know
which
mail
server.
So
when
you
start
with
a
one
to
one
communication
with
someone,
you
would
inform
him
which
mail
server
you
were
using
so
that
he
couldn't
that
person
could
send
you
messages
and
ask
for
a
push
notification
to
be
sent
to
you.
E
B
From
my
perspective,
even
if
we
would
want
it
to
like
couple
my
servers
and
push
notifications,
it
would
be
there
I
guess
to
figure
out
something
that
it's
it's
more
like
a
protocol,
how
this
communication
can
be
established
and
making
independent
from
my
servers.
That
would
be
idea.
I'm,
not
sure
yeah
I,
also
like
haven't
really
been
involved
into
designing
certification,
because
so
I
don't
really
know
the
details.
B
But
it
looks
like
from
at
least
from,
like
the
you
know,
my
perspective,
that
it
would
be
much
better
to
make
these
things
independent,
because
if
we
just
couple
them
with
any
decision
regarding
or
another
is
gonna,
you
know
change
the
way
the
other
Fink
works.
So
just
like
from
a
kind
of
engineering
perspective,
do
we
better
to
make
those
things
independent
and
kind
of,
even
if
we
need
some
communication,
then
we'd
better
to
just
do
this
on
a
protocol
level
rather
than
kind
of
coupling
it
together.
D
Yeah
I
thought
that's.
The
perfect
notifications
for,
like
the
centralized
infrastructure
would
be
that
the
client
itself
decides
if
it
shows
notifications
or
not.
So
it's
essentially
like
scraping
the
firebase
and
everything
and
just
using
classifications.
The
client
is
responsible
to
receiving
a
whisper
message
that
somehow
in
the
background
and
then
if
it's
like,
if
it
needs
to
then
just
show
local
notification,
and
this
will
also
solve
the
problem.
That's
all
notification.
Is
there
but
not
whisper
message,
but
that
will
require
some
rework
on
how
we
receive
messages,
the
background
and
whisper
messages.
D
E
D
E
There
is
a
github
issue
from
Eric
that
details
how
this
should
work,
which
is
basically
just
get
a
firebase
notification
to
trigger
the
logic,
to
fetch
messages,
and
then
it
can
populate
the
push
notification
with
further
detail
once
it
cuts
the
messages
from
whisper.
So
initially
it
just
says
you
receive
the
message.
Then
the
app
starts
in
the
background
and
populates
the
the
actual
push
notification
and
chose
it's.
D
E
Can
you
hear
me
so
I,
just
I
just
wanted
to
raise
the
like
the
brief
discussion
about.
Do
we
need,
at
this
point,
push
notification
at
all
if,
if
we
target
Android
platform
as
our
primary
platform
for
mobile
and
we
have
status
desk,
though
we
don't
have
the
problem
that
push
notifications
solve
right,
we
mainly
need
it
for
iOS
users
who
can
do
any
background
things.
E
So
we
rely
on
the
Apple
or
some
other
external
mechanism
to
deliver
messages,
and
if
we
can
solve
it,
maybe
not
efficiently,
but
we
can
solve
it
for
Android
for
now.
Without
you
know,
building
the
whole
new
stack
and
complexity
into
the
app
wouldn't,
be
it
a
better
approach
for
now.
What
do
you
think
I.
D
A
D
B
Yeah,
so
there
are
two
things
here,
just
something
to
like:
informative,
more
like
informative
thing
to
do
so.
Basically,
what
we
did
recently
was
that
we're
gonna
remove
our
patches
from
status.
Go
to
our
for
work,
go
atrium
and
yeah
I.
Just
right
now,
I
go
a
trim
is
used
in
with
Deb
just
like
regular
dependency,
and
you
don't
need
to
patch
the
vendor
or
directory
in
status.
Go
repository.
So
anyone
who
has
been
working
on
that
and
remember
this.
It's.
G
B
It's
it's
simplified
it
just
you
download
it.
You
update
dependencies
just
like
you
would
use
that
without
any
additional
Patras.
Anything
like
that
and
yeah
they
do.
One
more
thing
is
that
the
pro
request
is
pending,
but
Demitri
work
on
moving
whisper
from
go
at
room
to
status,
go
repository,
so
it's
gonna
be
synced
with
the
upstream,
but
pretty
much.
B
The
reason
was
that
we
just
were
growing
with
patches
for
whisper
like
significantly
and
also
dimitri
recently
has
been
working
on
a
writing
meeting
thing
in
whisper,
and
you
know
that
hood
that
that's
a
little
more
code
than
we
used
to
do
with
patches
and
it
just
was.
It
was
just
too
much
patches
to
maintain.
So
the
decision
was
that
decision
was
to
just
move
whisper
to
a
separate
package,
and
that
was
go
and
that's
it.
So,
just
just
an
affirmative
thing
for
everyone,
who's
who
has
been
working
on
status,
go
and
look.
D
Yeah
and
I
think
that
it
with
for
whisper,
it
makes
sense,
because
we
also
want
to
do
this
whisper
gateway
thing
yeah,
which
will
also
require
some
like
refactorings
of
what
whisper
dust
and
later,
we
can,
of
course,
up
stream.
It
and
I
think
that
the
idea
is
that,
since
there
are
external
companies
that
are
interested
in
that,
so
we
won't
want
to
up
students,
but
for
now
it
probably
would
be
much
easier
to
just
implement
it
in
status,
go
and
then
just
submit
a
badge
to
whisper.
Instead
of
making
a
cash
issue,
mm-hmm.
C
D
E
D
D
B
C
B
You
would
like
to
test
it
and
it's
it's
just
much
easier
right
now,
when,
when
we
have
a
the
packet
whisperer
package
in
the
repository
instead
of
you
know,
I'm
going
through
the
whole
patching
phase,
and
you
know
if
you
decide
to
remove
it
and
you
submit
it
or
other
patches.
On
top
of
that,
then
it's
a
problem
and
so
on
so
forth.
So
and
generally
it
is
to
simplify
the
whole
process
of
experimenting
and
but
yeah
it's
in
terms
of
Sigma
submitting
to
something
chancers
to
the
upstream.
It's
just
gonna
be
the
same.
B
D
B
D
Perfect
sense,
I'm,
certainly
different
generations.
So
how
do
we,
who
will
be
responsible
for
keeping
the
our
fork
up
to
date?
That's
like
free
base
in
all
the
patches
and
things
like
that
because
before
it
was
more
obvious
because
he
justs
update
the
upgrade
process
was
more
obvious
and
streamlined
and
it
was
obvious
which
patches
are
a
quiet
which
is
which
are
not
applied,
and
things
like
that
now,
it's
of
course
it's
faster
to
develop
new
features
and
that's
what
we
want,
but
it
will
be
a
bit
more
cumbersome
process
of
upgrading.
B
That's
a
good
point.
This
knows
any
specifics
about
it.
We
can
definitely
I
just
put
an
out
and
work
around
it,
but
yeah
I
mean
it.
Definitely
it
just
need
to
see.
We
need
to
figure
out
something
some
way
to
do
that.
We
do
not
have
any
process
like
official
want
to
upgrade,
get
we're
just
looking.
We
just
look
at
the
at
the
newest
version
and
if
it
appears
there,
we
just
upgraded
now
I
think
it
might
be
better
similar.
B
Just
that
the
problem,
the
small
problem
is
that
I
mean
we
can
we
can
like
make
it
another
point
when
you
upgrade
get.
You
just
need
to
look
at
that
for
request
the
twirler
marriage
in
the
whisper
package,
maybe
and
just
move
them
to
to
our
package.
If
that
makes
sense,
that
could
be
the
white
dude
but
yeah.
Definitely
we
need
to
yeah.
We
need
to
kind
of
document
that
so
it's
clear
how
it
happens,
but
yeah.
D
B
D
B
B
D
I,
just
was
I'm
pretty
sure
with
my
mom
like
like
forget
features
but
I'm
more
like
things
like.
Can
we
get
some
raise
condition
that
we
fix
locally
and
we
waiting
for
it
to
be
merged
upstream,
but
it's
still
open,
and
then
we
upgrade
to
the
media
and
we
want
like
it
would
be
good
to
make
sure
that
this
page
still.
D
A
B
I
think
that,
like
some
more
like
formal
description
of
the
solution
would
be
the
first,
because
what
I
just
described
today,
what
was
like
a
kind
of
a
vision
rather
than
a
specific
plan,
so
you
know
I
imagine
there.
Okay,
we
include
this
way.
It's
gonna
work,
but
yeah.
Definitely
something
more
concrete.
F
Yeah
I
will
discuss
it
a
bit
so
well,
I
guess
we
already
touched
a
little
bit
on
the
subject
with
the
previous
discussion
around
like
med
service
and
push
notification,
but
essentially
the
ILA
values
case
we
have
and
it's
a
new
scale.
We
have
been
discussing
with
a
lot
of
different
deaf
developers,
so
it's
pretty
important
and
interesting.
F
So,
the
the
the
idea
is
to
have
some
some
way
for
any
external
process
to
send
a
message
and
with
a
push
notification.
So
that's
based
on
some
discussion
between
ego
and
andrey
that
they
are
during
the
di
cottan
at
a
spelling,
and
the
current
option
we
are
considering
is
to
create
a
centralized
gets
away
to
to
achieve
that,
so
a
little
bit
similar
to
what
you
can
do
with
infra.
So
it's
very
not
like
the
best
solution
in
that
we
will
consider
in
the
future,
but
at
least
I'll
shorten.
F
F
Like
the
new
version
of
good
push
notification,
we
are
working
on
quickly,
add
red
dots
but
yeah.
That's
kind
of
the
the
current
state
of
the
D
ID
that
we
have
so
I
think
that
already
shared
discuss.
There
is
a
discuss
around
that
and
some
people
already
discussed
well
and
share
their
opinion
about
it,
but
maybe
just
for
the
sake
of
sharing
with
all
the
developers,
if
either
ego
or
Pedro
could
could
comment
on
that.
I
would
be
interested
in
hearing
what
they
think
about
it.
D
D
So
for
that
for
the
certifications,
I
think
it's
we
don't
need
to
share.
I
mean
I'm
opposed
sharing
any
kind
of
tokens
with
any
apps
and
I.
Think
if
we
have
for
this
phase,
if
we're
talking
about
any
kind
of
centralized
HTTP
like
service,
that
will
stand
whisper
messages,
then,
like
problem
solved,
because
this
service
can
be
kind
of
trusted.
If
you
run
by
us,
we
can
share
that
with.
A
D
C
A
Mean
that's
as
part
of
it.
That's
this
product
and
journey
and
that's
by
necessity
that
doesn't
mean
that
our
services
are
triggering
it
and
the
logic
around
has
to
be
centralized
because
you
can
still
have
that's.
Five
is
just
one
example
right
and
you
can
have
f-droid,
but
there's
all
kinds
of
mechanisms
that
you
can
use.
It
just
gets
it
right
now
we
are.
We
only
had
firebase
right,
but
it's
the
same.
Yeah,
multiple
clients,
you
have
you
have
I,
guess
your
band
or
your
desktop.
A
You
might
have
different
ones
and
then
you
have
different
ways
of
delivering
it
if
it's
firebase
or
App,
Engine
or
whatever
right,
but
but
but
we
should
not
sort
of
increase
the
sensation
by
having
a
single
point
of
failure
that
we
are
hosting.
That's
just
the
wrong
sort.
In
point
completely
to
me,
yeah.
D
Yeah
but
again
it's
it's
only
one
Barnabas
thing
push
notifications
themselves.
It's
one
part
of
the
I
think
and
the
second
is
I
think
like
we
like
being
able
to
send
whisper
messages
from
somewhere
like
using
like
in
fear
like
we
did
api's,
but
for
whisper,
which
is
not
really
yeah.
It
doesn't
work
right
now,
because
what
whispers
stateful
and
you
have
to
store
your
keys
somewhere
and
what
our
idea.
D
A
D
C
C
E
A
So
to
me
like
this
is
pretty
much
exactly
the
first
use
case
in
the
white
paper.
Why
don't
we
want
to
implement
it
in
a
similar
way,
as
described
there
and
had
actual
economics
and
in
implementing
in
a
felony
nice
fashion,
as
opposed
to
just
ignoring
economics
and
and
hosting
it
like
a
good
standard,
sass
2.0,
app
I,
don't
understand
why
that's
not
the
more
of
Turkey!
Oh.
D
Initially,
it
was
just
I,
don't
know
what
we
did
on
a
hackathon
was
just
sending
a
whisper
message
from
an
HTTP
endpoint,
and
that
was
without
storing
your
private
key
on
this
server.
That's
sort
of
what
was
done
right
because,
right
now
whisper
architecture
doesn't
allow
you
to
do
that,
so
it
doesn't
have
like
central
transaction.
E
Yeah,
so
do
I
understand
correctly,
it
could
be
done
now
if,
if
anyone
just
started
the
status
daemon
locally
and
talked
to
it
over,
you
know
we
have
library
like
SDK,
like
statuses,
decay
or
something
or
over
direct
IPC
whatever,
and
just
talk
to
this
local
instance
of
status
D
and
it
will
achieve
the
same
same
thing.
No
yeah.
D
They
already
yeah
if
they
want
to
host
it
like
this.
Yes,
but
the
second
part
was
also
that
we
were
contacted
by
in
fewer
themselves
because
probably
their
heard,
that's,
oh.
Why
we
in
fewer
doesn't
support
whisper
kind
of
question,
so
they're
very
interested
in
that,
because
they
want
to
host
whisper
and
whispered
into
habit
stateless
to
host
the
state
test,
because
they
don't
wanna
store
private
keys
of
anyone.
So
it's
not
any
different
from
the
protocol
is
like
talking
to
about
Adam
wrote
I.
D
Have
it
here
like
whispers,
a
protocol
right,
but
the
implementation
right
now
requires
you
to
have
a
private
key
on
the
note
that
sense
and-
and
so
the
idea
at
least
as
I
see,
it
was
just
okay
just
to
remove
this
requirement
and
that's
it
and
it
won't
be
much
different.
Then
and
that'll
be
the
same
envelopes,
the
same
proof-of-work
the
same
whatever
it's
just,
it
will
have
an
additional
API
method.
E
D
D
E
D
Well,
I
mean
you
can
of
course
ask
it,
but
I
mean
that's.
It's
more
convenient,
of
course,
to
have
something
like
this
API,
yes
for
convenience
reason,
but
otherwise,
why
would
we
I,
don't
know
even
have
status?
Clients
at
all
user
can
just
run
something
on
his
own
computer
and
send
messages
through
our
PC,
and
it's
perfect.
D
C
B
B
D
B
D
D
Yes,
if
we
implement
this
RPC
change
and
in
Fuhrer
like
actually
run
this
version,
which
they
as
I
understand
it
once
ok
will
work
both
in
meta
mask
and
in
status
and
yeah.
So
it's
and
also
it
would
be
possible
to
I
know,
send
messages
from
metamath
to
status.
That
might
be
a
good
thing
as
well,
but
essentially
all
the
apps
that
already
using
Fuhrer.
It's
like
we
are
concerned
right
about
decentralization
and
not
using
in
Fuhrer
and
not
using
things
like
that.
B
C
F
Yeah
I
think
here
it's
it's
important
to
see
that
the
main
use
case
is
notification
and
notification
being
sent
outside
after
that.
So
even
where
Muslim
statues
will
not
even
be
running
and
the
buttons
you
scaled
that
we
we've
been
discussing
so
far
because
yeah,
you
just
want
to
receive
a
notification
and
be
in
photo
of
something
any
time
without
having
to
just
make
sure
that
then
the
the
rights
that
is
running,
which
kind
of
defeat
the
point
of
an
education.
A
So
I
would
strongly
recommend
people
reread
the
white
paper
and
specifically
first
use
case,
which
is
about
essentially
this
class
of
problems
and
unless
there's
a
strong
reason
not
to
do
the
project,
I
would
just
evaluating
it
because
it's
actually
India
centralized
doesn't
require
central
solutions
and
it
has
economics
in
it
which
I
haven't
heard
anything
like
that.
I
would
just
like
reread
it
and
think
about
it.
How
we
can
do
it
properly,
I.
E
Yeah
I
strongly
agree
and
I
would
rather
get
rid
of
support
for
notification.
Then
then,
moving
from
the
centralized
solution
to
centralized
in
some
way,
I
mean
we
not
building
just
yet
another
messenger
right,
we're
trying
to
make
decentralize
things
like
a
new
thing,
and
if
it's,
if
it
means
that
we
we
don't
have
push
notification
that
that's
fine.
A
A
Also,
a
problem
of
you
have
two
stakeholders
and
then
a
third
one
who
sort
of
acts
as
an
intermediary
which
is
perfect
for
any
kind
of
time
delayed
mechanism
and
is
literally
something
where
someone
is
providing
value
and
know
this
providing
value
to
two
other
nodes,
and
that's
that's
where
the
utility
comes
in
and
doesn't
require
any
sort
of
essential
services
or
anything,
and
it's
actually
has
a
specific.
Yes,
in
the
use
case,
I.
C
D
D
A
Yes,
it's
probably
not
for
mobile,
but
this
ties
into
the
specification
person
to
idea,
and
it
also
ties
into
the
okay
around
this
intersession
I.
Having
started
running,
you
know
multiple
nodes,
which
is
where
it
touches
on
what
I
was
talking
about,
and
also
that
node
and
so
on
and
desktop
as
well.
Actually.
H
Comparing
how
the
system
is
right
now,
it's
when
you
show
yourself
in
a
toes
or
you're
going
to
receive
push
notifications
from
station
server.
This
is
something
that
the
user
agrees
on,
because
it's
hard-coded
in
the
in
the
application.
The
difference
in
this
push
notification
market
that
his
agreement
is
not
automatic.
You
don't
have
this
automatic
server
gives
you
certifications.
Instead,
the
system
will
make
me
to
find
an
old
to
bet.
You
trust
to
give
you
that
backwash
notifications
later
on
we
can.
H
A
C
E
A
H
That,
probably
the
APNIC
that
wants
a
push
notification
system
in
there.
The
app
needs
to
implement
some
some
script.
That
would
be
run
by
these
nodes
and
to
query
the
network
and
when
some
some
some
change,
some
state
change
happen
in
the
network,
then
it
will
make
another
create
a
message
and
sensual
specific
user.
H
D
What
which
is
essentially
what
we
wants
to
have
fixed,
so
you
can
sign
message
locally
and
then
push
it
somehow
into
whisper
network
without
having
to
have
like
if
I'm
at
the
app
owner
and
I
don't
want
to
run
a
server
with
a
whisper.
Not
then
I
can
use
in
fear
our
use,
whatever
else
to
send
it.
That.
D
H
D
A
H
The
design
of
your
diet
is
wrong
because
you
cannot
be
down
there
off
the
idea.
Of
course.
Some
people
do
it,
but
then
you
are
actually
wrong
in
this.
The
idea
of
we
want
to
make
centralized
applications.
These
notifications
are
for
the
centralized
application.
If
you
are
the
owner
of
the
application.
Probably
you
can
provide
your
yourself.
You
can
provide
an
old,
maybe
you
can.
We
can
give.
H
H
But
for
the
push
notification
probably
would
be
something
like
the
user
interface,
but
is
like
a
back
end
interface
that
is
used
by
these
push
notification
services
that
they
will
use
some
something
that
they
are
developing
like
like,
for
example,
for
the
CEO
search
engine
optimization.
We
have
like,
like
that,
robots,
dot,
text,
text
and
and
other
think
configurations
in
the
web
server.
We
can
do
for
push
notifications.
We
can
also
can
have
a
specific
file
that
is
run
by
this
node
by
these
status,
nodes,
of
course,
that
they
can
know.
H
A
C
Yeah
I
have
a
question
to
you,
guys
so
saying
that
you
have
an
owner,
but
why
do
you
well
one
developer.
I
completely
agree
with
even
arguments
that
you
need
to
run
status
and
status
did
not
yet
not
to
to
communicate
using
whisper.
The
fact
that
some
unpromising
fora
shouldn't
in
influence
our
decision-making.
Why
why?
Why
do
you
insist
on
it.
D
No
but
I
mean
just
to
like
lower
the
threshold
to
adapt
to
whisper
adaption.
Okay
yeah
we
can,
we
can
just
look
up
and
think
that's
okay,
you
have
to
do
that.
You
have
to
do
Y
and
of
course
all
the
developers
will
come
and
gladly
run
the
services
and
set
up
monitoring
for
them,
and
things
like
that
so
but.
C
D
I'm
a
separate
business
person
that
wants
to
create
a
service
using
quite
blockchain
and
the
more
we
want
more
business
people
and
more
businesses
using
watching
back
in
January
for
crypto,
like
for
crypto
adoption,
we
need
like,
like
coffee
shops,
we
need
like
betting
sites.
We
need
like
with
grocery
stores.
We
need
them
to
adapt
easier
or
whatever
crypto,
right
and
I
took
it
like
if
we
giving
them
more
convenient
way
to
build
something.
D
A
The
primitives
have
to
be
have
to
be
decentralized
and,
of
course,
people
can
build.
Semi,
centralized
citizens
on
top
of
us,
but
we
have
to
the
primitives
we
provide
should
be
use
of
proper
and
true
to
the
theorem
ecosystem.
That
doesn't
mean
that
we,
we
don't
allow
other
people
to
have
semi
nice
things,
but
that
has
to
be
what
we
provide.
Otherwise
we're
not
really
an
encapsulation
of
Afyon.
D
A
Mean
so
the
primitives
that
we
provide
status
as
whatever
we
are,
the
premiers
have
to
be
serviced
and
sliced
and
work
without.
If
you
turn
on
our
cluster
and
so
on,
like
we
should
still
have
the
same
uptime
for
communication
and
logic,
and
all
of
that
it
has
to
sort
of
be
designed
that
way
and
then
building
on
top
of
us
in
terms
of
solutions,
convenient
solution
and
so
on.
It
doesn't
mean
that
that
has
to
be
done.
Gently
centralized,
as
you
say,
and
that's
something
we
can.
D
I
mean
I'm,
not
so
keen
that
we
need
so
like
we
have
to
like
deploy
and
run
so
that
a
few
things.
Okay,
there
is
additional
API
to
whisper
right
and
there
is
a
deployment
of
our
own
cluster
with
this
API,
that's
other
with
the
ex-king
news
and
the
second
one
as
I
understand
is
yeah.
We
don't
want
to
do
that.
We
wants
so.
D
Instead
we
want
to
implement
this
white
paper
market
of
push
notifications
right,
but
is
there
anyone
who
is
against
having
two
more
API
methods
in
whisper
that
then
can
be
used
by
fewer
and
the
others
for
those
who
can
using
fewer
and
stuff?
So
we
don't.
We
want
to
deploy
them,
but,
like
in
theory,
takes
our
our
fork
and
deploy
that
and
so
more
people
who
use
in
fewer
anyway,
they
will
use
it
for
whisper
as
well,
and
but
we
don't
host
anything
ourselves.
D
D
White
least
in
our
browser,
please
say:
P
is
and
also
in
fewer
can
use
them
to
deploy
twisper
as
a
service
sink
yeah.
And
that's
what
that's
like.
That's
what
I'm
talking
for
and
I'm
talking
how
these
can
like
improve
the
adaption
and
things
like
that,
but
like
deploying
our
own
web
service
II,
and
that
might
not
be
a
great
idea.
A
I
A
D
D
You
started
to
arguing
against
something
that's
like
in
this.
The
whole
idea
so,
which
points
are
you
most
opposed
to,
which
point
see
you
think,
are
unacceptable
for
us
in
this
idea
like
there
is
like
this
new
to
API
methods,
there
is
deploying
our
own
cluster,
that's
exposed
to
see
Parramatta,
and
there
is
push
notifications
like
three
deploy.
E
Deploying
our
notes,
our
class,
not
the
cluster
servers
right
for
that
and
making
our
infrastructure
relay
relying
on
this
I'm,
not
saying
this
bad
solution.
But
it's
against
our
as
foscar
said
our
whitepaper
and
our
general
idea,
and
that's
basically
preventing
us
from
developing
other
solution,
which
is.
C
A
Would
even
said,
and
then
also
I
would
just
picture
adding
API
calls.
I
am
on
the
side,
I'm,
not
clear
what
the
implications
are.
I
guess.
My
only
general
concern
would
be
that
we
should
be
focuses
of
our
core
value,
propositions
and
core
business
and
maybe
Messer
time
work
in
that.
But
I,
don't
know
how
I
don't
know
too
much
about
specifically
what
the
implications
that
would
have
in
terms
of
compatibility
and
how
much
effort
it
will
be
and
and
all
these
kinds
of
things.
A
So
maybe
it's
reasonable
trade-off
or
not,
but
the
main
thing
is
around
hosting
infrastructure
ourselves
and
given
the
problem
that
was
stated
in
the
beginning
that
we
are
not
so
pursuing
it,
because
it's
to
me
it's
a
straight
fit
with
Dwight,
Dave
and
IA.
It's
a
shame
that
we
passed
this
work
before
the
better,
because
I
think
we
could
have
been
a
float
along
out,
but
I
would
really
like
us
to
revisit
it
and
do
it
like
it's
written
in
the
white
paper,
essentially
because
I
don't
think
we're
learned
anything
new
in
the
last
year.
D
My
idea
was
that
we
had
these
two
methods
and
we
exposed
them
through
our
browser
like
their
browser
and
then
since
inferior
is
interested
in
that
they
will
just
take
this
message
and
deploy
them,
and
we
don't
use
that
like
we.
They
just
to
take
this
from
us,
deploy
it
and
someone
else
will
use
them,
but
we
want,
but
for
someone
else
why?
Why
is
it
good
for
someone
else?
D
C
H
Problem
with
using
impurities
that
we
kind
of
limited
by
them,
I
guess
see
your
solution
that
we
just
open
these
two
optional.
Who
is
that
I?
Don't
have
any
problem
with
this
idea
in
specific
but
I
in
my
vision,
what
will
happen
is
that
in
will
run
like
I
start
to
snowed
in
start
serving
like
a
very
trustworthy
status
mode
in
future
we
develop
this.
H
These
mechanics,
that's
in
Fuda
in
anyone
can
run
so
we
can
kind
of
democratizes
this
type
of
services
we
may
make
anyone
be
able
to,
because
we
know
that
there
is
a
single
point
of
failure
and
we
don't
want
that.
We
don't
have
anything
against
people.
We
need
food.
We
just
don't
want
this
point
of
failure,
so
we
make
this
to
everyone
that
can
want
to
run
it.
A
I
I
mean
you
know
the
smart
contract
stuff,
quick
update
for
anybody
who
doesn't
know.
We
engaged
Sigma
Prime
a
auditing
firm
to
perform
an
external
audit
on
the
ENS,
smart
contract
user
name
registrar
at
Seoul,
and
they
have
we
did
this.
We
started
this
movie
a
week
after
your
F
Berlin.
They
got
it
back
to
us
this
past
week.
They
did
it
really
really
quickly
without
lack
of
quality
report.
So
we're
now
going
through
the
report
with
them
to
fix
the
issues
they
found
on
a
quick
overview.
I
Like
points
of
note,
so
overall
they're
really
impressed
with
the
quality
of
the
work
that
we
did
and
we're
very
happy
with
the
quality
of
the
audit
that
they
performed
us.
So
once
this
is
done,
I
imagine
Ricardo
will
have
most
the
fixes
done.
They'll
sign
off
on
these
on
all
the
fixes
create
a
more
finished
report
and
then
we'll
release
that
to
the
public.
To
show
that
you
know
this
is
what
we've
done.
If
you'd
like
to
see
it.
H
It
the
quality
of
the
Sigma
prime.
They
really
understood
what
we
needed,
so
they
really
thought
about
our
application.
They
did
not
simply
look
it
into
our
contracts
or
filings,
but
they
also
look
at
into
our
concepts
so
at
overall,
the
the
core
of
the
quality
of
the
the
report
was
great
and
regarding
the
issues
it
found
the
summer
where
there
were
medium,
but
idea
what
were
very
unlikely
to
happen
because
they
all
were
related
to
the
controller
of
the
contracts,
the
honors
would
matter.
Nope,
no
phones
at
risk,
an
ad
at
this
point.
A
I
I
have
another
one
real,
quick,
that's
outside
of
that
contract
stuff,
and
that
is,
we
are
signing
a
contract
with
threat
stack
to
have
some
monitoring
services
of
all
of
our
infrastructure
that
we've
considered
critical
infrastructure
over
time.
We'd
like
to
minimize
this
as
much
as
possible,
so
we
rely
almost
on
no
critical
infrastructure,
but
will
always
always
have
some
so
threat.
Stack
is
just
a
service
that
allows
us
to
see
who's
doing
what
on
those
computers
and
if
something
goes
wrong
we
can
mitigate
it
triage
it
quickly.
A
G
And
oh,
your
briefly:
we've
I've
just
moved
across
docs
to
use
different
static
site
generator
called
hexa,
because
that's
what
it
bark
was
using
its
lucky,
more
modular
and
extensible,
then
donkasaurus.
It
should
look
mostly
the
same.
What's
interesting
from
a
dead
perspective
is
that
Jakob
has
kindly
helped
me
set
up
a
particular
pipeline
to
make
contributing
to
this
repo
even
easier.
So
if
you
want
to
change
a
specific
page,
you
just
need
to
go
to
dogs
the
status
that
I
am
you'll,
see
like
an
edit
button.
G
You
can
edit
it
in
the
browser,
if
you
like,
I'm,
not
feeling
really
lazy
and
it'll.
Take
you
onto
the
develop
branch
specifically
developed,
builds
to
a
machine
that
we
hosts
and
is
hosted
at
dev
Docs
that
status.
That
I
am.
That
means
that
you
can
merge
or
changes
to
develop,
check
it
on
the
live
sites,
and
then
we
only
merge
the
master
from
develop
with
fast
forward
flag
once
we're
happy
that
the
changes
reflect
what
we
want
them
to.
So
it
should.
It
should
be
really
really
easy
to
contribute
to.
G
The
ducks
now,
like
I,
said
just
push
to
the
develop
branch
and
check
out
your
changes
on
dev
ducks
that
status.
That
I
am
that
means
reviewing
changes
and
maintaining
the
whole
thing
will
be
even
easier
and
I'll
be
working
with
various
different
people
and
pinging
you
over
the
next
few
weeks
to
make
sure
that
your
sections
are
two
dates.
Once
we've
got
a
good
content
pipeline
flowing
through
there,
then
we
can
identify
properly
user
journey
is
feed
that
you've
put
back
into
design.
Think
about
iterating
through
the
site's
and
figure
out.
G
D
Notice
that
we
from
the
next
nightly
there
will
be
a
light
client
option
in
the
night
list
of
status,
so
it
probably
will
work
very
poorly,
but
we
bring
him
back
and
yes
and
we'll
test
it
thoroughly
and
we'll
try
to
fix
everything
we
find.
So
we
are
not
really
using
only
in
theorizing
as
an
only
option.
So
far
up
stand.
Only
we
get
in
there.
D
There
are
enough
rope,
stand
notes
right
now,
because
I
do
know
there
was
some
problem
that
there
wasn't
enough
rope
stand
nodes
in
the
network
like
main
that
was
kinda
working,
but
Robson
the
people
just
weren't
running
this
or
some
Robson
nodes,
but
I'll.
Try
that's
a
big
thing
about
for
magnet,
I!
Think,
yes,
that
we
can
really
rely
on
discovery,
yeah.