►
From YouTube: Core Devs Meeting
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
B
B
B
C
A
A
Right,
I,
don't
please
write
in
the
chat
if
you
want
to
try
again
so
I
guess
with
that
moving
on
to
the
next
one,
SMIL
server,
availability
and
responsiveness.
So
the
context
here,
there's
the
last
week
we're
seeing
of
multiple
reports
establish
and
so
on,
of
bad
availability
and
responsiveness
of
multiple
mill
source
and
it
sort
of
led
to
some
frustration
and
so
on,
and
it
just
intuitively
seems
like
we
are
sort
of
not
meeting
sort
of
our
own
goals
in
terms
of
how
quickly
you
want
to
respond.
A
D
D
Too
large
I
don't
know
we
get
times
out
when,
even
when
the
all
the
messages
have
actually
been
sent,
I
don't
know,
but
so
when
one
of
the
thing
that
we
can
do
to
improve
that
and
andrea.adrian
is
already
working
on
that
is
to
allow
the
app
to
request.
Multiple
topics
at
once
and
also
Adam
or
Andrea
will
be
looking
at
sending
more
messages
at
because
right
now,
it's
sending
only
messages
one
by
one.
D
So
if
you
look
at
how
the
app
is
fetching
messages,
it's
pretty
slow,
which
is
why
so
many
requests
are
timing
out
and
on
my
end,
I'm
working
on
making
the
mail
server
fetch
make
one
request
after
the
other,
instead
of
also
moved
annually,
so
that
there
is
way
less
timeouts
and
also
change
mail
server.
Whenever
there
is
to
three
consecutive
time,
ups
I'm
done.
A
D
Art
happens
when
you
after
10
seconds,
you
didn't,
receive
all
the
messages
and
but
it's
just
I
think
it's
after
10
seconds.
We
should
have
received
all
the
messages
and,
if
I
understand
correctly
one
of
the
promise
that
it
sends
messages
why,
by
one
so
the
the
speed
at
which
messages
are
sent
is
pretty
low.
D
I
don't
know
if
there
is
some
kind
of
ping
pong
going
on
with
acknowledgement
of
received
messages,
but
that
would
be
why
it's
so
so
slow
if
it's
how
it
works,
because
then,
if
each
message
is
acknowledged,
one
by
one,
we
need
a
round-trip
for
every
message.
Maybe
Andreea
or
Adam
can
tell
us
more
about
how
it
works
exactly
because
I
don't
know
for
sure.
E
Yes,
yes,
so
I
can
say
something
about
it.
So,
first
of
all,
I
wanted
to
build
and
kind
of
like
something
like
a
tool
to
measure
the
performance
of
the
messages
and
wider
slow,
because
even
when
they
are
sent
one
by
one,
it
still
should
be
pretty
fast
and
there's
no
any
acknowledgement
or
anything
other
than
one
protocol
lever.
So
it's
xvp.
E
So
it's
just
like
you
know
just
under
protocol
acknowledgement,
packets,
but
that's
it
so
also
the
the
number
of
messengers,
not
that
huge,
because,
as
you
said
currently,
we
send
requests
for
each
channel
and
it's
not
like
thousands
of
methods
like
hundreds.
So
at
most,
so
it
should
be
fine,
so
I
don't
know
exactly
the
reason
which
is
behind
the
slow
passage
of
messages
from
my
servers
to
the
device.
That's
that
one
thing
to
look
into
and
I'm
I'm
working
on
this
right
now,
but
yeah
the
rest.
This
is
true.
Definitely
we
can.
E
We
can
send
envelopes
and
packets,
like
whisper
protocol
itself,
operates
now.
My
server
kinda
was
never
adjusted
to
that,
so
it
always
sends
like
one
one,
one,
one
one
one
envelope
at
a
time
so
that
that's
quite
slow
yeah
but
yeah
from
my
side,
its
first
figuring
out
exactly
why
and
why
it's
low
like.
What's
the
reason
and
because
it's
kind
of
we
don't
have
right
now
any
tools
that
would
tell
us
exactly
what's
what
we
know.
A
E
Yeah
I
can
check
it
quickly
now,
but
recently,
when
I
was
looking
into
that,
that
was
fine.
We
still
have
a
lot
of
space
for
like
spots
on
the
main
servers
to
connect
to
so
that
was
okay.
It
just
suddenly
got
slower
and
the
only
thing
changed
is
the
number
of
messages
in
the
channels,
so
I
assume
that
it's
somewhere.
F
So
I've
had
a
lot
of
issues
where
messages
the
stone
download,
and
so
that's
the
fix
to
fix
that
I've
been
logging
out
long
back,
you're,
not
changing
mail
servers,
I've
also
been
getting
explicit,
could
not
connect
like
tap
to
retry
type
errors
that
have
been
showing
up
and
a
lot
at
a
time
like
fetching
messages
is
the
shown
constantly,
but
jet
in
general
it
seems
pretty
hidden.
This
sometimes
messages
seem
to
be
coming
in
no
problems
other
times
I'm
getting
like
black
spots.
E
Yeah
I
can
say
that
I'm
recently
using
desktop-like
for
ninety
percent
of
my
time
and
it's
I
can
see
that
sleep
slow
in
terms
of
downloading
messages,
but
I've
never
seen
any
problems
in
the
errors
with
that
it's
a
bit
slow.
So
when
I
open
it
after
in
few
hours,
I
can
see
it's
like
you
know.
The
message
is
visible,
for
you
know
20
40
40
seconds,
but
they'll
all
they
never
seen
a
name
error.
So
how.
D
D
A
E
G
E
E
E
G
E
E
On
average,
it's
like
I,
said
it's
yeah,
so
there's
like
there's
no
spikes
that
are
as
high
as
eight
seconds,
but
there's
very
rare,
and
usually
you
get
the
response
in
lake
below
below
two
seconds.
It's
it's
still
like,
like
99th
percentile
is
like
three
seconds
so
on
the
server
side.
It's
fine
so
but
yeah
and
you
know.
G
F
Just
something
to
add
some
color,
so
on
analytics
that
status,
that
I
am
you
can
see
some
messaging
stats,
including
messages
per
day
being
collected
from
a
handful
of
public
channels
and
there's
a
link
that
you
can
click
that
will
show
you
what
channels,
but
so
far
we're
doing
around
2000
messages
a
day
just
from
the
channels
we're
monitoring,
but
it
could
be.
Oh.
B
F
E
B
E
E
H
A
So,
just
for
some
context,
so
this
is
about
Death
Note,
and
that's
not
is
this
for
those?
No.
No
it's
this
initiative,
where
it
makes
it
very
easy
to
be
a
persistent
node
for
any
kind
of
peer-to-peer
application.
So
this
could
be
like
running
your
own
full
node
or
status
node
in
this
case
or
I.
Guess
not
like
all
these
sets
of
apps.
That
need
some
kind
of
know
that
you
connected
to
and
it
has
so
script.
A
So
you
can
run
it
on
your
virtual
server
or
your
own
server
at
home
and
it's
based
on
pointers
on
and
they
have
this
admin
panel
and
you
can
connect
via
some
kind
of
VPN
solution
or
whatever
directed
to
it
and
and
the
idea
is
that
we
want
to
run
status
there,
because
it
sort
of
enables
us
to
have
many
sort
of
nodes
that
connects
us.
Mail,
service
and
boot
notes,
and
people
in
the
community
can
can
sort
of
run
them,
and
this
would
be
in
conjunction
with
something
like
desktop.
A
Would
that
no,
it
would
be
more
likely,
be
persistent
and
like
high
availability,
because
it
would
be
online
for
a
long
period
of
time
and
one
of
the
Horace's
that
we
went
around
750
nodes.
So
we
would
need
to
make
sure
that
this
is
smooth
and
easy
to
set
up
and
I.
Think
that's
Adam
England,
you
wanna,
add
some
stuff
there.
So
correct
me
if
I'm
wrong,
but
we
want
to
make
this
easy
to
run
and
also
get
multiple
people
record
halves.
C
That's
that's
exactly
it
so
I'm
arranging
a
call
with
add
admin
guys
to
basically
get
a
bit
of
tutorial
about
how
to
host
that
Nouri.
That
node
on
a
digitalocean
account
on
the
weekend.
I
went
through
the
process
of
trying
to
set
up
a
Mac
and
I
promptly
found
out
that
if
I
was
successful,
I
would
have
completely
nuked
Newton,
my
Mac.
So
so
don't
try
run
that
note
on
your
personal
computer
and
yeah.
E
Yes,
yeah
yeah,
just
just
wanted
to
add
to
that
that
yeah
I
tried
it
on
on
digital
ocean
actually,
and
it's
pretty
pretty
easy,
so
you
pretty
much
the
instruction
is
that
you
just
execute
some
commands
like
bash
to
comments
to
install
dependencies
in
the
node
itself,
which
is
a
little
bit
tricky
because
I'm
not
fond
of
executing
some
bash
scripts
get
from
the
website,
but
that's
their
instructions
so
yeah.
But
after
that,
it's
it's
fine,
so
it
installs.
E
You
need
to
configure
a
VPN
connection
to
that
to
that
node
to
the
to
your
server
and
entry.
That's
that's
pretty
much
it
so
it
starts.
You
just
follow
a
host
name,
open
that
in
the
browser
and
you
got
access
to
it
through
VPN.
So
it's
so
it's
secure!
It's
not
like
do
you
know
you
don't
need
to
do
anything
else
on
your
own,
your
server!
You
have
already
these
things
configured
until
the
blockchain
syncs
automatically.
It
start
syncing.
In
the
background.
E
So
after
a
few
hours,
you
have
that
full
node
up
and
running
I'm,
not
sure
if
it's
full
nodes
or
slick
it,
just
lights,
sinking,
I'm,
not
sure
about
it
and
yeah.
After
that,
you
can
just
install
status
node
by
clicking
one
button
and
you're
set.
So
it's
fairly
easy.
It's
really
like
kind
of
approachable,
so.
B
A
E
A
E
Don't
know
I
haven't
seen
it
yet
I
mean
you
have
the
the
IP
of
your
server
available
I'm,
not
sure
how
its
configured
in
terms
of
ports
so,
for
example,
if
because
all
these
things
run
in
a
docker
container,
so
I'm
not
I
haven't
checked
it.
Yet
if
those
ports
are
available
for
the
IP,
probably
not,
but
yeah
I'll
check
this
and
I
haven't
seen
any
instruction
on
their
on
their
website.
That
allows
you
to
use
this
stuff
without
VPN.
E
A
Since
we
can't
have
a
demo
what
about
like
making
like
a
short
like
set
like
lists
of
instructions
or
a
video,
and
then
we
can
get,
maybe
five
people
in
this
call
to
commit
to
trying
it
and
then,
if
it
works,
we
can
push
it
out
with
community
marketing.
Since
we're
like
two
weeks
away
away
from
end
of
quarter
and
and
proud.
E
Yeah,
in
my
opinion,
is
the
ability.
The
only
problem
right
now
is
that
they
are
kind
of
the
owner
of
the
status
node
package
that
it's
in
that
node,
so
because
you
can
define
your
packages
and
install
them
and
right
now
it's
it's
listed
by
default
when
you
install
that
now,
so
that's
cool.
However,
we
do
not
control
that,
so
they
based
this
on
the
most
recent
version.
But
when
we
want
would
like
to
update
this
package,
we
would
probably
need
to
contact
them
and
ask
for
for
doing
that.
E
So
I
got
another
thing
that
I
wanted
to
resolve,
but
other
than
that
I
think
it's
it's
fine
I
mean
all
those
steps
and
and
the
connectivity
works
fine,
so
I
think
we
can
definitely
prepare
a
video
at
least
four
quarters
now
to
follow
some
instruction:
how
to
run
that
and
test
it
and
later.
If
it's
turn
out
to
be
okay,
to
do
that
for
the
community
idea
along
the
like.
In
the
same
time,
you
can
work
with
them.
A
G
E
Configurable
server
so
right
now
it's
the
right!
Now,
it's
very
simple:
you
could
just
switch
between
whisper,
rely,
node
or
or
my
server,
so
you
can
operate
it
in
two
modes,
but
their
interfaces
as
possible
to
be
configured.
So
if
you
have
a
packet
and
you
have
more
options,
you
could
you
could
specify
them
in
the
package
and
they
will
list
in
the
UI,
and
you
can
change
them
and
pass
this
values
through
through
a
config
for
a
JSON
file
or
through
environment
variables.
G
E
Yeah,
but
that's
the
problem
with
this
thing,
that
that
packages
are
stored
and
managed
in
a
smart
contract.
So
you
I
mean
you
know
what
you
could
do
is
to
build
a
package
it
up
load
on
ipfs
and
pass
a
link
in
that
UI
to
install
it
and
run
it.
But
currently
it
won't
list
status
as
an
available
package
without
sinking
the
chain,
because
the
packages
definitions
are
stored
on
the
on
the
blockchain.
I
E
I
E
E
But
it
connects
to
so
yeah,
it
must
be,
it
must
be
public
and
the
port
yeah
I
guess
it
must
be
public
as
well,
because
otherwise
it
would
not
connect,
but
you
can
see
logs
in
in
depth,
know
that
it
can
connect
other
peers
and
and
pass
messages.
We
try
this.
Let
me
check
if
the
metrics
are
updated
and
they
are
so
actually
this
note
running
up
not
realize
messages
so.
B
I
G
And
because
most
of
the
peers
have
exactly
the
same
problem,
if
you
run
it
on
your
mobile
phone
behind
a
router,
they
also
can
connect
to
other
peers,
but
no
one
can
connect
to
them.
So
it's
like
two
completely
separate
world,
but
I.
Think
in
general
there
might
be
a
good
idea
of
trying
to
pierce
net
using
VPN.
It
also
might
be
it's
a
little
bit
heavy
weight
more
heavier
than
not
like
piercing,
but
it
still
it's.
I
G
A
B
G
G
G
A
A
A
A
That's
just
brown
wild
when
I
set
up
to
definitely
enjoy
a
lot,
but
I
had
like
a
static,
IP
and
I
built
it
and
I
could
use
it
with
the
VPN,
but
I
couldn't
find
like
an
e
note.
I
was
just
looking
for
an
e
note,
so
I
couldn't
connect
to
my
my
no
there
without
being
connected
a
BB
gun,
but
it
wasn't
possible.
I
I,
don't
understand
why
you
know.
A
I
B
I
C
C
Think
I
mean
once
that
loads
running.
It
seems
to
be
fairly
simple.
You
know
the
GUI
interface,
you
just
covered
click,
the
boy,
but
getting
that
nerd
up
and
running
is
fairly
I
mean
fairly
challenging
in
the
sense
that
you
need
to.
You
need
to
have
a
digitalocean
accounts
or
a
a
space
of
a
kind
of
lying
around,
but
something
a
little
bit
more
powerful
in
Raspberry
Pi.
A
C
G
Yeah
so,
as
you
read,
the
background
is
that
slack
was
very
good
with
having
different
types
of
integrations
and
since
we
are
moving
to
status,
one
of
the
big
like
sort
of
impediment
or
something
that
we
lose.
Is
this
different
type
of
automation.
When
I
don't
know
somebody
published
a
blog
post
or
there
is
a
new
PR
or
even
the
history
think
so,
and
we
used
to
have
these
bots
that
we're
using
specifically
for
testing
that
can
spam
channels
and
read
channels
in
go
before
so.
G
My
idea
was
to
maybe
reuse
this
SDK
or
something
like
this-
to
create
a
one
example
boat,
which
will
be
very
well
documented,
and
it
will
be
like
literally
like
an
example.
And
then,
since
we
have
a
hackathon
coming,
then
it
would
be
nice
to
have
some
dogs
and
this
example
to
and
to
have
some
hackathon,
maybe
ideas
about
which
kind
of
boats
we
want
and
essentially
implement
them.
But
there
is
also
this
this
constant
struggle
right
now
with
okay,
every
now
and
then
our
protocol
is
changing
and
we
also
want
to
rethink
our
protocol.
G
So
that's
sort
of
I
mean
maybe
the
perfect
timing
for
that
would
be
after
we
will
rebuild
and
implement
some
more
formal
protocol
right
now,
but
on
the
other
hand,
we
have
this
problem
right
now
with
like
enough
automation,
so
I
just
was
thinking
to
discuss.
What
do
you
guys
think
about?
How
should
we
implement
this
example?
Should
we
use
like
JavaScript
or
closure
and
use
the
same
code
as
status?
Does?
G
Should
we
just
reimplemented
and
go
or
like
rather
fix
the
existing
libraries
and
do
it
this
way
so
yeah
I
just
wanted
to,
and
in
general
I
want
to
raise
awareness
that
there
are
boats,
and
so
when
people
change
something
in
the
protocol,
this
boss
will
probably
break
so
it's
it's
a
few
times
in
the
chatroom,
which
was
a
kind
of
surprised
that
oh,
we
have
some
boats
that
are
doing
something
so
yeah.
That's
that
are
my
thoughts.
G
E
I
can
only
share
my
experiences
with
both
so
and
the
current
status
go
SDK
kind
of
assumes
that
you
have
node
status,
node
running
somewhere
aside,
so
you
either
use
something
under
server
or
you
run
it
locally
and
communicate
with
it
through
HTTP
protocol,
I,
guess
or
yeah.
Oh,
maybe
it
was
IPC,
inter-process
or
HTTP
doesn't
matter.
The
other
approach
that
I
was
trying
to
explore
essentially
was
to
write
bots
that
are
kind
of
the
truck
that
ran
the
status
node
embedded
it
into
it.
E
So
you
do
not
need
to
run
to
processes
and
communicate
with
them
just
one
so,
and
that
was
pretty
nice
in
my
opinion,
so,
and
but
that
would
imply
that
you
ran
that
we
run
this
as
the
that
we
write
this
thing
as
go
as
decay.
Pretty
much.
You
know
just
just
kind
of
experience
that
I
had
so
yeah
I
think
it
would
be.
E
Ideally,
it
would
be
great
to
have
both
so
either
you
ran
simple,
but
that
you
do
not
care
about
a
specific
like
a
different
process
that
surrounds
somewhere
aside
and
you
just
explore,
charts
and
whatever
you
want,
and
the
other
thing
is
that
you
have
a
permanent
node
running
somewhere
and
you
just
want
to
connect
to
it,
communicate
through
HTTP
or
some
other
product.
That
would
be
ideal
to
have
both
Fox
News.
G
J
See
that
status
node,
they
should
replace
servers.
So
imagine
that
wherever
you
are
server
role,
this
will
be
the
role
of
status
nodes.
But
we
agree
on
the
user
agree
which
server
they
choose
so
I
think
we
start
without
payments
and
later
we
implement
the
payments
for
the
servers
or
subscription,
but
I
think
that
is
basically
processing.
The
blockchain
information
Evans
and
push
notifications
has
the
white
paper
so
having
a
bot
starting
up
as
a
bot.
J
G
I
think
in
this
case
we
are
talking
about
specific
whisper,
not
a
blockchain,
so
frankly,
none
of
the
boats
that
were
requested
at
least,
and
they
were
about
like
smart
contracts.
Maybe
this
is
the
next
phase,
but
right
now
it's
were
mostly
about
okay.
We
have
a
new
blog
post
up
on
our
ghost
and
we
want
I,
don't
know
in
the
status
channel.
We
won't
automatically
have
a
message
about
it
or
someone
open
it
PR,
for
instance,
repository
and
in
another
channel
message
about
this.
G
J
About
it,
this
example
of
a
smart
contract
event,
because
we
have
a
multi-sig
and
I
wanted
that
when
I
create
a
transaction
that
Muta
see
using
that
event,
the
and
other
users
of
that
mood
seek
receive
a
message
automatically
from
this.
So
this
is
the
simple
user
case
they
can
can
describe,
but
I
see
that
that
we
can
start
only
with
whisper
but
I
think
they
dig.
That
is
the
roadmap
going
to
are
towards
events
in
blue
in
the
blockchain.
B
E
That's
a
good
point
because
also
also
shows
that
we
actually
is
both
because
for
some
people
who
want
to
interact
with
the
blockchain
and
write
BOTS
that
also
do
interact
with
the
Box
chain.
It's
essential
that
you
they
have
access
to
blockchain
and
if
we
just
go
with
a
way
that
you
write
about
with
embedded
whisper
now
you
need
to
use
in
for
our
RPC
endpoints,
because
you
know
it's
not
feasible
to
run
ephemeral.
Node
easily
I
mean
the
syncing.
Blockchain
is
fairly
fairly
slow,
so
yeah
I
mean
it's.
E
D
I
But
if
you
want
to
make
boat
in
JavaScript,
where
I
do
need
something
bindings,
just
allowing
ships
that
lose
D
process
and
connect
to
it
used
in
HTTP,
what
I,
just
I,
just
don't
understand
what
the
role
of
bindings
in
I
don't
know
what
what
are
the
bindings.
But
what
has
a
role
point
regarding
things.
D
G
I
G
A
G
For
us,
because
if
we
are
I
mean
we're
running,
go
on
so
I,
don't
think
it
will
be
possible
to
run
it
on
top
of
standard
whisper
implementation
and
still
have
everything
work.
So
that's
one.
What
was
one
of
my
actually
questions?
So
should
we
run
it's
like?
Should
they
be
just
whisper
notes
or
should
be
maybe
status
notes
because
being
them
like
a
status?
G
G
Yeah
and
imagine
that
we
just
right
now,
we
are
in
status,
go
as
the
gates
just
go,
and
it
has
stuff
like
joy
in
public
channel.
Read
messages,
send
messages
some
very
basic
API,
but
there
its
status
specific.
So
it's
essentially
reads
from
a
public
chat
rooms
and
the
rights
to
them
and
in
the
message
format
that
status
understands
and
using
topics
like
status,
users,
topics
and
things
like
that.
So
but
it's
implemented
in
go
so
I
I
was
thinking.
That's
this
API!
G
J
G
J
That
that
a
great
way
to
create
these
BOTS
is
using
our
scripting
language,
I'm,
not
sure
if
you
feel
aggrieved
that,
but
yeah
I
use
it
a
lot
of
software
that
to
do
extensions,
you
need
need
to
learn
our
simple
language
that
many
other
software's
use.
That,
for
example,
is
Lua
is
one
of
them.
Of
course
it
could
be
done
in
Java,
scripting
and
C
in
anything,
but
this
language
is
really
simple,
and
maybe
we
can
create
the
bots
in
this
in
these
simple
scripts.
G
E
I
also
just
just
getting
back
to,
for
example,
this
create
accounts,
account
stuff
and
other
things
yeah.
So
for
me,
if
we
decide
that
status,
node
is
required
and
we'll
just
use
our
PC
to
do
things
like
create
account
and
that's
kind
of
I
mean
that's
probably
a
short
short
path,
but
also
not
the
kind
of
dead-end.
E
Maybe
so
my
thinking
was
that
we
will
try
to
build
a
protocol
and
describe
it
in,
like
you
know,
a
documentation
and
anyone
can
write
like
recreate
the
Sphinx
in
any
other
language
and
be
independent
from
whisper
or
from
from
notes
at
all.
So,
for
example,
this
create
account
at
some
point
that
would
require
connection
to
whisper
node.
But
you
could
you
could
actually
create
a
get
account
using
this,
this
library
and
not
do
that
with
status
node.
So
that
was
my
impression
like
some
time
ago,
when
we
were
thinking
about
it.
A
I
kind
of
agree
with
that,
the
first
of
all
we
had
some,
you
can
do
that
cogeneration
with
a
buffer.
So
so,
if
we
use
protocol
buffers,
we
could
have
like
implementation
like
specific
languages.
Whatever
language
people
are
comfortable,
they
could
so
here's
how
you
use
our
protocol
and
then
maybe
we
just
start
with
go
because
that's
what
the
person
who
wants
to
push
this
once
you
use,
but
it's
sort
of
agnostic.
G
E
So
it's
like,
for
example,
at
what
I
would
like
to
test
it.
That
would
just
work.
I
know,
I
could
just
just
plug
in
a
bleep
p2p
and
just
connect
both
no
it's
truly
p2p.
That's
it
and
just
start
sending
them
unencrypted
messages
or
what's
whatever
and
be
able
to
communicate
in
that
way
and
if
I
decide
to
use
whisper
as
the
transfer
protocol
I
just
configure
that
library
with
a
whisper
IPC
file
or
whatever,
and
then
it
starts
to
use
whisper
and
talk
through
whisper
to
different
notes.
That's
that.
B
G
Maybe
they
should
be
like
multiple
generations
of
bots,
because
like
right
now
for
ourselves,
we
want
something
that
just
works
like
kudos
and
things
like
that.
It's
just
not
about.
Maybe
maybe
then
it's
too
early
to
have
a
hackathon
about
this.
Maybe
we
should
just
hack
something
ourselves
to
just
make
it
usable
like
to
make
status
more
usable
inside
the
company
inside
ourselves.
Don't
open
it
that
much
unless
we
have
this
actual,
like
protocol
design
and.
B
G
G
E
So
I
also
definitely
agree
that,
for
now,
what
we
want
is
to
have
some
BOTS
then
can
talk
to
public
channels,
probably
that
they're,
probably
that's
it
and
yeah
like
how
to
make
some
things,
and
we
definitely
don't
need
this
protocol
right
now,
but
it
would
be
at
the
same
time
a
big
right,
yeah,
somehow
envision,
all
this
stuff,
all
this
process
and
how
we
want
to
build
the
bots
right
now,
but
that's
not
like
the
final
state
of
it
and
how
did
Kenai
evolve
in
the
future.
So
it's
not.
E
G
G
A
What
do
people
think
of
what
the
caller
said
before
in
terms
of
interacting
with
contacts,
because
that's
like
form
of
bot
as
well
right
like
your
humans?
And
then
you
have
these
contracts
that
are
agents
as
well
and
not
some
bots
in
its
or
demon
sense.
But
in
the
computer
world
computer
sense
to
give
any
photon.
No.
A
G
I
mean
to
write
a
boat.
You
still
need
to
be
somewhere
to
end
this
boat,
that
it
sort
of
connects
these
two
worlds
right,
because
it's
takes
an
event
from
let's
say
a
smart
contract
and
then
notifies
whisper
about
this
event,
because
this
would
be
talking
about
right
like
when,
when
something
happens
to
smart
contract.
Somebody
signed
some
transaction.
There
is
a
message
in
some
channel.
G
J
You
are
the
message
to
trust
and
check
that
information
right
now,
because
we
don't,
we
don't
have
a
method
of
proofing
the
event
to
the
in
the
because
you
need
the
blockchain
sync
a
tool
to
check
the
information.
So
even
if
someone
says
I
said
maybe
in
few
future,
we
have
some
fancy
cryptography
to
solve
that
problem.
G
B
A
Any
other
thoughts
on
this
one,
you
I
question:
how
can
we
make
sure
that
this
is
sort
of
something
that
we
don't
spend
a
lot
of
time
writing
with,
but
stub?
We
should
make
it
you
will
make
it
well
documented,
and
then
we
let
the
communities
of
help
us
build
out
all
the
things
that
we
want.
Integration
for.
G
G
A
Think
we
probably
shouldn't
put
any
like
hard
dependencies
on
like
some
future
rewrite
protocol
instead
sort
of
assuming
that
what
we
have
now
is
what
we
have
and
then
we'll
see
in
a
future.
If
we
have
breaking
changes,
then
we'll
have
to
update
some
stuff
and
so
on,
but
that's
fine,
but
I
wouldn't
sort
of
assume
that
oh,
we
can
have
this
breaking
thing.
So
it's
not
worth
doing
anything.
That's
my
opinion
anyway.
C
J
G
C
Mean
and
then
with
this
I
mean
we
came
from
I'm,
just
thinking
about
it
from
a
user
perspective,
you've
got
like
no
idea
of
what's
happening
in
the
background.
How
would
the
client
then
display
it
like,
and
what
information
with
the
client
even
display
when
engaging
with
her
but
I
assume
the
best
user
experience?
Who
would
be
like
a
ninis
domain
name,
but
any
ideas
around
there.
G
B
C
Reason
why
I
mention
it
is
I'm
sure
if
you
guys
have
spent
a
bit
of
time
a
telegram
there's
like
a
ton
of
telegram
good
coin
spambots,
which
are
like
a
invest
in
this
Bitcoin
scheme,
sent
Bitcoin
to
this
this
key
and
then
we'll
like
invested
40
or
you
can
choose
to
invest
in
or
whatever,
there's
a
lot
of
that
so
I'm
just
interested
in
terms
of
like.
If
this
does
kick
off,
we're.
J
These
BOTS
would
need
to
like
send
a
message
to
too
much
people,
so
they
need
to
know
what
is
the
public
key?
They
are
saying
we
have
the
Nina's
usernames
that
will
help
them
figuring
this
out,
but
then
we
should
implement
something
like
the
tribute
to
talk
or
visibility
steak
on
public
chat
messages,
but
in
my
opinion,
you
should
not
defer.
You
should
not
know
if
it's
a
person
or
a
robot
just
by
looking
to
the
message.
So
this
is
a
feature.
G
Because
it's
the
same
as
and
the
protecting
from
phishing
should
be
the
same
like
because
right
now
somebody
can
try
to
impersonate
someone
from
status
team,
for
instance
on
the
public
channel
without
any
boats,
and
we
need
a
protection
from
phishing
period
or
from
spam.
But
it
shouldn't
be
like
different
from
the
both
agree.
C
A
G
Right
now,
yeah,
so
I
was
thinking
that
we
might
seems
right
now.
All
the
people
who
wants
to
start
with
automation,
at
least
I
heard,
are
essentially
from
the
go
side
of
thing.
So
I
think
that
we'll
start
with
probably
the
go
implementation
of
some
high-level
and
simple
API.
That's
and
we'll
try
to
focus
on
that.
It
won't
change
with
a
protocol
underlying
for
the
call
structure,
so
it's
like
maybe
probably
should
be
as
a
sort
of
business
oriented
or
whatever
we
go
so
there.
B
A
If
we
have
like,
if
we
can
write
up
like
some
high
quality
use,
cases
like
we
will
make
one
well
and
then
we
have
some
high
quality
use
cases
that
we
could
announce
that
the
hackathon
I've
had
like
five
large
bounties
or
something
and
people
can
try
to
and
kudos
or
whatever
I
think.
That
would
be
awesome.
I
G
F
G
C
G
Now
I
would
really
love
to
be
able
to
see
the
results
without
having
metamath
crying.
You
think
it's
told
ya,
because
right
now,
even
the
so,
maybe
it's
hard,
but
that
would
be
the
top
feature
for
me
this,
because,
as
of
now,
I
can
only
vote
them
on
mobile
in
general,
but
when
people
just
see
I
I
mostly
work
on
my
desktop.
So
when
I
see
you
have
comments
with
like.
Oh,
this
is
a
link
to
this
vote.
I
usually
open
to
the
desktop
and
nothing
happens
because
I
don't
have
three
installed.
G
C
H
H
B
B
A
B
A
C
E
I,
like
it
I
think
like
for
this
type
of
pings
about
in
the
school
I'm,
not
one
of
voting
for
anything
for
everything,
because
that
voting
is
not
the
best
way
but
for
just
selecting
topics.
I
think
that's
fine,
but
we
also
should
look
at
which
topics
are
necessary
to
be
discussed.
So
some
of
those
should
be
included
because
they
are
important
for
some
reason,
but
for
the
rest,
boating
some
sounds
sounds
good.
A
J
G
C
J
C
J
Lashing
this
lashing
reward
will
be
too
small,
I.
Think
tennis
city
for
this
lashing
is
also
too
small,
but
it
would
be
like
it
still
makes
a
problem
for
people
with
registering
beds
usernames.
So
if
you
make
the
smaller
you
make
the
you
make
easier
for
scratching
lots
of
names
and
like
they.
The
price
of
registration
is
just
because
of
security
like
to
reduce
this
and
this
brothers
in
hey,
give
a
reward
in
this
lashing,
so
I
think
we
can't
go
to
lower
than
10s
energy.
Only.
J
J
J
A
J
When
you
register
bad
username
you,
the
amount
you
do
paid
is
rewarded
to
whoever
inform
it
the
contract,
the
rule
that
you
that
you
broke
so
basically
all
say
ourselves
in
stages,
we'll
call
this,
but
anyone
should
be
able
to
close
it.
So
we
can
also
provide
a
know
that
you
can
that
you
will
recite.
So
that
would
be
when
we
implement
in
the
say,
to
solve
all
this.
So
we
can
discuss
this
later,
but
basically
using
the
status
notes.
They
will
send
this
message.
So
look.
This
is
a
you
infringing
username.
A
A
J
But
to
change,
we
need
just
a
consensus
stages
developers
because
it's
the
controller
that
needs
to
make
a
cow
in
it's
a
multisig
wallet.
So
in
future,
when
we
have
devoting
controlling
these
applications
and
when
you
have
democracies,
we
need
to
actually
a
proposal
to
be
accepted.
But
right
now
is
easier
to
change
there.
I'm.
A
C
C
A
Guess
by
default
it
should
be
up
to
the
swarm
and
then,
as
I
said,
that,
once
before
observe
it's
a
multi
stage
search,
there
should
be
some
kind
of
consensus,
and
if
you
want
to
change
it,
if
some
people
think
we
should
change
it,
we
can
take
it
up
in
all
equal
or
habitat
vote
or
whatever.
We
don't
have
to
decide
right
now.
I
call
it.
J
So
there
that
would
be
on
another
description
than
in
voting
the
apps
like
how
we
implement
that
voting.
Maybe
we
can
use
a
mean
like
if
I
say
with
with
10
status
and
others
say
it's
100.
They
can
take
on
me
not
on
that
or
they
mostly.
We
have
to
discuss
also
that
so
because
how
you
make
these
proposals,
you
make
a
lot
of
options
or
you
make
like
I
think
this.
A
Is
a
big
topic
and
to
do
it
properly,
we
probably
need
to.
We
can
discuss
a
different
call,
but
like
it's
like
a
big
governor's
topic
and
we
need
more
background,
I,
don't
think
we're
gonna
yeah
we're
not
going
to
solve
it
in
five
minutes
cool,
so
just
a
quick
announcement,
just
very
briefly
on
Sunday
Friday
Saturday
at
the
hackathon
in
two
weeks
and
we'll
all
sort
of
do
brainstorm
of
yours
and
so
on.
A
So
if
swarms,
the
teams
that
want
to
start
thinking
about
that
now
that
could
be
useful
and
then
the
whole
thing
is
nothing
can
read
more
about
it
discuss
and
then
we
all
seem
to
say
what
to
do
with
next
call,
because
I
think
next
one
will
be
in
Prague
or
the
people's.
What
should
we
skip
it
or
how
do
you
wanna?
Do
it.
A
B
Need
to
take
a
bit
more
than
ownership
of
our
build
process
and
Jenkins,
because
at
the
moment
we
rely
a
lot
on
own
yeah.
It's
been
an
exceptional
job
with
Jenkins,
but
I,
listen
personally,
I
feel
alive
and
enough
and
I
think
you
know
like.
Maybe
it
would
be
good
if
we
could
sort
of
like
if
you
could
share
just
the
robes
Newgate
has
access.
B
She
says
well,
the
most
common
problems
are,
you
know
like
witness
either
you
know
in
the
right
direction
so
that
we
can
sort
of
like
help
out,
because
at
the
moment,
if
Jacobi
is
offline
for
me
is
very
difficult
to
sort
of,
like
my,
you
know,
get
my
way
that
just
fix
the
deal.
Basically
so
yeah.
That's
all
I.
G
G
B
H
Agree,
a
Chickasaw
currently
has
no
way
of
running
builds
like
that
locally,
and
there
is
a
real
pain
and
ask
for
issues
with
Jenkins
bells
is
mostly
just
trial
and
error,
but
yeah
I
don't
think
there
is
a
good
alternative.
I
and
switching
everything
would
be
a
huge
up.
If
you
want
to
learn
about
more
about
Jenkins,
it
could
always
poke
me.
B
A
H
G
H
The
young
pinpoints
would
think
as
well
are
certain
Mac
OS
slaves,
which
I'm
actually
thinking
right
now,
because
before
it
was
all
manual,
so
the
different
drinkin
styles
will
strike
that's
different
depending
on
when
they
were
configured
because
there's
some
manual
steps
in
Jenkins
configuration
for
for
Mac
OS
of
this
for
us,
but
I'm,
trying
to
automate
it
right
now.
So
for
hopefully
future
all
the
Jenkins
lives,
for
my
guys
will
be
the
same
means
those
will
be
more
consistent.
Oh
my.