►
From YouTube: Core Devs Meeting #6 November 19th 2018
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
All
right
we're
live
on
YouTube,
so
welcome
everyone
to
the
seventh
chord
of
call
bit
of
a
hiatus
after
Def
Con,
a
few
things
talk
about
today.
The
first
is
around
so
if
they're,
new
team
structure
for
the
team
gauge
and
what
this
means
in
technical
terms
for
interface
and
so
on,
and
the
second
around
sums
of
engineering
process,
things
that
we
can
improve
and
finally,
on
the
PTP
integration,
then
we
got
some
time
after
we
can
talk
about
general
things
that
are
something
people
not
talk
about,
so
I
guess
eager.
B
Just
to
recap
of
someone
didn't
will
change
our
teams
slightly.
So
essentially
there
would
be
some
like
areas
of
responsibility
like
there's
a
key
card.
The
apps
wallet
chat
desktop
and
it's
poor
as
if
now
six
performs
terrorist,
and
so
all
the
responsibilities
that
used
to
be
in
the
Coliseum
in
the
big
coaching.
They
are
exploited.
These
ranges,
but
lesser
problems
occurred
a
couple
of
times,
so
what
I
was
trying
to
sing
conference
share?
Guess
what
I
wanted
to
discuss
here?
B
B
Is
this
that
the
code
structure
should
follow
that
in
structure
I
someone
remembers,
but
essentially
the
biggest
pain
points
before
and
probably
the
biggest
pain
point
right
now
and
the
biggest
productivity
penalty
will
get
in
the
places
where
we
have
to
synchronize
team
attempts
and
it's
especially
exaggerated,
so
Sweden
and
it's
not
sitting
in
the
same
office.
So
that
means
that
like
for
me
at
least
it
might
be,
it
might
make
sense
to
a
little
bit,
even
if
it
means
sometimes
week
at
work
but
to
separate
our
two
cents.
B
Quite
some
code
modules
and
things
like
that's
a
little
bit
more,
so
they
intersect
it
less.
So
we
don't
so
things
like
CI,
for
instance,
that
maybe
this
purpose
mobile
might
have
separates
CIS
women,
even
though
it
might
be
again.
Example
m4,
but,
and
then
piece
there
shouldn't
be
much
about
overlap
as
long
as
this
unit
is
working
should
be
that
much
more
like
if
some
team
wants
to
change
something
and
not
or
doesn't
I
think
so,
that's
essentially
all
I
wanted
to
point
out.
A
So
one
first
question
who
is
so
eager
soft
thinking
for
himself
to
be
the
team
for
this
new
core
team?
Who
else
would
I,
self-identify
or
some
form
of
I
had
team
leader,
whether
it's
chat
the
world
to
desktop
and
infinitives?
And
maybe
when
I
say
so,
your
point
of
view
still
some
vitally
is
for
desktop
Goran,
poor
wallet,
and
maybe
there
are
other
teams
to
be
informed
them.
I,
don't
know
exactly
what
the
latest.
C
F
From
my
point,
my
point
of
view,
specifically
with
respect
to
CI,
the
problem
isn't
modules
its
platforms
really
because,
right
now,
whenever
we
push
PR,
it's
like
they're
playing
blackjack,
you
know,
you
know
at
least
one
platform
build
is
likely
to
fail.
You,
you
only
don't
know
which
one,
if
you
know
what
I
mean
like
the
build,
could
be
perfectly
fine
and
it
could
fail
on
desktop
Linux,
for
example,
for
some
particular
reason.
That's
from
the
CI
point
of
view,
I.
A
Think
maybe
broader
than
the
CI
point
of
view
is
it's
also
just
ringing
up
to
interface
in
terms
of
responsibilities.
So
it's
very
clear
what
each
team
is
responsible
for
and
what's
up
that
it
this
looks
like
and
then
it
will
be
questions
like
there
more
detail
with
CI
and
so
on,
but
I
think
the
first
priority
should
be
to
figure
out
the
interface
in
terms
of
responsibilities.
A
F
So
we're
talking
about
internal
API
switch
is
something
I
think
we
agreed
to.
You
know,
follow
as
a
guideline,
so,
for
example,
each
module
would
have
its
own
API,
which
would
be
external
in
the
sense
of
the
module,
but
it
will
be
used
by
for
exemplary
modules
like
wall.
It
would
offer
an
API
for
for
chat
or
for
browser,
but
it
would
also
consume
the
API
of
idle
chat,
for
example,
but
I
don't
see
how
this
what
this
has
to
do
at
all
it
with,
for
example,
see
I.
B
F
B
F
H
H
B
B
A
D
Yeah
yeah
and
again,
like
guys
still
I
still
in
making
my
way
through
the
entire
code
base.
Yeah
it's
a
long
novel.
You
know
so
it's
taking
a
little
while,
but
I
think
it's
I
think
Yuki
they're,
like
different
I,
think
we
could
all
agree
that
they're
different
out
degrees
of
separation
right
and
then
and
I
think
the
person
speaking
before
me
was
alluding.
To
that
you
can.
D
D
First
of
all,
some
of
these
things
are
hard
to
separate
right
now,
right,
like
some,
are
really
hard
to
separate,
and
so
you
you,
you
start
at
the
code
base
level
and
you
know
within
the
same
repo
and
and
you
start
pulling
these
things
apart,
you
know
then,
and
then
you
see,
if
you
can
actually
even
you,
can
you
separate
it.
You
know,
you
know
some
things
may
not
be.
D
We
practically
made
separate,
and
some
things
may
be
made
separate
and
the
question
of
taking
it
to
another.
Repo
right
is
a
big
question
right,
that's
not
an
easy
that
is
not
necessarily
an
easy
lift
and
if
it
was,
it
probably
means
that
you're
only
separating
a
very
small
amount
of
functionality
right,
like
you're,
separating
on
a
thin
wrapper
around
the
the
goal.
Library,
and
then
you
know
what
what
do
you
gain
from
that
right
so
and
so
the
ideas
would
be
great
to
take
to
separate
on
a
larger
piece
of
functionality.
D
Right
we
want
to
separate,
and
so
at
the
highest
level.
Right
is
a
you
know,
react
component
which
you
could
drop
in
and
that's
chat
and
from
there
down
you,
if
you
can't
achieve
some
sensible
degree
of
separation,
you're
you're,
including
a
lot
and
you're
separating
a
lot
from
everything
else,
does
that
make
sense
to
do
people?
What
do
people
think
in
response
to
that
yeah.
F
F
How
would
you
deal
with
the
shared
stuff
with
components
and
things
like
that,
or
is
that
possibly
something
that
you
would
you
would
separate?
First,
for
example,
when
you
say
you
know,
one
part
that
strat
alone:
how
would
we
how
in
that
case,
would
be?
Would
we
deal
with
the
stuff
that's
shared
by
chat
and
wallet.
D
So
this
is
right
so
that
so
that's
coordinated
at
a
higher
level
right
so
like
like,
for
example,
I
think
it
was
Eric.
Just
did
a
lot
of
work
to
move
all
the
subscriptions
in
the
chat
and
the
chat
window
highest
like
he
moved
them
up
higher
right
and
then
basically,
essentially
making
the
coat
the
the
components
below
like
buying
buy
separately
right.
So
now,
all
the
components
below
are
stupid.
D
They're
just
getting
past
data
right,
they're,
not
subscribing,
they
knew
nothing
about
reframe
they're,
just
getting
past
data,
or
they
do
know
about
referring
because
they're
still
finding
they're
firing
events
to
reframe
and
stuff
like
that.
But
but
the
idea
is
just
making
a
bigger
component.
That's
stupid
right
in
in
the
sense
that
it
doesn't
know
about
the
greater
world,
but
it
does
know
we
do
know
it.
Fires,
events
and
it
accepts
data
right
so
that
it's
interface
internally.
Is
it
fires,
events
at
the
borderline?
Is
it
fires?
D
H
A
Hey
speaking
of
ideals,
you
would
be
really
cool
if
we
could.
It's
in
my
truck
I
love
changes,
but
if
you
can
sort
of
start
to
move
things
like
the
wallet
and
other
things
into
more
of
a
something
that
you
can
replace
like
it's
more
like
a
tap,
it's
useful
for
lots
of
things.
It's
useful,
both
from
as
a
general
separation
of
current
concerned
kind
of
point
of
view
that
we
are
working
on.
A
It's
also
useful
in
in
the
sense
that
if
we
have
a
network
and
people
might
want
to
have
a
different,
so-called
wallet
abstraction,
that's
not
as
wallet
centric.
Maybe
it's
more
around
collectibles,
it's
useful
for
absol
approvals
and
so
on
and
just
treating
the
transaction
signing
as
the
core.
And
then
the
wallet
is
just
another
DAP,
that's,
maybe
more
polished
and
so
on.
What
are
people's
thoughts
on
the
feasibility
of
moving
to
that
I
know
we
used
to
have
it,
but
then
we
moved
away
from
it.
Yeah.
A
That's
unspecified
right,
I
mean
it
could
be
just
no
I
mean
because
it's
more
of
an
idea,
as
opposed
to
specific
way
of
doing
it,
it
could
be
like
a
web
dab.
It
could
be
extensions,
it
could
be
like
some
reconnected
component.
I
would
say
that
something
you
a
useful
property
would
be
if
you
can
trust
minimize
it.
So
transaction
signing
has
some
kind
of
privileged
thing,
where
it's
clear
that
this
is
council
status.
A
So
you
can
honest,
install
like
extensions
that
are
maybe
not
as
trusted,
and
they
are
more
like
a
nice
UI
because
you
still
need
to
do
have
some
abstraction
that
sort
of
is
cure,
UX
for
transaction
signing
and
key
management,
and
so
on.
That
will
be
the
main
criteria
and
swap
ability.
Whatever
that
looks
like
is
unspecified.
A
F
A
A
F
I
I
So
as
far
as
that
line
of
thinking
is
concerned,
I
mean,
if
you
look
at
Android
as
an
operating
system.
You
have
both
like
the
facebook
messenger,
which
is
most
egregious
from
from
the
privacy
point
of
view,
and
then
you
have
on
the
way
down
to
Breyer,
which
which,
which
does
try
the
check
very
well
and
likewise
for
wallet.
We
could
have
a
multitude
of
wallets
supplied,
but
that
sort
of
depends
on
on
the
status
Korac
being
developed
with
the
secure
operating
system.
Mindset.
F
D
Right
I
mean
I,
think
so
it's
to
me,
it
seems
like,
like
the
last
two
comments
right
like
about
it
like
these.
Are
these
overarching
ideas
of
yeah
making
things
like
I
would
say:
that's
making
things.
There's
a
yeah
there's
an
ideal
right
of
making
things
super
general,
like
operating
system,
moving
moving
to
a
death
and.
D
These
things
are
like
these
things.
These
things
seem
to
conflict
right
with
some
fundamental
friction
points
with
where
we
are
right
now
and
so
I
I,
don't
know
like
in
terms
of
like
in
terms
of
these.
Actually,
these
seem
like
far-reaching
future
ideals
that
don't
that
don't
speak
to
you
know
what
we're
gonna
do
for
the
next
six
months,
like
now,
maybe
I'm
wrong,
but
but
in
terms
of
movement
like
well
I'm
moving
to
adapt,
it
is
an
ideal.
D
D
You
know
some
of
these
native
things
that
we
really
like
now
I
do
what
I
do
see
as
a
possibility
right
is
moving
the
eight
you
know
again,
making
the
api's
a
solid
boundary,
and
if
you
want
to
move
that
boundary
and
to
go
or
whatever
and
making
that
you
know
again,
you
know
in
breezing
up
the
protocol
to
the
point
where
it's
more
general
and
and
again
making
the
UI
less
knowledgeable.
But
but
right
now
I
mean
we're.
D
G
Say
something
I
guess
so
speaking
to
your
to
your
point
about
these
things
being
so
like
general
and
and
idealistic
let's,
let's
look
at
a
very,
very
specific
threat
model
at
the
moment,
which
I
think
is
brought
up
the
discussion
about
moving
the
wallet
to
adapt
and
that's
kyc
and
AML
and
plausible
deniability
about
whether
we're
a
financial
service
or
not.
So
if
it's
not
a
question
of
moving
it
into
a
doubt,
how
do
we?
G
How
do
we
accomplish
that
with
our
code
base,
that
we
create
some
sort
of
separation
between
statuses
of
company
or
as
a
dowel
and
and
having
a
wallet
that
provides
access
to
financial
services
where
it
looks
like
the
direction
is
going,
we
would
have
to
provide
and
collect
customer
data.
So
how
that
that's
a
real
practical?
You
know
use
case
about
that.
We
need
to
solve.
So
where
do
we
go
to
do
that,
based
on
our
code
base
and
your
from
your
point
of
view,
yeah.
F
G
F
Practically
one
of
the
sources
of
confusion
is
the
definition
of
a
wallet.
What
is
the
wallet
I've
seen
at
least
three
separate
three
separate
well
well
accepted
definitions,
first
wallet
as
a
safe
storage
for
your
kids.
That's
like
the
most
minimalistic,
a
definition
of
wallet
like
literally
a
kid
store.
The
second
one
is
a
client
that
you
use
to
access
a
theory
and
to
send
transactions
in
the
most
general
sense
of
all
transactions.
So
that's
that's
the
third
one.
F
F
So
practically
be,
for
we
start
this
discussion.
We
need
to
agree
on
when
we
say
wallet.
What,
specifically,
does
that
mean
and
to
what
extent
does
it
fit
into
any
of
the
existing
pigeonholes
made
by
made
by
the
regulatory
bodies
and
so
on?
The
one
interpretation
of
all
that
I've
seen
among
users
is
something
where
you
put
your
assets,
so
basically
one
guy
literally
yeah
one
guy
literally
asked
me.
He
said
something
like
I
migrated.
My
wallet
from
Toshi
to
status,
I
opened
my
Toshi
and
I
can
still
see
my
tokens
there.
F
Wallet
is
a
container
where
your
assets
are
stored,
which
is
also
kind
of
technically
incorrect
in
the
sense
that
the
assets
are
stored
on
the
blockchain
on
every
single
full
node
out
there,
and
the
wallet
is
like
a
key
store
or
a
client
app
that
you
use
to
access
those
those
funds.
So
when
we
say
wallet
we
need
to,
we
need
to
agree.
F
What
specifically
to
me,
do
you
mean
by
by
it,
for
example,
right
now
wallet
from
the
first
definition
is
in
a
key
store
is
stored
within
the
diac,
and
the
only
way
we
can
get
rid
of
it
is
to
export
it
to
a
cart.
Optionally
right
wallet
is
the
second
definition
client
app
for
deuterium
again,
we
need
to
agree
to
what
extent.
F
To
what
extent
do
we
need
to
keep
or
remove
features
in
order
not
to
qualify
as
a
financial
service,
because
you
know
if
you
want
to
interact,
if
you
want
to
mutate
a
chain
at
all,
you
have
to
create
transactions
in
the
technical
sense
now,
I
know
there
are
transactions
as
much
as
database
transactions
or
transactions,
but
still
they
are
called
transactions
all
right.
So
you
see
where
I'm
getting
with
this.
F
Practically
did
it's
like
it's
like:
we
need
to
set
a
bar
exactly
how
much
do
we
want
to
offer
natively,
and
how
much
can
we
remove
and
put
to
some
kind
of
an
extension
or
on
an
optional
upgradeable,
downloadable
or
whatever
part,
because
some
something
we
definitely
need
some
things
we
definitely
need
to
have
internally.
One
of
the
things
I
was
thinking
about
is
I
mean
names,
names
names
are
powerful
names,
define
things.
F
We
may
even
rename
some
some
of
the
concepts
which
could
be
even
necessary
from
a
UX
perspective,
because
the
harder
role
it
is
and
and
the
wallet
as
a
tab
in
the
app
where
you
can
see
your
balance.
Those
are
two
completely
different
things.
Both
are
called
wallet
and
wallet
to
someone
who
is
non-technical
is
something
that
has
to
do
with
money.
Therefore,
therefore
financial,
can
we
call
them?
Can
you
give
them
different
names.
G
G
Just
separate
those
talk
about
signatory
functionality,
notarization
in
that
sense
and
then
and
then
assets,
storage
and
evolved,
and
if
we
can
separate
those
things
so
that
we're
not
only
the
providers
of
them
and
those
those
things
live
elsewhere
in
the
app
our
application
just
excesses
them.
It
might
be
a
way
that
we
can
accomplish
that
yeah.
E
The
wallet
is,
is,
by
all
instances
of
purposes,
just
a
password
manager,
a
little
bit
of
functionality
around
it
in
terms
of
met
like
deriving
passwords
and
signing
things.
That's
literally
it
and
the
messaging
should
be
around
that,
because
the
actual
things
that
value
don't
live
anywhere
on
those
keys.
It's
just
that
we're
just
holding
the
keys
that
can't
access
the
things
of
value
and
that's
a
problem
like
the
definitions
of
these
things
that
were
worried
about
code
regulation
is
a
problem
with
the
regulatory
bodies
which
we
cannot
get
around
so
like.
E
How
do
you
move
forward
to
you?
Do
you
do
you
change
your
signaling
to
deal
with
the
regulation
and
you
your?
Do
you
signal
appropriately
to
the
technology
and
hope
that
they
change
I?
Don't
I!
Think
that's
that's!
The
question
here
is
because
the
whole
whole
Kuehl
peasants
issue,
calling
it
a
wallet
makes
it
worse.
Yet
it
they
have
their
they're,
more
ignorant
about
how
it
actually
works,
and
they
can't
they
can't
reason
about
it
appropriately,
and
but
how
do
you
move
forward?
E
Because
that's
what
everyone
calls
it
I
personally
am
very
strongly
at
the
opinion
that
we
changed
the
names
so
that
they
match
the
intuition
and
what
the
thing
actually
does.
Corleone
password
managers,
key
stores,
key
cards
etc
gives
the
users
a
better
if
it's
Richard
or
what's
going
on
so
that
when
they
say
I
recovered
my
wallet
in
status.
When
I
see
my
assets,
how
does
that
work?
If
they
ever
said,
I
brought
my
keychain
over
to
status
and
I
can
see
my
assets.
That
makes
sense,
because
I
still
have
my
keys.
E
F
Yeah
and
on
the
other
side
of
the
specter
of
the
spectrum
status
also
needs.
We
have
access
to
the
network,
so
you
know,
there's
like
a
zone
on
one
side,
there's
a
net
work
on
the
other
side
when
you,
if
you
remove
the
ability
to
transact
from
a
Nigerian
client
and
if
you
make
it
option
or
something
that's
achievable
through
a
plug-in.
F
E
The
net
case,
like
the
actual
financial
services
Liza,
then
giving
the
availability
for
that
for
that
option.
Right
like
that
file
system
access
is
the
actual
service,
because
that's
what
the
logic
is
of
getting
access
to
those
things,
whereas
the
keys
are
just
their
keys,
are
just
keys
right,
we
whisper
keys,
we
have
wallet
keys
and,
and
if
about
the
functionality,
they
all
look
the
exact
same,
and
so
the
the
app
itself
is
the
is
the
access
to
financial
services.
So,
if
someone
deems
that
what
it
is.
B
D
Isn't
there
I
mean
in
the
possible
deniability
that
letting
people
install
adapt
I
mean
the
you
know,
the
regulatory
bodies
will
move
these
goal
posts
continuously
and
we're
not
in
any
different
of
a
position
than
any
other
wallet
out
there
or
any
other.
You
know
like
these
goal.
Posts
are
gonna
move
and
who
the
heck
knows.
You
know
I
mean
the
what
if
here
and
the
timeline
here
again
is
moving
to
a
depth?
What
was
that
by
you
know
like
it
looks
like
there's
a
wallet
there.
D
G
Not
I
think
that
the
technicalities
of
whether
it's
adapt
or
not
in
the
eyes
of
someone
like
Apple
or
a
government
is
going
to
be
available.
They're
gonna,
put
the
hammer
down
on
status
could
get
the
pallets
operated
by
us.
Turning
around
saying.
Well,
technically,
it's
a
third
party
DAP,
and
this
it's
not
gonna,
make
a
difference.
You
can
access
this
stuff,
which
is
you
know,
let's
say
not
allowed,
is
going
to
be
the
end
ruling
I.
G
I
F
Yeah,
just
one
question:
we're
talking
about
two
separate
classes
of
attackers,
one
the
likes
of
Apple
and
the
other,
the
right
so
financial
regulatory
bodies.
Now
do
you
know
that
Apple
doesn't
actually
need
a
reason
or
a
cause
to
shut
you
down
as
per
the
Terms
of
Service,
and
they
don't
even
have
to
explain
that
after
they
thought
you
had.
A
Imagine
if
the
Apple
2
came
with
a
BitTorrent
client
pre-installed
that
maybe
it
would
be
banned
because
it
was,
it
was
sort
of
against
Coco,
privé
way
or
whatever,
like
it's
about
separating
out
piece
of
functionality,
and
if
we
take
the
idea
of
status
being
an
operating
system
for
a
theorem
seriously,
then
it
kind
of
implies
this
type
of
design.
To
some
extent.
G
I
Yeah
and
that
doesn't
really
prevent
us
from
developing
a
quality
chat
and
a
quality
wallet
and
a
quality
integrated
experience
between
just
these
two.
But
it's
it's
the
mindset.
It's
approached
culture
that
we
have
in
developing
and
that
will
lead
to
like
a
share
at
understand
of
the
fact
of
what
it
is
when
we're
building
a
glue
movie
and
every
one
of
us
making
small
decisions
that
push
in
this
direction
and
that
that's
important,
especially
important
in
the
decentralized
organization
right
that
you
have
this
shared
idea,
so
that
we
can.
I
You
know
if
Trust
is
the
right
word
here,
but
that
we
can
be
comfortable
with
allowing
each
and
everybody
to
decide
on
the
many
small
decisions
that
lead
up
to
a
quality
whole
like
for
that
to
happen.
We
need
to
have
this
shadow
understanding
of.
Why
would
white
as
we're
doing
and
how
we're
doing
it
and
why
it
is
we're
doing
it
in
that
particular
way.
B
Yeah
I
hope
I
understand
like
every
one
correctly.
That's
it's
it's
important.
That's
we
are
having
some
yeah
sure
go,
but
it
doesn't
mean
that
we'll
just
throw
everything
out
of
the
window
that
we
have
right
now
and
just
start
everything
from
scratch
and
just.
But
if
we
in
making
some
decisions,
we
should
technical
decisions
as
well
should
just
look
at
this.
Go
so
like
even
like
inclusion.
B
If
we
improve
separation
and
like
internal
API,
it's
like
between
the
modules,
even
even
if
they're,
technically
not
the
apps
and
not
accept
or
depositors
and
stuff
like
this-
will
learn
what
API
side
necessary
right.
So
otherwise
we
can
theory
ties
on
the
API
sits.
Well,
if
will
be
our
first
users
and
it's
like
implementing
some
internal
API
separation
between
components.
Then
it
would
be
much
easier
to
just
make
those
Journal
after
a
proper
security
or
content,
but
yeah
things
like
that.
J
Regarding
the
financial
risk
and
the
App
Store,
it's
just
wanted
to
add
that
our
legal
counsel,
Sonya,
is
researching
the
topic
and
working
with
other
arms
law
firms
on
a
an
illegal
determination
on
the
the
best
way
Krissi's.
If
this
and
possibly
you
know
it
could
make
sense
to
wait,
for
you
know,
precedents
with
other
apps
before
making
a
decision
right
now,
but
I
think
we
do
have
a
little
bit
of
time
and
that
there
is
some
work
on
going
to
look
deeper
into
it
from
compliance
and
regulatory
point
of
view.
K
I
wanted
to
ask
a
question
about
this
modular,
ization
and
component
stuff,
because
so
we
have
lots
of
functionality,
so
we
can
kind
of
imagine
it
all
on
the
two-dimensional
plane
so
to
speak
and
right
now
we
kind
of
a
split
it
along
the
y-axis.
So
we
have
different
layers
so
fact
that
correspond
to
different
parts
of
functionalities,
but
we
also
have
like
the
x-axis
that
corresponds
to
let's
say
that
levels
in
the
technological
stack
so
as
in
order
to
diminish
and
decrease
the
level
of
teams,
sync
required
as
Igor
explain
it.
K
So
it
would
be
really
nice
that
whoever
the
features
are
going
to
implement
it
by
multiple
teams
potentially
right.
So
we
both
need
to
ensure
that
the
feature
is
once
it
is
delivered.
It
is
either
a
separate
component
or
as
separate
as
possible
that
it
has
a
well
defined
API,
most
like
with
some
nice
I
could
actually
decision
records
attached
to.
It
is
a
documentation,
but
also
as
I,
understand
it
sort
of
application
steps
for
each
of
the
stakeholders.
That
would
say
the
Guardians
on
each
of
those
tech
stack
levels
right.
K
Is
that
something
that
we
are
having
already
or
are
going
to
have,
because,
given
the
at
least
two
dimensional
nature
of
the
product,
each
parts
of
each
feature,
they
have
to
basically
to
go
through
two
pairs
of
eyes.
Both
the
free
feature
implementers
themselves,
who
guarantee
that
the
feature
will
be
a
complete
piece
of
functionality,
wise,
but
also
from
the
respective
let's
say
key
holder
of
the
technology
stack
that
this
feature
will
will
touch
upon.
B
K
Was
that
whatever
features
will
implement,
they
need
to
have
like
really
strict
boundaries
and
in
terms
of
description
of
what
they
do
expected
inputs
and
outputs,
so
that
once
is
developed
and
yon
code
wise.
This
moment
multiple
fears
whatever,
but
this
has
to
be
yea
really
encapsulated,
not
not
only
combines
but
also,
let's
say,
kind
of
organizationally
or
documentation
wise,
so
that
people
know
what
we've
got
meant
it.
How
does
it
interface
with
other
parts
of
the
codename,
because,
right
now
we
just
have
like,
let's
say
an
unstructured
stream
of
PRS.
K
To
parse
this
dream
of
PRS
work,
they
used
give
history
to
understand
how
something
happened.
What
are
the
limitations
and
they
have
to
look
at
multiple
places
to
leave
out
all
of
important
information,
so
we've
given
this
little
implementation
for
features
that
we
minimize
this
time
required
for
searching,
and
you
know,
gluing
together
different
pieces.
Let's.
H
A
I
think
yeah.
We
were
talking
about
this
first
item
for
40
minutes,
so
I
think
it's
important
conversation
but
well
move
on
so
in
terms
of
following
up
with
this
sort
of
operating
system
that
wallet
API
versus
product
thinking
like
what
do
people,
what
are
people's
thoughts
in
terms
of
how?
How
we
best
should
follow
up
on
this
because
I
agree,
which
I
think
it
does
like,
was
talk
about
it
that
it's.
A
If
we
don't
have
someone
shared
mine
on
this
matter,
there's
a
risk
that
we're
pulling
aside
different
directions
and
that
are
some
sort
of
preclude
like
even
having
the
operating
system
thing
that
doesn't
preclude
us
having
product-focused
teams
and
these
types
of
things.
But
what
should
we
do
in
terms
of
continuous
conversation
and
then
coming
to
some
kind
of
shared
on
sunny
one
of
people's
thoughts?
It.
D
Would
be
interesting
if
there
was
a
like
either
teams
I
think
if
people
I
think
would
serve
communicating
quickly
more
quickly
and
effectively.
If
maybe
we
had
a
discuss
where
people
or
teams
posted
like
whiteboard
drawings
or
paper
drawings,
or
something
like
that
and
I,
don't
know
some
some
way
and
then
maybe
having
another
meeting
where
people
could
talk
about
the
specifics
of
what
they
have
in
their
mind,
because
you
know
I
think
there
may
be
15
different
pictures
in
people's
minds
of
what
what
this
means
and
how
a
good
look
right.
A
So
that
is
a
good
point,
so
whoever
sort
of
wants
to
have
some
ideas
rather
can
post
something
on
discussing
and
we
can
talk
about
there
and
then
we
can
bring
it
up
at
the
next
call
or
have
a
dedicated
call
to
sounds
sanguine
to
people.
I
know
if
having
designers
on
this
call,
but
whatever
everyone
can
just
draw
a
name,
is
paint
or
something
if
they
want
all
right
cool,
so
I
guess.
Moving
on
to
the
second
point,
unless
someone
wants
to
say
some
final
words
on
that
topic,.
A
Whoever
feels
divine
inspiration
can
do
a
synchronization,
definitely
cool.
So
second
one
is
around
sort
of
engineering
process,
failures
and
I.
Think
this
is
touched
on
several
topics,
but
idea
is
that
we
want
to
have
higher
standards
for
general,
Indian
practice
and
I
think
this
applies
from
several
ways.
One
is
in
terms
of
good
issue,
description,
stand
and
to
big
PRS
and
automated
tests
still
not
being
something
developed.
You
can
rely
on
too
many
pull,
requests,
open
and
and
so
on.
I
think:
there's
lots
of
things
we
can
talk
about
here.
A
I
know,
I,
think
patrols
had
some
general
thoughts.
I
guess
we
can
leave
it
as
an
open
table
in
terms
of
taking
this
opportunity
of
this
sort
of
reorganization
to
setting
ourselves
up
for
higher
standards
and
and
sort
of
evolving
as
an
Indian
organization,
and
not
just
going
back
and
forth.
Having
the
same
conversation
we
had
a
year
ago
when
it
comes
to
PR
sites
and
making
issue
problem
descriptions
and
these
types
of
things
I
don't
know
better.
Do
you
want
to
start
with
what
you
were
thinking
about.
H
Yeah
well,
one
thing
I
mentioned
this
morning
in
another
meeting-
was
that
we
need
to
up
the
level
from
the
beginning.
You
know
from
issue
description.
We
need
the
person
who
is
writing.
An
initial
description
needs
to
think
about
all
the
stakeholders
down
the
line,
so
could
be
the
system
tester
the
developer,
the
reviewer
try
to
make
it
as
complete
as
possible
in
terms
of
information
so
that
we
have
more
information.
H
H
If
something
is
not
specified
in
the
issue,
and
then
we
have
to
go
and
ask
no
private
management
or
the
person
who
wrote
the
the
initial
issue,
but
we
I
think
we
need
to
be
more
demanding
in
terms
of
the
quality
of
the
of
the
whole
process,
not
just
all
requests
but
from
beginning
to
end.
It
might
delay
us
in
the
short
term,
but
I
think
it
will
increase
the
speed
of
the
team
in
the
long
run.
H
H
H
Demonstration,
for
instance,
the
the
push
notifications
on
the
on
the
desktop.
You
probably
noticed
that
there
was,
you
know
the
lack
of
icon.
Then
there
was
this
default
icon,
which
showed
the
zet,
which
meant
nothing
for
status
users
and
and
also
the
tests.
The
text
that
was
showing
the
push
notifications
was
not
very
helpful.
It
didn't
say
who
sent
a
message
or
in
which
chat
I
mean
this
is
something
that
a
should
have
been
more
specified
in
the
issue
that
started
this
implementation
and
B
should
also
have
been
caught
in
the
in
the
pull
requests.
H
A
review
asking.
You
know:
why
do
we
have
this
particular
icon
and
not
the
status
icon
and
white?
So
I
guess
the
reason
was
that
the
pull
request
was
more
based
on
the
technical
side.
Okay,
now
we
have
push
notifications,
which
is
the
the
major
part
of
the
issue,
but
I
don't
see
a
real
reason
not
to
do
those
final
touches,
and
you
know
maybe
it
wasn't
in
the
specification
but
I
think
it's
something
that
needs
to
be
touched
in
the
in
the.
H
In
the
pull
request
review
saying:
okay,
I
see
all
the
work
is
correct,
but
we
could
also
have
this
because
it
doesn't
make
sense
not
to
add
the
status
icon
or
this
or
that-
and
this
is
of
course
a
very
simple
example:
it's
the
simplest
eye
I
could
find,
but
so
you
could
transpose
this
to
two
other
issues.
You
know
much
more
complicated
where
I
mean
the
lack
of
unit
tests
or
the
lack
of
system
tests,
you
could
say:
okay,
we
fix
this
issue.
H
An
extra
pop
hidden
could
be,
for
instance,
Roman
is
working
on
the
speeding
up
the
login,
which
is
great
but
I,
hope
that
we
will
write
system
tests
or
end-to-end
tests
to
verify
that
we
don't
have
a
regression
on
on
that.
You
know
it's
just
taking
the
taking
things
the
whole
way
and
I'm
sure
that's
a
Roman.
Is
it's
going
to
work
on
or
to
delegate
or
whatever,
but
this
is
something
that
is
very
important
so
that
we
don't.
H
F
Yeah,
it
does
one
thing,
I'd
like
to
add
to
what
you
already
said
in
the
can't
stress
it
enough.
We
need
to
have
screenshots
in
peers,
because
some
of
the
things
like
the
ones
you
described
become
very
obvious,
very
visible
to
anyone
who
is
reviewing
reviewing
it
if
they
can
only
see
a
screenshot
yeah.
Of
course
you
can
always
join
rantings
yourself,
but
screenshot
is
really
fast.
Think
I'm
actually
I'm.
H
Actually,
a
big
proponent
of
you
know:
if
you're
reviewing
the
codes,
you
should
run
it
to
yourself
and
you
might
find
issues
with
with
the
total.
It
was
just
the
case.
This
weekend,
I
was
working
with
Roman
to
to
test
this
database
migration
and
on
my
database
it
showed
issues
that
we're
in
Ch
wasn't
showing
up
in
the
news.
Migration,
so
I
think
we
need
to
get
more
eyes
into
into
the
full
request.
I
mean
for
the
testers.
H
Sometimes
the
technicalities
are
not
visible,
for
instance,
in
the
in
the
simple
case
of
the
of
the
push
notification,
maybe
the
tester
thinks
okay,
this
wasn't
implemented
it's
because
it
was
a
hard
technical
issue
that
you
know
the
developer
can't
easily
put
the
status
icon
on
the
notification,
so
it
will
be
left
for
later,
and
it
wasn't
the
case.
So
other
colleagues
who
are
reviewing
the
pull
request
have
more
information
to
decide
whether
the
pull
request
should
pass
or
a
little
bit
more
work
is
needed.
A
Think
one
general
point
is
also
I,
see
very
few
pull
requests
being
rejected
and
the
same
with
sort
of
questioning
it
problem
statements
and
so
on
and
I
think.
Maybe
that's
easy,
because
people
don't
want
to
seem
like
an
especially
if
it's
like
a
distributed
organization
and
feels
like
you're
attacking
the
person.
A
H
That's
a
good
point:
I
once
worked
in
a
team
where
one
developer
took
things
very
personally
whenever
in
a
pull
request
was
rejected
and
we
had
this
big
meeting
saying
you
know
the
theme
was
you
are
not
your
codes,
you
know
you
have
to
be
able
to
separate
no,
and
it
happens
to
me.
Sometimes
things
like
criticize
on
a
pull
request.
I
do
it.
You
know
some
other
time.
H
K
Well,
sometimes,
PRS
really
get
stuck
in
code
review
for
well
I
used
to
have
this
like
a
long
time
ago,
but
actually
quite
petty
reasons
like
might
be
some
naming
and
stuff
but
I.
Think
in
general,
the
question
of
personally
taking
barriers
close
to
heart
is
in
nineteen.
Ninety
percent
of
cases
is
basically
solved
by
proper
choice
of
words
and
stuff.
You
know
people
I
would
never
be
offended
by
they
are
being
projected
because
there
are
some
reasons.
I
would
rather
be.
You
know
a
bit
offended
if
there
was
some.
You
know
important,
appropriate
wording.
K
I
think
that's
applies
generally
to
everyone.
So,
but
from
that
I
don't
see
and
the
reason
it's
really
to
take
the
guards
personally
and
I,
don't
think
we
have
any
problem
trouble
with
regards
to
inappropriate
wording
at
all
so
yeah.
Maybe
that's
just
need
to
turn
on
a
switch
inside
the
habit.
Yeah,
it's
fine
to
to
be
a
bit
more
strict.
That's
it.
A
G
F
A
F
A
Something
I
think
we
do
something
wrong
like
in
one
of
the
PRS
it
it
looks
like
it
was
passed
on
to
QA,
but
then
it
ended
up
in
a
contributor
column
with
no
action
or
anything
for
the
reviewer.
So
it's
been
hanging
it
for
a
month
and
the
contributors,
probably
just
waiting
for
us
and
too
timid
to
ask
us
what's
going
on,
and
this
happened
for
two
to
PRS,
that's
more
than
month
old
I.
Just
looked
at
all
those
peers-
and
that's
just
us
dropping
the
ball
like
this-
is
not
the
contributors
fault.
A
This
specific
case
I
think
it
was
yeah.
It
was
one
person
who
proved
on
one
or
less
common,
but
but
I
think
that
the
point
is
more
that
the
idea
with
something
like
through
your
today
is
to
catch
these.
These
things.
So
looking
at
the
oldest
PRS
and
just
ping
me,
if
nothing
has
happened
and
having
some
maintain,
where
we
make
sure
that
these
things
don't
fall
behind
the
cracks,
and
sometimes
that
means
closing
a
pure,
absolutely
but
yeah.
H
F
A
H
F
H
F
And
it's
even
automated,
so
there's
no
ambiguity
around
reading.
It's
it's
definitely
not
personal.
It's
a
machine,
yeah
right,
yeah,
all
right,
the
the
other
thing
around
something
you
said
Pedro
like
when
we
receive
an
income
incoming
PR.
We
are
reviewing
it
and
maybe
everything
is
technically
correct,
but
maybe
we
want
to
add
something
to
the
scope,
because
it
makes
sense
at
the
moment
I'm
not
sure,
that's
a
fair
thing
to
do
to
bounty,
pr's.
I
F
F
A
F
A
F
Yes,
probably
probably
the
thing
is,
some
of
them
do
exist
for
a
reason,
so
maybe
we
could
introduce
labels
for
exceptions
like
keep,
keep
me
alive
as
long
as
we
don't
abuse
it.
But
right
now
it's
really
throwing
dust
in
in
our
eyes,
because
we
look,
for
example,
every
now
and
then
I'd
take
a
look
if
any
of
the
pull
requests
need,
reviews
and
that
sift
through
40,
different
pr's
and
they'd
only
be
able
to
review,
because
most
of
them
aren't
actually
waiting
for
reviews.
K
K
H
Getting
your
builds,
no
CI
builds.
You
start
testing
on
multiple
platforms,
for
instance,
if
I'm
developing,
let's
say
the
windows
support,
PR
I
will
start
seeing
it's
executing
on
the
build
server
and
also
testing
that
what
I
do
doesn't
break
Mac
OS,
which
I
don't
have
a
Mac
o
Mac
machine.
So
in
that
sense
it
helps
you
know,
and
it's
also
becomes
a
place
where
you
puts,
as
you
are
doing
things
you're.
You
know
doing
the
explanation
there.
A
L
Can
like
speak
about
this
I
will
try
to
be
as
possible.
So
there
are
several
examples
in
like
regarding
not
reliable
tests
so
67-64
they
are.
It's
really
a
rare
cloud
infrastructure
issue,
but
anyway,
I'm
going
to
contacts
all
slaps
in
order
to
find
a
way
of
avoiding
situation.
Maybe
some
status
called
before
running
the
test,
like
it's
a
caterer
on
now,
every
surface
table,
etc.
So
other
two
issues
of
basically
starts
the
same
issues
into
cars,
6731
and
6752.
L
One
of
them
is
relation
product.
It's
about
wrong
tree
direction.
After
setting
up
for
that,
after
doing
the
transaction
in
chat
and
it's
reduced
rate
also
linked
and
as
a
two
issues,
there
were
several
attempts
to
fill
them,
but
new
issues
that
came
up
so
we
are
going
to
disable
them
until
being
fixed
and
for
all
issues
in
pairs
which
are
violent,
cause
and
ones
before,
because
once
it's
like
the
shortest
occurrence
Ricky,
we
can't
predict
that.
L
A
L
Right
but
we
can
implement
that,
but
we
need
to
think,
but
first
of
all,
we
is
better
to
implement
desktop
test
against
peers.
First,
because
now
we
have
more
powers
for
desktop,
and
some
pears
for
mobile
I
mean
like
we.
We
need
not
to
block
only
by
mobile,
but
to
finish
with
desktop,
runs
against
power,
and
then
we
will
implement
this
feature
and
if
tests
false
felt
it,
it
will
be
just
to
say
what
until
being
fixed.
So
it
will
solve
the
issue
for
spoiler.
F
A
F
F
L
A
L
A
L
I
sent
one
because
you
need
to
understand
like
the
whole
scenario
of
the
test.
You
need
to
like
to
go
much
deeper
and
we
definitely
don't
need
to
waste
developers
time
for
doing
such
things,
but
the
only
things
which
can
help
is
adding
accessibility,
ideas
for
new
component
and
not
removing
them
from
old
one.
H
Yeah
I
mean
this:
is
this?
Could
if
we
want
us
to
go
to
that
level,
we
could
specify
this.
In
addition,
in
the
issue
saying,
okay,
this
accessibility
ID
will
be
added
since
we're
asking
to
ask
to
add
a
button.
Then
this
ID
will
be
used
and
then
both
the
developer
and
the
tester
know
what
to
write
against.
A
So,
if
there's
some
way,
we
can
something
we
can
do
in
terms
of
how
the
research
is
run.
What
tastes
we
have
enabled
how
how
random
this
sort
of
successes,
or
whatever,
whatever
it
is,
but
like
we'd
like
what
do
we
need
to
do
to
make
we
get.
So
we
get
to
understand
like
soon
and
it's
this
isn't
the
thing
that
we
keep
talking
about
every
month.
L
L
We
need
to
change
this
happen,
that
in
the
PR,
and
to
provide
the
commitment
side
zipper
and
from
our
site
I
think
we
will
add,
like
weekly
person,
for
taking
care
for
all
pairs
and
also
would
be
great
to
have
some
kind
of
a
status
bot
or
like
it
was
just
like
where
we
can
get
that
results
directly
to
channel
to
one.
It
would
be
really
helpful
so.
G
A
We
can
briefly,
as
so
so
I
guess
we
can
take
the
ultimate
test
if
we
can
said
sometimes
something
will
target
for
that
or
yeah.
This
coordinators
are
just
meeting
what
we
need
to
do.
I,
don't
know
what
the
status
header
do
you
want
help?
Could
you
help
out
with
the
guitar,
but
in
terms
of
changes
there
sure,
and
then
we
can
just
make
that
a
goal
to
make
it
blocking
as
soon
as
possible
and
then
start
to
build
on
that
once
we
have
hundreds
and
I.
G
Mean
sorry,
go
ahead,
no,
no
go
ahead.
Now
I
was
gonna,
say
the
one
thing
that's
not
in
the
agenda
that
I
think
is
quite
timely.
Is
okay,
ours
I'm
like
wondering.
If
there's
any
update
on
that,
I
was
supposed
hoping
that
some
parts
meaning
would
talk
about
that,
but
how
our
team
is
progressing
I,
don't
know,
could
want
to
do
it
ate
sink
in
the
like.
A
discussed
pose.
G
That's
fine
but
like
there
hasn't
been
much
progress
made
and
I,
don't
like
there's
not
a
lot
of
visibility
into
how
the
teams
are
operating
beyond
that
hack
and
veto
like
team
structures.
Things
like
that
is
that
being
documented
somewhere,
like
is
there
still
confusion
around
yeah
I?
Think
that
I'd
like
to
have
the.
A
G
G
A
A
G
Long
as
for
misses
being
a
every
team,
quanta
and
holes
of
amazing
team
structure
and
things
like
that,
we're
still
unclear
if
people
happy
with
how
it's
going
inside
dr.
burns
allocation,
I,
know
Anna,
it's
and
the
testers
are
still
finalizing
things,
but
other
than
that.
It's
been
a
little
bit
of
yeah
radio
silence.
So
if
things
are
in
motion
and
that's
fine
I'll
just
wait
so.
A
I
know
things
are
in
motion
with
cor,
but
it's
not
been
so
much
about
okay,
ours,
at
least
in
the
most
recent
cold,
but
that
seems
like
something
that
could
be
worked
on
this
week.
If
that's
something
you
want
to
do,
I
don't
know
about
other
teams
like
it's
a
chat
team
like
what
has
to
been
some
conversation
in
that
team,
whether.
D
J
Everyone's
still
kind
of
like
forming
and
like
what
is
this
team
on
a
stage
but
I
think,
as
we
have
a
little
bit
more
clarity,
we
can
start
to
look
at
okay
else
once
we
understand
what
each
team
is
doing
and
how
it
leads
to
the
other
teams,
okay,
but
we
should
have
a
much
better
idea
at
the
end
of
this
week,
possibly
into
next
as
well.
Okay,.
E
J
C
G
C
Be
sending
out
an
email
to
the
individual
people
in
the
team
just
asking
just
for
those
sort
of
feedback
on
like
what
they
want
status
to
be
what
they
think
is
like
in
other
competitors
that
we
want
to
like
features
that
we'd
like
to
see
in
and
that
sort
of
thing
are.
You
know
for
the
whole
application,
but
particularly
focus
on
the
segments
of
the
product
they're
working
on
I'll
do
that
later
today,
that'll
be
an
individual
responsibility
for
everybody,
so
everyone
who
AM
L
on
that
for
each
respective
team.
C
It
would
be
great
if
you
could
individually
send
those
in
once
we
get
that
I
all
go
review
it
and
then
try
and
it'd
be
nice.
If
we
can
then
go
from
that
to
identifying
which
of
those
features
are
like
low-hanging,
fruit
and
high
impact,
the
users
will
find
out
what
is
their
sort
of
difficulty,
manpower
or
ish.
C
A
N
C
In
lit
p2p
sounds
like
it
makes
a
lot
of
sense
when
we're
talking
about
replacing
dev
p2p
entirely
with
that,
but
that's
always
seemed
like
a
go.
Ethereum
thing,
I
guess
I'm
a
little
confused.
How
that
relates
to
status,
go
and
if
we're
running
like
if
we're
running,
def,
p2p
and
lip-lip
p2p
on
a
mobile
device
for
for
status
I-I-I,
it's
a
I'm
a
little
unclear
on
what
what
the
what
the
goal
is
there
well.
N
The
goal
is
to
connect
devices
to
devices
and
devices
to
each
idiom
network,
because
this
is
how
we
are
currently
running
our
applications
application
anyway.
So
if,
if
you
are
using
the
mobile,
you
are
still
connected
to
those
full
lots
and
we
can
connect
our
whispered
devices
to
a
service.
Those
devices
and
the
main
benefit
of
using
the
p2p
there
is
to
use
that
is
to
use
tools
that
they
already
have
to
bypass
Nanette,
for
example,
yeah.
N
N
M
We're
implementing
a
JavaScript
in
JavaScript,
whispered,
I
mean
mean
Barry
and
and
Richard,
and
we
also
want
to
integrate
to
live
PGP
and
in
the
context
of
the
JavaScript
project,
and
the
hope
is
that
this
way
that's
exaggerated
when
work,
they
could
connect
that
to
the
PDP
through
the
browser
and
but
the
problem
that
there
are
essentially
different
networks.
But
if
the
state
is
nodes,
support
both
and
we're
gonna
have
a
bridge
between
them.
So
this
way
people
potentially
could
use
the
apps
that
connect
directly
to
you.
So.
C
M
C
I
guess
my
concern
with
that
is
I
mean
that
sounds
great
and
I
understand
the
the
goal
there
like.
Ideally,
you
wouldn't
jump
between
networks,
so
the
Serenity
will
be
moving
toward
slippy
turpino,
like
ethereum,
will
be
moving
towards
lifted
p2p,
so
we
can
consider
dev
peter
peter,
be
deprecated
in
this
sort
of
medium
to
long
term.
C
I
guess
the
major
concern
that
I
have
with
with
being
admitted
on
the
mobile
application
would
be
the
resource
consumption.
You
know
we're
trying
to
put
so
much
efforts
into
actually
trying
to
make
this
usable
on
a
mobile
application,
but
maintaining
a
CAD
million
Network
is
pretty
difficult,
I,
don't
know
if
we
are
intending
to
do
that
or
if
we
want
our
static
peers.
If
it's
any
difference,
it'd
be
great
to
get
some
benchmarks
on
what
that
looks
like
if
you
want
to
roll
that
into
release
said.
N
Ok,
yes,
ok,
but
just
brief
comment
like
we
used
discovery
and
maintain
the
media
networks
is
only
about
discover
here
and
to
use
discovery
for
limit
amount
of
time
until
we
find
required
number
of
peers
and
after
that
we
shut
down
it.
So
there
is
no
additional
overhead
of
maintaining
another
kabylia
network,
maintaining
academic
network
all
the
time
so
to
be.
In
this
case,
it's
only
about
transfer
between
peers,
ok,.