►
From YouTube: Core dev meeting #47
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
Started
going
cool,
okay,
so
death
think
I
have
good
news.
The
I
am
done
with
the
development
of
the
recurrent
transfer
updates.
A
I
I
have
stumbled
upon
some
issues
on
the
CLA
wallets
Parts,
because,
basically
now
we
add
like
a
unique
ID
at
the
end,
so
that
you
can
specify
the
pair
the
issue
with
that
is
that
we
there
is
a
way
to
set
up
a
parameter
as
optional
and
somehow
I
cannot
get
it
to
work
the
CR.
Well,
it's
always
asks
for
it.
A
She
will
be
helping
me
doing
do
stuff
with
the
hive
mind
mostly
we'll
see
where
she
lands
in
the
end,
but
yeah
I
expect
the
next
month
to
be
a
bunch
of
unboarding
tasks
here
and
there
and
yeah
I.
Think
that's
it
for
me.
B
Okay,
great
well,
nice
talk.
Your
new
developers
joined
us
welcome
as
far
as
we've
been
working
on
a
bunch
of
different
things,
I
guess,
as
always,
but
on
the
Block
trade
side,
we've
been
working
a
lot
on
the
Dockers
for
both
Hive
and
half.
We
had
a
couple
problems
reported
with
related
to
half
where,
when
somebody
would
like
shut
down
their
half
node
or
if
it
shut
down
its
own
for
whatever
reason-
and
they
restarted
it
up,
they
had
a
problem.
B
Restarting
it
and
we
found
there
were
two
problems
that
were
tending
to
well.
One
was
the
most.
The
one
was
the
more
serious
problem
which
was
a
permissions
problem
or
sometimes
if
the
external
system
wasn't
set
up
the
docker,
the
internal
Docker,
you
know
processes
weren't
able
to
properly
access
the
external
files.
So
those
were
just
really
permission
problems
in
the
local
in
somebody's
local
system,
but
it
was
hard
to
troubleshoot
those
problems
because
the
errors
weren't
the
permissions
errors
weren't
showing
up
on
the
console.
So
we
really
made
two
kind
of
fixes.
B
One
fix
which
was
to
we
made
a
change
to
make
it
more
difficult
to
have
a
permission
problem
and,
second
of
all,
we
we
set
up
the
Dockers
in
like
an
attached
mode,
so
that
now
those
if
there
is
any
kind
of
error
like
that
not
necessarily
permission
problem
but
some
kind
of
area,
it'll
more
likely
show
up
on
the
console
where
someone
who's,
maintaining
that
half
server
will
recognize.
That
will
be
able
to
understand
what
the
problem
is
easier.
B
So,
along
with
that
we've
another
thing
we've
been
doing
for
for
half.
Is
we've
been
looking
into
how
to
control
SQL
resources
postgres
under
postgres?
B
This
is
going
to
be
an
important
feature
for
when
we
start
working
on
the
well
I
guess
we
could
say
we
are
start
working
on
the
smart
contract
platform
that
runs
on
top
of
half,
because
there
it's
going
to
be
critically
important,
that
smart
contracts,
don't,
for
instance,
run
too
long
or
take
up
too
many
resources.
So
we've
been
basically
working
on
this
new
new
sub
task
to
both
monitor
postgres
resources
being
used
and
also
to
terminate
queries
if
they
take
too
long.
B
So
I
think
so,
like
a
sort
of
like
a
query
watchdog
or
we've
also
possibly
referred
to
it
as
a
query.
Governor
this
this
task
of
monitoring
and
measuring
query
times
so
for
measuring
the
query
times
right
now,
we're
looking
at
using
a
PG
stats
extension
so
in
using
callbacks.
B
So
it's
probably
going
to
be
a
c
plus
extension
that
measures
when
the
query
starts
and
also
when
it
ends
and
then
can
signal
to
abort
a
query
if
it
takes
too
long
and
some
other
things
we'll
need
to
look
at
is
potentially
measuring
the
data
stored
by
stored
by
a
smart
contract.
So
again,
we
can
cost
that
out
appropriately,
so
that's
kind
of
where
we
are
on
the
smart
contract
stuff.
That's
it's
really.
B
Preliminary
phases
really
I'd
say
just
sort
of
because,
but
obviously
measuring
resources
used
by
smart
contracts
is
going
to
be
one
of
the
most
critical
things
we've
seen
in
the
past.
B
A
lot
of
smart
contract
platforms
have
had
severe
problems
in
this
area,
so
we
need
to
address
those
issues
early
as
possible,
other
things
going
on
or
we're
probably
going
to
deploy
a
new
version
of
develop
to
master,
in
other
words,
we're
going
to
do
a
new
release
soon
for
both
Hive
and
half,
and
so
these
releases
will
on
the
hive
side,
it's
really
the
major
changes.
There
are
to
help
half
itself.
So
we've
switched
to
this
new
mode.
B
We've
been
testing
all
this
stuff.
These
changes,
especially
the
binary
storage
changes
both
in
halfa
and
hive,
mind
and
also
testing
the
block
API
that
runs
on
top
of
this,
and
so
far
the
performance
has
been
good.
The
only
thing
we
still
yet
to
do
is
we
need
to
test
it
on
production
and
I.
B
There
have
been
a
number
of
improvements
to
half
along
the
way,
since
not
only
the
binary
storage
but
some
other
things
that
came
up
when
we
were
working
on
the
Block
Explorer
so,
for
instance,
we're
now
storing
some
of
the
the
dynamic
properties
data
is
being
stored
into
the
half
database
and
is
available
so
on
a
block
by
block
basis,
and
that
came
up
just
recently.
I
think
the
Marky
marker
somebody
asked
me
for
some
data
like
that,
so
it'll,
be
there
soon
and,
like
I,
said
it's
there
now
and
develop.
B
If
anybody
wants
to
work
in
a
development
Branch
before
it's
released
and
honestly
I'd
say:
that's
fine.
If
somebody's
experiencing
half
apps
I
think
they're,
it's
perfectly
fine
to
experiment
with
the
development
Branch
we
development
Branch.
If
it's
passing
tests
should
you
know
we
only
merge
stuff
and
develop
after
it's
passing
tests,
so
we
believe
it's
it's
good
to
use
for
at
least
development
of
software,
if
not
necessarily
production.
B
B
Openhive
chat
is
a
instant
messaging
Communications
protocol.
We
use
in
Hive,
so
we're
looking
at
integrating
that
capability
in
directly
into
condenser,
so
that'd
mean
you'd,
be
able
to
do
instant
messages
inside
the
condenser
AKA,
for
instance
hive.blog.
B
So
that's
been
going
on
for
a
little
bit
and
I
think
we're
close
to
release
on
that.
We've
done
some
preliminary
tests.
We've
got
it
working
inside
condenser,
but
what
we
need
and
it
works
on
its
own
without
condenser,
but
we
need
to
have
it
work
together
in
both
modes,
and
so
we
ran
into
some
problem
having
it
operate
in
both
modes
simultaneously.
So
that's
one
last
thing:
we
have
to
do
with
open
chat
but
I
suspect
the
next
week
or
two
that'll
be
finished.
B
Other
things
we're
working
on
is
how,
as
Martin
just
recently
noted
in
matter
most,
he
saw
a
new
repo
appear
called
sea
live
and
that's
where
we're
working
on
a
we've
been
working
for
about
a
month
and
a
half
now
experimenting
with
a
new
command
line
wallet
for
Hive
and
this
one's
it's
a
python
based
wallet.
So
this
is
basically
an
alternative
to
the
existing
CLI
wallet
that
we
have.
B
So
we've
been
designing
from
scratch
to
be
more
friendly,
I
guess
to
users
than
the
existing
CLI
wallet.
So
it
it
has
like
what
you
might
call
a
terminal
based
UI.
So
it's
not
a
graphics
based
UI,
it's
not
a
GUI,
but
it
has
if
anybody's
used
something
like
midnight
Commander
or
any
of
these
other
applications
where
you
have
a
full
screen,
but
it's
text
based
full
screen.
B
That's
the
way
this
thing
works,
so
you
can
obviously
it's
a
little
more
friendly
to
use
than
a
straight
command
line
where
it's
just
command
response,
command
response,
counter
format,
and
so
we
we
developed
this
new
application
in
part
just
for
our
own
needs
and
but
and
also
just
to
meet
needs
that
we
thought
customers
might
have
that
they
might
users
might
have,
but
they
just
found
the
existing
wallet
doesn't
work
very
well
for
so
some
of
the
things
we
can
do
is
better
transaction
analysis
being
able
to
save
transactions
to
files
and
also
targeting
offline
usage
for
like
cold
wallets,
and
things
like
that.
B
So
that's
I
guess
in
the
last
thing
I
mentioned
in
passing
is
the
block
Explorer.
We
lost
one
of
the
programmers
working
on
the
black
explorer,
so
we've
assigned
a
new
guide
to
that,
but
we
still
have
the
GUI
guys
working
on
it,
and
so
that's
we're
picking
that
up
in
the
coming
week
and
so
we'll
see
how
that
goes,
but
I
think
it's
going
decently.
Well,
that's
pretty
much
where
we're
at
now,
so
I
guess
I'll
pass
it
on
to
whoever's
here.
C
Awesome
this
month,
the
last
month,
I've
been
focusing
on
gns
and
I've,
released
a
bunch
of
notification
types.
Some
support
for
notifications.
C
That's
mentions
when
someone
mentions
you
in
a
poster
comments:
Auto
rewards,
curation,
rewards
comment,
benefactor,
rewards
field,
convert
requests,
among
others,
also
done
a
bunch
of
refactoring,
because
some
of
the
modules
had
too
much
code
in
them.
So
now
I
split
the
code
up.
So
every
type
of
notification
is
basically
it's
on
SQL
file
and
now
I'm,
currently
working
on
supporting
user
preferences,
which
will
allow
people
to
customize
the
kinds
of
notifications
they
receive.
C
B
These
these
customizations
would
this
be
done
through
custom,
Json
transactions.
C
C
Retroactive,
though
so,
when
you
change
it
going
forward,
that's
when
it
will.
C
But
yeah
that's
been
my
focus
this
month.
On
the
you
know,
experimentation
I've
been
doing
with
tokens,
I
have
a
white
paper,
but
it's
not
ready
yet
I'll,
probably
maybe
share
it
in
the
next
week
or
so,
and
I've
started
writing
some
code.
For
that.
C
C
B
That's
yeah,
that's
exactly
what
so,
basically
we're
storing
a
bunch
of
the
the
dynamic
Global
properties
from
the
dgpo
object.
Now
on
a
block
by
block
basis
into
a
table
into
a
table
and
in
half
the
driving
force
was
this
was
the
ability
to
translate
Hive
to
vests?
Do
hive
device
calculations
but
yeah?
That
was.
B
Okay,
I
guess
pass
on
to
someone
else
now.
A
Okay,
maybe
not
well.
F
Other
front-end
be
able
to
use
it.
B
Oh,
the
the
integration
is,
it's
certainly
open
to
anyone
to
use
open
chat,
I
mean
rocket
chat
is,
is
an
open
source
software
right
and
all
this
stuff.
All
the
integration
we're
doing
to
condenser
will
also
be
open
source,
so
in
theory
Sure
it
I
absolutely
feel
it
should
be
able
you
know
other.
We
won't
in
fact,
other
other
front
ends
to
use
it
high
front
ends,
use
it
for
sure
as
to
the
actual
details
of
what
they
need
to
do
to
integrate
it.
B
I
don't
know
that
I
couldn't
really
answer
that
right
off
hand,
because
I
haven't
looked
at
the
Integrations
being
done
being
done
for
condenser
I,
don't
know
if
even
Gandalf
would
have
an
island
on
that,
but
he
might
get
a
comment.
D
This
Authentication,
so
in
a
way
that
we
can
have
a
high
identity
on
the
chart
So.
Eventually,
we
would
like
to
use
some
fancy
features
like
end-to-end
encryption,
using,
hopefully
keys
that
you
have
from
hive.
F
F
Cool
but
it
never
been
used
a
lot
yeah.
So
to
be
honest,.
E
If
you're
talking
about
the
chat,
then
that's
that's
something
that
is
actually
not
really
they're
going
to
move
forward
with
the
open,
Peak
team
or
the
Peak
open
team
is
working
on
another
messaging
client
as
well
that
they're
going
to
be
hoping
a
lot
of
people
will
be
using
so
I
know,
there's
going
to
be
an
iteration
on
that,
but
yeah
the
one
that
you're
thinking
of
is
B
chat.
A
Yeah,
pretty
much
so
yeah
I
have
questions
regarding
the
hive
release
that
you're
talking
about
what,
because
I
get
the
changes
for
half
in
the
the
Json
binary
format.
But
what
changes
in
Hive
are
needed
for
that.
B
So
the
way
the
data
is
injected,
basically
right,
so
the
data
is
being
written
written
to
the
TAF
database
in
a
you
know,
a
slightly
different
way,
that's
being
written
this
binary,
instead
of
as
Json
data
when
it's
being
inserted.
You
know
the
insert
this
insert
queries
were
changed.
A
B
A
Good,
okay,
I,
don't
really
have
anything
more
I'll,
probably
send
a
bunch
of
messages
for
Gandalf,
slash
Baltic
when
he's
on
his
back
regarding
the
the
optional
format
in
the
CLI
wallet.
But
apart
from
that,
everything
is
pretty
clear
on
my
end,
I
don't
know
if
you
guys
have
anything
more.
A
A
A
Basically,
you
just
have
one
extra
parameter
when
you
create
a
recurrent
transfers,
which
is
like
an
ID,
and
if
you
specify
an
ID
that
already
exists,
it's
going
to
update
the
existing
American
transfer
and
if
you
don't
specify
one
it's
just
going
to
default
to
cad0,
and
when
you
use
the
API
to
fetch
the
recurrent
transfers.
It's
going
to
basically
show
you
all
of
them
with
the
extra
ID
so
that
you
can
know
which
is
which.
A
Sale
orders
on
the
under
attacks.
Okay,
you
know
you
create
an
ID,
so
IDs
don't
need
to
follow
each
other.
They
just
need
to
be
unique
and
it's
a.
A
Uit
is
not
okay,
because
it's
it
has
letters
in
it,
but
yeah
it's
it's
an
inch
32,
so
you
like
literally,
you
could
do
random
and
the
likelihood
of
it
being
the
same
is
very
low,
but
I
would
not
recommend
that
yeah
speaking
of
which,
last
time
we
thought
about.
We
talked
about
creating
a
unified
APR
for
transfers
and
recurrent
transfers
and
like
has
there
been
any
progress
on
that
front.
A
B
Not
that
I'm,
aware
of
I
think
it's
still
the
plan,
but
I
haven't
heard
anything
about
it.
Lately,
I
don't
know
Gandalf.
Did
you
hear
anything
about
it.
A
F
B
Okay,
yes,
that's
it
then
well
I,
guess
you've
emerged
and
developed.
So
you
saw
that
there
were
new
tests
associated
with
the
recurrent.
B
Yeah,
of
course,
okay
sounds
good,
I
guess
no
one
else
has
got
anything.
Then
this
short
call
today.
Yeah
awesome
thanks.