►
From YouTube: Hive Core Dev Meeting #37
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
A
Cool
okay,
so
devsync
last
time
we
met
was
a
month
ago.
What
did
they
do?
I
worked
on
the
txt
documentation
script
for
hivemind,
so
if
we
can
better
track
the
marks,
I
spent
a
bunch
of
time
working
on
unit
testing
high
mind
for
my
issues.
I
mean
my
features.
I
have
a
bunch
of
weird
things
where,
when
I
dump
the
thing
locally
up
to
five
million
blocks,
I
don't
get
exactly
the
same
result
as
the
the
test
results,
which
is
strange.
So
I'm
working
also
on
that.
A
I
did
a
few
back
and
forth
with
one
of
your
dev,
then
about
authentication
testing.
He
found
a
bug
with
lists.
Also
delegations
which
I
fixed
I
put
pull
up
pulled
up.
Is
this.
A
Yes,
I
think
his
icon
is
green,
yeah
yeah,
so
I
don't
know
so
yeah
that
has
been
fixed.
I
pulled
up
a
merge
request
yesterday
and
finally,
in
other
news,
I
am
starting
to
look
for
a
a
second
dev
to
help
me
out,
so
we
can
build
out
cool.
Yes,
I'm
following
your
track
soon
will
be
like
10
000
in
there,
hopefully
yeah,
no
but
yeah.
I
figured
that
to
give
more
insight,
basically
how
mine
can
be
done.
A
Most
of
the
testing
can
be
done
by
someone
else,
and
so
that
way
I
can
focus
on
building
features
and
offload
some
of
the
work
and
also
work
on
hive
d.
More
any
I
mean
it's,
it's
gonna,
be
I'm
gonna
have
a
trial
period
by
the
end
of
the
month,
so
I'll
try
that
out.
So
I
have
a
bunch
of
calls
about
that
and
I
and
I
wrote
something
so.
A
Exactly
sql,
but
the
idea
is
mostly
to
write.
I
mean
at
first
it's
going
to
be
rating
a
lot
of
testing
because
that's
the
best
way
to
involve
on
the
project
and
this
stuff
yeah.
A
Yeah
exactly
and
so
that
and
maybe
we'll
find
some
stuff.
So
that's
gonna,
be
you
work,
so
I'm
basically
writing
a
mentioned
documentation
like
dev
documentation
on
how
to
get
started.
Maybe
I'll
pull
up
a
contributing.md
thing
in
the
repo
and
I'm
writing
like
some
tasks
here
and
there
I
don't
know
if
I'm
gonna
use
gitlab
to
centralize
development
or
not
or
if
I'm
gonna
use
like
a
private
jarrah
yeah.
It's
work
in
progress,
but.
B
Okay
sounds
good,
I
guess
we've
got
a
lot
of
stuff
going,
but
just
cover
some
of
the
main
things
of
interest
of
late,
we're
still
working
on
some
of
the
dockerized
stockerizing,
the
half
applications.
B
So
that's
going
pretty
well,
we
started
testing
some
of
those
one
of
the
more
interesting
things
we
did
recently
is
we've
added
compression
to
the
block
log.
So
I
got
to
get
the
final
numbers
right
now,
but
I
think
we're
about
compressing
the
block
logged
about
50
of
its.
You
know,
I
guess
uncompressed
size
right
now
and
another
thing
we're
looking
at
is
possibly
keeping
the
compressed
across
the
peer-to-peer
network
too,
which
would
effectively
cut
the
peer-to-peer
network
traffic
in
half.
B
I
mean
that's,
not
really
a
pressing
problem
at
the
moment,
but
still
why
not?
If
we
can
and
it'll
also
save
compression
time
that
way,
basically,
the
node,
that's
the
node
that
initially
produces
a
block
will
compress
it
and
then
it
can
just
be
exchanged
compressed
among
all
the
nodes,
so
they
don't
have
to
re-compress
it
themselves
that
way.
B
So
there
is,
I
guess,
some
also
some
performance
improvements
on
each
at
each
node.
From
that
what
else
I
guess,
we've
been
working
on
the
peer-to-peer
network
in
terms
of
efficiency,
both
latency
and
in
terms
of
bandwidth,
and
so
at
this
point
we've.
B
I
I
mentioned
this
in
my
post
yesterday,
but
we've
we've
actually
got
we're
running
on
the
mirror
net,
so
we
basically
got
the
same
amount
of
traffic
that's
running
on
the
main
net,
except
that
our
block
time
offsets
are
now
all
showing
as
negative
across
the
board
from
all
the
witnesses,
so
that
that'll
give
you
an
idea
of
how
much
we've
reduced
the
latency
and,
just
in
case
anybody's
wondering
how
you
get
a
negative
block.
B
B
B
I
guess
oh
here's
another
one.
That's
kind
of
important
we've
managed
to
get
the
half
base
the
half
base,
the
new
half
based
hive,
mind
passing
continuous
integration
tests
now,
which
is
great
news.
I
think
so
next
we're
going
to
be
qualifying
and
benchmarking
with
a
full
replay
to
the
head
block
and
if
that
passes
then
we'll,
I
suppose,
shortly
thereafter,
we'll
merge
that
into
develop.
A
Cool
regarding
the
compression.
It's
only
for
downloading
right
because
not
live.
B
No,
that's
what
I'm
saying
right
now
so
right
now,
right
now
the
compression
is
done
when,
when
you
receive
a
block,
you
can
press
it
and
it
gets
stored
to
the
block
log.
So
the
compression
right
now
just
happens
at
during
the
storage
of
the
block
hog,
but
the
next
step
that
shouldn't
be
difficult
to
do
is
to
actually
exchange
them
in
compressed
form.
So
say
your
node,
your
block
producer
and
you
produce
a
block.
You
can
compress
it
and
then
transmit
that
compressed
block
across
the
network.
B
So
a
lot
of
the
work
of
the
past
couple
days,
we
actually
got
compression
going
a
little
while
back
a
lot
of
the
work
lately
has
been
optimizing,
the
compression
more
so
we've
been
generating
dictionaries
to
allow
for
better
compression
blocks,
and
essentially,
we
created
right
now
for
optimal
compression,
we're
regenerating
a
new
dictionary
every
million
blocks.
B
So
we
basically
so
you
know
now
we're
talking
about
60
some
million.
I
mean
60
dictionaries,
one
for
each
million
blocks
and
it
swaps
between
the
dictionaries
for
maximum
compression
in
each
of
those
block
intervals.
A
Okay,
I
see
neat
okay,
okay,
I
guess,
in
terms
of
I.
A
Topics
today,
yeah,
I
was
reviewing
the
I
cleared
up
a
bunch
of
the
the
old
issues
in
the
old
issues
that
I
created
in
the
youtube
channel.
I
see
girls
that
you
wrote
a
lot
of
cla
wireless
issues.
Are
you
gonna
tackle
them,
or
is
it
just
for
tracking
purpose
and
they're
open
for
everyone.
B
Maybe
he
didn't
he's,
I'm
just
saying
he's
probably
gonna
reply
in
the
chat
channel
because
he's
he
can't
dial
in
so
he
he
doesn't
have
voice.
A
That's
right:
okay,
replay
yeah.
So
during
testing
we
noticed
that
there
was
a
bunch
of
satellite
work
in
progress,
but
could
seem
working
on
that
tv.
Okay,
cool
sounds
good,
so
yeah
as
he
was
tested
as
a
you're,
there
was
testing
the
rca
api.
I
noticed
that
it
was
kind
of
confusing
because
we
use
a
lot
of.
Basically
everything
is
account
id
based,
but
research
using
account
name
base,
and
so
that's
basically
the
created
confusion,
even
in
any
other
because
he
was
like.
A
Oh
I
I
query
like
from
alice
to
alice
and
I
get
it.
But
then
I
query
from
alice
to
aa
and
I
don't
see
it
but
because
aa
has
an
id
superior
to
alice,
then
he
doesn't
see
things
because
internally
we
convert
everything
to
numbers,
I'm
thinking
of
making
some
improvement
to
that
api
because
I'm
pretty
sure
a
lot
of
devs
will
get
confused.
But
is
it
too
late
to
change
in
regards
to
the
test
nets
or
should
that
go
into
like
a
later
thing?.
A
Yeah,
I
kind
of
like
to
either
query
using
ids
instead
of
names
or
query
or
like
return,
the
ids
when
we
do
the
query
so
that
way,
it's
more.
A
B
I
see-
I
don't
know,
I
guess
maybe
maybe
throw
out
or
just
let's
create
an
issue
on
it
and
we'll
we'll
bring
it
up
for
discussion
with
everybody
else
involved,
because
obviously
whoever's
testing
it
and
all
that
should
be
involved
too.
A
Yeah
sounds
good,
yeah
and
then
basically
have
and
yeah.
Also,
I
guess
I
will
have
to
include
the
my
effects
into
the
testness,
speaking
of
which
so
what's
the
status
on
the
testnet,
we
have
the
mirror
net.
That's
running.
How
long
are
we
planning
on
having
that
running.
B
I
think
we're
right
up
until
the
hard
fork
will
we'll
pass.
I
mean
we'll
be
we'll
we'll
start,
we'll,
probably
start
a
new
one.
I
don't
imagine
before
too
long,
we'll
we'll
we'll
create
a
new
version
of
the
internet.
A
Okay,
well,
okay,
then,
by
next
week,
if
we,
you
guys,
are
having
a
version,
I
can
probably
fit
in
some
of
the
the
fix
for
the
bug
for
next
week.
Let
me
post
this
up
and
pelican
tests.
I'm
probably
gonna
try
to
run
a.
I
mean
I
will
run
a
node
and
it's
on
the
next
version.
A
Yeah,
that's
pretty
much
all
about
it.
Stu!
Nothing
came
up
interesting
so
far,
right
apart
from
the
june
at
least
rc.
A
B
A
A
A
Okay,
cool
and
yeah-
that's
about
all
I
had
for
today.
Do
you
guys
have
anything
else.
A
Yeah
anyways
so
yeah,
that's,
let's
give
it
awesome
ring.
So
that's
been
done.
Yeah,
okay,
I'll,
probably
walk
in
some
more
and
have
mine.
After
after
the
vlog
I
mean
on
the
unhide
itself.
A
I
need
to
think
with
the
community
on
some
of
the
features
they,
namely
the
the
burning
mechanisms
that
I
was
brought
up
a
while
ago,
we're
getting
instant
with
withdrawal,
I'm
just
afraid
of
the
effects
on
rc
and
other
other
components.
B
I
don't
know
if
him
watson
wanted
to
say
anything
about
his
work.
He's
working
on
right
now.
C
Yeah
hi
guys
I've
been
mainly
focusing
on
on
gns
lately,
I'm
not
sure.
If
you've
looked
up
some
of
the
documents
I
shared,
but
basically
the
idea
is
to
have
a
a
notification
system,
that's
tailored
towards
making
it
more.
You
know
customizable
for
the
user,
and
so
the
idea
is
to
have
notifications
and
then
users
get
to
subscribe
to
different
types
of
notifications,
and
so
the
system
is
well
open
source.
C
Yes,
but
it's
modular
in
that
you
can
have
any
new
dap
that
comes
up
on
hive
can
come
up
with
their
own
notifications
and
then
define
them
within
certain
files
in
the
system,
and
then
those
notifications
become
available
I'll
share
a
readme
file
that
I
updated
today.
C
Just
to
give
you
guys
an
idea
on
you
know
where
it's
at
right
now
as
far
as
plugin
plays
concerned,
I'm
working
on
making
it
more
customizable,
because,
right
now
the
blogs
aren't
really
fully
independent
of
each
other,
and
so
I
was
looking
at
like
port
ping's
case
use
case
right
now.
If
they
wanted
to
just
run
port
ping
apis
on
there,
we
need
to
make
it
a
bit
more
simpler
to
do
that
and
yeah
I'm
also
just
you
know
preparing
it
for
when
we
get
that
feature
to
filter.
C
B
C
Oh
awesome,
which
branch
in.
C
C
B
I
think
there
is
yeah
effectively
there's
two
kind
of
filters
you
can
use
one
you
can
filter
based
on
account
name,
which
is
that
works
just
the
exact
same
way
as
it
does
for
account
history.
So
if
it's
exact
same
syntax
in
fact,
and
then
the
the
format
for
the
regex
based
one,
which
is
basically
how
you
can
filter
for
operations,
it's
it's
it's
fairly
straightforward.
You
basically
put
additional
statements.
Well,
both
of
them.
You
put
additional
statements
into
config
that
ini
file
of
your
of
your.
C
Oh,
that's
cool
yeah.
I
think
that'll
probably
be
my
focus
going
forward
just
to
make
plug-and-play
align
with
that
customizability.
So
yeah,
that's
cool!
Oh
just
a
brief
update
on
other
stuff
that
I've
been
working
on
that's
off
chain.
C
So
I
moved
to
mauritius-
and
I
was
just
looking
around
and
trying
to
see
if
there
are
any
people
interested
in
crypto
and
blockchain
besides
and
there's
quite
a
number
of
developers
who
are
interested-
and
I
got
in
touch
with
someone
who
can
help
me
organize
regular
meetups.
C
So
I've
got
this
high
mauritius
meetup
thing
going
on,
but
I
think
we'll
have
our
first
one
in
like
two
weeks,
we've
got
a
venue,
but
I'm
still
trying
to
get
the
logistics
done,
but
I'll
update
you
guys
when,
when
you
know
we
move
forward
with
that.
And
yes,
that's.
B
B
Okay
thanks:
anyone
else
have
anything.