►
From YouTube: Jitsi Live Stream
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
And
we
are
life,
it's
been
already
six
minutes
so
I
suppose
we
can
just
start
as
usual.
We're
gonna
go
through
our
team
one
by
one
and
share
some
progress
and
there
would
you
open
my
bag
and
start
because
I
only
have
a
few
words
to
say:
oops.
A
There
goes
the
sit
room
so
in
the
last
couple
of
weeks,
I've
been
working
on
some
changes
to
ice,
which
will
which
are
supposed
to
make
it
possible
to
switch
between
different
interfaces.
A
Without
any
signaling
just
on
the
ice
level,
so
this
is
mostly
useful
for
mobile,
where,
if
you
are
on
a
Wi-Fi
and
the
Wi-Fi
goes
down,
you
want
to
seamlessly
fall
back
to
it,
you're
pretty
interface.
So
this
part
of
the
work
is
now
complete.
The
only
thing
that
was
required
was
some
minor
changes
in
ice
cool
j
on
the
server
side.
A
So
I
think
I
need
to
update
the
bomb
file
in
the
bridge
actually,
but
the
next
release
should
be
should
contain
this
and
then
for
you
to
meet
on
iOS.
It
just
works
out
of
the
box
and
I
don't
think
we've
tested
android
specifically,
but
I
have
no
reason
to
think
that
it
wouldn't
work
here
and
now
that
I'm
saying
that
I'll
make
a
note
to
be
tested
on
laundry.
So
that's
from
me
Brian
do
you
wanna
go
next.
B
So
we
were
in
the
kind
of
performance
investigation
space
we
had
started
with
his
idea,
this
optimized
chrome
that
could
use
a
pre
encoded
file.
So
that
way
we
could
get
more
chrome
instances
easily
running
on
a
on
a
server.
So
we
could
do
some
load
tests,
but
Aaron
and
George
had
found
that
they
were
bumping
into
performance
issues
even
at
some
smaller
conference
sizes
or
smaller
bridge
load
sizes.
B
C
Cool
I'm
happy
to
connect
I've
been
working
mostly
on
drew
brees
stabilization
tasks
to
make
the
recorder
work
slightly
better
and
scale
up
and
down
a
little
bit
better.
That's
made
the
progress.
I've
also
done
some
work
on
gateway
that
I
talked
about
which
allows
us
to
connect
in
a
life
size
type
hardware,
room
I
tried
to
join
today,
but,
of
course,
we're
going
to
competitive
little
cookie.
So
I'll
probably
continue
on
that
I'm,
not
probably
the
biggest
projects,
I'm
going
to
make
it's
really
better
before
this
call,
otherwise
I'll
be
helping
with
you.
C
Some
of
the
word
brian
is
talking
about
it
from
the
bottom
eight
in
the
bar,
but
it
has
been
quite
distracted
but
I'll
pass
it
on
to
george
me.
Yes,.
D
D
That
that
that
consumes
more
bonded,
so
we
are
so.
The
receiver
is
able
to
ramp
up
its
remedies,
and
so
we
can
scale
up
there.
We
can
scale
up
with
the
bridge
is
sending
to
your
specific
receiver
after
congestion
has
occurred
and
the
bridge
has
scaled
down.
So
so
this
work
is
now
done,
along
with
the
bit
rate
betrayed
allocation
algorithm
that
we
have
and,
along
with
the
GCC
implementation
that
we
have
the
bridge.
This
should
give
us
a
first
version
of
adaptivity,
and
next
thing
would
be
to
improve
or
valid
or
well
mostly
improve.
D
Our
disagreement
makes
make
sure
that
it's
working
reliably
in
different
scenarios,
with
with
different
channel
characteristics
like
last
May
and
and
things
like
that,
but
we
haven't
done
any
extensive
times,
testing
on
that
arm
and
another
neck.
Another
task
that
that's
going
to
happen
is
adding
temporal
scale
ability
to
be
p8
in
the
short
future.
Hopefully
so,
yeah,
that's
cell
phone.
E
On
the
mobile
front,
the
the
efforts
in
the
last
two
weeks,
and
probably
before
that
were
fixing
bugs
crashes
so
that
we
can
get
into
a
state
of
the
mobile
app
in
which
we
feel
comfortable
announcing
on
the
mailing
list
in
Twitter
Facebook,
whatever
channels
and
new
has
so
how
he
mostly
working
like
reducing
the
cpu
load
for
the
app,
for
example,
when
it
goes
to
the
background,
because,
like
the
the
user
lock,
the
screen
order
hit
the
home
button
so
that
we
can
mute
the
local
participant
and
drain
less
battery.
E
We
are
also
fixing
a
lot
of
bugs.
I
hounded
like
there
was
a
problem
that
yup
couldn't
stay
in
the
background
live
in
and
react
native
would
not
allowed
the
app
to
use
HTTP
requests.
So
we
had
to
work
around
this.
E
We
currently
have
one
problem
which
supposedly
stops
us
from
announcing,
and
we
need
to
investigate
is
that
do
with
the
latest
tab,
which
is
earning
the
late
legitimate
and
the
latest
video
bridge.
We
see
far
too
many
black
videos,
even
though,
where
participants
see
these
participants
like
the
mobile
display
since
black
videos.
So
we
need
to
look
into
it
before
we
announce
and
that's
pretty
much
the
work.
It's
been
just
small
fixes
here
and
empty
bugs
and
crashes.
E
We've
got
a
lot
planned
for
the
next
two
weeks,
though,
so
we
can
get
up
to
speed
with
weber.
She
seemed
react
native.
The
latest
releases
because
web
RTC
were
running
with
54
I
think
but
common
web
RTC
respectfully
is
at
56
at
the
moment
is
a
stable.
E
We
need
to
go
and
finish
this
work
that
we've
been
doing
for
the
past
two
weeks
to
mute.
The
remote
participants
as
well.
Not
only
the
local
participant
wind-up
goes
into
the
background.
This
is
something
that
so
has
been
taken.
Care
of
and
yeah
he's
got
a
peers
just
need
to
merge
this
thinking,
I
test
it
make
sure
that
it
works
otherwise,
at
least
the
next
week,
we'll
be
continuing
to
fix
bugs
and
crashes
that
we've
deemed
to
be
less
important.
E
Not
important
enough
to
block
an
announcement,
but
still
some
things
that
we
need
to
address
I
crashes
without
resolutions
or
just
some
errors
that
are
happening.
The
user
doesn't
really
see
them,
but
they're
still
happening.
So
we
need
to
take
care
of
those
underneath.
That's
pretty
much
our
thing,
I'm
also
working
on
piers
I've
got
a
PR
from
about
and
supported
browser.
E
So
the
current,
stable
version
of
meat
has
some
problems
with
and
supported,
browsers
and
we've
got
new
designs
for
those
pages
that
we
implement
and
I
just
need
to
merge
the
PR,
a
lot
of
changes
there
and
we're
doing
taking
into
account
the
new
designs.
The
mobile
web,
fixes
on
Internet
explorer
on
safari
whatever,
and
the
other
thing
that
I've
got
from
the
me
on
is
work
on
internationalization
in
the
reactor
been
respectively.
The
reactant
eight
optimal
power
because
currently
were
missing
internationalisation
ever
these
translations
and
the
man
has
appeared
at
the
meridian.
E
F
So
I
was
working
mostly
on
fixing
remaining
issues
with
the
peer-to-peer
stuff
and
then
I
created
PR,
and
it
goes
through
some
initial
review
and
then
we
decided
to
split
it
into
a
few
smaller
ones.
So
I'm
going
to
work
on
it
this
week
and
hopefully
before
the
next
call,
it
should
be
matched
so
yeah.
That's
it
from
me.
G
A
Okay,
I
just
have
one
more
thing
to
add,
because
I
forget
about
it:
we're
starting
to
implement
the
sender
side,
bandwidth
estimation
and
the
first
step
for
that
will
be
adding
the
transport
ccr2
be
header.
Extension
on
them
are
TCP
packet
for
acknowledging
and
then
sometime
in
the
future
will
be
doing
the
whole
Thunder
estimation.
A
So
I
think
that's
all
from
our
team
and
we
can
move
on
to
the
open
mic.
Do
you
think
we
only
have
garlic
and
a
fellow
youth
stir
this
time?
Do
you
guys
have
any
questions?
I.
E
H
The
one
thing
on
the
duty
reactor
thing
from
friday
and
I
have
some
issues
so
in
the
compilation
like
an
on
my
android,
I'm
getting
some
errors
and
I
mean
I'm
mailing
us
all
he's
trying
to
help
me,
but
today,
I
decided,
like
I.
Don't
need
you
tree
app
for
now,
just
to
make
a
call,
strats
initially
compatible
to
react
to
you
and
that,
once
you
punch
off
with
all
the
all
the
issues
with
window
and
wave
Krypton,
everything
I
can
then
poured
it
a
definitive.
Yes,.
E
So
that
was
my
concern
as
well,
because
they're
here
an
hour,
GT
meet
mobile
web,
does
a
lot
of
polyfills
that
we
need
of
lib
GT
meat
and
some
of
those
poly
fuse
happen
to
Akio
coast
at
the
constants
library
as
well.
But
if
you
wanna
do
a
react.
Native
port
on
your
side
probably
make
sense
to
j,
make
sure
the
d'harans
in
react
native
in
general,
not
yet
Joaquin
era,
where
we
do
have
a
lot
of
policies
and
you
probably
not
gonna
find
them
in
other
react
native.