►
From YouTube: Open Match Community Call August 2019
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
Just
as
a
point
of
recap,
we
had
the
open-sourcing
gaming
day
last
week
at
the
open
source
summit.
North
america
had
some
really
great.
You
know
feedback
on
open
match
with
some
folks
that
were
there,
so
that
was
really
helpful
and
very
much
appreciated
anybody
else.
Wanna,
add
thoughts
or
share
anything
else,
cool
that
they've
been
working
on
I.
B
A
Yeah
I
don't
looks
like
there
wasn't
anything
added
in
the
agenda
directly
I
knew
you
guys
said:
we've
had
some
conversation
and
slack
about
like
different
proposals
and
stuff.
So
I
didn't
know
if
that
had
been
resolved,
but
I
figured.
That
was
a
good
thing
to
discuss
if
it
was
still
an
open
issue
and
then
yeah
Caleb.
If
you
want
to
give
a
general
update
sure.
B
B
So
I
guess
a
couple
of
us
went
down
for
the
open
source
and
gaming
at
the
Linux
Foundation
open
source
summit,
and
we
presented
there
at
the
open
source
and
gaming
kind
of
session.
There
are
a
lot
of
really
interesting
talks
on
the
open
batch
side.
The
talk
went
really
well
and
we
got
a
really
like
a
large
amount
of
discussion
out
of
it
actually
at
the
end,
which
made
it
extra
worth,
including
a
couple
people
who
had
a
lot
of
experience
in
the
industry.
B
There
was
some
very
fair.
It
was
definitely
worth
it
because
we
got
a
lot
of
perspective
on
kind
of
the
needs
of
folks
who
are
interested
in
like
evaluating
open
match,
which
is,
it
seems,
like
there's
a
desire
to
have
that
those
samples,
or
at
least
the
business
logic
out
of
the
box.
So
we
should
definitely
take
that
as
feedback.
B
However,
right
what
I
was
confirmed
on
was
that
there's
entire
team
staffed
just
solving
this
problem
that
open
matches
trying
to
solve,
which,
like
the
infrastructure,
the
concurrency,
the
availability
of
matchmaking,
so
the
fact
that
entire
teams
are
being
staffed
just
to
solve
this
problem.
Right
at
these
larger
organizations
is
I.
Think
a
really
good
proof
point
for,
for
where
the
project
is
and
where
it's
headed
so
generally,
like
really
really
positive,
so
I'm
Scott
you
had
lunch
or
Ypres.
Look
you
have
much
to
add
there.
That's.
B
I,
don't
know,
if
other
than
getting
into
like
the
specific
notes
that
we
took
I,
don't
know
if
there's
much
value
in
going
through
other
feedback,
we
got
out
of
that
other
than
it
seems
like
we're
on
the
right
track.
Folks
are
learning
that
they
have
to
bring
some
parts
to
the
story,
but
for
folks
that
kind
of
have
those
parts
it
seems
like
a
like
a
good
place
to
be.
A
Yeah
I
would
just
reiterate
kind
of
that
call,
for
you
know
news
cases,
and
so
just
and
that's
really
helpful
to
us
that,
even
if
you're
not
yet
using
open
match
or
just
evaluating
it,
please,
you
know
file
a
issue
or
whatever,
like.
Let
us
know
kind
of
what
your
use
case
is
because
that
was
a
huge,
huge,
helpful
feedback,
so
sweet.
C
C
The
really
big
one
is
that,
thanks
to
some
feedback,
the
GRP
see
a
lot
of
them
are
now
streaming
instead
of
just
G
RPC
calls,
because
the
G
RPC
calls
have
limits
that
we
didn't
realize
that
are
too
low
for
a
lot
of
overmatch
workload.
This
is
before
megabyte
limit
yeah
the
four
megabyte
limit.
So
if
someone
discovers.
C
So
that
I
believe
is
all
merged,
and
so
that
limit
should
be
gone,
zero
point
seven
and
then
the
other
major
thing
we're
hoping
to
landfill
is
some
other
types
of
filters.
That's
not
quite
in
yet,
but
we
have
a
plan
for
getting
it
in
kind
of
a
query
way
and
then
we
can
jiggle
it
on
the
inside
to
get
it
to
a
better
say
internally.
C
C
B
On
that
note,
there's
actually
there's
there
several
yes
kind
of
new
proposal
issues
that
have
been
cut
on
github
for
the
purposes
of
expanding
the
API
expecially
on
the
configuration
side
as
well,
some
several
enhancements
on
the
filtering
and
also
just
like
the
querying
surface
in
general.
So
those
are
definitely
open
for
discussion
and
comments.
Welcome,
yes,.
C
B
C
Believe,
there's
an
issue
on
that
Ethan
worked
on
this.
We
held
off
on
doing
it
in
point
7,
because
we
had
some
discussion
about
how
it
should
actually
work,
and
we
were
hearing
from
people
are
talking
to
that.
It
was
very
easy
for
it
to
inject
it
themselves
and
they
could
inject
a
slightly
better
value
to
it.
C
C
Currently,
it
is
just
the
commit
history,
you
don't
have
any
like
systematic
way
to
degenerate
release,
notes.
The
way
we've
been
doing
it
is
the
person
cutting
the
release
candidate.
Well,
look
at
the
pulse
between
when
point
six
was
French
and
now
and
then
we'll
figure
out
what
the
notes
are
from
there.
So
I
guess
that
would
be
end
of
next
week.
We'll
have
some
really
concrete.
No,
it's
four
point.
Seven,
once
we've
already
cut
it,
yeah.
C
Don't
think
so,
just
overall,
oh
the
other
thing
we
have
been
working
on
in
point.
Seven
is
scale
testing,
but
I
have
not
been
the
one
doing
that
so
I
don't
have
the
numbers
that
was
Arabic.
He
is
out
this
week,
so
I
think
going
into
point.
Eight
will
have
some
more
scale
numbers
to
share,
but
we
don't
have
that
today
and
then
we're
gonna
use
that
to
drive
some
discretion
about
how
indexing
works
and
how
the
internal
like
state
store,
works
and
then.
C
We're
also,
as
just
a
heads
up,
probably
going
to
discuss
removing
assignments
from
the
API
and
building
something
next
to
open
mesh
to
handles
assignments.
Just
because
we've
had
some
feedback
of
some
people
just
want
the
open
batch
of
it
that
doesn't
have
assignments.
They
just
want
to
plumb
to
get
through
and
match
them,
and
they
have
everything
else
and
then
obviously,
a
lot
of
other
people
want
some
assignments
waiting
workflow.
So
that's
vague.
C
B
We
have
we,
the
integration
on
dot.
Six
actually
went
really
well
kind
of
uncovered
a
few
issues
which
we
have
loaded
up
on
to
to
get
how
about
as
mentioning
those
a
few
seconds
and
few
minutes
ago,
namely
the
the
API
format
around
struct
and
how
indexing
works,
would
love
to
hear
what,
in
this
one,
I
kind
of
feel
like
a
lonely
voice
a
little
bit
because
I
think
maybe,
like
our
use
case,
isn't
the
same
as
like
everyone's
use
case.
B
But
besides
that,
we
were
able
to
totally
integrate
into
our
back-end
and
then
deploy
it,
and
so
we're
actually
running
some
some
tests.
I
don't
have
the
numbers
today,
because
we're
still
in
process
of
standing
up
all
our
intestine
for
structure
for
testing.
You
know
the
that
that's
side
of
the
house,
but
what
we
are
seeing
is
we're
seeing
some
really
good
flow
through
through
open
match,
and
it's
actually
pretty
pretty
quiet,
which
is
nice.
We
haven't.
B
You
know
we
have
our
own
issues
to
solve,
that
haven't
actually
allowed
us
to
to
totally
stress
up
a
match,
which
is
a
good
sign.
We
have
some
very
natural
Headroom
there,
so
really
looking
forward
to
the
stress
testing
results
on
going
in
ending
dots
have
been
going.
It's
a
date
and,
of
course,
if
there's
anything
like
the
ability
to
like
run
it
or
have
instructions
for
running
it,
we
would
definitely
be
be
down
for
for
participating
and
getting
more
samples.
There.
B
B
Suppose
at
some
point
there
should
be
like
a
demo.
That's
like
hey,
like
we.
We
deployed
this
thing
on
top
of
a
match:
I
demoed,
something
at
open
source
summit,
but
unfortunately,
since
then,
I
have
broken
it
sad
to
say:
yeah
I
would
share
my
screen
and
and
kind
of
like
look.
I
can
click
around,
buy
my
unity
game
and
behind
the
covers
it's
open
match,
but
yeah
I'll
be
sure
to
have
something
fun
like
that
for
next
time,
yeah.
B
I
guess
just
to
kind
of
be
transparent
about
what
the
know
what
the
joke
is,
but
what
was
shown
was
we
hooked
up
a
Unity
game
to
a
matchmaking
system
that
was
using
open
match
as
the
the
underlying
infrastructure
and
we
simulated
several
players
running
and
connecting
and
getting
back
I
piece
from
a
prototype
function
that
was
deployed.
This
is
through
pulling
the
assignments
API
currently
and
what
was
really
happening
by
in
the
cover
is
when
you
do.
B
So
you
know
20
to
40
millisecond
time,
zon,
the
the
delete
and
the
gets,
and
then
the
create
was
something
like
it's
like
80
millisecond
average.
So
we
have
since
tried
scaling
that
and
for
the
most
part
we
see
really
good
linear
scale
out
of
those
api's
where
things
slow
down
is
you
know,
just
the
whole
system
has
tendency
to
turn
and
so
that
synchronization
context
window
will
be
really
critical
going
forward.
B
But
that's
where
we
want
to
get
that
sampling
in
from
the
stress
tests,
because
I
think
having
that
stress
test
package
dubbed
for
for
folks
to
go
in
and
just
do
the
readme
and
run
it
and
then
put
their
own
function
in
the
middle
it'll.
Give
us
a
ton
of
more
information
about.
You
know
what
how
the
intrinsics
of
that
function
effects.
You
know
the
the
scalability
of
open
match,
so
those
folks
will
need
to
measure
that
I
think
as
they
as
we
go.
A
D
We've
done
last
month,
or
so
so
I
upgraded,
so
I
was
on
I
was
on
dot
for
I.
Think
prior
to
this
meeting
since
last,
and
we
got
up
2.6
when
I
came
out
and
that
was
pretty
seamless.
Actually,
this
is
pretty
pretty
straightforward,
which
is
great
and
then
I
actually
set
up
a
listen
server
to
run
against
our
production
matchmaking.
So
it's
essentially
pulling
down
all
of
the
real
production
data,
but
then
it
just
instead
of
sending
the
reservations
out.
It
just
drops
them
on
the
floor
and
that's
actually
I'm.
D
D
So,
unfortunately,
this
meeting
is
like
a
day
early
I
just
pulled
Master
down
from
because
the
that
streaming
change
just
went
in
Monday
or
Friday
I
can't
remember
so:
I
haven't
actually
I
wrote
the
code,
but
I
haven't
actually
deployed
all
the
new
things
and
run
it,
but
I'm,
hoping
by
the
end
of
the
week.
Actually
I'll
have
a
listen
server
against
production,
which
would
be
pretty
cool
and
then
plus
one
super
stoked
about
the
attributes.
D
Properties,
change,
I
think
that
just
that
issue
just
got
created
like
a
couple
days
ago
with
I
mean
you
guys
are
all
familiar,
so
yeah
I
think
I
think
that's
pretty
cool.
We
have
just
our
personal
use
case.
We
had
just
have
like
a
ton
of
properties
that
go
into
the
matchmaking
function,
so
it's
just
like
this
drug
norm
is
struct
of
like
30
things
that
are
just
sort
of
jammed
in
there
that
you're
constantly
like
turning
into
it.
D
D
Changes
have
been
like
yes
breaking,
but
pretty
straightforward
and
like
incrementally,
going
from
one
to
the
next,
without
having
to
change
like
the
whole
system
and
stuff
which
has
been,
which
has
been
good,
so
amazing
work
guys
has
been
awesome
just
working
with
that
stuff.
I
think
that's
about
it.
Yeah
I've
been
I've
been
super
happy
with
last
last
month.
Some
good
stuff.
B
Fantastic,
it's
awesome
feedback
on
the
doc
stuff,
yeah
I
would
love
to
hear
more
about
you
know
or
be
it
github
or
just
the
normal
channels
or
paying
but
yeah.
It
would
love
to
hear
some
more
thoughts
on
kind
of
like
the
structure
that
gets
passed
in
and
why
the
binary
use
cases
is
like
good
or
bad,
because
I
know
that's
been
a
constant
debate
on
the
the
steering
committee
sighs
just
like
what
do
people
want,
what
do
they
need
and
we're
kind
of
just
like
guessing
so
it's
cool
to
have
like
a
concrete
example.
D
I
mean,
like
I,
said:
that's
just
our
specific
use
case
like
I.
Don't
know
how
other
people
but
I
mean
I,
don't
know
if
we're
necessarily
alone
in
this,
but
like
all
players
and
servers
go
through
our
director,
which
I
think
isn't
like
wasn't
like
the
original
plan.
So
I
think
you
know
that's
kind
of
different
from
what
a
lot
of
people.
D
D
D
A
Awesome,
okay,
my
dogs
are
done
giving
their
opinion,
so
anybody
else
have
anything
that
you
want
to
share
or
talk
about
before.
We
call
it
for
this
time,
see
we're
in
August
coming
up
to
September,
I,
guess
the
next
big
thing
at
least
the
Sun,
the
well
there's
all
things
open
is
happening
in
October
I
know
they
have
some
gaming
content,
but
not
a
ton.
A
It
kind
of
makes
sense
to
have
some
folks
there
if
you're
interested
to
go.
Let
me
know
just
a
coupe
con
in
general.
I'm
previously
I
think
it's
sold
out
last
year,
so
I
would
encourage.
If
you
do
want
to
go,
you
know
go
soon
and
then
the
other
one
is
hotels
are
also
getting.
You
know
a
little
hectic,
so
a
lot
of
folks
are
booking.
Now.
There's
also
Airbnb
things
like
that.
A
The
open
source
and
gaming
day
last
week
a
little
bit
easier
to
do
in
San
Francisco.
Since
we
already
have
space
allocated,
but
yeah
it'll
be
yeah.
It
would
be
nice
to
have
everybody
back
together,
but
if
you
know
of
any
events
or
anything
coming
up
between
now
and
then
let
us
know
we
would
love
to
you
know
either
have
a
presentation
or
just
have
a
little
gathering
around
that
as
part
of
the
open
source
in
gaming
day
and
Caleb
answer
I've
had
a
great
presentation
that
they
did,
but
we
also
have
another
deck.
A
That's
kind
of
like
an
overall
overview
of
open
match
that
that
has
now
been
stylized
for
the
open
match
brand,
and
so
that
is
like
it
looks
really
awesome,
which
is
what
happens
when
you
have
a
professional
to
your
sake,
and
so,
if
you're
interested
in
giving
a
talk
about
open
anywhere
reach
out
and
we're
happy
to
share
that
with
folks
will
help
make
sure
it's
updated
to
the
latest
version.
But
we,
you
know
we're
trying
to
build
all
this
content
that
other
folks
can
use
for.
A
C
Will
be
doing
that
this
week
and
I
mean
feedback
and
especially
not
is
not
feedback
of
their
feedback.
We
can
solve
it
this
way
or
I.
Think
the
solution
is
bad
or
good
or
whatever
is
good,
but
even
just
feedback
of
this
is
our
use
case,
because
then
we
can
look
at
all
these
cases
and
come
up
with
solutions
that
fit
those
these
cases.
Oh
yes,
yeah.
A
I
will
actually
you
say
anything
like
I.
Should
we
do
have
an
open
match,
Twitter
account.
We
are
open
underscore
match,
because
someone
else
randomly
claimed
open
match
years
ago
and
I
haven't
been
able
to
grab
it
yet,
but
all
that's
a
good
use
case
for
Twitter
it's
to
just
tweet
out
like
hey
here's.
This
proposal,
please
review.
So
that's
just
another
way
to
get
eyes
on
it.
So
we
can
do
some
of
that
and
then,
however,
else
and
folks
have
suggestions
for
how
we
can
get
more
feedback
on
these
yeah.
B
The
I
guess
an
interesting
note
is,
it
seems
like
at
least
from
the
slack
and
folks
I've
talked
to
a
lot
of
people
are
watching
the
project
like
a
lot
more
than
I
was
expecting
are
actively
watching,
but
not
actively
participating,
so
that'll
definitely
be
something
that
that
I
would
like
to
try
and
work
on.
It's
just
like
trying
to
get
involvement
from
around
just
the
community
in
general.
The
the
Twitter
thing
is
a
very
interesting
idea.
I've
looked
like
to
try
that
out
a
couple
times.
A
B
A
So
I
did
find
the
box,
for
if
you
didn't,
if
you
missed
the
oh,
and
did
you
also
all
see
what
Mark
was
giggling
about
and
his
self?
So
last
week
we
had
two
things:
our
little
private
jokes.
That
happened
was
trying
to
bring
open,
match
t-shirts
to
the
event
and
went
to
our
you
know.
Swag
storage
calls
it
at
the
office
and
the
box
was
missing.
A
But
that
was
something
that
mark
Menzel.
Who
is
the
lead
on
the
ago
Nets
project
had
been
working
on
getting.
You
know,
conversation
about
kind
of
this.
You
know
online
gaming
technology
aspects
at
GDC.
So
that's
a
thing.
That's
really
exciting
and
we'll
put
a
link
in
the
working
dock.
Whenever
I
can
find
that
link
again,
so
people
can
be
aware,
because
I
I
know
like
GDC
CFPs
and
everything
I
think
already
opened
or
they're
about
to
open
yeah.
C
They
open
it
like
the
end
of
the
month:
okay,
yeah
yeah-
that's
really
exciting,
because
it
will
be
game
technology
that
won't
be
too
like
engine
focused
it'll,
be
more
back
infrastructure
I
think
there
is
some
room
for
like
network
technology
for
games,
but
I'm
not
quite
sure
other
to
play
it
but
open
matching.
The
gunner
is
definitely
fit
into
that.
Okay,.
C
A
Would
say
also
like,
if
you're
thinking
about
submitting
to
GDC
or
would
be
interested
to
speak,
but
don't
quite
want
to
take
it
on
alone
that
you
know
we
can
either
start
it.
I've
seen
other
projects
will
start
a
master
dock
of
like
every
topic
that
someone's
submitting
to
a
conference.
So
everybody
knows
what's
being
submitted,
and
you
know
you
know
kind
of
compete
with
each
other.
A
They
always
get
like
a
session
flawed
at
GDC.
So,
at
the
very
least,
we
have
that,
to
kind
of
you
know
talk
about
open
source
gaining
and
the
specific
projects,
and
before
we
know
what
February
will
be
here.