►
From YouTube: CNCF Serverless Working Group 2019 11 07
Description
Join us for Kubernetes Forums Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
CNCF Serverless Working Group 2019-11-07
A
B
B
D
F
C
C
G
C
G
G
G
H
C
Later
Tommy
yeah
are
you
there
or
Scott.
I
G
C
C
C
H
C
J
E
C
C
C
C
We
have
obviously
the
stock
here
with
all
the
information
about
what
we're
presenting
and
stuff
some
people
have
made
some
updates.
The
presentation
I
think
it's
fair
to
say
it's
still
very
rough.
It's
not
completed
yet
I
have
gotten
pings
from
some
people
that
they're
still
actively
working
on
it.
So
if
you
want
to
see
a
near
final
version,
I'd
hold
off
a
little
bit
longer.
C
C
I
do
want
to
have
a
discussion
with
you
guys
at
the
end
of
the
call
today,
for
just
you
guys,
so
everyone
else
can
drop
assuming
you
can
stay
on
the
call
that
long
because
I
know
some
of
you
sometimes
need
to
drop
early,
but
I
do
want
have
a
little
bit
discussion
about
the
presentations
and
stuff
to
see
everybody's
at.
If
you
guys
are
okay
with
that,
we
do
have
the
face-to-face
meeting
scheduled,
please
if
you
are
planning
on
joining,
had
your
name
here.
C
The
link
to
the
dock
is
right
here
in
our
meeting
agenda
practitioners
summit,
nothing
official
for
us
going
on
there,
just
a
reminder
that
that
is
going
on
if
you
want
to
join,
that
is
a
day
before
the
coupon
event
itself,
so
that's
Monday
all
day,
still
working
on
whether
we
can
do
some
sort
of
celebration,
or
at
least
some
sort
of
meet-and-greet
for
us
to
get
together
to
at
least
say
hi
to
everybody
and
currently
each
other.
Alright,
anything
relative
to
coop
con
people
want
to
bring
up
all
right
excellent.
C
I
I
I
I
I
Some
new,
some
new
things,
not
just
Mantic
semantics
change
to
on
so
I-
think
our
goal
is
to
make
this
workflow
spec
more
I'll,
go.
That's
my
okay
I'll
go
Maya,
making
a
vendor
neutral
and
also
portable
across
the
front
arm
somebody's
platforms.
So
we
would
like
no
more
people
to
join
this
effort.
I
think
you
know
there
are
more
people
joining
in
this
effort.
You
know
we
can
make
this
more
comprehensive
and
more
generic
in
the
more
usual
a
vendor,
neutral
and
portable
yeah.
So
welcome
time
you
know
how
they
recommend.
I
You
take
a
look
at
the
new
spec.
It's
now
you
know
much
better
readable
format,
and
then
you
know
you
contribute
to
it.
I
think
it's
a
very
important
how
to
say
it's
an
important
component
or
important
part
of
the
service
platform,
because
country
most
a
platform
supports
only
a
function
execution
triggered
by
a
simple
event.
I
But
still
it's
a
you
know,
you're
the
most
support,
but
quite
some
applications
are
not
a
simple
function.
They
are
composed
of
multiple
steps
of
multiple
functions
with
each
function,
executing
a
different
step
and
those
function
could
be
triggered
by
different
events,
a
different
time.
I
I
C
C
So
a
couple
things
aside
from
obviously
Kathy's
been
leading
the
charge
on
the
workflow
spec,
but
recently
there
has
been
an
uptick
of
activity,
mainly
from
Red
Hat
they've
become
interested
in
the
spec
and
they've
been
working
with
Kathy
on
making
changes
to
it
and
that
sort
of
brought
back
up
the
entire
question
of
next
steps
for
the
specification
itself,
and
there
has
been
some
offline
discussions
about
going
back
to
the
technical
Oversight
Committee
to
propose
that
the
workflow
spec
be
sort
of
spun
off
into
a
new
independent
sandbox
project
very
similar
to
what
we
did
with
cloud
events.
C
However,
from
a
process
perspective
and
I'm,
not
100%
sure,
that's
the
way
it
has
to
be,
but
it
seems
like
that
makes
no
sense
to
me.
It
seems
like
that
this,
the
servlets
working
group
itself,
first
needs
to
agree
that,
yes,
we
want
to
fork
this
off
and
propose
it
as
a
new
project.
I
can't
I
met
anybody.
Zest
I'm
gonna,
say
no,
but
I
figured
that
the
the
working
group
needs
to
be
able
to
I'm
sorry
Cathy
or
whoever
takes
us
forward
to
the
TOC
needs
to
be
able
to
say.
C
C
So
what
I'd
like
to
do
is,
from
a
procedural
perspective,
kick
off
just
a
one-week
vote,
since
a
lot
of
people
are
not
online
because
we
told
them,
it's
gonna,
be
a
slow
week
to
start
a
one-week
votes
for
yes
or
no
on
whether
the
working
group
approves
of
Cathy
or
the
working
group
subgroup,
taking
the
spec
forward
to
the
TOC
as
a
new
sandbox
project.
Is
that
saying,
okay
with
people
or
is
there
a
different
process?
You
guys
think
we
should
follow
here.
C
C
So
one
yes
vote
is
enough:
if
no
one
else
boats
all
right,
any
other
questions
comments:
okay,
so
Kathy
I'm
hearing!
No,
no
one
object
to
that
process.
Is
that,
okay
with
you
in
terms
of
next
steps,
next
stencil
I
know
you
were
talking
about
going
to
the
TOC
and
it's
one
to
make
sure
that
I
got
everything
covered
in
terms
of
a
process
perspective
yeah,.
I
C
The
the
vote
is
just
on
whether
the
working
group
agrees
with
with
going
forward
to
the
technical,
Oversight
Committee
on
proposing
a
new
sandbox
project.
The
technical
oversight
committee
will
still
have
to
make
that
decision
themselves
and
they
could
vote
yes
or
no.
We
don't
know,
but
the
purpose
of
us
doing
it
is
because
the
working
group
as
a
whole
owns
the
workflow
work.
I
figure,
it's
it's.
They
should
have
the
right
to
say
yes
or
no
on
whether
to
go
forward
to
the
TOC
or
not
so
principal.
C
If
they
say
no,
then
we
have
a
decision
to
make
either
it
stays
as
it
is
or
if
you
feel
strongly
about
or
that
they,
if
the
workflow
subgroup
feels
strongly
about
it,
you
guys
could
choose
to
fork
the
project
and
go
forward
any
way
to
the
TOC.
You
you
just
would
not
have
the
blessing
of
the
working
group
of
the
of
the
service
working
group
as
you
go
forward.
C
I
C
L
C
Are
basically
the
two
choices
as
I
see
it?
Okay,
but
does
seem
like
with
the
increase
in
activity
you're,
getting
close
to
the
point
where
you
really
should
be
a
little
more
stand-alone
and
not
hidden
behind
the
surrealist
working
group,
because
I
think
that
would
also
give
you
more
exposure
to
as
a
standalone
project
and
get
more
people
interested.
So
I
think
it
could
have
a
lot
of
benefits
to
you.
Yeah
I
think
that
there
could
be
a
third.
E
C
Yeah,
that
is
definitely
true.
Yeah,
that
is
true.
Yeah
I
was
trying
to
remember.
I,
had
a
conversation
with
Chris
Anna
check
about
the
possibility
of
of
just
creating
a
brand
new
repo
for
it,
because
it
was
a
little
awkward
for
a
while
there
being
under
the
serverless
working
group
repo,
because
they
think
that
you
know
I
had
to
give
them
basically
access
to
the
entire
repo.
C
Just
so
they
can
merge
their
own
pr's,
which
was
kind
of
awkward
and
and
Chris
was
a
little
hesitant
to
propose
them
getting
their
own
repo
without
them
having
some
sort
of
official
status.
Someplace
kinda
like
what
we
did
with
cloud
events,
and
so
that's
why
we're
heading
down
the
TOC
path,
and
so
the
TOC
says
no
I'm,
not
sure
what
to
do
about
the
separate
repo
kind
of
a
thing.
Well,
we'll
have
to
decide
that
later,
yeah.
D
C
But
you're
right,
there
are
probably
other
variants
of
options
available,
we'll
have
to
see
how
it
plays
out.
Okay,
all
right
any
other,
oh
and
I
should
mention
just
to
reiterate
what
Kathy
said,
because
I
did
mention
this
last
week
and
has
a
homework
assignment
for
people
to
go
back
to
talk
to
the
respective
companies
to
see
if
you're
interested
in
the
workflow
stuff.
C
And
so
obviously,
if
you
are
please
reach
out
to
Kathy
or
just
start
participating
through
the
issues
and
PRS
and
stuff
through
they
work
through
the
service
working
group
repo
because,
as
you
said,
they're
looking
for
people-
and
it
is
getting
a
little
more
interest
now
from
Red
Hat,
all
right
all
right.
Any
other
questions
or
comments
on
workflow,
all
right,
cool
and
Thank
You
Kathy
for
the
update
sure.
I
I
would
I
took
ask
one
more
comment,
so
if
you
really
like
to
by
the
direction
to
help
guide
the
direction
of
the
workflow
I,
think
it's
very
good,
you
know
you
joined
their
effort
to
you
know
I
mean
to
either
review
it's
very
important
to
review
those
PR
and
to
provide
your
perspective
and
to
maybe
contribute
some
PR,
so
that
I
think
that
would
be
very
will
make
that
spec
more
I
mean
better
and
there
will
be
more
healthy.
I
mean
a
subgroup.
Yeah
I
would
really
appreciate.
C
Great
alright,
thank
you
alright.
So
we
have
no
PRS
to
review.
Obviously,
and,
as
I
said,
we
get
a
call,
we're
gonna.
We
decided
to
wait
until
kook,
on
and
after
kook
on
to
work
on
the
next,
my
next
items,
and
so
as
a
reminder
that
face
to
face
meeting
that
we're
gonna
have
in
which,
of
course,
is
this
list
here
is
where
we're
gonna
have
the
discussion
about
what
what
to
work
on
next.
So
please
do
your
very
best
to
try
to
join
that
conversation.
C
If,
for
some
reason,
you're
not
going
to
be
at
KU
con,
we
will
try
to
have
the
zoom
meaning
set
up
and
going
so.
People
remotely
should
be
able
to
be
able
to
participate,
I'm,
not
sure
how
great
what
time
is
it
yeah?
It's
not
the
greatest
time
for
European
folks
I
apologize
for
that,
but
this
was
the
only
time
that
they
get
us
a
room,
but
we'll
see
what
we
can
do.
C
Obviously,
if
you
have
some
input
in
terms
of
which
things
to
work
on
feel
free
to
add
comments
to
this
doc
here
down
here
in
the
bottom
section
and
we'll
try
and
take
those
into
account
and
with
that
I
believe
at
the
end
of
the
agenda,
are
there
any
other
topics?
People
I'd
like
to
bring
up
for
discussion
today.
C
C
Yeah
yeah
after
the
coop
car
discussion
will
have
the
SDK
call
I
forgot
about
that.
Sorry,
yes,
okay,
yep
and
to
Pina
either
yeah
me
all
right.
Jim
I,
have
you
Matthew,
see
there
I'm
you're
all
right
did
I
miss
anybody.
Cuz
I
thought
it's
some
day
battery,
but
I'm
late.
Oh
there
we
go
yeah
David
anybody
else.
C
C
C
C
C
C
Okay,
a
couple
questions
I
had
for
you.
First
of
all,
I
allocated
30
minutes
for
your
stuff
yeah.
Is
that
sufficient,
given
that
I
was
hoping,
you'd
also
kind
of
cover
the
demo
and
talking
about
what
we're
gonna
do
after
1.0,
because
I'm
not
sure
you
mentioned
1.0,
yet
in
your
charts
or
post
1.0,
no.
K
C
L
K
C
K
E
After
that,
we
then
have
one
introduction,
which
is
the
post
plans.
1.0
could
also
be
you
know,
potential
future
plans
right,
it
doesn't
mean
the
weights
forwarded
there.
It
seems
like
here's
what
we
will
be
doing
post
1.0,
as
opposed
to
here
possible
areas
like
we're
talking
about
with
a
certain
Louis
Kirkland
group
yeah.
K
K
Yeah
I
mean
there's
there,
we
have
some
actual
gaps
or
something
I
need
to
be
very
creative
here
right
we
have
a,
we
have
security.
We
have
security
that
you
probably
need
to
talk
about.
We
have
the
subscription
API
that
we
need
to
talk
about.
We
probably
need
to
have
a
schema
registry
of
sorts.
It
not
things.
Neither
are
things
that
we
need
to
invent,
but
it
would
be
good
to
have
an
opinion
about
which
ones
we
like
so
so.
K
C
K
C
Right
and
then
for
part
two,
which
was
the
rest
of
the
sessions
I
had,
is
initially
I.
Okay,
ten
minutes
per
person
and
here's
my
current
list,
I've
reached
out
to
Austin
I,
have
not
heard
back
from
him
yet
so,
obviously,
if
I
don't
hear
back
from
them,
they'll
remove
them
from
the
list.
Can
anybody
remember
who
mentioned
Tecton
triggers
I
can't
marry
with
an
E
or
E
or
somebody
else,
whether
you
Scott
do
you?
C
J
C
Okay,
I'll
reach
out
to
the
technical
folks.
Thank
you
like
I,
said
I,
just
can't
remember
who
took
the
AI
there
and
obviously
I
miss
the
boat.
Thank
you
alright.
So
my
question
for
you
guys
is
oh
I,
guess
one
question:
Jim
I
think
I
confirm
this
with
ladder
ever
just
like
double
sure
he
is
okay
with
doing
the
PayPal
stuff
right.
That's
the
current
plan.
Yeah
okay
cool
in
that
case
is
10
minutes
sufficient
for
everybody
on
this
list.
F
C
Anybody
else
have
any
concerns
with
ten
minutes.
I
actually
think
you
might
have
a
little
bit
more
cuz.
We
have
a
whole
60
minutes.
I,
don't
think
we
ran
out
of
six
sessions.
I
think
we're
down
to
five
already.
You
know,
I,
don't
know
if
Austin's
gonna
respond,
but
it's
around
10
minutes
enough.
Everybody
yeah.
K
N
C
C
A
E
C
Yeah
and
like
I,
don't
think
I've
mentioned
the
previous
calls
most
slides,
that
I
put
together
we're
strictly
a
copy
and
paste
from
our
previous
presentations
and
you're
under
zero
obligation
to
use
anything
I
put
in
there
whatsoever.
It
was
just
a
starting
point:
yeah
made
us
feel
like
we're,
making
progress,
yeah.
C
And
I
think
I'm
it's
in
this
someplace
I,
don't
know
for
sure
in
weather.
You
guys
could
actually
update
the
presentation
behind
this
link.
It
is
on
Google
Drive,
even
though
it's
not
a
Google
Doc
itself,
it's
a
PowerPoint.
If
you
guys,
cannot
update.
What's
behind
this
link,
let
me
know
and
I'll
try
to
figure
out
some
way
to
get
you
a
URL
to
the
drive
itself
or
something
so
you
can
actually
update
it
and
don't
go
through
me
to
do
updates
it's
just
I'm,
not
sure
I'm,
using
Google
Drive
correctly
well,.
E
C
C
K
C
K
K
C
K
C
C
K
L
C
K
K
K
C
A
C
H
K
K
C
C
C
M
Yeah
I
did
I
actually
wanted
to
use
the
SDK
to
do
some
actual
real
work.
So
I
picked
the
goal
going
SDK
and
raised
an
issue
on
it,
which
I
note
by
the
way.
This
is
a
couple
days
ago
that
a
friendly
response
this
morning,
but
I
just
wanted
to
suggest
a
statement
of
principle,
which
is
the
way
I
think
about
events
is
as
immutable
things.
You
know
once
I've
created
an
event
and
want
to
ship
it
around.
M
It's
got
an
ID
and
everything
I
would
expect
people
to
have
check
sums
on
it
and
pass
it
around.
Thus,
I
was
kind
of
concerned
in
the
golang
SDK
that
it
was
just
a
typical
getters
and
setters
thing
and
the
notion
of
having
an
existing
event
and
just
setting
the
type
to
something
else,
or
something
like
that
struck
me
as
weird
strange
and
puzzling
in
our
you
know.
Internally.
M
Here
we
tend
to
have
builder
or
factory
things
where
you,
you
saw
the
stuff
and
you
say
build
and
you
get
but
and
then
the
actual
API
for
events
has
no
setters.
So
you
know,
I
took
the
spec,
doesn't
say
anything
about
mutability
or
changing,
or
anything
like
that.
The
cloud
event
spec
I
mean
but
I
just
wanted
to
just
sort
of
suggest
that
I
think
it
would
be
a
good
thing
for
event.
M
N
There
is
what
there
is
already
there
is
already
an
event
context,
reader
interface.
So
if
you
wanted
to,
if
you
want
to
make
it
clear
in
your
code
that
certain
pieces
of
code
are
not
allowed
to
change
the
things
you
could
pass
those
around
instead
of
instead
of
just
passing
their
the
writable
events
at
some
point,
you
need
to
be
able
to
construct
these
things.
So
it
seems
like
it
makes
sense
to
be
able
to
get
and
set
the
values
when
you're
constructing
it,
though
so.
K
Okay,
I
can
see
for
this,
so
I'm
from
the
c-sharp
SDK
perspective,
because
if
that
was
a
principal,
we
would
have
to
employ
that
too
and
I
can
see
having
a
affecting
the
interfaces
which
then
you
can
put
on
top
of
each
other,
where
you
have
a
get
interface
first,
a
read-only
one,
and
then
you
have
a
rewrite.
That
effect
extends
that
one
and
the
event
implements
both.
N
J
M
E
C
J
J
N
M
Caching,
just
based
on
the
on
the
you
know,
some
subset
of
fields
would
be
nice.
You
know
if
I
know
the
source
and
the
ID
you
know,
can
I
just
cache
on
that
basis
or
do
I
have
to
hash
the
whole
event
and,
if
you're,
storing
it
into
something
like
a
modern
law,
Journal
based
database.
You
know
the
whole
idea
is
of
you
know:
append-only
immutable
data
structures
are
at
the
core
of
a
lot
of
modern
high-performance
systems
and
I
don't
want
to
come
across
like
a
crusty
old,
functional
programming,
geek,
but
I
am
well.
N
Let
me
just
scrape
up
some
of
that
crust.
There
ya
know
I
get
I.
Take
your
point.
I.
Don't
think
that
we
should
have
a
lot
of
kind
of
requirements
in
the
in
the
in
the
spec
about
that,
because
different
languages
treat
that
sort
of
thing
very
differently.
Some
languages
have
a
lot
of
support
for
immutability
go.
Doesn't
we
have
any,
except
for
defining
interfaces
that
don't
have
setters?
N
M
M
N
C
It
I
haven't
sort
of
a
higher-order
question.
Look,
so
let
me
try
understand
this
it.
It
seems
to
me
the
to
put
this
as
a
requirement
on
the
STK.
Almost
feels
a
little
bit
odd
to
me,
because
I
always
look
at
SDKs
and
sort
of
like
utility
kind
of
things
right,
because
even
if
the
SDK
treats
everything
is
immutable,
you
can
get
around
that
because
you
can
obviously
use
the
SDK
to
create
another
event
and
you
can
pass
it
whatever
field
you
want.
C
So
you
just
take
the
old
event,
change
it
or
field
you
want,
while
you're
calling
the
factory
of
the
new
event
and
then
use
that
as
you
go
forward,
your
processing,
so
I'm
wondering
whether
this
is
is
not
so
much
a
factor
of
whether
the
SDK
should
allow
or
disallow
I
think
they
can
make
that
choice
if
they
want
to.
But
if
that's
a
big
concern
for
a
particular
environment,
isn't
that
more
of
an
environmental
concern
and
a
choice
for
that
particular
business
logic
to
decide
how
they
want
to
manage
these
things.
M
M
E
N
Would
be
Oh
before
I
would
be
fine
with
having
again
her
only
version
of
the
event,
but
not
with
having
no
version
with
centers,
because
the
factory
pattern
is
a
useful
one,
but
there
are
also
lots
of
patterns
where
people
are
gonna
be
building
events.
You
know
over
time
where
they
pass
it
through
several
functions,
each
of
which
adds
some
data.
M
N
C
M
J
Know
the
little
tags
that
works,
because
the
the
marshal
has
to
have
access
to
the
member.
And
if,
if
it's,
if
it's
private
and
how
is.
J
M
K
J
D
K
M
M
K
C
K
C
J
Okay,
who's
me,
yes,
Scott,
yeah,
I,
I,
think
we're
open
to
anything.
I
was
asking
Tim
if
the
ability
to
support
all
the
mul,
multiple
versions
of
cloud
events
is
a
feature
or
are
you
only
targeting
v1
in
our
particular
case,
we're
not
looking
at
anything
other
than
v1,
because
we
were
bringing
a
lot
of
baggage
right
now
supporting
all
the
types.
M
J
N
Coming
with
swords
at
some
point,
we're
going
to
do
a
v2,
and
maybe
so
the
part
of
the
goal
of
v2
should
be
to
remove
stuff
that
we
don't
really
need
anymore
rather
than
just
to
add
features
so
that
we
have
a
simpler
people,
can
walk
up
to
it
and
see
more
easily.
What
is
the
the
path
we're
recommending?
They
take
yeah.
J
N
Just
surly
said,
since
there
Christie
and
people
in
the
room
that
make
it
we
might
get
a
reaction,
and
so
one
of
the
things
I
would
like
to
see
in
the
neck
and
in
the
v2
version
of
the
SDK
is
a
more
as
a
sender
receiver
like
API,
rather
than
what
we
have
now,
which
is
really
quite
modeled
on
the
HTTP
request/response
style
of
API.
Because
again
this
this
exactly
what
Tim
was
saying
about
encouraging
people
to
do
the
right
thing
you
know
eventing
is
not
supposed
to
be
a
replacement
for
HTTP.
N
It's
supposed
to
be
a
way
of
achieving
something
different
and
the
big
difference
is
that
you
should
be
thinking
about.
You
know
you
send
an
event
and
you
forget
about
if
you
don't
wait
for
the
result
of
that
event,
there
isn't
the
results
you
have
you've
got
potentially
got
some
error
errors
to
tell
you
if
the
event
didn't
get
sent
and
you've
got
some
quality
of
service
guarantees
about.
N
You
know
how
likely
it
is
to
be
sent,
but
you
you're
not
waiting
on
a
response,
and
likewise,
if
you're
receiving
events,
it's
not
because
you
sent
events.
It's
just
because
you're
receiving
events
and
I
think
that's
important
too
make
clearer
in
the
API,
because
that's
the
that's
that's
where
you
get
benefit
really
from
using
events
over
HTTP
is
when
you're,
not
you
know
when
you're
using
patterns
that
are
not
straight
request.
Response
like
I,
send
work
to
you
and
you
send
there's
also
somebody
else
and
I
don't
care
those
kind
of
patterns.
N
N
We
need
we
need
to.
We
have
we
have
some
use
cases
where
we
need
to
preserve
the
request
response
pattern,
but
I
would
like
to
make
that
a
kind
of
an
explicitly
different,
a
different
thing.
So
basically
you're
basica
API
czar,
a
sender
which
has
send
and
a
receiver
which
has
received
sort
of
like
read
and
write
io,
reader,
I
or
writer,
and
then
a
requester
where
you
send
and
you
get
back
a
receiver
which
will
eventually
give
you
the
response.
N
It
should
be
the
case
for
all
SDKs
I.
Think,
that's
because
that's
the
kind
of
the
pattern
we
want
to
provoke
for
promote
for
eventing,
as
you
send
an
event
and
usually
sending
and
receiving,
are
completely
different
things
and,
of
course,
in
an
application.
You
can
tie
them
together
to
achieve
things
like
request
response
or
whatever
kind
of
patterns
you
want,
but
that
they're
not
we
don't
assume
that
you're
going
to
use
it.
C
N
C
N
H
Yeah
I
keep
getting
feedback
on
the
SDK
internally
from
from
colleagues,
and
they
see
the
biggest
value
always
in
having
this
way
to
to
marshal
unmarshal
cloud
events
for
the
different
protocol
bindings,
but
they
usually
have
the
desire
to
control
the
whole
clients
and
and
everything
on
their
own.
So
all
those
sender,
receiver
and
so
on.
They
aren't
really
interested
in
this
I
have
to
say
they
are
fine
with
akin
to
Marshall
and
Marshall
yeah.
They
have
that
option.
They
can
click
down
to
just.
A
H
C
Okay,
one
procedural
question:
I
think
the
next
scheduled
call
that
we
have,
or
the
next
SDK
scheduled
call.
We
have
is
in
two
weeks
which
obviously
overlaps
with
the
coop
gum.
I'm,
assuming
we
don't
want
to
have
the
meeting,
because
everybody's
getting
busy
with
Kuk
on.
Would
you
guys
like
to
have
another
SDK
meeting
next
week
or
actually
I
was
gonna,
say
push
it
out
to
the
week
after,
but
the
week
after
to
be
Thanksgiving
for
the
u.s.
folks,
I
suspect
that
won't
be
good.
C
C
Okay
and
obviously
you
guys
chat
through
this
slack
channel
in
a
way,
so
we
don't
have
to
wait
but
I
guess
for
some
of
the
folks
who
don't
usually
hang
out
there.
For
example
Tim
you,
you
were
waiting
for
this
phone
call
that
bring
up
to
these
discussions.
Obviously
don't
wait
for
the
you
know,
for
the
full
month
bring
up
topics
on
the
slack
channel.
If
you
have
things
you
want
to
mention.
Okay,
all.