►
From YouTube: CNCF Serverless Working Group 2017-11-09
Description
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 21! Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.
A
A
D
D
D
D
B
B
F
F
F
B
All
right,
I
have
for
past
the
hours,
so
it
must
look
get
started
should
be
a
fairly
quick
call.
Today,
I
was
hoping
to
do
more
of
a
post-mortem
from
TOC
call,
but
unfortunately
I
don't
think
we
actually
completed
the
discussion.
They
had
actually
there's
one
more
full
TOC
causal
you
ever
had,
and
unfortunately,
that
meant
that
for
working
group,
group
updates
can
only
had
I
think
five
minutes
or
less
in
order
to
presents
the
serverless
stuff
and
he
actually
managed
to
squeeze
most
of
it
in
or
all
of
it
and
actually
under
time.
B
But
unfortunately,
what
that
didn't
leave
time
for
was
discussion
about
possible
next
steps
like,
for
example,
how
does
the
TOC
feel
about
us
looking
at
venting
and
stuff
like
that
that
we've
all
talked
about
so
I've
asked
Chris
to
add
us
to
the
agenda
for
next
week's
call,
and
just
for
you
guys
for
those
who
don't
know
normally
they
meet
every
two
weeks,
but
because
of
Thanksgiving
coming
up,
they
decided
to
shift
it
and
have
the
week
of
Thanksgiving.
This
call
next
week
on
Tuesday
I
believe
it's
a
11
a.m.
B
Eastern,
so
I've
asked
Chris
to
add
us
back
to
the
agenda
just
so
we
could
finish
up
that
discussion,
see
how
the
TOC
feels
about
us
continuing
some
of
the
work
that
we
proposed
and
can't
get
people
a
chance
to
actually
comment
on
what
can
present
they
because
I
don't
think
anybody
spoke
up,
whether
that
was
simply
because
they've,
nothing
to
say
or
because
everything
or
because
they
ran
out
of
time.
It's
a
little
unclear.
So
we'll
give
people
a
chance
to
speak
up.
Then.
C
E
That
was
the
same
to
my
left
with
as
well.
Moreover,
the
other
workgroups
today,
like
the
other
work
website,
I,
spend
time
and
I
guess
technically
this
isn't
the
worker,
but
is
in
the
open,
metrics
effort
that
we
our
last
call
earlier
this
week,
we
were
acknowledging.
We
really
didn't
need
for
us
to
have
an
events
format
as
well,
and
so
so
that
quote-unquote
working
group
is
also
potentially
looking
to
descend
upon
the
the
event
format.
Effort
as
well.
E
Essentially,
or
to
the
extent
that
we
end
up
like
I,
don't
know
to
it,
we
are
considering
a
second
sub
track
here
for
inventing
or
just
going
shifts
the
entire
conversation
to
eventing
later
the
extent
that
we
shift
entirely
to
inventing-
or
we
have
a
separate
call
for
inventing
you
know-
probably
so.
Okay.
B
Cool
yeah,
that's
actually
one
of
the
questions
I
had
for
the
TLC
is
what
would
be
their
preference?
You
know
brand
new
working
group
sub
working
group
or
just
repurpose
this
one
for
that.
That's
a
that's,
not
good
at
COC
to
decide.
I
think
right
for
anybody
else.
Who's
on
the
TOC
call
any
other
comments.
G
B
F
Thank
you,
hey
guys.
This
is
Austin
Collins
here.
I
also
feel
like
this
is
a
natural
progression.
I
think
that
there
are
other
areas
that
we
should
focus
on
as
well.
If
we're
going
to
get
into
the
topic
of
coming
up
with
some
common
specifications
at
all,
vendors
can
adopt
like
what
does
a
standard
or
what
doesn't
the
comment
function
package.
F
Look
like
I
think
that
there
are
a
lot
of
topics
here
and,
while
I'm
all
for
keeping
it
stable
and
kind
of
aggregating
everything
into
the
same
working
group
I
just
want
to
call
out
in
voice
early
I
think
that
this
will
expand
as
we
get
into
more
topics,
so
maybe
keep
it
simple
for
now
and
then
break
it
up
when
it's
when
they're.
Just
too
many
topics
to
have
a
single
conversation
or
a
single
call
based
on.
G
B
Any
other
comments
about
the
TOC
call
all
right.
Moving
on
I,
just
remembered
right
before
the
call
that
I,
never
I,
don't
think
I
ever
actually
wrote
down
sort
of
the
next
steps
for
us
in
terms
of
the
documents
I
think
we
talked
about
things
like
moving
the
the
document,
the
white
paper
itself
into
a
markdown,
so
it's
part
of
our
github
repo
and
then
people
can
you
shoot
PRS
against
it
to
make
changes.
I'm
I
want
to
make
sure
that
everybody's
on
is
an
agreement
that
that's
gonna
be
one
of
our
next
steps.
B
Once
we
get
agreement
from
the
TOC
that
they
do
actually
want
this
document
to
be
something
that
is
long-lived
and
and
not
just
tossed
out
whether
other
AIS
or
action
items,
people
can
think
of
that
I'm,
forgetting
because
I
think
we
agreed
to
keep
the
landscape
as
a
Google
spreadsheet,
but
is
well.
There
are
other
things.
People
can
remember.
B
Okay,
not
hearing
anything
else
when
I
try
to
forget
that
I'll
take
the
AI
to
make
sure
that
we
do
that
at
some
point
after
the
TOC
does
their
stuff
alright.
So
next
topic
is
the
eventing
stuff,
austin
and
you're
on.
You
want
to
update
us
on
where
things
are
relative
to
those
offline
discussions
that
you
were
planning
on.
Having.
F
Yes,
I
think
I've
been
working
with
all
the
people
who
have
expressed
interest
in
being
a
stakeholder
in
this
effort.
There
is
no
shortage
of
interest
and
there's
a
lot
of
people
from
major
cloud
providers
interested
now
we're
getting
a
lot
of
interest
from
SAS
companies
that
publish
events
or
have
web
hooks
to
publications
and
Lee.
Also,
if
you
could
connect
me
to
anyone
who's
interested
in
this
in
the
open,
metrics
group
that
would
be
great
I
think
yeah.
We
just
need
to
get.
F
We
just
need
to
get
organized
so
we're
you
know,
I've
been
doing
some.
Some
private
calls
with
all
the
people
who
seem
to
be
most
passionate
and
a
lot
of
the
a
lot
of
the
big
cloud
vendors
and
trying
to
get
a
good
sense
of
priorities
that
they'll
need
to
actually
implement.
Those
things
will
have
to
address
first
and
then
create
a
better
scope,
focus
area
for
the
specification
itself
and
I
hope
to
have
that,
after
kind
of
a
working
session
that
we're
going
to
have
next
week.
A
F
F
F
Something
a
new
version
of
the
specification
and
posted
to
github,
and
hopefully,
from
there
and
we'd
like
to
do
weekly
calls.
It
sounds
like
we
can
organize
those
within
the
service
working
group
which
will
help
out
a
whole
bunch,
and
you
know
open
it
up
for
public
comments,
suggestions,
feedback,
all
that.
B
F
F
My
goal
is
just
to
get
everybody
lined
on
a
simple
scope
and
get
find
the
common
priorities
that
people
have
turn
that
into
a
simple
scope
is
like
a
starting
point
for
this
whole
effort.
So
that's
kind
of
that's
like
the
basis
of
the
meeting
I'm,
not
sure
if
we
want
to
open
it
up
and
have
people
dial
in
yet
there's
just
just
going
to
be
a
lot
of
people
right
now,
so
so
I
think
we're
going
to
keep
this
one.
Private,
okay,.
E
F
To
be
honest,
we
haven't
made
any
schema
changes
at
all.
We've
just
been
focused
on
identifying
that
priorities.
Now
that
we
have
more
people
have
expressed
interest
in
this
and
I'm.
Drawing
out
that
initial
scope.
We
will
probably
get
into
the
schema
a
bit
next
week,
but
it's
still
all
going
to
be
very
early,
but
most
of
our
conversation
will
be
around
scope.
You
know
what
everyone's
priorities
are:
we'll
have
some
schema
changes
and
then
the
next
goal
is
just
to
get
it
out
on
github,
where
everybody
can
provide
input.
F
H
The
schema
changes
we
propose
some
schema
changes
to
the
API
Austin
we
as
in
Iran
and
myself,
specifically,
that
a
metadata
could
be
set
externally
to
the
data
itself,
so
staff
at
content
type,
where
you
don't
need
to
sterilize
everything
into
a
larger
JSON.
You
can
do
protobuf
and
others
who
ization
methods.
F
Sure
yep,
that's
what
you're
on
has
communicated
to
me
as
well,
and
we're
gonna
get
there
so
we'll
keep
working
on
it
again.
We
really
haven't
touched
any
schema
changes.
Yet
it's
the
conversation
has
just
been
about
getting
people
at
the
same
table
outlining
priorities
to
figuring
out
what
the
what
the
initial
scope
of
this
effort
is.
So
I
think
you
know
we'll.
D
B
F
Think
I
think
it'll
work.
Let's
start
there
I
think
that
people
could
talk
about
this
stuff
for
a
very
long
time.
I
believe
that's
that's
what
I
see
even
just
going
over
the
schema.
Is
that
there's
just
a
lot
of
stuff
to
talk
about
in
there
but
yeah.
Let's,
let's
use
the
service
working
group
call
first
and
we'll
see
if
we
need
a
break
after
supper.
Call
necessary.
Okay,
sounds
good
I.
B
F
B
B
B
it's.
It's
room,
1.
Looking
on
the
first
level
of
the
Convention
Center,
we
have
room
for
20
people
and
they
asked
us
if
we
wanted
to
advertise
this
and
we
basically
came
back
and
said
no,
not
because
we're
trying
to
keep
it
private,
but
because
we
do
want
to
have
a
productive
discussion
and
we
were
afraid
that
we
got
swamped
with
you
know
a
hundred
people.
Nothing
would
actually
get
done,
so
we
decided
not
to
advertise
it.
B
So
we
have
tables
for
about
20
people
we'll
have
a
screening
projected
or
if
we
need
that
and
what
I'll
do
is
I'll
start
a
Google
Doc
or
something
for
us
to
start
gathering
topics
for
discussion.
My
hunch
is
that
a
may
end
up
being
a
working
group
session
on
eventing,
but
we'll
have
to
see
how
that
plays
out.
B
F
Anything
else
I'm
also
in
support
I'm,
also
in
support
of
a
working
group
session
on
eventing
there
I
think
it'll
be
a
great
time
just
to
you,
know,
meet
everybody
face
to
face
and
I
sit
down
and
continue
to
work
on
this
thing
as
far
as
advertising,
it
I
think
it's
a
unique
opportunity
to
just
kind
of
do
initial
introductions
and
meet
people
who
are
interested
in
this
effort.
Overall
I,
don't
know,
I
agree
as
a
working
session.
Is
it
the
right
place
for
that?
F
H
F
B
Yeah
I
I
know
that
there's
at
least
one
session
at
the
conference
itself.
That
will
talk
about
this.
This
working
group-
and
you
know
what
our
future
plans
are.
So
there's
a
little
bit
of
advertisement
there
I,
don't
think
we're
not
going
to
mention
the
face
to
face
that
evening
again
just
to
keep
them
our
people
to
a
reasonable
amount,
but
you're,
probably
right
us
and
at
some
point
we
probably
should
make
some
sort
of
public
announcements
tempio
of
their
interest
in
the
inventing
side
of
things
that
they
should
join
the
working
group.
F
B
Only
the
other
thing
I
have
on
the
agenda
is
agreement
on
the
next
set
of
calls,
which
he
already
sort
of
touched
on
my
heart
agreement
that
we're
going
to
keep
the
weekly
meetings
the
same
timeslot
on
Thursdays
skipping,
the
one
in
two
weeks,
which
is
Thanksgiving
week
here
in
the
US.
So
the
attendants-
probably
pretty
slim
anyway,
but
is
there
any
comments
or
objection
to
heading
down
that
plan
or
handle
a
path.
E
Just
maybe
a
short
piece
of
clarification
and
I
get
where
Austin's
trying
to
I'm
almost
like
started
like
starting
a
fire.
You
gotta
be
really
gentle
with
how
much
oxygen
you
give
it
initially
or
not,
and
but
also
understanding
that
we're
on
the
tail
end
of
the
white
paper
effort
and
our
landscape
effort,
and
the
conversation
is
probably
naturally
turned
to
the
inventing
like
in
a
week.
Do
I
get
done.
E
If
you
go
to
start
that
other
document
and
write
up
some
topics,
maybe
this
will
come
out
of
there,
but
I
suspect
we'll
have
a
hard
time,
not
focusing
discussion
towards
events
next
Thursday
and
and
towards
even
except
that
face
to
face,
and
so
I
guess
just
assessing
some
reservation
on
how
many
people
in
care?
What
format
to
do
that
within
I
thought
I'd,
give
it
a
sort
of
a
pre
warning
that
I
suspect
will
be
hard
to
keep
people
from
discussing
so
anything
that
we
can
do
Austin.
B
E
Topics
and
that
people
will
probably
have
a
National
incarnation
to
bring
up
the
event
and
so
yeah
if
the
way
in
which
we
want
to
steward
a
shape
or
Stewart
the
eventing
discussions
on
you
know
now
or
what
kind
of
before
next
Thursday
is
a
good
time.
It
has
to
try
to
draw
the
box
for
how
how
we
would
like
for
us
to
have
that
sort
of
discussions
or
outline
what
some
of
those
goals
are.
Yeah.
F
Yes,
I
have
a
suggestion
here
to
I
think
we're
going
to
get
a
we're.
Gonna
have
a
clear
scope.
We
have
clear
priorities
by
next
week.
Maybe
I
could
have
some
time
to
chat
on
that
next
call
and
just
give
people
a
basic
introduction
and
the
reasoning
behind
why
we
think
this
is
a
good
scope
to
start
with,
and
then
we
could
kind
of
see
how
things
go
organically
from
there
sounds
good
to
me
all
right.
F
You
know
one
other
comment
on
this
again:
I
think
that
there's
room
for
other
kind
of
common
specifications
within
the
general
service
category
function
standard
function,
packages
I
know:
we've
been
working
on
this
stuff
for
a
while
I'm,
not
sure
if
anyone
else
wants
to
take
the
lead
on
these
ideas.
I
know
that
this
you
know
probably
some
of
these
topics.
B
Anyone
on
that
so
I
personally
wouldn't
mind
having
sort
of
a
list
of
additional
topics
that
we
want
to
consider
for
felonies,
the
S
word
harmonization
or
some
sort
of
level
of
interrupt
going
forward,
but
I'd
seem
to
recall
in
previous
calls.
There
was
a
little
bit
of
a
concern
that
if
we
try
to
tackle
too
much
it
would
be.
B
It
would
just
be
basically
just
too
much
to
do
all
at
once
and
paralyze
it
so
we're
looking
to
serialize
just
a
little
bit
now
doesn't
mean
we
have
to
be
completely
serialize
with
everything,
but
there
was
some
concern
that
I
may
spread
ourselves
too
thin.
If
we
try
to
do
too
many
things
at
one
time
and
that
eventing
was
probably
the
right
thing
to
start
off
first
but
I
do
think
it
would
be
useful
to
hear.
B
F
Right,
yeah
and
I
raised
a
lot
of
those
concerns,
because
that's
the
feedback
I've
gotten
directly
from
all
these
stakeholders.
All
this
is
a
lot
for
them
to
digest
and
to
consider
all
at
once.
So
definitely
a
in
favor
of
keeping
the
conversation
focused
on
events.
I
just
want
to
bring
it
to
everyone's
attention
that
there
are
other
concepts
that
we
should
be
considering
at
some
point
in
the
near
future.
D
Yeah
in
in
this
job,
even
if,
if
not
in
person
during
the
working
group
sessions,
we
could
do
something,
maybe
in
the
dock
asynchronously
so
I'd
be
happy
to
take
the
lead
on
some
of
that,
we've
been
thinking
a
lot
about
function,
signatures
and
and
standardized
inputs
and
stuff
like
that.
So
it's
something
that
I'm
happy
to
take
on,
but
I'm,
just
getting
my
feet
on
the
ground,
so
I
won't
commit
to
too
much
until
I
kind
of
get
a
figure
out
what's
going
on
fair
enough,
but
Doug.