►
From YouTube: CNCF Serverless Working Group 10.26.17
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
A
Interesting,
you
don't
have
Ken
a
Merc
you're
on
good,
okay,
okay,
throw
it
I
want
to
go
ahead
and
get
started;
first,
imagined,
okay,
so
with
the
agenda
today,
basically,
obviously
gonna
try
to
finish
up
the
white
paper
as
best
we
can.
Anyway.
There
are
a
couple
of
of
outstanding
comments
that
I
just
wanted
to
get
some
discussion
going
on.
Let's
start
with
the.
A
We
don't
really
need
to
do
it
inside
the
white
paper
itself,
especially
since
at
some
point
people
are
going
to
want
to
continually
add
it
or
add
new
things
to
this
list
and
which
means
more
updates
to
the
white
paper
and
I'd
rather
I'd,
rather
save
that
for
this,
the
spreadsheet.
That
would
go
up
to
keep
those
two
in
sync.
So
what
I
was
thinking
about?
A
Doing
was
erasing
all
the
stuff
in
yellow
here
and
just
combining
the
two
paragraphs,
but
before
I
went
off
and
did
that,
since
it
does
involve
product
names,
I
didn't
want
people
to
be
upset
or
anything,
but
I
want
to
get
your
guys.
Thoughts
on
that
before
I
actually
did
it
in
other
areas
that.
D
No
I,
but
one
comment
from
my
perspective,
Matt
here
for
first
time,
so
I
apologize
already,
but
I
think
that
we've
been
noting,
especially
inside
IBM
I,
think
it's
noted
by
some.
The
names
you
see
here
that
you'll
be
deleting
likely
is
that
there
needs
to
be
discussion
of
why
serverless
discussions
are
shifting
more
towards
the
name
function,
functional
workloads,
cloud
functions,
because
basically,
we've
acknowledged
that
it's
the
workload
we're
discussing.
Not
the
concept
that
you
don't
have
a
server
is
more
important
about
talk
about
the
workload.
D
A
E
The
reason
that
I
put
it
in
there
was
to
allow
people
to
align
what
this
document
was
talking
about
and
the
popular
names
I.
You
know
we
we
can
debate
popular,
but
the
popular
names
of
the
services
that
people
would
likely
be
able
to
draw
a
relationship
with.
So
you
know,
I'm
fine
with
us
removing
I
do
it
was
it
was
merely
there
to
be
able
to
have
people
start.
Think
oh
well,
you're
talking
about
is
something
along
the
lines
of
what
AWS
lamda
is
providing
so.
C
A
F
F
C
A
So
my
let's
not
talk
about
that
later,
but
since
you
bring
that
up
now,
Kathy
I
I'm
a
little
uncomfortable
with
giving
right
access
to
the
whole
world.
Obviously
I
would
rather
have
people
try
to
add
to
it
by
making
comments,
and
that
way
someone
on
the
work
on
the
working
group
itself
can
then
look
at
it
and
review
it
and
approve
it
now.
I
have
no
probably
giving
everybody
in
this
in
the
working
group.
In
essence,
editor
editorship
permission,
I
just
didn't
want
to
open
up
to
our
world
just
to
make
random
edits.
C
E
A
Let's
see,
let's
jump
to
I
think
that
was
actually
the
easiest
one.
Now,
let's
jump
to
something
a
little
more
exciting
have
people
taking
a
look
at
the
old
section?
What
is
service
computing?
In
particular,
you
know
bullet
our
favorite
bullet
number.
Two
here
are
people
okay
with
what's
currently
in
the
document,
or
do
we
need
to
discuss
the
sport
and
keep
in
mind
I,
don't
think
we're
looking
for
this
to
be
absolutely
hundred
percent,
perfect
we'd
all
would
obviously
don't
want
to
lie,
but
I'm
looking
for
people
to
say
yes,
they
can
live
with.
A
G
I'm
gonna
call
so
I
can't
see
the
exact
text
right
now,
but
I
thought
a
couple
of
things
that
we
talked
about
last
time
was,
we
will
remove
the
word
absolutely
and
we'll
add
the
word
usually,
and
there
was
a
discussion
around.
Usually
it's
kind
of
vague.
You
know
what
do
you
mean
usually
and
it
doesn't
give
any
proper
guidance,
and
we
also
said
that
there
was
a
disagreement
in
the
working
group
itself.
So
has
anything
changed
since
then
so.
G
A
G
A
I
I
Yeah
I
guess
this
is
like
hey
yeah
I,
don't
know
that
there
is
this
absolute,
just
sort
of
reinforces
that
no
and
we're
still
saying
no
cost,
which
implies
absolutely
and
I.
Don't
know
that
I
don't
know
how
quite
how
deep
my
heartburn
goes
here.
I,
don't
know
it's.
Probably
it's
not
very
deep,
but
I
guess
one
of
the
questions
that
I'd
asked
the
team
last
time
was
what
the
negative
ramifications
are
of
using
something
that
isn't
as
definitive
as
now
and
so
like.
I
It
knew
to
the
extent
that
it
would
it's
a
generally
or
usually
and
I
think
some
had
articulated
those,
and
there
was
you
know
there
were
certainly
a
number
of
us
that
were
more
comfortable
with
saying
no
no
cost
and
then
the
other
suggestion
that
I
had
made
as
an
outcome
of
that
was
that
maybe
we
want
to
take
this
from
the
identify
the
perspective
from
which
we're
speaking
and
I.
Guess
we
did
you?
Should
we
have
a
from
a
customer
benefit
so
actually
now
the
rereading
it
might
been
hey
that
really
helps.
I
Yeah
and
in
that
statement
to
it,
says
a
customer
benefit
so
which
I
think
helps
a
ton.
I've
said
I
wouldn't
have
open
my
mouth.
How
do
I
read
that
and
just
a
minute
ago,
that
said
I
think
that
one
thing
for
a
donor,
one
thing
for
us
to
consider
is
that,
to
the
extent
that
you
know,
functions
are
being
run
by
1%
of
the
world's
workloads
or
they
represent
that
it's
some
small
percentage
to
the
extent
that
grows
and
grows
and
grows
over
the
next
few
years.
I
I
got
to
imagine
that
those
service
providers
might
can
begin
to
consider
different
pricing
plans
and
how
it
is
that
they,
you
know,
charge
customers
for
consumption
of
those
services
which
you
know
still
may
never
include.
I
know
time
that
I
did
just,
but
it
might
things
might
change.
I
guess
is
what
I'm
saying
they.
J
Get
we
tried
to
cover
this
with
the
caveat
that
we
put
it
in
in
agreement
with
the
IRA
and
the
other
week
in
brackets,
right
after
that,
we,
the
exception
of
stable
storage
costs,
are
added
capabilities
functionality.
Slash
feature
set,
does
give
you
kind
of
a
catch-all
for
that?
What
we
consider
to
be
usually
your
representational
of
usually.
A
K
Had
a
suggestion,
this
is
Alex
by
the
way
mm-hmm.
Oh,
why
don't
just
change
rather
than
Oh
change,
that
to
minimal
costs?
One
idle
and
then
in
the
paragraph
say
one
of
the
greatest
benefits
is
that
there
are
no.
There
are
minimal
or
no
costs,
because
that
would
let
you
future
proof
it
a
little
bit,
because
if
people
do
start
charging
overtime,
it
still
looks
like
it's.
Fine
and
minimal
is
close
enough
to
know
that
I
think
people
may
be
ok
with
it.
I
don't
know
if
that
makes
sense,
I
think.
D
J
That's
my
fear
I
like
keeping
the
copy
out
there,
because
it
does
cover
the
points
that
yarn
tried
to
make
the
other
way
around.
You
know
what
would
be
the
exceptions
to
it
like
we
had
those
kind
of
ambiguous
terms
like
usually
or
sometimes,
or
maybe
it
looked
a
whole
lot
of
ambiguity
to
what
could
potentially
fit
in
that
bucket,
and
so
I
mean
just
putting
some
amount
of
catch-all
in
there.
I
gives
at
least
context
to
what
could
be
there.
I.
J
D
If
you
have
an
exception,
I
would
probably
have
not
lead
with
it.
I
put
it
as
at
the
end,
saying,
of
course,
these
things
don't
consider
blah
I
would
move
I
wouldn't
lead
with
it
immediately
after
the
in
building
title
I
would
put
it
I
would
take
that
parenthetical
section
and
add
it
more
towards
the
end.
It
possible
I.
C
H
A
H
L
This
is
Austin
with
the
open,
faz
project
and
I
think
if
it
comes
down
to
what
we
or
what
people
can
live
with
or
not
live
with.
I
think
some
of
the
verbage
here
still
would
probably
make
a
lot
of
the
open
source
projects
a
little
uncomfortable
but
I
don't
know
how
much
to
say
we
would
have
in
that.
That
would
just
be
my
opinion
from
a
open
source
perspective.
D
A
A
Going
once
done,
okay,
obviously
this
isn't
sentence,
don't
be
able
to
still
have
time
to
review
it,
but
it
sounds
like
we
may
be
okay
for
right
now,
until
people
go
back
and
think
about
it,
some
more
so,
let's
go
to
the
next
big
change.
I
want
to
discuss-
and
it's
right
here
so
Alex
put
in
a
proposal
for
condensing.
A
F
So
my
my
general
feel
for
this
and
actually
applies
to
the
next
three
next
two
sections
as
well
is
that
I
know
I
do
appreciate
the
yet
it's
I
do
appreciate
the
effort
to
condense,
but
I
think
there
was
quite
a
bit
of
text
in
the
original
that
kind
of
went
through
a
lot
of
community
editing
and
a
lot
of
compromise
and
a
lot
of
discussion,
so
I
wouldn't
want
to
just
remove
it
for
the
sake
of
conciseness,
I
think
there
was
some
important
information
in
there,
but
what
I
would
like
to
do
is
I
see
that
Alex
rewrote
a
few
sentences
here
and
there
that
really
better
written
than
our
community
efforts.
A
K
Mean
it's
really
a
community
paper
and
if
the
community
feels
that
way,
it's
it's
fine
I'm,
just
looking
at
what
I
created
was
from
a
product
marketing
perspective
of
what
I
think
would
be
easier
to
consume
it
easier
for
lay
people
or
people
that
are
maybe
less
technical
to
understand
and
that's
why
I
set
it
up
that
way.
So
so
I
mean
that
that
would
be
my
primary
comment.
The
other
thing
just
to
note
which
I
did
change
is
that
in
every
instance,
I
linked
to
an
actual
real
world
use
case
outside
I.
K
A
Okay,
because
I'm
I'm
inclined
to
take
a
look
and
see
Dan
what
it
would
look
like
if
you,
after
the
cherry,
picking
and
see
how
that
feels.
So
you
wanted
to
take
on
that
task
and
make
those
edits
and
I
guess
in
all
three
sections
and
because
there
was
oh,
the
other
yeah,
because
it's
digital
one
here
for
the
server
list
versus
others,
and
then
there
was
one
more
which
cloud
eight
at
the
point
model.
Did
you
use
I
think
this
is
Howell
three
had
the
same
questions
related
to
it
right
then.
That's.
F
H
A
F
A
Because
I
believe
I
just
scroll
to
make
sure
not
lying
that
once
we
get
those
at
it's
done.
Those
are
that's
it
for
comments
and
changes.
The
dock
itself
and
what
I'd
like
to
do
is
if
you
could
get
those
edits
done
by
like
sometime
tomorrow
dan
then
we
could
give
people
until
next
week
to
do
a
final
review
and
claim
success,
and
then
we
can
present
this
to
TLC
on
the
seventh
I.
Believe,
that's
not
fair
to
you.
Dan
works.
A
F
A
M
A
M
You
know
people
can
react
just
saying
you
know,
we've
talked
to
a
bunch
of
guys
and
they
all
you
know,
feel
cozy.
Does
the
fact
that
we
need
a
standard
doesn't
buy
us
anything
and
also
appreciate
you
know
we
raise
questions
in
the
slack
and
people
sort
of
address
them
and
are
not
very
ignore
the
technical,
ass,
okay.
A
So
let
me
ask
this
question
because
I:
don't
I
don't
want
to
put
people
on
the
spot,
but
given
that
we
do
have
basically
half
an
hour
left
the
people
who
put
forward
a
proposal
for
the
eventing
stuff
are
they
is
anybody
on
the
call
card,
the
bold
discussing
and
presenting
their
stuff
now
or,
if
not,
that's,
fine,
we
can
play
into
a
future
call,
but
if
somebody
feels
like
they
can
do
it
now,
you
know
we
have
a
half
an
hour.
Well.
M
M
The
you
know
encompasses
multiple
use
cases,
that's
on
one
end.
On
the
other
end,
we
need
to
encompass
existing
solutions,
are
already
in
the
market
and
find
a
way
to
reconcile
existing
solution.
You
know
just
like
Chris
said
last
week
is
yeah.
You
know
it's
a
nice
proposal,
but
I'm
definitely
not
gonna
change.
How
Amazon
is
doing
lambda
just
because
there
is
a
proposal,
but
we
can
come
with
some
proposals
that
may
may
be
able
to
encompass
with
the
existing
situation
and
extend
it
from
there,
which
is
what
I
tried
to
do
in
my
proposal.
M
They
also
presented
four
different
use
cases
that
I
think
are
very
valid
and
are
covered
in
the
service
white
paper.
You
know
like
IOT
and
machine
learning,
it's
there
and
an
imprinting
of
pictures
and
in
others
that
I
think
we
need
to
see
that
the
proposals
do
enable
us
to
work
around
those
use
cases
and
currently
this
the
proposal
doesn't.
D
So
I
just
understand
if
you
seriously
take
up
this
topic,
what
you'll
encounter,
because
I
did
this
as
part
of
a
a
cloud
and
ISO
standard
for
eventing
and
I've
been
format,
and
so
I
can
talk
about
why
it's
important
to
get
your
terms,
terminology
straight.
To
gets
this
in
the
semantics
aligned
and
to
have
a
model
you
talk
against
and
if
I
can
help
illuminate
what
a
journey
we've
been
through
and
catyph.
Maybe
that
will
open
people's
eyes
as
to
how
expansive
this
topic
really
is
and
what
you're
getting
into
you.
Okay,.
A
D
A
N
About
that,
my
headphones
are:
are
acting
it
higher
on
hey
Matt
you're
on
I.
Just
think
you
actually
just
a
few
minutes
ago
to
line
up
a
chat
with
you.
I
agree
these
these
specs
can
be
reconciled
and
I
want
to
work
with
you
on
that
I
think
it's
probably
easier
if
we
just
have
a
private
coverage
or
we
kind
of
go
through
the
details
and
Matt.
We
chatted
briefly.
I
continue
to
want
to
work
with
you,
of
course,
just
to
be
clear.
N
The
way
that
we're
approaching
it
with
our
spec
is
that
just
like
Matt
said
this
is
a
very
expansive
topic
and
it's
very
clear
that
when
you
get
a
few
engineers
sitting
around
the
table
to
talk
about
this,
that
they
could
talk
about
it
for
days,
probably
even
years,
our
focus
from
the
beginning
has
just
been
keeping
the
scope
very,
very
small
and
focusing
on
gathering
industry
support.
We
don't
think
this
is
going
to
go
very
far
without
industry
support
the
specification
that
we
put
out.
N
There
is
not
final
by
any
means,
it
was
just
a
a
simple
thing
to
put
out
there
to
start
the
conversation.
So
we're
open
to
lots
of
changes
and
stuff,
but
at
this
time
we
are
prioritizing
just
kind
of
working
with
some
of
the
leaders
in
this
space
and
really
just
trying
to
listen
to
them
and
figure
out
what
their
priorities
are,
what
they
need
to
have
in
there
in
order
to
actually
implement
this
so
open
to
chatting
with
you,
Iran
and
Matt,
of
course,
to
figure
out
how
to
make
a
better
specification
here.
N
A
So
let
me
jump
in
just
for
a
sec
because,
as
Ken
is
reminding
us
in
the
chat,
a
venting,
it's
not
technically
part
of
our
charter.
Yet
so,
if
we
have
a
discussion
and
it's
working
at
all,
it
simply
adds
sort
of
an
additive
thing
that
we
just
happen
to
be
a
bunch
of
people
interested
in.
We
don't
have
a
direction
from
the
TOC
yet
to
talk
about
hey
this
stuff.
Okay,
this
is
just
a
bunch
of
guys
or
gals
who
wanted
to
talk
about
stuff
so
back
to
a
little
bit
of
a
process.
A
Question
because
Chris
made
the
comment
in
the
chat
that
I
want
to
get
back
to,
and
he
was
wondering
whether
there
are
other
parts
of
the
spec
I'm.
Sorry,
not
the
spec
of
the
white
paper
that
we
want
to
talk
about
on
this
call,
rather
than
shifting
to
the
eventing
stuff.
So
let
me
open
it
up
again.
Are
there
other
parts
of
the
white
paper
that
people
want
to
discuss
right
now
on
this
call.
J
J
J
Yeah
I'm
trying
to
read
through
and
see
if
there's
any
other
sections
that
like
I,
know,
I'm
not
familiar
with
there's
a
couple
that
I
know.
I
haven't
read
as
in
as
much
death
and
I'm
I'm
willing
to
bet
that
there's
a
number
of
other
people
that
haven't
spent
as
much
time
further
in
the
document
as
well.
I
totally.
A
Agree
and
that's
why
we're
not
shutting
it
down
today,
but
I
also
don't
want
to
have
people
read
it
on
the
call
itself,
sure
alright,
so
last
chance
are
there
other
specific
areas
of
the
document
that
people
would
like
to
talk
about
today,
because
otherwise
you
know
you
guys
are
free
to
go
if
you
want,
but
we're
gonna
allow
people
the
you
know
the
remaining
twenty
five
minutes
to
start
talking
about
the
event
thing
stuff,
because
that
seems
like
it's
an
interesting
topic.
Honking.
F
F
A
C
C
Think
when
I
heard
that
you
know
and
Dania
is
going
to
I
mean
to
Ray
Reconciliation
all
of
this.
So
when
will
that
be
done?
So
we
can
have
a
final
review
because.
A
A
B
F
A
Correct
and
let
me
just
double
check
my
calendar,
so
let's
see
October's
nights
a
lot
I
keep
saying:
October
November
7th
is
a
week
from
Tuesday,
so
we
will
have
at
least
one
more
meeting
before
then
and
Ken
said
he's
gonna
start
working
on
the
slides
this
evening,
so
hopefully
by
next
Thursday's,
call
we'll
be
able
to
look
at
the
slides
as
well
as
finish
up
any
last
comments
on
the
document.
Okay,.
M
B
A
A
M
There
could
be
two
approaches.
You
know,
one
that
I
see
Austin
is
taking
is
saying
you
know
what
I'm
coming
with
a
proposal:
I'm
preparing
a
coalition
of
all
the
cloud
providers
and
I'm
going
to
discuss
with
them
the
proposal
and
then
we're
gonna.
Have
the
CNC,
F
larger
group
go
and
say
whatever
they
have
to
say,
I
think
the
right
approach,
which
then
they
serve,
pushes
a
certain
opinionated
approach
and
doesn't
let
the
cloud
of
those
cloud
providers
hear
all
their
aspects
and
maybe
other
people
have
to
to
share
with
them.
M
I
think
the
right
approach
is,
let's
bring
people
to
the
table
and
let's
discuss
the
proposals,
the
advantages
and
disadvantages
of
each
one
of
them
and
I
assume
we
can
reconcile
those.
You
know
I've
been
to
many
working
groups
in
different
areas.
That's
usually
how
things
are
done.
It's
not
like,
let's
corporate
coalition,
let's
put
people
around
the
coalition
and
we'll
bring
it
as
a
finished
goods.
N
Your
Honor
I
agree
I,
ready
to
add
that
conversation
with
you
so
I
think
we
should
line
up
a
separate
call,
I
think
that
there's
room
for
reconciliation
and
probably
a
way
that
we
could
kind
of
prioritize
out
all
these
features
so
that
when
we
bring
it
to
two
large
vendors,
we
already
have
a
lot
of
investment
in
this
space.
They
aren't
kind
of
overwhelmed
by
by
the
complexity
of
the
proposals.
M
Sure
I'm
100%
with
you
with
putting
things,
stay
in
stages
and
simplifying
them
and
again
I'm
sure
we
can
either
get
agreement
and
and
we'll
do
the
okay,
but
I
wouldn't
position
it
like
one
is
complicated.
The
other
one
is
not
complicated.
Okay,
I
think
there
are
even
simpler
ways
to
coexist
with
existing
events
and
then
sort
of
pushing
everything
to
a
schema
with
the
JSON.
But
you
know:
let's
take
it
offline
and
see
progress
on
that,
but
there's
a
general
comment:
I
think
this
approach
of
aligning
a
coalition
and
throwing
a
spec.
N
A
So,
in
terms
of
next
steps
in
terms
of
a
conversation
who
would
like
to
set
up
that
that
follow-on
call
or
would
you
like,
we'd
like
me-
to
set
it
up
as
a
sort
of
a
as
a
tangential
thing
to
this
working
group,
not
officially
a
section
about
COC?
Yet,
obviously
how
do
you?
How
do
you
guys
want
to
proceed
on
that?
Because
I
have
a
feeling?
A
M
D
A
M
N
D
A
C
A
suggestion
so
we'll
bring
back
to
the
group.
Could
you
first
talk
about
the
scope
of
the
event
like
you
know,
because
you
could
the
event
could
be
right:
they
even
make
a
data
format
or
the
event
to
the
somebody's
platform
or
the
event
format
to
the
backend
and
service
right
and
there
could
be
different
event.
Sources.
D
D
That
I
mean
that
speaks
to
my
point
of
you
have
to
have
your
semantics
and
your
model
aligned.
So
you
know
we're
talking
about
there's
different
types
of
events,
us
and
there's
dip
this
concept
of
originating
event
or
metric
event
versus
the
actual
that
internal
event
as
yarn
calls
it.
So
you
have
to
get
you
have
to
get
your
terms
alone.
You
have
to
scope
out
what
you're
intending
to
cover
yeah.
C
M
Those
catalyst,
so
you
know,
if
you
have
an
opinion
here,
just
put
it
in
writing
and
submit
it
as
well.
Again,
there
is
no,
you
know,
there's
nothing
that
precludes
anyone
from
this
group
to
just
go
and
add:
PR
is
an
existing
stuff
or
write
their
own
papers.
I
think
this
is
the
only
way
to
converge
is
that
everyone
presents
or
a
starting
point,
and
then
we
sort
of
collapse
it
the
one
one
joint
offering
all.
A
Right,
okay,
so
I
think
we
have
a
next
step
in
terms
of
the
through
difficult
puzzles
are
put
forward.
You
guys
are
going
to
set
up
meeting
and
talk
and
then
bring
back
the
conclusion
or
something
to
this
group
for
further
discussion.
Is
it
I
said
so
before
I
assume
an
answer?
Is
it
worth
presenting
anything
on
this
call
now
I'm,
leaning
towards
saying
no,
and
let
you
guys
have
your
discussion
first-
is
that
fair
I.
D
Mean
the
only
thing
I
present
is
the
concept
of
a
model
and
what
you,
but
what
you
need
to
consider
in
crayon
model.
So
when
people
submit
I
mean
if
anyone's
gonna
submit
suggestions,
you
need
to
actually
it
need
to
have
a
model,
and
you
need
to
understand
what
what
the
code
with
a
sub
data
you'll
need
to
eventually
captured
regardless
of
use
case.
You
burn.
D
D
Intended
to
do
did
not
have
time
so
I've
been
traveling
is
on
my
proposal
and
I'll.
Add
an
additional
set
of
pictures.
Actually,
two
slides
will
explain
a
model
for
filling
out
events
as
well
as
what
questions
need
to
be
answered
when
you
thought
events
so
I'll.
Add
those
two
diagrams,
those
slides
to
my
proposal
for
people
to
look
at
okay.
M
I
think
man
also
there
is
a
server,
broader
scope
of
events.
I
think
we
had
the
chat
on
that
already,
but
what
you
refer
to
event
is
not
a
hundred
percent.
What
I
referred
to
event
because
I
generalized
it
a
bit
more
to
you
know
like
an
HTTP
generic
HTTP
request
from
my
perspective
is
also
serve
an
event
and
it
won't
fall
into
the
model
that
us
represents
as
event.
So
maybe
we
need
you
know.
Maybe
the
one
is
a
message
and
the
other
one
is
a.
D
A
J
A
comment,
a
chat
just
just
about
the
concerns
about
that.
We
talked
about
back
and
as
a
service,
we
defined
back
into
the
service
a
little
bit
later
on,
but
then
we
don't
really
go
too
much
deeply
into
it.
I
don't
know
if
that
matters
to
anyone.
I
know.
At
one
point
there
was
someone
from
auth0
that
used
to
join
us
and
I.
Think
at
times
there's
been
people
kind
of
like
from
the
firebase
team.
J
M
J
J
Sorry
yeah
I
mean
throughout
the
doc.
We
refer
to
back
into
the
service
in
a
couple
places.
Don't
we
I
know
we
do
at
least
a
couple
of
times?
Yes,
so
I'm
just
I'm,
just
saying
it
like
we
refer
to
it
on,
of
course,
I
can't
see
pages,
so
I
can't
tell
exactly
where
it
is
and
what
a
service
computing
we
refer
to
it.
It
shows
up
in
the
definitions.
J
It
shows
up
in
one
of
the
examples
here
with
firebase
in
mobile
backends
I'm,
just
not
sure
if
it's
got
the
same
type
of
space
and
coverage
as
some
of
the
rest
of
things
here,
especially
like
in
the
comparison
section
like
we're.
Basically
just
comparing
fast
but
again
we're
using
the
umbrella
term
serve
lists.
A
Well,
III
didn't
see
honest
at
this
point
in
time,
I'm
inclined
to
say
the
way
I
was
in.
There
is
probably
sufficient,
since
no
one
fired
up
as
a
concern
yep
up
till
now
and
they're
trying
to
shut
this
thing
down
yeah.
But
obviously,
if
that
people
reviewing
the
doc,
if
you
know
pon
thinking
about
it,
they
think
there
is
a
glaring
hole.
That's
gonna
make
us
look
kind
of
stupid
by
not
saying
more
about
it.
You
know,
propose
some
text
emotion,
so
people
can
look
it
over.
J
A
I
know
that
Mark
and
Mark
and
I
put
forward
a
proposal
to
basically
I
think
it's
fair
described
it
sort
of
like
a
summary
of
what
we're
doing
here
in
the
working
group
and
that
one
was
accepted.
I
can't
remember
for
sure.
If
there's
a
service
track
or
not,
does
anybody
remember
there
is
a
serverless.
C
M
A
Stay
what
I'll
take
the
action
I'd
receive?
We
can
find
a
room
to
have
some
discussions.
Obviously
this
the
list
of
topics
will
vary
depending
on
the
status
of
things
right,
because,
if
we're
all
completely
shut
down,
obviously
doesn't
want
to
do
anything
about
of
anything.
There's
may
not
be
anything
to
discuss,
but
if
the
TOC
says
yeah
sure
so
that
make
sounds
like
an
interest,
I
think
often
discuss
it.
Then
yeah
using
the
face-to-face
time
be
very
good
to
help
jump-start
the
eventing
discussion.
But
let's
see
how
that
plays
out
figure.
A
A
At
once,
they
right
so
I
think
Tuesday,
there's
like
either
they're
a
couple
of
different
all-day
sessions
that
I
went
down.
Aware
of
so
I'd
be
a
little
bit
nervous
about
trying
to
schedule
something
for
Tuesday.
What
do
people
think
about
scheduling
something
for
either
Wednesday
or
Thursday
in
the
afternoon
or
potentially
evening?
Yeah.
A
A
N
A
N
We've
always
chatted
about
it
as
a
working
session.
However,
I
think
that
cloud
native
con
is
a
good
target.
I
think
we
should
definitely
try
and
work
together
to
have
something
that's
kind
of
somewhat
solid
before
that
conference,
so
that
when
we
get
together,
we
have
something
to
chat
about
in
to
iterate.
On.
M
C
A
We
only
have
five
minutes
anyway,
going
once
going
twice
all
right,
I
believe
we're
done.
Thank
you
guys
very
much
I'm
pleased
after
Dan
sent
out
those
notes,
hopefully
by
tomorrow,
saying
they's
done
with
the
reconciliation.
Please
find
some
time
to
review
over
the
next
week,
because,
hopefully
on
next
Thursday's
call
we'll
try
to
shut
up
so.