►
From YouTube: wasmCloud Community Meeting - 25 Jan 2023
Description
Welcome to the wasmCloud community! Tune in live where we discuss the latest developments in the wasmCloud ecosystem, WebAssembly standards, and break out sweet demos.
Agendas for wasmCloud community meetings can be found at: https://wasmcloud.com/community
A
Should
be
should
be
going
but
I'm
going
to
let
Zoom
finish
its
little
little
progress
bar
first
there
we
go
all
right,
hello,
everyone,
Welcome
to
wasm,
Cloud
Wednesday
for
January
25th
2023..
A
We
have
a
pretty
exciting
agenda
today,
no
bias,
but
I
am
going
to
be
giving
a
demo
right
to
start,
but
let
me
go
ahead
and
share
the
agenda
so
that
we
can
get
a
good
idea
of
what
we're
going
to
be
talking
about
today.
So
this
is
on
our
watsoncloud.com
Dev
repository.
This
is
kind
of
like
a
deploy
preview,
but
what
we've
got
first
is
a
demo
about
a
it's
a
distributed.
A
Wasmcloud
application
I,
finally
got
around
to
setting
up
my
Raspberry
Pi
again.
So,
of
course,
I
had
to
put
wasmcloud
on
it
and
I
set
up
a
little
instance
in
the
cloud
for
some
for
just
a
little
bit
a
little
bit
more
fun.
We're
going
to
be
talking
about
a
markdown
file
that
we
recently
published,
which
is
a
short,
medium
and
long
term.
A
Anybody
who
missed
the
missed.
The
thread
maybe
be
able
to
like
put
some
more
input
in
there.
So
without
further
Ado
I'm
going
to
go
ahead
and
get
started
on
the
demo.
So
you
all
have
seen
you
know.
Various
applications
that
we
have
working
with
was
on
cloud
running
with
web
that
are
running
with
logic
and
webassembly,
and
a
couple
of
capabilities
to
fulfill
the
non-functional
requirements.
A
Now,
I
really
wanted
to
show
because
I
just
set
up
my
Raspberry
Pi
I
one
wanted
to
run
wasmcloud
on
it
and
two
wanted
to
show
a
pretty
sweet
distributed
application
that's
running
across
you
know
when
we,
when
we
say
things
and
we
say:
hey
wasmcloud
Works
across
clouds,
different
edges,
you
know
works
on
your
local
computer.
I
really
want
to
show
that
working
and
what
that
really
looks
like.
So
what
I
have
here?
I
set
up
a
few
wasm
Cloud
hosts
ahead
of
time.
A
A
I
have
a
waslam
cloud
host
here
on
the
right.
You
can
see
this
like
my
local
network,
so
this
one
is
running
on
my
M1
Mac,
that's
right
here
and
then
on
the
left.
This
is
actually
my
Raspberry
Pi.
So
recently
we
kind
of
updated
some
of
our
aptitude
repositories
and
so
all
I
had
to
do
after
Imaging.
This
Raspberry
Pi
was
curled.
The
the
like
apt,
install
script
and
then
run
app,
install
wash,
which
is
pretty
sweet.
A
I
was
able
to
run
wash
up
for
with
with
no
issues
and
that
actually
used
to
be
kind
of
a
point
of
friction
in
the
Raspberry
Pi
area,
because
we
didn't
always
publish
a
target
for
arm
Linux,
and
you
know
you
had
to
kind
of
build
wash
from
scratch
and
that
took
a
while,
so
super
excited
to
get
that
working
and
what
I
want
to
show.
You
was,
first
of
all,
if
we
go
to
the
wasm
cloud
dashboard
here.
This
is
on
my
Local
Host
I've
got
one
like
I
said
arm
Mac.
A
This
one
is
connected.
It's
on
gcp,
it's
actually
on
an
x86
Linux
box,
which
is
cool
and
the
one
on
my
Raspberry
Pi
and
when
I
say
that
all
these
are
connected.
All
I
mean
is
that
I'm
running
a
Nat
server
on
the
gcp
instance
and
then
both
of
the
WASP
Autos,
that
I'm
running
on
my
computer
are
connecting
to
the
NAT
server
that's
running
in
gcp.
So
all
of
these
being
connected
just
means
that
they
all
share
kind
of
a
Nats
connection,
there's
various
strategies
for
that
and
everything.
A
But
that
is
the
wasum
cloud
lattice
all
the
things
that
run
in
this
lattice,
whether
it's
a
piece
of
logic
or
a
capability,
can
communicate
seamlessly.
Just
by
virtue
of
being
on
the
same
network.
So
I've
got
some
other
exciting
stuff
planned.
You
may
have
seen
if
you've
worked
in
the
Watson,
Cloud
Community
or
seen
one
of
our
examples.
We
have
an
example
called
the
KV
counter
and
the
KV
counter.
Hopefully
the
readme
does
Justice
is
just
a
webassembly
module
that
can
accept
HTTP
requests
and
then
an
increments,
a
counter
and
a
key
value
store.
A
And
then
it
returns
that
new,
that
new
value
and
so
there's
a
few
components
that
go
into
this
application
and
instead
of
kind
of
going
through
the
process
of
launching
them
individually
on
the
on
the
walls
and
Cloud
dashboard
I
prepared
a
wadham,
manifest
and
wadham
is.
Let
me
go
ahead
and
get
the
repository
for
you,
the
wazen
cloud
application
deployment
manager.
This
is
kind
of
what
we
use
for
declarative
applications
in
wasm
Cloud.
A
It's
got
a
very
fun
name
and
very
easily
easy
to
exclaim
and
the
way
that
it
works
is
really
similar
to
the
way
that
a
lot
of
people
deploy
declarative,
workflows.
Now
in
any
Cloud
native
application,
which
is
you
can
write
up
a
yaml
file
or
you
can
write
Json
if
you,
if
you
really
are
vehemently
against
DML-
and
you
say
these
are
the
components
for
my
application.
Please
make
them
run
on.
You
know,
according
to
a
specification,
this
many
running
on
certain
hosts
and
just
make
make
it
happen.
A
I,
don't
care
how
it
happens,
which
we
and
I
generally
really
like.
So,
let's
talk
through
this
manifest
a
little
bit.
I
have
the
wasm
cloud.
Kv
counter
application
or
actor,
and
it
has
two
link
definitions.
These
are
basically
saying
at
runtime:
I
want
to
link
to
an
HTTP
server
and
the
redis
capability
provider
and
you'll
notice
that
for
redis
I'm,
connecting
to
like
a
local
instance
and
for
HTTP
server,
I'm
listening
on
a
local
address,
and
this
is
going
to
be
important.
A
So
I
I
have
a
host
running
in
gcp
in
on
the
Raspberry
Pi
and
on
my
local
MacBook,
since
I'm,
showing
this
demo
from
my
local
Mac
I
want
to
make
sure
that
I
run
the
HTTP
server
on
my
Mac,
so
that
I
can
hit
the
local
endpoint
I'm,
going
to
run
the
actual
actor
on
the
Raspberry
Pi
and
then
I'm
going
to
run
the
redis
server
in
the
cloud,
because
that's
where
I'm
hosting
my
database
and
all
of
this
I'm
kind
of
sketching
together
for
a
demo
application,
but
really
what
this
is.
What
I'm?
A
So
here's
what
I
mean
by
saying
that
it
has
a
specific
specific
scaled
property,
so
I
want
to
make
sure
to
run.
One
I
want
to
run
one
key
value.
Counteractor
and
I
called
the
spread
on
Pi,
but
the
requirements
is
that
it
has
to
run
on
a
host
that
has
a
label
raspberry
colon
Pi,
very
fun.
A
If
you
look
at
my
actual
awesome,
Cloud
dashboard
on
the
Raspberry
Pi
host
I
have
that
label
Raspberry
Pi
and
then
here
I'm,
saying
that
my
wasmcloud
HTTP
server
has
to
run
on
a
host
that
has
the
laptop
Mac
label,
which
I
have
on
my
local
wasm
Cloud
host,
and
then
my
redesque
capability
has
to
run
on
a
host
that
has
the
cloud
gcp
label.
So
you
you
may
be
familiar
with
this
from
the
kubernetes
angle
of
like
taints
and
tolerations
being
able
to
say
hey.
A
Of
course
we
marked
this
as
experimental,
because
we're
constantly
iterating
on
this,
but
this
lets
you
take
that
manifest
and
apply
it
to
a
lattice,
so
this
essentially
scans
all
the
hosts
and
then
puts
things
where
they
need
to
be
so
we
can
create
this
application
by
saying
wash
app
put,
and
this
takes
the
yaml
Manifest
it
uploads
it
to
Autumn
and
then
puts
it
in
its
list.
So
there's
a
few
different
deploy,
States
the
default
as
soon
as
you
create
it
is
undeployed.
A
So
what
we
have
to
do
is
deploy
the
application
and
wash
app
deploy
works
with
the
name
and
the
version.
The
versions
are
important
because
you
can
upgrade
you
can
change
your
specification
over
time
and
deploy
newer
new
inversions.
So
this
is
all
going
to
happen.
Pretty
fast
Autumn
is,
is
pretty
dang,
quick
at
figuring
out
where
things
need
to
go
and
sending
out
the
appropriate
control
interface
commands.
A
So
as
soon
as
we
do
that
we
can
see
if
we
come
back
to
the
Watson,
Cloud
dashboard
kind
of
starts
gathering
information
figuring
out
what
it
needs
to
do,
and
then
there
call
it
done
so
we
create
the
link
definitions
to
say,
listen
on
port
8080
and
access
the
key
value
store
in
gcp
we
actually
took.
If
we
take
a
look
in
the
figure
out
my
terminal
stuff,
take
a
look
in
the
logs.
A
So
now
we
can
actually
go
to
localhost
8080
and
hit
refresh
and
get
our
really
fun
KV
counter
application.
So
every
time
I
hit
increment
here,
it's
sending
an
HTTP
request
to
my
local
capability,
which
then
forwards
the
request
to
the
actor
running
on
the
Raspberry
Pi
and
then
accesses
the
key
value
store.
That's
in
gcp,
which
I
don't
actually
think
I
have.
A
Oh
I
do
if
we
take
a
look
at
the
Reddit,
CLI
and
gcp.
This
counter
default
is
essentially
what
we're
incrementing
every
time,
which
is
70.
just
to
show
you.
This
is
for
real
for
real
connecting
to
a
database
on
the
other
end,
and
if
you
hit
refresh
it
gives
you
a
nice
kind
of
Sparkle
of
of
fireworks.
A
So
that
is
a
couple
of
things.
I
I
really
wanted
to
bring
wadham
back
into
the
spotlight,
because
declarative
applications
are
really
important.
As
soon
as
you
get
past
that
initial
experimentation
phase
and
the
ability
to
come
in
and
take
a
look
at
this
manifest
and
say
you
know,
maybe
this
isn't
living
up
to
the
scale
that
we
want
it
to
and
update
the
update,
the
individual
actor
to
run
25
copies
the
ability
to
Simply
change
it
there.
A
Instead
of
changing
your
whole
Ci
or
anything
like
that,
it's
really
important
and
the
fact
that
when
I
was
writing
any
of
these
applications,
this
is
actually
the
demo
app
right
off
of
the
GitHub
repo.
Of
course,
none
of
it
details
exactly
where
a
database
is
going
to
be
living
or
whether
or
not
this
can
run
on
a
Raspberry,
Pi
or
or
in
the
cloud,
and
really
wanted
to
Showcase
a
small
example
of
a
distributed
application
with
wasm
Cloud,
because
this
is
one
of
the
coolest
things
that
we
do
is
make
that
work.
A
So
yeah
happy
to
I.
Think
that's
about
the
end
of
my
demo.
I
have
to
answer
any
questions
or
talk
about
you
know
kind
of
the
setup
of
what
I
had
to
do
for
the
Raspberry
Pi
I'm
actually
happy
to
just
like
recreate
it
right
now.
If
you
want
me
to
because
it's
it's
pretty
quick
but
yeah,
I'll
stop
sharing
for
for
a
moment
and
just
kind
of
answer.
Any
questions,
I
didn't
look
at
the
chat.
If
anything
came
up
there.
B
C
One
of
the
things
that
I
think
would
be
a
pretty
interesting
sort
of
extension
to
this
demo
is:
let's
say
you
have
like
a
fleet
of
raspberry
pies
I'm,
assuming
that,
if
they
all
had
a
similar
tag
and
a
similar
set
of
capabilities-
and
let's
say
I,
have
five
you're
spinning
up
25,
replicas
and
I.
Just
then
I
have
my
thing:
running
I'm,
clicking
increment,
increment,
increment
and
I.
C
You
you
go
and
off
screen
decide
to
either
kill
the
hosts
on
four
of
them
or
unplug
for
your
Raspberry
Pi's
it'd
be
cool
to
show
and
I'm,
assuming
that
it
just
sort
of
will
say:
okay
I
need
25
replicas
to
form
for
my
house
or
down.
Let
me
spin
up
20
more
on
my
Raspberry
Pi
and
continue
going
from
there.
A
Yeah,
so
the
the
whole
bottom
reconciliation
Loop.
Basically
anytime,
you
make
a
meaningful
change
to
the
Watson
Cloud
lattice,
meaning
that,
like
either
requirements
aren't
met
anymore
or
you
don't
have
enough
to
fulfill
their
requirements
like.
If
you
had
two
Raspberry
Pi's
with
that
same
label
and
said:
hey,
run,
20
copies
of
that
actor,
and
so
it
you
know,
spreads
it.
It
does
10
across
each
for
example,
then,
as
soon
as
you
kill
that
one,
if
you
killed
that
one
host
or
like
unplug
unplug,
the
Raspberry
Pi
wadham
immediately
gets
that
host
stopped
event.
A
D
Yeah,
the
the
general
plan
or
the
general
design
is
that
if
the,
if
the
host
you
know
for,
for
whatever
horrible
reason
banishes
in
the
middle
of
an
active
processing,
its
request,
yeah
that
that
request
is
basically
gone.
D
You
know
if
the
request
was
you
know
something
like
it
was,
maybe
handling
a
message
from
a
durable
stream.
Then
you
know
your.
D
The
actor
would
then
fail
to
acknowledge
that
inbound
message
and
then
so,
when
the
actor
restarts
in
a
new
location,
it'll
just
pick
up
the
message
that
it
had
failed
the
process
before
so
I
guess
in
put
another
way,
wasmcloud
itself
isn't
going
to
be
able
to
pick
up
an
actor's
execution
in
the
middle
if
it
failed
on
a
host
that
dies
on
gracefully,
but
there
are
architectures
you
can
use
within
wasmcloud
that
will
make
it
so
that
that
actor
can
recover
the
unprocessed
work.
A
Yes,
Stephen
I
think
you
were
next
and
I.
Think
Oren
had
a
great
question
in
chat
too
cool.
B
A
Yeah
here
let
me
Let
Me
Go
and
show
you
so
I
and
let
me
see
this
is
what
I'm
running
right
here.
A
Nice,
so
I'm
gonna
go
ahead
and
just
stop
this
watching
flat
host.
So
essentially,
all
that
I
did
was
app
install
wash,
which
you
can
do
after
you
run
the
you
can
do
after
you
run
the
little
command
to
curl
down.
Where
is
it?
You
can
curl
down
this
Debian
install
script,
which,
just
like
adds
our
repository
to
your
to
your
little
list
and
app
and
then
the
actual
command
that
I
ran.
Was
this
one,
so
I
added
a
label
using
the
host
underscore
syntax?
This
just
takes
anything
after
that.
A
With
the
you
know,
key
and
value
raspberry
equals
pi
and
gives
it
a
label
raspberry
equals.
Pi
I
ran
wash
up
and
I
added
the
flag
Nats
host,
which
that's
the
IP
address
of
my
gcp
instance,
which
is
where
the
actual
map
server
is
running.
A
I
could
always
run
this
with
a
leaf
node
and
and
so
that
it
kind
of
extends
the
network,
but
just
for
this
demo,
I
figured
this
was
the
easiest
path
and
then
I
also
added
Nats
connect
only
to
make
sure
that,
if
I
typed,
my
IP
address
wrong
like
I,
couldn't
connect
to
Nets
that
it
wouldn't
start
up
a
Nat
server
on
its
own
and-
and
this
kind
of
just
ensures
that
we
that
I'm
not
accidentally
going
to
start
a
new
lattice
I,
would
only
start
a
waslam
cloud
host
connected
to
my
existing
lattice.
A
A
So
if
I
look
at
that
host,
you
can
see
that
I
have
Raspberry,
Pi
and
Community
call
fun,
so
the
the
labels
are
completely
arbitrary.
You
can
make
up
your
own.
The
only
ones
that
you
can't
change
are
these
host
core
ones,
which
are
core
they're
like
they're,
not
overridable,
because
they
describe
the
type
of
machine
that
it's
running
on,
which
can
be
really
useful
and
I'm,
not
really
sure.
If
there's
like
a
risk
to
people
overriding
these,
but
it's
just
not
something
that
it's
not
something
that
you
should.
A
A
Thank
you,
and
before
this
on
the
Raspberry
Pi
I
mean
if
you're
in
the
community
call
you
sell
my
or
if
you're,
in
the
community
slack,
you
saw
my
meme
this
morning.
A
I
installed
a
firewall
and
I
blocked
myself
out
of
SSH
connection
to
the
Raspberry
Pi
I
was
like
I
could
figure
this
out
or
I
could
just
re-image
the
whole
thing
so
I
just
used
the
Raspberry
Pi
installer
I
put
Debbie
and
Bullseye,
or
the
the
raspbian
that
has
Bullseye
under
it
and
then
started
from
scratch,
and
it's
like
apt,
update,
apt,
upgrade
I,
know
tailscale's
the
right
way
to
do
it,
but
come
on
I'm,
just
hacking
it
together,
I'm.
You
know
that
I
I
re-image
it
this
morning.
A
Yeah,
let
me
let
me
take
Oren's
question
really
quick.
It
does
wadham
specify
how
a
deployment
of
new
code
would
roll
out
and
wadham
doesn't
explicitly
specify
how
but
it's
just
a
complete
rolling
upgrade.
So
if
you
essentially,
when
you
deploy
a
new
version
of
an
application,
we
are
under
the
same
application,
we're
we're
doing
a
Reconciliation
Loop
across
the
lattice
with
the
new
constraint.
So
if
I
change
to
you
know
from
one
instance
to
let's
say,
let's
actually
do
a
a
new
version.
A
So
if
I
change
from
wasmcloud
KV
counter
0.4
to
0.4.1
and
has
the
same
number
of
instances
will
actually
take
those
and
and
stop
the
instances
of
the
old
actor
because
they
don't
belong
anymore
and
then
start
the
new
instances
of
of
the
KV
counteractor.
Essentially,
it's
like
a
diff
between
the
two,
a
different
between
the
two
different
versions
of
the
individual
application.
A
It's
also
worth
mentioning
that
the
it's.
It's
also
worth
mentioning
that
if
you
have
two
different
applications
that
have
similar
resources,
so
if
I
had
KB
counter
one
and
KV
counter
two,
those
two
individual
applications
are
managed
separately.
So
if
I
said
you
know,
the
first
application
requires
five
instances
and
the
second
application
requires
10,
then
we'll
be
running
we'll
be
running
15
because
you
are
doing
it
per
application
and
what
that
really
lets
you
do
is
take
individual
apps
as
they
are.
A
You
can
delete
one
completely
and
not
tamper
with
tamper
with
another
application
and
for
actors,
it's
less
applicable,
but
it's
really
important
for
things
like
HTTP
servers,
where
you
want
to
make
sure
that
you're
running
an
HTTP
servery,
no
matter
what
other
applications
happen
to
be
running
in
the
lattice.
A
All
right.
Actually,
a
great
great
question.
Great
questions
in
the
chat.
Cj
had
a
follow-up
around
failover
around,
like
deleting
instances
out
of
the
you
know
deleting
hosts.
A
What
happens
is
when
you,
this
is
actually
handled,
as
Kevin
said
by
the
Watson
Cloud
lattice
itself.
If
you
have
an
instance
of
one
actor
on
one
host
and
an
instance
of
an
actor
on
another
host,
those
are
kind
of
equally
likely
to
receive
invocations.
Just
because
we're
doing
Nats
round
robin
and
if
you
were
to
delete
one
host
out
of
the
lattice
entirely
that
one
actor
is
no
longer
in
the
basically
an
interest
graph
or
receiving
invocations.
A
It
automatically
fails
over
to
any
available
compute
and
that's
just
kind
of
something
that
you
get
out
of
the
box
with
wasm
Cloud.
You
don't
need
to
Autumn
for
it
and
you
don't
need
to
explicitly
configure
anything
to
do
that.
A
Bailey,
you
posted
a
question
from
YouTube.
That
was
a
comment
in
the
YouTube
Hey
YouTube.
The
question
was:
can
you
split
traffic
between
actors
via
wadham
and
I?
Guess
that
that's
kind
of
a
similar
question,
but
this
is
actually
handled
exactly
by
the
exactly
by
the
lattice
itself,
so
by
virtue
of
these
actors
being
stateless,
reactive
compute
that
are
connected
to
a
lattice
in
wasmcloud
extra
instances
of
an
actor
are
automatically
load
balanced
between
and
there's
some
there's,
some
extra
things
that
can
make
that
more
efficient.
A
That
things
like
Nats
does
does
for
us
like
making
sure
to
keep
traffic
local
instead
of
reaching
out
to
a
central
Nat
server
when
when
possible,
but
that
whole
system
is
what
allows
us
to
do
the
automatic
failover
or
what
allows
us
to
load
balance
between
different
instances
that
all
handle
the
same.
They
kind
of
scale
up
in
a
horizontal
or
vertical
way.
D
Yeah
you
kind
of
went
into
it.
There's
a
question
in
there
about
whether
or
not
wasmcloud
can
prefer
a
I
guess
a
beefier
host
or
to
to
configure
or
fine
tune
the
the
round
robin
stuff.
D
So
the
first
point
is
when,
when
traffic
is
distributed
across
actors,
it's
done
pseudo
randomly
not
necessarily
round
robin,
so
we're
not
for
a
number
of
reasons,
most
of
which
are
related
to
things
like
availability,
I'm,
trying
to
figure
out
how
not
to
go
down
a
rabbit
hole.
But
if
you
round
robin
traffic
in
a
distributed
system
like
that
and
something
in
the
round
robin
pool
is
causing
problems,
then
you'll
just
get
the
same
problem
over
and
over
again,
you
can
kind
of
poison
pill.
D
Your
whole
system,
so
having
the
pseudo-random
choice
is
a
slightly
is
a
safer
way
to
make
sure
that
what
you're
what
you're
doing
is
actually
going
to
get
delivered.
But
then,
if
I
want
to
control
the
shape
of
my
network,
traffic,
I
can
do
that
with
a
Nats,
Leaf
node
and
the
way
the
leaf
nodes
work
is
I
can
put
one
of
those
wherever
I
want
to,
and
it
basically
acts
as
a
network
traffic,
Gateway
and
so
I.
D
Can
everything
on
the
the
quote-unquote
inside
of
a
natch
Leaf
node
is
going
to
prefer
to
stay
inside
so
where
I
put
my
leaf
nodes
and
how
I
connect
them
to
each
other
can
control
exactly
how
the
traffic
preferences
go
within
my
network,
and
so
if
I
have
Leaf
node
set
up
and
I've
got
say
30
actors
on
the
inside
and
on
the
other
side
of
the
leaf
node
there's
you
know
one
or
two
copies
of
that
same
actor
within
the
leaf
node.
D
D
A
Yeah
awesome
questions
it's
so
fun
to
kind
of
dive
into
dive
into
intimate
mechanics
of
the
the
lattice
and
and
how
Autumn
works,
because
it's
just
so
much
taken
away
from
the
developer
needing
to
worry
about
from
the
beginning,
with
the
distributed
application.
It's
very
cool.
A
Well
that
took
up
took
up
a
nice
little
chunk
of
the
of
the
call
really
great
discussion,
and
thank
you
all
for
letting
me
do
a
little
demo
I'm
going
to
vote
that
we
move
on
just
a
little
bit.
I
want
to
make
sure
that
we
get
to
the
rest
of
the
agenda
because
it's
pretty,
we
got
some
important
stuff,
but
we
can
always
come
back
and
chat
more
about
lattice
or
Autumn
things
after
the
fact.
A
You
know
we
call
it
now
later
in
the
ideas,
so
the
kind
of
short-term,
longer
term
and
then
things
that
we've
kind
of
brainstormed
but
haven't
quite
found
the
correct
place
for
yet
let
me
go
ahead
and
share
that
road
map
and
Taylor
I'd
love
to
invite
you
to
come
on
and
chat
through
it,
because
I
think
that
you
essentially
drafted
this
one
up,
maybe
with
with
Bailey
as
well
but
we'd
love
to
have
you
chat
about
it.
E
Yeah,
so
we
just
kind
of
did
an
initial
pass
on
it.
We
wanted
this
to
be
a
community
thing,
so
the
I
want
to
make
sure
okay.
E
The
plan
here
is
to
make
this
kind
of
an
idea
of
where,
where
we're
going
as
a
project,
what
we
expect
people
to
people
to
to
see
coming
from
the
water
we
just
really
want
to
make
sure
people
knew
where
we
were
heading
and
what
we
wanted
to
try
to
do,
and
so
that's
that's
meant
to
be
very
clear
here.
E
E
E
Okay,
people
entertain
for
about
10
seconds,
okay
anyway,
so
in
the
in
the
roadmap
we
have
now
next
and
later
so
now
are
the
things
that
we
kind
of
have
a
a
definite
plan
for
what
we're
trying
to
do,
what
we're
going
to
be
working
forward
with
there
and
those
first
two
kind
of
items
are
working
on
a
SQL
V2
contract
just
so
that
it's
actually
more
production
worthy
is
the
main
concern
there
because
I
know
a
lot
of
people
want
to
use
that,
and
then
we
also
are
going
to
work
on
finishing
up
with
them.
E
So
what
Brooks
demo
today
is
still
very
like
Alpha
level,
and
so
we
need
to
like
basically
validate
some
design
assumptions,
making
sure
we're
supporting
custom
scaling.
Algorithms,
like
deployment
and
infrastructure
around
how
you
run
with
Dam,
guides
documentation
all
that
kind
of
stuff
and
then
like
also
removing
some
under
some
underlying
dependencies
that
we
don't
think
we
need
anymore
just
because
of
how
things
have
evolved
since
we
kind
of
started
the
thing
so.
E
The
only
thing
that
will
be
these
two
things,
the
only
thing
that
we
worked
on,
is
just
giving
you
idea
of
what
the
maintainers
are
doing
and
so
I'm
I'm
sad
Matt
left
before
we
got
to
this,
but
like
Matt
and
other
project
maintainers
we're
also
making
please
come
contribute
feedback
here
we
could
have
missed
things
of
the
various
sub
projects
inside
of
wasm
cloud
that
aren't
necessarily
covered
on
here.
That
should
be
covered
on
here.
E
So
this
is
open
in
a
PR
which
I'll
have
Brooks
link
out
in
the
meeting
chat
and
zoom
and
Bailey
in
the
in
the
streams.
So
you
can
go
comment
on
anything
that
you
have
concerns
about
or
whatnot.
So
the
now
is
what
we're
trying
to
work
on
like
right
now.
Next
is
something
that
we
consider
very
important
and
they're
they're
kind
of
like.
What's
on
Deck.
That's
sorry,
that's
a
very
baseball
term
and
very
U.S
Centric
I
guess,
but
I
play
a
lot
of
baseball
growing
up,
but
the
idea
is
right.
E
Like
it's
the
next
thing
up,
it's
not
actually
being
worked
done
right
now,
but
we
want
to
get
these
done
pretty
soon,
and
this
is
a
great
area
to
look
at
if
you're,
like
I,
really
want
to
contribute
to
wasm
Cloud.
This
is
a
great
thing
to
look
at.
It's
not
like
I,
said
comprehensive
by
any
means,
but
it's
kind
of
some
big
things
that
we
want
to
add
in.
E
So
this
is
stuff
like
adding
things
for
capabilities
for
like
examples,
documentation,
we'd
like
to
have
something
similar
to
some
of
like
what
you
see
in
the
JavaScript
ecosystem,
where
you
have
like
examples,
here's
how
you
add
logging
like
so
it
has
hello
world
with
logging.
Here's
how
you
add
this,
here's,
how
you
add
this,
and
even
that
like,
if
you're
doing
it
for
the
first
time
and
capture
most
of
your
instructions,
that
kind
of
stuff
would
be
very
valuable.
E
We're
gonna
work
on
V2
versions
of
various
contracts.
These
are
all
kind
of
similar
ideas
that
we
want
to
make
them
more
production
rating,
and
we
also
want
to
match
up
with
what
the
rest
of
the
wasm
ecosystem
is
using.
For
these
interfaces
we
are
also
looking
at
how
we
want
to
handle
like
large
file
stuff.
E
So
when
you're
doing
like
image,
processing
or
large
files
across
to
a
blob
store
that
we
can
handle
it
a
little
more
cleanly
and
nicely,
we
want
to
really
polish
up
the
go
SDK
work
on
making
like
link
desk
2.0.
So
just
having
like
better
things
that
can
give
you
hints
of
what
type
of
data
should
be
passed
in
to
configure
your
actor
and
then
we
also
want
to
completely
revamp
wasnbus
RPC
so
that
we
can
just
make
that
developer
experience
a
lot
better.
So
that's
the
next
category,
like
I,
said
those
are
pretty
concrete.
E
E
These
could
just
be
nice
ideas
that
turn
out
to
be
bad
ideas
down
the
line,
and
what
will
happen?
Is
we
take
these
and
we'll
move
them
into
next?
As
we
decide
we're
gonna
as
we
decide,
we
want
to
do
them.
So
this
is
stuff.
Like
we've
talked
about
enabling
JavaScript
actors
that'd
be
a
big
one.
We
really
want
to
do.
We
just
have
inspected
out
and
there's
a
lot
of
those
other
things
that
you
saw
and
then
the
yeah
and
by
the
way
that's
that's
a
good
point.
E
Kevin
I
should
probably
clarify
that
I.
Think
one
of
those
is
like
there's
some
other
things.
We
could
do
around
components
and
the
provider
sdks,
which
is
what
I
think
I
meant
there
and
so
I'll
go
fix
that
in
about
10
seconds,
might
stealth
edit
it
underneath
Brooks
and
so
anyway.
So
what
we're
doing
with
with
this
is
we're
going
to
like
try
to
get
like
unit
testing
things
in
place.
E
We're
gonna
have
the
component
things
which
I
meant
to
combine
those
lines
JavaScript
host,
which
is
something
we've
kind
of
half
had
for
a
while.
We
want
to
actually
make
it
real,
like
an
embedded
first
host
for
small
devices
or
embedded
devices
and
then
also
finishing
up
like
making
awesome
cloudhouse
multi-tenant.
We
kind
of
have
that
thing
there,
but
just
for
people
who
want
to
host
it
and
be
able
to
run
multiple
tenants
on
the
same.
Like
awesome,
Cloud
host,
we
need
to
to
tidy
up
some
of
the
work
there.
E
E
Like
I
said
not
comprehensive,
but
we
wanted
to
present
it,
so
people
in
the
community
could
start
going
in
and
contributing
comments
and
things
to
that
draft
PR
to
say,
like
I,
really
think
this
should
be
on
here,
or
can
you
clarify
this
or
whatever
it
might
be,
and
we're
going
to
try
to
keep
this
up
to
date?
So
this
will
be
kind
of
the
place
you
can
go
to
to
add
any
comments
or
concerns
or
ideas
to
in
in
the
near
future.
A
Hey
Taylor
I
have
a
quick
I,
have
a
quick
question
for
you.
So
you
know
these
things
are.
Are
things
that
we're
talking
about
working
on
from
the
maintainer
perspective?
How
do
how
are
we
planning
on
splitting
these
out
for
kind
of
like
you
know,
we
have
things
that
we
want
to
get
done,
which
are
the
some
of
the
core
product
or
core
project
things
and
there's
also
plenty
of
Open
Source
contributors
who
are
looking
for
kind
of
good,
first
issues
or
a
little
bit
bite-sized
pieces
to
take
and
and
work
on.
A
Are
these
going
to
be
broken
down
into
like
a
couple
of
different
issues?
So
are
we
gonna?
How
do
we
designate?
What's
you
know?
Is
there
something
that
only
a
maintainer
could
work
on
and
like
stuff?
The
community
only
is
work
on
like
how
does
how
do
are
we
going
to
do
that?
Separation.
E
Most
of
this
stuff
is
open
for
anyone
to
take
the
like,
for
example,
finishing
up
with
Dam,
and
the
sequel.
V2
contract
are
like
very
much
a
single
project
that
one
or
two
people
working
on
that
project
together,
like
it's
still
done
in
the
community.
We
push
all
these
things
up,
but
it's
just
it's.
It's
such
a
like
big
unit
of
work
that,
like
will
be
people
from
the
community
who
have
like
touched
this
before
and
been
doing.
E
It
will
probably
be
the
best
candidates
a
lot
of
the
stuff
in
the
next
category,
because
it's
not
already
specked
out
and
kind
of
have
some
ideas
around
are
perfect.
Things
for
people
to
take
I
like
I,
already
have
some
thoughts,
for
example
around
key
value
contract
V2,
but
someone
can
open
an
issue.
We'd
find
it
and
I'd
comment
on
what
I
have,
and
someone
could
just
take
that
people
want
to
add
examples.
Those
things
are
there,
there's
also
if
you're
just
looking
to
barely
get
started.
E
We
have
good
first
issue
tags
in
most
of
our
repos
and
so
you're
able
to
go
through
and
see
something
small
that
you
could
kind
of
get
your
feet
wet
with
to
do
it.
A
lot
of
these,
like
I,
said,
are
higher
level,
and
so
they
can
appear
daunting
but
like
if
you,
if
it's
something
that
interests
you
like
we're
all
on
board,
having
anyone
from
the
community
help
Implement
these.
E
A
I
think
I'll
add
one
more
sentiment
on
top
of
that
anytime.
That
there's
something
like
this
that
you
feel
like
you
could
contribute
to,
but
you're
not
100
sure,
if
you're
doing
it
the
right
way
or
if
you
know
you're
approaching
it
with
the
same
code
style
that
the
Watson
Cloud
projects
have
used
before
I.
Personally,
love
pull
requests
as
a
collaboration,
space
and
I.
A
Think
you
know,
most
people
on
the
maintainer
side
will
will
agree,
feel
free
to
open
a
pull
request
at
any
time
if
you've
got
a
half
halfway
there,
just
looking
for
a
little
bit
of
initial
feedback.
Pull
requests
are
a
great
way
to
start
looking
at
code
and
to
collaborate
on
code,
because
it's
out
there
in
the
open,
we
can
always
kind
of
go
back
and
forth.
A
These
pull
requests
are
not
meant
to
be
a
space
that
you
have
to
push
perfectly
finish
rubber
stamped
code
in
order
to
in
order
to
contribute
so
just
wanted
to
add
on
a
little
bit
more
there
and
Stephen
great
great
kind
of
follow-on
question
to
that
are
PR
drafts.
Okay,
absolutely
I
mean
you.
Can
you
can
do
a
draft
PR
or
you
can,
which
is
a
great
way
to
kind
of,
say,
hey
I'm
kind
of
working
on
this?
It's
not
quite
ready
for,
like
a
full
review.
A
Just
I
would
just
denote
in
your
PR
description
what
you're
looking
for
feedback
on,
if
you,
if
you
put
draft
since,
depending
on
the
person
depending
on
how
busy
everything
is,
if
we,
if
you
know
seeing
a
draft
PR,
it
may
not
get
fully
reviewed.
If,
if
it's
marked
this
draft,
because
we
may
wait
for
more
commits
to
come
in
something
like
that,.
A
A
Well,
then,
I
think
that
that
I
think
that
that
kind
of
wraps
up
the
roadmap
section
actually
Bailey
is
there
a
is
there.
E
E
If
we
want
to
stream
an
Apache
Arrow
like
flight
encoded
by
a
sequel
actor
like
would
those
encodings
be
spec
by
individual
actors
or
interface,
whatever
the
wasimbus
RPC
will
won't
affect,
that
was
the
most
RPC
is
purely
the
underlying
transport
layer
between
hosts
anything
you
built
like
it's
opaque
to
it,
you
don't
have
to
it,
doesn't
matter
what's
in
it,
so
if
you
have
specific
encoding
things
you
want
to
do
or
or
anything
around
that
that's
purely
around
how
you're
designing
your
application,
which
could
be
a
custom
interface
or
leveraging
one
of
the
current
interfaces.
E
It
would
need
more
details
to
kind
of
like
be
super
specific
about
it,
but
yeah
wasn't.
Bus
RPC
is
just
kind
of
it's
such
a
fundamental
layer
of.
What's
going
on
that,
we
want
to
make
sure
that
this
is
like
it's
easy
to
use.
It's
separated
out
for
actors
and
providers.
It
has
all
of
that
kind
of
stuff
worked
out
to
just
make
the
developer
experience
as
seamless
as
possible,
which
it
sometimes
has
a
lot
of
rough
edges
as
it
stands
now,.
B
A
Right
we're
almost
there
I
think
this
is
probably
a
good
good
time
to
move
on.
Let's
tackle
the
last
agenda
item
and
then
we
can
maybe
get
to
some
some
free
q
a
time
at
the
end
of
the
meeting.
So
we
had
a
really
fun
discussion
in
the
wasm
cloud
slack.
A
You
know
we're
looking
at
some
of
the
things
that
we
have
created:
names
for
and
set
up
documentation
around,
specifically
the
name
actor
for
our
webassembly
modules
that
are
composed
of
business
logic
that
interact
with
abstract
interfaces
instead
of
specific
implementations,
so
things
like
interacting
with
a
key
value
contract
and
then
at
runtime
fulfilling
that
capability
with
a
redis
implementation
or
a
hash
map
in
memory
implementation.
Things
like
that
and
actor
actually
Kevin
I
hate
to
Super.
A
Put
you
on
the
spot
here,
but
I
think
that
you
came
up
with
this
name
for
these
and
and
would
love
it
if
you
wanted
to
add
a
little
bit
of
color
into
like
what
the
name
came
from
what
kind
of
aspects
of
actors
or
maybe
actor,
models,
inspired
the
the
name.
D
D
If
you
look
at
the
definition
for
actor
model,
the
wasmcloud
actor
meets
all
of
those
rules,
except
for
the
fact
that
our
actors
can't
spawn
other
actors
and
that's
more
of
a
practical
concern
and
host
runtime
type
concern.
Then
an
academic
one
but
yeah
the
actor
model
has
been
around
for
a
very
very
long
time
and
when
I
was
trying
to
figure
out
what
it
was.
A
portable
piece
of
independently
Deployable
reactive
business
logic
should
be
called.
Actors
seem
to
fit
pretty
well.
A
Nice
I
just
had
to
had
to
type
out
that
that
acronym
portable
piece
of
yeah,
what
was
it
portable.
A
The
time
well
so
one
of
the
one
of
the
big
Inspirations
for
starting
this
conversation
is
that
when
people
hear
actor
a
lot
of
the
time
they
hear
act
or
model
and
depending
on
what
their
past
programming
experiences
is
like
could
either
be
they've
worked
with
an
actor
model
before
and
it's
been
awesome.
They've
worked
with
an
after
model
before
and
it's
been
not
so
awesome
or
they
haven't
really
heard
of
the
term
before,
and
so
it's
kind
of
a
net
new
thing
for
for
them
to
comprehend.
A
So,
really
just
doing
some
thinking
about
whether
calling
them
actors
is
properly
priming
people
for
what
act,
what
what
our
wasm,
Cloud
webassembly
modules
do
and
exploring?
Maybe
some
alternate
namings
things
things
that
came
up
in
the
thread
were
like
logical
components,
application
components,
functional
components,
Nano
Services!
Thank
you,
Stephen
I'm,
just
reading
off
of
your
bulleted
list
for
things
napkins
space
dust.
A
We
we
got
a
lot
of
things
in
there
that
we
were
thinking
about,
and
I
kind
of,
have
an
opinion
after
reading
through
the
entire
Thread
about
what
kind
of
the
proper
direction
is
in
my
mind,
but
Kevin
I'll.
Let
you
let
you
go
first
and
maybe
Steven
you,
you
too
so
can
I.
D
Yeah
I
think
you
know
it's
when
when
we
build
software-
and
you
know,
naming
is
always
the
hardest
part
right
that
and
rounding,
you
know
off
by
one
errors
and
cash
invalidation,
but
I
think
the
the
thing
that
is
I
guess
my
opinion
on
this
is
that
if
we
try
and
name
the
name,
these
actors,
something
that
potentially
feels
more
approachable,
we
also
run
the
risk
of
diluting
the
meaning
of
what
it
is
that
one
of
these
components
does.
D
So
you
know
if
we
were
to
call
them
Widgets
or
poppables
or
whatever
else
we
could
think
of.
If
we,
if
we
use
a
more
broadly
scoped
term,
then
it
might
not
be
sort
of
immediately
obvious
what
it
is
we're
trying
to
build.
Of
course,
the
other.
The
other
side
of
that
is
for
people
who
aren't
familiar
with
what
the
actor
model
is
or
that
terminology
and
that's
a
learning
curve
for
them.
So
I
think
it's
a
balancing
act
really.
B
Oh
sorry,
I
was
muted,
I
was
just
saying
when
I
came
into
this
I
didn't
really
understand
what
an
actor
was.
I
hadn't
worked
with
an
actor
model
before
and
when.
B
Was
I
really
did
like
the
the
terminology,
but
where
the
friction
for
me
was
was
understanding
how
to
categorize
like
a
grouping
of
like
an
actors
and
providers
together,
so
for
me,
I
do
like
the
term
actor
I
just
also
want
to
understand
if
there's
a
way
to
normalize
a
term
for
a
logical
grouping
of
of
these
components.
A
B
A
B
That
exactly
and
then
you
can
have
like,
like
a
function
as
a
service,
you
can
scale
more
readily
the
the
the
smaller
levels
of
components
based
on
what
your
application
needs.
B
D
Yeah
I,
just
I,
wanted
to
kind
of
address
the
issue
with
link
definition
and
scope
right
now.
Link
definitions
are
scoped
to
an
entire
lattice,
and
so
one
strategy
that
we've
been
recommending
is
that
if
you
want
to
have
multiple,
if
you
want
to
have
two
different
types
of
configuration
for
the
same
set
of
actors,
then
you
would
run
that
in
two
different
lattices.
D
And
if
you
are,
if
you
remember
Taylor's
roadmap,
there
is
some
time
on
that
road
map
to
kind
of
revisit,
link
definitions,
and
as
part
of
that,
we
might
be
reevaluating
the
scope
so
that
a
link
definition
is
more
I,
guess
more
properly
part
of
a
an
application,
rather
than
just
something
that's
Global
to
a
lattice,
but
we'll
see
where
that
goes.
A
A
Justin
is
currently
live
typing
his
thoughts
into
the
zoom
chat,
so
we
will
we'll
read
them
as
they
come
in
and
then
or
read
a
little
bit.
E
E
A
While
you're
getting
your
dots
down,
I
know,
you
probably
have
you
always
have
really
insightful
thoughts
for
these
types
of
conversations.
A
I'm,
gonna,
pivot,
really
quick
I
want
to
talk
about
a
couple
of
really
exciting
things
that
are
going
on
in
the
webassembly
space,
which
you
usually
kind
of
talk
about
General
webassembly
community
at
the
end,
but
I
want
to
specifically
talk
about
events.
There
are
a
lot
of
events
that
are
happening
in
this
year
and
a
few
of
them
in
kind
of
the
the
front
end
of
2023
that
we
as
a
waslam
cloud
team
and
probably
for
cosmonic,
which
some
of
some
of
the
wise
and
Cloud
maintainers
work
for.
A
We
will
be
at
a
lot
of
these
events
and
I
want
to
both
raise
awareness
for
them
and
encourage
everyone
who
is
working
in
the
webassembly
space.
Even
if
you
feel
like
your
your
application,
what
you're
working
on
is
small,
the
more
submissions
that
we
get
for.
These
types
of
events
really
help
grow.
The
web
assembly
community
and
I
promise
you
whatever
you're
working
on
in
the
web
assembly.
A
Space
right
now
is
interesting
enough
for
a
talk,
so
it
would
totally
encourage
you
to
submit
for
these,
so
I
actually
have
four
that
I
want
to
talk
about
the
first
one
is
cloud
native
wasm
day
this
one
is
going
to
be
in
Europe
in
Amsterdam,
so
anybody
who's
in
you
know
across
the
pond
I
know
we
have
a
really
great
Community
contributor
Giuseppe,
who
is
in
Italy
I
hope
that
they
can
make
it
over
to
Amsterdam.
We
can
meet
them
and
anybody
else
who's
over
there.
Submissions
are
still
open.
A
This
actually
just
kind
of
popped
up
the
other
day
for
cloud
native
wasn't
day.
So
please
check
out
the
call
for
proposals
that
is
due
on
the
12th
of
February
at
midnight.
Pacific
time,
there's
a
ton
of
different
topics.
This
is
going
to
be
a
really
exciting
was
in
Day.
Last
year,
wasm
day,
EU
was
a
really
fun
really
fun
conference
to
to
go
to
a
lot
of
us.
A
lot
of
us
actually
got
to
meet
for
the
first
time
there.
A
In
the
same
vein,
there
is
wazamayo
which
I
keep
calling
wazmio,
because
it's
kind
of
fun
to
mash
those
things
up,
but
why
is
myo?
2023
is
going
to
be
in
Barcelona,
that's
actually
at
the
end
of
March,
so
in
Europe,
a
little
bit
before
that
cfps
I
believe
are
no
longer
open
for
this,
but
you
can
still
get
your
ticket
and
we
have
personally
submitted
a
few
talks.
A
So
here's
to
hoping
that
we
can
get
in
and
talk
to
All
You,
Beautiful
People
about
the
fun
stuff
that
we're
working
on
in
terms
of
webassembly
conferences,
there's
also
going
to
be
the
webassembly
summit,
and
this
has
been
one
of
my
favorite
webassembly
conferences
that
have
happened.
It
started
if
I,
if
I
remember
right
in
2019,
so
it
was
right
before
right
before
the
pandemic
and
then
was
online
for
a
little
while,
but
I
believe
that
they're
actually
going
to
be
doing
things
in
person
in
London
and
then
also
live
streaming.
A
You
know
all
of
it.
As
usual,
one
of
my
favorite
webassembly
talks
that
I've
ever
seen
came
out
of
this
conference,
which
was
Peter
I'm,
forgetting
his
last
name,
who
used
webassembly
in
the
browser
to
make
music,
which
was
so
fun.
If
I
can
dig
up
that
link,
then
I'll
put
it
I'll
put
it
somewhere.
So
talk
submissions
are
still
open
for
this
one,
but
they
actually
close
next
Tuesday
so
make
sure
to
get
on
this.
A
If
you
are
planning
to
submit
a
talk
and
then
coming
up
soon
is
cloud
native
security
Con
in
North
America,
and
this
is
not
a
web
assembly
specific
conference,
but
our
very
own
Kevin
Hoffman
is
going
to
be
speaking
at
this.
One.
I
can
actually
probably
see
the
schedule
if
we
look
for
Kevin
there.
It
is
zero
trust
in
the
cloud
with
webassembly
and
wasm
cloud,
and
this
is
I'm
happy
to
let
Kevin,
if
you'd
like
to
add
a
little
bit
more
information
onto
this.
A
But
this
talk
is
going
to
be
all
about
the
different
levels
of
security
that
you
get
starting
with
webassembly,
so
not
Lawson,
Cloud
specific,
it's
all
the
denied
by
default,
sandbox
model
things
and
then
all
the
levels
that
we
add
on.
While
we
add
functionality
with
wozni
Cloud
Kevin
did
I.
Do
your
talk,
abstract,
decent
Justice,
yeah.
D
Yeah,
it's
first
I'm
just
trying
to
figure
out
where
that
the
picture
of
me
came
from
I
have
way
more
hair
than
that,
so
I'm
not
sure.
D
What's
going
on
yeah
a
clippinger
but
yeah,
the
basic
idea
of
mine
in
the
presentation
is
to
go
from
the
low
level
web
assembly
at
its
lowest
level
and
talk
about
the
security
there
and
then
just
kind
of
go
on
a
journey
from
there
all
the
way
up
through
the
various
runtimes
and
then
through
wasmo
cloud,
with
an
eye
for
how
security
works
at
each
one
of
those
levels.
A
So
this
one
is
going
to
be
really
sweet
if
anybody
can
make
it
to
this
conference.
I
think
it's
in
is
it
in
LA
can't
seem
to
Seattle
West
Coast,
all
the
same
sorry
yeah
it's
going
to
be
in
Seattle.
This
one
is
going
to
be
fun.
Hopefully
this
is
also
live
streamed,
but
I'm
sure
the
the
talk
will
be
uploaded
to
YouTube
after
the
fact
so
yeah
lots
of
lots
of
exciting
conferences
and
events
coming
up
in
the
year.
A
I'm
sure
that
there's
there's
going
to
be
more
and
more
announcements
about
conferences
that
we're
going
to
when
we're
talking
about
awesome
cloud
and
things
like
that.
But
this
is
starting
off
the
year
on
a
pretty
pretty
fun.
Note
I
know
that
we
ran
a
little
bit
over
time.
It's
202,
so
I
want
to
thank
everybody
for
being
here
for
this
community
meeting,
really
awesome,
really
great
questions
and
conversation
and
I'm
going
to
be
putting
together
the
notes
and
the
recording
after
the
fact.
A
F
Do
you
want
to
read
a
readout
and
so
that
you
know
everybody
doesn't
die
of
anticipation
of
what
Justin
was
typing.
A
Oh
yeah,
that's
right,
sure
sure
I
actually
haven't
gotten
to
read
it
myself,
but
I'm
just
going
to
read
the
whole
thing.
So,
given
the
history
of
the
actor
model,
this
is
all
Justin's
words.
I
didn't
come
up
with
whatever
brilliant
things
I'm
about
to
say
and
the
need
for
ongoing
education
for
many
computer
science
and
individuals.
A
I
actually
think
the
actor
model
may
need
something
like
what
the
Illustrated
children's
guide
gave
to
many,
which
I
think
is
like
a
kubernetes
Illustrated
children's
guide,
maybe
meaning
the
actor
model,
needs
to
be
personified
as
it
is
with
existing
language
and
terminology,
but
characterizations
that
make
it
more
approachable
which
yeah
I
think
in
the
chat
got
kind
of
a
resounding
ooh.
This
is
a
really.
A
This
is
a
really
cool
idea
and
I
I,
like
that
I
really
like
the
idea
of
you
know
with,
and
to
summarize
what
I
kind
of
was
going
to
say
before
the
discussion.
I
think
that
actor
is
a
very
fair
name
for
what
we
call
our
webassembly
modules
and
it
really
accurately
describes
what
they
are.
So,
even
with
some
preconceived
notion,
I
think
that
it
properly
describes
what
we're,
what
we're
doing
so.
A
I
really
love
this
idea
of
taking
the
education
route
and
taking
something
like
the
actor
model
and
adding
in
some
some
extra
fun
things
like
illustrations
and
things
like
that.
Foreign.
A
Prevented
death
by
anticipation,
I'm
gonna
go
ahead
and
stop
the
stream.
Thank
you
all
for
for
tuning
in
across
all
the
different
channels.
Anybody
who
is
hanging
out
here
welcome
to
hang
out
in
the
zoom
for
a
little
bit
longer.
If
you.