►
From YouTube: wasmCloud Community Meeting - 16 Aug 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
All
right,
hello,
everyone
welcome
to
the
wise
and
Cloud
community
meeting
for
Wednesday
August
16th
we've
got
a
pretty
packed
and
and
a
little
little
demo
stacked
a
community
call
today.
So
it
should
be
a
lot
of
fun
and
I
figure.
We
might
as
well
go
ahead
and
get
started
I'm,
giving
the
first
demo,
which
is
around
the
Autumn
actor
demon
scaler.
A
So
let
me
go
ahead
and
change.
My
share
really
quick,
I
think
I'm
just
going
to
share
my
whole
screen,
because
I'll
probably
go
back
and
forth
between
a
couple
of
different
things.
Let
me
do
this
here,
we'll
go
ahead
and
keep
this
there.
So
I
can
look
for
any
comments
in
the
chat.
Okay,
so
you
may
be
wondering
what
is
the
wadham
actor
name?
Demon
scaler?
It's
not
Damon!
It's
demon
as
as
Caroline
in
front
of
me.
Well,
the
demon
scaler
came
out
of
something
that
I've
been
wanting
to
do
with
wadham.
A
Ever
since
we
first
wrote
the
original
spread
scaler
and
if
you're
unfamiliar
with
the
way
that
wadham
manifests
work
in
terms
of
deploying
components
across
so
awesome,
Cloud
lattice
effectively
in
the
actual
manifest
for
for
an
application,
you
define
the
whole
list
of
components
that
you
want
to
be
running
in
a
lattice,
and
then
you
define
a
specific
trait
and
the
one
that
we
had
support
for
previously
and
and
still
will,
but
it's
called
a
spread
scalar
and
with
the
spread
scaler.
A
This
actually
has
support
for
defining
a
number
of
replicas,
so
you'll
be
able
to
do
something.
Like
you
know,
I
say:
I
want
one
copy
of
echo
running
in
the
lattice
or
110
or
100
or
whatever,
whichever
number
you
actually
want,
and
then
you
can
also
Define
a
certain
number
of
spreads,
and
this
is
a
relatively
simple
spread
configuration.
We
can
probably
look
up
a
more
complex
one
here.
A
Let
me
zoom
in
just
a
little
bit
more
because
I
know
it
always
comes
through
a
little
different
on
stream,
so
here's
an
example
of
a
more
complex
spread,
so
we're
saying
we
want
five
replicas
of
the
echo
actor
and
we
want
to
spread
it
across
hosts
that
have
labels
region
for
a
test.
So
we
had
some
fun
with
this
us
Brooks,
East,
U.S,
Taylor,
West
and
then
also
running
on
the
moon
of
course,
and
then
these
also
have
an
Associated
weight
with
them
effectively.
A
Now,
what
I
have
been
working
on
and
I
just
added
it
it's
in
a
work
in
progress
PR,
so
it's
not
out
yet
as
another
type
of
scalar
called
the
demon
scaler
and
what
the
demon
scaler
is
for
is
just
like
a
kubernetes
demon
set
it's
effectively
based
on
the
spread
that
you
define
going
to
spread
in
a
component.
Just
for
now,
it's
it's
just
actors,
but
it's
going
to
spread
this
component
across
every
matching
Post
in
a
lattice.
A
So
this
is
the
version
of
the
Manifest
that
I
have
deployed
right
now
or
the
one
that
I
actually
at
least
have
put.
And
if
we
look
at
our
one
host
here
in
the
on
the
awesome,
Cloud
dashboard,
it
is
it's
effectively
blank
right
now,
I,
just
I've.
Just
put
this
manifest
Echo
and
if
I
do
watch
app,
deploy
Echo
with
this
version,
we'll
see
that
Echo
HTTP
server
and
then
as
soon
as
that's
healthy,
the
link
will
come
up
and
now
we
can
hit
our
echoactive
now.
A
What
I
really
wanted
to
have
is,
depending
on
the
application
that
you're
writing
it
could
be
really
useful
to
be
able
to
dynamically
scale
this
application
based
on
what
infrastructure
comes
into
the
lattice.
So
if
another
host
comes
up
with
a
matching
set
of
labels,
we
may
want
to
spin
up
an
entirely
duplicate
copy
of
this
application.
A
Now,
for
this
demo,
I'm
just
going
to
keep
it
to
the
actor
because
I
I
don't
know
if
you've
worked
with
the
HTTP
server
provider
before,
but
that
opens
up
a
port.
It's
a
whole.
It's
a
whole
thing,
so
you
know
always
have
to
be
keep
that
kind
of
stuff
in
mind,
but
I'm
going
to
scale
this
actor
and
basically
say:
hey
I
want
one
copy
of
the
echoactor
to
run
on
every
single
host
that
has
this
label.
A
How
square.os
is
is
Mac
OS,
so
let
me
go
ahead
and
update
the
version
of
the
Manifest
and
then
the
type
here,
the
trait
for
the
demon
scaler,
is
actually
exactly
the
same
as
the
spread
scaler.
So
it's
effectively
like
a
drop
in
replacement,
but
now,
instead
of
running
one
total
copy
on
a
host.
That's
running
on
Mac
we're
going
to
run
one
copy
on
every
host,
that's
running
on
the
Mac,
so
I
can
deploy
this
actually
I'll
deploy
the
new
manifest,
which
is
Echo
and
nothing
is
going
to
change
right
away.
A
We
only
have
one
running
host,
but
the
demon
scaler
is
effectively
doing
its
job.
Saying:
hey
I'm,
going
to
run
this
actor
on
this
host.
However,
this
is
a
different
thing.
What
we
can
actually
do
with
watch
is
just
like.
We've
run
this
individual
host
here
by
running
wash
up,
we
can
run
another
host
which
will
automatically
connect
into
the
same
lattice
and
have
the
same
system
level
labels,
so
I
can
run
another
wazenfone
host
and
as
soon
as
that,
connects
we'll
see
that
another
host
has
popped
up
and
it's
already
running
the
echo
actor.
A
It
actually
does
it
so
quickly
that
I
can't
switch
over
to
that
tab.
But
if
we
look
at
the
all
host
view,
we
can
see
that
we
have
I
I
understand
that
this
isn't
the
best
visualization,
but
we
can
see
that
we
have
two
running
Echo
actors
and
then,
if
I
were
to
do
this
again
run
another
awesome,
Cloud
host.
A
Of
course
this
is
going
to
have
the
Mac
label
as
soon
as
that
starts
up,
Autumn
is
going
to
say,
hey,
that's
a
new
host
that
has
the
label
that
I'm
looking
for
and
then
start
up
an
echoacter,
so
I'm
really
looking
forward
to
coming
up
with
a
compelling
demo.
For
this,
like
I,
said:
I
am
working
on
this
for
a
plan
demo
for
the
upcoming
wasam
Khan
conference.
A
So
this
is
going
to
be
really
fun
to
be
able
to
say,
like
hey,
every
single
host
go
start
this
application,
but
this
is
what
the
Autumn
demon
scaler
is,
and
if
we
take
a
look
at
the
app
status,
we
can
see
that
it's
ready.
You
know
it's
running
all
the
things
that
we
wanted
to,
but
if
we
were
to
shut
down
one
of
the
hosts,
for
example-
and
we
lose
that
Echo
actor,
it
should
still
be
yeah.
It's
still
ready.
A
This
is
still
an
acceptable
status
for
the
Daemon
scaler,
because
it
is
running
an
echoactor
on
all
the
hosts
that
are
running
with
that
label
so
yeah
this
was.
This
was
a
pretty
exciting
thing.
I
I
have
to
be
honest.
It
wasn't
all
that
complicated
to
actually
implement
this.
You
can
go.
Take
a
look
at
the
code
on
your
own.
A
If
you
want
I,
won't
won't
do
a
deep
dive
here
just
for
time,
but
I
think
it
only
ended
up
being
I
mean
I
kind
of
I
was
able
to
reuse
a
lot
of
the
same
logic
from
the
spread
scaler
but
like
over
half
of
the
lines
of
code,
and
that
PR
is
just
running.
You
know
tests
to
make
sure
that
it's
working
well,
let's
see
I'm
gonna
go
ahead
and
stop
sharing
here
just
so
that
I
can
bring
the
focus
on
everybody
again
and
see.
If
there's
any
any
questions,
that's
the
that's!
A
A
How
would
you
handle
load,
balancing
I,
think
that
you
know
Kevin
and
Taylor
are
definitely
getting
at
the
the
point
in
the
comments
here,
but
around
the
HTTP
server
and
scaling
that
assuming
it's
not
all
on
the
same,
like
local
Port,
so
you're
running
it
on
different
machines
and
everything
like
that
effectively,
the
HTTP
servers
will
all
spin
up
their
own
listener
for
that
for
that
individual
actor
link
and
it's
not
something
that
I
think
we're
planning
on
I.
A
Think
I,
don't
think
we're
planning
on
doing
something
external
to
was
on
cloud
to
actually
load
balance
between
those
HTTP
servers
kind
of
like
if
you
have
different
machines
in
AWS
or
something
like
that,
but
I
think
we're.
Definitely
thinking
about
how
that
could
be
best.
That
could
be
best
done.
I
think,
ultimately,
that
just
comes
up
to
you
know
taking
the
endpoints
that
you're
listening
on
and
then
using
a
something
at
a
higher
level
than
awesome
cloud
like
in
the
infrastructure
side
to
to
load
balance
between
those
endpoints.
A
Any
other
questions
for
the
the
Daemon
scaler
demon,
scaler.
A
Sweet
well,
oh
Liam,
psyllium,
meshing,
probes,
comment
came
in
or
is
that
is
that
a
whole
separate
discussion
there,
oh
yeah
I,
think
that's
I,
think
that's
a
separate
thing.
A
All
right,
well,
I,
look
forward
to
showing
off
the
the
demon
scale
a
little
bit
more
plan
to
have
a
little
bit
more
of
a
you
know,
deploying
a
whole
application
demo.
Perhaps
I
can
show
that
next
week
or
the
week
after
as
long
as
it's
not
too
much
of
a
spoiler
for
the
for
the
conference
all
right.
A
Well,
then
I
think
we
can
go
ahead
and
move
on
to
the
next
item
in
our
list,
which
is
let
me
trying
to
coordinate,
getting
Kevin
co-host
so
that
I
don't
get
don't
get
yelled
at
in
real
time.
It's
going
to
be
the
concordance
event,
sourcing
code,
gen
and
documentation,
enhancements
preview,
so
Kevin
go
ahead
and
take
it
away
all.
B
Okay,
so
you
should
be
looking
at
a
chrome
window
with
some
stuff
from
GitHub
on
it
right,
that's
right,
okay,
cool,
so
the
for
I
guess
for
background
concordance
is
an
event
sourcing
framework,
that's
built
on
top
of
wild
and
Cloud,
so
that
you
build
your
event,
sourcing
Primitives,
as
was
implant
actors,
or
you
know,
webassembly
components,
and
then
concordance
takes
care
of
all
of
the
hard
work.
B
And
one
of
the
samples
that
we've
built
is
the
simple
bank
account
example.
That
shows
you
know
just
the
basic
Aggregate
and
how
to
do
you
know
some
really
simple
processing
on
it
and
if
we
go
through
the
code,
you'll
see,
there's
a
a
module
called
event
sourcing
that
and
what
we're
looking
at
is
before
some
of
the
enhancements.
So
I
want
to
kind
of
compare
and
contrast
between
the
the
old
and
the
new,
even
though
everything
is
sort
of
relatively
new
in
this.
B
In
this
version
of
the
example,
there's
a
build
that
RS
file
that
automatically
runs
the
Smithy
code,
generator
to
build
a
bunch
of
stuff
for
event
sourcing
and
then
there's
an
aggregate
service
that
we
Implement
for
this
bank
account
Aggregate
and
then
there's
a
pile
of
boilerplate
here,
where
we
basically
just
switch
on
the
data
on
the
type
name
of
the
various
events.
So
if
we
want
to
handle
a
command,
I,
didn't
I
forgot
that
that
thing
clicks
there.
B
So
when
we
handle
the
commands,
we
go
in
and
do
a
switch
on
the
various
commands
and
then
run
the
right.
The
right
function
same
thing
with
applying
an
event
is
we
apply
the
event
and
then
we
do
a
big
switch
on
the
event
data
type
and
we
manually
vote
deserialize
in
order
to
run
our
own
functions.
B
So
there's
a
couple
of
things
that
are
not
ideal
about
this.
The
first
one
is
obviously
that
we,
we
don't
have
strongly
typed
handlers.
So
there's,
no,
you
know
handle
create
account
command,
there's
just
handle
command,
and
then
we
do
a
switch
and
it
deserialize
all
that
stuff.
So
it's
a
lot
of
excess
typing
and
it
makes
most
of
our
code
most
of
the
useful
information
in
our
code
appear
below
the
fold
yeah.
So
that's
that's!
No
good!
B
The
other
thing
is
visibility
and
discoverability.
So
if
we've
experienced
this
firsthand,
where,
when
you're
building
a
bunch
of
event
sourced
components
like
Aggregates
and
projectors
and
which
commands
are
handled
by
which
Aggregate
and
which
events
are
handled
by
which
other
component,
the
only
way
to
see
that
is
to
go
into
the
code
and
look-
and
you
know
hope
that
this
information
is
accurate
and
so
that's
that's
no
good
either.
B
So
we've
come
up
with
what
I,
what
I
think
is
sort
of
a
a
better
way
to
deal
with
it,
and
so
the
first
thing
I'm
going
to
show
is
a
better
way
to
surface
the
event
flow.
B
B
But
again,
my
my
tolerance
for
friction
is
pretty
low,
so
even
having
to
edit
a
turtle
file
which
is
pretty
straightforward
is
is
not
good
enough.
So
what
I
wanted
to
be
able
to
do
is
actually
write
documentation
on
my
event
model.
So
what
I
have
here
is
an
open
source
product
called
event,
catalog,
just
making
sure
everybody
can
see
this
tab
when
I
switched
yeah
that
looks
good.
B
So
you
know
I've
got
these
events
for
a
sample
application
called
lunar
Frontiers,
where
we
have
an
event
sourced
lunar
rover,
and
so,
let's
see
I've
got
this
Rover
initialized
event,
and
so
let
me
just
walk
through
a
couple
of
these
things
in
in
here,
because
it's
it's
actually
pretty
impressive
and
I'm,
not
saying
that,
because
it's
our
project,
the
event
catalog
itself
is
a
pretty
cool
product.
B
So
we've
got
the
name
of
the
event,
a
summary
of
the
event.
Then
we
actually
have
some.
You
know
full
markdown
that
we
can
use
to
document
the
event
this
diagram
is
automatically
generated.
I
did
not
draw
this.
B
This
schema
down
here
came
from
a
Json
schema
that
I
put
in
the
appropriate
directory,
so
this
product
is
also
automatically
rendering
the
schema
for
this
event,
and
so,
when
I'm
looking
at
this
event,
I
can
see
that
the
Rover
aggregate
produces
this
event
and
they're
over
aggregate
also
consumes
it
for
its
own
state,
but
there's
also
two
other
consumers,
and
so
you're,
probably
thinking
you
know,
that's
great.
But
how
do
you
deal
with?
B
You
know,
event
versions,
and
you
know
I'm
glad
you
asked
Brooks.
Thank
you
so
over
here
you
can
see
that
we've
got
the
latest
version
and
we
also
have
this
thing
called
ve
0.0.9.
B
So
if
I
click
in
on
this,
not
only
so
now
I'm
looking
at
the
old
version
of
this
event,
I
can
see
that
I'm
viewing
an
old
version
of
the
event.
It
gives
me
a
big
warning
that
I'm
doing
that
and
you're,
probably
thinking
it'd,
be
great.
If
I
could
do
I,
don't
know
a
diff
between
the
old
schema
and
the
new
schema,
so
I
can
see
what
changed
and
again
Brooks
is
coming
up
with
really
great
questions
here.
B
If
I
do
view,
changes
on
this
I
actually
get
a
diff
of
the
schemas.
So
I
can
see
exactly
what
was
changed
between
you
know,
0.9
and
the
1.0
version
of
this
event,
and
this
kind
of
visibility
is
something
that
is
almost
completely
unheard
of.
You
never
see
this
when
you're
building
Event
Source
applications,
if
you're
using
Kafka
and
you
use
Avro,
and
then
you
use
a
couple
of
other
add-on
products
and
then
a
couple
of
plug-in
code
generators.
B
B
The
event
it's
a
little
slow
because
I'm
I'm
running
like
the
the
development
version
of
the
server
and
it's
this
actually
can.
This
is
designed
to
be
statically
generated,
so
I
can
run.
You
know,
npm
build
and
then
push
it
to
netlify
and
now
I
have
my
site.
B
B
Like
I
said
it's,
you
know
it's
a
little
slow
so
again,
this
this
diagram
was
automatically
generated.
I
didn't
do
this,
I
supplied
the
metadata
and
it
generated
the
diagram
for
me.
So
you
can
see.
I've
got
the
command
up
here
and
then
some
of
these
events
and
that's
great
but
there's
probably
a
better
way
to
visualize
my
entire
system
as
a
whole.
Again
thanks
Brooks.
B
So
if
I
click
on
all
events
and
services,
now
I
have
a
complete
holistic
view
of
my
event:
source
application-
not
just
you
know
what
goes
into
one
Aggregate
and
what
goes
into
another
one.
So
I
can
zoom
in
thing
to
do
and
the
what
we
what
we
usually
refer
to
as
antlines
the
this
animated
marching
stuff.
This
shows
that
the
event
flow
Direction.
So
you
can
see
that
they're
over
aggregate
is
emitting
all
of
these,
and
then
you
can
see
where
they
go
and
you
know
I
can
move.
B
Oh
yeah,
one
of
the
other
things
I
I
didn't
mention,
there's
so
much
stuff
on
this.
That
I
wanted
to
demo
that
I'm,
forgetting
all
the
cool
stuff.
B
In
addition
to
these
mermaid
diagrams
being
automatically
generated,
they're
interactive,
so
if
I'm
looking
at
initialized,
rotor
and
I
want
to
go
to
the
aggregate
I
can
just
click
on
the
aggregate
in
the
diagram
and
I'm
I'm
good
to
go
so
yeah
like
I
said
this
is
a
I
had
only
heard
of
this
briefly
and
for
whatever
reason,
I
didn't
go
into
too
much
detail,
and
then
Bailey
mentioned
it
the
other
day,
and
you
know
in
super
short
work.
B
I
was
able
to
put
together
an
entire
model
for
the
lunar
over
sample
that
we've
got
and
there's
another
cool
feature
on
this
again.
I
didn't
generate
these
These
are
generated
by
the
the
ad,
so
I
have
a
3D
visualizer
for
my
event
flow.
B
So
now,
I
can
see
that,
like
the
Rover
pilot
process,
manager
is
like
the
linchpin
of
everything
it's
receiving
all
these
events,
but
you
can
also
see
kind
of
these
little
clusters
where
You're
Building
every
has
you
know
the
construction
stuff,
but
it
also
feeds
in
here
and
then
you
can
also
zoom
out
and
find
these
islands
of
separate
functionality.
B
You
can
organize
things
by
domain,
so
I
could
do
like
you
know,
spaceships
as
a
whole
domain
and
then
put
all
the
lunar
stuff
underneath
it
if
I
wanted
to.
So
it's
not
I,
don't
have
to
have
everything
just
top
level.
B
B
B
So
one
of
the
other
things
that
I
realized
is
I've
got
this
documentation
on
this
event,
Source
system
I
know
what
the
event
flow
is
from
the
documentation.
I
know
what
the
schema
is
from
the
documentation,
so
my
next
thought
was
maybe
I
can
actually
reverse
engineer
code
from
the
documentation
so
again,
I'm
glad
you
asked
that
Brooks,
because
I
have
that
shown
right
here.
B
Okay,
so
can
everybody
see
Visual
Studio
code
right
now,
yep
all
right.
So
again,
it's
it's
worth
remembering
just
how
High
friction.
Things
were
back
in
the
old
days
and
by
that
I
mean
three
weeks
ago
in
the
old
way
you
had
to
manually
generate
your
event,
sourcing,
Smithy
stuff
and
then
build
these
giant
switch
statements
and
you
within
each
piece
of
code.
You
didn't
have
any
idea
what
the
rest
of
that
whole
event
flow
looked
like.
B
B
B
So
here's
I
I
point
it
to
the
website.
I
tell
it
that
I
want
an
Aggregate
and
I
want
the
Rover
version
of
that
it
generated
this
trait
and
now
I
have
all
of
these
strongly
typed
functions,
so
I've
got
handle
initialized,
Rover
I've
got
you
know.
So
this
this
handle
is
an
incoming
command
and
then
I
have
the
apply
functions
for
applying
the
events
to
the
aggregate
state
and
yeah.
So
the
the
workflow
now
becomes
a
model
and
brainstorm.
My
Event
Source
flow,
so
I
write
down
all
my
events.
B
I
write
down
all
my
commands,
I
build
all
my
Aggregates
out,
I
see
what
that
whole
thing
looks
like
visually
and
I
can
actually
see
it
in
3D
and
then,
when
I'm
ready,
I
can
just
use
this
one
macro
in
every
web
assembly
module
that
is
going
to
be
bound
to
concordance
and
that's
all
I
need.
So
in
old
versions
of
this
there
was
a
build.rs
file.
There
were
four
different
code,
generated.rs
files-
it
was,
you
know
it
was
better
than
nothing,
but
it
was
still
a
pretty
high
friction
and
now
I
think.
B
B
So
that's
all
I
had
for
at
least
the
planned
demo.
I
can
show
what
something
specific
if
people
want
to
take.
If
you
want
want
me
to
dive
deeper
on.
A
Done
silence,
Kevin,
I'm,
I,
don't
know
what
books
you
were
hearing
questions
from,
but
I'm
glad
that
he
had
the
good
things
because
I
I,
don't
really
you
know
you
didn't
tell
us
what
this
demo
is
going
to
be
and
I
don't
really
know
exactly
what
I
expected
I
kind
of
expected
this
like
this
one.
You
know
macro
here:
yeah.
B
Well,
last
week,
this
is
what
it
was
like.
This
was
the
extent
of
it
last
week,
except
instead
of
the
path
being
a
path
to
the
event
catalog
website.
It
was
a
path
to
that
hand-typed
Turtle
rdf
file,
where
you
know
I
specify
this
event
goes
into
this
aggregate.
This
event
goes
into
this
Aggregate
and
so
on,
and
that
was
tedious
and
boring
and
I
wanted
something.
B
Prettier
and
a
little
bit
of
backstory
is
I
tried
using
async
API
yaml
files
which
do
plug
into
the
event
catalog
website,
and
it's
not
nearly
good
enough.
I
ended
up
filing
a
couple
of
bugs
against
the
async
API
plug-in,
but
either
way
so
yeah
I'm
super
happy
with
being
able
to
write
my
documentation
first
and
then
guarantee
that
you
know
my
code
is
actually
working
against
the
latest
version
and
I
don't
support
I,
don't
I
mean
because
I
you
know,
I
wrote
this
yesterday.
A
B
Don't
support
selecting
it
but
like
I
would
imagine
that
if
I
want
to
like
it'll
default
to
latest,
but
you
know
if
I
want
to
pick
the
versioned,
you
know
a
specific
version
of
this
entity,
then
I
can
do
that.
A
Kevin,
the
only
well
really
what
I
wanted
to
say
is
that
I
can't
wait
to
actually
just
look
at
this
and
run
it
on
my
own,
like
I.
Would
love
to
I
would
love
to
start
like
from
an
idea
to
an
event
source
application,
like
start
with
a
template
with
this
one
and
then
be
able
to
go
from
there?
B
That
is
the
plan.
I
want
to
be
able
to
spit
out
a
new
event,
catalog
site,
and
then
you
know
with
like
hello,
world
or
or
something
in
it,
and
it
would
have
a
couple
of
like
an
aggregate,
a
couple
of
events
and
so
on.
So
that
brings
me
to
another
Point
like
this
stuff
here.
B
This
Rover
initialized
I'm,
currently
hand
coding
this,
like
this
data,
but
again
before
this
stuff,
before
I'm
finished,
pulling
the
duct
tape
and
bubble
gum
off
of
this
one
of
the
things
I
want
to
be
able
one
of
the
things
I
should
be
able
to
do
well.
I
I
can
because
I've
got
it
in
another
directory
running
but
is
take
the
Json
schema,
that's
sitting
inside
of
the
event,
definition
in
the
website
and
turn
it
into
a
rust
structure
and
verify
that
that
works.
A
Kevin
I'll
ask
one
one
quick
question
and
then
CJ
had
a
few.
Is
this
pushed
up
somewhat
like?
Is
this
in
a
repo
somewhere
that
people
can
take
a
look
at
or
is
it
still?
It's.
B
It's
not
it's!
It's
still
highly
volatile
like
so
just
to
give
you
an
example.
Two
days
ago,
the
active
PR
for
this
was
based
on
the
turtle
rdf
file
I
got
nerd
sniped
by
Bailey,
and
now
it's
based
on
this
website
stuff
and
documentation
first,
and
it's
a
an
infinitely
better
experience.
B
So
my
plan
is
to
just
make
it
so
that
this
stuff
is
usable
by
people
other
than
me
and
then
I'll
have
that
pushed
up
into
a
PR
and
then
merged,
and
then
we
can
just
kind
of
iterate
over
it
to
to
smooth
that
out.
B
Like
you
said,
I
I
want
to
be
able
to
go
through
the
process
of
creating
the
documentation
site
and
then
generating
the
code
and
like
that's
what
I
did
with
this
lunar
Frontiers
site
like
in
addition
to
coding
all
of
the
stuff
that
pulls
from
the
site,
I
was
able
to
actually
write
all
of
the
content
for
that
site
in
that
short
amount
of
time
too.
So
it's
all
super
easy.
A
So
cool
yeah
I
can't
wait
to
see
it.
Yeah
CJ
feel
free
to
come
on
if,
if
you
got
mic
access
but
otherwise
we'll
get
to
your
questions.
C
B
So
what
this
stuff
here
isn't
generated
by
the
macro
everything
generated
by
the
macro
is
currently
invisible
here.
This
macro
generates
the
Rover
aggregate
input
strut,
and
then
it
generates
the
trait
and
then
you
know
to
build
to
to
fill
this
out.
I
just
flip
the
trait
and
then
did
you
know,
Implement
method,
Implement
rate
members
in
terms
of
CI
and
CLI.
B
Thankfully,
there's
absolutely
no
CLI
involved
in
this.
Everything
that
you
need
is
done
by
the
macro,
so
the
the
only
real
tricky
part
is
in
ensuring
that
your
code
has
access
to
the
event
catalog
markdown
files.
When
you
build.
B
And
yeah
I
need
to
apologize
for
this
red
underline
here.
I
have
no
idea
why
Visual
Studio
code
won't
do
this.
This
code
does
build
and
like
I,
can
hover
over
this
and
it
says
I
can't
find
the
type,
but
yet
there's
the
documentation
for
the
type.
So
whatever
vs
code,
thanks
for
nothing.
D
I,
have
it
might
be
worth
it
if
one
of
us
tries
pulling
it
and
seeing
if
it's
it's
not
just
you,
you
know
what
I
mean
like
what.
If
what,
if
it's
an
issue
with
some
kind
of
macro
definition
in
in
the
rest,
analyzer
right
and
we
all
get
it.
B
I
had
that
I
had
that
plan
and
so
I
copied
the
code
all
over
to
another
Unix
box
or
another
Linux
box
and
I
opened
it
up.
No
underlines
it's
just
my
machine.
Thanks
love
it
okay,.
B
But
yeah
I
figure
sometime
early
next
week,
I'll
be
able
to
start
pushing
out
bits
and
pieces
of
the
the
stuff
that
I
did
here.
So
you
know,
I
got
cobbled
all
this
together
kind
of
quickly.
It
only
currently
supports
generating
Aggregates
this
way,
but
like
all
the
code
for
project
management
process
managers
and
all
the
other
components,
it's
all
sitting
there,
I
just
haven't,
you
know
exercised
it
so
I'm
sure
there's
bugs
in
there.
A
A
I,
don't
see
any
questions
on
the
live
stream
right
now,
but
does
anybody
else
have
any
other
questions
for
for
Kevin
here.
C
B
Like
I
said,
I'll
keep
everybody
posted,
I'll,
probably
end
up
doing
a
blog
post
and,
or
you
know,
video
walkthrough
of
this
when
it's
in
better
shape
and
I'll
keep
everybody
posted
during
the
weekly
meetings.
A
Yeah
speaking
for
everybody
here
and
tell
me,
if
you
disagree,
but
I
would
love
to
have
whenever
you've
got
progress
on
this
I
would
love
to
have
a
demo
in
the
community
call.
This
is
just
so
cool
such
a
cool
application
of
awesome
Cloud.
It's
it's
awesome
to
look
at.
A
Alrighty,
well,
we
are
getting
to
our
last
agenda
item
today
and
this
is
kind
of
a
soft
discussion
time,
but
I
really
want
to
try
and
Reserve
time
in
the
WASP
Cloud
Community
call
to
go
over
our
roadmap.
This
is
really
both
a
function
of
keeping
everybody
up
to
date
and
trying
to
keep
me
honest
with
the
things
that
we
are
keeping
on
the
roadmap.
A
If
you
haven't
oops,
if
you
haven't
seen
this,
this
is
in
our
documentation
site
now
under
under
this
top
level
section
for
roadmap
around
what
we're
working
on
as
developers
and
maintainers
of
wasm
cloud
for
the
you
know,
essentially
through
the
end
of
2023
and
our
top
level
things
we're
really
focusing
on
are
leveraging
webassembly
interface
types
for
all
of
our
interfaces.
A
You
know
kind
of
major
changes
in
the
ecosystem.
There.
There
was
a
desire
to
know
which
ones
are
really
more
of
an
effort
on
the
maintainer
side,
like
on
the
development
of
awesome
Cloud
side,
and
then
what
is
changing
the
experience
for
people
who
are
developing
apps
with
wasm
cloud,
and
this
is
really
important
to
know
because
for
something
like
leveraging
wit
for
our
interfaces
being
able
to
use
wit
for
defining
your
own
custom
interfaces.
A
It's
much
more
desirable
to
picking
it
up
now
and
defining
an
interface
in
Smithy,
which
will
be
you
know,
eventually
deprecated,
with
the
move
to
components
and
webassembly
interface
types.
So
I
really
just
want
to
be
clear
on
some
of
the
active
Milestones
that
we
have
that
the.
If
you
look
in
the
Watson
Cloud
Watson
Cloud
repository,
we
have
two
big
Milestones
here.
A
The
first
one
is
widify,
and
this
is
all
around,
using
wit
for
our
interfaces
defining
defining
interfaces,
using
wit
and
being
able
to
use
them
for
our
code
generation
like
for
providers.
This
is
a
really
important
one
to
watch
if
you're
looking
to
develop
applications
for
wasmcloud,
if
you're
going
to
create
your
own
custom
interface
for
an
application,
there's
certainly
nothing
stopping
you
from
doing
it
in
Smithy.
Today,
it's
just
a
good
thing
to
keep
in
mind.
A
Yeah
thanks
Siri,
it's
just
a
good
thing
to
keep
in
mind,
going
forward
that
we're
going
to
be
using
wit
for
our
interface
types,
and
we
would
be
really
happy
to
work
with
you
in
the
wazzle
cloud
slack.
If
you
wanted
to
get
started
today
and
then
move
to
wit
us
on
the
Watson
Cloud
side,
we
have
many
contracts
that
are
moving
to
wit,
so
we're
definitely
going
to
be
putting
out
a
guide
about.
A
You
know
a
migration
guide
from
a
Smithy
to
a
wit
interface,
the
other
larger
milestone,
which
was
updated
eight
minutes
ago.
There
are
many
things
going
in
and
out
of
this
Milestone,
all
the
time
is
getting
the
rust
host
up
to
feature
parity
with
wasmcloud
OTP,
and
this
one
is
really
something
that
we
are
focusing
on
on
the
maintainer
side
and
is
important
to
know,
but
not
really,
something
that
should
provide
any
hesitation
for
you.
Building
was
on
cloud
applications
effectively.
A
This
effort
is
all
around
you
know,
taking
the
rust
host
into
having
it
be
a
basically
a
drop-in
replacement,
either
on
the
lattice
or
on
the
command
line
for
the
OTP
hosts
and
I.
Just
want
to
note
that
some
of
these
issues
in
here
aren't
really
ready
for
taking
by
somebody
who's
kind
of
new
to
the
project.
So
like
this
one,
for
example,
I
just
filed
it
yesterday,
because
I
thought
I
knew
that
it
needed
to
happen
and
I
didn't
add
a
description
and
it
actually
is
a
duplicate.
A
So
we
might
be
able
to
close
that
one,
but
anyways
I
just
wanted
to
be
pretty
clear
about
the
efforts
that
we're
putting
in
across
the
the
wasmcloud
ecosystem
and,
if
you're,
a
developer,
developing
applications
for
wasm
cloud.
Now.
This
first
effort
is
one:
that's
really
important
to
keep
in
mind
that
we're
moving
to
wit
interfaces
this
second
one,
which
would
be
for
a
net
new
applications
that
you're
building
with
webassembly
components
which
we're
super
excited
to
to
add.
A
You
know
we're
going
to
be
using
standard
interfaces
instead
of
the
ones
that
we've
been
using
for
the
last
couple
of
years.
This
is
going
to
be
a
really
awesome
way
to
build
wasm
applications
in
general,
but
for
you
to
be
able
to
bring
that
to
wasm
Cloud
without
using
a
proprietary
interface,
and
so
these
Milestones
should
be
really
great
for
if
you
just
wanted
to
keep
an
eye
on
how
we're
how
we're
doing
on
these
and
see
how
they're
progressing
these
are
going
to
be
changing
rapidly.
A
But
that
is
essentially
essentially
where
we're
at
I
also
wanted
to
just
do
one
more
call
out
that
there
is
the
the
last
PR
for
validating
the
Manifest
is
going
to
be
merged
into
bottom,
just
a
little
bit
after
we
stop
this
call
and
then
solving
one
of
these
final
bugs
water
will
be
ready
for
a
release
candidate,
so
zero.
A
Five
rc1
should
be
going
out
pretty
soon
as
well,
which
will
include
some
of
the
things
we've
been
demoing
like
manifest
upgrades,
manifest
status
and-
and
things
like
that
so
wadham
is-
is
making
very
healthy,
healthy
movement
and
should
have
some
pretty
exciting
updates.
A
As
soon
as
0.5
is
released
fully
we'll
get
that
into
wash,
and
then
you
all
will
be
able
to
play
with
that
officially
in
the
next
next
version
of
wash
so
looking
at
the
development
roadmap,
that's
effectively
where
we're
at
I
don't
think,
there's
any
more
information
from
the
official
quote-unquote
roadmap
that
you
could,
then
that
would
you
know,
give
more
information
that
kind
of
what
I,
what
I
just
kind
of
spieled
about,
but
it's
all
looking
really
good
and
I'm
pretty
excited
about
it.
A
Sweet
then
I
think
we
can
probably
leave
the
we'll
leave
the
roadmap
discussion
there.
I'm
gonna
try
to
continue
to
update
people
as
we
are
in
each
Community
call
about
how
we're
moving
on
those
milestones
and
I'll
see
about
opening
up
a
milestone
for
the
for
the
Wazi
Cloud
effort.
I
know
that
there's
already
been
a
good
amount
of
movement
there,
like
some
of
the
we
do
already
support
some
of
the
interfaces
in
the
Watson
photos.
A
So
I
want
to
accurately
capture
where
we,
where
we
are
with
that
and
really
it's
just.
It's
really
exciting.
Looking
forward
as
we
start
to
build
applications
with
with
components.
So
I
can't
wait
to
can't
wait
to
have
you
all
try
some
of
that
and
get
some
Demos
in
here
of
building
webassembly
component
apps
with
awesome
clouds.
A
All
right,
Bailey
just
sent
me
like
45
links
here
in
the
zoom
chat,
which
I
did
open
the
last
little
section
in
our
in
our
community
call
agenda
is
just
to
always
remind
me
that
I
need
to
talk
about
the
wasam
con,
which
is
coming
up
here
in
just
a
couple
of
weeks,
actually
in
September,
6th
and
7th
in
Bellevue
Washington.
A
lot
of
people
on
the
bosom
Cloud
maintainer
side
will
be
here,
we'll
be
there,
which
is
which
is
really
exciting.
D
I
figured
I'd
give
a
shout
out
to
some
of
our
talks
too.
Specifically,
we've
got
a
lot,
so
Brooks
will
be
wow.
You
have
that
many
too
dang.
You
got
a
lot
of
work
to
do
yeah,
so
so
we've
got
just
those
I'm,
also
giving
a
keynote
Liam's
giving
a
keynote.
In
addition
to
those
other
three
talks
and
we're
running
a
workshop
for
cosmonic,
which
will
obviously
be
showing
off
a
lot
of
the
power
of
awesome
cloud.
D
That's
right:
I
am
moderating
the
host
panel,
so
we're
gonna
have
folks
from
Mozilla
and
Google
and
the
webassembly
chair
to
talk
about
how
the
webassembly
standards
and
implementation
for
browsers
impacts
really
a
lot
of
the
roadmap
and
an
impact
in
the
ecosystem.
So
I
think
that'll
be
a
really
fun
discussion.
A
And
then
I
know
that
we
also
you
know
not
not
too
shortly
after,
but
coming
up
at
the
end
of
the
year
is
going
to
be
Cloud
native,
wasn't
day
at
which
is
kind
of
a
regular
staple
now
at
kubecon.
So
the
co-located
events,
but
Cloud
native
wasn't
day
is
going
to
be
it's
going
to
be
pretty
sweet.
I,
don't
think
that
the
now
cfp
is
closed,
but
the
schedule
is
not
out
yet
as
far
as
I
know
so
that'll
be
that'll
be
coming
soon.
We'll
certainly
update
you
all.
A
See
is
there
anything
else
in
the
in
the
conference
world
coming
up,
Bailey
or
Liam
that
I'm
that
I'm
missing
know
that
did.
D
You
mention
componentize
the
world
we'll
have
remote
capability
there,
and
so
you
can
join
in
and
start
learning
how
to
use
some
of
these
tooling.
It's
going
to
be
focused
on
the
bike,
cut
Alliance
side
of
this.
So
a
lot
of
the
by
code.
Alliance
projects
are
sort
of
Upstream
to
awesome
Cloud.
D
We
rely
on
them
and
including
wasm
time,
which
is
our
runtime
there's
all
the
also
projects
called
Blossom
tools
and
Whit
binding,
and
you
can
use
those
to
basically
componentize
dependencies
throughout
the
entire
webassembly
ecosystem,
so
that
onboarding
to
webassembly
is
going
to
be
that
much
easier.
So
this
is
the
goal
of
this
is
to
basically
effectively
bootstrap
a
component
native
ecosystem.
A
So
Bailey,
as
soon
as
you
said
that
I
started
the
wheel
started
turning
in
my
head.
So
like
the
goal,
the
goal
of
this
or,
if,
if
I,
can
come
into
this,
like
it
componentize
the
world,
hackathon
like
I,
could
find
a
library
it's
like
only
written
in
Rust
that
I'm
super
passionate
about
and
I
could
turn
that
into
a
awesome
component
and
then,
like
you
from
a
not
Rusty,
language
could
be
like
I'm
going
to
use
that
library
from
only
rust
but
from
any
language.
Is
that
the
exactly.
D
D
D
I
also
think
it's
funny
because
you
you
had
mentioned
a
rust
project
called
names
earlier
this
morning
and
that's
on
I
have
an
Excel
or
a
Google
sheet.
That's
got
a
bunch
of
the
projects
that
I've
gone
out
and
and
the
rest
space
that
a
lot
of
people
rely
on
and
that's
one
of
them
that
is
in
my
hey.
D
You
want
to
get
started
and-
and
what's
super
easy,
like
really
shouldn't
be
too
tricky
to
convert
to
webassembly
names,
is
one
of
them
right,
because
it's
moving
strings
in
and
out
and
doesn't
really
rely
on
too
many
other
Hardware
kind
of
features.
Those
are
going
to
be
a
little
bit
harder,
but
yeah,
so
I've
got
I've
got
a
list
where
it
breaks
it
down
from
easy
medium
to
hard.
A
That's
super
cool
I
know
that
this
would
have
come
in
handy,
let's
think
so
much
Jordan.
What
was
it
like?
Maybe
a
year
ago,
when
tinygo
didn't
have
like
the
the
Json
packing
that
that
rusted
and
you
wrote
an
example
in
waslam
Cloud
that
did
an
actor
to
actor.
Call
like
a
tiny
go
actor
called
a
rust
actor.
Does
he
realize
it
and
then
bring
it
back?
Is
that
that
was.
A
C
A
That
is
such
a
big
props
to
the
people
in
the
in
the
tiny
go
development
Community.
It's
they
they're,
always
doing
awesome
stuff,
bringing
things
in
from
the
the
go
standard.
Library,
yeah
cool
I'm,
pretty
excited
for
that,
bringing
in
some
some
crates.
A
At
risk
of
putting
you
really
on
the
spot
Bailey,
how
is
the
component
has
JS
effort
coming
along
around
componentizing
a
JavaScript
component,
because
I
know
there's
like
approximately
500
billion
JavaScript
libraries
out
there?
Oh.
D
D
D
It
does
tend
to
be,
like
you
know,
a
few
days
behind
when
we
roll
out
new
stuff
but
I'm
expecting
us
to
hit
a
level
of
stability
within
the
bytecode
alliance
tooling
soon,
so
that
you
know
there's
you
know
the
lag.
You
don't
really
quite
notice
because
we're
not
changing
syntax
every
other
day,
so
we're
very
close.
D
I
had
guy
on
the
bike,
cut
Alliance
Community
stream
two
weeks
ago,
and
he
showed
off
another
project
that
he
works
on
guy
Bedford.
He
works
at
fastly.
D
He
is
the
main
engineer
behind
Jayco
and
componentize.js,
and
he
also
built
a
tool
called
wazivert
and
I
like
to
Hype
that
one
as
well,
because
that's
something
that's
going
to
be
a
really
great
pair,
that
I
I
think
will
bring
into
the
Watson
Cloud
ecosystem
for
virtualizing
things
like
the
file
system
and
other
things
that
wazzy
enables,
but
being
able
to
basically
do
that
without
having
to
actually
provide
a
file
system,
provide
something.
That's
a
virtualized
file
system.
A
Bailey
we
had
a
question
on
YouTube
which,
by,
if
I
understand
it
right,
they're
asking
well
I,
wanted
to
see.
If
you
would
consider
opening
up
that,
you
know
spreadsheet.
Basically,
that
you
have
in
you
know,
doesn't
have
to
be
now
or
in
its
current
form,
but
basically
turning
that
into
a
public
thing.
Just
your
notes
around
what
makes
what
makes
it
easy
to
turn
into
a
component
versus
hard
and
then
really
good
candidates
for
components.
A
That
would
be
a
really
great
thing
to
have
kind
of
as
a
reference,
especially
for
the
the
bacon
hackathon.
D
Yeah
totally
I
I
just
started
on
it
on
it.
You
know
I
I,
create
that
I
O
and
did
a
search
for
most
popular
downloads
and
then
I
quickly
realized.
There
were
several
that
don't
make
sense
to
be
a
component
that
would
be
like
Russ
macros.
Another
example
would
be
things
that
require
a
lot
of
ffi
kind
of
work
already
built
in
there's
a
lot
of
like
ffi
style,
binding
stuff
glue.
D
So
generally
speaking,
I
guess,
if
it's
language
Goo,
not
that
one
but
I
I,
think
my
goal
was
to
try
to
get
one
good
Library
example
of
of
each
sort
of
utility
that
people
are
interested
in
and
then
I've
got
separate
sheets
for
JavaScript
and
go
I
would
like
to
add
more
as
we
have
more
languages.
But
you
know
time
time
is
short
and
we've
got
three
weeks,
but
yeah
I
encourage
anybody
if
they,
if
they
have
a
good
feel
for
it
to
drop
something
in
here.
D
The
things
that
move
from
medium
to
hard
are
ones
that
are
going
to
require
a
lot
of
the
new
Wazi
apis
to
make
it
work.
So
that
takes
a
level
of
knowledge
of
the
ecosystem
that
a
lot
of
folks
don't
have
yet
the
things
that
make
it
medium
generally
I
was
thinking
things
along
like
mini
lines
of
code,
in
looking
at
what
dependencies
it
depends
on.
So
if
it
has
a
lot
of
dependencies,
it
usually
means
you
got
to
convert
those
Upstream
dependencies,
that's
kind
of
the
that's.
D
That's
the
sliding
scale,
I'm
hoping
to
find
more
easy
ones
than
not
right
of
having
those
enumerated
so
that
anybody
can
come
in
and
and
try
something
and
get
something
contributed
back
out
to
the
community
like
the
goals
for
other
people
to
be
able
to
create
basically
GitHub,
PRS
and
and
I.
Don't
know
how
to
say
his
name:
fincole
Finn,
Cole,
slash
names.
That
would
be
great.
A
Yeah,
oh
and
now
we're
at
time
but
I'll,
say
my
one
more
comment
and
then
and
then
wrap
up
I
guess
because
this
is
too
cool,
I
I'm,
looking
forward
to
the
the
what's
the
right
way
to
say
this:
the
the
flood
of
PR's
into
libraries.
It's
like
hey!
You
just
swapped
out
this
one
thing:
it'd
be
compatible
with
the
wasn't
component
and
here's
the
pr
to
do
it.
Then
it's
going
to
be
gonna,
go
harass
all
the
library
maintainers,
okay,
all
the
fun
stuff!
D
D
I
feel
like
there's
so
much
happening
right
now
in
the
space.
So
there
is
a
special
interest
group
within
the
by
code.
Alliance,
that's
around
guest
languages
and,
and
so
guest
is
like
that
thing
that
you
compile
to
webassembly
right.
D
That's
the
guest
side
and
the
host
side
is
your
webassembly
runtime,
and
so
this
is
focused
on
getting
Upstream
languages
to
have
webassembly
component
support
and
with
go
one
is
it
122
I
believe
now
has
wazzy
preview,
One
support
for
command
modules
and
now
actively
working
on
being
able
to
support
reactor
style,
we're
very
close
to
that
also
fitting
really
well
with
with
all
the
things
that
we
wanted
to
have,
but
there
are
several
different
patterns
that
can
broadly
be
applied
to
just
about
everybody.
D
What
what
binding
Generations
like
for
Zig
and
rust
and
C,
plus
plus,
like
a
lot
of
that
works
across
them.
But
there
are
a
lot
of
other
patterns
that
work
really
well
for
languages
like
Python
and
JavaScript,
where
they
need
to
bring
in
basically
The
Interpreter
or
compile
in
The
Interpreter.
So
there's
a
lot
of
techniques
there
to
make
that
performant
and
and
not
take
up
a
lot
of
memory
and
then
now
there's
one
for
go.
So
there's
a
basically
a
go
working
group
within
that
special
interest
group.
D
That's
just
focused
on
trying
to
make
something
feel
go
native
and
not
feel
like
you
just
slapped
a
bunch
of
cffi
on
on
a
project
so
that
that's
happening
right
now
and
if
you're
interested
in
getting
go
working,
I'm
looking
at
Jordan,
please
do
join
and
give
them
good
feedback
and
Jordan's
parser's
been
getting
a
lot
of
shout
out
in
that
space
too,
because
you
know
they
want
to
be
able
to
contribute
too
and
not.
Everybody
wants
to
cut
and
rest.
A
All
right,
everyone,
I'm
gonna,
I,
think
I'm
gonna
close
this
one
down
before
Jordan
comes
in
and
talks
about
not
wanting
to
coated
rust.
Sorry
for
giving
you
a
hard
time
thanks
everybody
for
for
coming
to
this
week's
Community
call
awesome
demos
awesome
awesome
progress
in
the
webassembly
space.
It's
really
just
seems
to
get
the
these
calls
get
more
and
more
exciting
the
as
the
year
goes
on.
So
thanks
all
we're
gonna
go
ahead
and
stop
it
now
and.