►
Description
wasmCloud is a platform for writing portable business logic that can run anywhere from the edge to the cloud, that boasts a secure-by-default, boilerplate-free developer experience with rapid feedback loop.
A
B
So
if
it's
inaccurate
don't
be
offended,
I'm
just
getting
it's
called
up,
but
everyone's
familiar
with
this
we've
added
this
community
meeting
drop
down
that
now
it
looks
a
lot
like
the
repo
that
liam
had
running
it,
tells
us
when,
where
and
and
whatnot
but
we're
gonna
starting
a
blog
here
and
we're
gonna
start
keeping
the
minutes
and
then
on
each
page,
I'm
hoping
to
have
the
actual
video
this
attendance
thing
off
of
github
handle.
So
one
of
the
things
I
need
from
everyone
is
so
I
don't
have
to
go
manually.
B
B
I
will
start
keeping
the
minutes
here
and,
as
you
can
see
in
the
bottom
left,
there
are
going
to
be
time
stamped
to
the
actual
youtube
videos,
so
people
can
go,
find
topics
a
lot
quicker
and
then,
at
the
same
time
we'll
be
tagging
them
so
that
the
search
feature
on
the
website
actually
works.
To
like
find
like
you
know,
you
know
if
we
want
all
the
demos,
the
demo
tag
should
pull
up
all
the
videos
with
like
demos
yeah
once
we
actually
get
it,
you
know
and
then
the
goal
is.
B
B
This
is
in
a
branch
called
community
on
the
wasmcloud.com
tac
dev
repo.
So
anything
that
you
think
is,
you
know,
maybe
change
a
little
bit
or
I
mean
I
have
to
work
on
these
fonts.
I
don't
really
know
how
to
do
that
in
hugo,
yet
to
make
it
a
little
more
readable,
but
whatever
will
make
it
easier
for
folks
to
find
you
know
what
we
talk
about
in
these
meetings
as
opposed
to
having
to
like
trudge
through
the
videos
manually
so
yeah.
That's,
that's!
Really
it
and.
B
A
B
B
C
B
A
C
C
Well,
I
think
he
goes
actually
nice
because
it
gives
you
a
lot
more
capabilities,
but
it's
more
like
if
you
want
to
do
raw
browsing
or
actually
searching
searching
for
documentation
inside
of
like
markdown,
might
be
difficult
depending
on
how
you
actually
template
it.
So
that's
a
consideration
when
you're
doing
this
tahito
website.
I
think.
D
C
D
E
A
A
B
D
Okay,
so
I
just
wanted
to
present
something
real,
quick
with
some
ideas.
I've
had
so
people
who've
been
in
this
caller
for
a
while
know
that
I
work
on
other
projects
in
the
web
assembly
space
as
well,
and
one
of
those
is
bindle,
which
is
a
storage
system,
that's
explicitly
designed
for
webassembly
that
can
easily
be
used
for
other
things,
and
I
wanted
to
demo
a
possibility
of
how
I
envision
this
working
for
before
wasmcloud
and
in
particular
with
provider
archives.
D
D
D
Okay,
so
let
me
go
ahead
and
increase
my
font
size,
real,
quick,
okay,
so
a
bindle
is
expressed
in
something
called
an
invoice,
and
these
invoice
have
have
a
strong
versioning.
That
is
very
explicit.
It
cannot
be
modified
after
it's
created
and
they're
signed.
I'm
not
going
to
go
too
much
into
the
benefits
of
it,
but
I
just
want
to
show
you
what
this
looks
like.
D
So
I
just
mocked
up,
for
example,
a
wasm
cloud.
The
wasmcloud
kb
redis
store
with
this,
so
I
actually
took
the
existing
kb
redis
provider
and
put
it
into
here.
I
can't
actually
load
this
into
a
wasn't
pod
host
right
now,
because
it's
functional
loading
as
a
file
is
tied
exclusively
through
the
web
dashboard,
but
we're
going
to
pretend
that
we
could
do
that
right
now,
just
to
show
the
example,
so
this
looks
really
complicated,
but
there's
just
a
couple
things
I
want
to
point
out.
D
But
these
parcels
are
individual
files
that
make
up
the
the
entire
whole
bundle
together
and
so
you'll
see
that
each
of
these
are
the
actual
binaries
for
the
providers
for
each
of
the
systems
and
they're,
grouped
into
specific
things,
which
I'll
explain
that
in
a
second
and
given
specific
features
that
allow
us
to
enable
or
turn
them
on
or
off.
And
so
when
we
go
down
to
the
bottom
of
it.
D
D
You
can
also
see
this
is
signed
by
me
who
signed
it
first
and
then
by
the
host
key
of
the
server
that
I
sent
it
to,
which
is
just
a
logo.
So
I
can
actually
go
and
validate
these
I'm
not
doing
that
in
this
process,
but
every
bundle
server
that
hosts
it
is
supposed
to
sign,
and
so
you
can
track
actually
where
the
bindle
has
been
before
it
hits
your
machine.
That's
an.
D
Benefit
that,
I
don't
think,
is
the
main
thing
here,
but
wanted
to
mention.
So
what
we
can
do
with
all
this
is.
We
can
then
select.
I
I
created
some
fake
test
binaries
here,
as
if
we
wanted
to
like
be
able
to
mock
out
the
key
value
store.
So
that
way
you
can,
when
you're
running
an
application,
you're
trying
to
run
an
actor
and
you
need
a
key
value
store.
You
can
say
I'm
just
running
this
in
test.
D
So
let
me
go
ahead
and
fetch
that,
and
it
will
it's
going
to
figure
out
this
os,
that
it
needs
the
specific
os
and
the
specific
thing,
and
it's
going
to
use
those
features
that
I
showed
up
here
where
it's
going
to
say.
Okay,
I
have
this
architecture
and
this
os
and
it's
going
to
filter
and
then
it's
going
to
only
pull
the
parcels
it
needs.
D
So
if
we
look
here,
you'll
see
that
it
figured
this
out
and
said,
I
need
to
pull
the
claims
because
that's
required
and
I
always
need
it
and
I'm
going
to
pull
the
x86
64
for
mac
os
and
now
it's
only
pulled
those
things.
So,
instead
of
pulling
this
huge
tarball
down
from
oci,
we're
now
saving
bandwidth
and
being
able
to
select
these
features.
D
So
the
the
first
major
thing
that
I
want
to
point
out
is
the
saving
of
bandwidth,
which
is
very
important
for
one
of
the
target
audiences
for
wazonfad,
which
is
edge
devices,
and
so
this
makes
it
so
you're
only
pulling
down
the
specific
bites.
You
need
and
not
the
whole
kitchen
sink
so
that
pulls
it
all
down.
So
that's
the
first
thing
now.
The
second
thing
is
what
I
mentioned
about
the
test
step.
Let's
say
I
wanted
to
test
something,
so
I
could
say
no
hey,
I'm
in
test
mode.
D
So
please
just
give
me
like
something
that
works.
I
don't
want
to
set
up
redis.
I
don't
want
to
set
up
that.
I
just
need
test
things,
and
so
I'm
going
to
just
pretend
that
like
there's
something
there,
but
I
would
specify
hey,
I
want
test-
and
in
this
case
I'm
enabling
the
test
group
rather
than
the
normal
group,
and
so
when
I
do
this
you'll
see
that
instead,
I
fetched
the
test
mac
os
binary,
plus
the
normal
claims
for
the
provider,
and
once
again
I
only
fetched
the
necessary
bits
and
nothing
more.
D
We
also
add
this
in
terms
of
storage
capabilities,
because
each
of
these
shas
is
unique
for
the
file
and
so
that
sha
will
be
used
across
things.
So
if
you
have,
if
we
have
20
different
key
value
providers,
we
could
still
have
just
one
set
of
mocked
key
value,
things
that
are
listed
in
every
in
every
bundle,
but
are
only
stored
once
on
the
server
and
then
fetched
accordingly.
D
So
anyway,
super
simple
demo.
I
was
really
hoping
I
could
actually
tie
it
into
the
wasm
cloud
host,
but
it
turns
out
you
can't
do
that
right
now,
but
this
is
something
that
I
think
would
be
a
much
better
and
more
efficient
way
for
how
we
do
things
inside
of
wasm
cloud
hosts
and
allow
us
a
lot
more
flexibility
and
having
to
say,
because
you
can
see
you
can
glue
together
all
sorts
of
complicated
features
and
things,
and
then
these
features
and
annotations
can
actually
be
used
to
schedule
things
smartly,
even
in
the
future.
D
F
I
had
I
had
one
question
or
my
first
thought
is
so
I
like,
I
see
that
you're
fetching
it
from
like
wasmcloud.dev,
slash,
kb
redis.
You
know
0
14.4.
That
may
just
be
like
a
mocked
up
thing.
How
do
you
actually
host
these
bundles
and
and
how
do
you
or
the
yeah,
how
do
you?
How
do
you
host
the
bundles?
Is
it
like
in
an
oci
registry,
do
you
do
you
launch.
D
D
D
Dev
group
could
push
invoices
up
to
that
specific
path
and
that's
that's
the
reason
that
exists,
so
the
storage
back
and
right
now
I
am
running
this
locally
and
this
is
using
an
embedded
database
locally,
but
the
bindle
has
something
called
a
provider
and
the
provider
and
not
to
be
confused
with
capability
providers
and
the
bundle
providers
can
be
implemented
to
go
against
any
storage
system
you
want.
Before
I
had.
I
left
microsoft.
D
So
if
you
want
to
have
like
a
whole
network
of
bundle
servers,
you
could
do
that
and
these
these
bundles
can
be
signed
by
people
who
review
them.
So
you
could
have
security,
people
review
them
and
then,
when
you
pull
down,
you
can
say
I
want
to.
I
want
to
make
sure
that
someone
who
I
trusted
has
verified
this
and
you
can
verify
it
cryptographically.
That,
like
is,
has.
A
Yeah
one
of
the
things
I
think
that's
been
kind
of
central
to
our
thesis
for
engaging
enterprise
customers
is
ensuring
that
the
things
that
we
build
are
compatible
with
where
we,
where
they
are
today,
how
does
this
technology?
How
does
this
standard
play
in
the
face
of
existing
artifactory
repositories
like
artifactory,
for
example,
from
jfrog?
You
know
things
that
you
would
we
typically
find
in
customers
places
that
are
hosting
containers
and
images
and
everything
else
is
this
compatible
with,
or
would
this
need
to
be
supported
by
it.
D
It
is
not
compatible
with
as
it
stands.
There
is
nothing
we
actually
taught.
Those
of
us
who
worked
on
bundles
have
talked
about
it
quite
a
bit
of
putting
a
like
oci
shim,
either
in
front
of
it
or
behind
it.
That
could
be
easily
doable,
especially
to
like
glue
it
in.
You
could
have
it
be
like
a
proxy
server
that
something
pushes
off
to
so
that
they're
accessible
as
oci
artifacts
and
as
this,
but
the
bindle
was
created
as
something
new,
because
oci
really
didn't
completely
fit
the
bill
for
what.
D
Should
look
at
doing
in
the
near
future
and
seeing
what
we
can
do
to
make
things
compatible
right
now
with
meeting
people
where
they're
at
and
kind
of
leading
them
and
giving
them
a
way
to
get
to
the
future?
And
so
I
think
one
of
those
is
like
an
oci,
a
layer.
We
put
in
front
of
a
bundle
server
that
converts
an
oci
thing
over
to
a
bundle,
and
then
they
can
still
access
it
through
it
as
an
oci
artifact
or
they
can
then,
or
they
can
slowly
switch
over
to
bindle.
D
A
As
a
as
a
distribution
mechanism,
I
mean
it's
definitely
a
powerful
concept
because
we
think
about
who's
life.
This
is
making
simpler.
It's
the
people,
creating
you
know,
distributable
components
that
they
would
want
to
generate
and
deploy
as
like
a
version
so
like.
Let's
say
we
released
wasmcloud
and
correct
me
if
I'm
wrong
here,
I'm
kind
of
restating
some
of
this
in
language.
D
D
You
can
imagine
as
people,
let's
just
say:
let's,
let's
give
an
example
here,
let's
say
I'm
building
the
business
logic.
For
my
for
my
business
and
as
a
user,
I
would
want
to
say
okay
in
testing.
I
want
to
do
this,
so
I
just
demonstrated
that,
but
also
let's
say
I
have
edge
devices
and
I
have
big
beefy
compute
devices,
and
so
I
have
an
actor
that
can
take
advantage
of,
let's
say
graphics
cards
and
do
ml
pipeline
stuff.
D
It
still
returns
the
same
results
but
like
let's
say
it's
like
a
machine
learning
thing
about
recommending
restaurants,
so
I
could
get
completely
accurate,
up-to-date
results
running
these
machine
learning
models
on
beefy
machines,
but
let's
say
I'm
just
trying
to
get
it.
I'm
also
trying
to
give
it
out
to
a
user
who
might
be
in
out
somewhere
in
more
rural
where
there's
only
like.
You
would
be
running
this
on
an
edge
distribution,
node
somewhere,
and
so
then
what
you
can
have
it
do
is
like
no.
D
I
want
to
deploy
this
version
of
the
application
and
have
it
select
something
that's
more
edge
edge
specific
and
it
would
pick
that
actor
versus
the
other
actor,
but
it
all
it
would.
The
bind
allows
you
to
say.
Please
pick
one
of
these
actors,
I
don't
care
which
one
you
just
need
to
do,
one
of
them
to
satisfy
the
requirements
I
have
here
and
so
then
you
can
choose
which
one,
and
so
it
really
enables
that
down
the
roads
too.
Obviously,
that's
future,
but
that's
something.
C
A
D
I
would
imagine
just
like
constructing
in
my
head.
I
would
upload
the
oci
artifact
and
then
my
shim
would
unpack
that
tarball
or
whatever
it's
basically
the
tarp
unpack
all
the
layers,
get
all
the
individual
files
and
then
push
them
up
to
a
bundle,
server
and
then
going
in
the
reverse
way.
If
it
needs
to
fetch
something
that
it
doesn't
already
have.
It'll
go
to
the
bundle,
server,
pull
the
artifacts
down,
reassemble
it
into
layers
and
send
it
back
down.
A
It
sounds
like
the
right.
The
right
path
forward
here
to
get
this
on
a
path
for
broader
adoption
would
be
to
make
sure
that
the
bundle
format
is
like
in
the
what
the
linux
foundation
that
who
owns
the
oci
right
standard
is
that
is
that
happened
or
is
that?
Is
there
a
path
to
that
or
it
was
this.
Just
like
the
you
know,
the
cncf
donation,
stuff.
D
D
Yeah,
that's,
and
so
I
think
that,
right
now,
it's
not
specifically
in
a
foundation
because
it
is
looking
for.
I
know
that
as
a
maintainer
bender
we're
looking
for
that
feedback
from
people.
People
have
been
interested,
but
like
this,
this
opportunity
of
working
more
on
wasm
cloud
has
given
me
the
ability
to
show
a
concrete,
real,
very
useful
opportunity
right
now
and
that's.
D
We're
looking
at
for
the
community-
and
so
I
think
something
I
would
ask
from
the
community
is:
if
you
are
doing
wasm
cloud,
would
you
find
this
useful,
especially
if
you're
deploying
stuff
not
just
writing
the
actors,
but
like
deploying
things
with
it?
Would
this
be
a
useful
technology?
What
are
the
rough
edges?
D
What
are
those
things
that
the
the
reason
being
we
want
to
solidify
that
in
bindle
before
we
ever
push
that
further
out,
we
want
to
make
sure
it's
solid
and
built
out,
so
this
is
what's
being
used
in
some
of
the
deus
labs
projects
right
now
like
hippo,
but
it's
also
something
I
think
many
people
could
leverage.
It's
designed.
A
G
C
G
We
deploy
a
capability
provider
right
now.
It
has
to
be
on
the
same
host
as
the
western
cloud
host,
which
is
which
means
you
have
to
do
finagling
if,
if
it's
running
in
docker,
but
the
capability
providers
need
configuration
files,
sometimes
we're
starting
to
get
more
complex
capability
providers,
and
so,
if
there
were,
maybe
we
could
pass
the
capability
provider
on
launch
the
path
to
the
bundle
file,
and
then
it
could
load
say
the
settings
at
these
or
sample
data
sets
and
and
load
them
there.
A
All
right,
no,
I
I
agree
steve.
I
think
there
are
some
compelling
use
cases
for
this
and
it's
a
good
idea.
I
think
the
you
know
there
there's
like
the
wasm
cloud
perspective
is,
is
that
you
know
this
is
this
could
be
a
good
distribution
mechanism
for
us
and,
I
think,
there's
a
bigger
motion
here
on
you
know:
how
do
we,
you
know,
put
a
thumb
up
and
say
like
if
this
were
a
standard
for
everybody
to
be
even
better?
A
F
Taylor,
just
a
just
a
statement.
I
think
the
you
know
it's.
It's
maybe
easy
to
gloss
over
the
first
point
that
you
made
about
pulling
down
only
the
resources
that
you
need,
but
as
we
continue
to
grow
like
our
list
of
targets,
you
know,
as
as
we
wait
for
like
a
better
mechanism
for
a
capability
provider.
D
Yeah,
you
could
even
stream
back
actors
into
interpreted
runtimes
and
then
run
them
through
like
that.
There's
there's
crazy
things
you
can
do,
but
I
think
also
like,
I
always
think
of
that
space
thing
in
terms
right
now.
It's
fine,
but
as
soon
as
docker
hit
even
a
little
bit
of
mainstream
people
were
shoving.
1.5
gigabyte
containers
into
things,
and
so
someone
will
do
that
with
wasmcloud.
D
A
A
A
I
I
dropped
a
link
into
chat
here
and
we
are
planning
a
fun
little
set
of
demos
here,
where
we're
going
to
be
demonstrating
all
the
new
features
and
a
new
hero
demo
that
we've
been
putting
together
around
the
java
pet
store
app.
If
you
like
the
love
my
little
art
there,
I
wanna
start
doing
a
little
a
poster
for
every
major
release.
We
do
that
comes
out
cloud
native
wasm
day
is
october,
the
12th,
so
we've
got
a
great
agenda
planned
for
this.
A
If
you
haven't
signed
up
yet
you
can
see
the
schedule
right
online,
but
I
think
we've
got
a
ton
of
great
speakers
and
talks
again
and
we've
tried
really
hard
to.
You
know
pull
in
just
a
range
of
interesting
content
across
the
webassembly
community,
not
just
blossom
cloud
here
and
just
another
call
out
on
wednesday
october,
the
13th
I'll
be
speaking
at
kubecon.
A
I
will
be
live
talking
about
server
side
webassembly
with
blossom
cloud,
and
this
is
going
to
be
the
sort
of
like
total
relaunch
of
our
0.50
experience.
What's
our
goal,
our
mission
statement
and
and
all
of
those
kind
of
things,
are
there
any
other
community
events
that
we'd
want
to
call
out
or
call
attention
to
at
this
time.
D
If
you're
in
the
cloud
native
space,
there's
a
lot
of
really
interesting
talks
that
get
rejected
from
kubecon
for
any
variety
of
reasons-
and
this
is
like
a
great
great
conference-
it's
all
virtual
this
year,
which
still
makes
me
a
little
sad
because
it's
a
really
fun
one
to
go
to
in
person.
But
it's
definitely
worth
going
to
and
just
kind
of
meeting
people
and
seeing
some
of
the
interesting
talks.
A
D
C
A
Yeah,
I
will
be
there
as
well.
I
fly
out
the
weekend
before
and
we'll
be
there
through
the
weekend
after
so
super
exciting.
I
hope
to
see
you
guys
there
at
all
the
events
and
at
this
point
steve
I
don't
know
if
we
turn
over
to
you
for
road
map
review,
I
mean,
I
guess,
with
0.050
we're
done.
We
can
just
wrap
this
up
and
I'll
head
home
right.
G
Yeah
we're
good
we're
in
good
shape,
maybe
brooks
might
want
to
talk
about
some
of
the
the
packaging.
This
is
available
in
the
various
distribution
points.
We
are
now
working
on
a
demo
for
coupon
you've
seen
the
the
pet
shop
pet
clinic
in
the
examples,
repo
and
it's
still
a
work
in
progress
and
then
putting
together
plans
4.51
so
again,
as
always,
we'd
love
to
hear
feedback
on
on
what
you're
working
on
and
what
you're
running
into
as
we
prioritize
the
dot.
51
that'll
be
a
big
help
to
us.
A
A
Okay,
yeah,
I
mean
if
we
just
have
like
other
calls
to
action.
I
don't
know
what
sort
of
engagement
we
have
on
that,
but
I
think
it
is
helpful
just
to
put
it
out
there,
maybe
even
just
or
just
make
sure
to
include
the
call
for
you
know,
feedback
and
stuff,
like
that.
I
know
with
the
dot
zero
five
o.
I
know
we
are
I'm
sure
there
are
bugs
and
opportunities
to
improve
the
documentation.
A
So,
as
everybody
runs
through
the
guides
and
everything,
I
really
appreciate
pull
requests
and
you
know
tickets
and
stuff.
Like
that,
it's
great,
you
know,
we
know
we
hit
the
we've
got
the
bug
on
mac.
Still,
I
think
we
got
to
get
the
package
signed
so
hopefully
we'll
get
that
that
cleaned
up
pretty
pretty
quickly
great
any
other.
G
Yeah
I'll
add
one
more
one.
More
specific
thing
in
the
request
for
feedback
is
as
part
of
this
release
and
using
smithy
for
the
interface
generation
and
code
generation.
We
prioritize
rust
as
a
language
de
facto
for
actors
and
capability
providers,
but
we're
also
interested
in
hearing
what
languages
you
guys
want
to
work
with,
and
the
platforms
you're
working
on
the
kinds
of
capability
providers
that
you
would
find
most
useful,
because
those
are
all
all
things
going
into.
The
priority
bucket.
A
C
Aisha
and
david
we're
gonna
turn
up
today
and
do
a
demo
for
it.
I
so
they
had
some
outstanding
issues.
I
will
not
steal
that
thunder
by
doing
the
demo
for
them
and
I
just
opened
a
pr
to
get
to
some
of
their
feedback.
Mostly,
I
think
it's
in
an
mvp
state
as
long
as
the
lattice
controller
actually
comes
along,
so
it
is
usable
and
it
doesn't
do
something,
but
there
are
parts
that
are
missing,
as
people
will
probably
push
it
and
find
out.
What's
going
on.
A
A
A
All
right:
well,
that's
awesome
everybody.
We
we've
been
building
up
to
this
kubecon
for
a
while,
so
I'm
really
excited
for
the
next.
You
know
just
two
weeks
of
of
content
and
fun
and
for
anybody
that's
gonna
be
out
in
la.
Let
me
know:
if
not,
I
will
talk
to
everybody
later
all
right,
bye.
Everybody
have
a
great
week.