►
From YouTube: TAG General Meeting
Description
TAG web site: https://tag-app-delivery.cncf.io/
TAG Slack channel: https://cloud-native.slack.com/archives/CL3SL0CP5
TAG git repo: https://github.com/cncf/tag-app-delivery
TAG meeting notes: https://docs.google.com/document/d/1OykvqvhSG4AxEdmDMXilrupsX2n1qCSJUWwTc3I7AOs/edit
A
B
Yeah,
so
it's
1003
I
guess
you
know
I'll
just
start
sharing
and
talk
a
little.
So
one
of
the
things
that
I
was
inspired
at
cubecon
from
the
speeches
and
stuff
was
to
see
how
we
can
bring
people
in
here
and
and
it's
very
Bottoms
Up.
You
know
so
it's
up
to
them
to
kind
of
Define
what
they
work
on
to
some
degree,
but
like
help
people
on
board
a
little
bit.
B
B
B
So
if
folks
are
okay
with
that,
if
we
could
get
started
and
I
would
just
ask
if
people
you
know
I
know,
I'm
I,
don't
know
everyone
I'm
sure
people
have
been
here
before,
but
if
people
that
want
to
introduce
themselves
take
a
second
and
and
tell
us
who
you
are
and
and
what
you
are
most
interested
in
I
guess
what
brings
you
here.
D
I
can
go
first
so
on
the
middle
back.
Mark
I
have
not
been
on
these
meetings
before
so
we
met
on
some
of
us
met
on
cuecon
two
weeks
ago,
so
I'm
involved
in
this
series
initiative:
Under
The
Continuous
review,
Foundation
I'm,
one
of
the
founders
of
that
project.
Together
with
Andrea
treaty,
we
will
join
anytime.
E
Hello,
I'm,
daddy
and
I
work
for
Apple
I'm
on
the
Spinnaker
technical
oversight
committee
and
I
got
involved
with
the
CDF
about
this
time
last
year,
really
looking
into
interoperability,
my
team
also
pops
a
little
bit.
You
know
we
attend
the
city
of
instig
as
well,
and
we're
just
looking
to
take
advantage
of
that
works
here.
B
Cool
nice
to
have
a
lot
of
CD
events
folks
here,
thanks
for
bringing
everyone,
I
see,
Justin
I
see
two
hands,
so
you
know
jump
in
and.
F
Sorry,
hi
I'm,
Justin,
comac
I'm,
then
I'm.
Now
the
cncf
liaison
for
this
tag
I've
been
on
layers
on
for
tax
security
for
a
long
time
now.
Actually
so
it's
nice
to
have
a
change
I'm,
the
also
the
CTO
at
Docker
and
so
really
interested
in.
F
You
know
the
application
side
of
things
in
that
you
know
kind
of
having
more
developer
focus
in
cncf,
which
has
been
kind
of
a
very
Ops
focused
in
many
ways
and
I
think
that
many
of
the
things
that
this
tag
you
know,
fall
in
the
room
matter
of
things
that
I'm
really
interested
in
I've
and
and
the
interfaces
like
things
like
get
options.
C
B
F
B
Something
yeah
cool
I
like
that.
You
mentioned
app
development
in
our
group,
I
I
think
that's
the
default
Landing
place
for
right
now
for
those
kinds
of
projects
is
with
us:
yeah
yeah
I've,
seen
I'm
just
going
with
what
I
see
on
my
screen.
C
My
guys
so
I'm,
just
seeing
I'm
based
in
France
first
time
in
this
meeting
and
I've,
been
able
to
met
some
of
you
guys
at
kubecon
a
few
weeks
ago
and
I'm
walking
and
I'm.
One
of
the
maintenance
of
the
project
which
is
named
Mike
rocks
dedicated
to
API,
mocking
and
testing,
and
we
have
submitted
to
join
as
a
Sandbox
project
so
waiting
passionately
in
a
queue
and
happy
to
to
have
a
discussion
with
you
guys.
B
B
Time
to
the
folks
that
weren't
at
the
kubecon
Meetup.
G
Hi
everyone
I'm
just
working
at
the
New
York
Foundation,
leading
to
continue
Foundation
and
similar
to
any
of
this.
The
first
time
I'm
in
this
meeting
but
I've,
been
following
the
work
I've
been
doing
over
the
last
couple
of
years
and
we
actually
came
and
presented
Potato,
Head
and
other
instances
you
have
within
the
tech
application
to
R6
and
over
the
past
couple
of
years,
I've
been
contributing
interability
area
before
I
joined
building,
Standish,
so
I'm
here
to
see
and
discuss
the
potential
collaboration
opportunities
between
tiger
pillory
and
C
demands
project
thanks.
H
Yeah
I'm
a
database
guy,
so
I've
been
trying
to
joining
various
open
source
communities
to
learn
as
Flamingo,
because
the
database
I
work
on
is
not
an
open
source
database.
H
So,
after
a
couple,
several
months,
I've
now
started
contributing
in
anything
database
related,
including
database
security
database
on
kubernetes
and
then
as
a
supply
chain,
security
that
this
area
I'm
interested
in
mostly
interest
by
now
when
it
comes
to
this
group,
also
learning
I,
said
I
think
I
think
it
would
Josh
to
I
want
to
contribute
to
the
the
translation,
localization
and
that's
one
way
of
my
learning.
Oh
just
one
comment
about
that.
They
are
in
the
paper
at
the
bottom
of
there.
H
There
is
a
matrix
of
tools
just
for
new
newbies.
It
would
be
great
to
I,
don't
know
where,
but
to
have
some
even
go
deeper
into
like
each
group
of
tools
where
they
are
and
what
the
difference
is,
because
it
I
I
just
want
to
it's
great
to
have
list
them
now.
It's
already
helpful
it'll
be
great.
If
you
just
go
into
deeper
in
like
a
diagrams
or
discussions.
B
Thanks
Victor
yeah,
actually
I
just
pasted
in
that
there
is
an
issue.
This
came
up
in
the
last
platforms
working
group
meeting
and
a
couple
other
discussions
in
kubecon
rationalizing
the
list
of
capabilities
with
the
landscape,
with
the
the
coverage
of
the
tags
so
yeah
and
and
also
going
into
each
one
of
those
has
also
come
up
in.
There
are
different
issues
in
our
in
our
repo.
So
if
you're
passionate
about
one
of
them,
that's
what
we
need.
We
need
people
we
in
leading
into
kind
of
the
next
thing
here.
B
Kubecon
I
think,
like
I,
said
one
of
the
biggest
earnings
for
me
and
other
people,
chime
in
you
know
better
support,
folks
that
want
to
contribute
enable
you
to
find
your
find
your
own
way.
Like
you
know,
it's
bottoms
up,
I
I
feel
like
so
it's
to
you
identifying
what
you
know
what's
missing
and
what
you're
passionate
about
kind
of
finding
the
match,
but
we
could
help
you
we
can.
We
can
advertise
what's
available
so.
B
Think,
that's
you
know,
shifting
into
the
kubecon
recap
here
before
we
go
to
see
the
events.
I
think
that
was
the
a
big
thing.
So
that's
why
we
wrote
the
contributing
to
our
we.
We
put
up
the
contributing
dock
with
four
or
five
ways
that
people
could
contribute
check
that
out.
Add
to
them.
If
you
have
or
ask
questions,
I
did
know
yeah.
B
Oh
thanks,
whoever's,
sharing
that
I
did
know
who
that
one
of
the
ideas
we
came,
we
discussed
in
the
meeting
which
I
wanted
to
just
kind
of
bring
up
again
was
because
we
talked
about
our
prototypes
like
potato
head
and
maybe
we'll
build
a
platform
prototype
and
one
of
the
things
was.
How
are
we
going
to
maintain
those
we
don't
really
want
to,
or
we
don't
want
to
guarantee
we
can,
but
we
still
want
people
to
come
and
tell
us
about
the
things
they've
built
so
open
an
issue
in
GitHub.
B
B
Another
thing
Abby
asked
me
to
mention
this.
We
are
in
the
platforms
group
starting
up
work
on
a
second
follow-on
paper.
This
is
like
tactics
to
act.
You
know
the
first
paper
was
kind
of
sell
you
on
the
idea
of
a
platform.
The
second
one
is:
how
do
you
measure
your
success
getting
deeper
in?
What
are
the
tactics
so
she's
organizing
a
meeting
next
week?
If
you
want
to
participate?
B
Another
big
discussion,
I
think
a
kubecon
is,
was
the.
How
did
tag
support
the
technical
oversight
committee?
There's
been
a
discussion
forever,
it
feels
like,
but
definitely
you
know
looking
over
the
sandbox
projects
and
things-
and
you
know
now
we
have
Justin
here.
So
that's
great
help
us
help
guide
us.
I
F
Yeah,
it's
not
yeah,
it's
not
not
officially
yeah,
but
it
should
be
soon,
but
yeah
I
mean
I.
Think
we've
always
there's
always
been
a
kind
of
trying
to
get
the
right
balance
and
I
think
that
the
I
think
we've
gone.
We
there's
definitely
a
feeling
that
there's
there's
definitely
the
the
balance
has
gone
too
much
back
to
sort
of
centralizing
things
in
the
COC
which
is
leading
to
bottlenecks,
which
is
a
bad
thing
about
having
a
better
role
for
the
tags.
F
C
F
Has
the
same
problem,
so
it's
kind
of
I
think
that
you
know
I,
think
that
you
know
I
think
we
need
to
get
that
balance
right
and
and
I
think
that
yeah,
the
working
with
the
project
in
particular,
is
something
that
we
still
haven't
got
right.
So
there's
there'll
be
something
soon.
I
I
B
B
No
that's
great,
and
one
of
the
things
that
came
up
kind
of
informally
in
discussions
was
this
new
at
least
the
the
new
Sandbox
submissions
are
all
in
one
place.
So
if
anyone
wants
to
go
in
there
and
identify
you
know,
projects
that
are
relevant
to
us
and
either
ask
them
to
come
present
like
Yasin
is
going
to
do
or
even
if
you
know
you
evaluate
it,
and
you
want
to
share
for
five
or
ten
minutes.
Oh,
this
is
the
new
idea
from
that
project.
You
know,
that's
a
that's
a
Bottoms
Up
way.
I
That's
actually
quite
a
good
idea,
just
because
sometimes
it's
not
completely
clear
which
project
the
tag
might
map
to,
and
vice
versa.
So
you
know,
as
a
representative
of
this
community,
you
could
say
hey.
This
would
be
a
great
thing
to
come
and
show
to
observability
or
to
to
app
delivery.
For
example,.
B
Yeah
yeah
and
you
never
know
what
ideas
will
come
out.
Kpt
came
in
like
a
couple
weeks
before
kubecon
and
I,
one
of
the
Google
guys
came
and
Google
people
came
and
presented
on
at
one
of
our
lightning
talks
and
like
an
idea
that
they
had
around
converging
on
that.
Whatever
I'm
not
gonna,
get
give
the
idea,
because
then
we'll
just
cause
the
discussion,
but
it
was
like
so
helpful
and
yeah.
B
B
A
As
a
tech
now
is
that
we
didn't
really
used
to
look
at
sandbox
projects
that
much
but
obviously
they're
not
nicely
available
and
at
least
inviting
them
to
present
and
having
a
discussion,
I
think
makes
more
sense
and
also
getting
a
dealer
than
people
from
other
projects
to
join,
because
very
often
the
outcome
is.
You
should
be
talking
to
this
project.
You
should
be
talking
to
this
project
as
well,
so
this
is,
and
then
we
did
it
before.
A
Kubecon
like
active,
reaching
out
to
people
to
projects
are
currently
submitting
and
they
were
very
but
very
positive
feedback
Josh
and
they
joined
in
so
I.
Think
that
was
good
and
we
should
keep
doing
this.
Definitely.
B
Cool
the
last
item-
I
guess
here
is
feedback.
You
know
we
can
discuss
a
tiny
bit
now
or
offline
or
in
the
slack,
but
you
know
we
want
to
know
what
you
all
thought
about
the
lightning
talks
about
having
a
boost
you
know,
I
can
share
my
thoughts,
but
that's
more
years.
How?
Well
you
thought
about
the
session.
I
A
I
think
this
goes
back
to
like
overall
organization
is
that,
like
everything
is
like
cram,
what's
the
black
cramped
into
this
one
day
and
it
was
I
mean
luckily
Josh-
you
won't
cheat
to
derive
a
lot
of
this,
but
there
was
so
much
going
on
in
parallel
on
that
one
day
for
for
tax
and
project
that
it
was
I
mean
we
had
like
very
great
attendance
like
a
robust
packed,
but
still
it
was
like
a
lot
going
on
in
parallel
and
I
know
with
more
and
more
happening.
A
Obviously,
at
kubecon
this
won't
get
easier,
but
it
is
a
challenge
for
people
I
think
to
join
two
sessions.
A
F
One
of
yeah
I,
think
I
mean
it's
a
big
problem
generally
I
mean
I
think
that
what
one
I
mean
tax
security
ended
up.
Having
started
starting
to
organize
events
outside
cubecon
as
well,
in
order
to
have
more
space,
so
I
think
that
you
know
there
are.
There
are
potentially
other
options.
I
know
that
I
mean
obviously
kubecon.
Has
you
found
that
everyone's
there?
F
F
For
all
preferences,
as
well
I'm,
not
sure
I,
don't
know
yeah
I
think
the
as
you
most
of
actually
most
of
it
was
one
day
announced
that
wasn't
too
much
of
it
was
on
the
Tuesday.
Hardly
anything
on
Monday
yeah.
H
F
B
B
It's
an
interesting
suggestion
about
tax
security,
because
next
week
is
you
know,
cdfcon
and
get
Ops
con,
and
maybe
you
know
app.
Delivery
should
focus
on
that.
We
were
kind
of
the
representative.
A
little
get
Ops
group
and
a
flux
beat
up
I,
think
but
I
kind
of
I
I
don't
know
I
kind
of
feel
like
we're
all
just
representing
pretty
similar
ideas.
So
maybe
we
should
emphasize
that
you
know
the
the
CD
conference
or
the
get
Ops
conference
like
security
is
doing.
Maybe
that's.
A
Yeah
and
I
think
also
the
boosters
are
very
important,
like
we
had
it
this
year.
Well,
we
had
one
last
time
as
well,
but
this
one
I
think
it
was
like
very
crowded
and
Joshua
spent
those
a
lot
of
time.
There
I
think
that
should
be
really
planned,
that
the
tags
have
dedicated
Booth,
where
people
can
be
like
right
now
we're
kind
of
like
competing
a
bit
with
the
projects,
but
people
did
show
up.
People
did
engage
more
than
they
did
in
the
past,
so
that
is
definitely
helpful.
A
B
Yes,
Abby
was
very
helpful.
Somebody
came
up
to
me
from
the
tax
security
conference
because
the
or
the
security
group
had
like
a
security
Village
at
the
person
that
came
up
to
me.
I,
don't
know
I
wasn't
there,
but
they
were
asking
like.
Maybe
we
can
partner
next
time
and
do
something
like
absolutely
did
here
in
the
same
area,
because
I
think
the
security
Village
was
far
removed
and
maybe
they
weren't
getting
as
much
interaction
as
they
wanted,
and
but
it
seemed
like
we're
both
pursuing
the
same
thing
like
we
want
to
be
present.
B
A
B
A
Cd
events
yeah:
we
have
first
of
all
welcome
of
the
City
events
and
CVS
people.
I
think
this
is
a
great
collaboration.
It
should
have
happened
quite
some
time
ago.
A
Obviously,
that's
a
very
good
overlap
on
the
work
at
the
City
events
that
people
are
doing
on
standardizing
all
the
communication
between
the
different
Tools
around
continuous
delivery,
which
is
closely
related
to
app
delivery,
I
think,
despite
us
being
in
different
foundations,
it's
a
very
good
point
to
collaborate
on
this,
especially
as
there
is
now
also
semantic
conventions,
work
Hotel,
semantic
convention
work
going
on
regarding
standardizing
metadata
for
continuous
delivery,
so
I
think
we
were
I
know
that
I
was
a
CD.
A
Foundation
has
a
wider
scope
than
just
Cloud
native,
but
I
think
it's
very
important
that
we're
aligned
on
reusing
work
that
has
already
been
done
there.
So
it's
a
great
great
to
see
the
additional
momentum
there.
I
know
the
City
events
has
been
around
for
a
while,
that's
also
quite
actually
to
be
people
from
Apple
and
others
not
not
joining
in
there
and
the
great
for
the
team
has
been
doing
so.
First
of
all,
I
think
I'll.
A
Give
everybody
here
a
chance
to
get
up
to
speed
on
what
you're
working
on
and
then
really
finding
out
how
we
can
collaborate
together
here
best
yeah
and
I
think
was
this
a
passive
ear
over
to
Andrea
or
fatty,
where
I
want
to
take
it
from
here.
J
Yeah
I
mean
I
I
could
give
a
a
quick
intro
to
to
see
the
events.
I,
don't
know
if
everyone
is
familiar
with
the
project.
Does
it
make
sense
like
got
a
couple
of
slides,
I
use?
Often
I
can
go
through
those
or
I
can
just
talk.
F
J
J
Okay,
can
you
see
this
the
slide.
J
Okay,
cool
yeah,
thanks
for
the
introduction
always
so,
as
you
were
saying,
I
mean
we
in
the
CDF.
We
have
this
CD
events
project
and
the
goal
really
is
interoperability,
so
we
want
to
have
a
common
language
for
different
tools
in
the
continuous
delivery
space,
so
they
can
communicate
for
the
events
seamlessly
and
yeah
and
I
agree
what
you
said.
Although
he
said
there
is
quite
a
good
deal
of
overlapping
with
what
the
app
delivery
tag
that
seems
to
be
covering
in
scope.
J
So
I
think
there
is
a
great
potential
for
collaboration.
So,
just
briefly,
this
is
I.
Think
the
the
image
that
the
picture
I
wanted
to
show:
that's
it's
not
100
up
to
date
and
we're
expanding
a
bit
scope
but
to
to
get
to
get
an
idea.
So
if
you
consider
like
a
traditional
python,
where
you
start
from
code
on
the
left
and
you
do
build
testing
signing
you
upload
some
artifacts
into
some
registry
and
eventually
you
deploy.
J
So
what
we
we
want
to
to
do
is
proceed.
Events
is
to
enable
communication
through
the
tools
that
live
in
the
different
areas
of
this
python,
to
communicate
for
each
other
right,
and
we
do
it
in
a
way
with
that
we're
we
want
to
use
declarative
events,
so
the
idea
is
here
is
to
have
an
architecture
where
each
component
is
sending
information
about
what
is
happening.
J
But
these
are
not
imperative
events.
So
it's
not
about
telling.
Please
start
this
pipeline,
it's
declaring
what
is
what
is
happening,
and
then
you
can
have
some
policies,
some
triggering
component
that
receives
the
events
actually
optionally.
It
collates
them
and
then
will
it
will
trigger.
Then
the
next
step
in
your
even
driven
workflow
and
I
mean
why
events
we
we
want
basically
to
take
the
benefits
of
event-driven
architecture.
So
this
makes
your
overall
City
pipeline
more
scalable,
the
different
components
decoupled.
J
It
gives
you
flexibility,
because
if
you
want
to
add
an
extra
block,
you
can
do
it
seamlessly
and
the
other
components
don't
need
to
be
necessarily
aware
of
about
it
and
it
can
be
policy
driven
and
the
other
advantage
of
using
this
kind
of
declarative
event
is
that
you
can
also
then,
send
all
your
events
to
a
common
broker,
and
then
you
can
collect
them
and
store
them
and
then
you're
accumulating
data,
and
you
can
crunch
this
data.
Basically,
you
can
build
metrics
out
of
it.
You
can
use
these
events
to
send
notification.
J
You
can
build
views
about
your
workflow
to
even
pipeline,
so
there
is
an
extra
block
here
to
the
right
that
is
not
displayed
in
this
feature
yet
which
is
Operation.
So
in
the
latest
specification
of
City
events,
we
added
events
for
incidents,
so
incident
generated
incident
result
of
this
kind
of
event.
So
we
want
to
extend
this.
The
scope
into
the
kind
of
continuous
operation
space
as
well.
J
Yeah,
let's
see
if
there
is
more,
you
know,
this
is
about
the
CD
event
0.1
release,
but
it
gives
you
an
idea,
so
we
have
events
grouped
in
different
phases.
So
we
have
generic
orchestration
events
about
like
python
orchestration
tools
and
then
we
add
events
by
software
configuration
management,
continuous
integration,
deployment
and
operation,
and
we
try
to
to
keep
in
mind
some
develop
metrics
like
the
Dora
metrics,
as
a
reference
where
we
are
defining
our
events
to
put
enough
information
there
that
allow
us
to
calculate
those
metrics
from
from
the
events.
J
Apart
from
that,
we
we
have
SDK,
so
we
have
SDK
in
go
and
Java
and
Python
and
yeah,
depending
on
our
yeah
users
and
adopters.
We
plan
to
build
things
like
JavaScript
past
or
more
sdks.
J
Right
so
this
was
work
work
in
progress
when
I
wrote
this,
but
like
incidental
remediations,
we
implemented
the
first
version,
but
something
that
we
are
still
working
on
is
to
include
in
the
specification
how
to
connect
the
different
events
to
each
other.
So
we
have
a
few
ideas
and
proposals,
and
so
we
we
target
that
for
the
next
release.
J
Yes,
so
correlation
exactly
so,
basically,
the
ability
to
have
like
a
shared
context
that
allows
you
to
kind
of
navigate
through
the
events,
and
we
have
one
approach
that
is
through
explicit
links
so
that
you
can
have
links
to
an
event
and
say
this
event
is
semantically
associated
with
this
other
event.
J
So
you
can
say
this
event
is,
was
generated
because
of
that
other
event
or
specify
some
other
kind
of
relationship,
and
this
is
an
idea
that
we
took
from
isil,
which
is
another
protocol
that
is
using
this
kind
of
approach,
and
another
thing
that
we
are
considering
is
having
some
kind
of
current
child
relationships,
similar
to
what
you
have
in
distributed,
tracing
so
or
possibly
a
combination
of
the
two.
So
here
working
on
on
this
and
of
course
we
are
also
open
to
suggestions.
J
Yes,
so
what
else
we
are
working
on?
So
we
we
started
looking
at
like
supply
chain
security
type
of
scenarios,
so
we
just
added
one
event
about
artifact
being
signed,
but
we
want
to
have
more.
So
we
want
to
extend
the
data
model
for
artifacts
to
include
things
like
asbomb
provenance
and
something
that
we
have
been
asked
about
is
also
signing
the
events
themselves,
and
this
might
be
interesting
if
you're
using
events
to
actually
trigger
your
workflow.
J
You
may
want
to
be
sure
that
your
event
has
not
been
tampered
with
to
avoid
triggering
things
that
you
should
not
be
triggering
like.
If
you
have,
if
you're
relying
on
an
event
to
do
a
deployment
in
a
staging
environment,
you
don't
want
to
do
that
with
the
link
to
the
artifact
to
be
deployed.
There
is
actually
something
that
was
injected
maliciously.
J
So
that's
an
area
that
we
we
plan
to
cover
yeah
I,
think
in
terms
of
community,
and
here
we
get
some
special
interest
group
as
we
call
them
in
the
CDF.
So
we
work
very
closely
with
the
interoperability
interest
group
and
we
have
connection
with
the
supply
chain
and
best
practices,
interest
groups
as
well.
J
We
started
discussing
also
with
the
psmi,
so
it's
a
value
stream
management
group.
They
have
some
wider
scope
density
events,
but
there
are
also
some
overlaps,
so
we
may
collaborate
with
them
and
of
course
we
are
very
keen
in
collaborating
with
this
community.
So
it's
not
only
CDF.
J
Here
are
some
companies
that
are
involved,
yeah
I,
think
I
took
them
enough
so
but
yeah
any
questions
so
far.
I
I
suppose
I
had
a
few
questions.
So
do
you
see
the
Eventing
system
being
effectively
a
signal
generator
for
observability.
J
I
think
yes,
so
we
could
see
USE
events
for
observability.
That's
one
of
the
use
cases
we
we
have
in
mind
and
yeah,
so
you
could
basically
one
of
the
use
cases.
For
instance,
we're
seeing
in
also
at
IBM
I
mean
different
teams
using
different
tools,
even
in
the
same
area,
so
maybe
some
use
Jenkins.
Some
will
use
Tech
tone
or
some
other
we'll
do
something
else,
because
there
is
a
lot
of
yeah
experience
from
the
engineers
in
those
tools.
So
they
are
keen
on
staying
with
their
tool.
J
And
if
you,
these
tools
being
able
to
report
what
you're
doing
with
a
consistent
language,
then
you
can
basically
observe
what
is
happening
across
the
different
tools.
So
you
can
have
one
single
build
one
single
dashboard
and
you
don't
have
to
I
mean
you
can
do
it
even
if
you
have
different
languages,
but
then
you
need
to
basically
adapt
the
signals
that
come
from
the
different
tools
and
yeah.
I
It
feels
like
as
well
what
you
are
saying
really
correlates
nicely
without
delivery,
in
that
the
composite
output
of
the
aggregate
of
the
events
creates
effectively
a
signature
right
in
that
I've
delivered
this
application
successfully,
because
all
the
events
along
sequence
have
been
composited
together,
and
this
is
my
signature,
so
there's
sort
of
a
security
element
there
as
well
right
when
you
trust
the
Eventing
system
has
articulated
all
the
bits
along
the
process.
We
could
have
a
really
nice
story
with
something
like
Potato
Head
or
with
this
platform.
F
J
It
is
I
mean,
as
you
said,
the
Assumption,
of
course,
is
that
you
need
to
be
able
to
trust
the
events
there
to
be
able
to
do
something
like
that,
but
that's
definitely
something
that
I
see
as
a
possibility.
I
don't
know.
If
Emil
you
have
some
other
comments.
D
I
So
it
was
like
there's
a
really
nice
story
to
tell
there
with
you
know,
a
CA
signing
events
and
delivering
an
application.
You
know
in
a
really
deterministic
way
and
showing
that
that
journey
and
that
kind
of
answers
the
observability
question
as
well
I
feel
like
we
could
definitely
work
to
make
some
demonstrations.
I
know
Josh
will
be
chomping
at
the
bit
to
do
something
there
as
well.
A
So
all
I
can
think,
there's
a
poster
like
one
issue
that
you
might
want
to
comment
on
where
we
will
try
to
get
this
into
semantic
conventions
for
open
Telemetry.
So
the
idea
is
not
to
create
events
but
have
the
proper
metadata
on
traces,
because
this
basically
allows
you
to
have
an
end-to-end
race
more
or
less
from
check
in
all
the
way
to
retirement
from
production.
A
There
are
still
some
challenges
to
be
solved.
I
think
there's
other
ways
to
do
this,
like
need
to
create
the
events
natively
in
kubernetes
to
don't
have
tools
to
subscribe
to
it.
A
It
has
this
double
I
think
this
double
app
project,
one
for
like
documenting
things
and
the
other
one
for
subscriptions
of
tools
and
on
the
potato
has
had
about
I
think
what
we
could
do
Josh
is
more
or
less
redesign
the
way
we
handle
examples
right
now
like
right
now
we
have
to
mean
this
individual
directories
and
the
maintenance
is
kind
of
not
at
these
in
order
to
testing
but
more
or
less
creating
a
CD
events
based
execution
flow,
a
very
simple
one,
and
then
individual
tools
can
actually
subscribe
to
take
on
certain
tasks,
which
would
allow
way
more
composability.
A
I
think
also
ties
nicely
into
the
work
of
the
platforms
working
group.
B
I
I
I
guess
I
I'm
totally
down
to
I,
hear
what
you're
saying
on
the
CD
events
and
adding
that
into
Potato
Head.
Could
you
elaborate
a
tiny
bit
on
where
that's
like
Platforms,
in
particular,.
A
If
we
build
that
prototype
of
a
platform,
I
would
for
I
would
like
to
see
the
same
events
being
issued
as
well
as
these
tools
interact
with
each
other
like
I
want
to
know,
can
image
get
got
pulled?
It
got
deployed
into
this
environment.
The
tests
got
executed,
operational
issues,
kind
of
occurred.
It
got
promoted
to
to
a
different
stage.
B
I
Would
it
be
interesting
for
you,
folks
to
have
some
collaboration
from
the
kubernetes
kind
of
angle,
emitting
events
like
that
that
could
be
something
I
could
certainly
be
interested
in
helping
with,
for
example,
if
you
could
tell
sorry,
if
you
could
tell
me
the
kind
of
events
or
I
could
read
through
the
spec,
we
could
certain
bridge
that
Gap
to
what
George
Josh
would
want
to
build
as
well.
J
I
mean
there,
there
are
some
certain
abstractions
and
in
kubernetes
today,
like
the
deployments
or
you
could
think
of
other
things
like
even
roll
out
from
Argo
rollout
or
those
kind
of
things
that
are
a
good
fit
for
City
events.
I
mean
we
have
deployment
type
of
events,
so
those
definitely
are
kind
of
things
that
we
can
and
we
can
consider
it
today.
J
Yeah
I
mean
we
don't
have
anything
really
like
right
now
in
scope
for
a
more
like
infrastructure
type
of
events,
but
it's
something
we
could.
We
could
discuss
and
see
if
it
makes
sense
for
for
the
scope
of
the
project.
B
E
Just
out
of
curiosity,
I
mean
what
I'm
hearing
is
this?
This
talk
of
event,
bus
versus
the
events
themselves?
We
don't
necessarily
have
to
say
that
the
event
bus
has
to
be
compatible.
I
mean
is
limited
in
compatibility
to
just
City
events.
So
that's
I
think
we
can
accomplish
what
we're
trying
to
talk
about
here
without
necessarily
saying
that
this
has
to
be
like
in
CD
event,
spec
and
scope.
E
But
the
idea
of
having
this
kind
of
like
thought
process
between
these
groups
and
bridging
that
information
to
build
some
sort
of
project
that
consumes
these
events
and
then
either
gives
us
visualization
or
observability
or
metrics
across
the
door
spectrum.
Is
you
know
really
what
I,
what
I
think
I'm
hearing
that
we're
trying
to
Target
as
a
group,
not
necessarily
saying
that
all
of
this
has
to
be
captured
in
CD
events,
I
think
it
does.
E
It
really
helps
in
terms
of
the
focus
especially
or
the
companies
that
are
not
just
you
know,
Cloud
native,
because
there's
a
lot
of
other
considerations
that
we're
going
to
have
about
the
events
themselves,
if
you're
doing
like
device
deployments
or
you
know
outside
of
just
cloud
or
even
just
bare
metal
servers.
E
There
are
some
things
that
we
want
to
consider
there
in
the
City
events
back
as
well,
but
having
this
concept
of
how
this
event
bus
is
going
to
perform
and
what
it's
going
to
do
and
what
it
can
span
and
scope
and
understand
that
I
think
is
is
an
interesting
conversation.
That
was
the
first
thing.
I
thought
of
when
I
saw
this
project
last
year.
Is
who's
going
to
build
that
event,
bus
system
and
how
inclusive
is
this
going
to
be?
And
you
know
what
are
the
Brokers
doing?
E
What
do
they
understand
and
could
it
be
modular
in
its
construction
and
I?
Think
that's
what
we're
talking
about
the
modular
bits
in
that
and
that
construction
of
that
process.
I
J
I
I
wanted
to
say
that
we
we
have
a
reference
architecture
initiative
within
the
CDF
as
well,
and
we're
definitely
interested
in
knowing
having
events
and
City
events.
Part
of
that,
so
that's,
maybe
something
we
could
collaborate
as
well
onto
is.
J
No,
it's
not
a
project,
yet
it's
just
an
initiative
that
we
started
and
we
we
need
to
build
it's
more.
An
idea
at
this
stage.
G
That's
reference
architecture
discussion
was
a
result
of
our
discussions
within
our
special
interest
group
best
practice
because
they
have
the
best
way
to
subscribe,
guide
published
on
their
website
and
when
we
have
conversations
with
people
who
want
to
see
things
in
a
more
realistic
manner.
That
discussion
started
in
Europe
and
maybe
we
should
better
reference
architecture
with.
You
know
how
people
can
bring
up
their
production
systems
based
on
different
architecture
and
follow
the
best
practices
and
perhaps
have
cd
events
part
of
that
architecture.
So
they
can
have
technology
couples.
G
You
know
production
systems
happy,
then
you
know
pushing
their
software
to
different
types
of
Target
environments,
Cloud
versus
device
and
so
on.
So
as
Andrea
says,
we
have
been
having
conversations
around
that
topic,
but
yeah.
It
will
continue
to
happen
potentially
under
space
and
risk
or
best
practices.
B
A
Yeah
is
there
some
like
very
simplistic,
General
emitter
of
events
where
I
could
say?
Okay,
these
are
the
events
that
I
want
to
get
emitted
from
different
tools,
and
then
we
because
then
we
could
have
like
all
the
tools
and
Potato
Head
attached
implementation
of
various
stages,
more
or
less
jump
in
there
and
subscribe
to
those
events,
which
also
would
make
our
testing
significantly
easier.
A
J
All
right
so
today
City
events,
they
have
a
general
schema
that
defines
the
payloads
and
we
have
a
binding
on
top
of
Cloud
events,
which
means
you
can
run
on
Json
body
over
http
and
that's
what
we
support
in
the
sdks
relatively.
But
you
can
yeah,
you
can
decide
any
of
the
binding
that
is
supported
by
Cloud
events
as
well
or
I
mean
you
could
even
introduce
other
binding
other
bindings
if
needed.
J
So
you
could
have
like
the
the
Json
blobs
transported
as
part
of
I.
Don't
know
open,
Telemetry
type
of
messages,
if
applicable
or
other
contexts,
so
right
now,
but
yeah
right
now.
It's
basically
this.
The
the
specification
is
Json
schemas
with
the
cloud
events
binding.
B
J
J
Yes,
so
right
now,
they're
very
lightweight
so
basically
produce
Json
blobs
that
are
compliant
with
the
Json
schemas
and
they
allow
you
also
to
parse
them
back
into
some
object
model
with
a
specific
language.
So
to
make
it
easier
to.
J
Yes,
that's
true
for
the
the
goal
line
SDK,
it's
a
work
in
progress
for
the
job
and
the
python
one.
G
Yeah
about
that
SDK
I
think
Andrea,
you
didn't
have
some
of
the
projects
that
are
already
on
their
way
to
adopt
the
device.
So
maybe
I
can
highlight
those
projects,
because
these
sdks
are
currently
in
use
by
different
projects
and
Java
SDK.
Yes,
it's
it's
this.
It
means
some
work
redonder,
but
last
week,
actually
the
very
first
version
of
the
Jenkins
plugin
for
cdms
was
published
on
plugins
website
and
also
nrfc
was
accepted
by
Spinnaker
Community.
But
this
is
with
us
from
student
cultures.
G
Is
why
I'm
putting
links
there
and
they
are
all
using
these
sdks
and
for
tecton,
for
example,
I
think
Andrea
you'll
work
with
that's
an
experimental
control
for
pectin
is
available
and
outside
of
CDF
again
I
believe
test
Cube
projects
which
is
kubernetes
made
to
test
framework.
They
introduced,
they
contribute
test,
Advanced
CB
lines
and
they
are
currently
implementing
CDI
support
to
within
their
project.
So
is
this
the
case?
If
more
projects
start
using
disaster,
they
will
become
better
because
everyone
will
find
something
is
missing.
I
J
So
in
the
proof
of
concept
we
have
used,
we
have
done
so
far.
We
used
like
Kennedy
serving
as
a
broker
for
cloud
events
because
yeah.
So
that
means
you
can
publish
events
into
into
one
bus.
J
J
I
I
B
B
D
That's
I
would
say
as
important
as
as
creating
implementations
at
the
moment,
but
for
sure
both
are
imported.
The
improve
the
concepts,
of
course
at
least,
is
very
valuable
to
show
that
it
actually
works
in
practice,
but
both
of
them
are
interested
too.
To
help
out
with
you.
E
And
if
we
get
to
a
point
where
we
want
to
discuss
the
event
broker
or
the
event
bus
or
what
that
system
actually
looks
like,
especially
if
we're
you
know
digging
through
an
expansive
modular
model,
I
personally
would
love
to
be
part
of
that
conversation
and
get
a
deeper
understanding
of
what
other
people
are
thinking.
D
D
So
when
it
comes
to
meetings,
sorry
well
I
pasted
the
link
to
the
community
web
page.
We
have
there
and-
and
it
might
be
hard
to
find
the
meetings
from
there,
but
we
have
in
the
City
events
protocol
group.
We
have
meetings
once
per
week
once
it's
on
Tuesday
afternoons.
European
time
I
mean
once
it's
on
Monday
midday,
European
Time.
D
D
J
Yeah
I
just
wanted
to
add
something.
I
think
one
aspect
is
sending
this
event
generating
this
event
and
something
I
did.
One
of
the
first
feature
is
like
the
incoming
side
or
like
taking
decisions
of
what
to
do
based
on
those
events,
so
that
area
is
not
something
that
we
we
have
done
any
standardization
work
yet,
but
it's
something
that
we
we
felt
from
the
beginning.
It
might
be
something
where
it
says
it
might
be.
J
An
area
will
build
something
as
part
of
CD
events,
so
some
kind
of
standard
way
to
receive
events,
apply
policies,
decisions
and
you
know
trigger
an
API
or
emit
another
event,
as
a
consequence
of
that,
so
there
is
a
lot
of
potential
of
thinking
that
area
to
have
something
like
that.
J
Some
component,
like
that,
some
of
the
tools
we
have
worked
with,
they
already
have
similar
logic
or
some
kind
of
logic
like
that
others
they
don't
and
they
would
need
to
build
it
from
scratch,
so
maybe
having
kind
of
common
solution
or
Standalone
component
that
could
deal
with
that
and
maybe
use
also
against
standard
technology.
You
could
think
I,
don't
know,
defining
your
policy
using
some
existing
policy
language,
for
instance,
so
that
that's
a
an
area
which
I
think
is
very
interesting
as
well
and
I.
Think
it's.
H
J
B
All
right
well
we're
getting
into
time.
The
only
other
item
on
our
agenda
today
was
Ouija
the
artifacts
working
group,
which
there's
a
link
in
there.
If
anybody
wants
to
check
on
the
current
state
Eloise
Alex,
do
we
have
more?
Do
you
have
anything
else
we
want
to
discuss
today
or.
I
I
had
one
item.
Yes,
it's
just
very
briefly
to
say
that
I've
opened
up
a
new
issue,
Josh
and
Ellis.
That
I
think
we
probably
need
to
revisit
the
tag
Charter.
It's
it's
four
years
out
of
date,
which
is
fine,
it's
a
good
Charter,
but
maybe
we
want
to
spend
some
time
in
the
next
meeting
just
identifying
if
there
are
any
particular
parts
of
vernacular
or
phrasing
or
technologies
that
we
want
to
bring
into
light.
A
Yeah
I
think
that
makes
sense.
Let's
go
over
it.
What
we
also
have
to
do
by
the
way
we
prefer
this
classic
at
kubecon.
We
need
to
get
into
the
culture
building
again.
I'll
prepare
something
for
the
next
meeting,
so
I'm
right
now
talking
with
zncf
what
the
official
process
this
line.
What
that
looks
like
I
have
some
insights,
but
I
will
share
more
in
the
next
meeting
in
two
weeks
from
now.
So
let's
need
some
more
prep
on
my
side
to
handle
it
properly.
A
That
shouldn't
be
too
hard
to
do
just
for
those
of
you
who
don't
know,
obviously
every
now
and
then
we
need
to
reload
co-chairs.
We
didn't
do
it
in
the
past,
because
we
had
coaches
leave
and
new
ones
coming
in,
but
now
it's
actually
time
to
do
this
as
well.
I
think
it
goes
well
together
with
the
work
on
the
charter
as
well,
and
I
will
do
actually
both
because
both
will
require
Toc
approval
and
we
might
but
I
want
to
do
them,
probably
together
at
least
alive
with
each
other.
A
All
right
yeah,
maybe
we
regroup
with
CD
events,
I,
would
say
we
let
this
sink
in
a
bit,
see
what
we
can
take
it
and
maybe
speak
again
in
a
month
or
so
that
everybody
kind
of
process
and
see
what
we
get
from.
You
I
think
it's
good
that
we
keep
talking.
A
I
think
it's
still
maybe
need
some
more
time
to
figure
out
the
details
of
how
we
continue
here.
But
it's
great,
we
are
talking
I,
think
especially.
The
work
on
semantic
conventions
in
open
Telemetry
is
definitely
something
to
align
on.
That's
obviously
not
Tech
app
delivery,
but
we
can
help
obviously
catering
brokerage,
also
with
the
observability
tag,
to
get
more
standardization
in
there
as
well,
and
what
we
put
on
traces
and
metrics
and
events
in
some
cases.
A
So
there
is
like
this
whole
issue
that
I
posted
in
here
that
was
already
started
out
of
the
the
captain
project,
but
I
would
have
to
like
it
go
further
to
like
see
the
events
see
how
far
we
can
get
there,
because
that
would
be
step
one
having
observability
in
a
standardized
way
for
CD
events,
I
think
it's
fine
and
it
should
fit
in
nicely
there
as
well
that
that
actually
might
be
a
good
starting
point,
because
that
will
benefit
everybody,
whether
it's
Cloud
native
or
non-cloud,
native,
to
some
points.