►
From YouTube: CDEvents / SIG Events Vocabulary - Sept 20, 2022
Description
For more Continuous Delivery Foundation content, check out our blog: https://cd.foundation/blog/
A
B
I
actually
had
a
question
that
you
might
be
able
to
answer
so
I
noticed
on
the
calendar.
It
says
this
meeting
happens
every
two
weeks,
but
looking
at
the
meeting
notes
or
the
agenda,
it
looks
like
it
happens
every
week
which,
which
is
it
it's
both.
A
I
can
show
you
okay,
so
if
we
bring
up
the
calendar,
I
can
share
my
stream.
Okay,
all
right
cool
Google.
Where
is
it?
There
is.
A
Yeah,
you
see
my
sharing
there,
I
guess
yeah,
but
we
have
this
meeting
on
the
Tuesdays
in
the
afternoons.
But
then
we
have
a
complimentary
or
additional
I
would
say
a
meeting
on
Wednesdays
every
other
week,
which
is
in
a
more
Asia
Pacific
time
frame
friendly
time.
A
Oh
both
of
these
are
actually
for
the
City
events
work
group.
They
are
not
named
exactly
the
same.
Okay
I
believe
they
used
the
same
Zoom
link,
I
think
so
so
both
of
these
are
intended
to
be
Project
work,
group
meetings,
let's
say
working
on
the
actual
spec
I
seriously:
okay
cool!
Then
we
have
the
the
Monday
Sig
meetings,
which
is
related
but
more
handling
the
administrative
stuff
on
the
Sig
itself
and
not
exactly
on
the
vocabulary.
Parts.
B
C
C
I,
don't
think
so.
Some
of
the
the
talks
were
published
almost
immediately
on
some
sort
of
virtual
platform
for
the
conference,
but
I
haven't
seen
our
talk
there
on.
C
It's
called
yes,
because
that's
how
you
saw
it
right.
Yes,.
C
A
Yeah
yeah
sure
there
it
is
yeah
everyone
in
the
meeting
here
who
didn't
catch,
the
the
record,
the
actual
live
presentation
by
Eric
and
Andrea.
You
are
free
to
go
in
here.
If
you
have
access
to
this
platform,
I
guess
it
might
need
to
be
registered
for
the
event
to
get
access
to
this
platform,
so
that
might
not
help
too
many
of
you
anyway,
I'm,
not
sure,
but
I'm
pretty
sure
it
would
be
posted
on
YouTube,
hopefully
answer.
Hopefully,
yes,
so
Eric.
A
When
talking
about
that,
did
you
get
any
comments
or
questions
or
discussions
based
on
your
presentation,
because
the
stream
I
watched
just
quit
directly
when
the
your
presentation
was
done
without
waiting
for
any
questions
or
anything?
Oh.
C
That's
a
shame:
yeah
there
were
some
questions
in
the
room:
I,
don't
remember
off
the
top
of
my
head
exactly
what
they
were
about,
but
there
were
some
people
who
showed
some
interest
and-
and
we
said
that
we
would
be
happy
to
talk
to
you
and
then
please
come
on
one
of
our
meetings
and
introduce
what
you
were
doing.
Maybe
on
Jay.
If
he
joins
today
can
remember
more
exactly
of
what
people
were
talking
about:
I
sort
of
blanked
out
I
guess
after
the
presentation,
okay,.
A
A
Yeah,
that's
great,
so
we
who
are
on
the
call
might
be
the
ones
who
will
stay
on
the
call,
then
I
assume
so
I
wasn't
really
prepared
to
drive
this
today.
But
since
Andrea
is
not
here,
I'll
do
my
best
to
see
if
we
can
at
least
go
through
the
agenda
that
Andrea
prepared
and
then
see
where
we
end
up.
A
So
we
can
see,
we
had
the
let's
see
Dora
metrics
demo.
A
C
A
C
The
only
thing
I
could
see
was
that
16
people
have
somehow
registered
for
the
talk
in
the
virtual
platform,
but
that
doesn't
mean
that
60
people
were
watching,
of
course,
but.
A
So
not
too
many
there.
He
didn't
you
yeah,
okay,
fair
enough,
so
unless
anyone
else
has
any
other
ideas,
I.
C
Do
remember
one
of
the
questions
now,
okay,
which
is
something
that
we
have
discussed
before
and
that
was
like.
Well,
two
questions
were
one:
is
anyone
running
this
in
production
and
to
that?
The
answer
is
no.
C
We
haven't
gotten
that
for
yet
we
are
mainly
doing
Concepts
and
the
other
question
was
like:
how
do
we
get
the
big
players
on
board
with
this
or
how
will
people
start
using
it
and
I
think
me
and
Andrea
answer
basically
which
we
have
discussed
before
as
well,
and
there
are
two
ways
we
can
go
about
this
or
two
ways
in
parallel
that
we
are
going
to
go
about.
C
This
one
is:
is
native
inclusion
of
sending
CD
events
and
receiving
City
events
in
existing
products
like
Jenkins,
plugins,
and
things
like
that
and
the
other
is
translations
where
we
just
hook
on
to
whatever
signals,
are
sent
by
other
larger
platforms
and
translate
them
into
City
events
and
then
pass
them
forward
right,
right,
yeah,
but
yeah
and
those
things
we
have
already
discussed
in
this
forum
before
foreign
for
us
anywhere.
B
Oh
yeah,
so
I
was
gonna
say
that
I
know
we
talked
about
headers
I,
don't
know
like
a
few
meetings
ago,
but
I
actually
wanted
to
revisit
that
at
some
point,
because
I
know
it
got
I
know:
Andrea
switched
back
from
binary
mode
to
to
this
bag
like
grab
bag
type,
but
this
kind
of
has
to
pertain
to
your
comment
about
getting
big
players
and
I
talked
a
little
bit
of
to
Andrea
about
this,
but
I
think
in
order
to
get
someone
that
is
of
Enterprise
level.
B
Like
of
you
know
of
someone
like
you
know,
like
apple
us,
for
instance,
or
like
AWS
from
that's,
where
I
was
previously
working
at
you,
you
would
need
to
pass
these
as
as
headers
I,
don't
know
what
that
would
look
like
still
because
you
guys
do
require
a
lot
of
metadata.
B
But
the
main
issue
for
this
is
that
if
you're
touching
payloads,
it's
going
to
be
a
pretty
big
deal
for
these.
These
type
of
companies,
unless,
like
you,
have
a
translation
layer
in
front
of
every
service,
that's
sending
as
well
as
every
service
that's
receiving,
which
I
think
would
be
really
difficult
to
do.
B
C
No
and
I
I
don't
see
that
necessarily
as
the
goal
of
the
City
events
project
to
replace
whatever
internal
signaling
existing
players
have
between
their
products
that
that's
not
really
our
goal.
Our
goal
is
to
to
promote
interoperability
between
different
vendors
and
also
to
promote
observability.
So
we
don't
to
me
it's
not
super
important
that
people
or
big
actors
start
replacing
the
things
they
are
already
doing
with
City
events.
B
But
that's
that's
the
issue,
though,
if
you're
touching
the
payload,
they
will
never,
they
will
never
adopt
it
honestly.
It
like
that.
That's
the
biggest
and
that's
why
I
wanted
to
go
to
headers
is
because,
if
you
say
okay,
these
are
now
in
headers
you
don't
you
don't
touch
the
payload
they'd,
be
more
willing
to
add
it.
On
top
of
it,
you
know
and
I
get
your
statement
of
like
being
additive
but,
like,
like
I,
said
touching,
the
payload
is
a
huge
deal
for
these.
For,
for
these
companies.
C
B
C
That's
one
way
of
doing
it,
because
if,
if
it
is
the
case,
as
you
say
that
the
structure
and
format
of
this
payload
is
extremely
important,
then
that
means
that
someone
else
is
relying
on
that
payload.
Looking
exactly
that
way,
and
that
means
that
we
would
then
be
trying
to
replace
something
with
seed
events,
which
I,
don't
think
is,
is
one
of
at
least
it's
not
a
goal.
That
is
particularly
important
to
me
to
me.
C
It's
important
that
people
who
want
to
to
do
interoperability
and
observability
can
start
to
see
City
events
produced
by
by
bigger
actors,
and
we
can
do
that
either
by
getting
those
actors
to
produce
seed
events
natively
or
through
translation
layers,
and-
and
it
doesn't
matter
to
me
if
those
translation
layers
sort
of
destroy
the
payload
so
that
it
cannot
be
received
by
whoever
was
receiving
the
old
payload,
because
that
solution
already
works.
We
don't
need
to
use
CD
events
for
that,
so
roughly
in
that
that's
how
am
I
thinking
or
yeah.
B
Yeah
no-
and
that
makes
sense-
maybe
I'll
talk
to
you,
maybe
more
in
slack
to
kind
of
give
you
you
know
kind
of
like
hash.
This
out,
you
know
and
kind
of
figure
out
like
what
maybe
the
best
approach
is
because
I
I
need
I
need
like
another
opinionated
someone,
that's
kind
of
against.
B
You
know
this
idea
to
kind
of
really
you
know
flesh
it
out
right
because
right
now
like
when,
like
I,
because
I
I
have
proposed
the
spec,
you
know
some
people
that
I've
worked
with
at
Amazon
as
well
as
internally
here
at
Apple,
and
the
biggest
concern
that
they've
had
so
far
has
been
touching
on
this
payload.
But,
like
you
said,
if
you're
saying,
oh,
maybe
you
send
it
as
a
separate
Channel.
B
I'm
curious,
like
I,
said
that's
kind
of
interesting,
but
like
I'm
curious.
What
that
would
look
like,
though,
because
now
like
every
end,
point
needs
like
a
duplicate
right,
so
yeah,
like
I,
said
we
can
hash
it
out
and
then
maybe
we
can
kind
of
figure
out
like
a
good
solution
for
this
and
then
kind
of
go
from
there.
If
that
works.
For
you.
C
D
B
Yeah,
so
one
so,
both
from
my
experience,
both
Apple
and
AWS
is
is
a
very
you
know,
is
a
very
non-breaking
change
type
of
company,
so,
like
any
service
that
you
have
is
going
to
work
basically
forever
for
all
customers.
B
So
changing
the
payload
is
a
huge
break
of
that
contract
and
then
getting
you
know
like
now,
if
you're
sending
extra
data
with
that
payload
as
well-
and
you
send
that
along
you
know
the
biggest
issue,
there
is
just
kind
of
like
missing,
like,
for
instance
like
if
I
send
it
like
service
a
to
service,
B
right
that
data
would
get
dropped
in
the
payload.
And
then
you
go
to
service
C
it.
B
You
know
that
they
lose
all
that
data,
but
let's
say
they
support
it,
but
with
headers
you
can
pass
headers
along,
so
you
know
just
buy
like
a
like,
typically
a
flag.
You
can
just
say
you
know
just
pass
these
header.
B
You
know
pass
these
headers
along,
so
the
barrier
to
entry
is
a
lot
lower
than
now
changing
all
these
models,
saying
oh
now,
support
this
New,
Bag
type
and
so
and
so,
like
I,
said,
the
barrier
entry
is
a
lot
lower
to
to
support
something
to
pass
pass
along
to
multiple
services
and
as
well
as
you
know,
it
being
not
a
breaking
change,
potentially
as
well
so
yeah.
D
Yeah
I
still
don't
get
it
sorry.
I
understand,
I,
understand
the
the
concept
of
you
know
you
don't
want
to.
If
you
have
data
that
you're
passing,
you
want
to
continue,
it
and
I
I
totally
get
that,
and
maybe
we
should
make
sure
we're
passing
all
the
data
that
anybody
would
use
but
yeah
having
I
I
I'm,
really
struggling
with
under
really
understanding
the
core
pain
point
other
than
missing
data.
D
B
No,
no,
no
that's
well.
I
said
that's
where
I
was
explaining
that
the
barrier
of
Entry
is
different
right
to
pass
a
header
along
is
a
lot
easier
than
changing
every
model
in
your
code
base,
like
you
know,
like
we
work
in
a
code
base
like
you
know
where
it's
like
hundreds
of
thousands
of
lines
of
code-
and
we
have
you-
know
hundreds
of
thousands
of
of
models
like
request
models,
response
models.
Do
you
expect
us
to
change
every
class
to
have
this
this
bag
right
like?
B
Instead,
you
know
it's
easier
for
us,
just
to
say:
hey
pass
these
headers
along
right
for
these
type
of
companies,
so.
A
Yeah
those
headers
to
all
those
calls
in
in
all
other
parts
of
the
source
code
instead
and
in
a
lot
of
different
places
in
the
code.
I'm
I'm.
Thinking
in
in
that
scenario,
that
you
should
have
some
kind
of
service
that
listens
to
those
messages
that
already
sent
and
just
translate
the
relevant
ones
with
the
relevant
data
to
Siri
events
as
a
one
single
Standalone
thing,
then.
A
B
I
think
that's
what
Eric
was
kind
of
alluding
to,
but
like,
but
passing
headers
along,
like
we
don't
have
to
change
that
like
basically
and
well
like
we
well
I
guess
it
depends
to
company
company,
but
in
our
like,
for
instance,
back
at
AWS.
We
could
say
just
pass
these
headers
along
from
service
to
service
and
it's
it's
that
easy
to
just
to
populate
those
through,
but
with
like
models.
For
instance.
That's
that's
a
lot.
That's
a
lot
more
difficult
for
us
and
then
the
biggest
thing
for
us.
B
It
also,
as
you
alluded
to
Tracy,
is-
is
security
like
having
some
something
touched,
the
payload,
especially
for
some
of
our
services,
that's
a
huge
yeah
that
that's
gonna
be
a
a
no
like.
That's
not
gonna
happen
with
an
apple.
To
be
honest,
so
security
is
also
a
very
big
issue,
but
also
like
breaking
changes,
because
we
don't
know
like
unless
you
know
like
we
don't
know
like.
B
What's
going
to
be
happening
to
that
payload,
you
know
I
mean
you
could
say
what
you're
doing
with
it
but
like
if
you
change
any
structure
of
it
and
it
breaks
any
customers.
That's
a
huge
issue
for
any
of
these.
These
these
companies,
right
because,
like
AWS,
adheres
to
contracts
like
saying
hey.
This
is
what
we're
sending
this
is
what
you're
receiving
and
if
that
ever
changes,
maybe
even
through
like
an
error
or
something
of
that
nature,
that's
a
huge
issue.
A
I,
don't
really
say
why
it's
a
descent
by
the
cloud
itself,
I
mean
it
would
be
sent
by
the
applications
running
in
the
cloud.
Rather
so
so
I
mean
for
this
picture.
For
example,
if
you
would
have
some
CI
service
running
in
AWS
use
that
uses
Technical
and
capital
for
example,
then
then
this
would
be
more.
What
happens
is
it's
more
the
application
layer
that
these
events
are
sent
and
not
on
the
the
infrastructure,
the
cloud
provider
layer
itself?
Oh.
B
Yes,
I'm
sorry,
yeah
yeah
when
I
say
customers
like
actually
so
this
is
actually
kind
of
a
different
semantic
meaning
than
what
most
people
are
used
to
so
back
at
AWS.
When
we
say
customers,
we
literally
mean
anyone
that
uses
the
product
and
that
might
be
even
internally.
B
So
this
is
like
if
in
for
instance,
service
a
service
B
like
these
are
are
like
person
that
service
B
that
uses
our
service
service
a
that's
our
customer
right,
even
though
it's
all
internal
to
like
AWS
or
apple.
So,
if
you're
changing
that,
you
know,
that's
that's
a
huge
deal.
If,
if
that's
what
you're
trying
to
say
there.
A
No
I'm
saying
that
it
will
change
anything
I
mean,
of
course,
if
you
deploy
an
application
on
AWS,
those
supplications
could
use
whatever
interface
is
externally
to
this
to
send
messages
or
RPC
calls
or
whatever
towards
other
services
in
in
the
same
cluster
or
some
other
cluster
and
AWS
I.
Guess
don't
care
about
that
too
much.
A
A
To
deliver
them
the
software
for
the
cloud
itself,
maybe
I
mean
the
what
are
AWS
software
in
in
the
production
of
what
you
call
it.
Software
production,
our
AWS
software
itself,
but
not
within
the
cloud
itself
as
such,
but
I
mean
if
AWS
was
the
software
for
AWS
was
produced
through
some
kind
of
CI
tool,
says.
B
For
the
confusion
is
because,
like
so
when
I
was
talking,
so
all
so
I'll
go
to
AWS,
because
that's
what
we're
talking
about
so
AWS
has
its
own
set
of
series
of
CI
CV
tools
and,
and
that's
what
I
was
talking
about,
and
these
CI
CV
tools
again
are
all
internal.
B
They
are
not
Forks
or
anything.
So,
like
you
know,
for
you,
you
know
the
first.
They
would
need
to
get
the
plugins
in
there
and
then
figure
out.
You
know
how
to
change
all
these
payloads
or
headers.
You
know
to
support
you
know
and
then
get
interoperability
with,
like
you
know,
like
maybe
other
CI,
CD
type
things
like,
for
instance.
B
If
code
build
wanted
to
support
this,
then
maybe
you
could
interface
with
now
the
internal
Amazon
CD
section
of
it,
because
that's
that's
kind
of
the
vision
that
I'm
seeing
right,
but
the
biggest
issue
here
is
again:
if
you're,
if
you're
changing
the
the
payload
that
I
don't
see
that
happening
one,
because
security
is
a
big
part
of
it,
but
also
the
potential
of
not
knowing
what
the
the
translator
is
going
to
do.
B
A
Yeah,
so
what
you
say
now
is
that
AWS,
for
example,
uses
their
own
proprietary
Solutions
as
cicd
tools
for
their
own
software,
so
they
they
don't
use
like
Jenkins
or
techton,
and
anything
like
that.
They
have
their
own
stuff
yeah,
but
is
there's
then,
but
but
you
say
that
they
might
be
interested
in
City
events,
because
they
want
to
use
some
common
visualization
that
is
or
metrics
collector
or
Trace.
B
B
Yeah
yeah,
so
the
one
thing
at
AWS,
that
was
a
major
pain
point,
is
actually
getting
AWS
to
work
well
with
internal
like
inter
the
internal
tools
as
weird
as
that
sounds
so
like,
for
instance,
if
you
can
now
like
to,
for
instance,
get
code
build
to
actually
run
as
the
CI
portion
of
this
like
pipeline,
that
does
CI
and
CD
was
very
difficult
to
do
in
AWS,
because
the
you
know
the
payload
and
the
you
know
the
events.
B
There
was
no
standardization
right
there
right,
so
they
had
to
write
these
like
adapters
and
plugins
just
to
support
this,
but
as
AWS
like
new
tools
or
cons,
new
services
are
constantly
coming
out.
You
know
like
with
Cloud,
watch
and
whatnot
and
to
to
to
support
you
know
any
of
these
Services.
You
know
they
have
to
spend
time
and
develop
it,
but
if
they
were
using
something
like
CD
events,
it
should
just
be
able
to
be
dropped
in
right,
I
mean
that's,
that's
the
goal
that.
B
D
What
you
know,
I
have
to
jump.
I
think
this
is
a
great
discussion.
Could
we
instead
of
moving
it
to
a
slack?
Could
we
just
create
a
discussion
in
our
whatever
in
our
GitHub
project,
on.
D
Issue
I
think
it's
a
better
place
for
these
kinds
of
discussions,
and
the
other
thing
that
I
just
want
to
point
out
is
one
of
the
security
is
an
interesting
topic
right
now.
Obviously,
but
one
of
the
things
that
I'm
hearing
where
people
are
starting
to
say
hey,
maybe
we
should
be
looking
at
an
event
based
CD
platform
is
the
fact
that
so
many
people
have
aren't
not
including
the
generation
of
s-bombs
through
their
process
and
they're.
D
Looking
at
having
to
update
so
many
workflows,
yes,
so
I
just
want
to
make
sure
whatever
you
know,
I
think
this
is
a
great
discussion
to
have
and
thank
you
for
bringing
it
up.
But
what
I
think
we
need
to
keep
in
mind
is
we
are
disrupting
this
process
and
we
need
to
continue
disrupting
it
because
the
way
we
do
it
now
doesn't
work.
D
And
there
is
you
know
you
have
to
be
brave
to
walk
into
a
new
world
and
say
this
is
a
better
way
to
do
it
and
it
might
require
a
lot
of
work
but
it'll.
Take
you
a
lot
farther
in
the
long
run
and
I
think
that
that's
the
goal
of
CD
events
is
to
really
fix
the
problem
and
not
patch
it.
So
that
would
be
my
concern.
Is
that
we're
trying
to
dumb
down
the
events
process
to
patch
it.
B
Okay,
yeah,
that's
definitely
yeah
I
like
that
call
out,
and
that's
definitely
not
what
I
want
to
do.
I
don't
want
to
patch
it,
but
I
also
want
it
to
be
useful,
for
you
know
for
things
like
AWS
and
and
and
apple.
But
the
question
is:
how
do
you
get
those
things
in
there
and
and
maybe
that
discussion
though
you
know
that
GitHub
issue
will
will
help
with
that.
So
yeah
I'll
definitely
do
that
later
in
the
day,
I
got
another
meeting
after
that.
So
but
I'll
definitely
do
that.
A
It
might
even
be
an
issue
we
can
continue
discussing
and
if
I
remember
correctly,
I
think
Andrea
wrote
an
issue
about
this,
but
it
might
have
been
Council
or
something
but
I'm
not
sure.
D
D
You
know
we
we
do
all
of
you
know:
I
run
the
I'm
doing
CD
F
meetups.
We
finally
are
starting
those
again
and
we
always
just
do
a
kind
of
we
just
talk
at
people
as
opposed
to
having
a
discussion.
D
I
would
love
to
be
able
to
do
something
like
this
in
a
CDF
Meetup
that
this
is
just
going
to
be
a
discussion
on
how
CD
events
should
be
built,
and
we
can
have
a
discussion
like
this
publicly
because
there's
so
much
learning
to
be
done.
I've
learned
a
lot
in
the
last
15
minutes.
So
thank
you
for
bringing
this
up
in.
B
Yeah
no
problem-
and
you
know
thank
you
for
the
push
because,
like
I
said
like
it's
hard
like
when
you're
you
know
trying
to
bring
up
an
issue
like
you
know,
like
an
apple
for
instance,
or
in
AWS
everyone's
gonna,
just
kind
of
agree
with
you,
because
they're
all
in
the
same
mindset,
but
bringing
it
here
and
again
that
pushback
is
going
to
really
help
flush
this
out
and
hopefully
be
for
the
better
right
for
CD
events.
That's
kind
of
the
goal
here,
for
me
at
least.
D
A
And
one
idea
is,
of
course,
that
you
should
be
able
to
not
just
interpret
between
systems,
but
also
replace
systems
without
I
mean
if
you
would
replace
the
consumer
of
these
events.
So
the
information
say
visualization
thing
you
shouldn't
need
to
replace
the
producer
of
the
event
as
well
at
the
same
time,
because
they
would
need
to.
They
should
understand
the
same
information
in
the
same
format,
so
I'm
a
little
bit
concerned.
A
If
we
have
a
lot
of
payload
that
is
currently
sent
in
these
events
within
AWS
in
those
proprietary
Solutions
between
those
and
then
we
somehow
switch
out
there,
the
whatever
orchestrator
to
tekton
or
something
in
that
system,
and
then
all
of
a
sudden,
the
the
consumers
of
that
information
would
require.
Take
them
to
produce
some
some
new
payload.
That
tecton
is
not
used
to
be
just
because
the
old
system
did
and.
B
D
A
B
Well,
yeah
thanks
thanks
for
for
challenging
it
again,
Tracy
appreciated
so
yeah
yeah.
A
A
Yeah
all.
B
A
B
A
So,
let's
see
we
are
still
leveling,
we
haven't
left
yeah,
so
action
items
from
last
time
we
had
a
ction
on
Andrea,
where
he
actually
made
the
since
all
the
action
anyway
about
the
document
in
the
subscriber
model.
So
the
action
is
not
or
the
documentation
is
not
written,
but
there
is
an
issue
for
it
now.
I
believe
this
one
should
be
added
to
the
project,
so
we
added
to
cr.1
as
well.
A
This
is
actually
very
much
needed,
I
would
say
so
we
so
it's
clarified
what
the
events
are
intended
for.
So
I
will
add
that
to
this
project
here
and
also
to
the
Milestone
0.1.
A
It
could
be
so
SDK
features.
That's
another
thing:
we
should
document.
What
are
the
general
features
required
by
any
SDK
that
is
supposed
to
be
compatible
with
City
events
and
examples
of
such
things?
We
should
add
to
those
requirements
or
what
kind
of
coverage
when
it
comes
to
unit
tests.
It
will
be
needed
by
this
case,
should
there
be
conformance
tests
towards
the
schemas
for
the
City
events,
and
they
should
pin
the
specific
version
of
the
specification
in
the
SDK
as
well.
A
And
well
I
didn't
take
it
serious
one
Journal
with
another
example.
B
When
do
we
plan
on
how
far
away
are
we
guess
I
guess?
Are
we
real
plan
on
releasing
like
a
tag
for
the
go
SDK
just
out
of
here,
or
is
that
still
pretty
far
out.
A
A
There
are
a
lot
of
things
in
to
do
here
and
quite
a
few
things
that
are
assigned
to
people,
and
also
those
designed
to
people
are
not
always
in
progress.
I
have
some
which
I
haven't
progressed
too
much
on,
so
I
would
hope
that
we
get
more
help
from
people
on
this
call
or
in
other
way,
to
help
out
with
this.
B
Yeah
I've
tried
I'll
talk
to
my
manager
again
but
like
the
problem
with
apple
is
they're,
like
so
finicky,
with
touching,
like
anything
like
code,
wise,
open
source,
so
I
I'm,
but
I'm
gonna
keep
pushing
for
it.
I've
been
trying
to
get
approval,
but
once
I
get
approval,
I
I
will
definitely
help
out
and
I'll.
Let
you
guys
know
when
I
do
yeah.
A
Yeah
because
of
course
we
we
want
to
have
this
announced
during
Cape
Condon,
our
city
Summit,
but
of
course
it
depends
if
we
can
manage-
and
there
is
not
just
this
I
mean
we
have
the
yeah
the
administration
and
said
we
need
to
to
do
the
announcement
itself,
with
some
blog
posts,
with
it
probably
and
PR
documents
and
such
things,
and
that
it's
not
just
on
any
quarter
either
so,
and
there
are
some
not
too
many
things
in
the
actual
spec
to
be
added.
I.
A
C
Yeah
so
I
think
Andrea
has
actually
done
quite
a
bit
of
the
work
for
yeah
for
for
changes
for
artifacts
and
for
services.
C
C
C
And
the
incident
events,
though,
are
not
started
yet,
but
so
that's
probably
a
bigger
thing
that
we
need
to
to
work
on.
A
Yeah,
because
we
have
said
that
we
want
to
be
able
to
support
the
Dora
metrics
fully
or
as
much
as
feasible
with
a
series,
one
version,
but
it
might
be
so
well
so
that
we
have
something
that
is
not
part
of
it
so
say
we
can.
We
can
provide
three
out
of
four
metrics
or
something
yeah.
That
could
also
be
a
way
to
handle
it,
of
course,
but
preferably
we
should
have
support
for
all
of
them.
A
Yeah
I,
don't
intend
to
go
through
all
of
these
different
topics,
but
you're
free,
of
course,
to
to
volunteer
to
take
on
any
of
these
sections
or
tickets
to
work
on
them
anyway.
Those
are
the
history
created
and
we
have
discussed
sorry
now
the
spec
changes.
A
The
order
is
there
to
talk
about
that
now,
and
this
is
what
we
talked
about
as
well,
so
we
had
a
change
in
the
artifact
events
here
there
are
some
some
updates
already
in
FCS
work
yeah,
which
will
eventually
be
the
provided
through
pull
requests
to
the
spec
itself.
A
That's
what
I
think
we've
covered
them,
but
what
is
part
of
the
agenda
here
is
anything
else.
On
your
mind,
that
we
should
use
this
time
for
them
that
remains,
or
should
we
try
instead
to
spend
the
20
minutes
on
actually
trying
to
do
something
about
all
these
issues?
We
have
ourselves.
A
B
Don't
have
them
too
much
so
yeah
like
I
I'd
vote
to,
because
that
gives
me
20
minutes
to
write
the
issue
so
that'd
be
perfect.
Yeah
can.
B
No,
it's
mostly
how
it's
being
passed
so
right
now
it's
it's
a
payload
like
it
touches
the
body
which
is
is
a
huge
issue
because,
like
for
example
like
there's,
some
services
like
we
can't
even
see
the
payload
so
getting
approval
for
like
a
translator
or
anything
to
touch
that
payload
is
like
a
really
big
deal
internally,
so
that
that's
like
one
potential
issue,
but
you
know
like
there,
but
there
might
be
Solutions
there,
but
the
thing
is
yeah.
B
Just
touching
that
body
is
generally
a
big
No-No
for
you
know,
for,
for
you
know
from
the
past
companies
that
I've
worked
at
one.
You
know
worrying
about
breaking
change
to
worrying
about
you
know.
What
are
they
going
to
be
doing
with
that?
Metrics
has,
you
know,
has
security
because,
like
it's
a
getting
security
approval
back
at
AWS
would
take
years.
You
know
like
it
takes
a
very
long
time
and
then
my
experience
here
at
Apple
also
very
very
long
time.
B
So
if
this
is
something
that
we
want
to
use,
it's
it's,
the
adoption
seems
really
really
like
the
barrier
to
entry,
for
these
type
of
companies
makes
it
very,
very
high,
which
I
don't
see
these
companies
wanting
to
use
it.
If
that's
the
case
is
more
of
my
concern,
there.
A
I
told
her
that
they
would
need
some
big
benefits
to
get
from
it
in
the
directly
and
that
could
be
Dora
metric,
so
it
could
be
visualizations
or
it
could
be
something
else
maybe,
but
they
need
to
be
huge
benefits.
I
see
to
be
able
to
to
allow
such
changes
internally.
Sure,
but
I
also
think
that,
even
though
there
might
be
a
lot
of
information
sent
in
these
messages
today,
yeah.
B
A
Yeah
actually,
but
then,
of
course,
all
standards
of
these
events,
he
that
all
the
centers
need
to
send
it
as
well
or
there
needs
to
be
some
acceptable
consumer
of
the
current
data
that
can
just
consume
service.
B
Yeah,
does
that
also
does
that
answer
your
question
Cara.
D
Yeah,
it
does
a
bit
I,
guess
I'm,
just
trying
to
to
figure
out
like
just
to
get
closer
to
and
I'm
thinking
of
more
concrete
example
so
like.
If
you
look
at
other
messaging,
just
like
grpc,
the
metadata
is
opaque
to
grpc
itself.
So
let
the
client
provide
information
to
the
server
and
vice
versa,
but
like
associated
with
the
call,
but
is
that
what
you
would
that
enable
CD
events
to
be
then
used
more.
B
Yeah,
so
to
kind
of
give
the
why
that
works
is
because
that's
chosen
up
front
typically
or
like
when
you're
creating
a
new
API.
You
know
you
can
choose
it
to
be
grpc
or
you
can
make
it.
You
know
potentially
additive
if
they
wanted
to
do
that.
But
but
typically
people
don't
just
add
grpc
I
mean
some
people
do,
but
I've
never
seen
it
internally
where
people
just
like,
let's
just
add,
grpc
on
top
of
it.
When
you
know
the
the
current
encoding
works
works.
B
Fine,
so
that's
the
biggest
issue
is
like
because
you're
start
you're
you're,
just
since
you're,
just
starting,
you
can
choose
it
to
be
grpc.
So
if
this
was
a
new
service
that
was
coming
out
of
like
apple
or
AWS
perfect,
we
can
use
CD
events,
but
the
problem
is
if
it's
only
this
one
service.
How
useful
is
that
you
know.
C
B
Okay,
so
so,
when
you're
choosing
like
you're
choosing
so
like
I'll,
go
like
you're
choosing
like
you're,
paying,
let's
say
like
your
payload
encoding,
so
we're
choosing
Json
up
front
right.
That's
similar,
like
we're,
choosing
grpc
up
front
right,
there's
no
breaking
changes
there,
there's
no
contracts,
there's!
No,
really!
Nothing
right!
So
like
this
new
service
now
can
just
start
sending
CD
events,
you
know
and
sending
you
know
whatever
and
whatever
encoding
in
whatever
format,
because
that's
chosen
up
front,
they
didn't
make
any
promises
to
any
other
services.
B
Or
you
know-
and
this
is
assuming
that
security
is
fine
with
this
right.
But
that's
the
biggest
thing
is
you
know
not
making
any
promises
to
all
these
other
services
and
that
this
is
what
this
payload
looks
like.
B
Because
now
well,
you
know
like
I,
said,
and
then
you
know,
like
imagine,
you
know,
you're
the
server
the
same
service
and
you're
like
oh
we're,
gonna
change
from
Json
to
XML.
That's
a
huge
issue
right
but
I
mean
it's
not
the
same
because
you're
not
going
that
far.
But
you
know
just
kind
of
give
you
an
idea
why
that
might
be
some
somewhat
of
an
issue
somewhere
right.
B
Cool
so
I
think
that's
all
I
had
and
I
almost
try
attending
the
the
other
meeting
that
you,
you
posted
a
meal
like
that's
on
a
Wednesday,
but
it's
like
four
a.m,
my
time,
so
that
might
be
a
little
bit
harder.
Yeah.
A
It's
a
bit
harder
yeah.
So
that's
next
Wednesday
but
of
course
we
could
have
ad
hoc
meetings
as
well.
If
we
need
to
discuss
this
in
any
more
detail,
I
I
I'm,
pretty
sure
we
won't
have
any
such
header
handling
of
the
or
putting
data
into
headers
as
part
of
zeroth
one,
but
but
for
sure
we
can.
We
can
look
into
it
for
zero.
Two
of
them
I
mean
we.
A
B
B
And
then
a
meal
and
then
I
think
maybe
like
and
like
I
said.
This
is
just
an
idea.
You
know
like
saying:
headers
is
you
know
like,
but
if
we
go
a
totally
different
route
like
like,
you
were
suggesting
like,
maybe
we
just
need
a
service
that
handles
these.
You
know
that
that's
fine
too,
but
I
just
want
to
make
sure
that
that's
outlined
in
the
spec.
B
So
when
companies
like
you
know
like
my
company,
Apple
or
AWS,
is
looking
at
this
they're
like
okay,
they
handle
this
use
case
we're
okay
with
this
you
know.
So
that's
that's
kind
of
more
of
my
concern
right,
yeah.
A
But
maybe
that's,
then,
what
we
should
try
to
do
for
president,
even
and
that's
just
the
short
paragraph
or
a
couple
of
paragraphs
around
that
and
for
the
use
case
itself,
you
can
mention
that
in
your
ratio
as
well,
if
you
like
yeah.
B
A
That
would
help
a
lot
and
will
decrease
the
confusion,
yeah
yeah
for
sure,
but
for
people
having
the
same
same
use
cases
in
minus
as
you
have
now
yeah.
That
would
be
great.
Actually,
so
please
don't
spend
the
last
next
14
minutes
right
in
the
ratio,
all
right
awesome.
If
we,
if
we
stop
the
meeting
now,
okay.