►
From YouTube: CHAOSS.Community.September.24.2019
Description
CHAOSS.Community.September.24.2019
B
D
It
was
actually
it
was,
it
was
actually,
it
was
super
profound,
initiate
repeated.
What
dawn
said
is.
Is
that,
instead
of
saying
reference,
implementation
examples
uses
three
different
implementation
headings
that
we
should
just
have
I,
think
one
heading
that
says
implementations
and
what
that
is,
can
vary
by
the
specific
metric.
D
You
know,
whatever
the
most
clear
explanation
of
the
metric
is
whether
it's
a
nobler
corridor,
Gomorrah
animal
impatient
for
both
for
an
auger
implementation
or
all
three,
whatever,
whatever
helps
make.
What
the
metric
is
more
clear
in
a
concrete
way
should
be
under
component
ation
I
think,
is
what.
E
E
E
D
Filters
are
filters
are
important
because
that
specifications
I
think
could
be
under
implementation
like
if
we
had
one
one
implementation
heading.
There
could
be
a
subheadings
in
there
for
visualizations
and
I
think
that
would
be
more
clear
because
filters
are
actually
something,
and
you
guys
others
to
me,
tell
me
if
I'm
not
and
I
wish
it
was
super-scary
plate.
Well,
we
we
had
this
discussion
about
whether
a
filter
wanted
to
call
a
filter
or
a
parameter
way
back
when
and
filters
are
essentially
the
expected
method
inputs.
D
The
fact,
essentially,
that
you
would
be
able
to
provide
a
tool
to
get
a
subset
of
data
showing
that
metric
right,
so
common
filters
are
repository
date
range
things
like
that
and
I
think
it's
that's
very
stuff.
It's
very
different
than
filters,
and
so
I
mean
then
visualization,
so
I'm
I
wasn't
there
for
the
discussion
of
merging
those
headings.
E
D
Feel
like
I'm
talking
a
lot
of
us
colleges,
but
I
think
filters
should
stand
on
its
own.
If
I
was
to
make
a
change,
I
would
say
filters
can
skimp
on
its
own,
because
it
is
a
very
specific
thing,
with
a
very
specific
purpose
and
visualizations
and
implementations
could
be
a
heading
or
even
just
implementations,
and
then
it
could
have
some
headings
that
were
things
like
product
implementations
or
not
product
known,
like
basically
grimore
lab
otter
example.
F
D
D
D
You
could
yeah
I
wouldn't
be
opposed
to
that
I
think
I
think
I
want
it
to
remain
clear
that
felt
like
it
built.
The
filter
list
is
intended
is
like
if
you're
gonna
claim
that
you're
implementing
the
automatic
that
you're
gonna
implement
but
filters
that
are
enumerated
definition,
that's
the
only
reason,
maybe
I'm
reluctant
to
take
filters
and
put
it
on
orientation,
but
that's
totally
pinned
antic
and
I.
Think
it's
probably
just
fine
to
do
that.
Okay,.
G
On
my
part,
I
think,
to
a
degree
that
kind
of
comes
back,
that
kind
of
comes
back
to
what
Matt
was
asking
early
on.
So
like
how
much
information
do
we
really
need
to
put
in
these
documents
so
filters?
Filters
are
important
for
the
augur
implementation,
but
is
it
important
for
the
metrics
definition?
D
Actually
filters,
I,
don't
think,
are
important.
Filters
are
important
or
I
sent
I
think
essentially,
instructing
anyone
implements
the
metric
on
what
the
filters
should
so
are
you're
a
bit
more
lab
Joe's
Python
source
lab
anything
when
it's
like.
We
want
like
we
don't
when
we
agree
on
a
filter
at
the
discussions.
I've
been
in
generally
focused
on
the
idea
that
the
metric
isn't
helpful
unless
you
can
filter
the
data
these
ways.
E
D
That's
where
I
am
and
I'm
Kevin
and
Kevin
daughter
persuading
me
to
think
about
it.
That
I
think
we
should
do
that.
I
mean
I,
think
about
it
way
too,
on
the
technical,
concrete
level,
sometimes
and
so
I
think
I'm
hearing
from
Kevin
and
Don
T's.
They
want
to
understand
the
metric,
that's
not
as
important
so.
E
I
can
tell
you
just
like
this
is
purely
anecdotal.
There's
no
research
behind
this,
but
when
I
talk
to
people
who
take
a
look
at
the
chaos
metrics,
you
know
who
are
kind
of
looking
at
the
metrics
oftentimes.
It's
just
I
was
taking
a
look
like
this
is
the
anecdotal
part.
I
was
taking
a
look
at
the
metrics
just
to
kind
of
see
what
what's
available
to
take
a
look
at
it's
not
about
how
I
would
implement
it.
E
It's
not
about
where
I
would
source
the
data,
but
it's
just
to
help
me
as
a
person
kind
of
get
my
head
around.
What
are
the
things
that
are
available
to
me
and
how
might
I
think
about
those
so
I
think
the
work
in
the
working
groups,
like
you
know
that
whole
question
metric,
but
the
focus
areas-
that's
been
extremely
helpful.
I
hear
a
lot
of
the
conversation
about
that,
but
I
it
usually
stops
them
out
there.
When
we
talk
about
metrics
and
then
the
whole
deployment
implementation
part
of
the
next
question.
E
So,
in
that
case,
it's
really
just
about
expressing
a
definition
here,
you're
the
things
that's
it
and
here's
why
you
would
use
it
like
the
objectives.
Here's
the
description
of
what
the
metric
is.
I
still
think.
Visualizations
are
great
because
here's
you
know,
while
you
could
take
a
look
at
this
metric,
meaning
the
implementations
to
air
you're
scratching
your
chin.
E
H
F
E
Can
I
am
I'll
make
a
proposal
just
kind
of
listening
to
everybody
here
it
doesn't
seem
like
anybody's,
like
completely
averse,
like
do
not
touch
the
structure.
It
doesn't
seem
like
anybody
has
that
concern
not
hearing
that
no,
but
what
the
optimal
structure
might
be
is
yet
to
be
determined.
That's
kind
of
also
what
I'm
hearing
I
don't.
E
To
make
that
campaign
promise
that
they
have
the
optimal
structure,
why
don't
I
propose
this
I'll
put
an
action
item
to
myself
because
I
don't
know
whoever
whoever
says
these
things
right.
If
you
give
somebody
something
in
front
of
them,
then
it
least
serves
as
a
starting
point
that
we
can
all
go
off
of
right
now.
Working
we
don't
have
something,
but
take
a
look
at
and
so
I'll
propose
a
structure
for
the
metric
lay
out.
E
The
definition
lay
out
based
on
what
I'm
hearing
today,
not
just
on
what's
in
my
mind,
but
what
I'm
hearing
today
I'll
also
propose
in
that
structure.
What
are
required
headings
so
basically,
some
cardinality
in
this
and
what
are
optional
subheadings
within
each
of
those
head.
Does
that
make
sense
so.
E
E
D
E
D
Only
thing
I
would
say
about
creating
some
headings
for
known
and
reference
implementations,
fine
I,
guess,
yeah
I
guess
we
know
what
I
know
what
a
reference
implementation
is.
I,
don't
know
if
everyone
on
the
column
do
it.
We
intended
it
to
be
prior
to
call
so
sometimes
like
what
these
headings
are
obvious,
like
description
and
objectives
other
times,
I,
don't
know
that
they
are
zhonya's,
particularly
frankly,
all
the
subheadings
under
implementation.
So
we
might
want
to
provide
some
guidance
about
what's
intended
under.
D
H
H
E
G
H
So
data
collection
strategy
and
I'm
gonna
call
it
data
collection
strategy,
because
when
we
talk
about
strategies,
that's
really
what
we're
talking
about,
how
we
collect
the
data
and
then
we
have
success
metrics,
and
this
is
where
it
becomes
a
little
challenging
when
diversity
and
inclusion
talks
about
success,
metrics
mm-hm-
that
is
what
a
metric
is
for
everyone
else.
So
I
mean.
H
E
So
then,
what
you're
taking
a
look
away
you
had
proposed
so
then
we
would
have
four
required
top
level
headings
for
every
metric,
which
is
description,
objective,
implementation
and
resources
period
and
I.
Think
objective
to
your
point.
Descriptions
and
objectives
are
the
same.
I,
think
they're
understood
the
same
between
the
working
groups
agreed
yes,
and
then
implementation.
H
H
G
E
G
B
G
B
D
E
It
yeah
I
was
gonna,
say
maybe
taking
a
little
I'm.
Sorry
Sean
go
ahead
and
finish
it
I'm
done
well.
I
cut
you
up,
then
that
didn't
mean
to
do
that.
Sometimes
I
should
be
tough
yeah.
So
maybe
in
this
list,
if
you're
looking
in
the
Google
Drive,
should
we
make
all
of
the
things
underneath
implementation
optional?
E
C
E
Okay,
so
then
another
so
then
another
comment
down
here
is:
if
we
have
these
two
known
implementation
and
reference
implementation,
I
think
we
have
to
be
particularly
around
known
implementation.
We
have
to
be
very
diligent
that
this
has
actually
been
implemented
in
a
tool,
not
that
it
can
be
implemented
in
a
tool,
but
that
it
has
been
implemented.
I
was
thinking
about
this
I
think
we
got
a
little
wax,
sometimes
that,
yes,
this
is
something
that
I
could
produce,
but
it
actually
hasn't
been
produced.
B
E
F
E
F
That's
just
my
thought,
like
I
mean
it's
great
that,
like
you
know
like
somebody
like
some
communities
actually
deployed
this,
but
I
mean
it's
good
to
know,
but
I'm
not
going
to
use.
It's
unlikely
that
I'm
going
to
use
exact
same
implementation
because
my
communities
are
and
I'm
using
different
data
source
or
whatever
the
reasons
right.
I
mean
I.
I.
Think
the
key
value
is
that
hey
here
is
sort
of
an
example
of
how
you
could
like
deploy.
This
I
mean
there
might
be
some
customization
needed
for
your
community,
but
I,
don't
I.
F
E
So
my
my
thought
here
was
on
visualizations
and
I,
hopefully
am
addressing.
That
visualizations
would
give
somebody
who
comes
to
the
metric
kind
of
like
what
you're
talking
about
just
a
way
to
kind
of
get
their
head
around.
What
it
might
look
like
mm-hmm,
you
know,
so
it
just
tears
a
visualization
from
augur
or
looser
visualization
from
your
lab
right.
The
the
known
implementation
is,
if
you
actually
want
to
deploy
this
kind
of,
as
is
grimore
lab.
Does
that
for
us?
So
that's
what
the
known
implementation
is
to
me.
E
So
the
visualization
is
something
that
I
could
just
look
at
quickly
on
the
document
to
just
kind
of
get
my
head
around
what
this
is
all
about,
then
there
might
be
something
to
be
said
for
knowing
that
this
metric
is
actually
deployed
in
all
your
kibble
or
more
lab
whatever
it
might
be.
What
do
you
think
yeah.
D
F
Yeah
all
right
or
the
other
way
we
can
do.
I
mean
I
mean
I'm
we
can
do
like.
We
can
just
call
it
implementation
and
use
a
generic
term
and
then
and
then,
if
it's,
if
it's,
if
it's
known,
then
we
can
note
that
right,
like
your
as
here's,
how
it's
actually
implemented
in
Jurmala
I've
been,
and
otherwise
you
don't.
You
don't
mention
it,
but
I
mean
I'm
okay
either
way,
but
I
don't
want
to
have
like
a
too
many
categories
under
implementation.
I
guess
that's.
The
only
thing
I
was
concerned
about
I
know.
E
D
E
Okay,
all
right
personally,
I
think
the
level
of
detail
that
we
need
is
a
sample
visualization,
that
if
somebody
is
coming
to
the
page
and
just
wants
to
get
their
head
around,
what
this
could
look
like
the
potential
for
it
and
then
just
links
to
the
tools
that
have
actually
deployed
this
thing.
This
thing
being
the
metric.
E
H
E
Okay,
well
all
on
I'll
at
least
put
together.
I'll
still
put
something
together
based
on.
Basically,
it
looks
like
I'm
just
gonna
copy
and
paste,
but
Geragos
typed
and
put
it
into
a
markdown,
something
that
we
can
talk
about
next
week.
Maybe
on
the
next
week,
is
actually
the
the
entire
monthly
call.
So
thanks
everybody
for
the
feedback
on
that.
I
did
have
one
other
things
so
I
had
reached
out.
If
you
recall,
we
had
talked
about
working
with
an
open
source
community
to
do
metrics
deployment.
E
Remember
this
so
actually
trying
to
deploy
the
chaos
metrics
as
published
and
I'd
reached
out
to
Kate,
and
she
was
happy
to
have
Zephyr,
be
part
of
that,
and
so
she's
connected
me
with
folks
at
the
Zen
project
as
well,
but
I'll
have
to
repaint
them.
So
so
the
premise
for
those
of
you
that
don't
know
is
we're
really
trying
to
take
a
look
at
how
the
metrics
that
we
have
released
in
version.
A
E
E
F
F
H
F
H
F
F
F
E
F
Sorry
yeah,
that
was
just
quick.
B
H
A
E
G
G
E
D
D
G
E
B
B
G
G
E
B
D
B
It
is
it's
super.
It's
super
helpful
and
I.
Do
that
with
loads
of
calendars,
I
have
like
kubernetes
calendars
and
some
other
things
like
that.
But
what
it
doesn't
do
is
well
well,
it
doesn't
block
your
calendar
off
so
me
as
an
employee
of
a
company
if
I
just
use
this
calendar.
People
are
gonna
book
me
in
meetings
right
over
top
of
these.
So.
C
B
To
have
it
separately
on
my
calendar
as
an
actual
calendar
item
on
whatever
calendar
people
try
to
schedule
me
on
on
and
that's
what
I
try
to
do
for
other
people
and
that's
why
I
invite
people
to
the
to
the
calendar,
partly
so
that
they
always
remember
it,
but
also
because
if,
if
they're
invited
and
that's
the
calendar
that
they
use
for
work,
then
people
won't
book
meetings.
Over
top
of
the
I
see.
D
Yeah
so
like
Kevin
I
think
it
would
be
at
all
to
give
like
basically
everyone
on
this
call,
or
at
least
all
the
working
group
chairs
access
to
invite
people
from
this
calendar,
because
that
would
solve
Don's
problem
and
take
it
off
of
your
list
of
responsibilities.
Would
that
solve
your
problem?
Done
I?
Don't.
D
I'm
sharing
my
screen
again,
it's
only
the
cast
calendar,
but
if
I
had
edit
rights
on
this
meeting,
I
could
then
edit
the
list
of
invitees
and
I
could
invite
all
the
people
that
I
normally
expect
that
normally
want
to
be
at
he's
working
for
meetings.
The
ones
in
my
case
are
evolution
and
risk
that
that
I
have
a
set
of
people
that
I
have
invited
to
a
meeting
on
my
calendar
and
so
giving
a
certain
number
of
people
of
their
rights
to
edit.
I
I
C
B
C
B
D
B
Probably
overkill
for
for
this
one,
but
because
because
all
the
meetings
for
kubernetes
are
tied
to
specific
special
interest
groups
and
each
special
interest
group
has
its
own
mailing
list.
When
you
join
a
mailing
list.
You
get
invites
to
all
of
the
meetings
for
that
special
interest.
Group
I
recently
joined
the
cluster
life
cycle
special
interest
group
and
got
like
a
bajillion
meetings
because
they
have
a
whole
bunch
of
sub
projects
and
many
many
meetings
and
those.
And
you
get
that
as
an
invite
from
from
some
calendar.
B
B
G
B
Do
which
is
probably
how
that
works,
but
it's
also
kind
of
overkill
for
a
project
the
size,
because
we
don't
have
a
million
special
interest
groups
and
and
there's
a
specific
URL
that
people
are
supposed
to
invite
to
the
calendar,
invite
that
puts
it
on
the
main
calendar,
and
sometimes
people
forget
to
do
that
where
actually
I
yeah
I
just
wrote
some
Python
to
like
pull
things
out
and
do
like
a
comparison
of
which
ones
are
where
and
we've
got,
we've
got
a
bit
of
a
mess.
So
it's
not
it's
not
perfect
for
sure.
We.
B
B
B
I
love,
I,
love,
Kevin's,
idea
of
having
a
calendar
that
you
can
put
on
your
calendar
because
it's
nice
that
I
can
see
all
of
the
other
meetings.
If
there's
one
that
I
occasionally
want
to
drop
in
and
out
of,
I
still
think
that
kind
of
the
people
leading
the
working
groups.
If
you
have
some
regular
attendees
having
it
on
your
calendar,
I
think
I
think
helps
and
you
can
invite
those
people
or
as
a
working
group.
You
can
decide
not
to
do
that.
B
H
B
B
I
I
will
say
the
only
thing
that
groups
at
I/o
does
slightly
nice
in
this
is
that
if
there
are
changes
that
are
made
on
the
main
calendar,
because
they
basically
you're
basically
doing
the
same
thing
here,
but
they
do
do
notifications
if
meeting
invites,
have
changed.
But
it's
not
everybody's
clients
a
little
bit
wonky
of
how
it
sucks
that
in
so
just
just
not
to
throw
out
more
trolling
ideas,
but
just
throwing
that
out
there.
Okay.
G
D
B
H
A
H
Don't
know
if
we
want
to
now
that
we
have
the
calendar
remove
all
of
our
different
working
groups
and
software
calls
and
everything
just
have
the
calendar
as
the
source
of
information
on
where
to
find
the
meeting
minutes
how
to
join
or
if
we
still
want
to
maintain
it.
And
then
we
should
add
the
software
under
us.
I
know:
Green
Moore
Lappe
has
its
own
menu
list
that
is
currently
not
on
the
participant,
page
and
I
believe
augur
has
a
slack
channel
that
is
currently
not
on
anywhere
on
our
website.
I.
E
B
E
H
D
D
E
Party
lunch
lunch
time
so
next
next
week
is
the
monthly
call
just
FYI.
So
if
you
have
agenda
items
that
you
want
to
bring
up
in
particular
just
send
them
my
way.
Okay
and
then
I'd
also
just
like
to
say
that
just
one
last
thing
Georg
has
been
elected
as
co-chair
generationally
totally
and
the
two
new
board
members
are
Daniel
and
Nicole
regulations.
Yes,
all.