►
From YouTube: CDF SIG Best Practices - Nov 15, 2021
Description
For more Continuous Delivery Foundation content, check out our blog: https://cd.foundation/blog/
B
C
So
I'm
up
to
my
third
test
for
that
one
event
at
the
moment,
but
yes,
feeling
symptom
three
so
cross
fingers.
It
was
someone
I
didn't
interact
with
like
linus
or
old.
C
C
C
D
C
So
yeah
I'm
happy
to
go
first,
just
a
bit
of
an
update.
C
C
What
are
some
of
the
most
strategic
things
we
should
be
focused
on,
so
we
did
a
bit
of
a
brainstorm,
went
around
the
table
and
asked
everybody
what
they
thought
were.
The
key
focus
areas
or
the
different
cdf
could
make,
and
then
we
distill
that
down
to
common
areas
and
try
to
get
some
consensus
and
pretty
much.
We
were
able
to
focus
on
deeper
discussions
in
three
areas
and
best
practices
emerged.
As
you
know,
one
area
that
there
was
general
consensus
that
folks
felt
it.
C
It
was
really
key
to
cdf
to
be
to
have
this
dynamic
body
of
work
that
does
evolve
and
this
that
does
speak
to
best
in
emerging
practices.
C
So
it
we
presented
the
deck
you
put
together
again,
nikola
because
it
turned
out,
even
though
we
have
shown
it
to
the
strategy,
call
a
lot
of
people
hadn't
seen
it
or
hadn't
acknowledged
it,
but
in
general
the
feedback
was
was
just
that
it's
great
people
love
the
the
way
it
breaks
down
the
different
perspectives
and
it
shows
an
acknowledgement
of
what's
needed
at
different
stages.
C
So
long
story
short
folks
were
generally
supportive
and
the
discussion
was
like
how
do
we
support
this
effort?
And
how
do
we
get
it
to
the
next
stage
and
that's
something
we
want
to
have
follow-up
discussions
because
it
feels
like
people
don't
necessarily
know
how
to
engage
or
just
telling
them
to
come
along
to
this
meeting?
Isn't
quite
working.
C
C
There
was
a
suggestion
as
well
that
we
should
brand
it
very
strongly
so
give
it
a
specific
name
or
something
a
bit
more
sticky,
so
it
has
a
way
of
referencing
it
that
people
can
latch
on
to,
and
I
don't
know
anything
we
can
do
to
to
brand
it
and
make
it
a
thing.
So
we
just
refer
to
you
know
not
just
cdf
best
practices,
but
whatever
that
I'm
not
gonna
make
up
names
now,
but
if
that
makes
sense
of
just
having
some
strong
branding
around
it.
C
So
it's
a
thing
and
people
refer
to
that
as
as
a
specific
thing
yeah.
So
I'm
basically
all
good,
and
my
job
is
just
to
figure
out
how
to
move
that
forward
and
how
to
connect
the
right
people
here
so
I'll
leave
it
there.
Any
questions
or
thoughts
related
to
that.
D
The
the
branding
is
an
interesting
idea.
The
the
first
thing
that
comes
to
mind,
though,
is
that
it's
there's,
like
you
know,
branding
that
you
do
because
it
is
work
we've
created
versus
branding
versus
what
we're,
what
we're
actually
doing,
which
is
we're
not
creating
work
as
much
as
we're
curating
it,
and
so
I
just
want
to
make
sure
that
we
avoid
inadvertently
some
form
of
broader
authorship
around
some
of
these
concepts
like
we're,
pulling
heavily
from
accelerate
and
state
of
devops
and
we're
pulling
heavily
from
other
sources.
D
We
want
to
encourage
people
to
participate
and
bring
their
ideas
for
consideration
and,
as
we
build
up
this
body
of
work,
we
don't
want
them
to
feel
like
oh
well
now
it's
you
know
copyrighted,
cdf
or
whatever,
not
of
course,
that
we
would
do
that.
But
I'm
just
saying
that
the
approach
has
to
be
very
thoughtful.
A
D
And
honestly,
I
think
we
could
turn
that
as
a
strength
right
with
treating
it
like
proper
research,
with
citations
and
and
and
also
you
know,
as
a
way
of.
D
Maintaining
that
balance
that
we
talked
about
before,
which
is
to
really
make
this,
be
the
neutral
content
and
using
like
hey
this
is
you
know,
you're
trying
to
bring
marketing
stuff
into
this
space?
Show
your
work.
You
know,
where's
your
research
that
shows
it's
the
best
practice
that
kind
of
stuff.
I
think
we
could
have
a
really
interesting
approach
to
it.
C
I,
like
that,
and
I
think
our
strength
as
as
kind
of
the
vendor-neutral
body
is,
is
where,
where
we
focus
on
like
we're
just
bringing
everybody
together
and
making
it
clear
how
everybody
can
contribute
and
everything
is
led
by
principles,
not
by
anything
else,
principles,
research.
E
I'm
just
going
to
say
plus
one
for
references,
I
think
they're
great,
for
people
who
are
essentially
end
users
of
the
stock,
because
it
enables
them
to
go
deeper
when
they
need
to
and
then
implicitly
really
helps
encourage
contributions.
Because
people
know
that
when
they
contribute,
you
know
their
name
and
their
work
will
be
on
it
and
it's
kind
of
a
version
of
self-promotion.
But
it's
good.
C
B
I
had
something
to
read:
yeah
I
didn't
have
a
ton
of
updates
last
week,
not
really
much
anything
I
I
did
have
doxie.
I
did
get
doxy
running
locally
on
a
vm.
If
you
want,
I
can
share
my
screen
and
kind
of
show
a
very,
very,
very
rough
skeleton
of
how
it's
working
so
far
go
for
it
all
right
and
once
again
I
don't
do
zoom,
so
I
have
to
figure
out
where
this
is.
E
B
There
we
go
now,
you
can
hear
me
right:
okay,
ignore
the
ugly
colors
and
stuff
and
placeholder
everything.
So
there
was
an
example
site
basically,
and
it's
it's
themed
around
goldilocks,
so
you'll
see
lots
of
like
weird
folders
yeah
here
and
ignore
the
ugly
colors
I
removed
the
picture
porridge
and
suddenly
this
thing's
now
this
orangey
brown.
But
the
idea
here
is
that
we
have
some
basic
navigation.
So
here
we
can
kind
of
have
you
know
what
is
this
site
about?
B
Basically
and
that
we
have
these
sections
here
and
you
know
we
have
a
place
where
you
can
learn
about
the
vendor
neutral
stuff.
We
have
a
community
where
we
have
more
specific
implementations
that
you
can
look
at
and
then
last
week
I
think
we
had
also
talked
about
having
you
know
some
specific
resources
around
things
like
you
know,
accelerate
some
of
the
other
books
that
are
in
progress
and
coming
out.
So
that
could
be
a
sort
of
centralized
place
here.
B
The
learn
area,
then
we'll
have
some
of
these
sections
that
we
have
in
our
draft
stock.
So,
like
the
version
control,
this
section
here
would
be.
You
know
again.
Placeholder,
but
the
idea
here
is,
we
would
talk
about.
You
know
if
you're
looking
at
organizational
maturity,
we
have
this
stuff
in
terms
of
dora
and
you
can
go
and
assess
your
maturity
and
what
areas
you
might
want
to
look
at
then
come
back
here.
B
It
can
also
reference,
for
example,
from
a
security
perspective,
the
emerging
salsa
framework
for
looking
at
supply
chain
security,
which
is
really
relevant
in
this
continuously
space
and
any
sort
of
other
assessment
tools
that
we
think
might
be
relevant
could
go
here,
and
then
we
can
provide
some
guidance
about
what
sections
you
might
want
to
look
at,
based
on
the
results
of
those
assessments
and
then
yeah
I
put
in
some
placeholders
for
these
various
they're.
B
Not
some
of
them
are
not
in
order
because,
but
you
can
see
I
have
like
there's
some
stuff
I
had
mentioned
in
my
slide
deck,
maybe
some
stuff
around
team
culture.
The
other
thing
I
saw
that
I
really
feel
like
kind
of
spans,
a
lot
of
these
areas
and
I'm
not
sure
how
we
wanted
to
address
it,
whether
it's
in
specific
sections
or
we
have
a
dedicated
section,
but
is
things
like
feedback
and
measurement
and
observability
and
all
that
kind
of
stuff?
B
C
So
I
do
think
we
have
referenced
that
in
the
the
doc,
but
maybe
we
haven't
put
it.
I
don't
recall
if
it's
under
specific
section,
but
I
do
know-
we've
had
a
lot
of
discussions
about
measuring
devops
success
and
yeah
conversations.
Yeah.
B
So
in
the
practice
doc,
we
kind
of
have
this
consistent.
Why
definition
description,
scope,
best
practices
and
then
some
stuff,
like
a
specific
case
study
with
tools
and
stuff?
And
I
guess
maybe,
if
we
have
some
kind
of
specific
heading,
around
yeah,
measuring
success,
measurement
measurement
or
feedback,
or
something
so
and
not
even
necessarily
success,
but
but
measuring
performance,
yeah,
yeah
and,
interestingly
enough.
C
On
the
other
part
of
cdf,
with
the
landscape,
we
were
having
these
discussions
about
landscape
categories
in
the
new
cars
actively
trying
to
figure
out
how
to
represent
those,
and
the
words
are
measurements,
metrics
and
all
the
way
down
to
observability
and
feedback
such
as
trying
to
position.
The
in
things,
which
overlap
a
lot
into
black
and
white
categories
is
definitely
interesting.
E
It's
definitely
complicated,
but
one
one
of
the
ideas
that
one
of
the
ways
in
which
I
think
about
this
is
really
like
a
long
chain
of
checking
for
sort
of
quality,
all
the
way
from
dora
metrics
to
testing
to
you
know
looking
at
your
pipeline
to
observability
when
in
production
and
are
you
thinking
for
the
best
practices
docs.
However,
we
call
them.
Are
you
thinking
of
having
them
all
together
or
are
you
breaking
them
into
separate
categories.
B
So
I
am
basing
these
placeholders
on
that
and
then
there
are
a
couple
extra
things
that
I
thought
I
wanted
to
discuss.
Things
like
like
team
culture,
things
like
feedback
and
and
measurement
where
that
really
is
something
that
you
might
from
an
implementation
perspective.
It
could
fall
in
any
of
the
sections
and
arguably
we
have
a
section
for
security,
but
arguably
that's
something
that
you
should
be
incorporating
into
everything.
C
E
Yeah,
okay,
so
you're
keeping
observability
and
testing
quite
separate.
Then
you
know.
C
C
B
C
And
I
think
yeah
we
have,
I
don't
know
if
it's
in
this
group
or
elsewhere
we
so
it's
it's
the
netflix
model.
They
talk
about,
develop,
deliver
operate,
and
so
we
ignore
the
develop.
We
don't
talk
about
architecting,
we
don't
talk
about
ids
and
we
to
some
extent
the
observe
we
don't
talk
about
observability.
We
don't
talk
about
site,
reliability,
engineering
as
much,
but
it
lots
of
people
intertwine
them.
So
we
might
have
to
be
more
explicit
if
that's
out
of
scope
or
not
yeah,.
B
Well
and
that's
where
we'd
have
to
tie
in
things
with
the
assessment
as
well
right,
because
if
you
go
to
over
to
dora,
it
does
cover
the
gamut,
and
so
if
what
you're
missing
are
things
like
instrumenting
observability,
we
might
come
come
back
here
and
say:
okay,
look
at
your
results,
and
this
is
the
stuff
we
cover
here
and
if
you're
looking
at
things
like
observability,
there
is
some
more
deep
dive
into
that
somewhere
else.
Right,
yeah.
A
So
I'm
a
little
bit
worried
that
we
might
be
looping
back
around
to
where
we
came
in,
because
you
know
we're
we're
talking
about
a
lot
of
things
which
are
really
trying
to
break
down
continuous
delivery
into
a
hierarchical
model
that
fits
into
a
website
and-
and
that
has
some
problems
associated
with
it,
because
it
it
makes
it
look
like.
A
A
We
need
to
make
sure
that
we're
structuring
this
in
such
a
way
that
we're
presenting
it
for
different
audiences
and
it's
easy
as
as
soon
as
you
come
into
the
landing
page,
to
be
able
to
pick
a
path
through
the
information
that
is
speaking
to
to
your
concerns,
rather
than
potentially
bamboozling
you
with
a
whole
bunch
of
technical
terminology,
you
don't
care
about
so
so
so
that's!
I
think,
that's
important
that
we,
you
know
we
think
about
that
again
and
and
how
we
can
get
that
journey.
A
You
know
in
this
example,
if
I,
if
I've
come
to
the
site,
because
my
concerns
are
all
about
the
neighbourhood
home
invasion
problem,
then
you
know
getting
a
detailed
picture
of
the
bears.
Tax
position
is
is
not
what
I
want
to
see.
Yeah.
C
So
nicola,
just
going
back
to
your
about
page
before
you
came
into
this
section,
you
showed
us
so
there's
the
landing
page
like
the
front
page,
and
then
you
had
a
was
it
about
yeah
and
this
would
kind
of
be
10.
000
foot
view
yeah.
It
feels
like
a
lot
of
the
tying
everything
together
for
me
would
fit
there
or
there
should
be
something
there.
That's
the
very
high
level.
C
B
This
structure
tends
to
be
very
hierarchical
and
people
could
come
in
from
anywhere
and
so
a
possibility
is
that
we
do
have
sections
for
things
like
light,
metrics
and
measurements,
and
we
have
a
page
that
kind
of
talks
about
the
large
picture
of
it,
and
we
may
not
get
into
details
about
things
that
that
is
of
observability,
but
we
we
can
link
into
various
places
that
might
cover
some
areas
of
that
throughout
the
docs,
like
not,
everything
has
to
be
linear
and
hierarchical
in
terms
of
how
we
how
we
build
this
out.
B
Might
not
have
a
lot
of
prescriptive
content
but,
for
example,
something
like
security.
It
really
is
sort
of
this
horizontal
thing
that
goes
across,
so
it
would
be
an
example
of
I'm
coming
in
and
I'm
responsible
for
say
compliance
with
my
within
my
organization
and
might
be
interested
specifically
in
the
security
stuff.
So
we
come
into
the
security
and
the
security
might
actually
point
into
some
of
the
best
practices
within
testing,
for
example
yeah.
So
this
overlap.
B
How
we
handle
some
of
these
sort
of
horizontal
kinds
of
things?
So,
even
though
we
might
have
these
detail
sections
and
the
best
practice
docs
that
cover
things
like
like
ci,
ci
and
cd,
maybe
we
have
some
sections
here:
maybe
we
even
have
like
yeah?
Maybe
we
have
some
sections
at
the
top
level.
I
don't
know
how
we
want
to
do
this,
but
that
covers
some
of
these
things
that
are
broader
and
more
spanning.
A
Let
me
say
this
a
different
way.
This
weekend's
twitter
topic
has
been
how
everybody
is
doing
devops
as
an
anti-pattern
by
setting
up
a
devops
team.
C
D
Can
you
hear
me
it's
just
coming
through
yeah,
hello,
okay,
sorry,
I
just
now
have
the
coffee,
so
I
know
that
there
was
some
concept
of
of
this
discussion
previously,
but
there's
I
kind
of
want
to
dig
in
on
what
terry
was
saying
a
little
bit,
because
I
think
there's
two
aspects
to
it.
There's
the
here
is
what
it
means
in
order
to
have
high
performance.
D
You
know
if
we're
going
to
use
the
accelerate
language
right
here
are
the
capabilities
that
you
have
enabled
right
here
are
the
philosophical
approaches
that
neces
that
are
necessary
in
order
to
enable
this
here
are
here
are
examples
of
the
mechanics
of
it
right
and
it-
and
this
goes
back
to
the
you
know
what
I
was
saying
about
the
landscape.
Is
that
being
able
to
like
break
things
out
into
categories
really
trips
over
the
idea
of
continuous
delivery?
As
this
ongoing
cycle,
it's
a
loop
and
it's
not
even
a
single
loop.
D
So
we
need
to
figure
out
what
that
sweet
spot
is
and
then
have
the
redirects
right,
and
this
is
where
tracy,
I
think
you
know
I
keep
coming
back
to.
How
can
we
build
these?
These
interesting
bridges,
for
example,
with
open,
telemetry
right
where
it's
like?
Okay,
if
you
really
want
to
dig
here,
there's
a
redirect
from
their
side.
D
You
know
it's
like.
If
you
really
want
to
understand
the
cd
stuff
the
redirect
comes
from
that
direction,
because
I
don't
think
we
can
completely
not
talk
about
it
right
up
elements
of
the
continuous
delivery
workflow
will
be
based
on
if
you're
doing
it
right
the
the
structured
reactions
to
what's
happening,
triggering
things
in
your
pipeline
so
like
the
whole
idea
of
a
canary
deployment.
How
do
you
know
whether
or
not
the
canary
is
still
alive
well
because
of
the
instrumentation
that
you
have
in
production?
D
That
tells
you
that
things
are
going
well
right
so,
and
in
order
for
that
for
that
knowledge
to
occur,
then
you
have
to
have
some
form
of
observability
in
place
right.
So
that's
so
I
think
there's
and
nicola
kind
of
kicked
me
virtually
if
I'm
saying
this
wrong,
but
like
there's
this
element
of
navigation
like
at
any
point
of
this
journey
like
as
you're
in
this
site,
there
should
be
an
easy
way
to
go
to
related
concepts
right.
D
I
don't
know
that
it's
a
left,
nav
thing
yeah
a
tagging
thing
right
what
what
the
best
way
to
do
that
was,
but,
but
I
think
that
the
the
overall
sort
of
organizational
or
persona
structure,
however
we
want
to
describe
it-
needs
to
be
pervasive
because
it's
the
spring
loop
of
overlapping
identity.
So
it's
like
not
even
do
we
not
want
this
to
be
linear,
I
don't
think
it
can
be
and
be
successful.
B
I
I
completely,
I
completely
agree
with
that
when
people
like
most
of
the
time
when
people
come
to
docs,
it's
because
they
went
to
a
search
engine
and
typed
a
query
and
landed
in
somebody's
documentation-
and
you
know
even
even
google's
documentation,
they
aren't
necessarily
searching
within
our
doc
sets.
They
are
often
coming
from
a
search,
and
that
means
they
are
landing
anywhere
and
they
aren't
that's
when
they
might
become
aware
of
how
our
table
of
contents
are
organized.
So
it's
an
organizing
structure.
B
It
helps
people
to
kind
of
see
how
we've
thought
about
how
these
things
are
related,
but
I
completely
agree
that,
in
within
some
of
the
pages
that
we
make
call
outs
to,
why
are
we?
Why
are
we
doing
this
particular
instrumentation,
as
part
of
as
part
of
you
know,
continuous
integration,
because
we
want
to
be
able
to
look
at
it
later
so
that
we
have
feedback
and
that
we
do
then
have
these
call
outs.
You
know
what
why
are
we?
B
Why
are
we
being
careful
about
how
we
use
third
party
repos?
Because
we
want
to
be
concerned
about
security,
so
integrating
some
of
that
stuff
and
not
necessarily
deep
diving
in
those
places,
but
then
pointing
you
whenever
we
need
to
yeah
yeah.
A
And
maybe
we
need
a
slightly
clearer
problem
statement
for
this
whole
piece
of
work,
because
if
we,
if
we
think
about
it
best
practices
means
different
things
to
different
people,
but
in
in
practice
I
I
don't
think
that
the
organizations
who
are
already
doing
continuous
delivery
well
are
going
to
immediately
come
to
this
resource,
to
use
it
as
a
metric
by
which
to
score
themselves,
because
they
already
know
this
stuff.
They
understand
it
and
they're
running
it
to
a
high
standard
they're
getting
the
feedback.
A
So
that
means
our
real
audience
is
the
people
who
have
yet
to
successfully
adopt
that
and
if
they
haven't
already
adopted
it
it's
because
there
there
is
a
a
a
problem
in
there
understanding
the
value
or
the
model,
and
and
probably
the
reality
of
that
situation
is
that
you're.
Looking
at
an
organization
that
has
you
know
an
organizational
process
that
they
built
under
prince2
in
about
1995
and
and
they're
approaching
the
problem
saying
we
want
our
prince2
waterfall
process
to
go
faster.
A
Therefore,
how
can
we
take
what
we're
already
doing?
Call
it
devops
and
get
a
silver
bullet
that
will
speed
us
up
and,
of
course,
that
that's
a
that's
a
real
challenge,
because
that's
that's
never
going
to
work
and
they
need
to
know
that
up
front.
A
Otherwise,
we
go
into
this
cycle
of
people
saying
they're
doing
devops,
but
they're
never
actually
doing
any
of
the
continuous
delivery
processes
and
then
failing
and
then
having
devops
become
a
banned
word
because
you
know
it
didn't
work
and
therefore
you're
not
allowed
to
speak
for
it
again.
C
Think
we
articulated
the
problem
statement
well
and
I
think
we
need
to
go
back
to
it,
because
I
remember
that
table
from
your
presentation
nicola.
We
could
almost
take
that
and
we
have
the
you
know,
do
I
buy
into
cd
and
for
that
one
it
was
yes,
I
do
and
then
the
journey
was.
How
do
I
share
what
I'm
doing,
and
I
think
so,
I
think
the
site
is
good.
It's
a
it's
a
good
start
and
we
just
need
to
go.
C
D
Also
argue
terry
at
some
point:
if
we,
if
we
get
to
the
point
where
we
have
attestations,
because
you
want
to
bring
your
project
in
you
know
all
screwdriver,
shipwright
or
whatever,
like
people
hold
up
like,
does
it
meet
these
criteria?
That
would
be
awesome.
That
would
be
a
great
problem
to
have.
A
B
No,
it's
it's!
This
is
a
different
deck.
Okay,
but
some
of
these
these
the
point
of
this
deck,
is
that
there
are
a
lot
of
these
buzzwords
here
that
we
are
familiar
with.
If
we're
in
the
software
development
space
and
a
lot
of
executives,
don't
know,
don't
care
about
their
buzzwords.
B
B
A
B
Look
at
it
from
this
perspective
that
you're
going
to
look
for
what
it
is.
That's
not
working
and
maybe
you've
already
identified
that
or
maybe
you
need
to
do
some
of
this
value
stream
mapping
stuff-
and
this
is
what
I
mean
this
deck-
is
pitching
dora.
But
the
idea
is
that
you're
coming
here
with
a
problem,
and
so
there
they
run
through
an
example.
Here.
I
think
this
is
a.
We
want
things
to
go
faster
problem
and
then
work
for
that,
but
the.
A
B
Is
that
you're?
Oh,
did
you
like
that
infrastructure
yeah?
So
the
idea
there,
though,
isn't
it
and
then
the
next
one
is.
D
D
B
Yeah,
so
you
know
it's
this
incremental
thing
rather
than
trying
to
fix
all
the
things
right,
but
the
point
is
that
the
idea
is,
you
might
be
coming,
and
you
might
already
know
that
what
you
want
is
to
make
things
go
faster.
But
what
is
what
is
it
that
you
want
to
make
go
faster
right
and
trying
to
walk
people
through?
B
Don't
try
to
boil
the
ocean?
Okay,
so
you
want
to
make
things
go
faster
to
make
things
go
faster.
What
part
of
it
do
we
need
to
make
go
faster,
here's
a
way
you
can
assess
what
might
be
causing
your
bottleneck
and
the
reason
why
things
aren't
going
faster,
or
at
least
the
top
reason
why
things
aren't
going
faster
fix
that
first
and-
and
so
that's
one-
that's
one
way
we
can
pitch
it,
but
that's
kind
of
what
I
had
in
mind.
B
A
That's
needed
to
drive,
continuous
delivery
and
get
buy-in
for
that
at
an
executive
level
and
funding
and
and
the
push
to
reorganize,
potentially
large
bits
of
a
company
in
order
to
then
enable
the
technical
delivery
teams
to
to
do
their
their
part,
and
I
think
the
challenge
that
we
have
is
that
if
we,
if
we
jump
too
quickly
into
the
you
know
the
technical
nuances
of
continuous
integration
with
that
audience,
they'll
immediately
switch
off
and
they'll.
They
won't
get
the
message.
A
So
we
need
to.
We
need
to
be
able
to
communicate
this,
the
cyclical,
continuous
delivery
process
in
business
language,
because
that's
actually
our
target
market
for
for
this
information.
D
I,
like
the,
and
I
think,
I'm
having
some
sound
problems,
I'm
maybe
coming
on
a
delay,
but
I
like
what
tracy
was
saying
earlier
around
what
was
tracy
saying,
but
oh
the
workflows
right.
So,
if
we
can
say
here
are
the
here
are
those
concepts
here
are
the
benefits
that
you
get?
Here's
how
you
got
that
benefit
right
with
with
some
concrete
examples
and
basically
extract
some
of
the
data
from
door
and
then
hopefully
get
more
data
from
partner
teams
within
amber
teams
within
cdf.
D
We
could
try
to
fill
out
and,
and
have
it
be.
A
data
driven
discussion
like
here
are
some
key
examples
of
what
what
you're
talking
about
terry
and
and
basically
have
them
as
like
white
paper,
or
you
know,
whatever
the
appropriate
forum
is
like.
Okay,
now
that
we
have
your
interest,
you
know,
go
get
your
tech
team
and
unleash
them
and
here's
the
directions
that
they
can
go
right.
Yeah,
but.
A
D
A
D
D
A
C
But
they
do
so,
someone
shared
this
with
me:
that's
minimum
viable
cd
and
it's
a
few
folks.
I
think
dave,
foley
and
fox
and
they've
put
down.
C
You
know
kind
of
headers
and
things
and
then,
when
I
see
that-
and
it
looks
like
we're
trying
to
build
something
similar,
but
it
doesn't
like
seeing
that
again,
it's
it's
a
list
of
things,
but
without
this
business
reason
tying
it
together.
So
it
feels
like.
Maybe
that's
what
we
are
on
track
to
to
do
the
same
or
more
of
the
same
off.
A
So
the
thing
that
I
that
we
keep
hitting-
and
it's
it's
here
again
it
says
cd-
is
the
engineering
discipline
of
delivering.
No,
it
isn't
continuous
delivery
is
a
business
methodology
for
delivering
products.
Okay
and
a
part
of
that
is
in
using
the
engineering
capability
to
automate
and
accelerate
that
process.
A
C
So
going
back
to
what
I
consider
our
problem
statements
where
we
have
these
different
personas
and
we're
building
towards
that.
It
does
feel
like
we
have
this
question,
but
we
don't
specify
whether
that's
the
engineering
perspective
or
whether
that's
the
business
or
both,
and
so
maybe
we
do
need
just
to
re-evaluate
these
and
come
back
in
because
like
what
I
feel
nicola,
but
that
does
definitely
cover
some
of
the
cases.
Maybe
I
want
to
adopt
cd
and
then
what
are
the
best
practices
for
certain
things,
but
there's
a
lot
more.
C
A
Yeah
and
it's
almost
it's
a
it's,
a
user
story
problem
really
yeah.
It's
you
know
as
ceo.
I
need
to
you
know
and
as
product
manager
I
need
to
it's.
It's
answering
those
questions
that
that
defines
the
structure
of
the
information
that
we
need
to
get
across.
C
As
another
dimension
this
we
have
a
new
rep
for
cdf
from
ebay
and
it's
someone
called
justin
abrams,
and
it's
really
interesting
because
he
more
than
anyone
I've
met
in
the
community
is
working
with
an
end
user
company,
but
not
so
much
on
the
engineering
side,
but
more
on
he's
trying
to
help
the
teams
work
out
what
they
should
be
doing
and
connect
that
to
leadership,
and
so
he
was
interested
in
putting
together
an
assessment
like
he
wants
an
assessment
across
ebay
where
they
assess
themselves.
So
they
know
for
all
the
different
teams.
C
We
have.
What
should
each
team
be
doing
so
maybe
he
might
be
a
great
person
to
bring
into
this
conversation
and
have
he
might
be
like
one
of
the
people?
Who
has
that
specific
perspective
of?
Would
this
be
useful?
What
do
you
need
to
have
those
conversations
with
leadership
about
what
each
team
should
be
doing
and
how
they
should
be
dealing
with
their
constraints,
so
I
might
suggest
hey.
C
Sounds
good
so
I'll
I'll
reach
out
and
see
if
he
can
make
this
time
or
we
can
set
up
something
specific
at
a
time
that
works
for
him.
B
Ahead,
go
ahead.
I
was
just
going
to
say
that
table
that
tracy
was
sharing.
I
mean
a
lot
of
that
was
stuff
that
that
jan
and
I
had
put
together
and
and
some
guesses
around
what
people
would
be
coming
in
looking
for,
but
that
that
was
the
reason
we
had
been
kind
of
probing
for
for
some
user
personas
and
like
what
are
the
you
know,
top
couple
of
things
that
they're
asking.
B
A
So
I
think
it
it
might
help
also
to
have
a
look
at
some
of
the
the
views
that
external
organizations.
You
know,
non-technical
organizations
have
on
this
perspective
and
the
best
way
to
do
that
is,
go
and
have
a
look
at
some.
Some
job
adverts
for
devops
developers
in
finance
say
because
what
you'll
get
is
senior
devops
developer
and
then
must
have
project
estimation,
skills
experience
with
gantt
charts.
A
A
So
there's
a
big
communication
problem
out
there
right
now
that
devops
is
a
magic
wand,
but
everybody
has
a
completely
different
view
of
what
devops
actually
is,
and
you
know,
there's
massive
miscommunication
going
on
outside
of
a
very
small
number
of
organizations
that
that
get
the
process
because
it
matches
their
organizational
process
at
a
business
level.
E
Yes,
but
it
it's
a
bit
of
a
reversal
to
what
you
were
saying:
tracy
about
justin
abraham,
so
it's
kind
of
skipping
over
terry's,
very
good
comment
right
now,
but
I
guess
we'll
get
back
to
that
from
my
understanding
of
his
what
he
was
describing,
what
he
was
doing
at
ebay.
E
They
are
using
the
door
metrics
widely,
but
what
he
would
like
to
create
internally
and
maybe
bring
to
the
public
you
know
in
open
source
is
something
that
sounded
a
lot
more
like
the
salsa
framework
like
something
really
very
clear-cut,
you're
at
level
zero.
If
you
do
these
you're
level,
one
you're
level,
two
and
and
sort
of
almost
almost
like
a
maturity
model
or
some
way
of
thinking
of
it.
E
That
way-
and
we
did
speak
about
how
previous
door
research
has
suggested,
as
you
were
saying
nicola,
where
you
have
like
a
set
of
capabilities
and
as
teams
you
look
at
where
your
bottlenecks
are
and
you
move
forward
there
and
choose
your
own
adventure
type
way.
So
it's
kind
of
two
different
approaches-
and
I
just
wanted
I'm
just
asking
out
of
curiosity
like
how
do
we
feel
about
that?
Do
we
feel
that
both
are
useful?
Do
we
feel
like
more
one
is
useful
like
how
are
we
thinking.
A
Well,
if
you,
if
you
have
a
a
a
very
rigid
business
process,
which,
for
example,
requires
large
amounts
of
legal
sign
off
on
every
change,
then
that's
always
going
to
prevent
you
from
accelerating
your
release
cycle.
You
won't
be
able
to
gain
any
of
the
the
real
benefits
of
this
cyclical
process
and
you
know
there
are
going
to
be
some
big
ones
in
here.
A
So,
for
example,
if
you're
delivering
machine
learning
applications
to
european
customers
every
time
you
do
a
release,
you're
going
to
have
to
get
a
third
party
to
come
in
and
assess
your
code
base
for
legislative
compliance,
so
that
will
probably
be
a
six
to
12
month
process
for
every
release.
B
I
think
that
comes
back
to
what
is
the
definition
of
devops,
though,
because
if
you
look
at
something
like
the
door
capabilities,
things
like
processes
that
that
are
very
bureaucratic
and
heavy,
are
you
know
the
opposite
of
that
is
one
of
the
capabilities.
Is
that
you
have
more
lightweight
process
and
you
have
the
process
where
you
need
it,
and
you
don't
have
the
process
where
you
don't.
So
we
can
still
point
to
some
of
those
things
and
say:
maybe
that's
where
you
need
to
start
some
of
your
journey
right.
C
Okay,
let's
posit
this
nichola
with
the
framework,
I
do
think
it's
useful
to
keep
to
have
that
or
to
use
that
as
a
place
to
iterate
on
the
concrete,
because
I
do
feel
like
at
the
end
of
the
day.
We
do
need
to
turn
this
into
something
concrete
and
get
the
feedback
and
make
it
more
useful.
So
do
you
have
a
path
to
even
sharing
that
or
like
a
place?
B
If,
if
we're
not
in
a
huge
rush,
what
I
would
like
to
do
is
take
the
weekend
when
the
u.s
is
on
their
thanksgiving,
because
that's
usually
a
nice
quiet
week,
and
I
want
to
use
like
one
or
two
days
that
week
to
kind
of
clean
this
up
a
little
so
that
we
we
don't
have
as
much
of
this
random
goalie
logs,
play
placeholder
text
and
then
put
that
in
github.
I
also
haven't
deployed
it
externally.
B
There
there's
a
there's,
a
product
called
netify
that
a
lot
of
products
used
to
just
kind
of
deploy.
So
I
want
to.
I
want
to
work
on
setting
some
of
that
up,
so
I
want
to
take
a
few
days
that
week
maybe
to
to
set
something
up.
That's
a
little
cleaner
than
what
I've
shared
here
and
then
yeah
and
then
maybe
put
that
into
github,
and
we
can
kind
of
iterate
from
there.
That's
okay.
I
thought
I
guess
the
other
thought
is.
B
You
know
if
we're
looking
at,
maybe
accelerating
things
I
don't
know
if
any
of
the
other
any
of
the
any
any
of
the
other
members
have
tech,
writers
or
anybody
else.
Who
was
interested
in
contributing
or,
if
like
whether
that
would
help
speed
things
up
or
not
like
that?.
C
E
C
D
What
I
was
hoping
is,
I
think,
is
once
we
get
the
repo
up
and
we
get
it
set
up.
Okay,
here's
the
thing
and-
and
that
will
make
it
a
little
bit
more
easier
to
sell,
hopefully
to
get
contributors
yep
that
that's
fair.
C
B
Yeah
well,
I
my
goal
is
to
like
have
that
in
a
place
where
I
can
then
check
it
in
and
then
once
we
hit
the
end
of
november,
then
yeah
to
hopefully
have
that
all
available
in
the
github
repo
that
you
set
up
yeah.
C
C
A
Might
be
worth
us
doing,
one
of
those
sort
of
user
story
exercises
where
we
just
put
on
the
heart
of
each
of
the
key
people
involved
and
then
capture
a
list
of
of
you
know
what
are
their
concerns?
What
are
they
interested
in
and
then
just
keep
keep
that
as
a
reference
document
to
say.
Look
these
are
the
questions
we're
actually
trying
to
answer
with
with
this
piece
of
work
and
then
come
back
to
it
and
say
right.
Have
we
answered
all
those
questions
to
for
each
of
those?
You
know
audience
members.
C
D
Awesome
all
right-
and
I
think
nicola
jan
is
gonna-
be
back
from
germany
soon.
So
hopefully
we
can.