►
From YouTube: CDF SIG Best Practices - Jan 25, 2023
Description
For more Continuous Delivery Foundation content, check out our blog: https://cd.foundation/blog/
C
Hello,
everyone
Let's
just
give
it
another
couple
of
minutes
for
people
to
join.
C
So
welcome
everybody
to
the
best
practices
special
interest
group.
D
D
The
special
interest
group
with
Tara
has
just
joined
us
and
our
Focus
really
is
on
communicating
all
things:
continuous
delivery,
helping
to
capture
an
overview
of
or
significant
information
in
the
space
to
try
and
provide
a
repository
of
information
where
people
can
go
to
to
learn
more
and
to
understand
what
things
are
important.
What
things
you
might
need
to
watch
out
for-
and
you
know
which
things
you
you
should
be
very
careful
to
avoid.
D
D
So
Tara,
would
you
like
to
introduce
yourself.
E
Sure,
good
morning
my
name
is
Tara
Hernandez
I'm,
a
VP
of
engineering
at
mongodb,
specializing
in
developer
productivity
as
Terry
indicated
partner
in
crime.
When
we
nerd
out
about
the
academic
elements
of
continuous
delivery.
D
Yeah,
sorry
fatty,
would
you
like
to
just
do
a
quick,
intro
and
then
ran
to
to
the
next
person
in
line
okay.
F
Sorry
yeah
welcome
everyone,
I
work
at
the
dunes
Foundation,
leading
the
contest
foundation
and
very
interesting
upcoming
conversation
about
collaboration
potential
collaboration
between
the
CDF
and
auto
Mainframe.
Thank
you.
C
G
In
Japan,
I'll
go
next
yeah;
okay,
all
right
hello,
everybody,
I'm,
Michael,
Bauer,
I'm,
product
manager,
working
at
broadcom,
particularly
I,
work
on
our
Endeavor
product
Endeavor
is
the
leading
software
change
management
system
for
Mainframe,
but
I
also
have
a
lot
of
experience
in
the
open
Mainframe
project
working
on
projects
such
as
Zoe
I
was
a
squad
lead
for
the
Zoe
command
line
interface,
which
allows
you
to
interact
with
the
platform
from
off-platform
Technologies
for
a
number
of
years
and
also
involved
in
the
Cobalt
programming
course.
A
Yeah
I
was
going
to
say:
I
was
going
to
wait
for
my
name
to
be
called,
but
I'll
jump
into
thanks.
Mike
Peter,
Russell,
I
work
for
broadcom
as
well.
Mike
is
on
my
team.
A
My
role
within
broadcom
is
leading
the
work
that
we
do
in
the
product
space
to
define
a
set
of
solutions
that
we
call
Mainframe
devops,
which
consists
of
solutions
that
could
involve
continuous
delivery,
the
adaptation
of
tools
in
the
open
space
to
the
Mainframe
and
so
forth.
I'm
also,
a
member
of
the
open
Mainframe
governing
board
working
closely
with
John
Martin
I.
Get
like
do
I
get
to
call
the
next
one
I'm
going
with
the
Yvonne
fire.
How
about.
H
Hey
eventually
I
work
for
I'm,
a
devops
engineering
for
one
child
and
Luis
and
I
we're
going
to
talk
next
well,
the
devops
engineers
that
worked
on
for
lunches
Z.
So
we
call
this
one
shot
on
Z,
so
creating
making
one
shot
to
work
on
Mainframe
and
we
got
an
email
about
this
meeting.
So
we
came
to
see
how
we
can
contribute
or
collaborate
Luis.
B
Hi
and
Luis
Mr
wire
senior
developers
at
Ivan
said
we
we
work
to
immigrate
brancher
another
approach
from
suse
to
s390x
so
happy
to
be
here.
E
E
Okay
party,
it
looks
like
you
dropped
a
couple
of
suggestions
into
the
agenda.
F
Yeah,
so
maybe
I
can
I,
don't
know
it's
like
I
put
them
John
and
I
had
a
few
conversations
past
couple
of
weeks
to
see
like
what
are
the
opportunities
between
contessary
Foundation,
open
Mainframe
commit
to.
You
know
have
like
explore
based
on
those
discussions,
I
just
added
a
few
topics
there.
It's
like
okay,
what
is
the
contingency
best
practices?
Our
community
has
been
developing
about,
maybe
territory.
F
You
can
memorize
that
and
maybe
highlight
two
things
about
that
work
which
is
like
collaboratively,
developed
stores,
no
source,
and
then
maybe
we
can
learn
what
open
firm,
Auto
Mainframe
project
is
about
and
then
have
a
conversation
about
the
Epica
weight
of
the
content,
Theory
best
practice
of
the
mainframes
and
any
other
relevant
topic,
Red
Cross
Country
survey
for
mainframes
and
content
to
talk
about
like
what
kind
of
opportunities
we
have
to
collaborate.
F
D
Let
me
explain
a
little
bit
about
what
we've
been
doing
and
and
and
and
then
how
things
might
fit
together.
D
So
we
we've
noticed
that
there's
a
there's,
a
fairly
big
challenge
with
continuous
delivery,
because
it's
it's
such
a
broad
scope
of
of
topic.
D
So
many
people
come
to
continuous
delivery
from
an
engineering
perspective
within
an
organization.
That's
already
adopted
that
as
a
methodology
and
so
a
lot
of
the
the
visible
focus
and
the
discussions
that
you
see
in
in
general
are
about
the
you
know:
the
engineering,
discipline
and
cicd
and
best
practices
in
build
pipelines.
Things
like
that,
all
of
which
is
extremely
important,
but
also
any
part
of
the
overall
picture
because
for
continuous
delivery
to
work
for
an
organization
it.
D
It
has
to
be
fundamentally
a
business
methodology
because
you,
you
need
to
have
all
of
your
teams
structured
in
such
a
way
that
you
can
support
that
style
of
delivery,
and-
and
this
is
a
major
sticking
point
for
a
lot
of
organizations
that
they
they
try
to
implement
purely
with
inside
an
engineering
department
and
then
get
a
a
really
effective
system
that
can
deliver
product.
D
D
Resources
are
to
provide
the
the
overarching
picture
of
how
all
of
these
problems
break
down
into
different
areas,
and
so
you'll
see
that
on
the
website.
There's
a
lot
of
discussion
of
commercial
practice
and
how
to
how
to
structure
your
business
in
order
to
do
continuous
delivery
and-
and
then
we
drill
down
into
each
of
the
engineering
sections
and
and
talk
about
individual
aspects
of
testing
and
employment
and
how
those
things
add
up
to
give
us
a
an
overarching
practice.
D
And
then
what
we
also
provide
within
that
set
of
resources
is
a
a
an
area
for
individual
teams
and
projects
to
to
publish
white
papers
on
their
specialist
technology
or
their
area
of
Interest.
So
the
main
repository
we
try
and
keep
as
technology
agnostic
as
possible
and
and
then
we
we
provide
an
embedded
area
that
allows
for
much
deeper
Dives
on.
B
D
So
I
would
be
very
interested
to
understand
what
some
of
the
challenges
are
in
in
from
a
Mainframe
perspective
and
then
how
we,
we
might
be
able
to
facilitate.
D
D
Well,
I
think
at
this
stage
it
would
be
good
for
us
to
just
have
a
broad
discussion
about.
You
know
where,
where
we
think
things
are
at
the
moment,
you
know
what
level
of
maturity
is
there
in
in
the
Mainframe
space.
What
what
challenges
are
there?
What
what
can
we
do?
That
would
help
to
to
improve
things.
A
Okay,
I
guess
I'll
start
in
Michael
or
John
murtick
or
whomever
can
chime
in
as
we
go,
the
the
big
market
share
or
solution.
That's
in
the
Mainframe
market
today
for
solving
the
challenges
around
what
would
be
called
continuous
integration
or
continuous
delivery
on
the
Mainframe,
as
distinct
from
applications
that
include
the
Mainframe
is
a
product
called
endeavor.
A
Now
we
in
broadcomer
buy
is
towards
Endeavor
because
it's
our
product,
but
in
fact
it
does
have
a
largest
market
share
right
and
a
number
of
Mainframe
clients
over
the
last
several
decades
have
come
to
rely
upon.
It
have
come
to
invest
in
it,
for
you
know,
creating
pack
creating
bills
packages
and
deployment
for
software
artifacts
on
the
Mainframe.
C
A
One
of
the
big
challenges
that
we
see
with
our
Mainframe
clients
is
well
a
couple
things
as
Mainframe
or
as
the
Enterprise
looks
to
Endeavor
continuous
delivery
outside
of
the
Mainframe
there's
sort
of
a
organizational
difference.
A
So
that's
a
challenge
within
broadcom
for
sure
and
I
know:
John
within
open
Mainframe
project
for
sure
we're
promoting
some
thinking
which
says
the
Mainframe
should
be
viewed
as
just
another
platform
right
and
it
in
so
far
is
solving
the
end-to-end
challenges.
You
can
include
the
Mainframe
in
your
continuous
delivery.
Your
Agile
development
practices,
your
overall
sdlc,
because
what
we
have
done
in
open
Mainframe
project
is
work
towards
making
the
experience
in
working
with
the
Mainframe,
like
the
experience
in
working
on
distributed
and
Cloud
environments
right.
A
So,
for
example,
when
it
comes
to
automating
processes
on
the
Mainframe,
perhaps
from
off
host
or
perhaps
integrating
tooling
that's
off
host,
we
have
something
called
the
Zoe
CLI
right,
which
interfaces
with
Mainframe
resources
like
could
go
into
detail.
He
led
the
squad,
but
you
know,
gives
a
CLI
based
user
experience
that
can
be
used
to
automate
tasks
and
access
to
resources
and
data
on
the
Mainframe
right
trying
to
think
where
to
go
next
with
this
Mike.
Do
you
want
to
chime.
G
You
know
the
AWS
CLI
to
automate
things
against
AWS
I
can
use
the
Zoe
CLI
to
automate
things
against
the
Mainframe.
Just
another
platform
I'm
interacting
with,
and
the
Zoe
CLI
also
has
it's
it's
integratable,
so
different
vendors
can
plug
in
their
products
into
it
and
offer
extensions
and
then
extend
access
from
all
these
distributed
tools
to
the
Mainframe,
for
example.
Now,
it's
just
like
also
like
promoting
a
lot
of
the
visibility
into
this,
because
I
think
just
generally
in
the
distributed.
G
World
folks
aren't
necessarily
aware
or
have
the
conversations
around,
including
Mainframe
in
these
processes,
but
yeah,
using
Jenkins
to
automate
processes
on
Mainframe
definitely
possible
using
popular
Ides
for
development
like
Visual
Studio
codes
possible,
and
that's
really
what
I
view
the
open
Mainframe
project
is
about
is
extending
the
ability
to
work
with
Mainframe
from
the
same
tools
and
Technologies
you
would
use
for
distributed
or
Cloud
development.
D
So
one
thing
that's
probably
worth
mentioning
here
is
we
have
a
sibling
project
called
CD
events,
and
that
project
is
focused
on
trying
to
come
up
with
a
a
technology,
agnostic
vocabulary
for
continuous
delivery
events,
so
that
we
can
solve
exactly
these
sorts
of
problems
where
you,
you
need
to
have
a
mixed
technology,
stack
that
you
would
like
to
have
some
sort
of
overarching,
continuous
delivery,
automation
process.
D
So
CD
events
is
hoping
to
provide
a
a
way
in
which
you
can
integrate
different
tools
with
a
Common
Language,
so
that
you
can.
You
can
have
a
shared
perspective
about
what's
happening
in
different
types
of
your
parts
of
your
pipelines,
and
you
know
what
events
are
going
on.
That
may
influence
the
behavior
of
segments
of
your
Pipelines.
A
I
think
that
makes
sense.
You
know
what
we
have
found.
I'm
sure
you
all
share.
This
view
is
that
devops
environments,
continuous
delivery
pipelines
are
multi-tool
multi-vendor
by
definition
and
the
part,
that's
missing
is
the
part
you're
speaking
about,
which
is
sure,
but
how
do
we
all
speak
about
them
so
that
we're
sort
of
on
the
same
page?
Right
and
yes,
you
know
we
would
want
on
behalf
of
open
Mainframe
project.
We
would
want
the
Mainframe
to
be
part
of
that
because,
like
we
said
it
should
be
no
different.
E
So
this
is
fascinating
because
one,
you
know
the
first
thing
that
you
said
Peter
was
around
sort
of
organizational
adoption
or
adoptability.
If
you
will
I
mean
so
in
some
ways,
it
really
doesn't
matter
the
runtime
environment,
we're
always
going
to
have
this
tension
about.
Why
are
we
investing
all
of
this
effort
in
order
to
do
this
thing
right?
What's
the
business
proposition,
which
is
always
Terry's,
you
know
big
fight
that
he
takes
like
no,
no
there's
so
much
business
value
here.
E
The
other
thing
is
that
within
mongodb,
so
I
joined
a
little
less
than
a
year
ago,
and-
and
this
is
the
first
time
since
I
was
in
college
back
in
the
90s
that
that
I've
been
involved
with
mainframes
again,
because
you
can
run
mongodb
on
Z
series
and
powerpc
mainframes,
and
so
you
know
I'm
looking
as
the
VP
that
oversees
the
dev
infrastructure
like
how
do
we
integrate
this
effectively
and
there's
a
similar
challenge?
E
I
think
when
you
think
about
like
Max
Stadium
right,
I
mean
AWS
most
recently
is
starting
to
offer
Mac
hosting
within
their
environment,
but
for
a
long
time
that
wasn't
the
case,
and
it's
still,
you
know
it's
still
kind
of
on
the
fence,
whether
or
not
that's
a
good
value,
add,
and
but
it's
that
concept
that
we
have
at
mongodb
around
Dynamic
versus
static,
hosting
I
think
that
the
industry
has
forgotten
that
there
is
more
than
just
you
know,
Linux
x86
or
maybe
Linux
and
windows
x86,
as
as
a
default
platform
when
to
I
forget
who
said
it
there's
a
huge
market
for
Mainframe
consumption,
on-prem,
not
cloud-hosted,
and
it's
not
likely
to
change
given
the
customers
that
are
likely
in
that
space.
E
You
know
government
agencies
financials,
where
there
are
frankly
laws
on
where
you
can
store
data
and
where
you
can
run
compute,
so
I
do
think.
There's
a
pretty
critical
opportunity
from
a
from
a
like
a
business
vertical
perspective
to
kind
of
drill
into
this,
with
even
more
concrete
examples
in
this
space.
A
Yeah,
like
I'd
like
to
yes
and
what
you
said,
our
our
view,
at
least
in
broadcom,
is
that
you
know
the
Mainframe
should
be
it's
one
of
these
right
workload
for
the
platform
sort
of
thoughts.
We
we
don't
believe
all
workloads
should
be
a
Mainframe.
There
are
certain
people
who
believe
you
know
all
the
workload
should
be
on
the
Mainframe.
You
should
do
all
your
new
development
on
the
Mainframe.
Our
point
of
view
is
like
look.
A
There
are
certain
workloads
that
are
optimized
or
that
the
Mainframe
optimizes
scales
secures
Etc,
but
there's
a
certain
amount
of
work
that
rightfully
should
be
done,
hosted
and
done
off
the
Mainframe,
and
that
really
sets
that
hybrid
function
really
sets
up
for
what
we
talked
about
earlier,
which
is
well.
If
that's
the
case,
you
got
to
get
these
two
worlds
to
work
together,
right
and.
E
B
A
Of
the
value
prop
or
the
problem
statement,
I
should
say
in
where
you
were
going
I
think
so.
E
E
Just
to
see
who's,
so
one
of
the
things
that
I
think
has
been
very
prevalent
that
I
that
I
think
about
in
the
continuous
delivery
conversation
is
how
tightly
coupled,
not
only
to
you
know
Linux
x86,
but
also
containers
right.
E
So,
even
if
you
look
at
the
CDF
offerings
right
that
you
know,
Jenkins
Jenkins
is
the
old
workhouse.
That's
been
around
for
a
million
years.
You
can
kind
of
do
whatever
for
good
or
ill
right,
but
then
tecton,
Jenkins,
X
Spinnaker
to
a
certain
degree.
You
know:
there's
it's
a
really
hardcore
Focus
on
containerization
and
container
orchestration,
and
a
lot
of
the
industry
is
really
pushing
hard
in
that
direction.
E
Now
I
happen
to
know
there
are
people
Elizabeth,
oh
God,
what
is
her
last
name
anyway,
IBM
Mainframe
nerd.
He
was
talking
about
running
kubernetes
orchestration
on
mainframes,
but
it's
sort
of
a
it's
more
of
an
it's
more
of
like
an
interesting
sideline,
rather
than
a
heart
from
what
I
can
tell
a
hardcore
business
investment
now
I'm
curious.
What
your
take
is
on
that
and
where
nope
it
starts
with
a
J
I
can
I
can
look
her
up
later.
E
Thank
you
has
cute
babies.
She
keeps
posting
all
over
social
media
I'm.
A
E
You
know
you
don't
have
to
give
details
because
we
don't
want
to
violate
ndas,
but
where
do
you
see
this
going
like?
Do
you
see
that
actually
building
momentum,
or
do
you
see
that
that
is
kind
of
distracting
from
maybe
more
critical
issues
from
the
perspective
of
of
continuous
delivery?
Because
a
lot
you
know
a
lot
of
people
will
start
conflating
them
and
they're
not
the
same
thing
and
by
conflating
I
mean
continuous
delivery
sort
of
future
looking
and
containerization
as
the
one
true
way.
A
Yeah
see
that's
what
I
was
going
to
react
to
First.
Is
this
idea
that
you
know
my
point
of
view?
Having
done
multiple
variations
or
stepping
up
to
multiple
variations
of
this
problem
in
different
roles?
Is
that
implementing
a
continuous
delivery
pipeline
does
not
mean
you
have
to
adopt
containers
right?
It
is
my
first
point
of
view
right
and
and
in
fact,
when
you
look
at
the
kinds
of
things
that
some
of
our
more
advanced
Mainframe
clients
are
doing,
to
bring
together
the
Enterprise
and
the
Mainframe
worlds
and
develop.
A
You
know
a
true
developer
platform
that
self-serve
and
allows
their
developers
to
spin
up
instances
or
gain
access
to
Mainframe
resources.
Whatever
their
job
is
right,
that's
all
meant
to
try
to
normalize
using
the
constructs
that
are
available
on
the
platform
right,
so
that
doesn't
mean
like
you
got
to
do,
containers
on
the
Mainframe
to
make
that
developer
platform
work
and,
in
fact,
the
clients
that
I'm
referring
to
they
haven't
implemented
it.
A
A
So
so
that's
my
first
thought
second
would
be
that
IBM
is
certainly
heading
in
the
direction
of
placing
a
big
big
bet
on
kubernetes.
A
Right
well,
it
also
led
to
the
acquisition
of
red
hat,
which
was
further
instantiation,
because
you
know
the
red
hat
everywhere
strategy
says
you
know,
you
need
a
kubernetes
abstraction
layer.
We
got
it,
it's
called
redhead
openshift,
it
happens
to
run
everywhere.
You
would
want
it
to
run,
including
on-prem
go
right.
A
So
you
saw
that
initially,
with
making
a
hosted,
Red
Hat,
open,
openshift,
available
on
systems
e
Linux
right
and
what
you're
seeing
lately
is-
and
this
may
be,
where
Elizabeth
is
coming
from
using
parts
of
the
kubernetes
technology
stack
to
make
it
possible
to
orchestrate
containers
on
Z
OS
right.
That
starts
to
get
into
some
really
gnarly
issues,
though
right,
because
the
Mainframe
at
that
layer
has
been
optimized
like
crazy
over
many
decades
to
run
workload
very
efficiently
in
an
isolated
way.
So
you
sort
of
scratch
your
head
and
say:
well.
A
Why
are
they
messing
with
that
right
or
are
they
or
is
the
idea
that
they
simply
want
to
make
it
possible
to
run
containers
in
an
orchestrated
way
on
Z
OS,
which
I'm
not
sure
who's
doing
that
right
now,
or
even
would
want
to
do
that
just
because?
Just
because
the
workloads
that
are
already
running
in
that
environment
are
highly
optimized,
don't
touch
them
right,
so
that's
I'm,
just
riffing
a
little
bit,
but
did
that
help
Tara.
E
Yeah,
it's
just
kind
of
interesting
because
if
you
so
one
of
so
one
of
the
challenges
and
actually
one
of
the
areas
where
we
have
gaps
and
Terry,
please
feel
free
to
jump
in
anytime.
Here
is
you
know
around
best
practices
around
security
and
compliance
as
it
pertains
to
life
cycle
right
and
continuous
delivery.
So
when
you
have
a
whole
industry,
you
look
at
Sig,
store
and,
and
the
companies
around
that
and
then
there's
a
new
one
coming
out
now.
E
I
think
Microsoft
has
a
thing
that
they're
trying
to
do
and
it
and
again
all
of
that
is
based
on
containers
and
all
the
interpretation
around
the
executive
order
and
s-bombs
in
provenance.
It's
all
around
the
container
right
and
the
tooling,
but
even
outside,
of
mainframes
container
adoption
versus
VMS,
for
example,
is
still
not
in
the
majority
right
and
then,
if
you
had
customer
use
of
mainframes
on
top
of
that,
I
think
there's
an
even
bigger,
Gap
and
well.
A
C
G
E
G
But
the
some
of
the
integration
points
we're
looking
at
are
things
tools
like
like
the
open
tools
like
dependency
track,
where,
if
they
get
an
s-bomb
from
a
Mainframe
application
or
an
s-bomb
from
a
distributed
application
or
whatever,
they
can
load
that
into
that
tool
and
then
be
notified
of
where
the
vulnerabilities
on
their
system
exist
and
really
just
integrating.
You
know
into
all
those
tools
that
can
read
and
analyze
s-bombs
to
give
customers
better
views
of
their.
D
D
This
is
an
asset
management
problem
that
you
know:
you've
you've
got
customers
out
there
who
are
building
products
and
those
products
depend
on
multiple
Technologies
and
what
they
really
need
to
be
able
to
do
is
maintain
their
product
as
a
as
a
as
a
contiguous
asset
where
they
know
what
state
it's
in
at
any
time,
regardless
of
what
parts
of
Technology
are
involved
in
building
that,
and
so
from
that
perspective,
the
the
Mainframe
problem
is
is
not
a
unique
one.
D
It's
it's
what
a
class
of
examples
like
Ai
and
machine
learning,
high
performance
computing,
where
you've
got
specialist
architectures
for
doing
a
certain
thing
more
effectively,
and
those
specialist
architectures
get
treated
differently,
and
so
what
we
want
to
really
focus
on
is
having
the
ability
to
span
The,
Continuous
delivery
ethos
across
all
of
those
technology
spaces,
regardless
of
what
actual
technology
is
going
on,
not
force
them
to
all
align
to
One
technical
implementation,
but
instead
provide
the
appropriate
abstractions
so
so
that
they
can
be
integrated
into
this
overarching
asset
management
structure
that
lets
you
deliver
with
confidence.
D
D
It's
not,
but
it's
certainly
something
that
you
see
a
lot
in
in
people's
General
processes.
You
know
one
of
the
things
that's
that's
often
challenging
is
that
for
for
people
who
are
working
within
the
industry
on
large
industry
products,
yeah
they're
already
a
long
way
the
way
they
are
with
best
practice
in
in
product
development,
and
they
live
in
this
ecosystem
that
is
surrounded
by
all
of
these
good
techniques
and
good
things
to
do.
But
then,
when
you
move
out
to
the
customer
end
of
things,
you've
got
people
building.
D
You
know
equally
large
real
world
applications
that
are
reliant
on
products
from
multiple
providers,
but
they're
not
exposed
in
the
same
way
to
you
know
to
this
in-depth
culture
of
continuous
delivery
and
devops,
and
things
like
that,
and-
and
so
you
know,
it
is
not
unusual
at
all
to
find
Mission
critical
applications
still
running
on
an
Excel
spreadsheet
on
a
server
on
somebody's
desk.
D
D
Big
step
to
take
people
from
there
to
you
know
iterating,
for
orders
of
magnitude
faster
than
they
are
currently.
E
D
A
A
No
I
was
just
going
to
ask
if
we
could
double
click
on
that
for
a
second,
because
I
suppose
I've
had
the
luxury
of
leading
engineering
teams
throughout
my
career,
who
were
of
the
mind
you
know,
I
want
to
write
code,
I
want
to
automate
I
want
to
you,
know,
I,
don't
want
to
do
anything
manual
and,
and
they
gravitate
they
gravitate
away
from
the
situation
you
were
describing.
D
I
think
there
there's
a
cultural
challenge,
which
is
that
the
these
these
techniques
propagate
through
exposure
to
them.
So
you
know
if
you're
working
in
the
Bay
Area
you're
gonna
get
exposed
to
all
of
those
ideas,
all
of
the
time
from
multiple
directions.
Right
and
also
people
are
moving
around
between
companies
and
taking
those
cultural
ideas
with
them.
So
you
get
this
wonderful
ecosystem,
that's
propagating
best
known
methods
within
the
ecosystem,
but
once
you
move
out
of
hot
spots
like
that,
you've
you've
got
still
large
companies,
but
they
tend
to
be
geographically
isolated.
D
There's
not
so
much
freedom
of
movement
and
and
so
people
will
grow
up
in
an
industry
and
they'll.
You
know
they
may
only
work
in
banking
or
they
may
only
work
in
insurance.
C
D
They
have
a
very
in-depth
knowledge
of
the
history
of
how
to
do
things.
Going
back
to
the
60s
and
70s
in
in
in
that
area,
got
very
limited
exposure
to
Alternative
models
of
the
world,
I
mean,
which
is
a
big
part
of
why
why
we
exist,
and
what
we're
trying
to
do
with
with
best
practices,
is
to
disseminate
that
information
so
that
more
people
get
exposed
to
it
and
and
can
potentially
benefit
right.
E
So,
even
in
the
Bay
area,
though,
because
I
like
to
have
my
counter
example-
I
mean
I
work
at
a
company
where
I
I
would
get
into
arguments
with
another
VP
of
engineering.
Who
would
say
things
to
me
like?
Why
should
my
developers
waste
time
writing
tests,
they
haven't
finished
implementing
yet
and
by
the
way?
That's
how
we
did
it
at
my
last
company
right
or
what
was
the
other
good
one,
it's
too
difficult
to
try
to
ship
stable
artifacts
more
than
once
a
quarter.
E
I
I
love
that
company
soon
after
and
then
I
I
heard
he
left
and
went
to
Amazon,
so
God
help
them,
but
I
I
also
think
it's
the
type
of
business
that
you're
in
you
know,
I
think
if
you're
I
think
b2c
folks,
perhaps
to
make
some
broad
generalizations
might
not
have
the
same
strength
of
culture
because
they
don't
have
the
compliance
or
slas
or
other
types
of
things
gaming.
This
is
gaming
industry.
E
In
this
particular
case,
you
know
it's
I
do
think
there
are
kind
of
these
cultural
Pockets
where
we
have
wildly
Divergent
cultures.
So
how
do
we?
How
do
we
smooth
that
out
a
little
bit
better,
particularly
as
the
the
industries
themselves
start
to
merge
right?
The
gaming
industry
now
has
Financial
elements
right.
E
You
know,
Unity
has
their
whole
platform,
for
example
around
monetization,
which
is
introducing
things
that
previously
were
not
necessarily
a
part
of
the
gaming
game
development
ecosystem
in
the
same
way,
I'm
I'm,
curious,
I,
guess
this
is
for
Peter
and
and
Michael
in
particular,
where
you're
talking
about
you,
know
Endeavor
being
a
platform
for
mainframes,
but
there's
a
lot
of
element
of
sort
of
Bridging,
the
Gap
between
mainframes
and
broader.
E
So
what
are
kind
of
I'm
curious,
like
what
are
kind
of
the
key
bridging
needs
that
you're
seeing
in
that
process
is
it?
Is
it
incorporating
certain
types
of
third-party
Solutions
software
Solutions?
Is
it?
Is
it
middleware
type
of
things
like
kind
of
what
are,
what
are
the
the
main
bridging
challenges?
You're?
Seeing
because
I'm.
A
E
A
It's
basically
a
synchronization
solution
which
would
allow
somebody
to
prime
a
git
repo
with
a
code
that
was
previously
under
sem
control
and
Endeavor
and
then
give
those
who
prefer
the
get
interface
or
the
get
experience
to
you
know
operate
on
that
code
as
if
it
were
on
the
Mainframe
and
then
the
solution
takes
care
of
synchronizing
back
and
forth.
A
So
that's
one
example
where
you
know
what
we're
trying
to
do.
There
is
continue
to
make
Endeavor
relevant.
You
know
to
emerging
technology
Trends,
you
could
say
well
look
dude!
It's
been
out
there
forever,
but
you
know
emerging
technology
trends
that
Mainframe
teams
are
running
into
and
finding
that
you
know
younger
generations
of
folks
that
they
hire
prefer
right.
So
that's
one
example:
another
example,
although
it
may
not
align
completely
with
what
you
are
getting
at,
is
we've
spent
a
considerable
amount
of
investment
producing
a
set
of
Visual
Studio
code.
A
Extensions
that
leverage
things
like
you
know
the
LSP,
that's
already
in
this
code
to
understand
code
or
languages
like
Cobalt
that
run
on
the
Mainframe,
but
that
would
allow
you
know
application
developers
to
continue
to
use
vs
code.
Even
when
they
want
to
operate
on
or
you
know,
operate
on
Mainframe
code
application
code
that
runs
on
the
Mainframe
I
think
the
last
example
would
be.
A
A
For
example,
we
have
a
solution
which
allows
you
to
leverage
test
Frameworks
like
mocha
jest
but
operate
in
or
run
test
K
or
cause
the
running
of
test
cases,
and
then
the
validation
of
results
on
the
Mainframe
and
bring
that
all
back
to
the
test
framework
running
off
the
Mainframe
Mike.
Do
you
have
any
other
examples
or
thoughts
on
that.
G
I
would
just
add
you
know
just
really
any
off-platform
tool,
or
especially
in
cases
of
scripting,
there's
a
lot
of
power
out
there
in
terms
of
the
open
source,
libraries
that
can
be
leveraged
and
now
applied.
You
know
a
big
thing
around
testing
Frameworks.
G
Is
there
isn't
much
in
in
terms
of
native
testing
Frameworks
on
the
Mainframe,
like
you'd
have
to
build
out
your
own
custom
assertions
so
like
the
value
you
get
by
integrating
with
those
testing
Frameworks
and
having
assertion
libraries
I
mean
it's
just
it's
a
huge
amount
because
you
can
imagine
developing
distributed
applications
where
you
had
to
test
without
a
testing
framework
or
assertion
library,
right
it'd,
be
it's
a
lot
of
manual
effort,
so
just
wanted
to
highlight
that
the
other
piece
I
like
to
think
about
it
in
terms
of
is
like.
B
G
They're
looking
at
going
into
the
artifact
repository
housing,
which
obviously
is
not
a
full-blown
a
solution
as
something
like
artifactory,
but
they
have
GitHub
releases
right,
so
I'd
say
like
if
you
could
reverse
that
and
think
back
to
when
Endeavor
was
created.
It
was
first
created
to
solve
all
those
problems
and
it
does
every
every
step
of
that,
but
now
we're
looking
at
allowing
it
to
be
more
integratable
into
using
the
tools
of
choice.
G
So,
if
there's
a
particular
piece
you
want
to
plug
into
Endeavor
like
use
GitHub
for
the
search,
control
management,
or
maybe
you
want
to
use
Jenkins
for
your
continuous
integration
pipeline
or
what
have
you
you
can
plug
into
each
one
of
those
sections
of
endeavor
and
for
places
where
you
don't
want
to
take
risk
of
changing.
You
can
continue
using
what
you
have
been
using
for
the
past
so
many
years.
E
It
it
makes
me
think
about
a
system
back
in
the
battle
days
that
tried
to
be
kind
of
all
singing
all
dancing
at
the
time.
I
don't
like
you
may
be
too
young
for
this,
but.
B
E
Was
something
called
clear
case?
I,
don't
know
if
Terry
and
Mike.
C
E
A
E
All
right
now
that
I've,
given
us
all
the
old
folks,
PTSD
I
I,
think
you
know
we're
seeing
everything
Cycles
right.
We
just
keep
coming
back
to
the
same
challenges
they're
just
maybe
the
the
details
start
to
adjust.
So
a
lot
of
what
y'all
were
talking
about
was
from
the
developer
side.
Great,
that's
what
we're
kind
of
here
for
but
I'm
curious.
E
If
there
are
any
customer
facing
elements
that
pertain
to
this,
like
from
an
artifact
management
or
runtime
support
operations,
management
that
kind
of
come
into
play
here
and
I'm
not
trying
to
lead
this
in
any
particular
direction,
but
I'm
thinking
about
you
know
the
devops
Spectrum.
You
know
if
you're
really
dialed
in,
for
example,
using
a
cloud
hosting
solution.
You
know
you've
got
the
ability
to
to
have
a
pipeline
of
production
information
that
gets
eventually
sucked
back
into
the
dev
space
right
so
that
you
have
improved
time
to
resolve
metrics,
for
example.
E
Right
so
is
it
are
the
systems
that
we're
talking
about
here?
Is
there
really
that
kind
of
sort
of
functional
crossover
between
Mainframe
execution
and
non-main
frame
execution?
Or
is
there
a
pretty?
Does
there
tend
to
be
a
pretty
solid
layer
between
those
two
things
with
maybe
just
like,
like
you
know,
a
proxy
or
API
interface
or.
A
Know
really
not
sure,
where
you're
going
with
that
I
think
the
one
thing
that
did
come
to
mind
as
you
were
talking
about.
It
is,
for
example,
we're
doing
more
work
within
endeavor,
specifically
to
make
some
of
the
make
some
of
the
data
and-
and
you
know
the
camera
and
handling
of
packages
and
where
those
packages
exist
and
make
that
all
more
visible,
I'm.
E
A
A
G
Yeah
I
was
gonna
well.
G
I
was
just
gonna.
Add
that
you
know
when,
when
we
say
like
Endeavors
the
SCM
for
Mainframe
we're
talking
about
also
not
just
Source
control
management,
but
software
change
management.
So
it
includes
seeing
these
changes
as
they
go
between
environments.
A
lot
of
that
today.
Isn't
these
artifacts
or
changes
moving
between
long-standing
environments.
That's
typically
how
things
have
been
done
on
the
Mainframe,
but
starting
to
move
into
the
interest
of
you
know
like
like.
G
If
we
want
to
provision
a
new
environment
that
looks
similar
to
production
and
you
know
put
the
artifacts
into
that
environment
and
test
them
before
moving
forward.
Endeavor
also,
you
know
includes
strategies
for
for
those
things
as
well,
but
like
for,
like
a
I,
don't
know
if
that
answers
from
the
SRE
perspective,
but
also
when
you
were
talking,
I
was
thinking
from
the
management
perspective
and
we're
doing
some
work
around.
G
You
know
we
see
sets
of
changes
as
they
move
across
environments
from
the
Endeavor
perspective
and
looking
and
visualizing
those
in
terms
of
Dora,
metrics
and
things
of
that
nature.
So
we're
doing
some
work
around
that
as
well,
which.
E
Was
my
very
next
question,
which
is
you
know
going
back
to
where
Terry
is
probably
about
to
drag
us,
which
is
you
know,
exposing
the
business
value
through
metrics
driven,
you
know,
assessments
on.
Are
we
improving
our
velocity
quality?
You
know
that
type
of
stuff,
so
it
sounds
like
that's.
That's
an
element
of
it.
C
D
Know
the
we
often
see
people
focusing
just
on
the
engineering
piece
and
then
wondering
why
they're
struggling
with
actual
adoption
and
customers
and
and
it's
because,
if,
if
you're
only
at
the
engineering
level,
you
can
only
sell
that
to
people
who
already
have
bought
into
the
concepts
and
and
so
you're
back
to
this
cultural
problem
again
and
and
so
we
try
and
encourage
people
to
think
about
this
from
a
commercial
perspective.
You
know
what
what's
the
business
value
of
this.
How
do
you
quantify
that?
D
How
does
this
improve
your
ability
in
general
to
quantify
the
business
value
of
the
assets
that
you're
producing
and
be
able
to
make
that
argument
that
pitch
so
so
that
you
can
really
get
the
message
to
the
audience?
That's
paying
for
this
stuff,
because
I
mean
you
know,
part
the
reason
why
these
techniques
are
so
popular
in
cloud
and
kubernetes
space
is
because
any
developers
can
easily
spin
up
a
container
and
start
demonstrating
stuff.
D
D
You
know,
there's
lots
of
people
doing
it,
but
if
you
want
to
grow
The,
Continuous
delivery
space
now
you're
going
to
have
to
evangelize
to
people
who
come
from
a
completely
different
commercial
culture
and
take
them
on
the
journey
from
day
one
where
something
like
a
dependency
on
Mainframe
might
be
seen
as
a
blocker
to
that
adoption.
A
A
A
D
Yeah
and
I
think
again,
I'll
say
this
is
not
a
special
case.
What
right
what
you're
experiencing
is
is
not
only
the
norm
today,
but
actually
from
my
position.
I
I
also
sit
on
the
irds
roadmap,
so
so
I
I
get
to
be
involved
in
a
lot
of
the
architecture.
D
Discussions
about
what's
coming
at
a
semiconductor
level
over
the
next
10
years
and
and
believe
me,
we
we're
facing
a
Cambrian
explosion
of
new
architectures
that
are
about
to
hit
us,
so
so
we're
not
going
into
a
future
where
everything
is
kubernetes
running
in
the
cloud
in
a
homogenous
way,
we're
heading
into
a
future
where
there's
this
explosion
of
dedicated
architectures
with
specialist
silicon
to
solve
individual
problems
and
there'll
be
billions
of
these
devices
sitting
at
the
edge
and-
and
so
this
is
a
great
first
example
of
a
repeating
pattern.
That's
that's.
B
C
A
D
Snowflake
at
a
time,
yeah
definitely
be
great
to
get
some
content
about
this
Mainframe
perspective
in
into
the
repository.
So
you
know
we're
we're
starting
to
show
how
you
can
look
at
this
from
a
different
perspective,
but
make
it
all
all
work
in
in
an
overarching
way
right.
Yeah.
A
D
Yeah
I
I
I
think
this
has
been
a
very
productive
discussion.
I
think
we're
we're
definitely
talking
about
the
same
things
and
the
same
challenges,
but
it
would
be
really
nice
to
to
get
some
sort
of
white
paper
that
we
could
put
up
on
on
the
site
and
and
include
in
our
discussions
as
a
as
a
starting
point,
if
you
like.
So
you
know
that
would
not
need
to
be
technology
agnostic,
so
you
could.
You
could
discuss
the
approach.
D
Is
that
you've
been
taking
with
with
your
Technologies
in
your
problem
space
and
and
use
that
as
a
jumping
off
point,
but
then
yeah
I
think
we're
also
open
to
broadening
the
narrative
in
the
in
the
wider
document
to
to
be
more
inclusive
of
you
know.
How
do
you
address
diverse
environments,
and
you
know
why
these
are
actually
important,
that
the
diversity
of
the
environment
is
what
gives
you
your
Competitive
Edge.
G
H
I
just
wanted
to
point
out
just
a
little
bit
about
our
experience
well
trying
to
support
because
we
we
talked,
you
talked
about
Peter
talks
about
open
shift
before
so
one
choice
like
a
direct
competition
to
open
shift,
and
we
got
a
couple
of
customers
who
were
asking
for
one
child
to
be
on
availability
on
Rancho
support
on
Mainframe.
So
that's
why
we
started
the
project
to
part
one
show
on
Main
on
s390x
and
it
wasn't.
We
didn't
face
a
lot
of
challenges.
H
E
D
Still
drinking
my
coffee
we're
just
heading
to
the
top
of
the
hour,
so
fussy
did
you
want
to.
F
Yeah
it's
so
it
was
actually
what
you
just
said:
Terry
the
white
paper
idea.
Maybe
it
would
make
sense
to
have
this
work
on
a
white
paper
together,
like
CDF,
open
Mainframe
projects
to
you
know,
elaborate
the
idea
a
bit
further
evangelize
the
idea
pass.
The
word
you
know
call
people
for
Action
saying
this
discussion
is
happening.
It's
time
to
join
now
it
is
where
you
can
find
us,
so
we
can
both
sit
there
and
open
Mainframe
could
promote
this
further
to
bring
more
people
to
this
decision.
D
Yeah
and
I
would
encourage
everyone
to
to
have
a
read
of
the
best
practices
site
if
you
haven't
already
and
then
we
we
try
and
encourage
people
if
they're
going
to
add
content
to
to
use
the
the
definitions
of
some
of
the
terminology
that
we're
building
on
with
within
that
repository,
because
you
know
we
we've
spent
many
years
arguing
over
the
meaning
of
of
certain
terms
and
sometimes
that
that
can
lead
to
a
huge
amount
of
confusion.
D
So
so
it's
important
that
that
we
we
try
and
standardize
as
much
as
possible
in
that
space,
you're
more
than
welcome
to
start
drafting
some
content
and
submit
PRS
to
to
our
Repository
or
if
you
prefer,
we
can
have
a
you
know
a
more
formal
discussion
about
how
to
proceed
in.
You
know
how
to
write
this
stuff
up,
but
we're
we're
very
easy,
going
and
open
to
to
input
from
everybody.
So
you
know
feel
free
to
move
this
forwards.