►
From YouTube: Dec 16, 2022 - Ortelius Feature Review
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
C
So
let
me
start
by
saying
a
lot
of
this
is
already
addressed
in
artillius,
but
I
I
thought
of
a
few
things.
So,
first
of
all
it
just
sort
of
is
a
positioning.
You
know
what's
different
if
you're
building
an
application
out
of
microservices-
and
you
know
the
first
thing-
is-
keep
that
on
that
first
slide.
If
you
would
Steve
aren't
libraries,
okay,
a
library
is
typically
something
you
grab
from
some
place.
It
gets
put
in
your
code
and
it
doesn't
change
until
you
change
it.
C
You
know
you
don't
have
that
same
control
over
a
micro
service.
So
you
know
artillery
is
there
there
are
things
like
you
know.
The
notion
of
versioning
is
is
pretty
well
addressed,
but
it's
it's
an
important
Point.
Another
I
think
point
is
and
as
far
as
I
know,
this
is
true
of
artillius
too
a
catalog
doesn't
actually
hold
the
services.
It
points
to
the
services.
C
Right
and
those
Services
might
well
be
external
if
you're
doing
a
travel
and
expense
system.
You
might
point
to
something
like
American.
Express
is
used
by
many
many
corporations,
so
you're
going
to
invoke
some
services
with
them
and
you
may
not
invoke
others
so,
for
example,
in
my
career
I've
worked
for
companies
that
let
you
accrue
your
American
Express
points
and
I've
worked
for
companies.
They
didn't
so
you
have
different
ways
to
interact
with
the
services
based
on
that
I.
Don't
know
if
you'd
get
free,
expedited
shipping.
However
I.
C
So
so,
anyway,
the
point
being
that
those
Services
may
be
somewhere
else
so
essentially
go
to
the
catalog
to
say
a
few
things
and
let's
go
to
the
next
slide.
Steve,
maybe-
and
there
are
only
two
I
promise
guys
so,
if
I'm,
if
I'm
building
an
app
first
thing
I
want
to
do,
is
even
before
building
it
I'm,
probably
architecting
it
is,
you
know,
are
there
Services
out
there
that
I
can
use,
because
if
you
think
about
a
large
organization,
the
thousands
of
applications
is
not
crazy?
C
Well,
maybe
it
is
crazy,
but
it's
not
unheard
of
by
any
means
and
if
you
think
of
those
applications
being
composed
of
services,
one
of
the
things
you
run
into
is
okay.
I've
got
to
build
my
new
app
I'd
like
to
understand
what
services
are
available
for
me
to
use,
so
I
probably
want
to
look
at
various
attributes
and
in
artillius
that's,
you
know,
there's
a
whole
set
of
attributes
that
are
well
well
described,
but
I
presumably
would
like
to
understand
things
like
where
that
service
comes
from.
Has
it
been?
C
These
days
is
oh
there's,
three
services
that
all
do
the
same
thing,
which
one's
highly
rated
now,
if
I'm
a
manager
of
a
catalog
which
is
over
on
the
right
hand,
column
I
may
want
to
deprecate
the
lower
rated
services,
but
first
of
all,
I
have
to
be
able
to
see
ratings
and,
and
in
fact,
give
ratings
I'd
like
to
understand
where
the
service
came
from
I,
don't
know
if
you
guys
follow
a
lot
of
cyber
security,
but
there
was
a
report
in
the
last
couple
days
of
some.
C
In
this
case
it
was
libraries
are
commonly
used.
It
purported
to
be
maintained
in
America
and
actually
were
maintained
in
Russia.
C
You
know
we
don't
want
to
get
too
U.S
Centric,
but
you
know
the
notion
of
where
do
these
things
come
from
matters
I'd
like
to
understand
again
when
I'm
designing
in
a
service?
What
about
the
non-functionals?
You
know?
Is
there
some
performance
Target?
If
the
service
is
something
like
look
up
customer
for
a
say,
a
travel
loyalty
program,
you
know
how
long
do
I
think
it
ought
to
take
and
and
by
the
way,
is
there
any
measurement
of
it
actually
taking
that
long?
C
C
You
can
get
into
a
bunch
of
other
stuff
like
security
of
services,
but
then,
once
I've
done,
that
I
want
to
be
able
to
select
services
such
that
the
app
when
it
runs
can
actually
bind
to
those
services
and
presumably
also
release
bindings
and
then
another
one.
That's
quite
Central
to
artillius
is,
can
I
understand
the
this
s-bomb
for
the
service
and.
C
Does
that
particularly
well,
you
know
s-bomb
understanding,
is
it
changing
and
so
on,
but
my
these
again
were
thoughts
that
someone
that's
using.
This
would
probably
care
about,
and
then
someone
that's
creating
a
service.
Well,
you
want
to
be
able
to
publish
it.
You
want
to
be
able
to
update
it.
You
want
to
be
able
to
say
we're
taking
it
away,
you'd
like
to
know
what
people
are
saying
about
your
service.
Is
it
rated
well?
C
Is
it
rated
poorly
who's
using
this
service,
I
have
run
into
cases
where
things
like
services
are
being
consumed.
You
also
see
this
with
data
structures
by
the
way,
and
you
can't
even
tell
who's
using.
We
just
know
people
are
in
their
apps
I.
Also,
if
I'm
publishing
a
service,
I
probably
want
to
say
something
about
the
security
rules
around
it.
Can
anybody
choose
to
use
this
microservice?
C
Do
they
have
to
authenticate,
in
some
fashion,
a
world
of
zero
trust
that
becomes
more
important?
I
need
to
protect
my
service?
What
about
usage
rules?
That
goes
with
my
you
know.
Non-Functionals?
Is
this
service
available
12
hours
a
day,
six
days
a
week?
What,
if
you
try
and
use
it
outside
those
hours?
C
C
C
C
So,
for
example,
can
anybody
publish
into
the
catalog
or
do
you
issue
some
form
of
pull
request
that
the
catalog
operations
manager
has
to
say
yeah,
okay,
it'll
get
published
what,
if
someone
wants
to
update
a
service,
if
you
get
into,
for
example,
regulated
Industries,
there
may
be
an
expectation
that
the
results
of
an
application
will
not
change
due
to
somebody
pulling
an
underlying
service
out
putting
in
a
new
one.
You
know
lots
of
stuff
around
that.
C
Dave
is
a
little
bit
separate
Sasha
there
there's
no
question
the
data
sovereignty
matters,
okay,
so
data
sovereignty
means
for
those
that
haven't
really
dealt
with
it.
You
put
your
data
someplace,
say
in
within
the
European
Union
yeah,
and
expect
that
you
should
use
governed
by
EU
rules
and
it
may
well
require
data.
Doesn't
leave
the
EU
yeah
exactly
it's
running
in
a
data
center
in
Frankfurt
yeah.
You
can't
wake
up
in
the
morning
and
discover
it's
now
in
a
data
center
in
South
Africa,
where
Sasha
will
do
evil
things
with
it.
C
Yeah
the
services
again
I
may
say
this
is
a
service.
If
you
look
at
anybody,
that's
doing
these
for
government,
particularly
for
ministries
of
this
defense.
C
But
you
know
when
you
think
about
things
like
s-bombs,
it's
fairly
unlikely
that
Auditors
are
going
to
try
and
find
all
the
microservices
authors
they're
going
to
want
to
go
to
somebody.
That's
more
of
a
central
point.
C
Tell
me
all
the
places
log
for
Jay
shows
up
in
my
organization,
which
is
a
problem
organizations
have
even
months
after
that
issue
came
up
yeah,
it's
true,
and
then
you
know
again
log
of
Sir
our
services
being
used.
Are
they
not
being
used?
Services
can't
be
maintained
for
free
if
they're
not
being
used?
Maybe
you
want
to
say
well
these
four
Services
all
basically
do
the
same
thing
we're
going
to
turn
two
of
them
off.
C
You
know.
So
what
what
I
was
thinking
about
is
how
do
I
map
that
back
to
what
I
see
in
artillius
and
then
there's
another
level
of
question,
which
is
what
things
ought
to
be
in
the
open
source,
artifact
ortilius,
and
what
things
would
be
well
left
for
someone
wanting
to
offer
a
commercial
service
based
upon
ortilius,
which
obviously
could
be
deploy
Hub.
But
it
could
be
somebody
else
too
so
anyway,
that
that
was
pretty
much.
What
I'd
been
thinking
about,
I,
don't
know
how
much
that
makes
sense
to
people
that.
D
Makes
a
lot
of
sense?
How
would
you
what
would
be
your
comment
on
PCI
Ally,
PCI
compliance.
C
D
F
C
C
C
But
you
know
is
this:
is
this
in
fact
compliant
and
there's
probably
going
to
be
some
again,
you
come
to
the
the
notion
of
service
Providence,
someone's
asserting
compliance.
C
Can
I
tell
who
they
are?
Can
I,
you
know?
Is
there
some
Providence
if
this
thing
is
being
run
by,
say
MasterCard
I'm,
probably
a
little
more
willing
to
trust
it
than
somebody
I've
never
heard
of
but
PGA
compliance
has
very
much
a
specific
industry
spin
to
it.
So
I'd
want
to
be
able
to
understand
whether
microservices
were
claiming
conformance,
claiming
conformance
to
a
version
that
sort
of
thing.
B
Yeah
and
I
think
that
that
claim
to
conformance
would
be
one
of
those
attributes
that
you'd
want
to
tag
onto
the
onto
that
version
of
the
service.
So
people
when
they're
looking
for
things
they
can
search
based
on
certain
compliance
level
and
Etc
at
that
level.
C
Thinking
about
this
from
a
ux
point
of
view,
I'd
probably
want
there
to
be
some
way
to
browse
the
catalog,
so
that
I
can,
when
designing
an
application.
I
can
see
what's
available
for
me
to
consume
and
then
tell
my
my
build
team
use
these,
but
that
requires
some
user-friendly
search
capability.
I
would
think.
B
B
Seen
it
most
people
most
implementations
of
a
catalog,
you
know
if
you
look
at
a
Docker
Hub,
even
like
Pi,
Pi
npm,
those
do
a
tagging
type
of
methodology,
so
they
they
tag
things.
And
then
you
can
say
give
me
all
the
these
that
match
this
tag
and
I
think
that's
something
that
we
should
look
at.
C
And
you
may
I
I
think
we
probably
do
want
to
be
able
to
allow
users
of
utilities
to
extend
the
tagging.
B
Now,
one
of
the
things
just
to
to
kind
of
throw
out
there
for
an
architectural
point
of
view
is.
B
When
we
have
the
the
the
blockchain
Ledger,
we
will
have
it'll
be
it'll,
be
two-sided,
one
would
be.
The
thought
is
an
open
source
type
of
blockchain.
B
That
would
be
like
all
the
open
source
packages
and
services
out
there
and
then,
if
you're,
like
Sasha
you're
working
for
a
bank,
you
would
have
your
catalog
for
your
internal
proprietary
applications
in
the
blockchain
for
that
The
Ledger
and
then
you'd
Federate,
the
two
together
to
pull
information
about
the
open
source
packages
that
the
bank's
consuming
from
the
the
public
Federated
open
source
Ledger.
B
So
that
would
be
part
of
the
architectural
Viewpoint
is
from
inside
the
bank.
I
can
do
a
search
and
see
all
the
applications
that
are
consuming
log4j,
but
from
the
outside.
If
I'm,
the
log
for
J
team
I
should
not
be
able
to
see
which
banks
are
using
log4j
and
the
reason
being
is,
if
you
know,
if
you
can
go
and
search
and
find
all
the
companies
that
have
a
vulnerability,
then
you
know
who
who
to
go
attack,
and
it
just
simplifies
your
your
process
of
narrowing
down
exploitation.
B
Point
you
know
points,
so
we
will
have
that's.
Why
I
said
it's
like
it's
two
different
viewpoints,
whether
you're
in
inside
of
a
company
looking
out
to
the
open
source
side
and
then
from
the
open
source
side?
What
you
can
see
from
from
that
point
of
view
now
there
are,
there
are
ways
to
use
self-sovereign
identity
in
a
third-party
type
of
Ledger
transaction,
where
you
can
actually
obscure
the
consumer
side,
so
you
could
tell
that
from
an
open
source
package.
B
I
have
you
know
two
thousand
companies
that
have
the
that
need
to
be
that
have
this
vulnerability
but
I,
don't
know
the
name
of
the
companies
or
any
of
that
contact
information.
So
you
can
you
can
obscure
that
information
through
SSI,
and
that
may
be
something
that
we
would
want
to
consider,
but
that
would
be
down
the
road
when
we
get
into
that
type
of
scenario.
A
Yeah,
that
was
me
I.
Just
since
we're
on
this
discussion,
I
went
ahead
and
shared
uploaded
the
red
hat
requirements
document
for
an
attribute,
catalog
and
we're
meeting
with
him
at
the
beginning
of
the
year.
I
think
it's
January
4th
Maybe.
B
Cool
yeah
that'll
help
getting
more
of
these
requirements
together
and
I.
Think
the
did
a
do
you,
the
need
to
rate
a
service
and
add
additional
attributes
to
a
service
is
something
that
we'll
need
to
kinda
figure
out
how,
from
a
a
ux
perspective,
how
we
can
do
that
type
of
functionality
around
a
particular
service.
A
Yeah
I've
been
I
mean
that's
something
I've
been
hoping
that
we
would
do
rating
a
service,
even
if
it's
just
as
simple
as
giving
the
ability
to
add
a
star
or
determine
as
security.
It's
a
security
risk
based
on
the
domain.
It's
in
based
on
how
many
applications
it's
being
consumed
by.
Maybe
there
can
be
a
complexity
rating,
so
I
think
that
one's
an
interesting
one,
because.
C
Microservices
are
built
by
a
commercial
partner.
Exactly
you,
you
have
some
ability
to
go
back
to
and
know
who
you
can
talk
to,
and
so
on
and
things
that
are
published
and
become
considered
useful
by
an
open
source
partner
of
some
sort.
A
Yeah
I
think
I
think
that
we're
going
to
soon
see
open
source
libraries
be
microservices.
Mm-Hmm
I
mean
why
not
we're
not
compiling
code
anymore.
So
if
I
compile
all
of
our
open
source
libraries
into
a
single
container
and
have
every
single
container
have
to
repeat
that
right.
C
But
there'll
be
people
that
don't
want
to
do
that,
because
they've
got
some
proprietary
intellectual
property
in
it
or
like,
if
you
think
about
a
service
that
these
are
obviously
lots
of
them
out
there
already.
But
if
you
think
about
a
service
that
gives
you
an
optimal
route
for
a
truck
making
deliveries,
if
somebody
came
up
with
a
really
clever
algorithm
that
optimize
not
only
time
fuel
costs
based
on
picking
up
all
the
information
on
fueling
stations
and
their
current
prices,
they
may
not
want
to
make
that
open
source.
B
That's
and
that's
where
the
the
Federated
Viewpoint
allows
us
to
separate
the
proprietary
from
public
right.
C
B
Wouldn't
necessarily
necessarily
say
open
source
public
would
be
open
source
plus
any
public
services
that
companies
are
deciding
to
support.
Like
you
said,
American
Express
may
have
a
a
public
service,
for
you
know,
doing
credit
card
ratings,
or
something
like
that.
So
that
would
be
I.
Think
that's
a
good
Viewpoint
to
have
the.
B
Oh
on
the
for
tracking,
like
the
metrics
of
a
service
on
how
well
it's
performing
it,
that
could
be
difficult,
because
people
may
take
an
Open
Source
service
and
run
it
locally
in
their
cluster,
for
example,
and
that
would
give
it
you
know
they
just
may
not
give
the
machine
enough
horsepower.
So
they
have
a
that
service
is
performing
poorer
than
another
company
that
gives
the
that
service
a
more
horsepower
so.
D
B
One
metrics
could
be
an
interesting
one
to
to
tackle.
C
So
that
might
be
the
sort
of
the
non-functional
stuff
almost
would
have
to
be
within
the
context
of
the
organization
consuming
and
running
it
right.
So
the
consumer
has
an
opinion,
the
provider
where
it's
running
said
well,
you
know
I'm
constrained,
because
Amazon's
charging
me
like
crazy
and
I
can't
keep
16
core
systems
sitting
around
just
in
case
you
want
to
use
them.
B
B
Yeah
and
I
think
that
may
come
into
the
ability
to
look
at
well.
There's
two
aspects
to
that:
one
is
the
SLO
SLA
angle,
the
other
one
is
from
the.
B
What
what
do
you
want
to
call
it?
So
you
have
the
SLO
SLA
angle,
saying
that
I
I
want
to
have
this
service
meet
this.
Have
this
expectation
and
I
have
a
an
agreement
about
that
service?
B
The
other
way
is
to
look
at
how
that
service
is
going
to
not
only
perform,
but
also
the
attributes
around
that
service.
So
I
can
use
something
like
a
policy
agent
like
open
Opa,
so
I
can,
as
I'm
rolling
out
this
service.
I
can
look
at
the
security
threat
level
against
it.
I
can
do
like
you
said
the
rate,
the
service
is
it
as
a
does?
It
have
a
good
rating.
B
You
know
the
lag
time
of
the
current
version
of
the
service
to
the
the
one
I'm
using
those
types
of
attributes
you
can
funnel
through
Opa
to
help
you
make
decisions
about
how
to
flag
that
or
or
you
know,
either
flag
it
and
stop
it
as
part
of
your
pipeline
process,
or
you
know
flag
it
and
just
you
know,
review
it
later
type
of
thing,
but
I
think
when
we
look
at
the
data
that
we're
collecting,
we
should
always
think
about
in
the
context
of
a
policy
agent
as
well.
C
I
think
that's
absolutely
right.
Steve
and
you
know,
if
you
think
about
things
like
utilities,
tend
to
have
a
if
you're
going
to
build
something
and
deploy
it
into
your
network.
Here
are
the
versions
of
things
that
we've
blessed,
and
that
means
your
devops
or
devsecops
pipeline
has
to
understand.
C
B
I
think,
looking
at
the
list,
I
think
for
the
most
part
we
are
addressing
in
some
fashion.
Most
of
this,
the
part
that
we're
definitely
lacking
is
in
the
rating
system
and
the
additional
attributes,
like
you're,
saying
well,
the
rating
system,
additional
attributes
on
how
to
find
things
and
the
ability
to
browse-
and
you
know
the
browsing
and
finding
services
world.
C
B
That
it
included
in
the
open
source-
and-
and
this
is
where
I'm
going
to
push
over
to
the
the
ux
UI
professionals
and
what
is
the
best
way
to
find
something
in
a
large
amount
of
data.
You
know,
is
it
going
to
be
tagging?
B
Is
it
going
to
be,
you
know,
free
form,
search,
you
know
what
is
what
is
the
best
way
to
organize
that
data
for
finding
what
you
need
at
that
level
and,
like
I,
said
the
the
ones
that
I've
seen
you
know
if
you
look
at
Docker
Hub,
just
the
the
core
Docker
blessed
images
of
17,
000
versions
and
tags,
so
that's
a
and
if
you
look
at
Pi,
Pi
I
think
Pi
Pi
has
230
000
versions
of
modules
out
there
and
I'm
sure
mpm's,
even
larger,
probably
into
the
millions.
B
So
those
those
are
catalogs
I
think
we
should
kind
of
look
at
and
see
if
what
they've
done
is
something
that
we
should
adopt
at
that
level
or
if
we,
if
we
need
to
have
something
more
creative
for
us
put
it
that
way.
B
So
what
do
you
mean
by
policy
the
same.
E
E
So
do
we
have
rules
for
that
I
assure
we
can
use
a
policy
engine
like
iverno
who
can
help
achieve
this?
This
is
one
example,
but
you
can
stretch
this
example
to
some
other
other
categories
as
well
like
I
want
some
of
my
like
the
Registries
or
where
I'm
deploying
the
images
to
it's,
not
from
the
red
hat
or
QA.
E
B
We
need
to
gather
all
the
the
data
so
and
expose
that
data,
so
people
can
make
a
policy
decision,
because
those
policy
decisions
are
going
to
be
controlled
by
the
company.
That's
using
that
that
service,
for
example,
I,
may
decide
that,
for
example,
I
may
decide,
as
as
my
company's
policy
I'm
going
to
allow
any
images
out
there
that
have
a
critical
vulnerability
to
run
where
another
company
may
say.
B
I
want
to
have
no
voter
I
want
to
have
the
next
level
down
medium
level,
vulnerabilities
or
low
level
vulnerabilities
are
able
to
run
so
when,
as
long
as
we
provide
that
information
to
the
end
user
and
allow
them
to
query
that
information
easily,
they
can
put
that
into
their
their
policy
engine
to
decide.
What's
going
to
happen,
whether
they're
going
to
block
something
from
being
distributed
to
a
a
production
environment
or
not
that's
kind
of
the
Viewpoint
that
I've
been
thinking
about.
E
B
Yeah,
so
that's
on
your
on
the
the
screen
here
on
the
bottom
left
on
the
authoring
part
when
you
as
a
microservice
developer,
you
want
to
set
up
a
new
microserver,
so
one
of
the
things
just
to
everybody
give
it
back
a
five
second
background
on
Backstage
backstage.
Basically
what
it
does.
You
register
your
service
with
backstage
and
as
part
of
that,
you
click
some
basically
check
off
some
check
boxes.
Saying
I
want
this.
B
This
back
this
new
service
to
be
written
in
node.js
I
want
it
to
be
published
to
kubernetes.
The
pipeline.
I
want
to
use
is
going
to
be
Jenkins
as
the
pipeline
in
image
pipeline
tool
and
you
hit
go
and
when
you
hit
go
backstage
will
actually
generate
you.
B
A
sample
node.js
application
for
your
microservice
generate
the
Jenkins
file
for
you
connected
up
to
your
your
clusters
that
you
have
and
actually
deploy
that
all
the
way
out
to
production
in
a
single
go,
so
you
actually
have
a
test
micro
service
running
in
production
and
all
the
plumbing
for
that
the
devops
teams
normally
do
is
done
by
backstage
now
we're
or
to
these
hooks
into
that
is
in
two
spots.
One
spot
is
in
when
the
the
pipeline
is
generated.
B
We
want
to
hook
in
the
ortilia
CLI
into
that
pipeline
that
allows
us
to
grab
when
that
that
service
is
built.
We
grabbed
our
our
catalog
information
from
there
and
then.
The
second
part
where
we
hook
in
is
on
the
deployment
side
when
it
gets
deployed.
We
want
to
get
notified
that
the
deployment
happened,
so,
let's
say
they're
using
as
part
of
that
process,
they're
going
to
attack
on
Argo,
CD
and
Captain
as
their
deployment
tool
and
their
their
gatekeeper
of
of
the
cluster.
B
We
would
then
hook
into
the
captain
side
to
get
notifications
through
captain
that
a
deployment
happened
as
part
of
that.
So
that's
how
backstage
and
ortillas
kind
of
get
hooked
together
and
the
the
backstage
part
is
basically
done
once
when
the
developer
kind
of
creates
our
updates.
You
know
they
want
to
create
a
new
new
major
version
of
a
service
back.
That's
where
backstage
gets
involved.
F
F
F
So
backstage
is
giving
me
a
platform
rather
than
just
a
solution.
It's
just
that
I
can
customize
that
as
per
minute.
I
think
is.
Is
this
because
then
I
can
change
it
to
my
organization,
need
of
how
I
am
developing
Services,
what
I'm,
using
Jenkins
or
something
else
I
can
still
change
it
as
per
what
I
do
right.
B
B
Exactly
and
that's
where
the
backstage
templating
comes
into
play
and
one
of
the
things
that
when
we
do
our
backstage
I
can't
remember,
they
call
them
plugins
or
add-ins
yeah
that
there's
that's
where
they
would
choose
to
include
artillius
into
their
template
as
part
of
that
process.
F
So
do
we
are,
we
I
mean,
is
auto
list
now
listed
as
a
backspace
in
the
marketplace,
so
the
catalog,
which
they
have
or
not.
F
B
So
that's
one
of
the
things
that
Brad
McCoy
is
working
on
in
in
Australia
is
there's
a
front-end
plug-in
and
then
there's
a
back-end.
B
Back-End
plug-in
and
we'll
actually
have
both
at
that
level.
From
from
when
I
looked
at,
the
I
talked
to
I
can't
remember
her
name
from
Roadie
who's,
the
commercial
or
the
SAS
version
of
Backstage.
F
Yeah
no
I
used
Woody,
Martinez,
yeah
I
think
the
Roadie
is
is
a
better
version
because
it
gives
you,
you
are
your
setup
and
everything
is
done
and
I
think
they
have
much
better
support
for
backstage,
because
any
help
you
need
and
so
I
think
the
the
essence
is
that
what
Tony
was
earlier
talking
about.
You
know
that
for
microservices
we
need
observability
fin,
Ops
or
cost
management.
We
can
get
a
view
of
different
things
who's
using
all.
F
That
is
something
if
we
plug
ourselves
into
backstage
life
cycle,
because
backstage
is
pluggable,
we
can
use
other
providers
to
do
those
things
right
or
or
we
are
thinking
of
autelius-
can
also
get
into
those
areas
and
start
doing
something
more
than
a
backstage.
A
B
Yeah,
so
we
would
have
our
our
front
end
and
back-end
plug-ins,
and
those
plugins
would
allow
us
to
get
inserted
into
the
customer's
pipeline
in
the
right
places.
So
we
can
then
start
Gathering
the
the
catalog
information
when
they
go
through
and
run
their
Pipelines.
A
Of
the
reasons
why
I
think
that
is
because
backstage
is
actually
part
of
the
problem
when
it
comes
to
creating
a
more
agile
devops
pipeline
because
it
generates
these
basically
pipeline
workflow
files
for
it.
If
we're
you
know
whatever
microservice
or
API,
that
you've
registered
with
it
now,
let's
say
that
the
company
wants
to
switch
from
sonotype
to
jfrog.
A
You
have
to
go,
and
you
have
to
update
all
of
those
already
generated
pipeline
files
manually
and
that's
why
CD
events
is
becoming
more
important
is
because
it
creates
a
more
agile
Pipeline
and
if
CD
events
is
out
there,
then
you
don't
need
something
like
backstage.
A
B
A
Don't
you
don't
name
it?
You
don't
hard
code,
the
name,
you
hard
code,
the
action
right
and
then
the
name
is
substituted
right.
A
B
Yeah
and
that's
where
the
the
complexity
comes
like
you
know,
we
we've
we've
seen
companies
where
they
have
15
QA
environments
for
one
project.
You
know
so
that's
where
it's
going
to
be
interesting
and
how
that
fits
into
that
solution.
A
A
We're
going
to
continue
work
with
the
the
CD
events
team
at
the
CD
foundation
and
really
look
to
see
how
that's
going
to
shake
out.
B
Yes,
and
and
for
our
implementation
like
on
the
let's
say
you
go
on
to
they,
they
have
they've
put
Captain
into
their
cluster
and
we
want
to
know
when
a
deployment
has
been
completed
we
would
have.
We
would
be
listening
for
a
cloud
event
from
Captain
to
tell
us
that
something's
been
completed.
So
when
we
talked
when
we're
integrating
with
these
other
tools
that
we'd
be
doing
it
on
a
cloud
event
type
of
payload
to
make
it
generic.
B
So
if
they
take
out
captain
and
put
in
something
else,
unless
they're,
just
using
Argo
CD
and
leaving
Captain
out,
we
could
listen
for
an
Argo,
CD
Cloud
event
saying
that
the
deployment
was
finished.
It
would
be
basically
the
same
event.
Just
it's
just
a
different
producer.
It's
the
same
same
Cloud
event
that
we're
listening
for,
but
a
different
producer,
whether
it's
going
to
be
Argo
versus
captain.
B
It
is
kind
of
twofold:
one
is
the
the
well
it's
the
developer,
so
the
developer
wants
to
create
a
a
new
service
and
when
they
create
the
new
service,
they're
going
to
create
it
in
a
particular
language
and
they
get
like
kind
of
like
get
a
stubbed
out
starting
point
for
that
that
new
service,
so
that's
kind
of
from
the
developers
perspective.
B
Now
that
developer
wants
to
get
that
new
service
out
to
production,
and
that's
where
the
devops
engineers
come
into
play
and
normally
the
devops
engineers
have
to
go
and
manually
create
a
Jenkins
file.
For
example,
then
they
have
to
manually
go
configure
a
namespace
in
the
kubernetes
cluster
and
then
they
need
to
go
and
configure
Prometheus,
and
then
they
need
to
go
configure.
B
You
know
stackdriver
for
looking
at
the
logs,
for
example,
or
grafana
to
watch
the
transactions,
so
the
the
devops
engineers
actually,
where
all
the
heavy
lifting
happens-
and
this
is
why
this
is
why
backstage
came
about
out
of
Spotify.
Is
they
had?
They
had
so
many
microservices
that
they're
creating
that
they
couldn't
the
the
devops
team
couldn't
keep
up
with
creating
all
these
manual.
B
Doing
all
these
manual
steps
to
onboard
somebody,
you
know
to
onboard
a
new,
a
new
developer,
a
new
service
for
a
developer
was
taking
a
week
going
through
and
pointing
and
clicking
and
copying
and
pasting
and
all
that
stuff.
So
that's
where
it
came
from
was
using
this
templating
idea
to
go
ahead
and
get
it
rolled
out
quickly
within
just
a
couple.
You
know
a
couple
minutes
when
you
point
and
click
and
say
go
so
it's
the
the
audience
is
typically.
B
The
devops
engineer
is
where
they're
gonna
allow
more
fewer
devops
Engineers
to
be
able
to
support
more
teams
for
the
developers
to
be
onboarded.
D
A
A
B
Yeah
in
in
that
you
have
the
devops
teams
and
the
product,
you
know
the
production
of
the
release,
managers,
the
production,
control
teams,
get
to
pick
and
choose
the
constraints
that
they
want
to
put
the
developers
into
so,
for
example,
the
the
production
control
can,
in
the
template
say
you're
only
going
to
deploy
to
AWS
we're
not
going
to
allow
you
to
go
to
azure.
B
So
that's
where
the
the
controls
come
into
play
through
through
backstage
it's
just
like
like
Ankara
was
saying,
is:
is
a
giant
templating
engine.
F
The
way
we
have
I
have
used
it
in
my
organizations
that,
but
there
is
a
there
is
a
part
of
the
ugly
part.
Is
that
because
it's
open,
you
know
you
end
up
customizing
it
a
lot
right,
because
you
are
extending
as
per
what
your
organization
needs
right,
because
every
organization
has
different
tools
Technologies,
so
you
end
up
customizing
it
a
lot
and-
and
you
need
to
definitely
that
platform
engineering
team
needs
to
learn
backstage
to
that
extent
that
they
have
to.
F
They
should
be
able
to
support
all
these
nuances
of
how
people
want
to
see
backstages.
So.
B
Yeah
and
and
that's
where
you
know
we
we've
we've,
we've
always
thought
you
know.
This
is
always
a
dream
that
companies
had
what
was
going
to
be
I'm,
going
to
standardize
on
Azure
Pipelines,
and
you
get
a
a
consulting
company
that
comes
in
to
write
a
project
and
they
want
to
do
everything
in
GitHub
actions
and
GitHub.
B
So
now
you
have
to
support
as
all
of
azure
pipelines,
plus
GitHub
as
your
your
platforms
now
and
that's
one
thing
that
people
always
thought
that
would
happen
is
everybody
would
be
on
the
same
technology
stack
in
the
same
sem
tools
and
the
same
and
end
time:
runtime,
runtime,
environment,
and
that's
just
not
the
case
and,
like
you
said
it's
it's.
The
the
customization
and
of
the
templates
is
where
you
put
a
lot
of
work,
but
once
you
put
the
work
in
in
place,
onboarding
people
through
those
templates
is
is
very
easy.
F
And
there
are
multiple
other
companies
now
Ops
level
there
is
like
humanitect
and
like
this
is
pretty
hot
pocket
that
we
are
building.
The
idps
like
we
are
talking
about,
but
I
think
the
value
is
like
you
said
to
Steve:
it's
it's
more
about
how
you
can
make
your
developer
more
productive
so
that
they
focus
more
on
writing
the
service
rather
than
something
which
is
around
the
service.
Then
you
don't
need
to
worry
about
that.
So.
B
Yeah
and
that's
the
that's
the
thing
that
you
you
don't
even
though
developers
love
the
script,
you
don't
want
them
in
their
scripting
Jenkins
files
and
and
having
different
Jenkins
files
for
every
single
service,
because
they
decided
to
do
something
slightly
different.
You
know
one
one
one
developer
may
want
to
script
all
the
docker
build
commands
where
another
developer
may
use
some
reusable
Jenkins
library
that
the
devops
team's
written
and
that
that's
the
type
of
thing
that
you
want
to.
B
F
Just
more
exciting
the
way,
the
way
we
have
implemented
in
one
organization
is
that
even
the
git
repo
you
don't
need
to
create.
Okay,
everything
is
done
so,
basically,
the
the
you
don't
have
to
worry
about
now
that
how
the
port,
how
you
are
setting
up
the
code,
everything
is
kind
of
automated.
Now
you
just
have
to
launch
that
utility
to
get
this
created
and
then
yeah
I
think.
F
B
Yeah
and
the
problem
still
exists
with
most
of
the
CI
CD
pipeline
tools.
Is
those
themselves
aren't
templated,
so,
like
Tracy
was
saying,
if
you
want
to
go
in
and
switch
out
a
security
scan
tool
for
projects
that
have
already
been
created
through
backstage
you're
going
to
have
to
go.
Do
those
manually,
you
know,
there's
not
going
to
be
another
automation
process
to
do
that.
B
And
from
the
artillius
side,
where
we
like
I
said
where
we
fit,
is
we
want
to
be
able
to
be
inserted
into
the
pipeline
into
the
right
places
automatically?
B
B
And
once
we
start
Gathering
the
data,
then
we
can
start,
you
know
doing
the
fun
things
like
we've
been
talking
about
today,
being
able
to
rate
something,
be
able
to
look
at
statistics
metrics.
You
know.
Those
type
of
things
is
where
we
are
able
to
pull
all
that
information
together
into
a
single
hub,
that
we
can
make
decisions
from.
F
I
think
the
answer
is
make
strike
because
it
all
depends
on
different
companies
like
I,
am
working
for
mostly
Financial
firms
and,
like
some
of
them
have
opted
for
it,
and
some
of
them
are,
you
know
they
said.
No.
We
want
to
do
like
one
company
is
a
banking
firm
and
they
decided
to
build
their
own
platform,
but
they
have
now-
or
you
know,
10
people
team
supporting
building
that
platform,
because
because
their
environment
is
like
they
have
around,
you
know
six
thousand
developers,
so
they
are
definitely
building
their
own
platform.
F
But,
like
the
other
company,
which
is
a
small
bank,
they
have
used
they're
using
backstage,
because
it's
easier,
and
especially
with
solution
like
like
Roadie,
it's
a
SAS
version
of
Backstage,
so
you
you
can
actually
even
don't
need
to
maintain
the
kind
of
thing
all
you
need
to
write
is
temperature
is
what
you
want
to
do
for
your
organization,
so
maybe
so,
but
I
I
think
connecting
the
dot
to
what
areas
I
think.
F
Are
we
gonna
be
then,
like
this
question,
Steve
that
we're
gonna
be
more
like
a
backstage
oriented
provider
or
something
where
we
are
competing
with
backstage
in
that
space
or
no.
B
We're
we're
gonna
be
adding
our
plugins
to
the
marketplace.
Okay,.
B
So
we've
we've
already
talked
to
the
Roadie
folks
about
adding
to
their
Marketplace,
and
they
said
when,
whenever
our
stuff
is
good,
to
go,
they'll
go
ahead
and
basically
help
us
do
a
pull
request
to
add
it
to
the
the
Roadie
Marketplace
and
as
far
as
the
generic
backstage
I
have
not
looked
into
that
on
what
we
need
to
do
on
that
front.
Yeah.
F
E
And
I
think
like
Steve,
do
you
know
anyone
from
the
backstory?
It's
a
good
good,
good
topic
to
discuss
on
the
paymentelia's
podcast.
Have
you
know
anyone
from
the
team.
B
Yeah,
well,
we
can,
we
can
probably
hunt
somebody
down
we'll,
have
to
look
and
see
what
they're
up
to
is
backstage
is
in
CNC
off
right,
yeah.
F
That
is
an
incubation
they
have
I
think
applying
for
graduation,
but
right.
B
All
right
we're
coming
up
on
the
end
of
the
hour,
any
other
questions.
Thank
you
Tony,
for
putting
this
together
and
you
know
getting
the
ideas
down.
B
B
Alrighty
happy
holidays,
everybody
thinks
don't
work
too
hard
through
the
new
year.
B
Full
discussion,
yep
and
I
will
be
throwing
things
out
onto
our
Discord
Channel
when
I
get
issues
created
and
stuff
like
that,
so
we'll
I'll
keep
you
posted
that
way.