►
From YouTube: Layer5 Newcomers Meeting (April 1st, 2021)
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).
B
A
C
Well,
hopefully,
there
are
days
when
we
just
have
two
persons,
or
sometimes
just
one
person,
so
fingers
crossed
yeah.
B
D
It's
a
good
in
some
respects.
Maybe
it
makes
for
a
good
study
hall
sort
of
like,
while
we're
sitting
here
waiting.
We
can
knock
out
a
few
other
items
all
right
actually
since
you're
here
and
we
don't
always
get
you
and
I
don't
get
that
that
much
time
to
catch
up
there.
D
It
might
be
worth
well,
especially
as
like
chico
africa
comes
around
like
pretty
soon.
By
the
way,
are
you
guys
having
a
call
later
today
on
that.
D
D
D
D
D
D
D
Okay,
cool,
hey,
hey
g
bro,.
B
D
Bro
help
me
out
with
your
last
name:
it's
o
y,
o
o
y
o.
D
D
D
I'm
sorry,
I
think
of
you
as
this
splendid,
robotic
woman.
E
Okay,
so,
while
we're
here,
it
seems
some
binaries
are
not
compatible
with
alpine.
Isn't
ghost
supposed
to
be
cross-platform.
D
Yeah
yeah:
that's
there
is
oh
okay,
so
there's
so
just
for
context,
for
anita
and
for
asuka.
D
We've
been
using
it
because
well
because
it
outside
of
the
golang
that
meshri
is
written
in
so
mesherie
server
gets
compiled
is
go.
It
is
written
and
go
and
it
gets
compiled.
D
D
D
So,
okay,
good.
So
why
are
we
using
ubuntu
this
big
heavy
image
for
measuring
it?
Just
takes
people
longer
to
download,
but
you
well
there's
a
couple
of
other
components
that
get
dropped
into
place
in
the
same
meshery
image.
So
we
said
the
ui,
the
meshre
server
and
then
there's
at
least
and
and
the
only
two
that
come
to
mind:
they're
both
load
generators.
D
D
Traffic,
it
will
do
you
can
give
it
different.
You
can
give
it
specific,
http
headers
to
help
mimic
different
types
of
requests,
but
if
you
wanted
to
do
tcp
or
just
straight
up,
tcp
or
grpc
or
udp,
even
it's
not
possible
with
the
current
load
generators,
and
so
that's
actually
an
outstanding
like
feature
request
for
the
load.
Generators
vishal
is
with
us.
That's
fantastic!
D
E
E
D
Well,
the
nice
thing
about
something
like
supporting,
tcp
or
grpc
is
that
it's
a
well
it's
meaningful
to
the
project
like
it's
significant
enough
that,
like
it's
a
line
item
on
a
release
like
oh,
you
know
now
supporting
this
thing
and
there's
some
learning
and
things
to
do
in
there.
It's
differentiating
for
the
project,
because
none
of
them
or
most
the
other
projects
in
this
space,
they
won't
go
that
deep.
D
You
know
as
deep
as
measuring
wood
in
terms
of
giving
you
all
those
options
and
the
other
reason
why
it's
interesting
is
because
or
kind
of
a
nice
project
is
because
well
it's
not
it's
like
there's
a
lot
of
prior
art
in
the
code
base
for
how
to
support
that
type
of
a
thing.
D
So
it's
not
like
having
to
bring
in
a
whole
new
architectural
component.
Rather
you
rather
you're
augmenting.
What's
there
so
it's
so
it's
actually
achievable
like
it's
a
not
not
yet
not
super
hard.
It's
moderate.
D
But
so
then
so
vishal,
one
of
the
things
that
grill
is
looking
at,
is
reducing
the
size
of
the
docker
images
from
s3
and
there's
two
binaries
that
outside
of
mesh
reservoir,
there's
two
other
binaries
that
we
put
on
there.
So
in
the
same
image,
one
of
them
is
wrk2,
it's
written
in
c,
plus,
plus
or
c.
D
I
think
it's
c
plus
plus,
but
it
could
be
c
and
I've
seen
just
a
quick
google
search
this
morning,
like
first
of
all,
I
thought
that
we
had
wrk
2
running
on
alpine
with
meshri
a
while
ago,
but
I
could
be
misremembering,
but
if
you
do
a
quick
google
search
for
wrk2,
that's
one
of
the
load.
Generators
on
alpine
you'll
find
a
number
of
public
repos
docker
hub
repos
that
have
wrk2
and
alpine.
So
okay,
so
it
must
there's
some
configuration
in
which
that
works.
D
Okay.
So
then,
I
think
there's
I
think,
there's
just
one
hiccup
left
like
in
that
that's.
There
are
two
other
generators
load
generators
that
meshery
supports.
One
of
them
is
fortio.
Maybe
we
should
be
writing
some
of
this
down.
D
Actually,
you
know
what
it's
in
the
docs.
One
of
them
is
for
it's
by
a
frenchman,
and
I
was
giving
a
talk.
I've,
given
you
know,
talks
on
the
load,
generating
capabilities
of
measuring
a
few
different
times,
but
one
of
the
times
he
was
attending
it.
It
was
so
I
think
it
was
dockercon
and
he
he
corrected
me
on
stage.
It's
not
it's
not
fortio.
It's
fortio.
D
He
he's
a
nice
guy,
he
was,
I
mean,
you
know
it's
his
project
and
so
I'm
sure
he's
like
you
know,
I'm
sure
it
like
rubs
on
his
ear
nubs
whenever
I
think
he
even
has
it
well.
I
guess
so.
It
looks
like
a
greek
word
and
this
one
fortio
and
part
of
the
reason
we
got
started
with
40
was
40
was
two:
oh
hey
check
that
out:
oh
yeah,
it
is
which
means
load
burden
right
because
I
guess
measuring
isn't
worth
its
salt,
because
otherwise
it
would
have
been
named
greek.
D
So
I
digress.
Sorry.
The
point
is
this
is
written
in
go,
and
so
we
measury
incorporates
fortio
as
a
library
and
and
compiles
that
go
down
to
the
right.
You
know
compiles
that
down,
and
so
it
runs
on
the
image
that
we
desire
for
toronto.
D
Okay,
so
then
the
last
load
generator
the
third
one
is
nighthawk.
So
we
talk
about
fortio.
We
embed
this
as
golang.
This
is
c
plus
plus
I
think
we've
been
able
to
get
that
compiled
down
correctly
nighthawk
it.
It
is
written
in
c,
plus,
plus
and
we've
had
and
compiling.
It
is
enough
of
a
headache
and
that's-
and
I
think
that's
an
accurate
and
fair
statement
is
that
we've
created
an
entire
project
to
address
it
actually
we'll
be
talking
about
the
project
at
and
it's
it's
yet
to
be
really
announced
and
released.
D
D
It's
just
just
getting
started,
there's
a
description
of
it
here.
When
we
started
this
project,
I
decided
to
try
to
get
it
going
much
more
in
context
of
the
cncf
than
anywhere
else,
and
so
we've
been
so
there's
a
cncf
service
mesh
working
group
meeting
later
today
in.
D
Four
hours
or
four
hours
and
40
minutes
it's
at
one
o'clock,
central
here
so
four
hours
and
40
minutes
from
now,
and
we
use
that
meeting
time
to
talk
about
get
nighthawk.
We
also
talk
about
service
mesh
performance.
D
We
do
a
number
of
things
in
there,
but
yeah,
but
part
of
the
point
of
this
project
is
to
take
the
load,
generator
called
nighthawk,
which
is
it's.
It's
a
load
generator
that
was
created
as
part
of
the
envoy
proxy
project
and
in
order
to
and
so
envoy
is
written
in
c
plus
plus
nighthawk
is
also
then
written
in
c,
plus,
plus
just
like
istio
control
plane
is
written
in,
go
and
fortio
is
written
in,
go
like
the
projects
are
very
related.
They
can't
written
in
the
same
languages.
Nighthawk
is.
D
We
have
a
number
of
plans
for
it,
hence
between
those
evil
plans
that
we
have
and
the
challenge
of
like
getting
it
built
and
trying
to
get
it
incorporated.
Like
tight.
You
know
more
tightly
integrated
into
measuring.
We
decided
to
create
this
project
the
nighthawk
itself
as
a
project
like
super
nerdy
and
interesting,
but
like
this
is
the
the
site.
This
is
the
this.
Is
the
project
page
like
there
isn't
a
project
page.
D
So
since
we
were
going
to
have
to
be
doing
a
lot
of
builds
and
integrating
with
mesheri,
we
figured
heck
shoot,
let's
just
let
everybody
else
benefit
as
well
and
help
uplift
the
project
by
giving
it
making
the
builds
that
we're
going
to
be
doing
available
to
everyone
like
once
we
start
having
to
build
the
project,
maybe
putting
it
out
into
different,
build
different
formats,
wouldn't
be
too
hard,
and
by
formats
I
mean
things
like
this,
like
I'll,
create
create
a
few
different
distributions
of
nighthawk
so
that
maybe
we're
creating
a
container
image
or
a
couple
of
them.
D
D
D
D
D
D
Abhishek
kumar
is
the
one:
that's
been
that's
of
the
layer.
Five
community
direct
directly
he's
been
the
most
successful
with
getting.
D
So
under
github
under
layer,
five
io
get
nighthawk.
That's
this
is
the
repo
for
the
project,
so
both
the
website,
the
jekyll-based
website,
is
here
and
it
it's
in
need
of
some
love.
I
mean
you
can
tell
from
looking
at
it,
but
then
the
workflows
that
actually
build
so
there's
kind
of
two
other
parts
to
get
nighthawk.
D
So
one
one
is
what
we
were
talking
about.
Just
workflows
like
how
do
you
build
this
thing
and
how
do
we
build
it
for
debian
or
ubuntu
or
rel
or
fedora
or
whatever?
D
Whatever
any
of
those
things
are?
I
have
to,
I
just
say
like
for
the
most
part,
I'm
very
keenly
interested
on
getting
a
nighthawk
build
that
works
best
for
mastery
and
focusing
there
and
getting
that
done
first
and
then,
however,
other
many
distribution
formats,
the
community
can
put
out
great
the
the
third,
the
second
like
the
third
component
to
this
effort.
So
there's
a
website
there's
the
building
of
the
project
of
nighthawk
to
various
architectures
and
the
third
one
is
well
the
integration
into
measuring.
D
Well,
I
think
initially,
though,
the
direct
the
direction
that
we
took
was
to
wrap
golang
around
nighthawk,
making
system
level
calls
like.
Basically,
you
know,
you're
using
nighthawk
to
binary,
like
you
would
on
the
command
line,
which
you
know
it's
fragile.
It's
not
a
it's.
You
know
yeah,
it's
not
perfect,
but
nighthawk
itself.
It
actually
has
so
I
have
to
go
back
and
review,
but
we're
using
grpc.
Now
between
mystery
and
nighthawks,
we're
using
grpc
to
communicate.
D
I
have
to
go
back
and
review
whether
or
not
the
grpc
sitting
in
front
of
nighthawk
is
nighthawks
grpc
service.
I
believe
that
it
is
which
means
that
the
calls
that
you're
making
those
remote
procedure
calls
they're
being
executed
directly
by
nighthawks
c
plus
code.
I
believe
that
that's
the
case
at
one
point,
though,
we
had
had
a
go
wrapper
around
it,
and
so
I
think
this
repo
still
has
some
of
that
go
code.
D
D
The
other
thing
that's
going
to
happen
is
we'll
get
a
presentation
of
last
time.
We
got
a
presentation
from
red
hat
and
rancher
on
submariner.
Oh
linker
d
is
up
for
graduation.
Damn
it
I
gotta
go
go
approve
the
yeah
all
right,
so
we'll
do
link
or
d
graduation
today
we'll
get
a
presentation
from
kg
k8gb.
D
D
D
Which
to
me
is
like
nice:
I
want
to
smack
somebody.
This
is
stupid.
It
took
six
months
for
ambassador
to
get
through
the
proxy
that
you
know
the
microservices
proxy
that
wraps
envoy.
D
Yeah
we
already
have
the
one
from
heptio.
What
is.
D
F
It's
not
a
problem,
okay,
so
another
question
like:
what's
the
difference
between
cncf
incubation
graduated
and
the
other
ones,
yeah
sandbox
yeah.
D
Great
question:
well
that
changed
recently
by
the
way:
yeah,
okay,
that
changed
recently,
so
there's
there's
three
levels
of
project
maturity
within
the
cncf,
so
it
used
to
it
used
to
be
that
there
was
one
and
then
started
to
you
know,
set
these
levels
of
graduation
these
levels
of
maturity,
so
that
primarily
for
kind
of
a
couple
of
purposes
like
one
purpose,
is
to
hint
indicate
to
the
world.
To
let
the
world
know
that
certain
projects
are
considered.
D
You
know
very
mature,
like
very
production,
ready
or
mature
or
like
at
various
stages
in
their
life
cycle,
very
boring,
maybe
or
or
still
volatile.
The
other
reason
why
the
levels
were
created
is
to
apportion
the
spotlight
like
so
to
like
there's
only
at
cubecon.
Last
time
we
ended
up
giving
four
talks.
D
I
think
we
turned
down
the
fifth
because
just
couldn't
do
another
one
and
but
generally
that
never
happens
like
to
to
we're
so
involved
there
that,
like
we
end
up
speaking
on
a
lot
of
stuff,
but
for
most
it's
really
hard
to
get
a
talk,
and
you
most
people,
usually
don't
and-
and
my
point
is
there's
only
so
many
slots
and
the
more
projects
that
there
are
the
more
contention
there
is
over
the
spotlight
over
the
marketing
spotlight.
D
D
And
I'll
say
that
I
disagree
with
this
decision
and
was
vocal
about
it
of
one
of
the
few
times
I've
kind
of
disagreed,
with
the
direction
of
some
of
the
things
that
we've
done.
D
It
was
to
take
sandbox
since
dramatically
change
its
entry
level
requirements
to
to
like
almost
nothing
but
yeah.
So
so
so
it's
changed
and
I'll
describe
it
as
it
is
today,
which
is
well
the
two
things
that
I
had
said
about
what
a
project
gets
and
what
it
tells
the
rest
of
the
world.
D
And
then
the
the
require
the
entry
requirements
into
each
one
are
specific
in
some
respects,
they're
a
little
bit
what's
written
on
paper.
Some
of
it
is
maybe
easier
to
achieve
than
you
might
expect.
D
Some
of
it
is
about
to
be
a
very
heated
debate
with
respect
to
linker
d's.
Graduation
linker
d
is
submitting.
I
I
need
to
review
it
today.
Actually
before
we
meet
they're
submitting
to
move
from
incubation
to
graduation
and
one
of
the
requirements
is
that
literally
what's
written
on
paper
is
that
they
have
to
the
project
has
to
have
at
least
two
contributors
from
two
different
from
different
organizations
that
are
employees
of
different
organizations,
and
I
think
the
spirit
of
the
word
contributor
in
that
sense
was
meant
to
be
maintainer.
D
Only
kubernetes
kubernetes
was
the
only
graduated
project
for
maybe
a
year
or
more,
then
I
think
prometheus
was
next
and
and
linkery
they
don't
have
they
all
of
their
maintainers
are
from
one
company
that
hadn't
always
been
the
case,
but
over
the
life
cycle
of
the
project,
some
people
have
come
and
gone.
D
D
Yeah,
so
there's
there's
other
yeah,
so
speaking
of
like
yet
another
load,
balancer
or
hey,
there's
a
lot
of
projects
that
are
fairly
that
fairly
well
overlap.
That
part
of
the
spirit
of
part
of
the
philosophy
there
is
that
the
cncf
like
internal
to
the
cncf.
We
don't
make
we're
not
king
makers,
meaning
we
don't
say
oh
of
these
four
projects
that
do
the
same
thing
there.
That's
the
one
that
that's
the
king
right
there.
That's
that's
the
one!
D
That's
sorry
bump
my
mic,
that
isn't
the
intention,
and
so
the
cncf
does
welcome
in
competing
projects
and
naturally,
over
time,
one
or
more
end
up
emerging
as
being
successful.
D
In
our
case,
like
with
part
of
part
of
measures,
focus
is
on
multiple
meshes
mescheri's.
True
value
is
isn't
on
supporting
multiple
meshes.
It's
on
helping.
You
operate
a
service
mesh
with
confidence
and
doing
all
kinds
of
fantastic
things.
A
lot
of
things
that
we
haven't
talked
about
broadly
in
the
community
yet,
but
some
road
map
that
I
think
would
blow
your
hair
back.
D
It's
cream
on
the
top
that
the
things
that
we
do
will
just
work
with
any
mesh.
That's
awesome,
so
the
reason
I'm
mentioning
multi-mesh
and
king
making
of
projects
is.
That
is
that
I
don't
believe
that
we'll
the
world
will
that
will
consolidate
on
just
a
single
mesh.
D
D
I
bet
none
of
you
are
using
this
silly
mic
that
I
just
hit,
not
because
this
is
a
great
one,
but
just
because
the
world
is
made
up
of
all
kinds.
Why
are
we
having
to
compile
nighthawk
to
different
os's
because
there's
a
bunch
of
os's,
because
the
world
is
made
same
thing
with
service
meshes
like
like
it'll
it?
It
is
and
will
continue
to
be,
a
multi-mesh
world
like
there's,
there's
no
question
and
if
anybody
wants
a
deep
explanation
of
that,
come
come
see
me.
D
It
is
the
case
that
there
will
be
three
about
that
really
enjoy
much
of
the
focus,
one
that
takes
more
of
the
market
share
than
than
all
the
rest,
all
the
rest
combined.
That's
just
like
a
general
principle
of
economics
actually
like
in
most
markets.
That's
generally
the
way
it
is,
if
you,
some
of
you,
might
be
able
to
spout
off
the
top
of
your
head,
what
browser?
What
web
browser
I'm
using
safari
right
now,
I
bet
none
of
you
are
using
safari
at
the
moment.
Maybe
one
of
you.
D
I
mean
not
yeah
by
the
way,
I
wouldn't
judge
your,
whether
you're,
maybe
using
brave
or
like
what
it
means
opera
but
but
yeah
anita
like
to
what
you're
saying
just
between
the
six
of
us.
There
were
three
web
browsers
and
the
other
three
didn't
say
anything,
and
so
he
like,
though,
that's
just
a
great
example
of
like
over
time,
what
used
to
be
the
world's
most
popular
web
browser
internet
explorer?
Why?
D
Because
most
of
the
world
most
of
us
are
dunces
on
the
computer,
and
we
most
of
us
had
windows,
and
that
was
just
like
the
blue
e.
That
was
like
the
internet.
That's
you
know,
and
over
time
stuff
shifts
and
the
organizations
behind
some
of
the
meshes
are
so
heavily
invested
in
them.
That
and
they're
pitted
against
each
other
as
well.
That
they'll
always
be
there.
D
I
mean
I
shouldn't
say:
they'll
always
be
there.
I
mean
some
will
die
off,
but
it
wouldn't
consolidate
to
a
single
thing.
Kubernetes
is
a
bit
of
an
anomaly
in
this
sense.
D
Is
the
single
because
kind
of
the
single
standing
container
orchestrator?
Technically,
it's
not
like
hashicorp
still
makes
nomad.
They
still
yeah.
D
Yeah
openshift
used
to
be
on
well,
it
used
to
like
technically
not
even
use
docker
and
they
had
switched
to
rebasing
on
kubernetes
and
openshift
v2
and.
D
D
Anyway,
boy,
you
asked
me
a
question
I
can
just
blabber
on
for
like
40
minutes.
F
Okay,
another
question
you
can
use
measuring
to
install
service
meshes
right,
yeah,
okay,
great.
D
D
D
I'd
probably
have
a
smirk
on
my
face
about
what
these
idiots
are
doing.
On
measuring.
C
D
Wasting
all
this
time
with
compatibility
with
other
meshes
and
then
like
having
to
focus
a
lot
of
time
on
beginner
type
flows
of
like
installing
them
and
doing
simple
life
cycle
management
of
them
like
yeah,
I
gotta
say
yeah,
it's
been,
it's
been
kind
of
boring
and
it's
been
it's
frustrating
because
we're
not
getting
to
the
super
interesting
stuff.
Just
yet
we're
about
there,
but
part
of
what
they're
missing
is
that
it
will
be
a
multi-mesh
world.
They're
missing
that
and
it's
a
long,
long
term
strategy
on
that
they're.
D
Also
missing
that
it's
going
to
take
people
another
decade.
People
will
still
be
coming
to
service
measures
for
another
decade,
docker's
been
here
eight
years
a
hundred
people
just
came
to
docker
in
the
time
that
we've
well
not
a
hundred,
but
maybe
maybe
a
hundred
have
come
to
the
docker
for
the
first
time
since
we
started
this
call
like
people
are
still
coming
to
it.
For
the
first
time
they
take
people
a
long,
darn
time
to
get
to
a
service
mesh.
It
takes
a
while
long-term
strategy
works
out
a
bit
better.
D
Why
am
I
saying
this?
Oh
yeah,
because
part
of
like
being
able
to
install
them
is
like
helping
people
evaluate
them,
helping
people
interoperate
them
so
helping
people
evaluate
them,
understand
them
and
then
over
time
build
confidence
with
them
and
over
time
the
functionality
of
measure
will
just
get
more
and
more
sophisticated.
D
Is
that
another
effort
we
have?
Is
the
service
mesh
patterns
book.
D
And
in
this
book
well,
there's
a
github
repo
surface
mesh
patterns,
we're
still
kind
of
working
on
that,
but
surface
mesh
patterns.
This
book
will,
I
have
no
idea
when
it'll
come
out,
takes
a
long
time,
but
we're
building
out
a
bunch
of
patterns
about
how
to
do
certain
things
in
service
meshes
and
what
the
pattern
is,
how
to
do
it
well
and
it's.
The
pattern
is
agnostic
of
the
mesh.
So
it's
not
like
it's
a
link
or
d
cookbook,
or
something
like
that.
It's,
oh!
D
So
it's
like
a
design
pattern
for
service
missions,
yeah,
exactly
there's
it'll,
probably
it'll.
Unless
something
weird
happens,
it'll
be
a
two-part
book
series
like
it'll,
be
the
first
one
on
foundational
30
patterns
and
the
first
one
30
patterns
in
the
advanced
one
and
jubril,
since
this
is
gonna,
be
mesh
agnostic
and
we're
gonna,
and
we
want
the
book
to
have
a
long
shelf
life.
You
know
they'll
be
exact
and
you
wanna,
like
reinforce
every
pattern
with
an
example
of
like
how
to
do
this
thing.
D
D
Yeah
the
book
just
came
out
in
early
release.
I
actually
I
don't
think
I
really
mentioned
it
broadly
there's
the
first
couple
of
chapters
are
out
on
the
o'reilly
platform.
Now
second
chapter
is
nothing
but
measuring,
and
then
from
there
on
out,
mystery
will
be
mentioned
in
every
single
other
chapter.
D
What's
our
philosophy
like
or
that
I'm
giving
this
as
an
example
of
like
part
of
the
mentality
toward
long-term
strategy
on
the
projects
being
successful,
it
takes
a
lot
of
faith
and
it
takes
a
lot
of
to
one
of
our
conversations
before
it's
nice
to
see.
D
Amia
back,
but
tons
and
tons
and
tons
of
people
come
through,
say,
hi
take
off,
never
come
back,
we
invest
in
them,
try
to
help
them
teach
them
stuff,
they
leave
their
comments,
and
for
some
folks,
like
me,
is
a
great
example
is
like
hey
life,
changes
and
or
just
organically,
and
very
naturally,
like
focus
shifts,
folks,
come
come
and
go,
and
and
what
I'm?
D
I
wasn't
really
looking
to
comment
on
that
as
much
as
I
was
trying
to
say
that
keeping
the
faith
long
term
that,
like
doing
investing
and
writing,
is
yet
another
book
is.
D
Hopefully
it
is
like.
Hopefully
it
really
helps,
as
people
will
be
coming
to
learn
this
stuff
for
a
long
time,
a
lot
of
times
they
turn
to
a
book
from
o'reilly
or
out
elsewhere
to
learn,
and
in
this
case
mystery
will
be
right
in
their
face,
so
we'll
service
mesh
performance,
which
we'll
be
talking
about
a
fair
bit
in
the
book
and
service
service.
Mesh
interface
smi,
which
you
know
mesh
re,
is
the
conformance
tool
for
so.
D
We're
taking
a
different
approach
and
building
community
so
yeah.
The
first
two
chapters
I
think,
are
out
in
early
release
on
the
o'reilly
platform
and
then,
like,
I
think,
technically,
the
book
was
supposed
to
have
been
published
by
now,
but
I've
got
so
much
going
on
just
takes
a
long
time.
D
So
speaking
of
I'll
be
inviting
some
people.
I've
opened
up
invitations
in
the
past
for
people
to
participate
in
in
the
book
and
there's
different
ways
to
participate.
But
I'll
say
this
I'll
I'll
reach
out
I'll.
Ask
you
if
I
think
that
you
can
be
additive
to
it.
D
But
yeah:
no,
it's
a
lot
of
work.
D
Oh
so
gibral,
you
know,
based
on
our
last
conversations,
I
I
encourage
you
to
jump
into
get
nighthawk
like
it
needs
help
and
actually
of
you
might
be
able
to
help,
like
you
know,
of
the
things
that
you
already
know
and
that
you
were
kind
of
naturally
already
suggesting
to
be
fixed
that
that
last
standing
issue
with
nighthawk
boom
there's
a
project
for
trying
to
build
it
differently.
D
It's
a
good,
a
good
excuse
to
go,
make
a
mark
on
get
nighthawk
as
a
matter
of
fact
today,
part
of
what
I'm
gonna
do
as
soon
as
I
get
off
is
go
to
go
nominate
three
maintainers
on
the
get
nighthawk
project
and
there's
probably
room
for
more
there's
room
for
maintainers
across
a
number
of
the
projects
that
we
have.
D
If
you
guys
are
ever
curious
like
for
the
most
part,
there's
kind
of
two
places,
you
can
look
to
see
who's,
maintaining
what
one
of
those
is
under
there.
Five
io
slash
community,
slash
members
now
that
I
say
this,
this
might
be
out
of
date.
So
if
you
go
to
maintainers,
I
think
we're
we,
who
do
we
just
promote
to
navendu
yeah
navendu,
just
got
promoted
to
a
mastery,
ctl
maintainer.
He
needs
to
be
here.
D
We
should
take
a
note
actually
anita,
like
I
don't
know,
I
need
if
you
either
don't
mind,
creating
an
issue
to
promote
I'll,
send
you
his
name,
the
vendor
or
and
or
just
like
taking
on
the
issue
yourself
you're.
C
C
D
Okay,
you
know
I
misspelled
his
name.
It
doesn't
have
the
a
on
the
last
bit
but
yeah
yeah.
We
need
to
get
him
promoted
to
a
maintainer
or
he
is.
We
need
to
reflect
that.
That's
one
way
to
look
at
where
the
maintainers
are
another
way
to
look
at
where
the
maintainer
or
who's
maintaining.
What
is
I'm
curious
is
in
github,
I'm
curious
for
each
of
you.
D
If
you
go
to
the
layer,
5
io
and
you
were
to
click
on,
I
don't
know
if
you
all
yeah,
I
think
you
should
see
teams,
you
might
not
see
all
of
the
teams,
but
you
should
see
some
of
them
like
there's
one
there's
a
number
of
maintainers
teams
and
you
can
see
some
of
the
different
people
for
the
different
areas
that
they.
D
Maintain
so
the
maintainership
stuff
like
like
we,
I
think,
actually
as
anita
and
ruth
and
others
potentially
work
on
a
community
handbook,
there's
a
very
purposeful.
D
Journey
that
we
try
to
take
contributors
on
and
just
sort
of
growing
in
recognition
and
growing
in
scope
of
responsibility
as
they
go.
We
have
you
know
it's
actually
there's
a
flow
chart
for
like
there's.
It's
pretty
specific.
D
I've
been
falling
down
on
it
a
lot
because
we
don't
have
a
dedicated
community
manager
and
we
have
about
25
like
what
do
you
call
them
ordained
community
managers,
none
of
which
you
know,
I
don't
fault
any
of
any
anyone
for
it,
but
none
of
which
other
you
know
other
than
a
couple
of
them
will
actually
do
some
of
the
tasks
anyway.
Anyway,
point
is
part
of
that
that
isn't
well
defined
is
like
what
is
it?
What
do
you
have
to
do
to
find
ultimately
achieve
becoming
a
maintainer?
D
The
unsatisfying
answer,
but
it's
very
simple
and
true,
is
like
how
do
you
you
know?
How
do
you
get
added
to
the
github
to
this?
How
do
you
get
invited
to
this
team?
How
do
you
get
your
picture
on
the
you
know?
If
you
wanted
a
community
profile,
how
do
you
get
one
of
those
if
you
wanted
to
have
the
twitter
account,
follow
you
or
retweet,
some
of
your
stuff
like
how
does
all
the?
D
How
do
you
get
sent
a
book
or
like
any
of
that,
there's
a
bunch
of
rewards
that
are
kind
of
laden
latent
in
there?
How
do
you
get
an
internship
either
with
layer
5
or
through,
like
there's
open
positions
in
the
the
service
mesh
working
group
that
I
created?
We
need
a
tech
lead
there.
We
need
a
tech
lead
in
the
sig
network.
D
How
do
any
of
those
things
get?
It
comes?
Well
you
just
you
you
do
them,
you
start
maintaining,
and
then
then,
then
you
know
it's
not
like.
Then
you
just
get
acknowledged
for
doing
those
things.
D
I
mean
it's
not
like
you
get
promoted
into
the
roles
like,
oh
hey.
We
think
this
person
is
going
to
do
that
thing
and
we
should
they
look
like
a
good,
let's,
let's
promote
them
and
then
hope
that
that
happens
like
no
hey,
you
just
do
it
and
then
you're
recognized
and
then
you're
uplifted
and
supported.
So
by
the
way
for
those
who
haven't
been
in
industry
in
your
career
for
a
long
time,
that's
generally
how
it
works
in
corporate
in
organizations
as
well.
D
Corporations
that
like,
if
you
want
that
promotion
to
a
manager
or
something
is
like
how
do
you
get
it,
and
this
is
this-
is
weird
when
I
say
it,
but
you
just
start,
you
start
doing
your
role,
you
start
managing
people,
it's
like.
Well,
how
could
you
possibly
start
managing
people
if
you're,
not
you
know,
organizationally
positioned
to
do
such
a
thing?
It
didn't
report
to
you
all
right,
come
get
an
internship
here
and
I'll.
Tell
you
how
it's
gonna
take
a
while.
D
Anyway,
man,
I
just
I've
droned
on
any
other
questions
for
today,.
F
Oh
yeah,
I
was
asking
the
other
day
the
diagrams
for
adapters
yeah,
oh
yeah.
It
I
find
a
bit
confusing
yeah.
Let's
take
it.
Let's
take
a
peek.
D
D
Oops,
oh
wow,
I
forgot
we
were
doing
some
draft
work
on.
This
is
kind
of
ugly,
so,
okay,
so
mysteries
architecture
includes
any
number
of
adapters.
Without
me,
droning
on
and
saying
things
that
maybe
you
already
are
familiar
with
gibral.
Do
you
want
to
do
you
want
to
try
to
explain
it
to
me
and
then
I'll
characterize
it
as
you
go.
D
And
or
let
me
I'll
kick
I'll
kick
start
you
with.
D
Well,
with
the
the
notion
that
each
of
the
service
meshes
themselves,
a
lot
of
them
have
overlapping
functionality
like
a
lot
of
them,
will
allow
you
to
configure
a
rate
limit
right
so
that
a
service
doesn't
get
overwhelmed
with
a
bunch
of
requests.
So
you
can
it's
kind
of
a
common
feature,
but
but
not
all
the
service
meshes
like
all
these
logos
like
they,
they
have
dif.
You
know
they're
different
right.
They
have
some
of
the
same
overlapping
capabilities,
but
not
all
the
same,
and
and
they
try
to
be
different
right.
D
D
Smi
is
two
different
surface
meshes,
meaning
it
just
defines
a
lowest
common
denominator:
specification
set
of
specifications
for
interfacing
with
service
meshes.
That
was
confusing.
What
I
just
said.
Probably
actually
this
is
maybe
the
best
way
of
trying
to
explain.
Measuring
adapters,
I
would
say:
measuring
adapter
is
that
analogy
to
terraform
is
very
makes
a
lot
of
sense.
D
If
mesh
reserver
is
terraform
and
in
terraform
you
can
load
up
any
number
of
providers,
because
maybe
you're
you're
doing
things
on
aws
or
on
azure,
so
you
can
use
different
specific
providers
to
interface
with
different
technologies.
D
So
one
for
each
mesh
now
there's
a
lot
of
overlap,
and
we've
tried
to
here
recently
one
of
the
maintainers
michael
gefeller
had
written
a
blog
post
like
fairly
recently
talking
about
how
we've
tried
to
improve
the
developer
experience
of
like
of
managing
and
sustaining
10
different
adapters
like
gosh.
You
have
to
do
these
things,
get
all
this
stuff
10
times
over.
How
do
we
be
intelligent
about
this,
and
so
that
that's
what
this
talks
about.
D
E
Actually,
I
think,
that's
all
yeah,
like
I
get
how
this
works
so
in
my
mind,
measuring
makes
calls
to
the
adapters,
as
the
user
interacts
with
through
the
ui
or
the
command
line,
so
measuring
server
would
make
calls
to
be
on
the
adapters
and
get
information
from
there
and
I'm
not
sure
what
the
operator
does.
E
Okay
is
this
to
keep
it
in
sync
with
the
cluster?
I
think
you
mentioned
this
a
while
back.
E
Yeah,
so
I
guess
that's
all
I
got
thrown
off
when
I
saw
this
this
platform
on
the
right
here
so.
D
E
B
D
Exactly
see,
that's
actually
one
of
the
reasons
why
there
will
be
multiple
meshes
is
like
linker
d.
Does
a
super
fantastic
job
of
being
lightweight
and
kubernetes
native
and
kubernetes
only
great,
if
it's
all
in
kubernetes
great,
if
it's
not,
maybe
this
is
the
one
for
you
or
this
one
or
this
one
or
this
one
or
this
one
like
a
lot
of
some
of
the
other
ones.
Do
things
outside
of
kubernetes?
D
But
any
anything
else
for
today,
as
a
matter
of
fact,
actually
I
gotta
go.
I
should
I
shouldn't
ask
that
dude
bro
great
questions,
man
thanks
for
making
it
an
interesting
hour
for
me.
Thank
you.
D
It's
nice
to
it's
nice
to
have
asuko
and
anita
and
adina,
see
you
guys
tomorrow
for
the
community
call.
Hopefully.