►
From YouTube: Layer5 Community Meeting (Nov 19th, 2021)
Description
Layer5 Community Meeting - November 19th, 2021
Join the community at https://layer5.io/community
Find Layer5 on:
GitHub: https://github.com/layer5io
Twitter: https://twitter.com/layer5
LinkedIn: https://www.linkedin.com/company/layer5
Docker Hub: https://hub.docker.com/u/layer5/
A
All
right
so
we
have,
I
think
we
have
a
couple
of
newcomers
on
the
call
today,
so
let's
open
the
floor
for
their
introductions.
So
if
you
are
new
on
this
call,
we
have
this
tradition,
where
newcomers
get
to
introduce
themselves
to
the
community
and
get
introduced
to
the
community.
A
So
if
this
is
your
first
time
on,
this
particular
call,
please
introduce
yourself
so
that
we
can
get
to
know
you.
B
A
That
sounds
good,
so,
yes,
that
sounds
really
good.
Welcome
to
the
community.
Thank
you,
nice
to
have
you
here.
A
We
also
have
dpr
mukherjee
on
the
call
dp,
and
this
is
your
first
time
right.
Yes,
yes,
you
are
so.
B
Hey
everyone:
I
am
a
junior
from
india
and
I'm
pretty
new
to
the
project
and
I
come
from
a
web
dev
background.
So
this
things
are
pretty
new
to
me,
the
cloud
native
ecosystem,
but
I'm
willing
to
contribute
to
the
project,
because
I
kind
of
find
it
pretty
appealing
to
me
so
yeah
looking
forward
to
it,
I'm
looking
forward
to
meet
you
all
get
to
know
you
all.
Thank
you.
A
Thank
you
and
welcome
to
the
community.
A
I
guess
that
was
all
the
newcomers
so
welcome
again
to
the
community
and
let's
get
started
with
today's
meeting,
so
welcome
everyone
to
the
life
community
meetings
today
is
the
19th
of
november
and,
like
every
other
meeting
in
the
airfare
community,
this
meeting
is
also
recorded
and
the
video
will
be
published
on
youtube.
A
So
if
you
want
your
two
seconds
of
fame
on
the
internet,
this
is
your
time
and
we
have
a
couple
of
announcements
today.
So
the
first
one
is
that
push
who
is
on
the
call
this
time
so
pierce
has
been
nominated,
as
has
been
accepted,
accepted
as
the
measuring
machine
maintainer
for
measuring
cli.
A
So
we
had
the
nomination
last
week
and
if
you
haven't
seen
these
the
slack
thread
it's
in
the
announcements
channel
and
there
has
been
a
an
email
thread
as
well
so
pusha.
Everyone
has
really
good
words
about
you.
So
congratulations,
and
would
you
like
to
add
something
to
it?.
C
A
Sounds
awesome
pierce,
so
I
I
guess,
like
most
of
you,
are
familiar
with
the
pioches
work,
so
one
of
his
most
impactful
work
here
was
taking
the
test
coverage
from
0
to
33
or
like
up
almost
close
to
40
percent
now
in
measuring
ctl.
A
A
C
By
the
way,
sorry
to
interrupt
a
spoiler
alert,
I
have
been
already
maintaining
a
lot
of
components
secretly.
So
this
is
kind
of
an
official
thing
by
the
way.
A
Nice
and
we
also
have
another
announcement,
abhishek
kumar,
who
has
been
contributing
to
the
community
for
for
a
really
long
time,
has
accepted
his
role
at
layer
5
as
the
as
a
software
engineer
and
he'll
be
joining
us
next
year.
A
All
right
so
moving
on
to
the
topics
for
today.
First
up
we
have
the
machary
and
smp
at
cubecon
china
lee.
Would
you
like
to
talk
about
it.
D
So
you
know
each
so
measuring
so
by
the
way
for
congrats
piyush
for
one,
and
I
missed
the
second
thing
of
what
pu
should
so
congrats
piyush.
Now
that
his
headphones
were
back
in,
I
missed
the
second
part
of
what
push
had
said,
but
I
think
it
had
something
to
do
with
the
fact
that
push
has
been
contributing
in
a
few
different
areas
like
and
creating
a
number
of
things
beyond
the
cli
and
so
yeah
fanta,
which
is
fantastic
yeah.
D
If
you
haven't
tried
measuring
ctl
perf
in
the
last
couple
of
weeks
go,
I
encourage
you
all
to
go.
Try
that
command
mastery
ctl
perf
and
give
your
feedback
to
piyush.
D
There's
there's
a
lot
to
it,
and
so
so
yeah
yeah
go
go,
go!
Give
it
a
try,
go
complain
to
him
about
it.
I
guess
yeah,
it's
avi
was
on
a
call.
Abby
kumar
was
on
a
call
just
a
bit
earlier
and
so
isn't
on
this
call,
but
congrats
to
him
for
landing
a
full-time
role.
Here,
that's
great!
It's
entirely,
based
on
his
contributions
and
engagement
in
the
community.
Abby
is
wrapping
up
an
lfx
internship
with
player
five
now
and
so
clearly,
that's
been
successful
kudos
to
those
guys.
D
So
speaking
of
lfx,
speaking
of
cncf
things,
the
meshery
and
the
service
mesh
performance
project,
two
of
the
projects
that
we
spend
quite
a
bit
of
time
on
there's
other
projects
in
the
community,
actually
there's
a
lot
of
projects
in
the
community,
but
those
two
are
prominent
in
part
in
so
much
as
they
were
recently
adopted
by
the
cncf.
So
there's
a
couple
of
talks
about
them
at
kubecon
china,
which
is
just
around
the
corner,
one
of
the
layer,
five
mesh
mates,
anita
ihuman
she's,
giving
a
talk
there.
D
She
navigated
her
way
through
the
double
bite
characters
to
to
to
get
her
talk
accepted.
So
that's
great
she'll
talk
about
all
that
serving
or
solving
the
service
mesh
adopters
dilemma.
Certainly
one
of
the
things
that
measury
does.
D
And
so
she'll
introduce
some
folks
to
to
measure
we'll
also
have
a
measuring
and
service
mesh
performance
project
office
hours,
I'm
at
kubecon,
china,
we'll
also.
We
also
run
the
cncf
service
mesh
working
groups.
We'll
have
a
talk
there.
D
D
So
there's
there's
a
lot
of
requests
that
we
receive
to
give
talks
at
various
meetups
various
conferences,
various
podcasts
I
just
received
two
this
morning
and
for
my
part
I
don't
have
time,
but
but
very
much
so
like
to
connect
any
of
you
that
are
here
with
those
opportunities
so
so
go
get
busy
go,
learn
something
go
go,
make
an
impact
and
and
you'll
probably
get
called
on
more
than
you
care
to
be
at
some
point
for
sharing
your
knowledge,
sharing
your
experience,
good
any
questions
about
kubecon,
where
we
go
to
the
next
topic.
D
We'll
also
so
kubecon
eu
is
next
up
after
kubecon
china.
Just
for
those
that
are
curious
about,
like
you
know,
we've
a
member
of
this
community
has
spoken
at
every
kubecon.
D
There
has
been,
if
not
once
or
many
many
times
over
and
so
we'll
continue
to
so
we'll
be
talking
at
kubecon
eu
as
well,
we'll
have
even
more
to
share
with
with
folks
there
one
of
the
projects.
That's
I'm
going
to
jump
ahead
to
the
mesh
map
beta
sign
up.
That
needs
a
link,
but
we'll
be
no
doubt
talking
about
mesh
map
as
a
plug-in
to
mesherie
at
kubecon
eu.
D
D
More
people
signing
up
than
we
can
then
the
team
that's
working
on
that
can
handle.
Does
anybody
know
how
many
individual
users
there
have
been
of
mesherie
actually
cloud?
D
There's
there's
a
lot
and
we're
speaking
of
a
thousand.
Some
of
you
might
have
noticed
that
mescheri
hit
a
thousand
stars
this
week.
That's
fantastic!
D
It's
pretty
awesome.
I
remember
I
don't
know
how
many
times
we've
talked
about
stars
kind
of
a
funny
thing,
because
they
don't
really
mean
that
much
we're
already
2011,
that's
great,
but
even
though
they
don't
necessarily
mean
that
much
as
a
thing
as
like
a
like.
D
Usage
of
open
source
projects,
so
more
and
more
projects
have
a
bit
of
telemetry
or
analytics
built
in
to
help
attract
some
of
that,
but
sometimes
it's
hard
to
get
feedback.
Sometimes
a
user
might
try
one
of
the
projects
bump
into
an
issue
and
then
not
report
it,
and
actually
the
issue
goes
unknown.
D
You
know
which
is
bad
that
that's.
That
speaks
to
the
importance
of
some
things
that
piyush
in
particular,
has
done
around
testing.
So
unit
testing
has
been
a
focus
for
push
also
documenting
the
rest.
Api
for
meshri
has
been
a
focus.
We
have
an
ongoing
focus
every
two
weeks.
There's
a
measuring,
build
and
release
meeting
that
talks
about
or
where
we
get
into
testing
integration.
D
D
Those
tests
are
quite
important
so
that
we
can
well
not
have
that
bad
experience
that
I
was
talking
about
where
the
a
user
might
use
your
open
source
software,
not
report
the
bug,
and
you
wouldn't
know
about
it
so
having
those
tests
is
very
important.
A
couple
of
you
are
working
on
github
actions
at
the
moment
to
help
overcome
that
and
help
automate.
That
so
ashish
is,
I
think,
have
you
been
able
to
review
jubril's
recent
attempt
to
refactor
us
away
from
bash.
C
E
Yeah
I
I
saw
that,
but
I
was
I
was
doing
something
else,
so
I
postponed
that
procrastinated
that
so
I
would
look
into
it
by
by
tonight.
D
Sounds
good
sounds
good,
but
but
that
effort
in
general
that's
kind
of
another
example
of
what
I
meant
by
the
fact
that
there's
a
bunch
of
projects
around
here
that
that
project
has
to
deal
with
github
workflow
github
actions.
I
think
the
community
has
four
or
five
now
published,
so
there's
a
lot
going
on.
So
even
if
you
don't
know
golang
or
react
or
rust,
there's
lots
of
other
automation
to
do
other
other
scripting
languages.
D
D
Yeah
very
good
josh,
yep,
nice
yeah,
so
so
actually
josh
do
you
wanna
like
so
I'm
digressing
from
the
bullet
points
that
we
have
here,
because
because
some
of
your
new,
so
it's
good
to
give
you
a
general
introduction
and
for
the
some
of
the
rest
of
us,
it's
a
good
reminder
of
things
that
we
have
going
on,
that
we
need
to
pay
attention
to
and
do
and
actually
for
josh.
D
F
F
Josh,
take
it
away
yeah.
First
of
all,
I'm
adding
a
intra
page
navigation
in
in,
in
just
click
on
any
of
the
call-outs.
F
I'm
adding
a
intra
page
on
the
right
side.
There's
a
little
issue
on
the
safari
just
so
just
completing
it,
and
I'm
also
adding
call
out
to
which
link
to
the
handbook
in
the
community
page
and
the
newcomer
newcomers
page,
and
I
think,
on
learning
paths.
Also.
D
D
All
right
very
good,
any
any
initial
comments
from
some
of
the
things
the
topics
that
we
just
covered.
Anybody
have
questions
or
comments
on
those
things.
D
Okay,
just
as
I
go
to
hand
back
the
microphone
I'll
say
that
so
there's
a
collection
of
engineers
from
hpe
who
are
asking
for
kind
of
an
intro
to
measuring
and
overview
of
what
it
would
do.
We
need
to
get
some
time
on
the
calendar
or
find
a
spot
on
the
calendar
to
do
that,
we
will
likely
either
do
that
in
the
newcomers
call
on
thursday,
which
will
be
early
for
those
pacific.
Sorry
jared.
No,
we
won't
move
it.
D
Although
that's
a
good
request,
we
might
end
up
doing
it
in
the
front
in
this
community
called
next
friday,
which
will
be
a
little
bit
later
and
easier
for,
hopefully
everyone
to
join,
but
we'll
we'll
probably
wait
for
some
feedback
from
yogi
yogi
porlo
was
the
hpe
engineer
who
was
asking
there's
a
number
of
hp
engineers
in
the
slack
hpe
had
acquired
seitail
a
little
while
ago,
cyto
was
a
startup
behind
spiffy,
inspire
spiffy
and
spire
have
a
prominent
position
or
function
in
the
cloud
native
ecosystem,
but
in
service
meshes
in
specific.
D
D
That's
on
the
mesh,
a
unique
identifier
that
can
then
be
used
to
build
authorization
around
also
mutual
tls,
so
bi-directional
authentication
and
encryption
between
every
service
on
a
mesh,
so
the
spiffy
inspire
are
broadly
pervasively,
used
across
service
meshes,
there's
actually
in
the
service
mesh
landscape
on
layer,
five
dot,
io,
slash,
landscape,
you'll
find
a
list
of
all
the
service
meshes
that
are
using
spiffy
and
or
spire,
and
so
because
hpe
had
acquired
saitail
they
are.
D
They
continue
to
be
stewards
of
those
other
cncf
projects,
spiffy
inspire
and
they
want
to
put
in
the
time
to
help
integrate.
Spiffy
inspire
more,
are
directly
into
measuring
as
part
of
mescheri's
architecture
and
that's
something
for
the
maintainers
on
the
call
to
give
us
consideration
to
to
think
about
as
to
whether
or
not
that's
a
good
architectural
addition
come
talk
with
me.
If
I
can
help
explain
that
a
little
bit
more
as
to
why
that
might
make
sense.
D
D
D
D
D
A
Yep,
thank
you
lee.
Before
we
move
on
to
the
next
topic.
We
also
have
another
newcomer
on
the
call
deep.
Would
you
like
to
introduce
yourself.
B
Yeah
yeah,
so
hey
guys,
my
name
is
deep
and
I'm
in
my
free
final
year
of
my
btec
in
computer
science
and
as
I've
recently
started
my
journey
in
open
source.
I
want
to
contribute
in
different
projects
and
I'm
glad
that
I
found
layer
5.
I
hope
that
I'll
learn
a
lot
of
new
things
in
this
community
yeah.
That's
it.
A
Thank
you.
Thank
you
deep.
Thank
you
for
introducing
and
yep
welcome
to
the
lyrify
community.
There
are
a
lot
of
areas
that
you
can
contribute
to
like
yeah
as
you
explore
all
the
different
projects
in
the
airfare
that
we
do
in
the
life
community.
I'm
sure
you
will
find
areas
that
that
you
find
interesting
and
you
can
contribute
to
all
right.
So
moving
on
to
the
next
topic,
ashish
has
a
demo
on
displaying
patterns
validity
on
ui,
so
yeah.
A
E
Okay,
so
we
came
up
with
this,
you
can
go
at
this
link
and
we
discussed
this
issue.
I
think
in
there
was
a
filter
meeting,
so
the
small
feature
that
that
we
were
going
to
add
was
that
currently
we
do
not
okay.
E
So
this
is
the
patterns
ui
and
currently
we
do
not
differentiate
between
the
patterns
that
can
be
deployed
or
and
the
ones
that
cannot
be
deployed
and
the
reason
for
not
deploying
I'll
go
over
them
after
I
demo
the
thing
so
I've
added
this
column
just
for
the
demo
purposes.
This
should
they
should
not
be
taken
as
something
that
I'm
proposing
the
I'm
actually
proposing
something
else.
E
I'm
proposing
that
the
patterns
that
can
not
be
deployed
should
be
grayed
out
or
something,
but
so
because
I
didn't
want
to
go
back
and
flow
between
the
json.
So
I
just
added
this
field
so
and
the
this
status
and
not
supported
can
be
misleading,
so
I
will
give
a
little
bit
of
context
to
those
who
do
not
know
about
it.
E
So
meshi
has
a
concept
of
patents,
so
you
can
basically
deploy
your
service
meshes
as
patent
files
and
when
messy
and
messy
server,
basically
delegates
the
responsibility
of
the
patents
to
different
components
like
different
adapters
or
sometimes
to
itself,
and
for
delegating
that
responsibility.
The
adapter
should
come
up
and
register
their
components
and
register
their
capabilities
that,
yes,
I
am.
I
am
able
to
handle
this
particular
component
so
that
meshi
can
figure
out
okay.
So
if
I
get
this
pattern,
I
have
some
adapter.
E
I
have
something
that
has
registered
its
capability
to
actually
handle
that
pattern
and
at
the
back
end
we
have
a
mechanism
where
we
can
actually
dry
run
and
we
can
find
out
whether
we
can
actually,
whether
there
is
an
error
in
the
particular
whether
there
is
an
error
in
the
pattern
file
error,
as
in
that,
whether
the
components
that
have
been
described,
whether
the
things
that
have
been
described
in
the
patent
file
are,
can
they
be
basically
are
they
registered
in
the
database
or
not,
for
example,
for
example
like
if
I
have
gateway
or
sto
so
has
istio
registers.
E
Its
key
has
istio
registered
its
capability
to
handle
something
known
as
gateway,
so
all
of
this
is
with
message
server.
So
all
I
was
proposing
was
that,
just
before
we
send
these
patents
to
the
ui,
we
can
lazy.
Just
this
is
just
one
line
of
code.
We
can
easily
just
add
some
metadata
to
the
to
the
to
the
pages,
and
that
metadata
is
two
fields.
First,
is
that
we
can
change,
the
name
can
support
two.
It's
it
is
set
to
true.
E
If
there
is
some
error
on
the
basically,
if
there
are,
if
there
are
some,
if
there
are
some
error,
those
can
be
typos
or
something
or
maybe
there
is
no
component
that
can
handle
that
particular
pattern
and
the
the
along
with
this
particular
field.
We
also
have
the
error
messages
like,
for
example.
If
I
talk
about
what's
the
name
of
this,
let
me
reload
the
page
so
that
I'm
sure
that
I'm
looking
at
the
latest
json
okay.
E
So,
for
example,
I
have
this
particular
this
particular
this
particular
pattern
file,
and
here
the
error
messages
says
that
okay,
so
you
don't
have
the
containers
values
required.
E
So
you
don't
have
this
particular
array
set,
so
we
can
use
this
error
message
to
actually
have
some
kind
of
you
know:
question
mark
type
of
thing
around
the
around
the
patterns
that
we
have
grayed
out
and
the
user
can
click
on
those
and
realize
that,
oh,
I
haven't
added
the
containers
field
so,
for
example,
that
here
the
this
particular
this
particular
type
is
application.
Application
is
a
machine
native
component,
so
it
should
be
supported,
but
it's
why?
Why
is
it
not
supported
or
why
is
it
grayed
out?
E
So
the
they'll
figure
out
that
okay,
I
haven't
added
the
containers
feel
to
it.
So
now,
if
they'll
just
go
back
and
add
the
let's
add
the
containers
field,
and
when
I
update
the
pattern,
it
will
reload
and
it's
taking
time
to
reload,
so
it
will
reload
and
see
now
it's
so
it
says
supported,
so
it
can
be
not
non
grayed
out
or
greened
out
for
that
for
that
matter
and,
for
example,
for
another
for
another
app
example,
I
have.
I
have
not
deployed
this
to
adapter
right
now.
E
That's
why
these
two
components
are
not
supported,
but
if
I,
when,
when
I
deploy
the
issue
adapter
when
I
start
this,
your
after
the
hd
adapter
comes
up
and
registers
its
capabilities
with
the
message
server
that
I
can
handle
the
components
now.
So
when
I
I'll
reload
the
page
now,
the
the
another
patent
file
that
was
not
supported
will
be,
as
you
can
see,
it's
it's
supported
now,
so
supported,
not
supported.
We're
not
going
to
have
to
have
these
supported,
not
supported.
E
Whatever
was
proposing
that
we
can
have
graying
out
of
these
components
now.
The
second
part
is:
what
do
I
mean
by
basically
not
supported?
Not
supported
means
one
of
these
things.
First,
that
either
you
have
some
kind
of
typo.
The
second
is
that
there
is
no
component
that
has
registers
it's
registered
its
capabilities
for
that
particular
patent
file.
Now
there
is
one
caveat
to
this.
E
The
caveat
is
that
the
way
this
validation
works
is
that
you
are
not
guaranteed
even
after
it
says,
supported
or
it's
not
grayed
out,
you're
not
guaranteed
that
you'll
deploy
it
and
it
will
be
deployed
successfully
because
the
dry
run
is
limited
to
three
stages.
It
does
not
test
whether
you
are
actually
you
know
if
you,
basically,
if
you
deployed
it,
will
go
successfully
or
not
so
in
future.
E
At
the
adapter
side
we
have
either
helm,
chart
type
installations
or
manifest
type
installations.
In
both
of
these
cases,
we
can
dry
run
and
the
adapter
can
actually
send
back
the
message
that
whether
it
will
be
able
to
deploy
this
particular
patent
file
or
not.
E
So
this
all
the
components,
all
the
patent
files
that
will
not
be
grayed
out
will
actually
it
will
actually
mean
that
you
can
actually
go
ahead
and
deploy
it
and
the
adapter
will
successfully
deploy
it
and
if
the
adapter
is
down,
it
will
also
be
reflected
here
that
those
things
would
be
grayed
out,
because
if
the
adapter
is
down,
the
grpc
call
would
fail
so
that's
kind
of
an
extension
over
adding
another
stage
to
it.
E
Up
until
this
point
we
can
have,
I
mean
I
mean
three
stages
added
to
it,
so
that
you
we
can
detect
typo
and
the
we
can
detect
adapters,
those
adapters
which
have
not
registered
their
capability
to
support
a
patent
so
and
instead
of
these
this
particular
column.
We
can,
you
know,
gray
out
stuff
for
the
ui
people.
We
just
need
to
basically
take
care
of
this
particular
field.
If
this
is
set
to
false.
E
We
can
carry
out
that
thing
and
we
can
have
a
question
mark
with
this
error
message,
for
example,
invalid
workload
type,
so
that
user
can
click
on
their
message
and
can
actually
do
something
about
it.
So
yeah
anyone
has
any
feedback
to
it.
D
Yeah,
you
know
by
the
way,
as
as
everyone
digests
this,
if
you
couldn't
tell
like
the
the
thing
that
pushes
it
or
jeez
I'm
looking
at
piyusha's
announcement
and
sorry,
the
thing
that
ashish
is
solving
for
here
there
were
a
litany
of
considerations
that
he
went
through
and
in
my
mind
they
all
boil
down
to
two
letters.
D
We'll
just
I'm
not
gonna
answer
the
question
actually
I'll.
Just
you
guys
can
all
guess
that
what
those
two
letters
are
in
the
chat
and
if
and
whoever
gets
it
right,
I'm
going
to
send
a
I'll.
Send
you
one
of
these
shirts.
E
And
there
are
not
a
lot,
there's,
not
a
lot
of
code,
things
that
that's
required,
so
this
is
basically
in
future.
If
you
don't
want
that,
we
can
just
comment
out
this
line.
What
I
do
is
I'm
doing
it
as
suggested
completely
lazily,
so
there
is
no
change
in
database
or
anything
just
before
sending
that
response,
I
unmarshal
it.
E
Do
all
the
modifications
take
out
the
pattern
file,
do
all
the
checks
and
then
marshal
it
back
to
you
know
then
send
it
back
so
yeah,
there's
not
a
lot
of
code
change,
just
maybe
10
lines
of
code,
so
keep
in
mind.
D
E
D
Yeah
and
yeah,
and
in
part,
like
you,
know,
educate
me
here
or
educate
the
rest
of
us
here
like
is.
Is
it
true
that
one
of
the
challenges
that
people
will
have
with
a
go
routine
is
like
in
firing
them
all
off?
Is
that
you
know
you're,
not
necessarily
the
the
order
in
which
the
response?
You
know
those
individual
routines
complete?
You
know,
isn't
necessarily.
D
D
You
know
it
doesn't
matter
because
you're
just
because
you're
just
trying
to
show
people
as
quickly
as
possible
which
which
patterns
are
deployable
and
which
ones
are
not.
D
D
E
When
the
user
clicks
on
update
patent,
two
calls
are
made.
First
is
to
actually
post
the
new
pattern
and
a
subsequent
call
is
made
to
refetch
that
all
all
of
these,
or
basically
to
refresh
the
entire
patent
page
to
refresh
all
the
patterns
and
when
we'll
refresh
all
the
patterns,
those
dry
errands
would
be
made
again.
So
yeah.
That
would
be
completely
dynamic.
D
Nice
refactoring,
the
sorry
there's
a
new
there's,
a
service
mesh
evangelist
that
just
followed
the
mashery
twitter
handle,
which
is
I've,
never
seen
that
title
before
a
service
smash
evangelist.
That's
fantastic!
D
Random
side,
note
let
me
if
you
don't
mind,
let
me
go
back
and
ask
so
re
phrase,
reposition
that
question
and
ask
more
in
context
of
subscription
like
real
time
notifications
over
websockets
that
or
a
websocket
that
those
evaluations
were
performed
right
now.
It
sounds
like
all
of
those
evaluations
need
to
complete
before
the
paid
before
data
is
given
to
the
ui,
but
is
that
the
case?
Yes,
that
is
the
case.
D
Okay
is
that
do
you
know,
is
that
user
interface
based
on
using
graphql?
No,
it's
not
okay,
okay,
all
right!
So
there's
another
open
area
of
need,
so
I'm
not
sure
who's
sharing
the
screen,
but
if
we
can
go
back
to
what
ashish
was
talking
about.
D
Oh,
the
patterns
table
yeah,
okay
measure
you.
I
can
again
for
context
as
I
make
this
next
ask
or
suggest
that
if
anyone
is
into
react.js
and
javascript,
there
should
be.
You
can
reference
okay.
So
so
this
ui
needs
to
stop
using
rest
and
it
needs
to
start
using.
Graphql
meshware
has
two
apis:
a
rest-based
api
and
a
graphql-based
api
they're
they're.
Both
of
them
are
documented
in
mesh
redox.
D
D
That
this
ui
should
be
converted
to
use
so
anyway,
this
isn't
exactly
what
you
were
looking
to
talk
about
or
have
feedback
on,
but
since
no
one's
since
I
don't
have
to
send
it
a
sweatshirt
to
anyone
today.
I'll
say
that
in
part,
what
I'm
furthering
the
conversation
on
is
the
same
thing
that
I
think
that
you
were
more
or
less
taking
this
initiative,
for
which
is
ux.
D
This
is
about
ux.
This
is
this,
is
the
core
of
what
ashish
was
talking
about
from
a
number
of
different
perspectives
is
like
how
do
how
does
a
user,
who
might
have
any
number
of
patterns
service
mesh
patterns
that
they
want
to
deploy
at
the
time
at
which
they
submit
the
the
request
to
to
deploy
the
pattern
or
to
apply
the
pattern?
D
D
So
shouldn't
we
intimate
to
them
somehow
that
you
know
they
can
look
at
the
pattern.
They
can
edit
the
pattern,
but
if
they
want
to
click,
apply
they're
just
going
to
get
a
failure
back.
So
you
know
like
so
don't
even
do
that
to
start
with.
That's
in
part,
what
he's
going
for
he's?
Also
going
for
the
ability
to
for
users
to
like
to
look
at
a
given
pattern,
see
that
it's
supported,
but
also
maybe
before
they
actually
apply
it
and
affect
you,
know
pervasive.
D
You
know,
broadly
or
impactfully
affect
their
infrastructure,
potentially
in
a
negative
way,
like
maybe
they're,
applying
an
anti-pattern,
and
they
didn't
know
it.
It's
supported,
but
they
didn't
know
it
was
an
anti-pattern,
so
they
apply
it
and
then
they're
screwed.
After
that
they
have
to
go
undo
that
quickly.
D
D
G
D
Hey
jared,
if
I
could
quickly
ask
you
of
so
so
your
relative,
you
know
so
you've
only
been
around
the
community
for
so
long,
you're
still
digesting
the
projects
and
all
the
things
that
are
going
on
and
and
and
there's
there's
just
it's
a
big
old
fire
hose.
I
wanted
to
ask
of
part
of
the
area
the
aspects
that
intrigue
you
is
that
more
toward
front
end
or
back
more
toward
golang
or
more
toward
react.js
or
one
area
versus
the
next.
D
Very
good
guy
got
it
yeah
thanks
for
that.
This
is
good
to
have
in
the
back
of
my
mind
and
the
other
maintainers
minds,
as
as
they're
coming
across
issues
and
potentially
looking
for
opportunities
to
get
you
hooked,
you
know
find
a
foothold
for
you.
So.
D
D
I
mean
like
it's
not
a
trick
question
or
I'm
not
trying
to
be
mean
or
actually
trying
to
get
everyone
on
the
call
like
engaged
and
and
learning
something
kind
of
thinking
through
stuff,
like
the
smarter
that
you
will
get
you
all
are
the
better.
The
projects
will
be
the
better.
The
projects
are
the
smarter.
You
will
be
so
well
she's
just
like
facing
this
well
here.
Let
me
put
it
like
this.
Let
me
let
me
so:
can
we
go
back
to
ashish's
screen,
please
sorry
to
keep
pulling
us
back.
D
I
think
we
have
a
little
bit
of
time.
The
who
all
has
placed
an
online
order.
D
Okay,
okay,
let's
just
all
right,
you've
replaced
an
online
order
and
you
get
in
you
know
you
get
a
confirmation
email.
It
says
it
says,
thanks
for
your
order,
we'll
be
shipping
that
shortly,
I
suspect,
you've
gotten
out
you
probably
also
you
might
have
gotten
the
one
where
it
says
you
know.
Thank
you
kindly
for
your
order.
You
know
we
will
get
to
that
when
we
actually
have
that
in
stock.
D
Sorry,
that's
on
back
order,
we're
happy
to
take
your
money
and
hold
that
for
you,
though,
don't
worry
about
it.
It'll
be
sometime
between
now
and
next
year
that
we'll
have
you
know,
refresh
the
stock
and
then
we'll
send
it
to
you.
Anybody
gotten
one
kind
of
like
that.
G
D
There's
no
doubt
that
you
have
so
well
was
your
response,
something
like
again,
not
quite
a
circuit.
It
was
like
what
the
hell
like
it
was
like.
Hopefully
your
response
was
like:
okay,
hey,
that's
reasonable.
I
get
it.
You
know
so
either
like
it
was
black
friday
jared
was
over
there
with
his
trigger
finger.
He
got
the
deal
before
you
did
and
then,
but
you
know,
they're
going
to
get
it
back
filled
after
you
know
like
like
here's.
Here's
this
challenge
like
the
same
thing
that
would
cause
error.
D
Geez
I
apologise
is
dealing
with.
Is
it's
very
much
like
it's
just
an
ordering
system?
It's
you
know
what
it
boils
down
to
so
so
you
can
make
the
analogy
to
shopping
like
hey,
if
they're
out
of
order
like.
Why
did
you?
Let
me
order
the
thing
in
the
first
place
like
I
wouldn't
have
sent
in
the
money.
Now
I
want
a
refund,
so
I
can
go
order.
It
somewhere
else
to
get
it
because
I
needed
it
for
the
birthday
party
tomorrow
or
whatever
like.
D
Actually,
if
you
make
that
a
generic
statement,
a
generic
kind
of
a
problem,
it's
about
it
is
what's
the
right
word.
It's
about
resource
reservations,
it's
it's
about
resources
and
availability
and
reservations,
and
this
type
of
like
we're
just
continuously
all
of
us
that
are
working
on
software
working
in
one
of
like
10
different
computer
science
paradigms
just
in
like
the
10
000
different
flavors
of
the
same
darn
thing.
It's
like
there's
a
resource,
there's
a
finite
number
of
the
resource,
the
the
the
number
changes
and
and
that's
what
this
is.
D
If
you've
ever
received
an
if
you
are
on
your
wireless
network
right
now
and
you've
got
a
dynamically
assigned
ip
address.
Dhcp
an
ip
address,
address
management,
ipam
systems.
Dhcp
has
that
built
into
a
protocol
that
says
that
goes
through
a
series
of
very
talkative
communication
that
says:
hey,
router,
where's,
the
ip
where's,
the
ipam
system,
the
ip
address
management
system-
hey
it's
over
there.
Here's
the
address
to
that
great!
Oh,
it
turns
out.
That's
not
a
different
subnet.
Did
you
know
that
we'll
have
to
talk
about
dhcp?
D
If
anybody
knows
how
this
this
works,
but
like
anyway,
dhcp
doesn't
naturally
cross
subnet
boundaries,
so
it's
cloud,
it's
just
doesn't
work
with
cider.
Naturally,
you
have
to
have
a
router
send
along
a
helper
pass
along
an
address.
Anyway,
you
have
to
have
a
router
help.
You
get
dhcp
requests
across
domains,
and
so
the
point
is
I
want.
You
know
what
I
won't
take
you
through
the
whole
sequence
of
what
happens.
But
the
point
is
like
there's
a
finite
number
of
ip
addresses.
You
have
a
run
out
of
them.
D
You're
sitting
there
on
the
network.
You
get
this
self-assigned
169
dot,
something
something
reserved
address
where
you
can
just
basically
play
with
yourself.
You
can
like
it
that
only
feels
so
good
for
so
long
and
then
you
want
to
play
with
others,
and
so
you,
like
resources,
yeah
pardon
my
you
resource
reservation,
only
plays
or
is
an
important
thing
that
you
see
all
around
and
yeah.
You
can
and
that's
what
ashish
is
trying
to
do.
D
He's
basically
saying
there
could
be
any
number
of
reasons
right
now
he's
only
addressing
like
really
kind
of
one
reason,
but
there
could
be
any
number
of
reasons
why,
if
you
click
to
apply
that
pattern
like
we
already
know
it's
going
to
fail.
So
if
we
know
that,
for
we
know
that
there
might
be
a
number
of
different
reasons
to
the
extent
that
we
can
pre-qualify
the
request
before
someone
submits
it
great
like
we
know,
it's
gonna
fail.
A
All
right,
thank
you,
ashley,
sir,
so
we
we
have
some
areas
where
people
can
engage
and
where
we
need
active
participation
to
get
things
moving,
and
you
can
find
links
to
these
things
in
the
meeting
minute.
So
to
give
a
brief,
walkthrough
one
of
the
areas
that
we
are
currently
working
on
is
on
get
nighthawk
which,
which
is
which
is
basically
night
of
the
load
generator.
So
we
are
working
on
adaptive
load
control
in
measuring,
so
adding
that
feature.
A
So
this
this
is
one
area
that
people
can
jump
in
and
participate.
So
they're
like
there
are
some
active
works
going
around
in
this
area.
Another
area
is
in
measuring
ctl.
So
if
you
are
more
interested
in
golang
and
the
back
end
things
measure,
ctl
is
a
good
place
to
start
contributing
golang,
so
measuring
ct.
So
we
have
this.
This
ask
is
basically
to
improve
errors
in
measuring
ctl,
so
to
add
contextual
errors.
So
this
basically
boils
down
back
to
user
experience
improvements.
A
So
as
users
of
cli,
we
should
get
the
users
the
proper,
proper
user
experience.
So
some
some
work
has
been
going
around
in
this
area
to
improve
the
errors
that
that
are
thrown
to
the
user
so
that
they
are
more
contextual
and
people
can
actually
easily
debug
it
or
like
go
through
the
other
code
and
find
what
the
issue
is.
A
So
there
are
a
couple
of
open
issues
and
pull
requests
that
address
this
issue,
so
you
can
take
a
look
and
into
that
and,
like
start
contributing
like
make
more
contributions
around
areas
that
that
that
still
remain
unfixed,
and
we
also
have
another
area
in
the
measuring
ui.
So
measuring
c
ui
is
undergoing
some
restructuring,
basically
to
add
state
management
with
redux.
So
there
is
a
separate
feature:
branch
in
the
measuring
repo
called
measuring
ui
restructuring.
A
And
another
area
is
on
topics
so
so
basically,
this
would
revolve
around
creating
content
on
lab
the
layer,
5
website
and
we
have
a
channel
in
slack
called
block
kitchen.
So
if
you
are
interested
in
in
writing,
content
and
technical
writing
and
yeah.
A
That
sort
of
thing
you
can
jump
it
on
the
blog
kitchen
channel
in
slack,
and
we
can
help
you.
We
can
assign
you
some
issues
to
some
docs
or
some
some
content
to
write,
and
you
can
pick
it
up
all
right.
So
all
of
these,
the
links
to
all
of
these
is
in
the
meeting
minute,
so
you
feel
free
to
check
it
out.
A
Alright,
moving
on
to
our
final
topic
of
the
day,
saying
is:
would
you
like
to
talk
about
your
work
on
switching
the
community
forms
to
the
website.
C
A
So,
while
saying
fixes
this
microphone,
maybe
we
can
or
say
this
back,
which
is
fine,
now
yep
much
better.
E
So,
as
of
now,
our
community
is
using
a
google
based
newcomers
form
where
the
newcomers
are
asked
to
fill
this
form
and
they
give
they
get
the
access
to
advice
and
other
community
resources.
So
we
are,
we
have
been
planning
and
we
are
planning
to
shift
to
the
website
based
form
the
base
form
which
is
in
our
layer
5
website.
So
the
discussion
is
going
to
be
about
that
and
the
updates
on
that
is.
This
is
done
using
an
automated
system.
E
So,
as
of
now,
the
automated
system
is
working,
pretty
fine
and
also
the
data
is
being
updated
to
a
google
form,
as
well
as
the
mails
are
being
sent,
and
the
access
is
also
being
given
to
the
newcomer
there's
a
a
bit
a
small
bit
of
work
of
integration
from
the
current
form
to
this
web-based
form
remaining
other
than
that
everything
is
working.
Fine
and
the
tests
are
perfectly
working.
D
D
D
It's
a
little
more.
It
feels
a
little
more
classy
to
have
a
web-based
form.
D
You
get
access
to
resources
in
the
community,
so
you
get
on
a
mailing
list.
You
get
access
to
a
google
drive
that
has
lots
of
project
assets
and
design
specs
and
lots
of
things.
D
And
some
other
things
happen,
you
get
sent
messages
about
resources
and
what
to
do
next
and
some
steps,
I'm
hopefully
anyone
who's
joined
recently
like
well
like
when
you
join
the
slack
as
well.
You'll
get
a
private
message
that
says:
here's
a
bunch
of
here's
like
two
months
worth
of
things
to
go,
look
at,
or
at
least
a
couple
of
weeks
of
things,
and
so
that
automation
is
what's
being
shared,
and
that
and
zane
is
just
improving
that
so
zayn.
It
sounds
like
you're
ready
for
review.
D
D
D
D
You
were
wondering.
Why
do
you
want
to
know
my
shirt.
B
B
Did
you
end
up
with
access
to
the
community
community?
No,
no!
I
didn't
the
new
community
member
on
slack
or
on
the
website
on
the
google
form.
B
D
D
Horrific
all
right
vendo,
that's
all
from
me,
do
we
have
anything
else.