►
From YouTube: Infrastructure Group Conversation (Public Livestream)
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
B
Yes,
as
moving
into
more
practice
management
as
a
product
manager,
my
question
would
be
do:
is
it
possible
to
shadow
some
of
our
Suri's
on
how
they
work
on
github.com
on
kubernetes
and
who
would
be
the
right
people
to
share.
A
Okay,
thank
you
for
that
question.
So,
as
martin
points
out,
Jarvis,
kaarvok
and
Devon
are
probably
good
folks
to
shadow
and
I
think
this
would
actually
be
something
that
more
folks
could
take
advantage
of.
So
we'll
make
some
update
to
the
handbook
on
how
we
can
structure
this
a
little
better
than
just
sort
of
randomly
tagging
on
people
so
reach
out
to
those
folks
and
then
we'll
make
something
a
little
bit
more
structure
to
the
handbook.
C
Victor
I
can
also
invite
you
to
join
our
weekly
demos,
so
delivery
team
does
kubernetes
migration
demos
where
we
go
through
the
weekly
work
that
we
accomplished
and
we
kind
of
go
through
the
goals
that
we
set
go
through.
What
we
achieved
where
we
can
adapt.
You
can
find
the
invite
on
the
develop
team
meetings
calendar
there,
and
you
also
can
find
the
recordings
on
YouTube
on
filter
Channel.
We
have
a
playlist
specifically
related
to
the
kubernetes
migration.
You
can
check
that
out
as
well.
D
E
B
F
That's
it
yeah,
thanks
to
hear
you're
transitioning
now
to
fellow,
but
I
want
to
thank
you
for
the
improvements
you
and
the
rest
of
the
team
achieved
over
the
time.
You
were
in
the
leadership
role,
it's
it's
night
and
day
compared
to
where
we
were,
and
we're
now
and
now
able
to
to
have
availability
that
is
comparable
to
our
competition,
which
is
amazing,
I,
don't
think
we're
stopping
here.
A
G
A
So
we
discussed
this
at
the
clubhouse
meeting
this
week
and
the
feeling
is
this:
for
a
long
time,
we
didn't
do
any
real
management
of
our
costs.
So
a
year
ago
we
had
an
initiative
to
go
cut
some
costs
and
get
some
savings.
We
succeeded
in
that,
but
in
doing
so
we
took
our
sort
of
our
focus
off
availability,
so
we
went
through
another
slump
over
the
summer,
and
now
you
know
recently
we
picked
that
back
up,
but
what's
happening
is
we've
we've
hit
most
of
the
low-hanging
fruit
we've
dealt
with.
A
A
We
really
need
to
think
about
management.
Where
we
have
this
framework,
we
have
some
targets
that
we
want
to
hit
and
as
product
and
development
and
infrastructure
make
decisions,
we're
making
it
we're
making
those
decisions
within
that
framework.
So
it's
I
think
it's
an
evolution
of.
We
always
knew
we
needed
to
care
about
the
money,
but
we
did
some
trade-offs.
I
I
think
you
put
I
was
just
gonna,
say
I.
Think
Davis
put
it
really
well,
it's
it's
not
so
much
that
these
cost-saving
well
cost-saving
things
are
no
longer
something
that
we
can
tackle.
It's
just
as
was
mentioned
additionally
like
we
have
to
build
a
framework
or
a
roadmap
around
how
we
want
to
tackle
these
things,
to
look
at
them,
not
only
through
the
lens
of
like
okay.
I
Let's
we
have
a
whole
bunch
of
things
that
we
can
do
to
achieve
some
of
these
savings
in
the
in
the
near
term,
but
we
also
have
to
look
at
the
like
a
more
holistic
roadmap
of
like
how
do
we
do
things,
including
like
talking
to
vendors,
to
find
additional
discounts?
How
do
we
enable
essentially
Davis
to
do
these
magic
of
like
analyzing
data
and
fighting
some
technical
solutions
that
we
can
tackle?
H
B
In
the
past
weeks,
I
was
thinking
about
having
at
least
some
of
the
review
apps
like
github.com
that
are
status,
apps
KNAT,
which
is
kind
of
sketches.
You
know
on
top
of
kubernetes
I,
don't
know
if
we
have
it
today
as
a
feature
or
not
I,
don't
know
how
closely
running
this
review
apps
are,
but
if
you
can
do
you
have
any
ideas
around
that
like
do
we
have
it
already?
Is
it
costly
or
not,
I.
C
Ups
obvious,
because
money
like
the
current
ones
we
have
but
I,
don't
know
whether
our
review
apps
actually
support
chemica
native
out
of
the
box.
I
know
that
the
way
the
review
apps
are
configured
right
now,
basically
is
connecting
to
the
community
cluster
directly
rather
than
through
K
Natives.
So
I'm
I,
don't
I,
don't
know
whether
the
product
supports
it.
H
Yeah
Lincoln
issue
here
there's
an
issue
about
review
app
costs,
but
overall
it's
not
a
huge
person,
another
CI
cost.
So
it's
not
a
big
concern
today,
but
I'll
pass
the
issue
in
here
for
you
thank.
F
Call
slight
fees,
a
safety
over
velocity
I
looked
at
at
the
issue,
and
it
seems
to
be
that
we
need
correct
pre
and
post
light
text,
which
is
probably
kind
of
Acts
are
work.
If
you
want
to
implement
the
change,
I
wonder
whether
it
impacts
velocity
assertive
in
my
view
of
velocity,
is
kind
of
the
volume
of
changes
you
can
do
in
a
set
time
period
but
saying
deployments
per
month.
F
I
think
if
you
have
good
pre-flight
and
post-flight
checks,
you're
gonna
increase
your
velocity
because
you're
less
likely
to
end
up
in
some
unknown
state
where
there's
a
big
problem,
and
now
everything
is
stopped
for
the
whole
training
stuff
for
a
few
days,
as
you
figure
out
what
to
what
to
do
so,
I
wonder:
I,
wonder
why
why
it
was
set
up
as
a
as
a
conflict?
Maybe
it's
safety
over
time
spent
deploying
a
single
change
or
something
like
that.
I
think.
A
The
messaging
and
I
can
certainly
update
that
to
reflect.
That
is
think
safety
first.
So,
whenever
we're
getting
ready
to
make
a
change,
the
cost
of
that
change
may
seem
simple
at
first,
but
if
the
change
goes
sideways
and
off
the
rails,
then
that
cost
is
going
to
be
a
star
could
be
astronomical,
so
sometimes
we're
making,
as
we
think
through
changes.
A
We
are
making
decisions,
not
necessarily
building
these
assumptions
into
the
automations,
and
so
the
idea
is
to
make
sure
the
automation
understands.
This
is
my
initial
state,
and
this
is
the
state
on
with
under
which
I've
been
designed
to
operate,
and
if
none
of
these
conditions
are
met
and
the
automations
are
actually
protect
the
environment
before
it
tries
to
to
execute
the
change,
and
you
know
we
found
instances
where
some
assumption
we
made
was
incorrect
and
so
the
animation
sort
of
proceeded
and
then
that
caused
some
downtime.
A
So
the
idea
behind
this
is
it's
not
about
having
a
checklist
that
you
follow
on
on
a
change
issue,
but
it's
about
building
this
into
the
automation
and
over
Tonja,
and
that
building
this
library
of
we
changed
checks
and
verification
checks.
And
then
you
just
pick
what
you
need.
So,
when
you're
building
that
library
there's
probably
a
little
bit
of
a
slowdown,
because
you
know
you
are
investing
the
time
in
creating
that
those
checks.
A
But
then
over
time
not
only
is
your
speed
faster,
but
your
confidence
that
what
you're
doing
is
correct
and
that
the
automation
is
going
to
the
right
thing
when
it's
not
is
is
very
high.
So
yes
in
essence,
you
spend
less
time
sort
of
double-checking
yourself
so
yeah.
We
can
change
the
warning
on
that
to
reflect
the
fact
that
is,
it's
mostly
about
make
sure
that
we're
building
those
checks,
because
over
time
this
won't
be
a
question
right.
Well,
if
one
over
the
other
we're
building
text
that
have
safety
safety
measures
built
in
great.
F
I
think
safety
first
is
a
great
way
to
say
it.
I'm
I'll
propose
another.
One
measure
twice
cut
once
it's
a
it's
a
coma
saying
in
in
in
English,
but
I
like
it,
because
normally
it
means
that
you
measure
twice
before
you
make
a
cut
in
this
case,
we're
doing
something
else
who
have
pre-flight
and
post-flight
checks.
So
we
measure
twice
one
up
front
and
one
after
the
factor,
it's
kind
of
twisting
and
existing
saying.
So
that's
that
there
will
be
my
proposal
for
how
to
say
it.
A
The
general
idea,
this
all
came
because
I
run
into
this
article
about
Avalon
tea
safety
on
the
New
York
Times,
and
they
talk
about
human
factors
and
then
I
started
thinking
about
okay.
When
I'm
making
a
change
there
is
those
factors
are
affecting
the
way
I'm
making
decisions.
So
how
do
you
counteract
those
in
an
Atalanta
that's
actually
hard,
but
in
our
environment,
where
we
can
essentially
control
everything?
It
should
be
far
easier.
So
yeah
I'll
update
the
language
to
reflect
that
Thanks.
F
F
J
C
A
F
J
A
So
I
don't
see
any
other
questions
or
comments.
You've.
No
one
wants
to
have
anything
else.
Then
we'll
wrap
up
okay,
Craig
Isis,
making
a
complimentary
understands
background
so
anyway.
Thank
you
all
and
again,
thank
you.
It's
been
a
it's
been
a
fun
year.
I
have
doing
this,
then
I
look
forward
to
to
the
next
face,
so
the
next
one.
It
should
be
actually
our
new
VP
I
doing
this.
So
thanks
again,.