►
From YouTube: Release Management Think Big #2
Description
Welcome to our second Release Management Think Big session. We dove into some thoughts around how to improve Environments and where that fits in the plan for Release. See our direction at https://about.gitlab.com/direction/cicd/#release.
A
A
Okay,
I'm
gonna
assume.
Yes,
sorry,
you
all
have
a
link
to
this.
It's
our
think
big
release
management
agenda
and
we're
hoping
to
bring
in
a
little
bit
more
of
the
vision,
product
roadmap
conversations
to
our
technical
teammates,
but
also
improve
partnership
and
forward-looking
vision
between
design,
UX,
research
and
engineering.
So
that's
what
these
think
big
sessions
are
for.
Last
meeting,
we
talked
a
little
bit
about
some
of
the
ten
years
three
years,
one
year
horizon
where
we
want
release
management
to
go
a
little
bit
about
target
market
and
I.
A
Think
from
that
conversation,
I
distilled
two
Direction
items
that
I'd
like
us
to
start
having
conversations
about
and
potentially
doing
some
tech,
debt
reviews
on
or
thinking
what
is
what
is
truly,
the
minimal
viable
thing
that
we
can
do
to
make
this
possible
and
get
lab.
So
those
two
items
are
building
up
our
secrets,
management
philosophy
with
hashey,
core
vault
and
then
the
second
one
is
improving
environment,
variable
management
and
really
environment
management
in
general,
so
I'm
happy
to
start
with
environments,
because
I
feel
like
this
is
a
little
bit
more
in
our
own
control.
A
On
this
it's
within
get
lab.
We
don't
worry
about
like
a
partner
or
building
an
integration
with
an
external
party,
so
I
think
from
my
perspective,
I
have
a
lot
of
issues
and
bugs
in
the
backlog,
regarding
the
experience
of
environments
and
I
also
have
a
lot
of
issues
that
were
created
by
mark
around
how
environments
are
expected
to
you
behave
based
off
of
how
our
competitors
are
using
variables.
I
would
love
to
hear
your
guys's
perspectives
and
what
you
think
we
should
do
as
far
as
kicking
off
a
experience.
Research
for
environments,
management.
B
A
A
C
C
A
D
C
I
know
in
particular
like
just
with
my
own
experience
that
the
UI
of
that
section
is
really
kind
of
clunky
like
it's
really.
It
doesn't
work
well
on
small
screens
and
it's
like
a
lot
of
very
common
use
cases
in
putting
a
really
big
string
into
the
value,
the
variable
value
field,
and
it
doesn't
really
expand
nicely.
So
a
lot
of
little
UX
improvements
that
could
make
that
that
grid
a
lot
nicer
to
use
just
in
its
current
state.
You
might
at
first
thought
on
that.
B
Yeah
there's
overlaps
with
the
the
item
we
were
discussing
yesterday
during
the
p.m.
UX.
Seeing
that
verify
already
started
working
on
this
I
think
eight
months
ago,
I
didn't
have
a
proposal
for
improving
the
UI
UX
of
that
page
because
as
far
as
I
know
and
as
far
as
what
I
have
worked,
we
didn't
really
look
into
what's
next,
so
we
just
added
the
title
bubbles
and-
and
that
was
it
but
like
we
know
now,
there
are
gonna
continuously
iterate
on
this
page.
A
So
I
just
linked
the
issue
there,
the
one
that
you're
referencing,
so
it's
the
CI
CD
environment,
variable
scope
to
specific
environment
pipeline
job.
This
issue,
title
doesn't
really
reflect
I
think
what's
happening
inside
this
issue,
but
just
like
what
hi
on
I
was
saying.
There's
this
greater
UX
proposal
on
the
verify
team
side,
but
I'm
trying
to
understand
is
the
verify
team
going
to
execute
against
it
or
we're
going
to
execute
against
it
and
so
I'm
waiting
for
if
verify
is
going
to
be
implementing
it.
So
I'll
confirm
on
that.
A
B
Think
some
of
them,
let
me
see
which
issues
can
you
go
up
to
so
I
can
see
the
tile.
B
Yeah
you
mean
the
proposal
from
verify
right,
yeah,
that's
this
one
yeah
this
one
I
think
you
solve
a
lot
of
our
usability
problems,
but
I,
don't
really
I'm
not
familiar
with
the
full
scope
for
this.
This
feature
so
I'll
say
maybe
it's
best
to
to
into
what
the
proposal
have
now
and
what
are
the
pieces
are
missing.
A
A
For
us
to
to
really
gauge,
if
that
proposal
will
even
satisfy
everything,
I'm
worried
that
we're
only
taking
what's
been
given
to
us
from
some
of
some
other
team
and
then,
if
we
start
implementing
it,
it
might
not
actually
address
all
the
things
that
we're
needing
to
execute
against
so
I
think
about
environment
variables.
It
touches
a
lot
of
different
personas.
It
goes
throughout
the
entire
CI
CD
process.
It's
not
just
release
so
I'm
curious
on
how
we
think
we
can
approach
that
better.
B
E
A
E
But
here's
the
thing,
if
you're
an
assumption
and
that
assumption
happens
to
be
wrong.
Your
validate
on
something
that's
yeah,
kosher,
yeah
I
mean
at
least
that
gives
us
a
place
to
start
so
I
feel
like
what
we
had
to
actually
do.
Is
you
need
to
go
validate
those
assumptions
to
make
sure
that
we're
in
the
right
direction
or
look
at
Dimitri's
work
to
see
if
he
did
talk
to
people?
If
you
talk
to
people
I'm
good,
if
he
didn't,
then
I
have
a
little
bit
of
concern
about
the
validity
of
it.
I
do.
A
D
A
Fair
that
kind
of
goes
to
it
when
Nathan
was
saying
which
is
like
the
UI
is
clunky
and
long
strings
aren't
easy
to
implement
like
if
we're
just
trying
to
use
a
field
as
it
would
be
intended
for
entering
in
a
variable,
regardless
of
whether
or
not
that
variable
is
an
assumed
length
or
an
assumed
size
it
should.
You
should
be
able
to
enter
in
I.
Think
it's
not
any
type
of
variable.
Yeah.
D
E
And
it
might
help
to
think
about
the
jobs
to
be
done
as
something
they
need
to
accomplish,
not
necessarily
using
our
UI
at
all.
It's
here's.
My
task
I
need
to
do
this
thing
and
I
need
to
do
it,
because
it's
going
to
make
my
life
easier
or
whatever,
and
then
we
need
to
take
that
as
an
abstract
and
then
look
to
see.
How
can
our
systems
support
that
almost
UI
agnostic,
like
not
to
think
about
the
UI
immediately.
D
A
Example
in
one
one,
one
use
case
of
our
variables
are
to
use
that
to
contain
secrets,
and
that
is
actually
a
problem
because
it's
perceived
in
the
market
as
not
being
secure.
So
what
we
want
to
solve
the
job
to
be
done
is
as
a
dev
off
personnel
I
want
to
make
sure
that
my
CI
and
CD
are
secure
as
possible.
A
E
Think
I
would
focus
more
on
what
are
they
trying
to
do?
What
what
is
the
goal?
What's
our
job
as
it
relates
to
the
environments,
what
what
do
they
use
it
for
what
can't
they
use
it
for
and
that
could
follow
that
can
where
that's,
where
the
I
have
trouble
with
this
I.
Don't
like
this!
This
doesn't
do
this
kind
of
falls
in
to
that's
what
I
would
focus
on
to
try
to
get
that
and
then
out
of
that,
your
jobs
to
be
done
will
come
right
out
of
that
right.
A
A
D
C
C
A
So
one
problem
that
I'm
hearing
when
I
talk
to
customers
is
that
I
don't
want
to
deploy
my
kubernetes
clusters
using
gitlab
because
of
this
perception
that
get
lot
of
isn't
secure
and
a
big
part
of
that
is
the
environment.
Variable
management
and
I
think
I'll
just
open
up
an
issue
for
us
to
kind
of
talk
about
that.
A
little
bit
more
openly
is
there's
going
to
be
a
couple
ways
that
we
can
approach
solving
that
again.
A
Partnership
with
fault
may
help
that,
but
then
there's
this
whole
host
of
customers
that
are
using
secrets
within
get
lab
housed
in
environment
variables
that
aren't
worried
about
the
security
issue.
So
I
wouldn't
want
to
position
this,
as
you
have
to
migrate
from
environment
variables,
because
Ghaleb
is
insecure,
so
I
want
us
to
kind
of
think
about.
A
D
B
A
From
a
regulated
industry
view,
it's
not
as
refined
in
how
you
would
use
like,
for
example,
the
t-mobile
gave
I
think
we're
recording
this.
Damn
it
sorry,
a
customer
referenced
that
they
can't
use,
get
lab
to
deploy
their
super
secure,
kubernetes
clusters,
because
they
they
have
to
use
the
git
lab
root
user
to
execute
all
the
commands,
and
in
order
to
do
that,
that
permission
level
would
be
exposed
to
everybody
and
the
company
that
would
be
deploying
and
that
permission
level
is
not
acceptable
for
them.
So
they
don't
deploy
using
Gila.
A
A
D
Does
yeah
and
I
think
that's
the
I
think
that
kind
of
drives
a
kind
of
questions
we
need
to
start
thinking
about
right
to
me.
That
is
the
like.
What
is
the
real
underlying
problem
here,
because
yeah
we
can
clean
up
the
UI,
but
that
last
one
about
you
know
you
just
might
want
them
gone
that
that
might
be
a
different
approach.
A
D
D
C
A
This
is
actually
like
way
upstream
of
any
roadmap
where
I
would
see.
This
coming
to
fruition
would
probably
be
at
the
end
of
2020
and
to
2021,
so
we're
like
starting
super
far
and
advanced,
and
then
we'd
schedule
like
quick,
win
issues
that
we
know,
we
need
to
fix
to
improve
immediate
yeah,
immediate
needs
of
our
of
our
customers,
but
to
my
to
my
view,
I
think
environments
aren't
being
leveraged
as
powerfully
as
they
can
be
and
I
think
also.
A
We
could
build
a
really
strong
executive
dashboard
using
something
like
environments,
and
this
executive
dashboard
would
be
something
very
sexy
to
sell
in
our
market
and
it
really
opened
up
a
great
potential
for
us
so
going
into
this
I
want
us
to
think
about
not
only
the
user
experience
of
interacting
with
with
variables
with
secrets
with
environments.
But
how
is
somebody
who
is
purchasing
gitlab
that
wants
to
verify
that
we're
going
back
to
their
ROI
as
a
business
and
that
we're
contributing
to
the
success
of
their
organization
with
something
like
environments?
A
So
we
have
to
maybe
address
the
user
concerns
with
the
UX,
verify
18
proposal
and
then
kind
of
rethink
about
how
does
the
environments
and
operations
dashboard
play
with
the
environments
layout?
And
what
can
we
do
with
environment
variables
just
span
the
entire
CI
CD
workspace,
so
that
when
sales
goes
to
market
they're
able
to
come
in
with
something
that's
really
compelling
and
differentiating,
but
I
agree.
I
can
position
this
a
little
bit
more
clearly
against
the
roadmap
and
vision.
So
I'll
do
that
better.
Next
time,.
E
E
Yeah,
yeah
and
I
think
anything
else
like
links
to
anything
else.
That
would
help
like
or
like
to
tell
us
who
are
the
people
who
would
be
working
with
us
on
these
things
because
I
know
Dimitri.
But
you
know,
as
the
team
grows,
it's
gonna
be
harder
and
harder
to
figure
out
who
are
the
people
who
we
need
to
work
with,
so
that
might
be
helpful
to
another
nice
out.
That
was
great.
C
A
I'll
asked
him
on
how
he
sets
up
his
meetings
too,
since
he's
the
I
think
you're
one
of
the
other
pans
that
are
running
this
his
when
I
read
his
agendas,
they're
always
so
like
rich
with
actionable
content,
and
so
I'm
wondering
what
he's
doing
to
to
get
that
kind
of
result,
not
that
ours,
isn't
it's
just
whenever
outside
looking
in
it.
It's
just
look,
look
different,
all
right,
perfect!