►
From YouTube: High priority merge request discussion
Description
Discuss adding a high priority merge request process (to primarily give us a process to meet a critical customer need)
A
A
B
It
sounds
good,
so
the
problem
is
I
understand.
It
is
that
we
currently
rely
on
a
regular
review
process,
even
regardless
of
how
much
of
a
priority
there
is
on
on
whatever
the
feature
or
the
bad
fix
or
whatever
it
is
to
deliver
results.
So
this
is
not
the
first
time
we
had
a
situation
like
this,
and
we've
always
had
a
bit
of
an
issue
on.
How
do
we
rush
things
under
the
under
the
deadlines
and
what
you're
suggesting
with
with
the
process
here,
is
to
create
a
quick
or.
B
Like
a
more
efficient
way
of
handling
a
merge
request
as
a
response
to
a
high
criticality,
merge
request
requested
by
a
customer,
what
usually
ends
up
happening?
Is
that
like
whenever
we
do
one
of
these,
and
whenever
there
is
a
retro,
there
is
usually
a
planning
issue
that
caused
the
reason.
That
is
the
main
reason
for
a
high
priority,
merge
request.
So
usually
it
goes
from
the
issue
not
being
scheduled
or
issue,
not
being
scoped
well
enough
or
when
we
started
implementing.
B
We
ran
into
more
odd
roadblocks
which
expanded
the
further,
which
meant
that
we
cannot
finish
an
item
within
the
timeframe
that
we
promised
so
I'm
gonna
use
the
one
that's
kind
of
prompted
one
of
these.
It
seems
that
to
me,
at
least
from
what
I
read
is
that
the
original
issue
was
moved
from
a
milestone
to
milestone.
Then
the
customer
came
in
and
said.
Well,
it
needs
to
happen
in
this
milestone
or
we're
out,
and
then,
when
we
put
some
efforts
to
actually
get
the
feature
result
or
created.
B
B
We
recognized
fairly
quickly
that
there
is
like
a
smaller
item
that
we
could
ship
to
satisfy
the
need
for
for
this
specific
customer
unique.
That
is
not
the
point
where
we
need
to
look
for
a
solution
to
me.
The
point
is
earlier
in
the
cycle,
whether
why
did
we
share
jewel
it
too
late?
Why
did
we
push
it
from
one
to
another
milestone?
B
Why
didn't
we
scope
down
the
feature
immediately
like?
Why
didn't
we
recognize
exactly
what
would
be
the
smallest
possible
thing
that
we
need
to
share
and
then
only
then
look
into
what
was
those
like
final
couple
of
work
days
like
whether
we
would
even
come
to
that
position
if
we
were
to
look
at
it?
So
sorry,
I
just
wanted
to
finish
like
very
quickly.
B
So
that's
why,
for
me
the
the
process
says
it
is
written
right
now
is
so
large
and
complex,
and
it
is
so
easy
to
abuse
it
and
push
the
problem
on
the
end
of
the
pipeline
instead
of
to
resolve
it
somewhere
at
the
beginning
or
the
middle
of
the
pipeline,
and
that's
why
I'm
approaching
this
like
I,
don't
know
whether
it's
very
strongly
but
I,
don't
believe
this
is
a
solution
that
we
should
be
trying
out.
First,
no.
A
That
was,
it
took
that
amount
of
time
to
do
the
work,
and
it's
not
that
the
developer
working
on
it
was,
you
know,
inefficient
or
anything
it
was
just
it
was.
The
scope
was
too
large.
The
scope
initially
was,
that
is
what
product
management
said
was
needed
to
go,
live
on
the
feature
that
that
that
that
dependent,
that
that
was
a
critical
portion
of-
and
it's
it's
a
one
time
migration
job.
So
we
had
trouble
for
you,
it
more
iteratively.
A
If
it's
gonna
run
once
per
project,
both
in
get
lab,
calm
and
self,
hosted
customers,
and
then
it's
never
run
again
and
we
couldn't
release
it
iteratively
and
have
it
run
in
production,
because
it
would
have
only
done
part
of
the
job
it.
As
you
say,
this
was
the
team
discussed.
Can
we
make
this
in
actually
I
think
it
was
the
original
author
that
came
up
with
the
idea
that
say
I
think
perhaps
they
said
the
scope?
Is
this
big?
A
You
know
it
was
moving
all
specific,
all
vulnerability,
data
historically
and
that's
a
particular
concern
from
the
maintainer
z--
as
it
should
be
because
of
the
high
load
it
can
put
on
the
database
and
causing
potentially
instability
for
gitlab,
comm
and
large
self
hosted
customers.
So
that
was
very
valid.
Concern,
said:
okay,
I
think,
with
the
other
times,
I
think
we
can
actually
do
a
subset
of
the
work.
Previously
proud
management
was
not
in
favor
of
that
as
a
minimum,
viable
change
and
with
where
we
were
at
we
said.
A
Okay,
Park
is
like
some
customers
would
be
unhappy
with
not
migrating
all
the
data,
but
to
try
to
get
this
one
over
the
finish
line
anytime,
whether
it
was
the
1210
release
or
future
releases.
He
we
discussed
the
product
manager
and
they
were
good
with
reducing
the
scope
on
what
the
product
management
requirements
were,
which
lowered
the
load
which
made
the
merge
request.
A
You
know
being
20/20
when,
when
we
saw
that,
mr,
when
the
when
going
back
and
forth
many
many
times
so
the
development
when
the
developer
was
working
on
it
for
a
very
long
time,
that's
not
necessarily
a
bad
thing,
but
it's
kind
of
a
yellow
flag
to
look
into
so.
And
then,
when
the
back
and
forth
between
the
the
author
and
the
maintainer,
they
were
different
main
containers
involved
over
time
based
on
the
different
domains
and
people
vacation
time
etc.
Is
we
should
we
should
have
earlier
said?
You
know
what?
A
But
that
saying
hey
when
we
see
a
trend
like
X
or
a
trend
like
Y,
take
a
step
back
and
we
want
so
that
will
help
to
avoid
the
specific
situation.
Some
of
the
specific
situations
that
occurred
with
this
merge
request.
The
concern
is
that,
even
if
we
do
that,
this
kind
of
situation
may
happen
again
where.
B
A
A
We
could
we
can
the
the
concern
about
that
is
when
we
discussed
it
in
in
Christopher's
staff
meeting
amongst
those
on
his
report
to
him,
and
also
there
were
some
comments
in
the
issue
from
some
others
saying
this
kind
of
thing
could
happen
again,
and
so
the
specific
causes
have
been
happening
in
this
case.
That's
fine.
The
idea
for
this
high
priority
merge
request.
Let's
say
we
avoid.
We
do
great
changes
to
avoid
those
specific
things
happening
ever
again,
which
would
be
great.
A
You
know
that
that's
it's
a
big
goal
to
have,
but
you
know
we
can
aspire
to
do
that.
There
could
be
other
things
where
we
don't
have
a
process
to
or
the
process
is
not
effective
to
detect
things
that
are
going
in
this
direction
where
we
might
want
or
need
a
high
priority,
merge
request
process
and
then
we're
going
to
be
at
the.
A
If,
if
that
happens
and
there's
concerns
that
it
will
based
on
other
things
that
have
happened
in
the
past
unrelated
to
the
you
know,
the
the
vulnerability
migration
one
and
there's
concerns
about
it
in
the
future,
then
we'll
be
at
the
state
of
some
theoretically,
we
we
have
a
hype.
We
have
a
merger
quest.
We
really
want
to
get
into
a
certain
release,
and
things
happen
that
made
it
to
that.
A
We
can
either,
for
example,
not
meet
that
rep
that
that
financial
goal,
or
we
can
ad
hoc,
try
to
run
a
high
priority,
merge
request
process
without
having
something
documented
in
advance,
and
the
concern
is
that
if
we
don't
have
something
that
we
talk
about
and
debate
and
has
been
really
good
debate
on
this
one
in
advance,
that
will
be
it
will
potentially
and
I'm
gonna
work
my
hardest
to
make
sure
anything
on
my
team's
never
trigger
it.
Right.
A
I
can't
guarantee
that
but
I'm
definitely
I
so
very
important,
but
without
a
documented
process
to
follow
that
we
talk
about
and
kind
of,
make
in
good
shape
in
advance,
we'll
be
in
this
ad-hoc
mode
again
in
the
future.
Potentially
and
it's
it
may
have
the
same
impact
on
the
people
causing.
Do
we
do
the
financial
goal,
or
do
we?
You
know
have
do
this
thing,
ad
hoc,
which
creates
significant
stress
for
everybody
in
bulk,
that
that's
the
that's
the
reason
for
this
merger
I.
B
B
Whether
we
can
scope
this
down,
whether
like
understanding
the
actual
business
needs
more
and
then
on
the
other
side.
What
I
see
is
when
some
of
these
decisions
get
made
by
people
with
that
come
from
a
higher
level
position,
it's
really
hard
to
push
back
on,
but
it
is
almost
impossible
to
push
back
on,
which
then
creates
another
level
of
stress
of
that
at
that
level,
where,
if
I
feel
it's
delivering
this,
that
might
reflect
bad
on
me.
B
So
I'm
just
gonna
do
whatever
it
takes
to
not
fail
on
this,
which
means
lower
security,
lower
availability,
lower
everything
that
you
are
trying
to
protect
against
and
then,
ultimately,
you
pay
the
cost
somewhere
down
the
somewhere
further
down
the
process,
because
now
the
you
got
it
out
of
your
queue
right,
like
you
merged
it.
Now
it's
releases
problem
whether
this
is
going
to
be
deployed
or
now
that
I
release
deployed
it.
B
A
Processing's
would
a
process
this
process
won't
like
it
could
have
did
have
and
could
in
the
few
which
we
did
ad
hoc
and
could
have
negative
unintended
side
effects
and,
like
you
mentioned
like
like,
like
pressure,
which
we
definitely
intent,
did
not
intend
with
this
ones.
But
pressure
to
approve
before
things
were
ready
from
a
quality
perspective
which
we
definitely
would
did
not
want
to
see
happen
or
pressure
to
prioritize
work
over
family
and
and
and
friends
and
health.
A
You
know,
especially
in
times
of
covet
19,
we
tried
to
make
it
clear
to
do
not
sacrifice
those
things,
but
just
because
the
prop
where
it
was
coming
from
people
may
have
felt
conflicted
about,
and
that
is
definitely
was
not
was
intended.
We
didn't
want
to
sacrifice
any
of
those
things
so
I,
don't
you
know
it
can
have
negative
unintended
consequences
that
could
create
a
bigger
problem
than
the
one
we're
trying
to
solve
in
the
first
place,
exactly.
C
B
C
C
What
I
haven't
seen
is
like
a
proposal
around
what
I'll
call
the
screwdriver
process,
which,
whatever
it
is,
we
think
is
the
right
thing.
The
other
thing
is
is
I,
think
it's,
it's
very
fair
feedback
to
say
that
right
now
we,
the
the
person
who
determines
this,
is
larger.
The
manager
working
the
situation.
C
If
that's
the
primary
concern
folks
have,
we
could
make
the
DRI
for
using
this
capability
much
higher
right
like
it
could
be
myself
or
we
could
even
make
it
Eric
as
an
example
to
say
you
know
this
is
this
is
something
we've
you
that
we're
gonna
use
sparingly
and
it
kind
of
kind
of
might
mirror
like
what
we
do
for
headcount
resets,
which
is
we
have
a
process
there,
but
when
we
use
it,
we
really
don't
like
using
it.
The
other
you
know
the
third
optional
throughout
here
is
you
know
we
could
wait
and
see.
C
This
happens
again.
I
think
this
is
a
percentage.
This
game,
no
matter
how
good
we
get
at
planning-
oh
I,
had
a
good
we
get
a
prioritization
reality
is,
is
we're
gonna?
Have
this
situation
come
up
again,
just
strictly
because
of
the
volume
we're
doing,
and
you
just
gotta
kind
of,
like
you
kind
of
have
to
realize
that
that
you
know
like
I,
agree
that
we
should
push
back
as
much
as
possible.
We
should
are
CAA.
C
B
B
So
for
me,
while
this
is
important,
I
always
have
to
wait
between
the
two
do
I
break
somewhere,
which
I
know
are
open
source
for
really
care
about,
and
then
also
some
customers
as
well,
that
an
might
not
be
on
the
critical
path
of
the
feature
that
we
are
trying
to
ship,
but
they
can't
upgrade
when
we
ship
a
feature
in
a
patch
release
right.
That
is
something
that
I
continuously
have
to
wait
for
myself.
Alternative
I,
fully
agree
with
you
like,
however,
good
we
are
good
at
planning.
We
are
never
gonna,
get
it
100%.
B
How
I
would
see
this
process
actually
work
out
is
when
ever
there
is
this
critical
type
of
customer
interaction?
I,
don't
understand!
Why
are
we
not
leveraging
more
of
the
Future
groups?
I?
Don't
I
really
don't
remember
how
that
process
was
called
now
really
named.
It
probably
a
couple
of
times,
but
we
created
small
feature
teams
that
were
created
and
disbanded
as
soon
as
the
future
got
delivered
so
created
as
at
the
beginning
of
of
the
other
sprint.
And
then,
when
something
got
delivered
the
teen
got
disbanded.
B
B
But
at
least
it
was
very
clear
and
very
visible
that
we
were
shipping
an
MVC
and
we
were
continuously
moving.
Things
forward
think
something
like
that
for
high
impacting
features
should
be
implemented
prior
to
an
actual
high
priority,
merge
request
where
someone
pulls
the
line
and
says
okay.
Now
it's
time
to
go
in
this
case,
we.
A
C
That
I
think
merits
point
is:
is
you
start
three
weeks
back
with
lined
up
I?
Think
that's
I,
think
that's
I
think
that's
fair
feedback
right
is
this
like
like,
rather
than
treating
this
as
with
a
week
to
go
we're
gonna
we're
gonna
apply
pressure.
The
thought
process
is,
is
that
three
weeks
back,
we
basically
kind
of
adjust
on
it
and
basically
see
that
the
minutes
anyways
fair
feedback
like
we
knew
we
know
ahead
of
time.
C
So,
like
you
know,
the
question
was
is
when
do
engage
and
I
just
want
to
be
respectful
of
everybody's
time,
because
you
said
it,
you
said
it:
it's
a
more
efficient
I.
Don't
actually
think
it's
more
efficient,
I
think
it's
faster
there's
a
big
difference
between
those
right
efficient.
Is
that
everybody's
getting
what
they
need
done
done
faster
says
we're
getting
something
very
specific
done
at
the
cost
of
other
things
right
and
that's
that's
the
one
thing.
That
is
why
I
want
to
see
this
use
sparingly.
C
I,
just
I
just
know
that
we're
gonna
have
these
types
of
discussions
at
some
point
right,
key
features
and-
and
you
know,
what's
the
bar
I,
you
know
I
have
a
gut
that
says
you
know
the
bar
is
money's
on
the
line.
I
get
the
aspect
that
we
have
other
customers
and
other
considerations.
I
totally
agree
with
that
by
the
way,
but
you.
C
B
Define
exactly
who
is
going
to
be
your
reviewer
based
on
you
know
knowledge.
You
know,
whoever
is
on
that
critical
path
you
define
it
earlier
and
that
automatically
becomes
a
high
priority
request,
because
you
are
already
in
that
group,
if
you're
already
in
that
group.
That
indicates
that
it's
a
high
priority
anyway,
and
there
is
someone
who
is
driving
the
feature
itself.
So
you
have
all
the
stakeholders
involved
early
on
and
then
we
will
come
to
a
situation
where
that
is
going
to
take
time.
C
A
A
A
A
Y
is
worried
about
B
and
it's
gotten
too
big
and
it's
been
gotten
too
prescriptive,
so
I
think
a
simplified,
much
simplified
version
of
it
is
in
order
as
well
so
senior
director
of
dev
approval,
a
simp,
much
simplified
process
and
also
specifically
highlighting
to
involve
the
maintainer
z'
as
early
as
possible.
I
think,
would
be
key.
B
And
then
also
indicate
you
also
indicate
that
whoever
ends
in
that
group
right
that
was
approved
automatically
has
that
as
their
one
of
the
top
priorities
that
they
have
to
do,
because
they
are
in
that
group
right
and
then
it's
not
a
high
priority
merge
request
two
days
before
the
deadline.
It
is
a
high
priority,
merge
request.
That
is
a
team
effort
to
try
to
deliver
under
the
certain
deadlines.
I
like
it.
Thank.
A
A
Definitely
not
intentionally
and
don't
sir
commit
anything,
especially
unintentionally
would
do
to
not
what's
written,
but
by
people
reading
between
the
lines
and
assuming
they
should
be
doing
things
that
was
not
intended,
but
then
they
get
the
impression
they
should
do
and
that's
not
what
we
could
see
happen.
So
I
think
that's
I,
think
that's
I
think
that's
great.
C
Hey
can
I
can
I
just
say.
Thank
you
to
both
of
you
Wayne
thanks
for
taking
us
on
I
know
this
hasn't
been
the
easiest
task.
Yes,
you
do
and
Marin
thanks
for
the
Kim,
candor
and
feedback
I
really
appreciate
it,
because
this
is
how
we
make
things
better
and
make
sure
that
it's
acceptable
to
everybody
involved.
I
really
appreciate
both
of
you
work
on
this
with
with
me.