►
From YouTube: 2020 12 10 Database Group Meeting
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
Okay,
jumping
into
the
rice
framework
for
the
database
team,
so
we
now
have
a
single
source
of
truth,
the
spreadsheet
for
all
the
rights
framework
for
the
database
team.
A
lot
of
this
was
copied
over
from
the
dock,
and
giannis
did
a
great
job
of
breaking
it
down
by
epics
and
issues.
It
was
a
good
idea,
so
we
have
scores
on
most
of
these
this
one.
We
did
not
score.
I
think,
oh,
no.
A
We
didn't
score
because
it
needs
to
be
broken
down
because
there's
quite
a
bit
in
there,
but
I
don't
think,
there's
an
urgency
to
break
this
one
down
because
we're
not
going
to
touch
it
for
a
while.
So
we'll
just
leave
that
one
alone
giannis
were
there
issues
in
here
that
we
should
start
scoring
now.
Is
there
any
semblance
of
order
in
which
you
put
these,
or
did
you
just
drop
them
in.
B
Oh
so
I
just
dropped
them
in
the
beginning.
I
tried
to
also
link
the
epic
so
that
we
know
which
issues
are
in
epics
and
then
fabian
today
in
our
course
moved
the
ones
that
have
epics
at
the
top.
So
the
order
is
issues
that
have
ipx,
and
especially
I
have
the
two
that
most
probably
we
will.
We
will
plan
either
way
for
13.00,
the
defense,
migration
and
the
second
step
of
step
of
a
converting
events.
Id
everything
is
an
order
there.
We
have
to
decide
and
also
below
those.
B
There
are
the
two
issues
on
the
the
ieds,
which
I
assume
are
really
important
and
we
have
to
discuss
with
them.
So
andreas
can
provide
more
everything
else
is,
then
there
are
the
issues
that
craig
added-
and
I
also
brought
in
all
the
issues
we
had
in
our
board
for
13.8
and
any
pending
is
from
13.7.
A
Thank
you.
So
I
guess
the
question
is:
do
we
want
to
run
through
this
synchronously
or
do
we
want
to
do
an
async.
B
B
C
B
What
I've
seen
in
the
memory
team?
What
they
did
was
that
everyone
feel
the
feeding
their
scores,
and
then
there
was,
you
run
a
median.
B
A
We
started
out
async
and
then
the
ones
that
were
pretty
close
together
in
score.
We
didn't
really
talk
about.
We
spent
some
time
on
the
ones
that
were
pretty
far
apart,
because
there
were
a
couple
that
were
way
far
apart.
So
that's
how
the
memory
team
did
it.
B
So
the
the
only
thing
that
we
have
is
that
13.88
is
pretty
close
and
with
friends
and
family
day
and
everything
we
will
have
to
finalize
the
planning
issue
and
decide
which
issues
we're
going
to
have
for
13.8
and
run
the
kickoff.
So
maybe
yeah.
I
don't
know
if
we
can
at
least
discuss
that
or
we
could
do
a
synonymously.
C
I
find
sync
pretty
good
to
you
know,
discuss
any
like
if
there's
a
huge
difference
between
things,
but
it
might
take
a
while
to
really
get
those
numbers
so
getting
maybe
the
numbers
first
and
then
identifying
what
we
want
to
talk
about
right.
D
A
It's
fine
too,
I
mean
you
can
always
put
that
too
to
say
I
would
need
more
information
that
actually
happened
on
a
couple
of
memory,
team
issues
and
then
those
were
things
that
we
talked
about
synchronously.
So
let's
do
that.
I,
I
think,
probably
what
makes
sense
everybody
just
kind
of
make
their
own
copy
of
the
spreadsheet,
and
then
we
can
put
it
on
the
agenda
for
tuesday
to
come
together.
B
Okay,
could
we
go
through
which
ones
where
we
plan
to
add
for
13.8,
so
that
we
can
move
forward
with
the
planning.
A
A
C
Seeing
sorry
before
we
before
we
get
into
that
also
seeing
the
next
topic
about
those
database
peaks
that
we
observe,
would
it
make
sense
to
sort
of
keep
that
in
mind
for
the
next
milestone
as
well
and
so
reserve
some
time
for
that,
because
I
think
that
is
we
continue
to
do
it
that
way.
There
is
a
couple
of
findings
that
in
our
attention
as
well,
we
will
need
to
spend
some
time
on
those,
even
if.
B
A
Yeah
yes
agreed,
and
the
question
I
had
here
on
this
one
as
well
is:
what's
the
other
filter
we
should
add
on
here,
so
we're
only
looking
at
new
ones.
I
was
thinking
we
do
like
milestone
is
empty.
B
B
C
And
I
think
many
people
are
are
getting
those
those
things
and-
and
I
can
imagine
everybody's
quite
quite
looking
at
this
stuff
and
to
some
extent
I
think
it's
not
even
not
even
clear
at
least
to
me,
but
what
the
significance
of
those
things
are.
It
is
like
those
are
all
good
findings.
What
what
is
sort
of
the
overall
significance
in
fixing
those
issues
it's
hard
to
tell
until
somebody
looks
at
that
so
and
that's
probably
going
to
be
a
good
part,
is
going
to
be
on
us.
B
A
B
End
if
we
get
a
severity,
one
issue
means
that
we
stop
everything
and
work
on
that.
So
it
took
me
a
while
to
understand
that
there
is
a
different
meaning
of
severity,
one
for
those
issues.
B
A
C
So,
just
just
to
give
some
context,
so
this
is
a
background
job
that
basically
recalculates
permissions
for
a
particular
user.
C
So
whenever
something
changes
you
add
a
group,
you
remove
a
group
remove
or
change
group
membership
and
those
kind
of
things,
then
the
the
permissions
to
access
those
projects
they
change
and
what
we
do
is
basically
do
that
in
the
background
schedule
that
job
review
all
and
recalculate
all
the
permissions
for
that
user
and
sort
of
swap
things
out,
so
the
job
is
basically
two-phased.
C
The
first
phase
is
recall:
recalculate
all
the
permissions
for
all
the
projects
that
the
user
has
access
to
and
then
second
phase
is
swapping
swapping
out
the
data
on
the
table,
and
what
we
see
here
is
the
the
first
phase.
That's
the
read
phase,
and
this
is
a
very
heavy
query-
that
sort
of
tries
to
catch
all
those
cases.
Lots
of
union
unions
going
on-
and
this
is
this-
is
showing
up
now
and
is.
This-
is.
A
This,
the
one
that's
being
exacerbated
by
our
one
customer
that
is
running
this
quite
a
bit
and
is
a
big
camera.
C
It
could
be,
there
is
a
theory
there,
but
it's
not
been
proven
yet.
I
think.
C
B
Okay,
so
some
other
issues
guys
that
was
the
problem
but
yeah
and
for
for
what
I
can
say,
andreas
can
correct
me.
B
This
thing
with
the
recursive
namespace
queries
and
the
other
problem
with
permissions
going
all
over
the
place
is
there
is
a
problem
we
have
in
general
with
the
with
our
with
our
design,
and
we
know
about
it.
This
is
exaggerated
here
with
the
authorized
project
worker
for
sure,
but
this
is
also
a
longer
discussion.
C
Yeah
that's
right,
and
for
that
particular
topic
I
can
see
there
is
two
sort
of
things
that
we
might
want
to
look
into.
One
is:
can
we
do
we
see
anything
that
would
immediately
improve
that
query,
because
that
would
be
the
easiest
way.
We
add
an
index
somewhere,
something
like
that
maybe
hard
to
do,
and
we
may
not
work
out.
But
that's
that's
one
avenue.
C
The
second
one
is,
I
split
it
out
into
a
separate
issue
is
something
that
is
also
applicable
in
general
about
side
kick
jobs,
psychic
shops
they
all
go
to
the
primary
and
what
we've
discussed
is
sort
of
having
having
a
way
to
label
jobs,
making
them
or
marking
them
as
read-only
jobs,
so
they
can
actually
use
a
replica,
and
in
this
context,
what
we're
discussing
is
allowing
the
read
phase
of
that
job
to
go
to
the
replica,
which
has
certain
problems
with
consistency,
because
the
replica
can
always
lag
and
yeah
there
is
there
is.
C
You
know
we
have
to
be
sure
that
we
are
reading
up-to-date
data,
particularly
for
this
job,
which
is
security
which
has
security
implications
right.
I
think
those
are
the
two
two
topics
that
I
I
see
for
for
that
issue
and
the
bigger
one
like:
how
do
we?
How
do
we
actually
want
to
calculate
those
permissions?
How
do
we
manage
those
permissions?
I
think
that
that
really
needs
to
go
together
with
the
I
can
group
the
access
group
and
I'm
sure
they
or
they
already
have
good
good
insight
into
what
is
happening.
A
C
A
A
B
Is
this
the
one
that
you
added
to
the
index
that
you
proposed
an
index
for
andreas,
or
is
this
another
one.
A
A
C
Those
are
all
the
pretty
simple
queries
for
record
lookups,
so
I
guess
most
of
the
time
it's
one
record.
It
comes
back
there
is
this
change
about
to
happen
with
the
query.
C
Caching,
you
know
where
we
enable
the
query
cache
to
work
together
with
the
load
and
at
least
for
those
n
plus
one
queries
that
might
have
a
have
a
good
impact,
and
then
what
jose
is
talking
about
here
is,
I
think
so,
aside
from
another
like
can
we
can
we
do
any
sort
of
additional
caching
to
those
queries
where
we
don't
hit
the
database
for
for
them,
and
this
is
obviously
a
bigger
topic.
We
don't
have
any
any
caching
going
on
for
that
at
the
moment,.
A
I'm
okay
figure
out
another
way
to
filter
those
out.
Then
just
leave
that
for.
A
Now
so
thinking
about
13
8,
do
we
just
want
to
come
back
and
we'll
address
it
on
tuesday,
when
we've
gone
through
the
first
pass
of
the
the
rice
framework
on.
B
These
yeah,
but
what
are
we
going
to
add.
C
A
A
C
B
The
audit
events
will
go
to
certain
point.
Eight
am
I
right,
patrick
because.
D
C
And
the
more
or
the
issue,
with
more
with
a
higher
priority
that
we
discovered
was
the
one
about
internal
ids
and
consistency
right
was
was
the
other
team?
Picking
that
up?
Are
we
gonna?
Look
at
that.
C
C
A
B
To
them
and
that's
on
the
plan,
okay,
I
think
that
this
should
not
create
a
new
work
for
us
or
a
lot
of
work
for
us
for
13.8,
but
let's
start
discussing
with
them.
C
A
A
No
they're
still
looking
into
it.
They
understand
that
we're
waiting
for
guidance
on
how
to
set
up
a
server,
but
now
there's
there's
no
news
on
their.
C
I
don't
know
if
we
wanted
to
add
more
details
to
sort
of
a
a
design
document.
We've
started
to
write
this.
I
think,
describing
how
we
how
we
intend
to
do
that
testing.
C
A
Are
you
talking
about
the
blueprint
or
yeah?
No,
I
haven't
touched
it
in
a
couple
weeks.
C
C
A
Yeah,
I
still
need
to
get
an
update
to
that
blueprint.
However,
so
I'm
not
sure
if
there's
an
issue
specific
to
that
or
if
we
even
need
one,
it
might
just
be
duplicate
work.
A
A
C
I
haven't
really
put
anything
on,
but
it's
similar,
like
you,
said
over
the
holidays,
maybe
a
day
before
that,
or
so.
If
that
works
out,
but
I'm
gonna
be
around
most
of
the
time.
A
B
That's
the
problem
nowadays,
I'm
not
sure
about
which
way
around
what
she
wants,
but
yeah.
Exactly
aside,
I
think
that
if
we
assume
that
this
is
a
milestone
short
by
a
week,
this
is
not
for
everyone.
A
A
So,
let's
keep
talking
async
about
what
we
should
bring
in
you,
honest
you
and
I
can
polish
it
tomorrow
when
we're
talking.
It
sounds
like
there's
a
lot
of
little
things
that
we
need
to
do
for
next.
Milestone,
mostly
bug
fixes
improvements
and
proving
like
the
primary
key
migration
works
out.
Fine,
so
that
we
can
continue
on
the
next
one
and
we
have
reduced
capacity.
So
we
may
be
fairly
close
to
what
we're
going
to
do
for
13.8
based
on
what's
in
there
but
yeah.
B
B
A
What
was
left
so
already
answered
that
question:
nope
yay
for
the
memory
team
update,
yep.
A
All
right,
I'll
I'll,
think
I'll
put
fun
beyond
to
make
sure
that
gets
in.