►
From YouTube: 2020 02 18 Memory Team 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
B
Lots
of
stuff
going
on
with
contribute.
There
is
a
contribute
slack
channel,
and
there
is
this
handbook
page
so
read
on
stay
up-to-date
with
what's
going
on,
it
changes
pretty
frequently
the
team
day.
We
still
need
to
get
some
feedback
on
that
on.
What
folks
want
to
do?
I
linked
over
to
what
the
distribution
team
is
talking
about,
I
heard
from
Rachel
that
their
team
is
just
gonna
kind
of
hike
around
the
city
and
they
don't
have
a
real,
strict
agenda
with
what
they're
doing
so
any
ideas
are
open.
B
C
C
C
B
Are
fun
yes,
so
our
team
day
will
start
around
10:30
and
then
we'll
have
the
team
dinner.
The
team
dinner
location
is
already
picked
because
there's
a
thousand
of
us
or
so,
however,
many
descending
on
Prague,
the
contribute
coordinators
have
set
aside
times
and
places
for
dinners
so
that
there
weren't
a
bunch
of
last-minute
reservations
whomping
these
restaurants,
so
the
engineering
teams
have
all
been
assigned
times
and
places
for
their
dinners,
and
it's
in
more
details
will
emerge
soon.
I,
don't
think
they've
released.
B
What
restaurants
are
where
we're
supposed
to
be
meeting
yet
so
alright,
but
yeah
go
back
to
the
issue
that
I
have
linked
here.
If
you
have
any
other
ideas
outside
of
what's
been
suggested,
please
feel
free
to
add
them,
and
if
we
have
a
couple
that
are
close
like
we
have
a
clear
winner,
then
I'll
just
book
it.
If
we
have
a
couple
that
are
pretty
close,
then
we
can
have
none
off
well.
C
B
Cic
being
so,
I
think
everybody
here
has
done
it
before,
but
yep
til
the
end
of
the
month
to
expense.
An
evangelism
dinner
and
last
update
I
have
an
interview
for
a
memory
team
product
manager
later
today,
I
know
there's
at
least
one
other
candidate
in
the
pipeline,
so
starting
to
make
some
progress
there.
A
C
So
this
came
up
in
the
first
much
requested,
yeah
much
requests
we
sent
for
both
inserts
kind
of
that
was
the
preparation
thing
and
someone
pointed
out
I
think
during
the
maintainer
review
it
was
pointed
out.
Oh
there
is
this
new
relation
that
we
have,
which
might
benefit
from
this
and
either
George
from
import
group.
I
knew
anything
about
it.
I
think
is
pretty
recent
edition
so
with
for
now.
Just
to
keep
that
mr
moving,
we
created
a
follow-up
ticket
for
this
because.
C
Considered
right
now
in
the
import/export
schema
file
that
we
use
so
so
right
now,
I,
don't
even
know
what
it
is.
So
it's
called
a
much
request,
context
commit
and
contact
tip,
so
I
think
someone
should
probably
spend
some
time
actually,
first
of
all
looking
into
should
this
be
part
of
export,
because
Kearney
is
just
ignored,
I,
don't
know
what
that
means,
but
yeah
so
either
our
team
or
I'll
be
important.
Like
group
managed
group
should
probably
look
at
okay,
even
if
it's
just
like
a
first
super
quick
assessment,
you.
B
D
C
C
There
was
some
discussion
about
this
somewhere
else,
because
we
do
have.
We
do
have
some
like
data
points
that
we
export
into
logs.
So
you
can
see
them
in
Cabana
and
Cabana
has
a
feature
as
well
where
you
can
create
visualizations
and
only
very
like
spend
like
20
minutes
on
this
or
so,
and
I
felt
a
bit
restrictive,
so
I
I
think
I
would
probably
towards
doing
this.
But
yeah,
like
you
know,.
B
C
D
D
C
F
D
C
C
C
C
Too
low
because
they
were
configured,
this
was
kind
of
like
yeah,
we're
kind
of
shooting
ourselves
in
the
foot
here,
because
they
just
in
a
sentence
that
were
basically
configured
to
run
on
a
single
thread
and
and
by
default.
If
you
don't
scale,
the
pool
manually
I
think
the
default
value
is
0
or
1.
So
so
our
scaling
logic
took
the
maximum
of
this,
which
is
one
which
is
not
enough
to
run
these
things
right.
So
it's
like
kind
of
like
silly
little
things
that
we
didn't
think
about.
C
You
know
or
assumptions
that
we
made
of
cases.
That
would
not
happen,
but
clearly
they
are
happening
and
yeah.
We
don't
detect
this
currently
and
it
took
them
a
lot
of
time
like
the
engine
engineers
I
sorry's.
They
spend
hours
like
trying
to
figure
out
what
was
going
on.
You
know.
We
also
need
to
account
for
all
that
time
spent
debugging.
These
issues,
mm-hmm.
B
B
B
Does
this
work
feel
pretty
bigs
I
feel
like
it's
gonna
spill
over
beyond
twelve
nine,
four,
indeed
gays
on
in
general,
because
if
it
is,
then
we
need
to
get
the
import
team
involved
in
this.
Oh
no.
C
It
will
definitely,
but
so
that
the
two
main
areas,
so
we
thought
there's
probably
two-
were
tracks.
One
is
really
focusing
on
kind
of
nd
Jason
per
se
and
there's
already
an
open,
mr,
which
is
pretty
far
that
implementation
is.
This
is
actually.
D
C
Really
great
job
there
there's
just
like
a
couple
of
things:
it's
not
ready
to
be
shipped
tomorrow,
maybe,
but
so
this
whole
first
work
track.
What
make
sure
that
this
is
something
that's
actually
emerge
Abul
and
we
can
put
that
in
front
of
well
not
in
front
of
users,
maybe
but
like
get
that
into
production
and
in
a
way
that
is
compatible
as
we
currently
have.
C
A
C
B
B
E
D
This,
mr,
the
finish
not
done
about
this
aspect,
but
I
didn't
merge
it
yet
I
hope
that
I'm
gonna
use
that
this
week,
because
my
discussion
going
this
one
and
I
emerged.
Some
other
changes
related
to
Puma
and
I
need
to
look
exactly
what
is
the
state
of
that,
but
I
likely
gonna
close
this
issue
as
we
solved
as
of
now.
This
is
this
is
my
perception
right
now
Frankie.
We
can
close
this
issue,
but
III
need
to
exactly
look
what
was
the
status
of
that
stuff?
D
D
D
D
D
Our
sorry
III
think
like
that,
that
ish
is
much
smaller
change,
which
is
like
only
updating
the
default
settings
to
what
we
consider
today,
I
the
best
and
defined
by
our
it's
more
like
the
investigation
to
understand.
If
there
is
something
better
that
we
should
do
so.
I
think
that,
like
this
way,
is
like
updating
exactly
where
the
ground
30
shows
the
line
did
change,
because
code
be
changed
to
sensor
defaults.
D
The
forum,
please
ascended
to
me,
I'm
gonna,
comment
exactly
on
the
state
of
how
to
resolve
the
confusion
between
these
and
these
other
ones.
F
C
C
Often
like
this
could
have
even
triggers
right,
because
it's
just
so
specific
to
a
very
large,
deport
and
I
I'm,
not
sure,
like
it's
a
bit
difficult
to
say
when
that
this
is
considered
safety
to
you.
On
the
other
hand,
like
you
know
what
I'm
actually
thinking
come
here.
What
do
you
think
about
this?
Because
what
once
we
do,
the
any
Jason
changes?
C
C
C
B
D
Camille
I
would
like
us
to
like
to
continue
talking
about
the
nd
Jason,
and
the
second
truck
is
like
defining
how
we
can
avoid
supporting
legacy
format
in
the
most
simple
boring
iterative
way.
I
want
to
push
on
like
figuring
out
that
maybes,
not
gonna,
be
super
perfect,
but
maybe
we've
30
by
zero.
That
cannot
happen
in
three
releases
from
now.
We
could
remove
a
lot
of
legacy
code
from
the
import/export
yeah.
B
D
So
so
like,
we
are
probably
on
a
very
tight
timeline
given
between
9,
because
I'm
really
hoping
that
for
us
to
finish
that
in
1299
make
decision
and
make
the
smallest
possible
decision
that
we
can
make
and
also
like
implement
that
and
kind
of
finish
all
development
work
in
the
12.9
and
maybe
have
some
extra
work
be
floating
out,
but
do
not
move
our
development
working
into
twerking
time.
That's
kind
of
my
intent,
shortening
the
cycle
and
speeding
the
iteration
here
so.
D
This
is
way
what
I'm
really
looking
for
I'm
I'm
certain
in
dozens
of
our
possible
problems,
but
I
hope
that
is
it
all
of
us
to
say
we
13.0.
We
just
stopped
supporting
legacy
for
months.
This
is
my
intern.
If,
if
we
can
be
a
formative
of
of
that
goal,
and
it's
good
enough
solution,
I
would
ask
for
us
doing
that,
if
possible,
okay.