►
From YouTube: 2023-08-08 Product Analytics Group Sync
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
C
A
Better
all
right,
these
issues
have
been
here
for
an
open
for
some
time,
so
I
think
unless
anyone
has
anything
they'd
like
to
call
for
those
we
can
just
move
on.
A
We've
got
the
issue
applying
breakdown
for
visualization
and
error
handling
for
dashboard,
visualizations
I
know
someone
was
looking
I.
Think
Alan
was
looking
at
this,
but
I
don't
see
it
assigned.
So
unless
anyone
has
anything
we'd
like
to
add
about
this,
we
can
move
on
all
right
and
we've
got
updating
the
dashboard
style
to
match
the
latest
designs
which
isn't
ready
for
development.
So
as
soon
as
anyone's
available
to
pick
that
up,
please
feel
free
to
sign
it
to
yourself.
A
A
Great
this
is
the
customer
facing
of
structure,
release
issue
that
calls
out
of
what's
remaining
to
get
out
to
customers,
including
security
improvements
and
the
new
issue
as
well,
for
connecting
to
college
Cloud,
which
I
will
cover,
as
it's
got
its
own
issue
further
down
the
list.
But
as
far
as
the
next
two
issues,
we've
got
follow-ups
for
adding
missing
filling
breadcrumbs
for
analytics.
A
Dashboard
Pages,
as
well
as
the
with
the
order
by
modifier
that
we
want
to
make
to
the
couch
driver
Rob
is
out
on
PTO,
so
I,
don't
think
we
have
any
updates
there
unless
anyone
else
does
cool,
then
last
but
not
least,
we
have
connecting
clickhouse
Cloud
analog
Stacks.
This
is
our
new
development.
A
As
of
last
week,
we've
got
access
to
clickhouse
Cloud,
a
managed
service
by
clickhouse
that
has
recently
released
on
gcp
and
so
I'm
looking
to
connect
to
that
to
our
production
cluster
as
well
as
use
that
for
the
ingesting
gitlab.com
data,
and
so
hopefully
this
should
help
us
in
a
couple
of
different
aspects.
How
to
further
agenda
point
which
I
guess
I
might
as
well
just
explain
now
in
terms
of
the
security
review,
because
if
we
use
clickhouse
cloud,
we
wouldn't
be
ourselves
hosting
Red,
Data,
so
I
think.
A
As
far
as
the
analytics
stack
is
concerned
for
sasstor.com,
we
really
only
have
to
worry
about
how
we
Transit
or
Tran
traffic
Red
Data
to
clickhouse
cloud.
Clickhouse
cloud
itself
has
been
approved
for
red
data
usage
already
and
as
far
as
the
analytic
stack
development
is
concerned,
I
think
everything
we
need
as
far
as
continuing
to
secure
the
transit
of
data
is
in
review.
A
A
Cool
moving
on
to
in
review
yeah
and
you
have
Cube
query
rendering
adding
visualization
options
to
schema
anything
to
call
out
here.
C
A
Awesome
Ellen's
got
the
next
one
for
sensitive
fields,
which
I
know
have
some
conflicts
with
another
issue
which
I
think
has
since
been
resolved.
So
hopefully
we
can
get
this
through
the
review
process
and
then
Rob
has
a
Mr
for
the
analytics
stack
to
remove
all
the
digital
related
Services,
hello,
I
think
you've
already
reviewed
this.
A
C
No,
this
is
just
a
small
but
part
of
an
ongoing
effort
to
improve
our
loading
States
across
dashboards.
Currently,
we
have
a
lot
of
spanners
everywhere,
which
is
not
performance.
Compliance,
we're
moving
towards
skills
and
motors.
D
Yeah
there,
let's
give
a
notification,
basically
I,
want
to
deploy
them
and
then
see
if
it's
about
the
non-non
default
user
about
the
security
improvements
yeah.
This
is
after
verification.
We
can
close
it,
but
I
expect
some
follow-ups
based
on
how
the
security
check
will
be
if
yeah.
A
Got
it
sounds
good
and
I'll
take
a
look
at
your
outstanding
merge
requests
as
well
when
I
get
the
chance,
I
believe
the
last
bookmark
was
product
analytics
query
spelled
to
load
I
just
moved
this
one,
because
this
was
in
a
different
project
and
I
moved
it
over
to
analytics
Stacks.
So
that's
not
actually
done
yet.
A
So
we've
got
a
couple
issues
here.
Oh
one
completed
equating
the
configuration
screen
for
configuring
product
analytics
post
initial
setup
James,
you
completed
the
usage
view
user
story
map
anything
you
wanted
to
call
it
here.
E
The
implementation
issue
is
there,
there's
a
clunky
wireframe
that
I
put
together,
showing
what
that's
going
to
look
like
and
started
carving
out
the
different
use
cases
that
support
for
the
MVC
thanks
everyone
who's,
taking
a
look
and
added
questions,
but
going
to
start
getting
that
work
scheduled
here
in
the
next
Milestone
or
two,
and
it's
a
nice
Kickstart
to
the
development
work
as
well.
A
Awesome
cool
and
then
I'll
always
added
a
loading
state
to
the
share
dashboard
listing,
so
we're
not
just
waiting
for
cube
queries
to
return
back
before
we
see
the
dashboards
with
a
blank
screen.
So
that's
a
great
Improvement
we've
got
to
create
a
custom,
dashboard
flow,
NBC,
design,
issue
James.
Anything
you
wanted
to
call
out
here.
E
Now
that
was
just
closing
out
the
design
part
of
things
that
was
Kevin
thought
packed
the
implementation
issue.
You
should
have
the
design
attached
to
it
now,
so
you
can
start
working
on
dashboard
designer
using
the
existing
visualizations
out
of
the
box
and
then
we'll
start
working
on
exposing
visualization
designer
in
its
current
state
either.
So
they
can
start
to
save
new
visualizations
and
add
them
to
custom
dashboards
as
well.
A
Awesome
and
I
thanked
everyone
on
the
dashboard
designer
issues
to
kind
of
start
breaking
that
down
from
its
current
state.
So
please
take
a
look
if
you
have
a
moment,
because
that'll
be
a
focus
for
the
quarter
which
I'll
get
to
for
okrs,
but
also
for
the
next
Milestone,
so
cool
and
then
Leo
you've
got
the
next
issue
here.
D
So
last,
in
the
last
Milestone,
this
initial
analytics
worker
was
using
Jitsu
and
in
the
last
Milestone
we
actually
converted
to
no
op.
This
was
a
cleanup
MR2
yeah
cleaned
up
that
worker
got.
A
C
Should
now
be
a
lot
less
fun
to
use
in
terms
of
just
being
boring
and
working,
but
I
think
I'm
excited
to
work
with
whoever
we
get
as
a
new
ux
counterpart
on
just
improving
this
in
general,
because
I
feel
like
popovers
on
Hover
is
not
the
right
thing
to
use
here
and
that
relates
to
another
closed
issue
that
I'll
talk
about
later.
A
But
that
is
the
now.
This
Gap
here
would
be
the
differentiator
between
being
able
to
actually
hover
over
it
to
click
the
link
or
not.
So
we
since
resolved
that
so
thanks,
John
allele
you've
got
an
issue
here
for
viewing
the
shared
analytics
dashboard
without
permission
to
the
500
error.
Page.
A
Cool
and
then
We've
tweaked,
the
Border
error.
Excuse
me:
error,
border
radius
for
dashboard
panels.
I
want
to
thank
you
for
that
and
then
yeah.
You
have
the
next
two
issues.
C
Yeah,
so
the
retry
button
was
added.
This
is
just
a
small
quality
of
life
like
ux
helper,
basically
to
allow
people
to
retry
individual
panels
instead
of
having
to
refresh
the
entire
dashboard.
This
is
especially
useful
while
we're
still
running
into
stuff.
Like
query
timeout
Studio,
like
memory
limits,
the
following
issue
is
a
gitlab
UI
issue
that
I
close.
This
won't
fix
because
after
doing
some
more
ux
research
into
whether
or
not
this
should
be
something
we
do.
C
I
came
to
the
decision
that
we
shouldn't
do
this
I,
don't
think
we
should
universally
increase
the
delay
limit
to
a
half
a
second,
because
what
happens
especially
on
the
dashboard
that
is
completely
errored
out.
Is
you
move
your
mouse
cursor
over
the
dashboard
and
you
suddenly
have
like
eight
pop-overs
lingering
on
your
screen?
C
A
Makes
sense
yeah?
Perhaps
we
can
just
have
it
over
the
icon
or
something
or
some
actual
specific
trigger,
as
opposed
to
the
entire
thing.
F
A
Can
figure
that
out
with,
hopefully
a
new
designer
yeah
cool
awesome,
then
that
wraps
up
the
billboard
review
any
final
thoughts.
A
All
right,
then,
James
I,
think
you
or
if
you
have
a
read-only,
so
I
will
read
it
for
you.
Unless
you'd
like
to
read
it.
E
A
Awesome
all
right,
I've
got
the
next
one
I
just
kind
of
wanted
to
provide
a
little
bit
of
some
visibility
into
some
work
in
progress.
Okay,
guys
for
engineering
I'm
a
week
late,
but
you
know
better
late
than
never,
but
hopefully
looking
for
any
feedback
regarding
this
should
hopefully
come
to
no
surprise,
at
least
for
the
first
objective
in
terms
of
getting
out
to
customers.
I
know
we're
all
wanting
it
and
I
think
we're
making
some
great
progress
there
to
get
closer
and
closer.
A
But
of
course
not
only
do
we
want
to
get
out
to
the
experiment
phase,
but,
as
James
mentioned
in
the
product
okrs,
we
want
to
get
deliver
the
beta
and
so
really
around
that.
So
the
first
key
result
for
delivering
to
customers
is
getting
out
there
and
getting
getting
in
the
hands
of
external
customers,
but
for
the
data.
This
is
where
things
get
interesting.
A
In
the
engineering
side,
it
is
definitely
a
little
bit
more
infrastructure
focused
in
terms
of
how
do
we
actually
set
up
monitoring
alerting
so
that
we
know,
and
we
need
to
get
our
direct
our
attention
to
scaling
the
services
so
that
we
can
continue
to
help
handle
customers
use
throughout
the
experiment
and
beta
phases
we've
already
started
discussing.
How
would
we
actually
expose
this
and
how
allow
users
to
opt
in
James
and
I
have
been
discussing?
You
know,
P.
A
We
have
an
issue
for
the
experiments,
often
similar
to
how
people
opt
into
AI
experiments,
currently
being
able
to
have
that
toggleable
or
controlled
feature
flag,
so
that
we
can
actually
control
people
can
opt
in
and
out
if
we
find
that
performance
or
load
is
is
becoming
too
much
for
us
to
handle.
A
At
the
same
time,
we
are
getting
to
the
point
where
we
need
to
start
actually
really
understanding
the
costs
and
how
we
can
actually
optimize
performance
to
know
how
much
the
service
can
handle
our
clusters
can
handle
and
how
we
can
scale
it
as
well
as
how
much
it
costs
so
that
we
can
really
wrap
up
the
conversation
around
packaging
on
pricing
and
then,
of
course,
how
we
want
to
start
to
present.
A
As
we've
seen
with
the
user
usage
view
that
uses
quota
user
story
map
that
James
has
been
working
on
and
how
we
actually
bring
that
visibility
to
customers
to
actually
understand
how
much
they're,
using
as
well
as
to
eventually
start
exploring
how
to
close
that
loop
with
billing.
As
we
work
towards
GA
to
actually
be
able
to
let
them
know
how
much
they've
used
and
how
much
they
owe
us
for
consumption-based
usage
pricing.
A
So
there's
a
lot
more
supporting
issues
in
epics
I
have
to
come
to
this,
but
I
just
wanted
to
call
that
out
and
that
we
would
like
to
I
think
once
we
get
out
to
to
customers.
I
think
we'll
have
a
lot
more
momentum
to
kind
of
snowball
into
maturing
the
platform,
so
we
can
open
this
up
to
everyone.
A
Furthermore,
we
of
course
still
wants
to
deliver
improvements
to
what
we've
built
so
far
in
product
analytics.
So
the
main
focus
here
is
completing
the
experience.
We've
got,
CMA
driven
dashboards
and
we've
got
the
ability
to
do
it
manually,
but
we
want
to
complete
that
experience
in
terms
of
making
sure
it's
easy
to
use
from
a
user
perspective,
and
so
that
means
building
a
UI
around
it.
A
So
this
should
come
to
shouldn't
be
surprising
as
we're
already
starting
to
pull
in
the
dashboard
designer
MVC
and
exposing
visual
designer
as
it
is,
but
I
wanted
to
call
out
that
we
still
want
to
make
sure
we
can
kind
of
complete
the
UI
based
experience,
your
own
product
analytics
and
custom
dashboards,
so
yeah
and
then
something
new,
that's
kind
of
coming
out
with
the
recent
launch
of
gitlab
Duo
is
that
the
AI
framework
team
and
everyone
in
the
borrow
request,
is
kind
of
starting
to
contribute
to
building
an
API
so
that
other
features
other
groups
can
bring
in
their
context
specific
domain
to
actually
help
Duo
do
more,
and
so
we've
already
started
experimenting
with
natural
language
querying
and
using
it
through
Optics
Cloud,
2
model,
and
things
like
that,
and
so
the
natural
next
step
of
that
is.
A
If
the
a
framework
team
is
building
an
API
on
top
of
Duo,
how
can
we
leverage
Duo
to
potentially
perform
queries
through
clickhouse,
passing
it
through
Cube,
or
you
know,
we're
looking
at
schema
validation?
Can
we
help
explain
the
visualization
schemas,
and
things
like
that?
So
this
is
more
just
exploratory.
Just
to
kind
of
see
how
we
could
potentially
leverage
that
as
it's
maturing
within
gitlab,
so
instead
of
us
directly
using
the
models
we
can
use,
Duo
which
actually
perform
based
on
the
user
input
determines
what
tools
to
use.
A
So
given
the
context
of
product
analytics,
we
can,
you
know
potentially
leverage
Duo
there,
but
those
are
the
okrs
that
I
have
so
far
open
to
Dots
here
in
the
call
or
any
feedback
afterwards.
Please
feel
free
to
comment,
but
just
wanted
to
call
that
out.
A
F
Yeah,
sorry
to
budget
your
meeting
today,
I'm
here
to
try
and
throw
a
provide
a
little
bit
more
coordination
from
infrastructure
side
and
provide
support
to
the
team
here
for
any
infrastructure
needs
and
I
are
working
together
on
the
infrastructure
side.
F
Reliability
specifically
and
I
really
want
to
make
sure
that
you
know
the
product
we
and
the
features
we
developed
here
are
also
successful
for
money
infrastructure
to
interview
so
from
time
to
time,
I'll
be
looking
at
the
agenda
and
also
joining
the
meeting
to
make
sure
that
we
are
providing
all
the
necessary
support
and
if
there's
additional
help
from
that,
Nails
need
a
platform
just
himself.
We
can
see
how
do
we
Fast
Track
such.
F
B
B
However,
I'm
thinking
that
there's
probably
some
things
that
I'll
be
able
to
continue
to
work
on
as
long
as
I
get
unblocked
by
this
particular
issue
that
I
opened
earlier,
which
is
issue
49
in
the
analytics
Tech
project,
which
is
I
request
to
have
the
current
publication
of
the
analytics
stack,
Helm
charts
pushed
into
the
Ops
registry,
which
is
what
infrastructure
uses
to
deploy
from
for
their
releases
into
their
gke
environments.
B
I
think
all
it
really
requires
is
a
is
a
change
to
or
in
either
in
addition
to
the
existing
CI
pipeline
stage
for
publication
in
the
analytics
stack
or
simply,
the
creation
of
a
mirror,
a
push
mirror
from
the
gitlab.com
project
to
the
ops.gitlab.net
project
and
there's
a
token
available
for
that.
Push.
If
requested.
A
Okay,
I'm
happy
to
dig
into
this
I
guess
I'm
not
as
familiar
in
creating
a
mirror
to
this,
but
it
seems
like
we
would
have
the
credentials
to
be
able
to
do
so.
So
I
can
follow
up
with
you
offline,
perhaps
to
see
what
we've
done
from
there.
A
Okay,
cool
awesome,
any
other
final
topics.
Anyone
would
like
to
discuss
all
right.
Well,
it's
good
to
see
everyone.
Everyone
gets
10
minutes
back,
have
a
good
rest
of
your
Tuesday
rest
of
your
week
and
take
care.