►
From YouTube: 2023-02-14 Product Analytics 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
Hello,
everybody
and
welcome
to
the
product
analytics
sync
for
February
14th
2023
I'm
going
to
go
ahead
and
share
my
screen
and
we're
going
to
do
the
build
border
review.
A
All
right
so
I'm,
just
gonna
hide
all
the
zoom
stuff.
Hopefully
everyone
can
see
my
screen
all
right,
cool,
so
open
issues
deploying
podcast
analytics
to
gitlavas
the
project
itself,
specific
opt-in
users.
We
still
only
have
three
opt-in
users
at
this
point.
A
We're
going
to
do
some
more
announcements
if
anyone
else
would
like
to
opt
in,
but
we
may
end
up
actually
kicking
this
out
depending
on
how
much
feedback
we
get
and
and
maybe
focus
our
efforts
on
deploying
onto
the
public
handbook,
for
example,
so
yeah
the
59
planning
issue,
I,
don't
think
needs
to
be
covered
at
this
point
max
you
have
the
product
analytics,
export,
Spike.
B
Investigation,
we're
gonna
have
a
sync
you
me
and
Sam
later,
just
to
make
sure
on
the
same
page
and
decide
what
we're
going
to
do.
Next.
That's.
A
Correct
we
also
have
an
open
issue
for
The
Collector
host
to
differentiate
that
from
the
configurator
I.
Think
we'll
just
need
to
break
that
down
and
schedule
that
in
but
that's
open
if
anyone
has
any
available
capacity
there,
the
capacity
there
we
have
a
discussion
around
the
query,
designer
I
believe
we've
already
settled
on
the
name.
A
I
think
we
also
have
a
strategy
in
terms
of
how
we
want
to
prefer
kind
of
deploy
the
first
implementation
on
production
and
we'll
probably
want
to
handle
it
if
we
do
end
up
having
multiple
versions
of
it
or
variants
of
that,
I
think
that
part
of
the
strategy
does
need
to
be
documented
and
I
have
a
draft
somewhere.
So
I
will
try
to
publish
that
as
soon
as
possible,
and
we
can
probably
close
that
discussion
out
so
yeah.
A
C
C
It's
good
that
we
now
have
an
idea
of
what
we
are
going
to
do
with
the
listing
page,
because
this
also
impacts
how
we're
gonna
update
and
how
we
want
to
migrate
from
our
current
onboarding
process
to
the
new
listing
page,
which
is
going
to
be
shared
sort
of
all
noted
that
in
the
implementation
plan,
so
yeah,
whenever
both
the
back
and
the
front
end's
approved,
we
can
move
this
onto
greater
scheduling.
A
C
That
yeah
they're
all
likely
slip.
It's
also
an
issue
that
was
I,
think
created
during
the
milestone,
and
it
was
just
added
in
case
we
had
the
capacity
but
yeah.
A
It's
fine
I'll
leave
it
here
for
now
and
then
I'll
move
it
over
to
the
next
milestone.
Once
we
close
this
one
out,
we
have
the
dashboard
issues
for
audience.
Behavior
I
did
finally
file
the
issue
to
start
instrumenting
custom
events
for
the
internal
handbook.
A
I
think
we
can
I'll
assign
both
of
these
to
myself
actually
and
just
make
sure
there's
nothing,
there's
no
other
follow-ups
to
file,
and
we
can
close
these
out
since
these
are
effectively
out
there
and
we've
got
date,
range
filters
and
all
that
kind
of
stuff
which
we'll
speak
to
later
set
up
data
persistence
with
Jitsu
writer
server
service
I
have
not
made
any
movement
since
I've
been
working
on
standing
up
Devon
staging
clusters
I've
been
trying
to
resolve
a
compatibility
issue
with
autopilot
clusters
so
that
we
don't
have
to
worry
about
scaling
them,
at
least
for
the
purposes
of
non-production
use.
A
So
I'll
just
have
to
publish
these
changes
a
little
bit
later,
but
that's
a
little
bit
lower
priority
in
terms
in
relation
to
finishing
the
onboarding
instructions.
So
I
can
get
Rob
and
Max
to
help
out
with
some
infrastructure
tasks
so
yeah.
That
is
that
issue
and
then
yawn.
You
have
the
next
one
for
cube,
coloring,
rendering
for
improving
widget
responsiveness
and
consistency.
C
Yeah,
this
is
one
of
the
two
issues
that
I
took
over
the
other
ones.
The
learner
sure
I
have
not
had
a
chance
yet
to
look
at
it
this
week,
I
will
probably,
after
this
meeting
since
my
other
thoughts
come
out
in
review.
A
Got
it
thanks
for
the
update
for
you
guys
too,
and
then
run
you
have
this
issue
here
with
replacing
product
analytics
dashboard
listing
with
new
shared
dashboards
listing
at
the
project
level.
D
Yeah
I'm
having
some
R-Spec
fun
times
a
bit
alternating
between
passing
and
failing
pests,
so
that's
always
fun
so
I'm
trying
to
debug
that
yeah
still
working
on
it
still
making
good
Headway
and
you
spread
it
out.
Let's
start
getting
reviews,
so
if
I
don't
make
too
much
more
Headway
on
this
aspect,
then
I'm
going
to
start
doing
that,
so
I
can
focus
on
the
getting
reviews.
While
trying
to
fix
these
aspect
bugs
got.
A
C
C
So
yeah.
This
is
in
review
with
Rob,
we'll
see
what
he
has
to
say,
and
hopefully
not
so
much
but
yeah.
A
Exciting
no
I
think
that'll
be
also
nice
just
for
us
when
we're
testing
things
when
you
know
that
Source
data
gets
old
and
we
want
to
look
back
at
a
different
time
frame
and
just
to
copy
and
paste
that
that
parameter.
A
So
thanks
for
that,
we
also
have
funnel
analysis,
which
Max
I
could
probably
just
speak
to
this,
where
we,
you
need
to
verify
the
changes
here,
but
that's
kind
of
blocked
on,
like
the
cluster
being
able
to
publish
those
changes
there,
having
some
issues,
publishing
it
to
the
production
cluster
and
not
not
wanted
to
disrupt
the
product.
The
product
production
dashboards.
Excuse
me,
so
that's
that's
effectively
blocked
on
me.
A
There
there
are,
of
course,
like
the
following
implementation
issues
that
need
to
be
created,
but
but
it
comes
after
verifying
that
it
all
works
in
the
where
we
need
it
to
work.
Max
that
I
miss
anything
there.
B
Nope,
that's
it
I'm,
pretty
sure
it
works,
but
yeah
we
we
can
test
it.
I've
got
some
I've
got
a
couple
of
test
projects
now
that
have
defined
funnels.
So
once
the
changes
are
available,
we
can
test
that
they
work
got
it
cool.
A
B
Yeah,
it's
the
sort
of
the
other
half
of
it.
Sorry,
the
other
half
of
it
from
Jitsu
to
track
events
which
is
now
done
in
production
and
works.
Fine.
D
Very
simple,
replace
the
term
widget
with
panel
across
our
documentation,
schema
backend
front
end
everywhere,
so
don't
use
the
term
widget
please.
From
now
on.
I
beg
of
you.
A
B
No,
thank
you.
Sorry
will
I
go
away.
No!
Sorry,
just
that.
It
now
works.
We've
sort
of
got
a
a
hack
in
there
so
that,
instead
of
joining
the
trapped
events
table
to
sessions
table
now
the
sessions
Cube
itself
just
queries
the
database.
So
there's
no
need
to
join,
which
makes
the
queries
faster
anyway,
so
yeah,
it
should
now
all
be
functioned
as
expected.
B
C
Something
I
also
like
to
call
out
there
is
I've
created
a
documentation,
update
Mr
to
note
that,
when
occurring
or
measuring
track
events
or
sessions
that
you
can't
mix
the
dimensions
with
either
so
using
sessions,
you
have
to
use
session
Dimensions.
C
This
is
also
something
that
I
think
going
forward
like
in
our
visualization
designer
that
we
should
make
sure
that
that
is
applied
and
that
if
there
is
a
find
some
way
of
like
doing
it
in
our
schema
or
one
way
like
with
Json
schema
version,
number,
seven
there's
actually
a
new
feature
coming
in,
where
you
can
add,
like
ifs
to
schemas.
C
So
we
can
like
actually
track
like
if
someone
has
in
their
query
configuration
sessions
that
we
can
like
test,
that
they
don't
have
any
track
events
dimensions
and
vice
versa,
but
yeah.
This
is
something
that
I
just
wanted
to
call
out
for
people
to
be
aware
of
When,
developing
new
features
and
yeah
also
something
that
I've
documented
for
ourselves
and
our
users.
A
C
Here,
no,
please,
it's
live
on
the
internal
handbook.
Please
go!
Try
it
out,
try
to
break
it
if
you
find
any
way
to
break
it.
Please
let
me
know.
A
Great
then,
I
think
that
about
wraps
up
the
billboard
review
unless
anything,
if
there's
anything
else,.
E
C
A
E
Right
thanks
so
I
wanted
to
call
out
that
we
finished
our
final
plan
for
1510,
so
I
wanted
to
walk
through
that
briefly,
there's
a
link
to
the
issue
there
if
you'd
like
to
follow
along
at
home.
E
So
this
is
going
to
be
running
us
from
the
18th
to
the
17th
of
March
and
really
what
I
wanted
to
highlight
as
our
Milestone
objectives
for
1510
is.
This
is
going
to
be
the
one
where
we
complete
the
internal
preview.
Epic,
so
I'm
really
excited
about
this
as
a
goal,
because
we've
been
working
in
there
for
so
long
and
we're
at
the
point
where
it
seems
realistic
to
actually
Mark
that
thing
as
close,
which
is
super
exciting.
E
Additionally,
we
want
to
get
a
back-end
POC
completed
for
funnels
if
that's
not
already
done
and
also
get
our
plan
finalized
for
What
shipping.
That
NBC
funnels
might
look
like.
As
a
second
theme,
you
can
see
some
of
the
issues
that
we're
going
to
be
potentially
carry
over
from
15.9,
depending
on
how
progress
goes
over
the
next
week
or
so,
and
then
our
key
feature
item
is
again
going
to
be
that
internal
preview,
epic
I,
won't
go
through
all
the
individual
issues,
but
they're
marked
on
1510.
E
So
if
you
filter
or
look
at
the
board
for
15,
10
you'll
see
them
there.
After
that,
these
are
going
to
be
stack,
ranks
like
we
always
do
so
work
top
to
bottom,
please,
but
we
want
to
do
this
one
it's
marked
in
15
9,
so
this
might
not
actually
need
to
be
here.
We
have
an
issue
about
adding
custom
events
in
the
internal
handbook
right
now.
E
E
After
that,
we
want
to
start
working
on
the
column
chart
visualization.
This
is
a
really
foundational
way
to
look
at
data,
especially
data
that
is
non-continuous,
which
is
what
a
lot
of
people
use,
LINE
charts
for,
and
that's
not
really
what
you
should
be
using.
So
this
is
really
going
to
be
a
key
visualization.
We
expect
users
to
use
quite
frequently.
D
E
So
just
funnels,
so
whenever
we
talk
about
Sankey,
we
should
all
think
of
that
as
a
completely
separate
sort
of
visualization
versus
a
funnel.
So
a
funnel
is
really
intended
to
be
a
single
input
and
a
single
output
for
a
data
flow,
whereas
a
Sankey
diagram
is
a
multi-source
multi-sync
data
flow,
so
you
can
have
data
coming
from
multiple
places
going
multiple
places.
E
A
Sankey
diagram
really
could
be
visualized
as
a
superset
of
a
funnel,
but
we
don't
need
to
approach
it
that
way.
We
should
start
with
just
the
funnel,
because
it's
the
the
simpler
one
people
understand
that
more
I
personally
really
like
sankies
but
I,
know
a
lot
of
people
find
them
confusing.
That
might
be
something
we
don't
even
get
to
for
a
long
time.
If
our
users
can't
think
of
something
actionable,
they
want
to
actually
do
with
it.
So
we
should
think
about
those
as
separate
things.
E
F
One
important
thing
here
is
that
this
was
already
decided
and
discussed
I.
Think
six
eight
weeks
ago,
with
the
foundations
team,
we
are
not
doing
kitlab
UI
hops
for
charts.
We
are
going
to
directly
use
them.
There
is
a
chart
component,
but
you
can
access
the
e-chart,
and
this
is
the
preferred
way.
I
need
to
check
if
the
foundation
team
has
already
updated
the
docs
on
this
so
far,
but
we
are
not
going
to
re-implement
now
one
chart
after
the
other
from
e-charts
into
Kettle
API
and
then
use
it
in
gitlab.
F
We
are
always
going
to
use.
There
is
a
specific
gitlab
UI
chart
component,
which
simply
has
access
to
all
the
e-chart
stuff,
and
this
is
basically
the
way
forward.
They
are
even
thinking
about
replacing
the
old
launchers.
This
has
created
huge
overhead
and
didn't
help
at
all,
so
we
can
go
directly,
skip
gitlab,
UI
and
use
it.
There.
F
Yes,
they
have
already.
There
is
a
component
on
gitlab
UI,
which
is
simply
I,
think
it's
called
charts,
which
basically
is
simply
the
wrap
around
the
charts,
but
I
I
will
double
check
tomorrow
if
they
already
update
the
doc,
because
the
dock
currently
says
create
a
chart
component,
but
this
is
horrible.
This
is
a
yeah.
There
is
a
chart
component,
which
is
simply
gives
you
one-to-one
access
to
each
others.
E
So
if
anyone
does
have
any
other
additional
comments,
questions
feedback
always
happy
to
have
an
async
discussion
as
well
feel
free
to
take
a
look
at
this
when
you
get
time
so.
I
also
have
the
next
point.
I
wanted
to
signal.
Booster
merge
request
that
I
put
in
recently
now
that
we
are
getting
close
to
wrapping
up
what
we've
been
calling
the
internal
preview
up
to
this
point,
it's
going
to
be
time
to
start
working
what
we've
been
calling
the
customer
preview,
but
customer
preview
is
not
a
commonly
used
term
in
gitlab.
E
It
means
a
lot
of
different
things,
depending
on
who
you're
talking
to
the
context
of
the
conversation,
and
so
this
merge
request
really
tried
to
replace
and
clarify
that
into
release
phases
into
something
that
should
be
more
aligned
with
what
our
gitlab
terminology
is
around
Alpha
closed
beta,
open,
Beta
GA,
as
well
as
a
first
external
use
release.
E
So
this
merge
request
is
the
first
step
here.
I
haven't
gotten
around
to
update
epics
and
issues,
but
I
will
be
doing
that
to
reflect
what
is
in
this
merge
request.
So
that's
the
the
background
context.
If
you
see
those
changes,
I
also
want
to
point
out.
This
does
not
necessarily
directly
correlate
to
a
specific
Milestone.
We
don't
have
a
milestone
attached
to
Alpha
closed
beta,
open,
open
Beta,
yet
we
will
get
there,
but
this
merge
request
is
not
proposing
those
at
this
time.
F
E
D
Yes,
cut
tangently
related
to
it,
just
because
we've
had
previous
conversation
in
the
analytic
session
meeting
around
replacing
Jitsu
with
snowplow,
which
is
absolutely
fine,
but
obviously
that
has
an
impact
on
self-managed
customers.
If
they're
reviewing
the
stack
to
make
sure
it
meets
all
their
legal
compliance
requirements
or
even
if
it's
on
gitlab.com,
necessarily
just
making
sure
that
the
data's
stored
in
a
way
that's
safe
for
them
and
their
business
practices.
D
So
I
was
just
wondering
if
we're
planning
on
where
we're
planning
on
doing
that
snow,
plow
swap
in
relation
to
this
new
release,
schedule
sure.
E
It's
going
to
take
us
some
time
to
figure
out
who
that
first,
external
user
is
going
to
be
we're
going
to
have
to
identify
that
person
talk
to
their
legal
team,
see
when
we
can
get
Cycles
in
their
development
to
even
do
it,
so
we
will
move
forward
with
starting
those
conversations.
Now,
even
if
we
haven't
completed
the
snowplow
migration,
all
those
will
be
framed
around
hey.
This
is
pre-alpha
everything
is
subject
to
change,
including
the
Jitsu
stuff.
E
We
will
definitely
have
this
buttoned
up
by
GA,
because
we
don't
want
to
be
using
Jitsu
and
GA.
I
would
prefer
that
we
have
it
wrapped
up
before
we
call
the
alpha
complete,
but
definitely
what
we're
calling
beta,
which
is
essentially
feature
complete,
will
be
snowplow,
snow
plowified
as.
E
D
D
A
F
Of
the
advantage
that
we
yeah
okay,
like
a
big
advantage,
that
we
will
get
from
snow
plows
that
we
might
even
can
have
a
customer
who
doesn't
add
tracking
but
rather
gives
us
data
slices
from
S3.
F
So
this
is
definitely
something
where
snow
plow
can
can
help
us
a
lot,
because
this
might
mean
that
they
don't
even
need
to
inject
directly
with
the
collector.
But
the
collector
gets
the
data
done
from
somewhere,
so
or
snowplow
sends
it
out
their
existing
environment
and
sends
it
to
our
staff
plus
their
existing
stuff,
and
they
basically
can
run
that
stuff
side
by
side.
So
this
is
this
is
a
huge
flexibility
Advantage
compared
to
to
previously
so.
A
Yeah
and
part
of
I
think
Global
Health
in
terms
of
actually
determining
when
we
can
actually
switch
over
is
putting
together
the
plan
which
sound,
busty
and
I
are
currently
working
on
because
of
the
different
strategies
we
have
with.
You
know:
gitlab
managed
environments
and
private
environments
as
well
as
self-managed,
and
so
you
know
vastly
had
alluded
to
before
in
this
proof
of
concept
that
we
could
potentially
just
replicate
our
existing
pipeline
that
we're
using
for
service
ping,
and
that
gets
us
to
our
first
customer
very
quickly.
A
But
then
you
know
we
have
to
just
like.
First,
we
need
to
have
that
plan,
so
everyone
is
aware
of
it
and
understands
it,
and
then
we
can
figure
out
how
that
that
matches
with
our
overall
strategy.
But
in
addition
to
that,
that'll
also
help
us
determine
the
migration
strategy
of
switching
over
from
Jitsu
to
Tim's
point
of
how
we
can
ingest
stuff
from
S3
or
we
can
just
switch
over.
A
You
know,
like
the
actual,
a
records
itself,
so
there's
no
change
needed
from
customers
that
are
using
it
using
a
sending
events
already
so
that
it
just
goes
over
to
snowplow,
and
then
we
just
have
to
backfill
with
the
existing
data.
Once
we
have
that
updated
because
we'll
have
to
update
the
cubes
schema,
have
to
transform
the
previous
data
so
that
it
all
fits
and
has
it's
one
complete
set.
But
all
that
just
needs
to
be
discussed
and
we
just
need
to
put
the
plant
together.
A
And
for
that
to
happen,
we
just
have
to
know
what
is
our
kind
of
a
near-term
roadmap
for
for
snowplow
and
how
that
lines
up
with
how
that'll
fit
in
with
our
phase
releases.
Here.
A
All
right,
then,
it's
show
and
tell
time
Young.
C
Yeah
not
much
to
say
about
this:
it's
already
covered
in
the
close
issues,
but
I'll
repeat
yeah.
Please
do
go
click
on
that
link
and
try
and
break
our
UI,
and
let
us
know
if
we
need
to
create
any
bug
issues.