►
From YouTube: 2022-07-18 Analytics Section 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).
B
All
right,
hello,
everyone
welcome
to
the
first
I
believe
this
is
the
first
official
meeting
for
the
analytics
section.
Hopefully,
everyone's
got
the
agenda
open.
We've
got
a
bit
to
cover,
but
yeah
I
just
wanted
to
welcome
everyone.
Thanks
for
the
Emoji,
appreciate
the
support
so
yeah
as
far
as
just
kind
of
housekeeping
items
I
wanted
to
let
everyone
know
that
we
have
a
new
group
for
the
section
links
included
in
the
agenda.
B
I
know
there
are
I,
know,
product
intelligence
has
had
their
own
group
and
stuff
organized
so
Elena
I
mentioned.
If
you
wanted
to
move
things
over
seems
like
we
could
just
go
ahead
and
do
that
right.
C
Yeah
I
think
so.
I
was
just
looking
at
the
private
reports
and
I
was
a
bit
concerned
to
not
make
them
accidentally
public
without
making
sure
that
everything
is
okay
there.
But
oh
I'll
take
a
deeper
look
and
add
my
notes
in
the
issue
and
posture
there:
privacy.
B
Should
be
retained,
I
I'm
not
able
to
actually
create
anything.
That's
not
everything
I
create
under
the
section
is
private
I.
Don't
even.
A
B
Option
for
for
creating
public,
so
that
should
be
okay,
but
we
could
we
could
test
it
out
if
we
need
to
cool
awesome,
so
it'll
be
good
to
have
everything
in
one
place:
I'll
be
working
on.
D
So,
just
to
realize
one
thing,
I
think
the
question
to
ask
would
be:
is
going
to
be
git
lab
maintainers,
who
are
members
of
the
gitlab
org
project
who
aren't
get
that
invoice.
So
that
would
be
the
one
group
that,
even
if
that
group
is
private,
would
still
have
access
to
that
project
that
we
just
need
to
clarify.
I.
Think
that's
okay,
but
I!
Think
just
walking
through
that
logic.
That's
the
group
that
we
should
ask
the
question
about.
B
E
First
of
all,
although
hello
to
everyone
and
thanks
for
joining
I,
just
wanted
to
give
you
a
brief
update,
so
Alina
is
still
acting
em
and
is
doing
a
fantastic
job
there
thanks
again
for
your
all
your
support,
but
we
are
also,
at
the
same
time
trying
to
hire
a
new
engine
management
for
that
role
and
are
currently
in
the
final
steps
for
making
an
offer.
E
I
really
hope
that
we
can
get
this
out
of
the
door
in
the
next
24
hours,
so
everyone
cross
your
fingers,
but
then
we
should
have
a
really
nice
em
coming
in
for
this
team
and
as
soon
as
I
have
everything
signed.
Hopefully,
then
I
will
basically
give
you
another
update,
what
the
timeline
looks
like
and
who
the
person
is
yeah
over
to
you.
Dennis.
B
Yeah,
so
we
finally
thanks
to
Tim,
got
the
positions
and
Rex
approved
from
all.
You
know:
Finance
recruiting
all
that
stuff,
so
our
product
analytics
is
looking
forward
to
back
in
engineers
and
two
front-end
Engineers
as
well.
So
the
internal
job
postings
are
live,
we've
had
some
interest
internally
and
so
we're
we're
receiving
those
which
is
great
but
at
the
same
time
I'm
working
on
the
job
description
so
that
we
can
get
the
public
job
postings
out.
And
so
that's
my
goal
for
basically
as
soon
as
possible
this
week.
B
But
by
the
end
of
this
week,
we'll
at
least
have
the
public
posts
for
us
to
start
accepting
applicants
from
that
time
and
then
it's
on
to
Kenny
yeah.
D
That's
me
yeah,
so
the
PM
for
product
intelligence
was
just
opened
last
week,
that's
Amanda's
backfill
and
the
PM
for
product
analytics.
It's
been
open
for
I
think
about
three
weeks.
We've
got
three
candidates
in
the
pipeline.
D
Some
of
you
might
be
on
interview
panels
for
those
candidates,
so
I
think
one
was
in
interviews
last
week,
I'm
going
to
also
drop
the
job
family
links
for
the
like
gitlab
handbook,
content
of
the
job
posting
and
then
I'll,
see
if
I
can
dig
up
the
Greenhouse
more
public,
where
you
can
directly
apply
links
to
but
they're
both
open
internally
and
externally.
I'd
love
to
have
internal
candidates
for
either
one
of
those
positions.
D
E
Yeah,
so
I'm
really
excited
to
share
a
couple
of
things
today
that
we
have
been
working
on
in
the
analytics
team
over
the
last
couple
of
weeks.
E
So
you
might
be
aware
of
the
original
plan,
which
was
to
basically
integrate
post
talk,
which
is
like
an
open
source
analytics
solution,
and
that
was
the
original
plan
or
basically
Mission
briefing
to
see
that
we
can
get
this
included
imported
introduced
to
gitlab.
To
some
extent.
This
was
basically
stopped
very
early
due
to
a
couple
of
limitations,
including
license
and
problems
there.
E
So
what
we
did
back,
then,
is
take
another
look.
What
are
the
options
that
we
have
out
there
that,
where
we
can
jump
start
basically
the
whole
topic,
but
at
the
same
time
not
start
completely
from
scratch
and
to
the
other
side?
E
Also,
please
don't
have
something
that
is
basically
done
directly
in
gitlab
that
people
can
use
and
where
we
can
actually
rethink
to
some
extent
the
whole
analytics
space
from
the
perspective
that
we
are
doing
so
we
went
ahead,
did
a
lot
of
looking
at
different
projects,
pocs
open
source,
libraries
and
so
on,
and
today
I'm
quite
proud
to
show
you
a
couple
of
things
that
are
so-called
Vision
POC,
which
means
that
this
should
share
with
you
the
vision
where
we
want
to
go,
and
some
things
are
working
already,
which
is
also
nice
to
Simply,
see
how
the
technology
Works
how
easy
it
is
to
integrate.
E
On
the
other
hand,
some
things
are
completely
fake
and
are
basically
showing
you
just
what
we
want
to
accomplish
over
the
next
couple
of
months
and
on
one
hand
with
analytics
team
but
tightly
related
to
the
product
intelligence
team,
and
this
is
something
that
I
think
we
can
see
and
transform
over
the
next
couple
of
weeks
and
months
is
where
do
those
two
teams
basically
handled
in
the
different
topics,
both
on
one
hand,
starting
to
dog
food
as
soon
as
possible?
F
E
So
what
do
we
have?
I
was
looking
at
a
couple
of
different
libraries
and
setups
that
we
were
that
are
around
in
the
whole
realm
of
analytics
and
also,
to
some
extent,
take
a
look
at
a
lot
of
the
different
products
that
are
on
the
market.
What
advantage
do
they
have,
but
also,
on
the
other
hand,
What
complications
do
they
have
or
what
problems
do
they
have
in
our
days
with
the
architecture
that
they
might
have
chosen?
E
A
couple
of
years
ago,
then
I
talked
to
a
lot
of
you
in
the
product
intelligence
team
took
a
lot
of.
It
looks
at
different
issues
and
road
maps
and
everything
that
was
implemented
and
especially
taking
a
look
on
all
the
hard
topics
that
we
currently
have
getting,
basically
trekking
into
the
product
on
one
hand
and,
on
the
other
hand,
getting
something
out
of
that
data
and
making
it
visualized.
E
Nowadays,
a
lot
of
analytics
products
have
the
problem
that
they
have
data
in
a
SAS
solution,
which
means
that
they
have
already
the
problem
that
they
can't
determine
very
easily
where
the
data
is
stored.
So
there's
a
huge
topic
going
on
in
the
European
Union
around
is
Google
analytics
legal
or
not
data
privacy
and
so
on.
E
So
the
idea
is
that
that
we
basically
provide
them
with
so-called
tracking
components
that
they
can
go
ahead.
As
you
can
see,
okay,
I'm,
coming
in
I
want
you
to
use
kitlab
for
tracking.
E
And
they
we
set
for
them
up
a
tracking
component
I
think
we
need
a
good
name
for
it.
It's
in
reality.
It's
set
up
either
Docker
or
kubernetes,
which
consists
of
three
components.
E
So
what
the
idea
is
quite
simple,
so
you
go
in
on
your
workspace
and
say:
I
want
to
start
tracking,
and
you
can
say
Okay
I
want
to
start
a
new
tracking
component
in
which
data
storage,
so
in
which
data
region
you
can
start
in
the
US,
so
you
can
start
in
Europe
and
we
go
ahead
and
basically
set
up
an
instance
or
a
mesh
which
contains
three
Technologies
on
one
hand,
G2,
which
is
a
toolkit
for
Data
Tracking
and
data
transformation,
which
contains
a
couple
of
other
sub
libraries.
E
As
soon
as
you
have
clicked,
okay,
I
go
ahead.
I
come
in
I
want
to
select
the
use
case.
I
want
to
do
that.
Mobile
analytics
I
have
no
own
data
store,
so
we
are
going
to
set
up
clickers
for
this.
You
go
ahead
and
set
up
a
region,
and
we
will
then
basically
the
next
step,
go
ahead
and
do
a
setup
in
our
Cloud
environment
to
set
up
an
instance.
Why
is
this
important?
E
First
of
all,
the
whole
first
person
story,
but
on
the
other
hand,
what
we
can
provide
here
is
we
have
tons
of
self-managed.
The
problem
with
self-managed
is
that
self-managed
most
of
the
time
is
not
100
available.
So
if
you
want
to
do
tracking
with
your
your
application,
this
would
mean
that
your
application
needs
a
100
data
connection,
all
the
time
to
your
gitlab
instance,
which
is
a
very
hard
thing
to
do.
So.
E
The
idea
is
to
basically
deploy
this
into
a
into
a
data
store
into
a
cloud
service,
so
that
also
self-managed
can
go
ahead
and
say:
okay,
I'm
connecting
my
application.
There
I'm
connecting
myself
managed
to
the
tracking
instance,
and
we
can
both
get
there
and
get
the
data
and
have
there
the
uptime
that
we
need
for
the
whole
tracking
story.
E
On
top
of
that,
this
opens
up
a
really
nice
business
model,
which
is
very
much
directed
because
that's
the
hard
part
for
analytics
companies
is
the
more
events
that
you
basically
track,
the
harder
it
becomes
and
the
more
expensive
it
becomes
the
whole
story.
So
this
would
be
very
hard
to
say.
Okay,
free
tier
can
give
you
how
many
events
and
so
on.
So
this
means
really
that
you
have
a
expense
that
is
based
on
the
amount
of
events
that
we
are
going
to
trade.
For
you.
E
This
is
the
main
tracking
part
classic
web
tracking
app
event
tracking.
You
can
do
everything
with
it.
It
has
already
tons
of
different
sdks
included
that
we
can
build
on
top
of
it.
The
other
thing
that
I
saw,
especially
that
we
are
struggling,
that
all
the
others
are
struggling,
is
that
we're
trying
to
some
extent
do
counts
for
things
that
we
anyhow
have
in
our
user
in
our
production
database.
So
we
have
in
the
production
database
how
many
projects
do
we
have?
E
How
many
namespaces
do
we
have,
on
the
other
hand,
almost
every
product
or
web
product
out?
There
has
some
sort
of
users
groups
to
some
extent
or
different
membership
statuses,
so
something
that
is
modeled
very
closely
connected
to
that.
So
the
idea-
and
that's
something
that
Jitsu
is
able
to
provide
us-
is
to
do
a
data
transformation
pipeline
so
that
you
are
not
only
starting
this
tracking
box
with
just
trekking
ahead,
but
that
you
basically
tell
the
tracking
the
tracking
component
hey.
This
is
my
data
storage
that
stores
all
my
production
data.
E
Please
also
ingest
from
this
data
source.
Also,
these
tables
this
information
Etc,
and
we
can
do
this
very
easily,
because
Jitsu
is
basically
providing
us
this
under
the
hood.
Whereas
there's
a
system
called
air
byte
which
is
providing
tons
of
different
data
transformation
pipeline.
So
a
classic
example,
it
even
Imports
gitlab
by
the
way,
but
you
can
go
ahead
any
sort
of
database.
You
can
basically
import
all
the
time,
but
you
can
also
have
Shopify
WordPress,
all
the
classic
things
that
are
basically
going
along
to
your
product,
where
you're
going
to
track
stuff.
E
We
should
have
a
better
understanding
what
we
might
be
able
to
connect
there,
so
that
you
can
then,
and
the
product
manager
can
go
ahead
and
can
segment
at
a
later
Point,
hey
I
want
to
see
only
analytics
for
ultimate
users.
I
want
to
have
everyone
that
is
basically
part
of
our
project
group,
that
is
more
than
a
thousand
people,
and
so
on.
This
gives
us
way
more
possibilities
that
we
wouldn't
have,
or
basically
all
the
time
do
in
a
very
different
and
complex
way
to
connecting
those
two
data
storages.
F
E
So
that's
the
other
idea
behind
it
is
that
we
can
go
ahead
and
import
into
the
same
actually
cow's
data
storage,
also
your
product
data
and
now
I'm
happy
to
show
you
a
little
bit
what
we
are
already
able
to
do
with
this,
and
what
we
have
in
the
same
tracking
component
is
also.
We
have
a
cube.
Def
instance
run
Cube
def
is
basically
our
cube.js
I'm,
not
sure
100.
E
What
the
naming
is
for
videos,
it's,
basically
a
business
intelligence,
open
source
Library,
it's
a
complete
setup
that
provides
you
with
different
apis,
different
front-end
components,
tons
of
data
analysis
topics
and
has
really
nice
Integrations,
so
that
you
can
basically
connect
it
from
your
application.
But
we
also,
we
would
be
also
able
to
support
from
day
one
all
the
funky
bi
tools
that
are
ahead
of
it,
so
that
all
the
data
scientists
that
you
still
have
in
your
team
can
could
still
work
with
the
data.
E
So
it
supports
Apache,
superset
and
tons
of
other
bi
tools
that
are
around
to
connect
to
your
own
data
and
that's
the
main
part
of
the
story
that
you,
as
a
company
or
own.
This
data
directly
and
no
one
else
in
between
good.
What
do
we
have
based
on
Cube
Dev,
is
that
you
have
a
very
easy
query:
language
that
you
can
build
components
with,
and
you
basically
set
up
in
the
playground.
Take
that
Implement
take
it
to
the
view
component
and
you
have
very
easily
and
very
fast
those
queries
in
your
product.
E
So
this
is
already
live
data.
You
can
see
that
this
is
basically
mainly
run
through
the
aspects
that
are
running
against
the
test,
instances
which
are
then
cooked,
creating
tracking
data
all
the
time.
So
this
is
why
we
had
already
41
000
sessions
of
headless
Chrome
in
there
and
working
on
on
that
platform,
basically
and
providing
us
with
data.
E
The
big
topic-
and
that
is
something
I
saw
that
you
did
in
product
intelligence,
which
I
think
is
a
very,
very
good
thing
for
nowadays.
Web
application.
I
think
this
is
a
perfect
pattern,
is
to
not
only
have
you
have
on
one
hand
the
URL,
which
is
the
classic
pattern
that
every
tracking
page
out
there
has
and
is
tracking
against.
But
we
also
have
this
page
identifier
or
page
name
and
I,
think
this
makes
it
makes
much
more
sense
nowadays,
because
you
have
a
view
or
a
page.
E
That
is
basically
then
used
and
represented
in
tons
of
different
URLs,
and
then
you
have
a
very
hard
time
going
back
in
time
and
say:
okay,
all
those
three
thousand
seven
hundred
us
that
you
see
here
is
all
just
one
view,
but
I'm
I'm
ready
most
of
the
time
brought
interest
in
that
view,
so
default
process
was
let's
take
this
and
make
this
a
first-class
citizen,
so
that
we
are
not
that
we
can
go
ahead
and
basically
a
product
manager
can
go
in
and
can
say
exactly
the
the
tracking
stuff
that
you
that
you
are
using,
because
what
I
did
here
is
I
integrated
the
teacher
tracking
code
directly
into
gitlab
the
instance.
E
E
That
is
happening
in
the
groups
realm
and
we
can
go
ahead
and
do
also
Stars
so
that
you
can
say
everything
in
groups
and
below
and
we
can
build
full
trees
so
that
product
managers
especially
can
go
ahead
and
can
grow
and
say
just
give
me
that
information,
because
I'm
not
interested
in
the
whole
package,
but
rather
I,
want
to
have
a
clear
focus
on
something
the
other
big
part
is,
and
this
is
classic
demo
time.
I
need
to
reset
the
view.
Otherwise,
it's
not
going
back
to
the
original
data.
E
E
For
sure
the
other
thing
is
that,
as
said,
what
we
can
do
is
with
the
connection
between
tracking
data
and
product
data
is
we
can
track
and
segment
based
down
on
premium
uses
ultimate
users
and
so
on,
and
can
basically
go
ahead
and
segment.
The
data
against
that.
E
E
Full
time
against
Cube
and
the
stuff
that
we
can
see
here
now,
what
the
idea
is
quite
clear,
I
think
tracking.
Nowadays
we
are
a
skit
lab.
We
have
this
one
product
for
everything
story.
So
why
shouldn't
we
simply
go
ahead
and
not
trade
page
views,
events
clicks
and
stuff
like
that,
but
something
that
we
already
do
in
the
product
is
we
do
error
tracking.
So
this
is
something
I
strongly
believe,
has
a
big
impact
on
everything
that
is
regarding
analytics.
E
E
Against
this,
so
what
you
can
see
here
is
basically
in
a
session
replay
and
I
would
love
to
show
you
the
test
data
that
I've
provided,
because
what
we
can
do
is
already
in
this
test
instance
and
I
think
the
data
was
already
pushed
out.
We
can,
we
can
track
the
web
page
performance.
So
what
we
are
able
to
do
there
is
we
can
track.
E
How
long
did
each
page
take
loading
and
we
can
basically
highlight
okay,
those
pages
very
slower,
that's
where
we
had
a
funnel
drop
off
and
so
on,
and
we
are
already
tracking
front
end
exceptions,
so
you
can
see
basically
in
the
session
recording
are
that's
basically
one
of
the
pages
that
had.
E
And
you
would
see
an
error
in
here
in
between
and
you
can
see
the
different
clicks
and
all
the
events
that
we
are
already
tracking
through
the
cruel
implementation
and
basically
simply
sending
that
over
to
Jitsu,
and
that
gives
you
much
more
insight
and
not
only
for
the
product
manager,
but
also
for
us
as
engineers
and
and
you
X
researchers,
because
you
get
full
insight
into
a
full
session
recording
here,
taking
it
from
there.
What
you
can
also
do
with
G2
based
data
is
you
can
do,
of
course
funnel
analysis?
E
This
is
also
something
we
can
transform
to
through
the
data
platform,
so
we
can
basically
Define
different,
funnel
steps
and
other
interesting
parties.
As
said,
we
have
all
the
information
imported
now
into
our
clickhouse
database,
which
is
basically
our
data
warehouse
here,
and
this
is
also
something
you
can
query
against,
so
you
can
go
ahead
and
say:
okay,
I
want
to
see
a
list
of
all
the
gitlab
name
spaces
on
that
test.
E
Instance,
let's
see
if
it
works,
no,
it
doesn't
I
need
to
refresh,
and
you
can
basically
go
ahead
and
start
building
full
queries
and
dashboard
views
against
the
data
that
you
are
basically
creating
because
also
fully
supports
full.
E
A
E
Is
to
take
all
those
views
that
you
can
build
yourself,
you
can
say
add
to
the
dashboard
and
you
stand
and
then
go
ahead
and
basically
get
this
into
dashboards.
Axel
has
worked
on
this
part
and
he
has
already
took
a
couple
of
experiments,
so
you
can
basically
go
ahead
here
and
go
and
start
the
edit
mode.
E
F
F
D
I
have
some
questions.
It's
so
much
Tim,
that's
awesome!
Thank
you.
First
of
all,
I
just
wanted
to
highlight
Tim
and
I
had
discussed
this
last
week
and
I
actually
only
saw
like
a
fraction
of
what
you
just
presented.
So
it's
kind
of
mind-blowing
I
think
this
really
fits
our
vision,
especially
you
know
the
focus
that
Tim
started
on
on
making
sure
that
users
can
control
their
own
data.
D
That's
a
competitive
differentiator
for
us
as
a
product,
but
also
a
kind
of
critical
moment
in
the
product
analytics
space
that
I
know
the
product.
Intelligence
team
is
really
familiar
with
that
more
and
more
users
are
looking
to
and
in
some
cases
required
to
own
their
own
data.
So
I
had
a
question
about
data
pipeline
definition
and
where
it
should
live,
and
Dennis
and
I
had
a
dialogue
about
it.
So,
thanks
for
answering
my
questions,
Dennis,
but
thanks
for
the
verbalize,
your
response.
B
Yeah
I
mean
it's
gonna,
there's
a
lot
of
detail
that
I'll
need
to
cover
in
less
than
the
minutes
we
have,
but
yeah
most
most
of
what
we
Define
here
will
be
able
to
be
defined
in
Version
Control
somewhere.
What's
some
things,
we
will
need
to
look
into
I
know
with
the
JavaScript
transformations
of
Jitsu.
B
Their
API
is
a
little
less
documented,
theirs
but
I'm
sure
there's
a
way
where
we
can
synchronize
that
so
that
you
know,
if
you
have
Transformations
being
version
controlled
in
gitlab,
you
could
control
that
in
Jitsu.
B
But
that's
like
those
are
some
of
the
odds
and
ends
we'd
have
to
figure
out
there,
but
for
most
part
when
it
comes
to
Transformations
or
even
dashboards,
and
things
like
that,
we
want
to
make
sure
that
they're
like
there's
a
way
to
define
them,
because
obviously
we
want
to
be
able
to
have
a
customizable
experience
for
users
as
well.
That'll
be
important
to
support.
D
Yeah
yeah
I
think
just
another
point
that
I'll
highlight-
and
this
has
been
kind
of
lesson
learned
in
the
app
section
in
general-
is
being
able
to
trace
the
lineage
of
something
like
we're.
Building
tools
for
developers.
Developers
want
to
be
able
to
know,
oh,
like
there's
an
easy
button
here,
but
if
it
breaks
that
I
can
like
Trace
what
happened
and
how
to
fix
it
and
where
I
can
contribute
to
fixing
it.
So
that's
really
the
principle
that
I'm
I
want
to
make
sure
we
we
lean
into.
D
We
can
make
it
very
easy
by
default,
but
I
want
to
make
sure
that
yeah
in
case
something
breaks,
there's
like
a
break
glass
option.
Yeah
cool
Tim
I
had
just
one
clarifying
I.
Think
in
the
demo.
You
use
the
word
product
data,
but
the
actual
word
on
the
screen
was
production.
I
think
we're
talking
about
products.
Data
yeah
is
that
true,
yeah
cool
I
can't
remember
where
it
was
I
think
it
was
in
the
onboarding
portion
where
you
said
like
yeah.
D
Do
you
have
production,
yeah
and
then
Tim
you
and
I
had
talked
about
this,
but
I
want
to
highlight
just
for
the
rest
of
the
team.
I
think
this
is
really
exciting.
You
can
create
some
amazing
delightful
experiences.
You
wouldn't
get
in
any
other
products
analytics
capability
that
Beyond
attaching
your
product
data.
You
can
also
like
we
can
by
default,
attach
your
kind
of
software
delivery
or
devops
platform
data.
So
you
can
think
about
things
like
showing
annotations
about
when
releases
happened
or
when
a
feature
flag
was
clipped
right
in
your
analytics
dashboards.
D
Today,
that's
a
super
exciting
for
me,
maybe
I'll,
just
the
last
one
was
kind
of
open-ended
that
we
can
talk
about
Tim
about
on
by
default,
but
I
want
to
make
sure
not
for
the
POC
or
for
initial
dog
fooding,
but
that
we
think
about
what
it
means
so
that
a
developer
doesn't
have
to
before
they
can
start
using
this
spin
up
a
kubernetes
cluster,
so
that
might
be
providing
organizational
wide
kubernetes
clusters
for
tracking
that
others
can
use,
and
then
I
was
super
excited
to
see
the
session
recording
you
know,
I
was
thinking
about
as
a
PM.
D
What
new
capabilities
would
I
be
really
excited
to
get
my
hands
on
and
I
think
that
is
especially
important.
It
gives
you
a
kind
of
qualitative
understanding
of
user
behavior
that
we
don't
get
access
to
today.
D
So
some
of
those
are
comments.
Some
other
questions,
thanks
Tim
for
the
other
questions.
E
For
going
so
for
going
so
long
and
running
short
on
time,
but
yeah
I
think
this
is
this
gives
us
a
lot
of
opportunities
that
we
can
think
of
really
connecting
a
lot
of
thoughts
here,
because
I
I
hated
the
analytics
that
the
sessions
thing
didn't
work
out.
E
Give
me
a
second
and
I
simply
raised
limit
on
the,
because
what
happened
is
that
this
was
now
deployed
to
or
it
was
pushed
to
gitlab,
and
this
is
starting
all
the
pipelines
and
the
pipelines
are
creating
tons
of
sessions,
and
this
is
basically
pushing
out
the
stock
here.
Let
me
see
if
I
can.
I
have
now
changed
the
limit
on
the
query
so
that
we.
E
There
you
can
then
actually
see
that
when
there
was
an
exceptional
race
that
we
already
could
connect
collecting
all
the
same
data
that
we
are
tracking
and
collecting
right
now
already
with
snow
plow,
no
I
think
it
was
pushed
out
completely.
E
I
will
send
this
afterwards,
the
link
to
it,
because
the
target
is
really
to
take
this
and
update
our
instances
there,
due
to
a
cube
instance
today
with
the
updated
schema
and
deploy
exactly
this
branch
that
we
have
here,
that
I
demo
to
date
and
have
this
again
on
the
gitlab
test
instance,
and
then
you
can
play
around
with
it,
as
mentioned
it
breaks
very
easily
so
far,
but
on
the
other
hand,
we
have
a
lot
of
brown
stuff
here
with
puke.
E
B
I'll,
just
briefly
speak
to
it,
but
there's
a
link
included
to
the
repository
if
you
want
to
check
it
out,
but
basically
what
I?
One
of
the
things
I've
been
working
on
is
kind
of
a
a
dev
kit
for
getting
this
tracking
instance
or
whatever
we're
going
to
call
it
Jitsu,
quick
house
and
Cube
into
a
way
that's
manageable
for
us
to
to
also
develop
against
locally,
but
also
that
might
lend
itself
into
how
we
operationalize
it
in
terms
of
deploying
it
onto
kubernetes,
because
it's
a
single
Docker
compose
file.
B
So
we
could
use
compose
and
then
start
building
a
Helm
chart
on
it
to
basically
see
how
it
works
and
test
out
how
we
want
to
do
this
in
terms
of
like
what
you
saw
in
the
onboarding
in
terms
of
how
we
wanted
to
play
this.
But
for
now
it
works
as
a
development
environment.
If
you
want
to
connect
your
GDK
to
it,
so
I'd
love
for
anyone
to
test
it
out
and
tell
me
what
I
haven't
documented
that
there's
already
a
couple
steps
that
I've
missed
but
feel
free
to
check
it
out.
E
E
Next
step
is
defining
now
from
what
we
see
that
we
should
be
capable
of
is
really
Define
the
next
step
to
make
this
reality
as
soon
as
possible,
with
exactly
paths
also
how
to
get
the
real
data
of
gitlab
that
we
are
tracking
already
right
now,
some
segments
out
of
it
or
track
just
groups,
etc.
Those
will
be
some
of
the
next
steps
to
yeah
get
this
into
the
product
as
soon
as
possible.
B
All
right,
then,
thanks
everyone
from
for
going
a
couple
minutes
over
enjoy
the
rest
of
your
Mondays
and
your
week
and
we'll
see
everyone
next
time.