►
From YouTube: 2022-08-01 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).
A
Hello
and
welcome
everyone
to
the
first
day
of
August
and
our
next
analytics
session
meeting
Amanda
you've
got
the
first
point.
A
Yeah
so
last
I
remember:
we
were
in
the
offer
generation
stage.
I
was
actually
trying
to
help
DL
in
terms
of
in
terms
of
some
of
the
numbers
there,
because
Tim
is
out
I'd
already
pinged
her
to
see
where
we're
at
with
the
offer
generation
because
I
it
looks
right
now
that
we
haven't
gone
through
the
approval
chain,
and
so
we'll
definitely
want
to
do
that.
So
I'll
I'll
follow
up
with
you
on
more
with
more
details
as.
A
I
can
thank
you
cool,
then,
while
I
read
through
this
wall
of
text,
this
is
a
good
opportunity
for
anyone
else
to
add
anything
to
the
agenda
if
they
would
like
to
just
have
a
couple
of
different
updates
for
product
analytics.
So
right
now
outlining
the
internal
preview
of
product
analytics.
That
is
the
first
iteration
of
a
larger
epic
that
I
started,
but
then
had
Sam
fully
build
out
and
got
a
lot
of
good
feedback
from
Sam
and
Kenny.
A
Regarding
that
so
feel
free
to
check
all
of
the
epics
out,
there's
tons
of
links
in
here
tons
of
reading
material,
but
basically
for
the
first
iteration.
We
want
to
make
sure
that
we
get
our
tracking
stack
converted
over
to
kubernetes
and
deployed
into
our
gcp
environment.
That
way
we
can
actually
start
instrumentation
for
gitlab.com
scoped
to
get
lab.org
to
start
getting
some
production
data
and
also
see
how
that
will
scale,
as
we
continue
to
ramp
that
up
once
we
have
that
we'll
have
the
production
data
that
we
we
need.
A
So
you
start
building
out
what
we
would
think
is
kind
of
a
good
first
iteration
for
a
dashboard
and
start
getting
that
in
front
of
an
internal
GitHub
team
members
to
see
what
their
feedback
is.
But
we
will
also
be
dog
fooding
it
ourselves,
because
once
we
collect
instrumentation
on
it,
we
should
see
how
people
are
are
using
it.
So
we
get
some
immediate
dog
food
happening
there,
but,
like
I
mentioned
there's
a
lot
of
links
here.
I
think
we
have
a
really
good
plan
for
I.
A
Think
we
have
the
next
seven
iterations
kind
of
detailed
out.
They
definitely
get
a
bit
more
vague.
A
The
further
we
go,
but
it's
at
least
a
really
good
plan
for
the
short
term
and
then
we'll
continue
to
flesh
that
out,
as
time
goes
by
I,
just
want
to
say,
give
a
special
shout
out
and
thanks
to
Sam
for
creating
all
the
other
option,
epics
and
issues
and
assigning
those
to
me
as
well
as
the
kubernetes
fund
we've
been
having
last
week,
I
think
we've
made
a
lot
of
good
progress
and
honestly
I
think
we
should
be
able
to
get
in
there
this
week.
A
So,
let's
see
where
we
go
like
I
mentioned
my
plan
for
this
week
is
to
get
that
stack
working
in
gcp
I
have
to
remember
how
we
make
changes
to
our
content
security
policy
because
I
believe
that's
controlled
by
cloudflare.
Otherwise,
the
instrumentation
won't
work
but
yeah.
It's
all
looking
good,
there's
an
epic
specific
to
the
tracking
stack.
If
you
all
would
like
to
follow
along
other
updates
include
hiring.
So
we
have
those
two
Recs
open
one.
First
senior
back-end
engineer,
one
for
senior
front-end
engineer.
A
A
So
I'm
really
excited
to
have
more
more
help
there
on
the
analytics
side
of
things
and
then
in
case
you
missed,
it
have
a
couple
of
other
links.
You
probably
saw
these
in
the
slack
Channel,
but
just
kind
of
wanted
to
signal
boost
these.
We
have
some
interesting
user
interviews.
It's
been
interesting
to
see
the
different
experiences
people
have
been
having
with
product
analytics.
A
So
if
you're
curious
about
how
others
are
using,
it,
then
yeah
check
those
links
out
and
then
finally,
Sam
has
made
a
number
of
edits
to
our
Direction
page
to
clarify
use
cases,
we're
not
pursuing
different
personas
and
a
number
of
other
different
edits
that
were
announced
in
the
channel.
I
just
wanted
to
call
that
out.
If
you
wanted
to
re-review
that,
but.
C
And
on
those
we're
definitely
still
refining
those
personas
and
use
case.
So
if
you
have
questions
or
comments
or
feedback,
please
share
them.
So
we
can
continue
to
iterate
on
that,
because
I
put
a
lot
of
these
Mrs
together
as
I,
was
getting
up
to
speed
and
had
some
questions
on.
How
do
these
differ
from
our
normal
git
lab
as
a
whole,
personas
and
I?
B
B
What
tool
are
you
using
for
that
for
specifically
for
the
instrumentation?
Is
this
new
plow?
Is
this
reddis
like?
What
do
you
use.
A
Tools,
the
right
word
yeah,
so
it's
going
to
be
the
Jitsu
JavaScript
library,
that's
going
to
be
hopefully
plugged
into
our
Jitsu
sorry
Jitsu
rolled
out
onto
gcp.
That's
part
of
the
work,
because
we
want
to
understand
how
all
this
is
scaling
to
production
traffic
as
well.
That's
that's
another
side
of
this.
In
addition
to
building
the
internal
preview,
we
want
to
make
sure
that
this
all
this
stuff
is
able
to
scale-
or
at
least
we
have
a
plan
for
it.
So
Jitsu
is
the
short
answer.
B
What
what
is
Jitsu
can
somebody
give
me
a
non-technical
answer.
A
Just
a
way
to
submit
events
and
collect
them
and
python
into
whatever
data
warehouse
you
want.
It
is
snow
plow
without
Snowflake.
B
A
Right
now
we're
just
going
to
stick
with
page
views,
one
of
the
edits
we
made.
We
weren't
sure
exactly
what
events
we
want
to
track
yet
and
we'd
rather
just
you
know,
have
an
idea
of
what
we
want
to
track
specifically
rather
than
trying
to
collect
data
on
every
possible
action.
So
page
views
is
just
standard
at
that
point,
so
we
just
want
to
start
there
and
then
figure
out
okay.
A
What
are
we
actually
curious
about
and
then
maybe
set
up
some
funnels
and
things
like
that
based
on
that,
but
as
far
as
user
driven
interactions
and
things
like
that,
there's
no
there's
no
plan.
Yet.
Okay.
A
So
the
tracking
stack
consists
of
Jitsu,
which
is
the
data
collection
or
and
piping,
and
then
click
house,
which
is
where
the
data
will
be
set
to
and
stored
and
then
we'll
be
testing
out.
Cubes
just
see
how
it's
able
to
query
and
set
up
the
schemas
for
that
and
then
from
that
point
we'll
be
setting
up
dashboards
to
to
consume
from
the
cube
API.
A
C
B
B
A
Yeah
so
I
mean
yeah,
that's
actually
a
really
good
point
and
I'm
glad
you're
bringing
it
up.
This
is
not
intended
to
replace
any
instrumentation.
We
have
set
up
right
now.
This
isn't
to
support
us
building
the
user-facing
analytics
features
so
that
they
can,
you
know,
potentially
implement
it
for
their
own
products
right
now.
A
What
we're
doing
is
just
setting
up
the
stack
making
sure
it's
set
up
we're
running
our
environment
and
that
we're
able
to
collect
that
so
that
eventually
we
can
get
other
customers
into
that
stack
so
that
they
can
start
collecting
on
things
that
they
want
like
for
their
own
products,
perfect.
So
we're
dog
fooding
for
the
purposes
of
making
sure
that
it
works
for
customers
before
we
say:
hey
customers
play
with
this
and
then
find
out
that
it
doesn't
scale,
doesn't.
D
B
Think
it
might
yeah
it
I
think
that
it
might
so
immediately
when
I
heard
dog
food
and
I
thought.
Oh,
my
God,
how
is
this
gonna
affect
everything
we
do
so
I'm?
That
puts
my
mind
at
ease,
but
once
it
gets
a
little
bit
more
sophisticated
and
you're
actually
at
feature
level
or
event
level,
rather
than
just
page
views.
It
would
be
good
to
kind
of
replicate
what
some
groups
have
been
doing
right
and
then
do
side
by
side
to
see
where
the
gaps
are
yeah,
not.
A
Definitely
I
mean
it's
going
to
be
good
to
see
how
we're
doing
instrumentation
across
the
board
at
gitlab
and
see
how
that
could
be
implemented
with
with
product
analytics.
There's
nothing
stopping
us
from
replicating
that,
but
I
want
to
be
clear
that
the
intentions
not
to
like
replace
anything
at
least
right
now,
the
the
goal
is
just
one.
It's
a
good
opportunity
for
us
to
see
if
this
works
and
get
this
running
on
gitlab
scale.
A
Ramp
it
up
from
like
one
percent
traffic
and
go
from
there
because
yeah
we,
we
literally
have
we're
in
progress
of
making
this
work
from
a
Docker
compose
file
to
now
kubernetes,
and
so
we
want
to
make
make
sure
it
scales
or.
A
B
A
Like
that,
this
is
literally,
and
that's
that's
going
to
be
something
that's
worth
reminding
as
well.
Is
that
there's
a
distinction
between
us
using
it
to
drive,
feature
development
and
then
building
the
feature,
as
opposed
to
like
there
were
a
lot
of
discussions
before
about
well
what
data
do
we
want
to
track?
And
it's
like
well,
let's
remember
that
this
is
just
for
us
to
build
the
features
for
other
customers
first
track
what
they
want.
This
is
not
you
know
intended
to
place
any
system,
at
least
this
point
in
time.
D
Yeah
thanks
Amanda,
so
since
this
is
awesome
that
we're
making
so
so
Speedy
progress,
but
I
just
wanted
to
point
out
one
concern
if
we
are
going
to
enable
any
kind
of
additional
tracking
on
github.com
I
want
to
make
sure
that
we
are
aligned
and
following
our
commitment
towards
the
professor
protection
and
the
third
point
of
this
commitment
is
about
the
US,
announcing
anything
that
we
are
gonna
track,
and
so
this
it
is
a
kind
of
user
tracking,
because
wider
Community
can
interact
with
the
github.org
group.
D
So
it's
it
can
be
a
data
that
is
not
only
product,
GitHub
employees,
but
also
the
wider
Community
I.
Wonder
if
you
should
announce
this
in
some
form
of
blog,
post
or
or
whatever
kind
of
different
announcement
that
we
are
gonna,
enable
new
tracking
and
what?
For
and
what
gonna
be
tracked.
So
we
are
not
creating
the
so
we
are
not
breaking
our
policy,
because
this
can
severely
impact
the
trust
that
we
are
very
hard
working
to
get
to
gain
from
our
community
to
give
us
data.
A
Agree
right
now,
there
is
no
intention
to
scope
this
Beyond
get
lab
team
members.
This
is
just
a
way
for
us
to
get
production
scale,
data
at
least
from
how
we're
using
it
as
opposed
to
what
Tim
was
doing,
which
is
running
CI
pipelines
to
simulate
traffic
on
the
proof
of
constant
instance.
So
as
far
as
I
under
like
for
one,
even
when
we
do
open
it
up
or
gitlab
team
members,
I'll
definitely
want
to
notify
that
hey.
A
This
is
something
that
we're
looking
at
doing,
but
right
now
there
are
no
plans
to
actually
expand
that
Beyond
gitlab
org,
it's
just
one.
One
thing
we
really
wanted
to
do
is
be
able
to
like
open
it
up
to
something
that
was
real
like
operating
in
production,
and
so
that's
why
we
picked
gitlab.com.
But
the
intention
is
not
to
start
at
least
with
this
internal
preview
and
for
at
least
the
next
few
iterations
to
to
open
this
up
to
anyone.
That's
not
part
of
gitlab
right
now,
so
I.
B
B
B
If
it's
internal
projects,
if
there's
still
the
core
people,
so
it
wouldn't
hurt
just
to
make
that
clear,
yeah
on
the
commitment,
page
I
think
it's
as
simple
as
like
Sam.
If
you
could,
on
the
page,
that's
linked
have
just
a
subsection.
B
That's
like
product
usage
data
collected
for
product
development
as
the
sub
header
and
just
say
there
will
be
times
that
during
the
course
of
developing
for
our
product
analytics
whatever
and
link
to
the
vision
page
that
we
may
collect
unidentifiable
data
for
use
of
our
development
process,
it's
not
taken
out
of
our
product,
blah
blah,
blah
and
I.
Think
that
that's
enough
and
then
I
think
you
are
free
to
continue
as
long
as
you
stay
non-identified.
B
B
C
Nikolai's
I
hope
I'm
pronouncing
your
name
correctly.
Is
it
mikolak?
Is
that
it?
Okay?
That's
your
your
point
earlier
I
mean
I
I.
Think
there's
a
benefit
in
over
communicating
what
we're
doing
here,
just
because
it's
better
to
be
transparent,
and
it
also
generates
excitement
for
this
new
direction
of
round
product
analytics.
I.
Think
a
blog
post
would
be
perfect
to
not
only
show
people
what
we're
doing,
but
also
gives
us
an
opportunity
to
say
this
is
the
future.
C
We
envision
we're
going
there
please
contribute,
so
they
can
see
it's
a
public,
what
we're
doing
as
opposed
to
a.
Why
do
I
have
this
cookie
now
or
this
tracking
link
and
a
link
to
something
that
looks
more
like
a
legal
document
inside
the
handbook
so
agree,
I,
think
a
blog
post
is
great.
I
think
handbook
updates
are
great.
We
definitely
need
to
be
explicit
rather
than
people
discovering
this
on
their
own
and
inventing
that
story
themselves.
B
C
A
B
Yeah,
that's
right
on
what
we've
learned
Sam
from
for
those
who
have
been
around
a
while
from
Telemetry
gate,
2000
and
whatever
is
people
were
most
angry
about
the
failure
of
communication
of
proper
and
thorough
communication.
B
They
also
didn't
want
to
share
their
data,
but
they
were
imagining
either
filling
in
the
gaps
that
they
didn't
have
the
information
of
what
was
going
to
happen
because
we
didn't
give
that
information,
and
so
they
were
angry
about
things
that
we
weren't
even
intending,
or
at
least
at
that
moment,
weren't
intending
and
so
the
more
we
can
over
communicate,
I
think
the
better
and
just
very
transparent-
and
this
is
a
really
good
story.
This
is
adding
to
our
feature
set.
This
isn't
actually
taking
more
information
from
somebody.
B
We
might
use
some
in
purposes
of
core
purposes
of
development,
but
ultimately
the
end
goal
isn't
to
take
more
privacy
from
the
user,
it's
to
give
developers
more
Tools
in
their
toolpit.
So
it's
a
really
nice
story.
I
love
the
ideal.
Blog!
Isn't
that's
right
on!
Let
me
know
if
there's
anything
I
can
help
with
it
sure.
C
Yeah,
we'll
definitely
get
something
put
together
before
we
do.
A
full
rollout
I
think
we
have
a
lot
of
steps
to
get
there
before
before
we're
at
there,
but
it's
definitely
something
to
keep
in
mind
and
we'll
definitely
do
before.
We
push
the
button.
A
Yeah
for
sure,
there's
a
it's
happening
fast,
but
we
want
to
make
sure
we're
not
tripping
over
ourselves
too,
but
yeah
right
now.
The
focus
is
just
make
sure
it
can
run
on
gcp
and
then
we
can
actually
figure
out
the
details
of
yeah
how
we
want
to
roll
it
out.
But
that's
that's.
B
A
Yeah
yeah
sure,
but
that
would
be
a
good
step
towards
while
we
prepare
like
the
outward
communication
totally,
this
effect
specifically
I,
don't
think
exists
yet.
But
what
we'll
say
is
that
yeah,
we'll
probably
want
to
start
in
staging
as
we
can
put
together
the
blog
post
and
then
communicate
it
out
as
we
roll
out
to.com.
So
yeah
great
points
appreciate.
C
The
discussion
instance
from
the
dog
footing,
topic
I,
also
added
an
issue
link
to
where
we'll
want
to
put
these
product
analytics
tracking
pieces.
In
whatever
screens
we
put
together
to
show
dashboarding
reporting
the
issues
brand
new.
So
it
doesn't
have
a
lot
there,
but
you
can
follow
along.
That's
where
we'll
be
doing
some
of
that
dog
food
and
work.
A
Yes,
we
are
looking
for
feedback.
There's,
there
are
tons
of
Ethics.
There
are
tons
of
readings
to
be
done,
a
lot
of
Link
spaghetti.
So
if
there's
a
point
of
clarification
or
just
general
feedback,
you
have
please
feel
free
to
chime
in,
but
it's
it's
exciting
going
from
no
no
epics,
and
now
we
have
a
lot
of
epics.
So
there's
a
lot
more
clarity,
we'll
keep
we'll
keep
working
to
drive
more
of
that,
but
definitely
looking
forward
to
some
more
feedback
from
everyone.
A
Well,
that
was
my
wall
of
text
I'm
happy
to
talk
or
see
if
anyone
else
would
like
to
share
anything
else,
that's
going
on
with
them.
Otherwise,
if
there's
nothing
else,
we
can
get
some
time
back.