►
From YouTube: 2020-09-10 Knowledge Weekly 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
All
right
welcome
to
the
knowledge
team.
Weekly
sync:
we've
got
an
agenda
here.
Let's
kick
off
with
the
mini
stand
up.
B
Okay,
I
think
I
can
voice
it
because
both
look
and
are
not
present,
so
backhand
is
all
merged
for
design
to
dos,
and
as
for
tom,
I
would
say
that
all
the
front
end
part
is
merged
as
well
he's
saying
one
major
merger
quest
left,
but
it's
already
merged
today.
The
only
thing
we
have
to
do
is
to
reload
a
feature
flag
because
front
end
was
delivered
behind
the
feature
flag,
and
I
think
we
can
safely
say
that
it
will
be
delivered
within
134
and
on
to
marcus.
C
And
so
for
group
vikis,
and
I
had
one
deliverable
for
this
post-
receive
stuff
that
one's
merged
and
at
the
moment,
I'm
working
on
features
specs,
and
I
have
two
automatch
requests
to
fix
the
attachment
uploading
and
also
to
add
the
license
check.
I
realized
that
at
the
moment
it's
just
an
ee
feature,
but
it's
just
free
it's
like
available
in
all
licenses,
so
I
did
the
check
to
actually
check
for
start
the
license
and
those
two
are
nebu
and
once
those
are
merged,
I
think
we
can
enable
it
on
staging
and
stuff.
B
A
Cool
are
we
still
on
track
for
release
next,
really
like
for
rolling
out
in
thirteen
five.
C
C
A
A
A
That
would
be
awesome,
and
it
might
I
know,
there's
also
those
configuration
like
with
confluence.
Like
I
posted
in
below,
we
were
able
to
pull
like
112
projects
or
using
confluence
where
it's
just
it's
like
a
variable,
like
account
almost
on
the
setting
being
on
or
something-
and
I
know
we
don't
have
the
setting,
but
but
maybe
we
could
also
just
set
a
variable
like
that
or
something
that's
doing
a
count.
C
B
C
A
Because
yeah,
then
we
could
have
a
graph
for
like
wiki
page
views
across
everything
right
and
then
split
it
yeah.
That
would
be
awesome.
So
if
we
need
an
extra
issue
for
that,
let's
make
sure
that
that
goes
in
for
13.5
and
then
so
when
we
roll
it
out,
we
truly
do
we
can
see
how
many
people
start
using
it.
C
A
C
So
I
guess
we'll
have
to
think
about
that.
Yeah.
A
I
guess
we're
gonna
kick
that
off
in
a
couple
weeks,
but
yeah,
I'm
gonna
we've
been
very
clear
that
we're
doing
mvc1
up
through
the
management
chain
as
well.
So
I
will
kick
it
off
and
say
that
it's
like
group
level,
wikis,
mvc
and
we'll
we'll
say
what
you
can
do
and
what's
still
planned
and
then
what's
still
planned,
is
probably
going
to
be
further
out
on
the
roadmap,
because
we've
got
other
things
that
we
need
to
focus
on.
C
A
Yeah,
that's
that's
a
good
point.
We're
not
definitely
not
rolling
out
a
complete
parody
with
projects,
so
we
got
to
make
sure
that
that's
clear,
okay
and
yeah.
We
might
be
the
like
I'm
trying
to
see
if
we
could
maybe
even
be
the
headline
of
the
next
blog
post.
It's
not
up
to
me,
but
I
have
asked
the
marketing
team
to
consider
it
because
it
is
the
second
most
uploaded
issue
right
now
at
gitlab,
so
that'll
be
pretty
big.
B
D
So
the
I
implemented
the
adding
the
working
design
contributions
to
the
contributions
calendar,
unfortunately
that
with
already
quite
an
expensive
query
to
calculate
the
contributions
calendar,
especially
for
heavy
users
like
we
see
in
good
lab
and
it's
kind
and
it
this
expands
that
and
makes
it
even
more
expensive
and
I
think
we're
back
we're
now
blocked
on
that
performance
issue.
So
I
just
need
to
kick
that
back
and
particularly
get
an
input
from
product
about
what
we
want
to
do
about
that.
D
There's
lots
of
information
on
the
voter
request
there
about
what
issues
we've
experienced,
what
the
different
avenues
we've
explored.
What's
the
potential
future
paths
are,
but
either
way
is
going
to
require
substantial
extra
time
to
make
this.
A
D
Performant
for
the
kind
of
things
we
want
to
do
with
it,
so
there
are
a
few
different
things
in
there.
It'd
be
good
to
get
either.
Give
you
a
bit
of
time
to
read
through
that
and
to
know
how
we
want
to
perceive
whether
whether
something
we
want
to
spend
more
time
or
not.
A
And
so
is
there?
Are
there
like
weighted
options
in
there
like
for
me
just
to
understand
like
we
could
fix
it
this
way,
and
it
will
take
this
number
of
points
or
we
could
fix
it
this
way,
and
it
would
take
this
number
so
that
I'm
just
sort
of
clear
on
what
would
how
how
much
would
go
into
the
capacity.
A
Oh
alex,
I
think,
you're
on
mute.
I
just
realized.
D
If
you've
got
at
the
bottom,
you
should
see
there's
an
alternative
approaches
section
and
one
of
the
right
near
the
near
the
end.
In
a
conversation
with
andreas
and
the
the
cheapest
one
would
be
fetching
this
data.
We
could
fetch
data
on
the
moment
fetching
a
year
at
a
time
if
we
bet,
if
we
do
a
quarter
or
a
month
at
a
time
we
could
which
queries
will
definitely
be
a
lot
cheaper.
They
will
require
front-end
changes
or
display
changing
how
much
of
a
calendar
we
display
at
once.
That's.
D
Going
to
be
the
have
a
visible
impact
of
the
user,
definitely
requires
ux
input
and
product
input,
but
overall
that
would
be
the
cheapest,
the
lowest
hanging
fruit.
A
few,
maybe
three,
four
points
front
end.
D
If
we
want
to
do
a
d
normal,
if
we
that
the
most
robust
solution
is
listed
there
under
a
new
denormalizing
table,
that
would
be
probably
about
five
points
on
the
back
end,
at
least
and
would
require
a
new
new
back
end
table.
You
require
new
workers,
new
queues.
It
would
require
a
quite
a
long
database
migration,
and
so
the
events
table,
I
think,
is
the
the
largest
table
we
have
in
the
database
must
require
a
new
table.
D
That
is
a
view
on
that.
So
it's
going
to
be
also
a
large
amount
of
storage
and
postgres.
We
need
to
consider
whether
we
want
to
dedicate
that
much
okay,
disk
storage
as
well,
because
that
could
be
expensive
and
it
would
also
require
at
least
a
few
different
word
quests.
In
order
to
get
there.
D
A
D
It's
quite
an
old
piece
of
code,
it
was
obviously
originally
focused
around
around
pushers.
Now
has
has
accumulated
lots
of
other
event
data
inside
it.
The
immense
table
is
used
throughout
the
organization
we
populate
with
lots
of
different
kinds
of
events
yeah
and
the
calendar.
D
I
don't
think
any
one
team
owns
the
contributions
calendar
so
currently
there
is
a
single
query
that
reads
a
year
of
data
for
each
user.
When
you
view
your
your
contribution
calendar-
and
this
is
just
reading
out
of
what's
populated
with
that.
A
Okay
and
they're,
not
it's
dynamic,
they're,
doing
queries
like
this.
It's
not
like
everybody
is
writing
a
line
of
text
into
like
a
static.
D
Well,
it's
a
single
it's
a
single
sql
query,
but
yes
every
time
you
view
the
events
calendar.
We
issue.
One
of
these
large
queries
for
100
days,
365
days
of
data
for
each
it's
crazy
for
the
user.
It's
quite
an
interesting
way
of
going
about
it.
D
Postgresql
does
a
really
good
job
of
caching
this.
So,
although
it's
quite
an
expensive
query,
it's
run
times
are
performant
when
you
hit
the
hot
cache,
but
once
the
data
drops
out
of
cache,
it
gets
a
lot
slower.
So
postgres
gets
a
lot
of
that
transparency
transparently
for
us
and
for
less
active
users,
they
have
fewer
events
and
the
query
performs
a
lot
faster.
B
A
Yeah
I'll
read
through
it,
and
I
know,
there's
also
been
talk
about
making
the
user
feed
more
robust.
The
descending
list
of
what
you
last
touched,
which
I
believe
is
on
your
profile
page,
could
be
wrong
but
like
in
the
from
a
ux
perspective
when
you
land
into
gitlab,
one
thing
gitlab
lacks
is
that
descending
list
yeah,
so
that
this,
the
underlying
structures
of
that
whole
system
may
need
to
be
reevaluated
that
it
so
that
it
could
be
quicker.
A
D
Diversion
is
not
a
big
issue
so,
but
if
there
isn't
that,
you
think
could
do
with
clarification.
Just
let
me
know
I'm
very
happy
to
write
up
something
on
different
levels
if
that's
required,
but
we-
and
I
think
this
also
needs
some
input
from
different
areas
of
product
as
well.
So
this
doesn't
affect
just
us.
It
affects
obviously
merge
requests
and
source
code
and
planning
and.
D
That
a
user
writes
a
comment
so
these
all
these
are
all
things
that
will
end
up
on
the
contributions
calendar
so
having
input
from
other
users
other
products
teams
about
how
they
want
the
contributions
calendar
to
work
and
whether
they're
prepared
to
support
this
level
of
work.
A
Yes,
useful-
and
I
think
that
we'll
want
to
unify
on
these
objects
like
what
is
a
user
feed
or
what
is
the
true
single
source
of
truth,
history
of
what
someone's
done
in
gitlab
and
then
what
gets
displayed
up
into
the
the
ui
and
just
be
very
clear
on
that.
A
But
I'd
also
yeah
definitely
want
their
input,
because
I
don't
want
to
change
something
if
there
might
ever
like
each
of
those
other
teams
might
be
having
the
same
discussion
we're
having
right
now,
but
since
we
are
trying
to
improve
the
speed
and
the
performance
of
git
lab
overall
in
this
quarter,
this
might
be
something
we
look
at
where
we
could
contribute
a
new
architecture
that
would
help
speed
things
up
for
everybody.
On
that
page.
D
B
D
It's
not
the
landing
page
for
any,
so
not
the
landing
pages
that
your
project
list
or
other
your
issues
list
or
something.
But
it's
quite
an
expensive
query
and
we
could
cut
maybe
90
90
95
of
the
time
off.
There
would
be
quite
significant.
A
I
pulled
numbers
on
the
confluence
integration,
sorry,
the
confluence,
gitlab
plugin
being
turned
on
on
projects
and
instances,
and
so
we
had
two
variables
that
I
have
access
to
in
sisense
and
it
looks
like
there's
112
projects
that
have
turned
that
on
and
44
instances
what
I
don't
know
and
I'm
gonna
try
to
dig
into
it
more,
and
maybe
you
all
would
know
how
to
query
this
out
of
our
data
set,
but
I'm
not
able
to
split
that
number
from
self-hosted
versus
versus.com.
A
A
A
But
in
size,
since
there's
a
chart
that
a
colleague
of
mine
had
where
you
could
pull
any
usage
ping
value
into
the
chart,
and
so
we
were
looking
at
that
yesterday
I'll
get
the
link,
but
I
don't
think
that
these
usage
ping
variables
have
the
nuance
of
like
any
metadata.
I
think
they're
just
a
number.
A
Okay,
so
that
will
be
an
interesting
number
to
see
how
that
ends
up
aligning
the
feedback
I
have
got
so
far
on
the
that
integration.
I
have
been
asking
you
all
in
slack,
which
has
been
really
good,
that
they
want
to
use
self-hosted
with
it.
So,
even
though
we're
building
only
confluence
for
dot
com
and
for
confluence
cloud,
we're
already
having
people
say,
oh,
I
want
to
use
it
with
confluence
server
or
on
my
self-managed
instance.
A
So,
right
now
we're
gonna
keep
investment
just
around.com,
but
like
if
this.
If
we
do
keep
hearing
this
again
again,
we
might
shift
and
say:
okay,
we're
gonna
explore
having
these
two
things
talk
together,
I
would
say
from
some
of
the
conversations
I've
been
in
probably
ninety
percent,
though
we
would
just
stick
with
dot
com,
ninety
or
ninety
five
percent.
A
So
I
gotta
try
to
pull
those
numbers
and
just
see
how
many
of
our
customers
are,
what
the
split
is
between
them
between
them.
C
Just
remember
if
somebody
has
production
access,
they
can
just
query
this
from
the
database,
so
you
can
just
ask
somebody
with
production
access
then
subtract
it
from
the
other
number.
Then
you
have
to
self-hosted
instances.
A
C
A
A
A
I
know
that
they're
looking
at
that
as
an
avenue,
so
potentially
it's
there
in
some
anonymized
way.
That
would
help
us
see
it,
but
I'll
ask
them
about
that
and
see
if
we
could
do
a
query
from
license
world
as
well.
A
A
A
All
right,
I
think,
that's
it
for
now.
I'm
also
going
to
reply
on
the
our
kpi
issue
that
we
have
for
designs
being
able
to
view
them
update
them.
A
I'm
a
bit
behind
this
week
on
my
to-do's,
but
that's
my
focus
today
and
trying
to
make
sure
that
we
have
the
definition
for
that
short
up
and
then
also
I'll,
be
reaching
out
to
you
all
just
so
that
we
have
it
weighted
and
scoped
and
make
sure
we
do
have
the
telemetry
there
so
that
it
by
13
6
at
the
latest.
A
So
this
last
bit
of
the
month
and
actually
today
we're
doing
a
pi
review
as
a
product
team,
so
we'll
be
looking
at
them
and
just
making
sure
that
those
are
ready
to
go
and
be
scheduled
and
worked
on.
Are
there
any
questions
about
that
around
that
one
telemetry
issue,
I'm
gonna,
see
if
I
can
find
it
here
in
my.
C
A
Yeah
I
was
noticing
that
yeah
we
have
been
pretty
active
on
that,
but
the
data
team
is
overwhelmed
doing
a
lot
of
these
all
at
once
so
yeah.
I
think
I'll
just
read
through
this
issue
again
and
just
see
where
we're
at
and
then
I
might
go
ahead
and
just
draft
the
chart
in
sizes
and
then
see
if
we
can
get
it
pulling
through.
C
A
C
A
Yes,
so
we
yeah,
so
we've
chosen
designs
to
be
the
rep
representative
right
now,
as
our
mbc
from
our
group,
the
most
correct
gmail
for
us
would
be
deduped
unique
users
that
have
touched
wiki
or
designs.
A
So
at
some
point
those
two
might
come
together
to
get
our
mvc
up
we're
putting
designs
up,
because
we
think
that's
the
number.
We
can
move
the
most
and
get
the
most
growth
to
like
to
watch
it
move
up
in
terms
of
adopting.
But
that's
a
good
question.
Yeah,
like
the
whole
organization,
is
going
to
run
off
monthly,
active
users
and
then
there's
all
the
the
the
letters
on
the
front
like
g's
group.
A
They
don't
have
f
now,
but
yeah
we
could
say
feature
mao
would
be
like
designs,
wiki
design
system.
Those
would
each
get
their
own
like
feature
or
feature
mouse,
but
no
one's
really
doing
that
at
this
point,
and
then
actions
is
another
one
that
we
could
look
at
where
we're
looking
at
an
individual
action
like
an
update
or
a
create,
or
something
like
that
or
a
comment.
A
Yeah
and
if
we
added
it
in
the
future,
it's
going
to
be
millions
and
like
designs
are
teensy
tiny
right,
so,
like
we'd
also
have
to
figure
out
like
how
to
make
the
charts
show
that
so
that
it
doesn't
dwarf
the
design
growth
and
we
understand
what's
coming
from
wiki,
what's
coming
from
design,
so
I
think
keeping
them
separate
right
now.
Even
though
designs
is
the
main
one,
we're
still
going
to
look
at
wiki
the
same
way,
it's
just
not
going
to
be
our
official.
A
Our
group
stat
all
right,
we're
at
time
here
or
overtime
thanks
everyone,
I'll
message
you
in
slack
or
on
the
issues
and
have
a
great
day
bye-bye.