►
From YouTube: Kubernetes SIG Contributor Experience 20190607
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,
hello,
everybody
welcome
to
the
June
5th
edition
of
contributor
experience.
My
name
is
Elsie
Philips
and
I'll
be
hosting
the
call
today.
I
would
like
to
kick
off
this
morning
by
reminding
everyone
that
we
do
have
a
code
of
conduct.
The
TLDR
on
it
is
please
be
excellent
to
each
other.
I
feel
like
more
of
the
fine
print
go
ahead
and
look
that
up
I,
don't
believe.
I
saw
any
new
contributors
on
the
call
today,
but
if
there
are
any
go
ahead
and
shout.
A
B
Nothing
much
to
report
I
just
got
to
schedule
the
things
in
YouTube,
which
I
usually
do
at
the
end
of
the
day
at
the
end
of
meaner
contributors.
So
those
will
go
out.
I
wanted
to
bring
something
up
to
the
group,
but
maybe
we
could
talk
about
it.
We
can
punt
if
we
need
to
still
debating
on
whether
we
should
do
West
Coast
time
zone
meetings,
because
the
East
Coast
European
friendly
ones
easily
have
like
twice
the
amount
of
participation.
C
Like
we
can't
even
get
the
traffic
information
from
YouTube
on
the
where
those
people
are
coming
from
timezone
wise
to
because
I
also
have
I,
also
I
mean
at
least
on
meet
our
contributors,
I
think
I
think
West
Coast
people
like
we
say
we
don't
like
morning,
but
for
this
kind
of
stuff
we
might
actually
like
the
morning
because
it's
just
kind
of
kicking
back
and
it's
not
like
thinking.
If
you
know
any
means
like
I,
don't
know
so
I
mean
I.
C
B
And
our
West
Coast
time
zones
are
like
1
p.m.
like
after
people
are
coming
back
for
and
like
they're,
going
back
to
work,
they're
not
gonna,
go
like
hang
out
on
YouTube
I
would
love
to
have
a
more
like
hey
it's
a
Thursday
night
or
whatever
you're
just
getting
off
work.
Why
don't?
You
come
hang
out
in
office
hours,
but
for
me
because
I'm
on
the
East
Coast,
we
would
need
like
a
West
Coast
streamer
by
far
so
I'm.
C
B
C
B
D
That's
gonna
be
the
best
thing
to
do,
because
the
settings
are
different
per
computer.
We
can
try
and
export
them,
but
it's
only
exporting
them
for
that
computer.
If
I
do
something
with
my
laptop
and
send
it
to
George.
It's
gonna
break,
like
he's
just
gonna,
have
to
reconfigure
everything
anyways
and.
C
Contacts
quickly
for
people
on
the
line
right
now
that
have
no
idea
what
we're
talking
about.
We
are
doing
like
a
YouTube
streaming
team
because
we
have
a
lot
of
streaming
programs
at
this
point.
So
we
have
the
community
meeting
meet
our
contributors
office
hours.
These
are
the
program
updates
that
we're
doing
right
now.
So
sorry,
I
just
wanted
to
give
that
context.
I
know
people
join
or
joining
that
I,
don't
necessarily
know
so
just
make
sure
we
give
that
context.
C
A
C
Us
to
meet
our
contributors,
which
is
actually
your
section.
We
had
our
June
edition
this
morning.
It
was
really
fun,
I
feel
like
they're,
getting
funner,
which
is
awesome,
that's
the
point
and
yeah.
If
anybody
on
the
line
wants
to
come
on
for
July,
we're
only
doing
one
right
now,
it's
the
same
problem
that
office
hours
is
having
it's
not
necessarily
a
problem.
It's
just.
We
feel
like
a
better
time
slot
for
the
you
know,
other
you
know
another
addition
would
be
it
post,
4:00
p.m.
Pacific
time,
so
that
and
that's
still
in
the
cards.
C
B
B
D
C
If
there's
anybody
on
the
call
right
now
that
thinks
that
our
chairs
and
our
technical
leads
of
our
six
and
working
groups
need
no
information
this
week,
that's
super
critical
I
think
there
was
like
release
information
that
went
out.
I,
remember
Claire,
for
instance,
said
something
to
the
sig
leads.
If
there's
anything
that
needs
to
be
reiterated
or
disseminated
to
all
of
our
SIG's.
Let
me
know
now
put
it
in
the
doc
forever
hold
your
peace.
The
end
is.
B
C
Think
Jonas
is
still
out.
I
don't
want
to
give
a
summary
I
think
he
should
do
that,
but
Barcelona
summit
was
awesome
and
then
we'll
get
an
update
from
Jonas
when
he's
back
I
think
that's
next
week
actually
and
then
San
Diego
summit,
we've
started.
Organizing
awesome.
If
everybody
in
here
is
interested
in
shadowing
I,
think
we
have
a
ton
of
shadow
rolls
available.
C
C
The
hope
is
that
you
will
either
lead
the
contributor
summit
the
next
time
or
lead
a
section
of
the
contributor
summit
the
next
time
and
it's
sort
of
mentoring
by
doing
but
I
do
have
a
issue
as
soon
as
I'm
done
talking
that
I'll
link
in
the
document,
so
that,
if
you
would
like
to
volunteer
for
the
san
diego
cobra
Nettie's
contributor
summit,
please
see
us.
We
have
not
met
or
kicked
off
anything
just
yet.
C
E
E
E
The
CN
CF
is
going
to
provide
a
shuttle
bus
from
the
Intercontinental
for
people
to
attend
the
contributor
summit,
which
will
be
at
the
convention
center
and
also
what
is
it
a
hundred
power
boxes,
because
apparently
the
Convention
Center
is
an
adequate
power,
so
the
CN
CF
is
having
to
bring
in
portable
battery
packs
for
everyone.
It's
gonna
be
a
lot
of
fun,
yeah,
the
the
things
you
find
out
when
you
show
up
in
a
convention
center
for
the
first
time.
Oh.
E
C
E
C
E
Yeah
yeah,
no,
that's
not
that's
not
sort
of
current
contributor
of
content.
We
could
possibly
use
some
of
that
for
new
contributor
workshop
for
the
sort
of
basic
take
on
how
testing
works
for
that
matter.
A
lot
of
people
can
do
the
basic
sort
of
walkthrough
love
use
of
meteor
PR
and
here's
what
happens?
The
this
was
more
to
do.
An
in-depth
of
hey,
I'm,
adding
a
new
feature
to
kubernetes
and
I
need
to
figure
out
how
to
add
it
to
the
ete
tests.
E
A
G
A
A
Let's
talk
about
our
one
issue,
that's
in
the
blocked
column
and
then
move
on
renamed
release,
notes
and
traffic
labels
to
conform
to
category
subcategory
norm.
Okay,
can
anybody
talk
about
that?
One
I
had
a
peer
open
for
it,
but
I
haven't
updated
it
because
it
was
in
the
middle
of
the
release
cycle
I'm,
not
even
sure
if
they
want
to
do
this,
but
I
can
make
revisit
it
and
make
sure
that
it
doesn't
remain
in
the
block
column.
At
least
cool
that'd
be
great
all
right.
A
A
A
Okay,
maybe
we'll
go
to
the
next
topic
and
the
whole
circle
back
around
yeah,
yeah,
custom,
flocculating
message:
George
yeah,.
B
So
can
you
guys
hear
me?
Yes,
yes,
sweet
so
Katherine
was
working
on
a
PR
to
have
a
welcome
message
when
a
new
user
joined
slack,
you
know,
hey
check
it
out
and
I
was
like.
You
know
what
those
slack
loading
messages
are
cute.
We
should
totally
like
use
those
instead
of
like
whatever
stuff
they
put
on
there,
so
I
I
dug
in
there
it
ends
up.
We
can
do
a
hundred
and
fifty
custom
loading
messages.
You
can't
put
links
in
there.
B
Unfortunately,
that's
kind
of
really
a
bummer
and,
depending
on
you
know,
network
conditions,
it
might
be
slow,
so
I
started
to
put
things
in
there
like
check
out
meaner
contributors
or
you
know,
join
cute
careers
if
you're
looking
for
a
job
that
sort
of
thing
so
I,
mostly
just
wanted
to.
Let
people
know
that
a
that
exists
and
we
I
put
about
10
in
there.
So
we
have
like
a
hundred
and
forty
useful
things
that
I
think
we
can
put
in
there.
B
We
have
a
method
in
the
community
repo
for
for
you
to
file
requests
for
slack
stuff.
So
I
was
thinking
just
like
saying,
hey.
If
there's
a
cool
message,
just
like
one
of
the
slack
happens
now
or
better
yet
do
it
in
github,
so
we
can
track
it
for
useful
things.
You
know
our
programs,
you
know,
don't
you
know,
don't
forget
to
register
for
cute
like
that
sort
of
thing,
I
figure,
you
know
they're
benign
people
have
to
read
something
anyway,
might
as
well
make
them
useful.
A
B
B
Would
be
great,
I
find
a
picture
or
links.
I
thought
I'd
be
awesome
that
it's,
like
you
know,
click
through
to
find
a
mess
like
you
can't.
Unfortunately,
it's
all
plain
text,
inspirational,
pin,
Hawken
quote
yeah,
so
there's
there
is
one
that
I'm
gonna
ask
him
permission.
The
problem
is:
is
that
a
tributes
a
quote
to
me?
We'd
have
to
make
him
an
admin.
Put
his
one
quote:
I
think
we're
thinking
about
the
same
quote
and
then
D
adding
him.
Maybe
we
can
figure
out
a
way
to
do
that
or
something
but
I.
B
H
Me
now,
oh
great,
okay,
I'm,
so
sorry
about
that
I'm.
So
sorry
that
my
my
video
camera
doesn't
work
for
some
reason,
but
Linux
no
I'm
on
Mac
I,
just
maybe
it's
my
monitor,
plug-in
toolchain
thing.
Whatever
I
don't
know:
hi
I'm,
Hamill
I
work
at
github
on
the
machine
learning
team.
H
The
way
I
kind
of
got
connected
with
some.
Some
of
you
folks
initially
was
we've
been
collaborating
with
some
open-source
maintainer
to
build
tools
that
might
help
automate
some
of
the
workflow
and
kind
of
ease
the
pain
and
we've
been
trying
to
do
it
so
that
you
know
we
can
validate
our
ideas
and
sort
of
in
real
time
sort
of
iterate
and
and
have
things
that
can
eventually
make
it
into
the
product,
but
also
make
a
difference
for
maintainer
and
so
kind
of
an
idea
was
mentioned.
So
we
had.
H
We
have
this,
so
we've
been
starting
a
couple
of
different
projects,
so
one
of
them
is
we
try
to
start
simple,
so
the
first
one
is
trying
to
automatically
label
issues
for
maintainer
x'
that
might
help
so
that
might
help
a
triage
and
I
know.
There's
probably
thousands
of
things
that
you
would
rather
have
improved
with
github
that
are
probably
burning
topics
that
could
be
improved.
H
H
Okay,
so
this
is.
This
is
how
you
install
it.
It's
a
it's
a
github
app.
So
what
that
means
is
on
the
marketplace.
There's
a
link
in
them
in
the
notes
to
this
page.
So
don't
worry,
I've
already
installed
it
on
my
own
repos.
Obviously
so
it
doesn't
say
install,
but
this
green
button
will
say
install
for
you
if,
if
you're
trying
to
do
this
on
your
own,
but
the
way
it
works
is
pretty
simple.
H
H
It
has
like
three
different
types
of
labels,
so
labels
bugs
it
labels,
things
that
are
questions
and
in
in
feature
requests
and
you
can
alias
the
labels
so
that
they
conform
to
your
repository.
So
this
is
an
example.
You
know
so
someone
open
this
issue
and
then
what
happens
is
this
issue?
Label
block
comes
in
and
if
it's
confident
enough,
if
it
knows
what
it
feels
like
you
know,
you
know
we
can
classify
this
issue.
H
Then
it'll
it'll
go
ahead
and
label
it
and
then
also
a
link
out
to
this
more
information,
and
then
we
collect
feedback,
and
that
allows
us
to
retrain
this
and
make
it
easier,
and
so
that's
how
it
works.
You
can
install
it
and
uninstall
it
any
time
without
too
much
pain
with
you
know,
going
to
this
marketplace
page
and
that
I
showed
and
again
you
can
alias.
So
this
is
how
you
alias.
H
H
B
Is
yeah
this
looks
awesome
by
the
way.
Do
you
have
a
sense
of
failure
rates
when
people
report
when
the
machine
learning
is
wrong?
What
do
you
track
that
yeah.
H
I
tried
yeah,
we
track
that,
so
we
see
a
like
a.
This
is
like
a
dashboard
that
is
surface
to
the
public
and
we
have
one
for
every
everything,
so
we
collect
it
and
not
everybody
gives
us,
does
a
good
job
of
giving
us
feedback
on
every
single
thing,
but
we
see
like
failure
rate
of
about
like
false
positive
rate
of
about
five
percent
right
now,
we're
being
very
kind
of
conservative.
So
if
we're
not
confident,
then
the
the
bots
will
not
label
an
issue.
H
I
H
We
started
with
those
three
things,
because
those
are
the
things
that
the
maintains
were
working
with:
that's
we're
kind
of
on
their
list
of
burning
things
they
wanted
to
triage.
However,
if
you
know
you
there's
other
things
that
would
be
happy
to
retrain
a
model
on
on
those
four
kubernetes.
If
there's
something
that
is
helpful,
there.
I
Yeah,
so
my
my
big
thought
here
is
so
we
be
kind
of
competing
products
as
I
see
it
to
what
you
got
here
is
issue
templates
and
we
use
those
fairly
extensively
for
this
exact
thing,
where
we
to
be
able
to
classify
into
those
those
categories
and
and
for
example,
in
our
main
Cabrini's
kubernetes
repo,
we
have
four
template
bug
and
has
been
request,
support,
request
and
failing
tasks
and
three
of
those
fall
into
those
three
categories
that
you
mentioned.
Where
do
you
see
as
the
advantage
like?
H
Don't
see
that
yeah,
so
if
you're
using
issue
templates
and
they're
fairly
structured
process,
I
frankly
I
don't
see
that
much
of
an
advantage
of
using
the
label
bot.
You
know
it
depends
on
your
triage
process
right
now.
If
there's,
if
you
already
have
a
way
that
you're
you're
able
to
filter
and
sort
of
select
issues
easily,
and
you
don't
really
need
the
labels
because,
let's
say
they're
getting
shepherded
already
with
your
current
process,
then
yeah,
maybe
the
advantage
is
not
doesn't
seem
that
high.
H
B
Leads
to
a
follow-up
question,
I
think
I
have
about
this.
So
one
thing
I
always
mess
up
is
I
file,
an
issue
I
filed
a
template,
but
I
forget
to
label
all
the
right
things
and
then
the
bots
like
yells
at
me,
don't
forget,
but
like
I'm
thinking
in
my
non-engineering
had
if
95%
of
the
things
I
file
are
under
sink
and
Trebek's,
can
something
like
this
kind
of,
like
figure
out
based
on
machine
learning?
What
labels
to
at
least
recommend
me
maybe
before
yeah.
I
So
beyond
those
categories,
the
next
kind
of
like
classification
step
for
us,
is
splitting
it
up
based
on
special
interest
group
and
being
able
to
take
a
look
at
which
team
within
our
very
large
structure
would
actually
be
handling
the
triage.
So,
while
from
a
from
an
outside
perspective,
if
I'm
like
an
outside
person,
just
stumbling
across
KK
and
I
want
to
create
an
issue,
I
can
usually
have
a
very
good
sense
of
like
am
I
trying
to
report
a
bug?
Am
I
asking
for
a
new
feature
like
that?
Somebody's?
I
That's
something
that
at
least
from
an
outside
perspective.
A
lot
of
people
will
understand.
Oh
I
know
I
kind
of
know
which
basic
category
of
filing
it
into,
but
without
kind
of
understanding
how
Cooper
needs
to
split
up.
It's
hard
to
know
exactly
which
special
interest
group
is
going
to
be
handling
that
with
which
cut
which
team
within
kubernetes
is
going
to
be
doing
us,
or
we
also
have
labeled
based
on
like
different
areas
of
the
project,
again
to
like
split
up
in
triage.
I
Okay,
these
type
of
requests
are
all
grouped
together
where
I
could
potentially
see
machine
learning
being
very
helpful
in
things
like
if
there
was
something
that
was
able
to
kind
of
like
stock
up
hey
so
based
on
previous
triaging
decisions.
You
know
if
it
mentioned
this
file
or
these
keywords.
This
would
be
a
good
candidate
for
lumping
into
contributor
experience
or
say
note
or
signet
work,
or
things
like
that.
That
kind
of
triaging
is
something
that
it
won't
move
way
more
difficult
to.
Self-Classify
I
see.
H
I
Yeah
I
think
that
that
is
for
is
our
particular
model.
Well,
I,
like
you
know,
this
is
just
my
opinion.
So
if
anybody
has
differing
differing
opinions,
please,
like
do
feel
free
to
speak
up,
at
least
from
what
I
see
of
our
triage
model.
Things
like
a
max
would
be
extremely
helpful
to
us,
as
opposed
to
this
very
baseball.
I
know
you're,
very
early
kind
of
in
this
process,
you're
trying
to
make
something.
That's
generic
across
lots
of
different
projects,
but
I
know
for
us.
The
issue
template
works
fairly.
I
Well,
for
that
basic,
like
three
bucket
classification
bug,
feature
request,
support
question
we
actually
in
our
repo.
We
do
not
want
support
questions,
so
we
are
feature
our
issue.
Template
for
support.
Questions
is
like
please,
don't
log
them
in
github.
Go
over
here
for
support,
so
like
disabling,
like
disabling
our
templates
and
moving
over
to
another
system,
would
actually
probably
be
a
regression
for
us.
That
makes
a
lot
of
sense.
B
I
always
feel
like
the
bot
is
always
like,
with
the
content
of
my
issue.
I
feel
like
sometimes
feels
common
census.
You
know
like
the
contributor
guide
should
be
translated
in
15
languages
and
then
I'll
go
do
something
else
and
then
I'll
get
an
email.
That's
like
you
forgot
to
put
area
slash
contributor
guide
as
your
label.
You
know,
and
that
feels
like
something
that
could
yeah
awesomely.
You
know
catch
things
like
that.
You
know
it
definitely
be
the
tuning.
The
model
to
is
actually
the.
E
Yes,
it
would
be
lovely
if
we
could
actually
do
some
automated
sig
classification,
because
I
would
say
that's
one
of
the
most
new
be
hostile
things
yeah
when
they
actually
file
an
issue
and
it
and
they
get
an
immediate
notification
from
the
bot.
You
must
attribute
this
as
you
do
a
sig
or
you
know,
and
that
often
requires
a
lot
of
knowledge
on
the
part
of
the
new
person
that
they
don't
necessarily
have.
B
I
Thing
Mary
Esther,
the
Train
thing
there
is
George
is
filing
issues
so
for
us,
when
we,
when
somebody
opens
up
ER
and
we
know
exactly
which
files
they're
touching,
we
can
do
things
like
assign
interviewers
us
to
like
testing
approvers.
We
could
also
like
our
current
automation,
can
assign
labels
like
if
you
touch
X
directory
automatically
adds
these
three
labels
to
it.
I
So
we
could
do
a
lot
of
that
kind
of
stuff
with
owners
files
in
our
current
automation,
if
they're
opening
yard
the
problem
is
issues
don't
touch
file,
so
that
gets
really
tricky
as
far
as
doing
those
those
kind
of
like
triaging
label
suggestion
things,
because
there
isn't
a
direct
tie
between
an
issue
and
the
source
code
file.
Gotcha,
okay,
I
was
just
wondering.
D
H
J
H
Yeah
we
have
just
to
kind
of
catch
everybody
up
like
we
all,
so
we
got
when
I
first
had
a
discussion
with
kubernetes
community,
everyone
sort
of
said.
We
don't
need
this.
We
really
want
no
difficult
with
notifications,
and
so
we
opened
up
a
separate
thread
with
with
a
product
manager
at
github
who's
been
connected
with
some
of
you,
folks
and
sort
of
trying
to
work
on
work
on
that
yeah.
B
H
B
E
B
H
B
A
Okay
and
let's
go
to
Paris
or
at
the
meeting
proposal.
C
Yes
mo
fYI,
that's
awesome!
Thank
you.
So
much
for
coming
I
really
appreciate,
like
the
github
feedback
loops
that
we've
been
seeing
with
all
of
your
teams.
It's
been
super
awesome,
so
I
just
wanted
to
say
that
meeting
proposal.
So
we've
been
doing
this
song
and
dance
weekly
now
for
over
a
year,
and
we
clearly
have
a
lot
of
things
to
talk
about.
There
are
tons
of
work
streams
going
on
in
this
group
and
I
wanted
to
propose
that
we
move
to
more
of
a
cluster
lifecycle.
C
You
know
very
similar
to
stand-ups
that
you
may
or
may
not
have
you
know
at
your
work
or
whatever,
but
then
the
sub
projects
would
ultimately
meet.
They
can
either
do
once
a
month,
Lok
stand-ups
they
could
do
once
a
week.
You
know
they
could
do
bi-weekly.
It
just
depends
on
what
they
want
to
do.
As
far
as
their
work
streams
are
concerned,
I
wanted
to
I'm
gonna,
send
out
something
on
the
mailing
list,
but
I
wanted
to
run
it
by
here.
C
C
Any
questions
concerns
comments.
Christoph
said
+100
this
and
by
the
way,
like
I
said
proposal,
I
mean
like
this
is
a
gonna
go
into
action
next
week,
like
we're
gonna
talk
about
this
also
there's
gonna
be
a
list
that
are
sent
out
with
like
what
is
a
sub-project
owner
to
us,
because
that's
something
that
the
project
also
we
need
to
work
on.
It's
like
what
is
a
sub
project
owner
for
this
we're
identifying
sub-project
approvers
at
the
root
of
the
sub-project.
C
D
Only
dissent
I
have
isn't
really
dissent.
It's
there
are
a
lot
of
meetings
and
I
am
very
concerned
about
scheduling.
That's
the
only
thing
that
concerns
me.
I
I
am
all
for
structure
and
hell
yeah.
Let's
do
it.
I'm
just
I
personally
have
concerns
that
there's
gonna
be
a
point
where
it's
like
well.
I
can't
make
any
meetings
because
I
have
other
meetings.
Yeah
I
feel
I.
C
I
think
that
this
will
I.
This
is
what
I'm,
hoping
by
the
way
and
I've
done
some
experiments
and
other
smaller
work
streams,
I'm,
hoping
that
this
means
that
you
don't
feel
like
you
have
to
come
to
the
bi-weekly,
because
the
bi-weekly
is
just
to
stand
up,
and
it's
like
the
sub-project
folks
who
are
like
heavy
in
the
streams
and
merging
stuff
will
be
on
the
by
weeklies.
C
B
From
from
the
SEL
ones
I'm
sitting
in
to
because
they're
focused,
they
really
break
their
stuff
down.
It's
like,
like
their
meetings,
end
up
being
shorter
because
it's
like
here's,
our
ten
minutes
stand
up.
Boom
boom
boom
boom
boom
and
then
one
of
them
and
they
can
rotate.
That
out,
is
the
one
that
goes
to
the
main
meeting
later
and
here's,
like
you,
know:
hey
federal
government,
here's
all
the
stuff,
the
state
government.
That
kind
of
thing
you
know,
as
opposed
to
feeling
like
you,
have
to
sit
through
a
mega
board.
Yep.
C
I
The
regular
cadence
of
contributor
experience
make
sure
that,
like
all
the
projects
are
on
the
same
page,
but
individual
work
groups
may
not
need
to
meet
as
frequently
maybe
may
choose
alternative
ways
to
meet
may
meet
in
a
more
ad
hoc
fashion.
But
as
long
as
like
inside
the
sub
project,
you're
communicating
and
the
sub
project
is
communicating
up
to
the
sig.
Then
yeah
it
may
result
in
fewer
meetings.
Hopefully
it
does
not
result
in
more
and
if
it
does,
that's
definitely
something
that
we
should
talk
about
it
readable
here.
Yeah.
C
That
is
not
the
intention
whatsoever,
like
the
intention
is
for
us
to
work
faster
and
be
able
to
communicate
better.
That's
the
intention
here.
So
all
right,
well,
I
didn't
hear
any
major
objections.
I'm
gonna
go
ahead
and
with
the
email
on
the
mailing
list
and
we'll
do
our
usual
song
and
dance
there
and
then
take
action.
Like
I
said
this
is
like
July
and
stuff
like
that.
It's.
C
C
F
I
C
F
K
C
So
when
I
talk
about
marketing
team
I'm,
not
necessarily
talking
about
like
a
CN
CF
marketing
team
I'm
talking
about
people
that
are
going
to
market
the
work
that
we
do
within
this
SIG
because
communication
is
get
is
the
most
important
thing
that
we
can
provide
to
this
organization.
So
if
stuff
like
this
interests,
you
please
reach
out
and
speak
up,
because
this
is
definitely
the
forefront.
I
mean
you
can
see
some
of
the
role
books
they
like
to
do
in
sections.
C
That
means
yes
to
do
it,
but
the
other
group
too,
that
we're
trying
to
spring
off
ASAP
is
the
triage.
Excuse
me,
the
triage
team
that
will
give
Makita
and
other
folks
who
triage
heavily
in
the
repo
some
relief
and
then
also
some
insight
to
get
some
like
mentoring
opportunities
going
there.
So
that's
all
I
have
to
say
roles
are
getting
created,
please
tack
on
also,
if
you
think
like
we
should
have
a
team
for
something
like
talk
about
it,
that's
what
these
meetings
are
for.
So
that's
my
spiel
I'm
sticking
to
it.
E
Just
a
quick
appeal
to
the
group
I
put
a
graph
in
for
tomorrow's
contribute
community
meeting.
There's
one
of
the
few
people
who
regularly
checks
dev
stats.
One
of
the
things
I
noticed
is
that
since
February
we've
been
closing
according
to
have
stats,
we've
been
closing
PRS
faster
and
by
faster
I
mean
substantially
faster,
as
in
the
average
you
know,
existence
or
meeting
existence,
time
of
a
PR
between,
say,
November,
2018
and
March.
2019
went
down
by
half
the
I
can't
through
the
data,
pinpoint
anything
in
particular.
That
would
lead
to
that.
E
It
seems
to
be
sort
generally
distributed
motive.
Anybody
knows
it's
particularly
associated
the
changes
particularly
associated
with
February,
so
if
anybody
knows
something
that
happened
in
February,
that
would
have
caused
people
to
start
closing,
PRS
faster
or
for
that
matter,
opening
more
short
lived
to
be
ours.
These
would
be
please.
Let
me
know
about
that
before
I
present
it
on
a
recorded
YouTube.
L
L
E
I
mean
what
I'm
seeing
seems
to
indicate
that
we
started
getting
more
PR
we've
gradually
ramped
up
the
number
of
years
that
we
have.
We
started
getting
more
PRS,
but
like
the
additional
PRS
that
we
give
get,
are
shorter
lifts,
as
in
we
still
have
long
lived
PRS,
but
we
have
a
lot
more
short
live
PRS.
That's
what
the
distribution
looks
like.
L
L
E
B
You
know
and
I
just
think
it's
after
cube
con
now
give
someone
like
kind
of
like
a
cool,
hey,
y'all
are
doing
really
great
check
this
out,
because
sometimes
a
project
is
so
big
and
we
feel
like
every
PR
takes
like
two
years
to
get
in.
It's
just
nice
to
you
know,
show
people
some
data
and
be
like
look.