►
From YouTube: Weekly Package Team Meeting 08-26-19
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
So
I
want
to
share
with
everyone
that
the
survey
that
I've
been
working
on
putting
together
is
going
into
Qualtrics
later
today.
What
this
means
for
everyone
is
hopefully
tomorrow,
I
will
have
the
survey
ready
and
I
would
really
appreciate
everyone
on
the
team
if
you
could
go
through
the
survey
on
yourself
and
just
kind
of
record
what
your
experience
was
like.
B
This
is
just
a
chance
for
me
to
test
out
to
make
sure
you
know
the
questions
make
sense
that
they're
lining
up
with
you
know
your
expectations
as
an
engineer
and
I'm
hoping
to
get
that
wrapped
up,
so
that
we
can
actually
share
it
next
week.
So
that's
just
a
heads
up
on
reviewing
it.
The
other
thing
I
would
like
to
talk
about,
and
maybe
we
could
take
this
to
a
sink
conversation
is
more
around.
How
do
we
get
this
survey
out?
B
We
need
to
get
a
lot
of
people
a
little
bit
over
a
thousand
to
get
a
good
confidence
rate.
So
that's
a
lot
of
people
so
trying
to
figure
out
creative
ways
to
get
more
people
to
take.
The
survey
would
be
really
helpful
if
I
can
get
everyone's
insights
on
that,
so
just
want
to
give
everyone
a
heads
up.
That's
coming!
That's
gonna,
be
this
weekend
next
week,
so
just
kind
of
think
on
it
well
touch
base
from
there.
B
They're
gonna
send
it
to
the
existing
panel.
The
panel
isn't
very
large,
so
this
is
something
that
we'll
probably
have
to
find
some
other
creative
ways
to
get
people
to
take
the
survey,
whether
it's
pinging
people
that
are,
you
know,
writing
on
our
issues
or
you
know,
start
sharing
it
on
Twitter.
There's
a
couple
different
options.
A
Okay,
cool
all
right,
so
the
next
item
I
have
is
reviewing
the
12.4
milestone.
So
I
have
the
link
included
there
to
just
pull
up
the
issue
list.
If
you
want
to
pull
it
up,
I
have
it
sorted,
I
know
Dan
hates.
When
I
share
my
screen,
but
I,
it's
I
have
it
sorted
to
manual.
If
you
look
in
this
sword,
drop-down
box,
the
manual
sort,
will
pick
up
the
sort
that
I
included
in
the
the
board
list.
So
it's
in
priority
order.
A
We
could
go
through
each
item.
There,
sound
good
okay.
So
the
first
item
on
there
is
the
docker
distribution
pruner
from
experimental
to
GA.
This
could
be
an
investigation
task.
This
is
a
Dan
and
I
were
talking
last
week
about
whether
or
not
we
should
prioritize
forking
the
doctor
code
and
optimizing
the
garbage
collection
code,
that's
there
or
taking
what
Camille
has
already
done
with
this
docker
distribution
or
distribution
pruner
code,
which
is
working
for
some
customers.
A
Sara
I
mentioned
she's
using
this
for
some
people,
Don
support
calls,
but
it's
I've
seen
where
there's
issues
with
back.
It
causes
issues
with
backups
or
some
other
things
where
it
just
doesn't
work,
but
we
thought
this
would
be
the
fastest
way
to
improving
the
garbage
collection
story
for
some
of
our
big
customers,
some
that
we
met
with
last
week
that
are
having
you
know:
55
terabytes
of
storage,
that's
reasoning
in
their
registry
and
they
just
can't
use
the
current
version
of
garbage
collection.
A
C
So
DNC
Jack
yeah
I've
been
looking
through
the
code
and
the
list
of
issues
most
of
the
issues
in
that
project
kind
of
at
best,
a
year
old,
so
I
think
I've
been
trying
to
go
through
those
issues
to
see.
If
there's
anything,
we
can
sort
of
start
addressing
relatively
quickly,
but
we
might
end
up
having
to-
and
this
is
the
issues
in
the.
C
Visit
the
package,
but
if
the
pruna
sorry
they
contain
a
registry
pruna
bucket
ashish
and
pruna
or
whatever
he's
called
it,
and
so
I've
been
trying
to
go
through
these
issues
to
figure
out
than
anything
that
can
actually
just
kind
of
go.
Hey,
let's
go
do
this
and
it
seems
like
a
lot
of
them
goes
back
and
forward
with
people
who
are
trying
to
contribute
so
I
think
I
still
need
to
go
through
those
issues
a
bit
maybe
Tim
and
anyone
else
is
interested.
C
A
Okay,
great.
The
next
item
is
a
net
new
package
manager,
integration
with
nougat.
This
is
one
of
our
I
think
it
is
our
stages.
Highest
requested
feature,
that's
something
like
127
votes
now,
I've
had
this
a
few
customers
recently
request
this
as
well.
I
still
need
to
groom
this
issue,
so
it's
a
little
out
of
date,
but
that's
the
next
item
that
we
need
to
work
on
and
I
should
have
that
done
this
week
for
sure,
but
expect
it'll.
D
I,
don't
think,
there's
anything
in
particular
one
way
or
the
other.
Certainly
like
you
know,
more
information
is
better
I,
think
it
yeah
like
as
a
sort
of
having
this
top-level
issue
that
can
eventually
be
broken
down
by
whoever's,
working
on
it
as
needed.
I
think
that's.
What's
a
good
good
place
to
start,
okay,
cool
and.
C
Tim
I
think
the
other
big
lesson
that
we
should
take
away
from
how
we
working
on
Conan
was
that
it
took
a
lot
of
investigation
up
front
by
Krause
to
get
to
a
point
where
we
could
kind
of
move
forward
and
even
start
breaking
things
up.
I
think
we
want
to
I
think
you
would
have
done
this
already,
but
it's
worth
mentioning
is
to
say
that
we
should
have
had
some
sort
of
investigation
effort
included
upfront
so
that
people
have
the
opportunity
to
do
that.
A
I
think
we're
thinking
that
any
time
we
do
one
of
these
integrations
it
would
take
two
milestones:
one
for
sort
of
the
investigation
and
getting
the
API
stubbed
out
and
then
the
second
one
for
finishing
it
and
testing
it
and
getting
it
working.
But
that's
what
Steve
and
I
talked
about
of
these
so
yeah.
A
C
C
So
that's
the
goal
of
the
investigation
not
to
fully
document
all
other
possibilities,
because
that
ends
up
putting
us
behind
a
whole
set
of
the
investigation.
It
yes
delivers
knowledge
and
help
us
understand
how
to
make
decisions
and
that's
awesome,
but
we
should
focus
that
investigation
effort
on
how
do
we
deliver
an
MVC
a
most
minimum?
Viable
change?
That's
actually
useful
in
this
area
from
this
investigation,
because
then
we
can
like
make
sure
we're
not
just
sort
of
researching
for
a
month
or
whatever.
There's
things.
A
C
A
B
A
D
And
I
did
create
as
a
side
note.
It
created
an
issue
for
kind
of
just
discussing
like
when
we
are
implementing
new
package
systems
in
general
sort
of
you
know
things,
we've
learned
along
the
way
and
maybe
some
notes
about
how
we
can
break
things
up
in
a
general
sense
of
when
we
attack
each
new
one
I'll
throw
a
link
in
here
as
well.
That
would.
A
But
basically,
if
you,
if
you
click
the
the
issue,
the
idea
is
that
on
the
group
list,
we're
not
doing
any
sorting
for
the
package
of
the
packages
list
and
we
can
so
I
think
that's
something
that
Nick's
already
started
to
tackle.
Actually
the
next
one
on
there
is
checking
for
docker
images
before
renaming
a
group.
A
So
if
you
remember
there
there's
some
issue
where,
if
you
have
a
bad
view
of
a
special
character
in
your
project
group
or
branch
name,
it
will
break
the
connection
to
docker
and
which
is
obviously
a
problem
and
we
have
logic
in
there.
So
if
you
go
to
change
your
project,
name,
you'll
get
a
warning
and
you
could
see
it
in
the
issue.
What
happens
now
you
get
like
a
big
red
banner.
It
says
cannot
rename
project
because
it
contains
registry
tags
that
we
should
just
do
the
same
thing
for
group.
A
D
Might
also
be
able
to
get
easily
grouped
with
I
think
there's
another
issue,
that's
pretty
much
the
same
thing,
but
for
NPM
and
maven
packages,
because
the
same
thing
will
happen.
Is
you
rename
the
group
or
the
path
to
the
group?
It'll
just
break
the
packages,
the
links
to
the
packages
so
I'm
assuming
I,
think
I
looked
into
it
briefly
and
it's
like
all
of
that
code
lives
in
the
same
spot.
So
it
would
be
easy
to
change
all
of
them
at.
A
D
C
So
it's
probably
worth
leading
just
in
here
sending
it
to
Tim
like
you
requested
and
then,
if
we
can
get
it
or
we
feel
like,
we
have
enough
capacity
to
do
both
and
we
would
we
would
put
it
in
I.
Don't
know
what
the
what's
Joan
you
might
be
able
to
speak
to
the
Gil
obvious
way
to
do
this.
I'm,
given
that
you've
been
here
longer
than
the
rest
of
this
yeah.
E
A
A
A
Documentation
that
we
have
for
the
all
packaged
registries
so
just
as
a
small
help
to
change
in
the
front-end
and
then
I'll
go
ahead.
Sorry
did
you
say:
you're
gonna
grab
that
oh
shoot
I
think
we
should
give
it
to
Nick.
So
because
if
it's
on
me,
then
I'm
gonna
be
responsible
for
it
and
if
I
don't
get
it
done
yeah
and
they
could
do
it
much
faster
and
I.
B
C
A
Know:
okay,
all
right,
so
three
more
I
have
the
authenticate
for
with,
for
CI
job
token,
for
Conan
and
for
NPM
I'm,
not
sure
the
priority
of
those
I
think
we
could
I
know
NPM
people
are
at
because
customers
are
asking
for.
It's
definitely
a
much
higher
requested
feature,
but
I
know
once
people
start
using
Coenen
will
get
it.
C
A
C
So
I
think
this
is
going
to
be
good
value
generally
so
I
like
that,
though
we're
sort
of
prioritizing
it,
it
feels
like
we
have
a
discussion
a
little
while
ago
on
pricing
and
what
we
should
have
in
the
the
free
tiers
or
you
know,
CEO.
That's
changing
so
I
feel
like
some
of
these
things,
where
it
might
be
worthwhile
at
an
off
Tim
and
you're
thinking
in
this
in
this
respect,
but
it
might
be
worthwhile
sort
of
having
a
think
about
whether
some
of
these
things
are
working
well
together.
C
A
Know
I
mean
well,
we
were
probably
I.
Think
the
discussion
for
me
should
be.
Should
the
package
registries
be
an
enterprise
feature
or
a
consumer
feature
in
my,
in
my
opinion,
they
I
think
I
agree
with
Jason's
point
in
that
issue
that
they
should
be
a
consumer
feature,
but
that
is
she's
still
going
through
discussion
and
some
higher-up
strategy
to
figure
that
stuff
out.
But,
in
my
opinion
out
I
would
love
it
for
her
tobe
consumer.
C
Yeah
it
feels
like
to
me
and
and
I
don't
want
to
like
get
too
ahead
of
where
we're
supposed
to
be
anyway.
But
to
me
it
feels
like
they
probably
should
be.
Then
I
might
not
have
felt
that
way
a
couple
of
weeks
ago.
It
feels
like
they
should
be
consumer
and
it
feels
like
some
things.
Like
you
know,
the
one
of
the
items
we
have
to
discuss
in
here
should
be
sort
of
enterprise
level,
the
security
scanning
stuff
for
the
actual
proxy
coming
up
there,
I'm
good
yeah,
I.
A
E
B
A
Be
front-end
but
it'll
definitely
it'll,
probably
be
very
close
to
the
same
front-end
that
we
have
for
and
cream
and
maven
yeah
just
different
metadata
that
fits
into
those
columns.
Conan
was
a
little
bit
different
because
we
had
a
expandable
settings
piece
and
I.
Don't
think
nougat
will
have
that
it
seems
a
little
like
they
don't.
It
doesn't
carry
as
much
of
that
metadata.
But
okay
I'll
know
more
in
the
next
couple
of
days
as
I
as
I.
A
Look
at
it
I'm
just
been
struggling
to
get
it
running
on
mic
Peter,
because
it's
I
have
a
Mac
and
not
a
Windows
machine
and
so
I'm
just
working
through
some
of
those
kinks
they
have.
They
have
a
something
you
could
use
called
Sol
or
something
like
I
forget
what
it's
called,
but
I
just
wanted
to
make
sure
you
know
I've
just
been
struggling
with
it.
So
yeah.
D
E
A
They
have
something
have
another
download
that
you
have
to
do
and
use
it,
and
then
you
can
use
it
from
Visual
Studio
or
you
could
use
from
the
command
line.
I
think
we
want
to
discover
the
command
line
in
this
case,
and
this
in
this
case,
but
yeah
we
do
have
to
decide
between
core
and
standard
I,
think
I
forget,
which
one
we
should
lean
to
it.
Whichever
one
is
like
the
more
widely
accepted
now,
we
should
use
yeah.
A
A
Ok,
yes,
like
I
mentioned
the
next
one
on
there
is
the
dependency
firewall,
which
is
a
net
new
category
that
came
up
from
a
conversation
with
Sid
and
Kenny,
where
they
were
talking
about
us
doing
something
like
this
and
building
off
of
the
dependency
proxy
to
build
in
some
additional
security
features.
The
you
know,
J
frog
has
their
x-ray
product
and
Sunnah
type
has
a
firewall
product
everybody's
incorporating
this
stuff.
You
know,
NPM
Enterprise
is
trying
to
build
their
whole
value
chain
off
of
security
scanning.
A
So
it
seems
like
something
that
we
should
do
and
I
created
a
new
category
and
some
content
for
the
strategy
and
Direction
page
along
with
some
skeleton
issues
that
we
could
review,
but
I
just
wanted
to
make
sure
everybody
was
aware
of
it
and
had
the
chance
to
in
the
issue,
provide
any
feedback
or
thoughts
before
it
got
merged
or
if
you
had
any
questions,
I'd
be
happy
to
answer
what
I,
what
I
can?
Even
though,
knowing
it's
not
we're,
not
gonna,
get
to
this
until
2020.
A
A
A
And
it's
not
it's
not
quite
right,
I,
guess
naming
why
so
we
we
ended
up
at
dependency
firewall.
We
could
always
change
it
if
we
had
a
good
justification.
I
like
it,
because
it
helps
brand
that
section
that,
like
if
we're
talking
about
making
dependency
proxy
and
dependency
firewall,
the
enterprise
features
and
futures
that
you
pay
for
I,
think
it
helps
brand
them,
and
it
says
exactly
what
it's
doing,
which
I
think
is
also
the
gitlab
way.
So
I
thought
said:
did
a
good
job
with
naming
it
but
yeah
week.
C
But
what
I
am
looking
at
here
is
I
think
there's
a
couple
of
things:
one.
We
ought
to
be
really
careful
that
we're
trying
to
leverage
the
security
teams
work
as
much
as
possible
here.
I,
don't
want
to
be
recreating
the
wheel.
Ideally
in
my
head,
we
we,
as
a
package,
gets
loaded
into
the
proxy
it's
scanned
by
those
same
sets
of
tools,
if
possible,
because
you
know
trying
to
maintain
sigh
understanding
of
how
this
stuff
fits
together.
C
Is
it's
gonna,
be
a
lot
of
a
big
lift
for
a
small
team
like
us,
I'm
open
to
make
sense
right?
Sorry,
that
was
a
very
dismissive
way
to
put
it
if
it
doesn't
make
sense
that
I'm
happy
to
hear
arguments
other
way,
sorry
and
and
I
think
I
think
thing
I
was
gonna
say
is
I
think
we
ought
to
be
discussing
that
the
comments
of
the
eraser
and
the
dependency
proxy,
that
I
can
just
add
a
flag
to
the
death
distribution
registry,
have
run
it
in
proxy
mode
and
what
the
dependency
proxy
does.
C
C
So
that
could
we
do
you
want
me
to
add
an
issue
to
investigate,
or
do
you
wanna
add
that
issue
or
it
doesn't
have
to
be
in
here,
but
we
should
I
think
we
ought
to
be
discussing
this
before
it
before
it.
Sort
of
so
police
testing,
a
bunch
of
method
in
dependency
proxy,
just
to
make
sure
we're
clear
on
what
we
should
be
providing
and
in
our
intention
is
to
provide
further
support
for
partying
in
the
dependency
proxy
for
other
stuff
other
than
doctor
distribution
registry
or
a
k2
container
registry.