►
From YouTube: Package Managers Weekly 2019-08-13
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
A
B
You
know
I
have
some
options
for
what
I
do
next
week.
It's
no
longer
super
constrained
to
like
it
must
be
a
and
then
B
and
C.
So
I
have
some
thoughts,
but
if
anyone
like
has
any
priorities,
mostly
I,
don't
know,
Andrew
Andrew
has
any
priorities
of
anything
you'd
like
to
see
from
me.
Then
then
that
would
be
helpful.
Basically
like
the
PR,
the
PR
is
required
to
make
the
IPS
things
usable
will
hopefully
are
will
end
soon,
regardless
of
how
much
time
I
put
into
it.
B
It's
basically
like
just
it's
like
this
close
just
a
little
bit
of
back
and
forth,
and
so
after
that,
it's
like
all
right,
cool
like
what
next
and
I
can
either
start
working
on
more
performance,
related
pub/sub
things
which
needs
to
get
done
and
have
ongoing,
PRS
or
I
can
work
on
things
like
make
sure
that
your
arcane
s,
things
are
pinned
our
pin
locally
and
we'll
restart
publishing
them.
When
you
restart
your
daemon,
they
all
have
to
get
done,
but
if
there's
a
priority
ordering,
then
you
get
to
know
that.
A
C
D
A
A
A
A
B
I've
gotten
used
to
some
of
the
bouncing
in
the
sense
like
I'm
gonna,
one
thought
I
had
and
I
talked
with
Stephen
a
little
bit
about.
This
is
seeing
if
there
is
as
part
of
trying
to
like
separate
out
the
IPS
stuff
into
a
separate
library
in
which
I
should
add
on
here.
As
a
thing
to
do,
it's
not
really
a
feature
as
much
as
like
make
things
easier
for
people
that
plus
the
pub/sub
stuff,
see
if
there's
ways
I
can
make
this
like
slightly
reusable.
B
Relays,
given
that
everyone
is
behind
an
app-
and
maybe
we
shouldn't
do
that,
because
you
should
just
know
who
they
are
so
there's
like
a
few
options
about
that
and
there's
an
issue
I
can
link
to,
but
it's
sort
of
I'm
just
trying
to
keep
like
another
goal
in
mind.
Given
that
the
file
system
package
manager,
IPS
goal
is
not
NEADS
has
much
fewer
has
was
sort
of
designed
to
have
much
fewer
IQs
demands,
because
at
the
time
it
was
super
unreliable.
E
E
B
B
E
Doing
testing
testing
stuff
and
proving
proving
JSI
PFS
tests
and
that's
his
main
okay
I've
heard
a
quarter.
I
feel
like
we've
flagged
this
in
a
previous
core
implemented
meeting
but
like
this
is
blocking
and
we
would
like
for
it
not
to
be
blocking,
and
so
like
yeah
I
like
it
feels
like.
We
can't
land
this
super
nicely
until
we
have
a
spec
for
it,
because
we
can't
communicate
what
it
is
and
how
it
works.
F
D
B
Yeah,
that's
fine.
We
can
flag
that,
although
I
don't
think
it's
gonna
be
like
necessarily
just
a
couple
of
days,
because
there
hasn't
been
a
crazy
amount
of
review
on
the
thing
and
so
before
it
becomes
this
back
and
gets
moved
over
to
start
doing,
go
things.
We're
gonna
need
to
have
some
review
time
and
make
sure
we're
all
yeah.
D
B
B
A
C
B
A
So
knowing
worked
following
up
with
lots
of
people
which
doesn't
feel
like
work,
but
it's
important
work
and
maybe
I'll
mail,
all
of
you,
some
cookies
or
something.
If
we
finish
a
thing
and
can
say
like
I
pianist,
is
good
faster.
You
know
many
people
would
be
super
happy
about
that
lots
and
lots
of
people.
A
B
A
Okay,
anyone
else,
who's
that
or
who
would
like
to
be
up
for
the
week,
even
if
you're,
just
saying
I'm
good
to
go,
I
know
what
is
happening
this
week.
Just
let
us
know.
F
I
can
go.
Thank
you
yeah,
so
had
a
chat
with
Steven
yesterday
about
taking
on
the
performance
work
around
adding
files
and
go
I
haven't
had
a
chance
to
start
yet
because
part
of
the
say
there
was
an
issue
opened
on
the
Jedi
beavers
repo
about
some
of
these
having
problems
adding
large
folders
to
like
to
chose.
Forgive
us
something
somewhere
is
eating
all
the
memory,
but
it's
not
a
memory
leak
because,
after
a
while,
the
memory
that
it's
consuming
gets
garbage
collected.
F
So
what
it
is
is
chairing
this
refactor
from
pool
streams
to
a
think
iterator
something
somewhere
in
the
memory
usage
profile
has
changed.
That
means
that
all
the
stuff
is
now
bucking
up
into
memory,
which
was,
in
his
case,
was
causing
the
process
to
get
killed
because
he's
running
it
on
a
machine
that
didn't
have
a
whole
lot
of
memory
and
is
a
virtual
machine.
My
suspicion
is
with
no
swap
space
configured
good.
F
So
how
did
it
work
chat
without
and
we're
gonna
try
and
remove
a
lot
of
the
misdirection
that
occurs
to
a
stream
of
bytes
between
you,
adding
it
and
it
getting
into
IP
FS,
which
should
make
it
a
little
easier
to,
because
I
can
see
that
from
taking
lots
of
heap
dumps
and
comparing
them
all.
The
data
is
backing
up
in
node
streams
like
in
the
streams
implementation
in
node
core,
so
there's
basically
a
there's
a
function
in
there
that
says
like
can
I.
Can
I
omit
this
data
or
should
I
buffer?
F
It
Mena's,
buffering
it
and
crane.
This
enormous,
linked
list
structure
stretches
on
forever,
so
yeah
we're
either
not
assuming
it
quickly
enough
or
we're,
probably
not
this
human,
basically,
which
means
it
all
gets
backed
up,
but
we
should
be
able
to
say,
don't
consume
this
data
this
quickly
to
the
stream.
The
high
automatic
basically
says
you
know:
I
don't
want.
This
is
the
size
of
Trump
I
want,
don't
give
me
too
much
data
and
that
is
getting
lost
somewhere,
and
so
we
convert
between
async
iterators
and
streams
and
pool
streams.
F
Quite
a
lot
in
the
current
code
base,
but
not
too
badly
in
the
file
import
path,
which
is
when
I'm
looking
at
the
bit
that
is
outstanding
on
that
is
the
actual
importing
the
files
from
the
ipfs
layer
before
it
reaches
the
unix.
The
first
manipulations
I'm
trying
to
take
out
all
the
pool,
streams
and
stuff
in
that,
and
so
it
be
also
a
sink.
It's
races.
And
so
that's
what
I'm
doing
at
the
moment,
sort
of
emergency
everything's
on
fire
kind
of
affair.
A
G
Yeah
so
I've
been
working
quite
a
bit
with
banner
in
their
collaboration,
so
this
week,
I'm
hoping
to
get
to
a
point
where
we
can
run
some
benchmarks
on
their
test
lab.
So
we'll
have
something
compare
about
compare
against
as
we're
making
improvements
to
bit
swap
in
the
meantime,
I'm
also
working
through
some
idea,
Stephen
about
how
we
can
make
this
work
perform
better
for
this
use
case
and
just
in
general,
so
yeah
just
on
this
week.
Any
questions.
G
H
Am
I
audible,
yes,
perfect
sure
so
this
week,
I'm
just
basically
going
to
be
continuing.
The
work
I've
been
doing
I'm
a
little
behind
on
it,
just
because
of
some
unexpected,
like
issues
in
particular
at
the
moment
not
to
get
too
technical
into
it,
but
like
I'm,
dealing
with
an
issue
where
the
core
API
is
giving
me
back
some
data
that
I
don't
expect
it's
like
extra
data.
So
it's
a
little
strange
but
yeah
I'm.
It's
it's
like
this
highly
concurrent
thing.
H
So
I
have
to
constantly
be
stepping
through
the
debugger
in
an
arduous
way,
so
I'm
a
little
little
behind
on
that.
But
plan
to
you
know
continue
on
this
and
get
it
done.
H
A
A
I
guess
there's
me:
okay,
I
think
it's
an
update
which
is
I'm
still
thinking
a
bit
between
the
cluster
stuff
and
the
stuff
which
is
making
my
brain
hurt.
But
let's
see
so
this
week,
I
finished
that
service
blueprint
at
least
enough
for
first
drafts
like
to
share
it
with
the
world
and
write
up
a
thing
on
the
package
managers
Agrico
that
describes
what
that
isn't.
Where
it's
going
and
as
part
of
that
see,
if
we
can
link
oh,
these
will
not
see.
A
I
will
link
all
these
efforts
and
a
summarized
way
into
the
bits
of
that
chart
or
not
like
how
everything
fits
together.
So
hopefully
we
have
a
little
bit
of
a
better
overview.
We
can
share
a
more
accessible
overview
of
why
we're
doing
with
things
that
we're
doing
for
the
broader
community
and
for
us,
because
I
can't
keep
my
head
straight
from
one
day
to
the
other,
as
you
can
hotel
and
I
Oh
Molly
I
owe
you
a
couple
things
around.
Okay
are
some
stuff
I'll
get
there
sorry
I.
A
A
A
B
C
One
thing
I
I
run
into
this
week:
it's
the
word.
Discovery
has
different
meanings
depending
on
the
community,
the
people
that
you
don't
think
there's
very
much
not
looking
at
discovery
from
a
peer
find
me
page
find
me
content
that
please
have
was
more
like
how
do
I
search.
Think
search
is
the
other
word,
although
and
there
being
quite
overloaded
as
well
in
that
I,
don't
Navy
identifies
I
want
to
be
able
to
ask
my
ps4,
so
I
need
to
work
out
what
they
are,
and
rather
then
I
know.
C
B
So
there
are
some
schemes
that
do
like
in
the
same
way
that
for
pub/sub
we
have
like
here's
a
channel
and
then
there's
a
bunch
of
peers.
You
can
also.
There
are
schemes
where
you
have
like
DHT
is
and
you're
putting
like.
You
know,
I,
you
know
my
name
is
you
know
my
name
is
Joe
and
then
there's
like
a
thousand
people
who
like
claim
their
name,
is
Kjell,
and
then
you
figure
it
out
from
there
and
you
sort
of
it's
like
tag
based
search.
B
A
C
A
package
managers
conference
happening
in
October
I
found
about
about
two
minutes
after
booking
my
flights
for
a
holiday
on
the
same
day
last
night,
it's
invite-only
but
I
think
I
can
kind
of
say
like
well.
If
you've
invited
me
and
I
can't
go
then
invite
this
person
instead
and
someone
recommended
that
I
say
that
Ares
should
go
I
guess.
C
The
main
thing
is
that
they
shouldn't
just
be
like
oh
github
package
repository
is
going
to
solve
all
our
problems
and
totally
maybe
have
like
an
opposing
point
of
view
there,
rather
than
just
someone
who
wants
you
have
to
replace
NPM
and
that'll
solve
all
of
our
problems:
people
organizing
it
get
up
the
people
behind.
You
have
package
repository
because
they
now
have
vested
interest.
C
C
Yeah
I
mean
that
that
areas
like
highlighted
are
like.
Oh
all
of
this
wants
a
package
made.
Yeah
hold
it,
but
it's
also
like
this.
Those
are
these
are
bad
bits
of
packet
vengeance.
They
probably
need
to
be
fixed
to
it,
and
you
can
definitely
like
just
getting
people
in
the
same
room
together
might
end
up
being
like.
Oh,
my
god,
business
photos
bonds.
The
way
that
funds
us
working
on
practical
matters
will
be
will
be
undermined
if
get
out
take
control
of
this.
So
that
might
also
happen
and.
D
C
So
three
dates:
people
tend
to
buy
artifactory
or
some
type
nexus
or
impairment
prize
or
packages,
PHP
Enterprise,
alongside
their
github
enterprise
and
so
get
up
alike
and
get
lab
already
have
built
in
NPM
registry.
So
there's
a
lot
of
chasing
whatever
get
lab
doing
as
well
and
so
there's
like
and
they
built
it
like
three
years
ago,
and
they
struggle
to
ship
it
until
very
recently.
So
there's
like
what
they
want
to
happen
and
what
actually
happens
per
this
there's
a
long
period
of
time
in
between
these
things,.