►
From YouTube: Open RFC Meeting - Wednesday, Nov 13th 2019
Description
In our ongoing efforts to better listen to and collaborate with the community, we're piloting an Open RFC call that helps to move conversations and initiatives forward. The focus should be on existing issues/PRs in this repository but can also touch on community/ecosystem-wide subjects.
GitHub Agenda Issue:
https://github.com/npm/rfcs/issues/66
Meeting Notes:
https://docs.google.com/document/d/14wOHOIWF5D08wywfRMz_MYQyRF0NQIkA5J9F6VzwEbs/edit
A
Welcome
everyone
to
another
open
RC
car
at
p.m.
RC
car
appreciate
everybody
jumping
on
today
again
for
anybody
that
missed
the
Lina.
It's
link
is
in
and
feel
free
to
add
yourself
as
an
attendee
and
we'll
go
through
essentially
the
agenda
that
is
posted
there,
as
well
as
the
open
issue
on
the
RCS
repo.
Oh,
it's
a
little
clipped
that
better
step
back
for
audio
okay,
and
so
again
you
can
go
add
yourself
as
an
attendee
there.
We
will
essentially
be
going
through
that
agenda
going
forward.
A
If
you
want
to
get
anything
on
the
agenda,
we
actually
are
automating
this
now,
so
ensuring
that
a
issue
or
PR
on
anyone,
the
NPM
projects,
has
the
label
agenda.
It
will
get
surfaced
at
these
calls,
and
so
everything
you
see
on
the
agenda
today
essentially
was
an
item
that
was
labeled
with
that
with
that
agenda
label
and
going
forward.
B
A
Yeah,
we
can
add
that
for
sure,
so
digging
into
the
first
agenda
item
essentially
issue
number
58.
This
was
a
a
poll
that
we
ran
after
our
last
call,
essentially
trying
to
find
another
alternating
time.
So
we
changed
the
open,
RCE,
bi-weekly
initial
call,
time
to
2:00
p.m.
on
what
2:00
p.m.
Eastern
on
Wednesdays,
and
it
seems
like
that
is
the
sort
of
what
the
majority
of
folks
are.
Okay
with
so
this
poll.
This
is
a
second
poll
and
it
looks
like
the
outcome
is
essentially
let's,
let's
stay
with
2:00
p.m.
A
Eastern
on
Wednesdays
going
forward,
so
we
don't
have
to
do
any
work
there.
We
essentially
just
aren't
going
to
have
alternating
time
slots.
We
can
revisit
this
in
the
future
if
we
want
to
but
I
think.
The
action
item
here
is
for
me
is
to
remove
the
label
here
from
the
agenda:
I,
remove
the
agenda
label
and
then
to
close
out
this
this
poll.
A
C
Yeah,
so
this
is
a
a
reasonable
request.
I
think
it's!
It's
basically
saying
that,
since
you
could
never
reuse
that
same
version
number
again,
if
you
fetch
a
tarball
that
has
been
unpublished
rather
than
returning
404,
we
should
return
a
410
status
code.
It's
a
change
to
the
registry,
so
it
does
kind
of
fall
outside
of
the
the
purview
of
this
team
somewhat.
C
We
should
basically
resolve
that
to
a
new,
a
new
version
of
the
same
package
right,
a
new
version
that
that
meets
the
requirement
part
of
the
challenge
with
this
right,
then
the
other,
the
other
workaround
that
people
have
today
is
you
just
delete
your
lock
file
and
run
NPM
install
again,
which
can
which
can
result
in
updating
a
whole
lot
of
things?
That
might
not
be
what
you
want
right.
You
want
to
still
stay
sort
of
as
locked
as
possible.
C
C
C
Well,
so
it
has
the
idea
of
rolling
back
reification
process
if
there's
a
failure
in
reifying,
but
essentially
there's
there's,
there's
three
sort
of
main
stages
of
it
install
the
first
is
we
or
I
just
call
it
four
main
stages
of
the
install?
The
first
is
we.
We
read
the
packages
on
bisque
like
load,
the
actual
tree.
You
skip
that
friend
MCI,
because
we
just
delete
it
assume
it's
not
there.
C
Okay,
even
though
I
have
a
lock
file,
I
want
to
use
the
lock
file
as
basically
the
entire
ideal
tree
and
then
add
this
one
module
and
go
through
the
the
tree
building
steps
for
that
one
module,
and
so
we
have
to
fetch
the
metadata
for
it
and
then
incorporate
it
into
the
tree
and
then
the
third
phase,
once
we
have
that
ideal
tree,
is
to
fetch
everything
and
then
reapply
it.
Oh
I,
guess
that's
everything
and
then
the
fourth
phase
is
reify.
C
You
know
take
all
of
the
things
that
I
fetched
all
the
term
old,
I
fetched
and
dump
them
out
onto
disk.
So
the
the
the
challenge
that
this
suggestion
imposes,
which
I
think
is
not
it's
not
a
bad
idea.
It's
just
I'm,
just
calling
it
out
as
potentially
more
more
complexity
that
it
appears
is
if,
in
that
third
phase,
where
we're
fetching
all
the
tar
balls,
all
the
artifacts,
we
encounter
a
410.
We
essentially
have
to
roll
back
to
the
second
phase
of
building
the
ideal
tree
and
sort
of
restart
from
that
point.
C
D
I
think
this
is
like
an
NPM
seven
type
thing
and
also
that,
like
I
think
it's
like
perhaps
some
optimizations
that
can
be
made
there
and
they're
like
ideal
tree
phase,
you're
figuring
out
what
urgent
satisfy
what
and
like
a
short
it
should
be
request
to,
like
I,
mean
I'm
a
hash
time
when
that
particular
algorithm
takes
both
short
fetch
to
the
right
issue.
To
say
like
well.
I
would
like
this
version
because
it
satisfies
it,
but
does
it
actually
exists?
And
if
the
answer
is
true,
then
keep
going.
D
But
if
the
answer
is
false,
then
we
got
a
410
and
then,
like
you,
get
you
basically
suss
out
what
exists
and
what
doesn't
exist
before
you've
been
trying
to
finish
the
tar
balls
and
and
then
do
the
reification,
but
I
mean
that
and
then
it
bloats
that
step.
That's
that
prints
elements
that
none
of
us
he
saw
there
yeah.
C
So
that's
just
that's
just
the
complexity
added
by
you
know.
If
we
work
to
accept
this
I
think
the
bigger
question.
The
sort
of
semantic
question
is:
if
I
have
a
lock
file
or
a
shrink
wrap-
and
it
says
you
shall
have
this
particular
tar
ball
with
this
particular
integrity
value
and
then
that
tar
ball
is
gone
like
should
we
do
fix
that
problem,
or
should
we
surface
that
problem
to
the
you.
C
D
C
One
way
to
thread
that
needle
is
to
say,
like
you
know,
there's
a
separate
command
that
will
sort
of
I
re
design
the
tree
when
it
encounters
a
410
yet
or
what
it
fails
to
fetch
a
tarball
which
would
be
probably
the
ideal
user
experience
right
like
we.
You
know
we
surface,
we
say
like
hey,
I,
couldn't
I
couldn't
run
this
install
because
in
the
fetch
phase
one
of
the
tarballs
returned
to
410
run.
It
would
run
this
command
and
it'll
fix
the
problem.
A
Yeah,
like
that,
we
wanna
essentially
label
this,
then
for
MPM,
seven
and
I
think
it
feels
like
it
needs
some.
Some
work
in
terms
of
that,
like
flushing
out
that
specific
implementation
against
this
RC
like
if
you
want
to
give
that
feedback
Issac
in
a
comment.
I
can
properly
label
this
and-
and
we
can
essentially
backlog
this
discussion
until
we're
ready
to
essentially
take
on
the
work.
C
D
If
a
this,
like
should
really
be
broke
up
into
two
bits,
I
think
there's
like
the
registry
bit
but
I
mean
this
is
on
being
a
little
bit
semantics
and
like
and
work,
but
like
the
registry
should
be
updated
to
do
just
what
Isaac
suggested
that
if
the
package
name
has
been
completely
removed,
then
you
get
a
404
and
if
the
particular
turbo
doesn't
exist
and
you
get
a
410
and
then
there
should
be
a
like
an
RFC
for
like
the
CLI
to
handle
those
types
of
equipment.
Responses
from
the
registry,
I
think
yep
right.
A
Awesome
so
PR
number
23,
add
singleton
packages,
RFC
I
know,
there's
been
a
lot
of
discussion
about
this
for
quite
a
while
Isaac.
Do
you
want
to
maybe
speak
to
this
initially
also
I
know.
Jordan
may
have
some
insight
into
also,
like
you
know,
the
work
that's
been
happening
in
modules,
team,
arounds
and
discussions
around
singleton
packages.
D
Anybody
else,
too,
looked
at
this.
What
I'm,
I'm
confused
I
haven't
read?
They
basically
want
to
know
like
if
oh
I
did
read
you.
This
I
didn't
citizens
that,
like,
if
your
module
is
in
a
singleton,
you
should
put
a
flag
in
your
packages
and
to
denote
that
it's
a
singleton,
because
we
should
be
handling
this
appropriately
or
installing
it
once
or
something
along
these
lines
to
help
I.
D
Think
the
a
good
example
of
this
is
the
jerky
see
module
is
a
singleton
and
usually
Kajal
ERP
see,
there's
like
a
node,
G
RPC
module
that
it's
a
singleton
encountered.
That
said,
my
last
work
where
like,
if
don't
pin
versions
you
end
up
with
like
a
few
versions
of
it
and
depending
where
you
require
its
a
being
a
little
bit
strange,
but
I
haven't
knowledge
to
say
that
I
haven't
read
through
this
completely
I
think
to
have
a
well-formed
ID
on
them,
I
mean.
C
D
B
What
like
react
and
the
insulin
and
babble
and
the
the
core
module
of
each
ecosystem
cluster
is
already
doing
it's
just
there's
no
perfect
way
to
do
it
so,
like
I,
don't
know
about
the
specific
RFC,
but
the
concept
that
like
react-
or
you
know
yes
limbs
for
whatever
should
be
able
to
say
like
if
you're
depending
on
me,
make
me
appear
dependency
unless
you're
the
top
level
laugh
because
otherwise
they
have
that
dependency
hell
anyway,
just
without
any
programmatic
support.
Well,.
C
C
One
of
the
one
of
the
dependencies
of
tap
uses
uses
ink
to
generates
pretty
reports
on
the
CLI
Inc
uses,
react
version
16
in
order
to
generate
all
of
its
various
stuff
and
requires
that
all
components
be
reactive,
sixteen
components,
but
there's
no
fundamental
reason
why
you
can't
use
tap
to
test
a
react
15
site,
because
it's
not
actually
loading
ink
in
the
same
process
as
tests
are
running
right.
It's
kind
of
outside
of
the
process
boundary,
and
so
it's
a
completely
separate
node
process.
C
C
But
I
could
imagine
cases
where
a
similar
thing
exists
for
Babel
right.
You
have
some
tool,
that's
using
battle,
that's
loading,
it
programmatically
and
it's
happening
outside
of
the
kind
of
the
same
process
wall.
It's
the
main
application,
that's
using
babel,
like
especially
if
you're
using
babel
and
build
scripts
like
there's,
really
no
reason
why
you
can't
have
multiple
copies
of
it.
If
each
build
script
tool
has
its
own
different
version
of
babel.
Like
that's
fine.
C
Would
somebody
tried
to
eat
the
first
time
somebody
tried
to
use
tap
in
a
react.
15
application
was
it
just
you
know,
because
it's
a
peer
dependency
and
Kim
doesn't
do
anything
with
it.
Inc
got
voiced
it
all
the
way
up
to
the
top
level.
T
report
was
underneath
no
tap
and
had
a
version
of
react,
no
tab
because
tab
specified
react
16
as
a
depth,
so
it
couldn't
be
d,
duped
and
everything
sort
of
broke
and
blew
up.
D
D
C
C
I'm,
very
you
know,
having
having
gone
down
this
road
with
peer
dependencies,
a
very
hesitant
to
add
another
thing
that
gives
module
authors
more
control
over
the
tree,
shape
of
an
application
where
their
modules
being
used
and
where
they,
they
necessarily
lack
context
to
be
able
to
make
that
decision.
Right.
Like.
C
C
E
B
A
I
could
I
couldn't
find
any
reference
like
in
either
modules
package,
maintenance,
working
groups
or
even
I.
Think
kind
of
searching
around
I.
Think
that'd
be
I,
mean
that's
I'm,
also
in
the
camp
that
that
sort
of
I
think
solves
the
problem
that
you're
concerned
about
Isaac
in
terms
of
the
consumer
being
the
one
who's
dictating
the
the
shape
of
the
tree.
So
they
are
actively
adding
that
flag
on
install
right,
prefer
Peter,
so
I
think
I
would
solve
that.
Can
somebody
maybe
take
the
action
to
give
that
feedback
and.
A
A
The
signal
tone
one
and
then
okay,
awesome,
okay,
great
and
if
you
can
as
you're
doing
that,
just
ensure
that
if
it
hasn't
already
been
done
yet
that
the
agenda
labels
removed
as
well
so,
okay,
let's
keep
moving
on
here.
We've
only
got
two
or
three
items
left,
so
PR
number,
twenty
to
add
feature
to
showed
it
sees
on
particular
of
a
particular
dependency.
A
A
C
E
C
Out
same
reason,
NP
MLS
is
slow
because
it
it
reads
the
package
tree
and
that
does
a
bunch
of
analysis
on
it
and
it's
just
it
takes
a
long
time.
It's
a
lot
of
stuff
to
read.
Sorry
I've
got
a
what
we
should
probably
do
as
or
what
I
what
I
intend
to
do
is
write
an
RFC
to
describe
how
NP
MLS
should
change
and
NPM
seven
and
I
think
that's
this.
C
This
will
either
be
easy
to
do
or
unnecessary,
but
if
you
look
at
how
like
your
yarn
list
works
and
some
other,
you
know
how
P
and
PM
list
works,
it's
it's
essentially
just
looking
at
the
lock
file
and
then
doing
the
the
listing
operation
based
on
that.
So
there's
no
real
need
to
fully
realize
the
are
fully
loaded
the
package
tree.
Every
time
we
run
at
P
MLS.
It's
just
sort
of
that
way
for,
like
two
reasons:
I.
D
C
C
Yeah,
it
will
tell
you
if
there's
an
unmet
period,
F
there's,
no
reason
why
we
need
to
be
reading
the
node
modules
folder
in
order
to
get
that
in
photo
right
like
in
the
in
the
package.
Lock,
we're
gonna
have
a
list
of
what
everybody's
dependencies
are
and
then
what
versions
are
currently
on
disk
and
or
at
least
what
versions
were
currently
installed,
and
we
can
do
whatever
analysis
of
that
tree
that
we
want
it's.
D
B
Anytime,
anyone
uses
NPM
link
or
anyone
just
edits,
node
modules
after
install
or
anyone
does
NPM
install
and
then
separately
does
like
NPM
install,
no
save
some
stuff.
I
just
need
for
this
see
I
run
run,
they
won't
get.
It
would
I
agree
with
Isaac
that,
like
in
the
sense
that
the
default
case
probably
can
only
look
at
the
log
file,
but
I
very
much
would
want
the
ability
to
opt
into
checking
node
modules
on
disk
as
well
like
try
harder.
C
Yeah
the
Thoreau
case
yeah
another
possibility-
and
this
is
this-
is
a
trick
that
trick
that
PNP
abuses,
if
you
it
doesn't
handle
manually
editing
but
PMPM,
has
a
lock
file
that
is
in
the
in
the
root
of
your
project.
It
has
a
separate,
lock
file,
which
is
in
the
root
of
your
node
modules,
folder
yeah.
C
One
that's
in
your
root
of
your
node
modules,
folder,
even
if
you
do
know
save
like
it,
always
keeps
that
one
up-to-date
right.
That's
basically
just
like
it's
it's
notes
of
like
what
did
I
last
put
where
and
so,
even
if
you
install
something
with
no
save
or
you
install
something
that
happens
to
be
an
invalid
DEP
for
somebody
else
like
it
will
it'll
still
update
that
kind
of
hidden,
lock
file.
D
C
C
D
That
idea,
because
then
it
makes
that
the
efficiency
of
dipping
between
the
two
things
like
well.
What's
in
my
node
modules,
folder.
Well,
it's
whatever's
in
that
like
hidden,
not
hidden
but
like
separate,
lock
file
and
then
what's
what
am
I
actually
expecting?
Well,
that's
at
the
top
level,
bakla
and
you're
gonna
do
all
that
logic,
anyways
at
that
top
level,
lock
file.
So
even
if
you
say,
there's
no
save
like
you
just
don't
do
that
logic,
the
top
level
lock
file,
but
you
it's
already
that
elantra
starting
baked
in
right.
D
A
C
A
C
Yeah
I,
just
I,
just
wanted
to
basically
get
as
far
as
building
up
the
ideal
tree
to
make
sure
that
I'm
tracking
the
information
that
we
actually
need.
But
this
will
also
make
it
a
lot
easier
to
actually
run
NPM
LS
much
faster,
because
we'll
have
all
that
information
enough
information
to
build
up
the
tree
very
quickly.
Right.
A
A
Sounds
like
lockfile
v2
sounds
good,
and
then
you
can
essentially
I
mean
this
is
just
backlogs
like
we
want
to
unlock
this
capability,
but
it's
backlogged,
so
just
adding
those
those
labels
would
be
good
if
anybody's
on
there
or
I
can
do
it
after
this
call
and
then
taking
off
the
agenda
as
well.
For
this
issue,
number
22,
okay,
moving
on
PowerShell
scripts
for
installed,
binaries
I
think
we
had
some
internal
discussion
about
this.
A
A
C
It's
a
good
question,
I
think
it's
a
bug
in
PowerShell
the
bucket
windows.
Don't
write
yourself
about
it.
No
I
mean
it's
it's
it's
hard
to
say
whether
that's
even
really
a
bug
right
or
if
it's
error.
If
it's
just
look,
these
things
require
additional
permissions.
If
you
don't
want
them,
take
take
PS
one
out
of
your
default
script,
executing
set
or
whatever,
but
so.
C
I
really
don't
know,
I,
don't
know
it.
I
I
hesitate
to
comment
because
I'm,
not
a
Windows
user
and
I,
feel
like
I'm,
taking
I'm
taking
other
people's
advice
and
points
of
view,
because
they
are
Windows
users
and
then
other
Windows
users
coming
yelling
at
me.
Like
why'd.
You
do
this
thing,
so
I'm
not
sure
what
the
correct
answer
is.
I
mean
I'd
be
happy
to
remove
it
like
it's
still
skin
off
my
back
yeah.
E
A
E
E
E
C
A
C
E
A
A
There's
two
two
issues
left
and
I
want
to
get
to
make
sure
that
we
have
time
for
Jordan's
asked
to
give
up
some
discussion
on
I
guess
V
to
work
of
NPM
fund.
Do
you
actually
want
to
jump
to
that?
Instead
of
going
over
these
two
there's
PR
number
18,
Interactive,
Auto,
resolver
and
issue
number
65,
which
is
essentially
a
request
for
an
RFC,
so
I
think
that's
pretty
straightforward.
I.
B
Yeah
yeah
I,
just
I,
mean
I'll,
do
I'll,
take
the
time
to
write
up
an
RFC
of
course,
and
I'll
probably
write
the
implementation
because
it
should
be
trivial,
but
basically
I
want
to
make
funding.
Take
a
thing
or
an
array
of
things
wear
a
thing
what
it
currently
takes
and
that's
because
packages
have
multiple
maintained,
errs
and
single
maintainer
is
like
myself,
have
multiple
funding
sources
and
I
don't
want
to
have
to
make
a
landing
page
for
those
that
one
sucks
yeah.
A
B
A
B
A
Think
Isaac
I
saw
mhmmm
comments
and
I
mean
you
can
expand
on
it.
Isaac
yourself
had
some
sense
like
we
want
to
keep
the
scope
very
limited
initially,
and
it
also
makes
it
harder
for
us
to
do
anything
with
that
information
then,
like
the
website,
so
like
I
know
that
actually
right
now,
this
shape
I
just
made
a
PR
for
bubbling
up
the
NPM
fund
information
on
the
package
pages
on
our
website
on
fmjs,
calm,
and
if
the
shape
is
now
can
be
array,
it
makes
that
a
little
bit
harder
for
us
to
direct
people.
A
B
I
mean
I
would
assume
that's
the
same
as
like
the
sponsor
button
on
github,
which
is
that
they,
you
know,
and
they
of
course,
privilege
to
get
out
sponsor
thing
but
effectively
the
rest
of
them.
It's
just
a
list,
and
so
I
assume
that,
as
long
as
the
design
accommodates
and
that
simple
items-
which
of
course
is
the
part
I
can't
help
with
then
the
implementation
should
be
relatively
straightforward.
B
On
top
of
that,
so
I
can't
speak
to
the
design
difficulty
but
and
obviously
there's
a
little
bit
of
design
difficulty
that,
if
I'm
doing
the
CLI
implementation
I
would
have
to
resolve
about
how
the
printout
works
with
multiple
items,
but
assuming
that
that
could
be
resolved
and
it
feels
viable
to
me.
But
that's
why
I
was
trying
to
get
you
know.
You
focus
means
before
I
took
the
time
to
write
stuff,
I.
A
Mean
I'm
I'm
all
for
that
I
looked
at
that
opportunity
also
to
highlight
potentially
I
look
at
the
funding,
as
in
when
I
wrote
the
initial
RC
to
be
utilized
in
both
the
sense
of
future
funding
opportunities,
as
well
as
the
existing.
Let's
say,
backers
are
patrons
like
a
field
in
which
you
could
also
be
identifying
those
people
that
potentially
help
fund
the
development
work.
Of
that.
Let's
say
you
know
a
package
version
and
so
yeah
providing
allowing
for
multiple
entries
there
I
think
is
a
natural
next
step
and
like
how
it
looks
we
can.
A
A
C
D
C
Right
so
yeah,
so
they
have
a.
What
they
have
to
do
is
basically
go
go
through
the
entire
package
tree
that
you
upload
and
make
a
index
or
an
inventory
of
every
package
by
a
given
name
that
exists
within
that
tree
and
then
link
that
to
the
the
advisories
that
we
have
in
our
on
our
side
on
the
server
side
and
do
this
ember
calculation
to
see
if
it's
a
version
that
gets
that's
resolved
against.
C
C
That
causes
problems
when
you
have
bundled
dependencies
or
you're
depending
on
a
get
repo,
because
we
will
redact
that
out,
there
was
an
RFC
or
I
think
just
a
pull
request
of
the
clay
to
specify
out,
like
you
know,
audit
redact
or
something
I
think
first,
that
they
did
it
as
a
environment.
Variable
and
I
was
like,
oh
god,
no,
like
this
needs
to
be
a
proper
command
or
not
at
all
a
proper
config
or
not
at
all,
but
I
think
there's
either
a
request
for
RFC
or
an
RFC
to
not
redact
with
with.
A
C
B
Mainly
it's
that
the
like
and
I
was
talking
to
some
folks
to
get
up
today.
Well,
like
the
majority
of
see,
these
are
false
positives,
like
literally
every
catastrophic
backtracking,
regular
expression.
Cbe
is
irrelevant
for
any
linter
plugin,
for
example,
and
like
I,
just
don't
want
to
waste
even
or
that
yeah
and
I
mean
I
can
do
it.
I
can
handle
dead
abs
by
using
productions,
but
like
I,
actually
would
like
audit
mornings
about
my
dev
deaths.
B
It's
that
many
kinds
of
warnings
don't
apply
when
they're
used
within
dev
deaths
and
like
I
would
love
a
way
to
say
that,
like
this
warning
within
anywhere
within
the
tree
of
this
package,
ignore
it
it'd
be
even
better
if
I
could
define
those
things
as
a
maintainer.
So
I
can
publish
a
package
and
say
this
warning
when
it's
used
by
me
it
should
be
suppressed
and
then
nobody
else
has
to
see
it
ever
that.
C
You
you
want
to
you
want
a
way
to
like
affirmative
lis,
attests
that,
like
I
I
as
the
app
maintainer
or
or
as
a
library
author
have
I
have
looked
at
this
advisory
with
this
particular
number.
I
know
that
I
am
NOT
vulnerable
to
it,
because
I'm
not
using
the
thing
in
that
way.
So
when
the
the
thing
to
consider
there,
when
it
comes
to
library
authors,
is
you
probably
want
a
way
to
like?
C
B
It's
right,
and
then
they
can
to
American
city
could
have
a
way
to
indicate
like
I.
Accept
this
maintainer
is
override,
never
bother
me
with
it
again,
for
example,
but
yeah
I
me
that
you'd
also
don't
want
to
hand
out
trust
authors,
especially
you
know.
In
the
case
of
someone
like
I
jack
me,
a
package
or
something
we're.
A
Just
be
mindful
time,
I
think
we're
pretty
much
at
at
time
here
and
I
want
to
say
a
big
thanks
for
for
everybody
jumping
on
today.
The
me
notes
will
be
posted
quickly
after
this,
and
the
work
thing
of
the
lodge
will
be
on
YouTube.
So
again,
the
next
time
we
will
begin
will
be
getting
together,
will
be
in
two
weeks
same
time
same
place
and
thank
you
everywhere
for
shopping
on
today.