►
From YouTube: Node.js User Feedback Initiative
Description
A
All
right
welcome
everybody,
so
we're
back
after
a
bunch
of
exciting
user
feedback.
In
you
know
public
sessions
we
had
the
public
feedback
session
with
folks
giving
feedback
around
benchmarking
working
group.
You
know
that
was
really
successful
and
compelling,
and
we
had
the
in-person
session
with
the
team
at
Walmart
labs
that
Joel
Chen
brought
the
his
team
with
them
and
we
were
able
to
share
with
you
know
the
community
session
index
comp.
A
You
know
a
bit
more
perspective
about
how
Walmart
is
using
node
and
that
was
really
fun,
but
today
we're
gonna
dive
in
and
and
follow
up
with,
Greg
Wallace
of
the
node.js
foundation
and
see
how
we
can
land
the
benchmarking
working
group
survey.
So
before
I
get
into
the
meat
of
what
we're
discussing
today.
Let
me
make
sure
that
I,
let
everyone
sort
of
introduce
themselves
so
I'm
Dan,
saw
the
champion
of
the
user
feedback
initiative,
community
committee
member
and
know
Jess.
Oh
gee
attorney.
B
A
D
A
In
github,
we
want
to
make
the
data
that
we
collect
for
user
feedback
and
and
for
other
working
groups
inside
the
node.js
project
available
to
everyone
to
do
some
analysis
to
do
you
know
to
be
able
to
evaluate
trends.
You
know,
we
believe
that
you
know
not
not
just
the
sort
of
information
from
the
the
project.
These
surveys
is
useful,
but,
like
really
you
know
making
that
data
accessible
to
everybody
for
decision-making
is
an
important
part
of
what
we
do.
So
you
know
that
you
know
add
some
interesting
implications.
A
You
know
we're
an
all-volunteer
group
and
you
know
we
don't
have.
You
know
a
lot
of
business
structures
in
place
to
sort
of
protect
that
data.
So
you
know
Greg
I
know
you.
You
brought
some
some
data
with
you
today,
uh-hmm,
let's,
let's
dive
into,
we
know
what
are
the
outputs
of
the
survey?
What
do
we
have
to
sort
of
deal
with
in
the
data
that
we
get
from
the
completed
survey?
Yeah.
E
Yep
totally
so
there
basically
are
three
three
files,
so
we
we
used
Survey
Monkey
to
to
to
run
this
survey
and
Survey
Monkey
is
you
know
fine,
it's
like
it's
good
right,
it's
somewhere
in
between,
like
Google
Forms,
on
a
sort
of
low
self-service
end
and
like
some
really
sophisticated
survey
software
that
you
know
real
pros
use
if
they're
doing
massive
surveys,
Survey
monkeys
like
it,
has
had
a
lot
of
features.
E
So
basically,
you
know
just
to
kind
of
I,
don't
want
to
Wow
I,
don't
want
to
rehash
everything,
but
just
to
remind
everyone
right,
one
of
the
things
that
we
did
was
we
worked
as
a
group
to
ensure
that
the
way
we
were
asking
the
questions
would
make
the
analysis
of
the
results
as
C
as
possible.
So
you
know,
whereas
you
know
initially,
there
were
a
lot
of
very
open-ended
questions
which
would
have
required
us
to
read.
You
know
a
sentence
or
a
paragraph.
You
know.
E
We
closed
those
up
and
so
for
all
of
the
questions
like
that
that
we
closed
so
there
was
just
a
list
of
options
that
respondents
can
pick
from.
We
cranked
out
a
PDF
report
out
of
Survey
Monkey
that
basically
has
just
those
answers
represented
as
bar
charts
right,
so
really
easy
to
kind
of
quickly
look
at
and
get
a
get
a
feel
for.
E
There
were
maybe
like
four
or
five
questions
that
were
in
the
you
know,
opinions
of
you
all
right,
better
left
as
open-ended,
and
so
you
know,
Survey
Monkey
doesn't
know
what
to
do
with
that
data,
but
they
do
have
like
these
little
built-in
analyzers
so
for
those
ones,
I
created
a
quick
PowerPoint
because
they
don't
include
that
in
their
standard
PDF.
So
a
quick
PowerPoint
that
just
has
like
word
clouds
for
those
those
questions
with
like
the
most
frequent.
E
E
E
You
guys
Oh
yep,
yep,
so
yeah.
This
is
what
you
get
out
of
them
and
I'll
go
ahead
and
make
those
little
margins
smaller.
So
this
is
so.
This
is
the
you
know
the
thing
that
that
Survey
Monkey
cranked
out,
okay
and
so
for
you
know,
almost
all
the
questions.
You've
got
just
a
nice
simple
way
of
viewing
the
data
that
summarizes
it
in
a
hopeful
manner.
But
yeah
question
like
this,
where
we
had
an
open-ended
and
so
it's
blank
and
so
for
all
of
those
I
did
and
I'm.
A
A
B
E
E
So
right
so
for
for
anytime,
there
was
a
question
that
that
Survey
Monkey
was
really
not
present,
this
in
PDF
these
little
supplementals
right
dude.
So
right.
So
what
are
the
top
five
node
modules
that
you
use
most
often
one
tooth.
These
are
the
little
word
clouds
for
for
those
five
right,
so
you
can
risk
you.
Can
you
can
pick
out
trends
right,
ready,
alright,
so
that
was
question.
Five
question.
Seven
was
also
one
that
was
kind
of
open-ended.
E
F
A
D
E
E
You
know
from
a
summary
perspective
right,
if,
if
you
know
the
beat
the
audience,
whether
that's
or
somebody
you
work
with,
is
really
you
know
not
somebody
who's
going
to
be
doing
their
own
analysis
with
the
data,
but
is
interested
in
what
the
results
are.
This
combination
of
the
the
PDF
and
the
PowerPoint
that
should
be
right.
E
It
will
provide
that
that
level
of
visibility
into
the
data,
but
there
are
obviously
people
who
you
know
are
very
comfortable-
would
would
like
to
be
able
to
do
custom
analysis,
so
you
know
I
want
to.
For
example,
you
know
you'll
want
to
look
at
a
segment
of
respondents
and
then
see
how
they
answered
a
specific
set
of
questions.
So
I
really
am
very
interested
in
knowing
how
companies
that
are.
You
know
larger.
You
know
more
active
using
more
applications
right
how
they
responded
to
different
questions.
E
So
you
know
that
kind
of
thing
lends
itself
to
doing
like
crosstabs
and
stuff
like
that,
and
so
for
that
you
need
to
have,
and
so
we've
provided
you
with
the
Excel
spreadsheet
or
the
Excel
data
file,
and
we've
ensured
that
there's
no
PII
in
there
I've
asked
a
couple
of
my
colleagues
who
are
up
to
their
eyeballs
in
like
gdpr
and
all
this
sort
of
stuff
to
just
give
it
a
look
and
there's
nothing
in
there
that
can
be
traced
to
any
individual
respondent.
E
So
I
think
we
can
be
very
comfortable
putting
this
data
out
in
the
public
domain.
But
knowing
that
you
know
there
there's
no
way
for
anybody
to
be
identified
right
by
the
by
the
answers
or
by
the
information.
That's
in
there.
So
the
three
three
outputs,
if
you
will
so
two
of
them
kind
of
go
together.
That's
this
PDF
and
then
the
PowerPoint.
The
PDF
has
the
majority
of
the
questions
displayed
in
the
five
and
sort
of
images
like
this.
E
But
for
when
there
was
a
question
like
this
one
here
or
you
know
this
one
where
SurveyMonkey
really
doesn't
know
how
to
visualize
it
locks.
Anyway,
we
we
provide
a
summary
view,
just
switching
over
to
point
in
here,
so
we've
done
that
for
you
here
so
again.
This
is,
if
PowerPoint
and
or
PDF,
for
very
suitable
for
folks
who
are
really
just
interested
in
understanding
what
the
trends
are
at
a
high
level
and
for
anybody
who
wants
to
go
deeper.
B
E
E
F
E
Yeah
I
mean
you
know,
I
find
it's
one
of
those
funny
things.
It's
like
there's
like
the
long
tail
right.
There's
most
people
are
going
to
be
perfectly
satisfied
with
the
PDF
and
the
PowerPoint,
but
there's
going
to
be
people
who
are
going
to
have
edge
things
that
they
really
want
to
understand
and
having
the
excel
is
going
to.
Let
them
do
whatever
sort
of
analysis
they
want
to
do,
and
you
know
maybe
there'll
be
people
who
will
take
that
in
the
insights
that
they
produce
back,
that
might
be.
A
task
is
hey.
E
F
A
So
great
what
what
I'd
like
to
propose?
You
know
thank
thank
you
for
sharing
the
data
with
Michael
and
myself,
but
you
know
as
kind
of
a
point
of
process
as
and
as
a
attempted
to
model
how
we,
you
know,
take
responsibility
and
how
we
sort
of
you
know
allow
the
the
foundation
to
take
the
the
responsibility
of
you
know.
Managing
you
know
this.
This
data
and
the
responsibilities
of
you
know
certifying
in
a
way
that
we've,
you
know,
prepared
the
data
in
a
way
that
is
ready
for
public
consumption.
F
A
D
A
B
E
A
Bet-
and
these
are
you
know,
if
file
additions,
and
you
know
not,
you
know
edit
its
to
code,
so
you
know
full
understanding
that
that
you
know
you're
working
through
that
and
and
don't
worry
like
if
you,
if
you
make
mistake,
we
can
sort
it
out,
will
back
you
up
and
you
know
help
help
make
that
happen.
Cool
thanks,
great
well,
thanks
again
to
the
to
Greg
and
the
node.js
foundation.
For
you
know
making
that
happen.
A
F
C
B
B
E
And
there's
a
there's
actually
a
tool
that
I
found
on
the
web.
That
will,
you
know,
allow
you
to
determine
the
the
confidence
percentage
that
you
can
have
based
on
the
population
size
and
sample
size.
So
the
population
size
is
like
the
whole
universe
of
people
right
or
entities
that
you're
you
know
could
could
respond
right
and
then,
and
then
your
population
or
your
sample
size.
So
you
know
yeah
I
mean
I.
Think
300
almost
300
is
a
good
sample
size,
I
do
and
I'll
I'll.
Do
that.
A
A
So
sorry,
I
died,
I
had
a
call
that
I
had
to
hop
on
to
sake.
Are
we
ready
to
transition?
I
would
like
to
if
you're
comfortable,
moving
on
to
another
topic
I'd
like
to
in
order
to
propose
something
propose
a
a
couple
things,
but
I
really
want
to
make
sure
that
that
I
propose
something
that
you
know
began
to
sort
of
form
in
a
discussion
with
with
Chris
around
you.
What
would
hit
what
is
his
major
contribution
to
the
open
source
ecosystem
he's
a
maintainer
for
mocha,
and
you
know
he
was
like
yeah.
A
D
Nice
I've
got
I've
been
talking
with
typescript
and,
and
so
you
know,
there's
been
a
I've
gotten
positive.
You
know
response
from
from
from
the
idea,
so
you
know
it
pretty
much
just
about
everybody.
I've
asked
and
said
yeah.
This
sounds
great
and
I
want
to
want
to
participate,
so
yeah
I
mean
I.
Could
I
could
try
to
find
some
more
people,
but
I
mean
I
feel,
like
you
know
these
user
feedback
sessions?
You
probably
don't
want
to
too
many
people
right,
so
I
do
still
think.
A
C
F
A
You
know,
given
that
we
need
to
you
know,
do
the
sort
of
public
call
for
participation
I'd
like
to
propose
not
the
next
meeting,
which
would
be
in
two
weeks,
but
in
two
meetings
from
now
so
March
thirtieth.
We
could
begin
to
invite
folks
to
that
session
and
we'll
allocate
that
as
a
public
session,
that.
A
A
B
F
B
F
We
have
a
lot
to
discuss
about
that.
That's
kind
of
our!
If
we're
gonna
do
a
next
survey,
that's
one
that
I
think
might
make
sense,
but
we'd
have
to
go
back
to
the
Diagnostics
team
and,
and
today,
I'm
willing
to
take
the
action
to
I
still
have
the
action
to
sort
of
collate.
What
came
out
of
the
summit,
but
one
of
the
things
out
of
that
is
all
taken
action
to
talk
to
them
about
how
we
put
together
a
survey
yeah.
B
Yeah,
absolutely
and
I
think
that's
a
very
good
group
to
get
a
survey
from
as
well
I
just
this
is
me
kind
of
going
into
a
little
tangent,
I'm
I'm
wondering
if
the
website
working
bird
that
sorry
the
website
redesign
initiatives
would
be
interested
in
something
as
well
specifically
around
how
people
use
the
web
site
the
nodejs
site
and
what
they
would
want
to
use
it.
For
that
really
aligns
with
something
one
of
the
pain
points
we
were
feeling
yesterday,
which
was
you
know.
D
Hey,
can
I
button
really
quick,
yeah
I
sorry,
I
responded
to
I,
don't
know
whenever
this
benchmarking
survey
came
out,
I
was
like
you
know,
I
was
having
trouble
answering
it
because
of
reasons.
I
did
mention
that
yeah,
you
know.
I
have
previous
to
to
IBM.
I
worked
in.
You
know.
Basically
online
surveys.
I
worked
in
online
surveys
for
eight
years
and
so
I
have
some.
D
D
B
F
Tyranny
we
might
want
to
capture
somewhere
a
list
of
external
reviewers.
Just
so
you
know
I
think
we'll.
Definitely
you
know
once
we
put
the
question
once
we
get
a
set
of
questions
from
a
particular
group,
we'll
review
them
in
this
in
these
meetings.
But
if
there's
people
who,
like
you
know,
if
there's
people
who
basically
volunteered
like
Christopher
to
volunteer
to
review
but
may
not
be
attending
all
the
meetings,
we
want
to
make
sure
we've
got
it
documented
somewhere.
So
we
don't
forget
to
send
it
to
them
right.
B
C
We
are
first,
first,
of
course,
I'll
make
a
highlight
over
here,
but
it
should
be
a
start
as
the
meeting
notes
so
like
we
have
the
folder
right
now
the
meetings
and
probably
we
we
need
to
have
like
a
word
outside
of
our
meeting.
We
have
somebody
over
here
in
food.
Somebody
else
can
help
us
or
something
like
that.
Yeah.
B
Just
like
in
the
the
community
committee
repo,
we
have
a
few
like
a
fix
three
separate
lists.
One
of
you
know
the
members,
one
of
the
individual
membership
representatives
and
then
one
of
the
emeritus,
and
so
you
could
just
do
the
same
thing
where
you
create
effectively
the
same
structure
where
you
just
have
a
heading,
and
you
just
have
two
people
and
then
we
can.
You
know
PR
those
people
and
as
we
get
more
and
more
yeah.
B
D
F
B
Should
probably
I
don't
know
it's
a
rough,
it's
a
rough
thing,
because
I
was
looking
back
recently.
You
get
some
of
the
the
minutes
from
the
calm,
calm
stuff
and
it
seems
like
like
there
were.
There
are
things
in
the
and
user
feedback
saying
that
we
need
to
address
before
we
push
it
further,
but,
like
I,
don't
know
on
that
was.
B
F
You
know:
we've
had
a
couple
sessions,
we
got
about
five
people,
but
we
could
probably
use
more
participants
yeah.
So
you
know
I,
don't
know
if
we
need
to
do
it
right
away,
but
it
was
I
do
still.
Since
it
was
on
the
agenda.
I
thought
I'd
mention,
you
know,
I
think
we
still
want
to
be
making
like
hey.
We,
you
know,
might
be
now
hey.
F
B
F
I
can
I
can
give
you
my
standard
explanation
for
that
we
can
talk
separately.
If
you
want
to
as
well
but
sort
of
my
understanding
is
you
know
the
user
didn't
feed
the
end?
User
feedback
is
like
anybody,
who's
interested
anybody
who
either
uses
known
or
even
doesn't
use
node
because
they're
having
problems
like
we're
just
interested
from
anybody
who
will
give
us
feedback
on
on
know,
Jess,
the
enterprise
advisory
group
is,
is
cast
as
a
group
of
large
enterprises
who
have
large
deployments
of
nodejs,
so
they're
kind
of
like
a
special
subset
that
says.
F
B
F
For
well,
if
we
know,
if
we've
got
some
changes,
we
need
to
make
and
we're
condone.
You
know
we
don't
think
they're
cember
major
or
breaking,
but
we
want
somebody
to
test
them
out.
That
would
be
a
group
that
we
could
go
to
and
say.
Well,
you
know,
can
you
try
these
changes
in
advance
and
because
they
have
large
deployments,
they
should
be
able
to
be
able
to
do
that
for
us,
okay,
so
in
the
end
it
might-
and
you
know,
as
we
were
talking
earlier-
we
want
to
get
a
bunch
of
different
constituents.
F
You
know,
we've
got
just
people
who
have
no
js'
and
we
kind
of
started
with
the
general
one
in
production
not
being
able
to
use
it.
Whatever
we're
we're
I
think
gonna
have
a
specialized
one
around
tooling
and
I
could
see
the
enterprise
advisory
group
just
being
a
more
specialized
one
like
that
for
for
enterprises
with
large
deployments,
gotcha.
B
F
I
know
there's
still
some
fuzziness,
but
that's
kind
of
like
if
I'm
asked
that's
how
I'd
explain
the
difference.
We
do
I
think
you're
still
right
in
terms
of
there's
still
some
stuff
to
work
through
on
the
enterprise
Advisory
Group
in
terms
of
organization,
because
there
was
some
thought
that
they
were
gonna
be
like
self-organizing
or
whatever
I
think
we
should
almost
set
a
meeting
to
sort
of
work
through
those
specifically
yeah.
B
B
F
F
B
F
I
think
then,
we've
now
kind
of
covered,
but
I.
Somebody
if
Dan
was
here
I'd
ask
him
to
take
the
action
to
set
up
a
meeting
on
the
enterprise
advisory,
so
maybe
maybe
Mihai.
You
can
put
that
into
the
minutes
that
we're
giving
dan
that
that
yeah
yeah
that
action
to
set
up
a
meeting
to
work
to
work
through
that
make
sure
that
you
know
we're
all
on
the
same
page,
because
obviously
there's
still
a
few
questions
and
then
to
define
like
how
do
we
move
it
forward.