►
Description
A
A
Yes,
it
sounds
like
we
have
no
announcements
this
week.
So
let's
move
on
to
the
issues
that
have
been
tagged
for
TSC
agenda.
The
first
one
is
2
to
5
8
5,
which
is
how
did
you
feature,
detection
and
I
know
rich
you've
been
spending
a
fair
amount
of
time
trying
to
round
up
the
discussion
on
that.
So
why
don't
you
take
that
away?
Yes,.
B
B
So
since
there
are
people
who
who
object
to
the
release
going
out
as
it
is,
if
I
understand
our
process
correctly,
it
cannot
go
out
unless
we
either
achieve
resolution
here
and
now,
or
we
take
a
vote
so
as
of
right
now,
it
looks
like
if
I
understood
everybody
who's
weighed
in
on
the
issue
correctly,
we
have
eight
people
who
are
proponents
of
just
shipping
the
release
either.
They
believe
that
this
is
the
right
way
to
do
that.
B
Api
and
we'll
figure
out
feature
detection
later
or
we
won't
need
it
or
they
believe
that
it's
okay,
if
we
need
to
add
you,
know,
make
your
P
function
later,
that
people
can
check
the
existence
of.
If
that's
the
way,
we're
gonna
do
future
detection,
it's
okay,
to
add
that
as
an
ally,
we
can
have
both.
So
there's
eight
people
of
the
28
members
of
the
TSE
who
advocate
for
that
as
best
as
I
can
tell.
B
There
are
two
people
who
are
opposed
to
moving
forward
with
releasing
this
feature
without
figuring
out
the
feature:
detection
issue,
the
larger
feature,
detection
issue.
First
and
then
there
are
two
people
who
I
believe
are
kind
of
abstaining
like
I,
don't
really
play
with
either
so
I.
Don't
know
if
targis
has
anything
he
wants
to
add
or
I
know.
B
D
So,
with
regards
to
person
yeah
just
to
be
clear,
I
have
no
problem:
shipping
them
recursive,
mcdr
I,
don't
want
to
ship
now
with
the
risk
of
having
to
change
it.
How
we
do
it
later,
you
know,
and
then
how
that
being
confusing,
or
have
us
end
up
having
to
support
two
different
ways
of
doing
exact
same
thing.
D
So
I
would
like
to
have
that
question
result
before
it
actually
goes
out
into
a
release.
We've
done
that
with
similar
things
in
the
past,
or
we
weren't
quite
sure
if
the
fuel
is
quite
ready
to
be
exposed.
The
way
it
is
so
I
seen
a
problem
doing
it
holding
off
on
that
now
on
the
feature,
discovery
thing
in
general
I
prefer
the
release
not
to
be
held
up
while
we
handle
that
you
know
where
we
discussed
that.
So
what
I
would
like
to
see
is
just
to
have
this
one
feature
removed
from
this
release.
C
B
E
D
F
E
C
A
D
So
you
know
having
both
maker
with
the
recursive
option
and
make
derpy
option.
It's
just
two
ways
of
doing
the
exact
same
thing
and
just
doesn't
make
sense
right.
It
really
should
be
one
or
the
other.
That's
issue
number
one
and
then
the
issue
of
how
we
discover
this
thing
is
to
me
a
secondary
right.
D
B
B
F
B
C
B
C
B
Think
I
think
you
need
to
decide
how
we're
gonna
do
feature
detection
unless
we're
gonna
unless
we're
going
to
feature
detection
in
a
one-off
way
with
make
derpy.
But
you
have
to
side
how?
What
is
the
acceptable
way
to
do
feature
detection,
because
that
affects
the
API,
because
if,
if
the,
if
the
most
desirable
way
to
do
feature,
detection
is
check
if
F
s
dot
property
exists,
then
that
determines
you
know
how
we
do
it
here.
So.
D
There's
really
nothing
different
about
this
conversation
versus
the
you
know
whether
we
can't
add
new
modules
until
we
have
the
namespace
thing
figured
out
or
other
types
of
similar
conversations.
So
I
think
at
some
point,
when
you
don't
have
a
step
back
and
come
up
with
a
more
general
policy,
so
the
stops
being
a
one-off
as
things
keep
coming
up.
D
I
am
absolutely
fine.
If
we
want
to
say
that
we
won't
do
we
won't
ship,
mcdr,
PE
or
mcdr,
with
that,
you
know
until
we
get
this
figured
out,
but
just
yeah.
It
just
needs
to
be
acknowledged
that
this
is
the
feature.
Detection
issue
is
larger
than
just
that.
One
API
and
we
have
shipped
other
similar
things,
almost
identical
types
of
situations
where
new
options
without
having
this
issue
result.
B
So
I
would
add
them
that
the
feature
detection
issue
is
one
that,
based
on
the
conversation
I've
seen
so
far
in
the
issue,
tracker
I,
do
not
anticipate
consensus.
I
think
that
there
are
definite
people
who
are
definitely
strongly
in
favor
of
checking
for
property,
and
there
are
people
who
are
strongly
in
favor
of
avoiding
that
that
that
is
an
anti-pattern
that
that
you
know
anyway.
I
won't
get
into
the
arguments
and
I.
B
I
I
Just
the
the
amount
of
users
for
each
of
those
is
substantially
different,
and
so
I
can
fire
if
our
policy
or
if
our
solution
doesn't
doesn't
take
into
account
that
there's
gonna
be
those
differences,
something
that's
you
know
used
by
small
percentage
of
the
user
base,
which
is
something
that's
used
by
like
7080
percent
of
people
and
most
modules.
You
know
we're
not
gonna
arrive
at
something
that
actually
makes
sense,
and
that
conversation
isn't
really
happening
right
now.
B
Yeah
I
mean
and
I
would
argue
that
that
that
that
may
mean
that
the
correct
resolution
is
that,
in
fact,
ad
hoc
discussion
each
time
this
comes
up
is
in
fact
the
right
solution
that
some
features
make
sense.
This
way,
some
features
make
sense,
they're
not
even
bother
feature,
detection
and
other
features.
You
know,
etcetera
I
mean
it's
painful,
but
it
might
actually
be
the
correct
approach.
G
There
is
a
recursive
mcdeere
NPM
package,
currently
right,
and
so
I
would
imagine
that
most
people
use
that
if
they
want
to
do
recursive
mcdeere
rather
than
rolling
their
own
right
and
so
yeah
chair
away,
we
go
it's
only
recursive
mcdeere
that
that
needs
to
really
change
right,
because
most
people
will
not
want
to
change.
They'll
just
be
happy
knowing
that
that
it's
doing
the
right
thing
in
the
background,
yeah.
J
B
B
And
I
guess
the
other
thing
is
I'm
gonna
be
pestering
everybody
to
like
drive
the
future
detection
issue
to
a
resolution.
Even
if
the
resolution
is
we're
gonna
do
this
ad
hoc
I
hope,
that's,
not
a
resolution,
but
you
know
it's
better
than
like
dithering
and
not
making
a
decision
for
three
months
so
right.
A
A
B
Let's
move
on
because
I
don't
think
we're
gonna
achieve
a
resolution
of
anything
else
in
this
I
think
we
need
to
take
the
larger
discussion
back
to
the
issue.
Tracker
and
and
I'll
be
emailing
people
who
haven't
weighed
in
and
count
you
know
and
hope.
If
we
miraculously
arrive
at
consensus,
awesome
but
I,
don't
think
that's
gonna
happen.
I
think
this
is
gonna,
come
down
to
a
vote.
A
D
I
think
the
other
one
is
is
closed.
Like
discussion
is
happening
on
the
multiple
resolves
now
there
are
still
some
outstanding
to-do
items
on
it
from
what
I
saw
in
the
discussion
this
morning.
So
I
think
at
this
point
we
just
need
to
say
it's
not
blocked
as
moving
forward
and
just
let
it
run
its
course
to
get.
I
Out
so
I
would
have
one
thing
to
add
and
that's
that
Gus
only
closed
the
issue
because
he
was
not
willing,
so
you
continue
rebasing
it.
So
if
somebody
does
feel
strongly
that
the
other
one
is
a
better
one,
then
that
can
be
resurrected
still.
There
was
no
official
sort
of
agreement
that
it
should
be
closed
and
also
could
we
promote
Matteo
yep.
A
A
A
Okay,
the
next
one
on
the
list
is
buffer,
a
runtime
defecate
buffer
constructor
in
sync
mode.
So
this
is
two
two
five
eight
four
and
there
is
the
related
buffer,
runtime
deprecated
buffer
constructor
everywhere
by
default
to
one
three
five,
one
I
think
Nikita.
You
probably
want
to
lead
us
on
those
I.
L
Think
that
the
duplication
for
Norfolk
instructor
everywhere
she's,
my
default,
which
is
to
1
T
y1,
is
very
unlikely
to
land
in
the
necessities.
But
hopefully
we
can
manage
in
12
and
for
the
second
one
I
filed
urges
to
mediate
solution
and
it's
a
work
in
progress,
but
currently
I
think
that
it
needs
some
tests.
L
A
B
A
F
B
L
E
L
L
A
So
if
I
sort
of
recap
to
135
1,
is
it
fair
to
say
which
is
bifurcated
everywhere
you
know.
Is
it?
Is
it
fair
to
add
this
comment
to
the
issue
which
is
discussed
on
TAC
meeting
Nikita
mentioned
that
it
won't
make
him
to
11
dot
X,
so
hoping
for
12
dot.
X
taking
off
the
agenda
should
be
reacted
when
we
believe
it's
ready
to
make
the
case
for
12
dot
X,
so
that
fair.
F
A
J
A
A
B
A
Maybe
I'll
take
it
out,
I!
Think
it's
fine,
a
Nikita
mentioned
you
won't
make
it
xi.
Dot
X
taking
off
the
agenda
should
be
reality,
wouldn't
believe
it's
ready
to
make
the
case
for
12x
perfectly
okay,
so
I'll
do
that
I'll
remove
the
TSC
agenda
label,
so
we
don't
get
it
next
week
and
then
four
to
one
three
five
one.
A
A
B
K
Asking
for
questions
so
there
is
a
big
issue,
so
you
can
ask
this.
Remember
the
TAC
members
that
are
listed
in
there.
Some
questions
I
will
be
running
there.
I
will
be
making
the
questions
we
beyond
all
run
last
by
ourselves,
more
or
less
and
during
the
last
coffee
break
of
the
of
not
interactive
of
the
first
day
have
not
attractive.
So
we
have
to
see
you
all
there.
F
A
K
B
K
That
was
that
was
about
it.
So
yes,
oh
there,
it's
another
one
anyway,
yeah
I'm
really
bad
with
this
number
with
those
issue
numbers.
So,
as
you
can
see
so
now
it's
fine,
it's
the
the
the
selection.
It's
been
done.
We
have
an
agenda,
people
are
being
notified
and
it
would
be
published
on
the
main
as
part
of
the
converse
program
as
well.
So
we
will
have
short
talks
and
there
will
be
a
nice
area
where
people
can
just
sit
there
and
hack
and
do
things
if
they
want
to.
B
D
A
Modules,
we
don't
have
miles
an
API
I've
added
there.
So
really
the
focus
is
on
async
worker
evangelism
examples
and
test
suite.
So
you
can
see
our
overview.
One
thing
on
this:
the
examples
we
are.
We
already
have
sort
of
reinvigorated
the
note,
a
tonic,
note,
add-on
examples
and
merged
all
the
API
stuff
into
there.
Cuz
this
a
heads
up
next
is
open,
SSL
evolution.
We
don't
have
rod,
workers,
I,
think
Anna.
You
did
join
so
any
update
on
workers.
Yes,.
M
I'm
here
actually
find
a
few
things,
so
one
of
the
things
that
we
did
merge
this
week
is
inspector
support
for
workers,
both
so
for
one,
the
inspector
API
that
we
have
for
inspecting
the
current
current
state
of
the
engine
we
merged
that
and
I
think
but
I'm
not
quite
sure.
Full
inspector
support
for
workers
in
general,
for,
like
chrome,
debugging,
was
merged
yesterday
or
today.
M
M
A
A
Okay,
error
messages:
I,
don't
think
we
have
Joey
today.
No
core
promise
API
is.
D
A
A
E
Yes,
so
v8
7.0
is
almost
ready.
There
are
just
two
commits
that
I
hope
we
can
have
back
ported
upstreaming
v8
with
instead
of
floating
the
patches.
After
that,
the
PR
will
be
ready
to
land
and
I.
Think
I
will
open
an
issue
too
to
ask,
or
or
maybe
just
in
the
PR
to
ask
for
lending
it
before
it
becomes
stable
in
chrome,
so
that
we
can
start
testing
it,
because
that's
the
version
that
we
will
have
in
notes
11.