►
Description
A
A
B
David
that,
because
there
was
a
disagreement
about
whether
they
should
match
in
this
pie
chart
actually
there
is
some
progress
going
on.
I
mean
the
movie
repo
about
that
and
I
think
the
edge
the
situation
norge
most
likely
change
over
a
week.
So
perhaps
we
can
just
postpone
that
for
a
week.
Okay,
so
basically.
B
D
B
A
D
D
Don't
think
we
should
override
concerns
by
voting
on
them
personally,
I,
don't
like
I,
wouldn't
like
to
see
that
to
become
a
common
practice.
Talk
to
Ruben
over
at
the
collab
summit,
though-
and
he
pointed
out
that
I
haven't
been
in
the
tasty
very
long
and
I'm-
not
paying
a
lot
of
attention
to
these
things.
D
But
voting
is
not
something
we
commonly
do
it's
a
pretty
uncommon
situation,
so
it's
not
gonna
become
our
regular,
a
regular
pattern,
a
couple
people,
you
know:
don't
respond
to
a
PR
and
suddenly
there's
a
vote
and
it
gets
slammed
in
anyhow.
Then
I
don't
have
any
problems
with
what's
going
on.
Okay,.
E
So
the
alternative,
a
couple
of
options
here,
there's
basically
three
ways:
two
approaches
I
think
one
is
to
just
you
know:
let
it
play
itself
out.
It's
probably
at
an
impasse.
Pr
PR
eventually
gets
closed.
Another
is
Ruben
could
be
inclined
to
invoke
the.
Please
explain
your
your
objection
and
then
the
if
the,
if
you
know
Brian
but
Brett
well,
that
already
happened
and
Brian
responded.
So
you
can't
really
do
that.
I
guess
with
you
know.
E
It's
to
like
try
to
facilitate
discussion
and
I
guess:
do
you
have
a
feeling
on
what
the
most
appropriate
course
of
action
at
this
point
is
I
can
recap
those
quickly
again
if
you'd
like,
but
basically
like,
leave
it
alone
or
get
involved
to
try
to
like
work
something
out
between
the
parties
giving
us?
And
personally
let's
say
you
was
in
the
TSC.
F
I
have
a
question
about
that
PR.
My
understanding
is
that
we
we
that
the
PR
seeks
not
to
remove
tracking
for
for
for
any
sort
of
performance
characteristics
right.
It
seeks
to
reduce
duplication
in
the
things
we
benchmark.
So
if
the
PR
accomplishes
that,
then
we
do
not
lose
track
of
performance
characteristics
in
any
way
right,
and
yet
we
shortened
the
the
runtime
of
that
particular
benchmark
or
sort
of
Bank
right.
D
D
To
explain
okay,
this
is
why
this
this
is
why
this
this
way
this
and
Brian
has
not
yet
come
back
to
describes
what
specific
change
in
the
fairly
large
PR.
He
has
difficulty
I,
don't
I
I
assume
that
he
doesn't
object
to
everything,
but
I,
don't
know
exactly
what
he
does
object
to.
So
if
it
comes
to
a
vote
right
now,
it'll
probably
just
get
merged.
D
G
A
I
mean
I,
see,
I,
see
a
lot
of
abstentions
and
I.
Think
part
of
it
is
that
it's
like
looking
at
this
one
on
its
own
is
a
bit
difficult.
It's
more
that
there's
a
philosophical
choice
that
we
need
to
make
here,
which
is
around
like
do
we
actually
say
that
the
default
has
to
be
less
than
a
particular
limit
to
make
it
practical
to
run.
A
G
E
Just
you
know,
rather
than
this
one
big
PR
to
to
crunch
down
the
benchmark
run.
Time
is
you
know
individual
PRS
for
each
file
and
you
know
be
easier
to
review
and
easier
to
determine.
You
know
how
likely
is
it
that
this
change?
Will
you
know
Sarah,
but
anyway,
this
sounds
like
we
have
options
and
it
sounds
like
there's
nothing,
that's
yes.
He
needs
to
do
in
this
particular
moment
other
than
leave
it
on
the
agenda
for
next
week.
In
case
it
completely
stalls
out
or
an
impasse
is
arrived
at
right.
E
F
E
A
G
A
So
I
like
in
that
context,
it's
like
if
it
if
it's
if
we
get
into
the
state,
we're
like
you
know,
there's
there's
an
objection,
but
the
person
who's
objecting
isn't
actually
interacting.
Then
I
agree
like
we
sort
of
fall
into
that
that
category
as
long
as
it's
going
back
and
forth,
I'm,
not
sure
saying
next
week
as
a
deadline,
at
least
in
my
mind,
because
the
other
wives
are
saying
we
have
to
get
this
in.
You.
E
Know
yeah
the
only
the
only
the
only
the
only
I
mean
the
only
time
where
we
need
to
have
a
vote
is
if
there
is
literally
an
impasse.
We're
like
every
both
sides
feel
like
they've
made
their
case
and
neither
one
sees
the
other
ones.
You
know
you
know
nobody
agrees
and
they
both
agree
that,
like
there's,
not
really
a
lot
of
room
for
compromise,
they're,
not
gonna,
agree,
and
you
know
that's
that's
the
time
to
vote
and
and
I'm
okay.
E
If
this
conversation,
you
know
goes
back
and
forth
for
weeks
like
it,
you
know,
I
mean
just
as
long
as
it's
actually
happening
and
then
it's
time
to
either.
You
know
if
Rubin
stops
participating
it's
time
to
close
the
PR.
If
Brian
stops
participating
it's
time
to
land
the
PR,
unless
someone
else
is
objecting
or
agreeing
with
Brian,
then
they
can
step
in
and
participate.
But
you
know
it
had.
It
had
stalled
out
completely
and
conversation
seems
to
be
restarted.
F
And
and
let's,
let's
also
not
forget-
maybe
I'm-
maybe
I'm
oversimplifying
this,
but
suppose
we
land
the
PR
and
as
a
result,
we
lose
track
of
an
important
metric
and,
and
we
accidentally
introduced
a
regression
right,
that's
still
not
the
end
of
the
world.
We
can
always
fix
it
as
soon
as
we're
aware.
You
know
it's,
it's
not
it's
not
like
we're
we're
shooting
ourselves
in
the
foot
fatally.
You
know
if.
E
A
That's
why
I
was
just
sort
of
thinking
that
if
we
it's
we
may
get,
you
know
hey
this
one
lands
and
they'll
be
the
next
one
and
we
may
get
into
the
same
thing.
It's
almost
like
we.
We
need
to
figure
out
the
meta
issue
of
how
do
we
solve
the
fact
that
they're
too
long
running
and
are
we
and
can
we
get
consensus
on
what
you
do
in
that
trade-off
between
plus
coverage
and
running
in
a
reasonable
right,
I'm.
E
Kind
of
okay
winging
it
like
you,
know
we're
always
gonna,
be
adding
more
benchmarks.
There's
always
gonna
be
benchmarks,
being
added,
it's
just
the
fact
of
life
and
we
almost
never
take
them
out.
You
know,
except
when,
like
an
API,
you
know
a
you
know,
a
breaking
change
happens
that
makes
it.
It
makes
a
particular
benchmark
irrelevant,
because
that
API
doesn't
exist
anymore.
Something
like
that,
but.
E
A
E
What
I
wouldn't
want
is
I
wouldn't
want
like
we,
you
know,
we've
decided
our
benchmark
should
never
take
more
than
120
minutes,
like
you
know,
come
up
with
like
a
specific
metric,
because
you
know
that's
that's
gonna
change,
depending
on
what
machine
you're
running
on
you
know,
as
we
add
more
api's
blah
blah
blah
blah
blah.
You
know
like
it
has
to
be
something
I,
don't
know
what
it
would
look
like
is
the
thing
yeah.
A
E
A
A
A
The
challenge
has
been
is
that
they
run
way
too
long
and
generate
way
to
know
any
numbers
to
reasonably
like
show
something
that
you
can
look
at
say:
oh
there's
a
problem
or
not
yeah
or
to
run
them
every
night
like
we
do
the
other
ones
right
like
so
there
was
some
discussion
and
udom
was
gonna.
Look
at
like.
Can
we
run
them
all
once
a
month
or
something.
E
A
E
E
A
E
A
A
Moving
on
to
the
next
issue,
then,
is
reevaluate
meetings
times,
which
was
707.
That's
actually
already
been
closed
and
we've
I've
updated
all
the
time
so
that
meetings
going
forward
after
this
one
are
scheduled
in
the
calendar
and
will
be
scheduled
so
I
believe
that
it
ends
up
being
well
anyway.
That
times
are
there
it's
it's
a
tie.
It
tighter
cluster,
you
know,
I,
think
it's
a
range
of
six
hours
or
so
Thank.
A
Thank
you
very
much
update
the
Charter
that
one
basically
I
think
the
status
name
is
the
last
time
it
needs
to
be
reviewed
by
in
the
next
CPC
meeting,
which
we
should
do
hopefully
on
Monday
I.
Think
is
the
next
meeting
so
we'll
see
if
we
can
get
that
done
tracking
issue
for
updating
on
the
board
meeting.
We
don't
have
miles,
so
we
won't
do
that.
A
Don't
have
miles
and
I
guess
actually
related
to
that
which
maybe
I
should
have
tagged,
is
election
for
the
board
representatives,
the
discussion
in
the
issue
and
the
administering
yesterday
and
let
people
self
nominate
for
a
week.
I
haven't
seen
any
comments
since
then,
so
I
guess
I
should
probably
just
send
out
some
some
reminders
to
the
TSE
and
to
come,
come
to
do
that
and
then,
depending
on
who
self
nominates
I
think
we
were
deferring
who
how
we
would
vote
to
see.
You
know
what
kind
of
candidates
we
had
and
so
forth.
A
A
I
would
think
that
it
might
as
well
just
be
posting
in
the
same
issue
that
was
open.
That
works
for
everyone,
I'm
happy
for
other
suggestions,
but
I
was
thinking.
We
could
just
use
this
name
issue
I'll,
send
an
email
to
both
TC
and
Kham
Kham,
saying
you
know,
based
on
the
discussion
here.
You
know
we're
now
opening
for
a
week
for
people
to
self
nominate
in
this
issue
makes
sense
everybody
here:
okay,
so
now
flipping
over
to
the
strategic
initiatives.
A
F
Well,
we
we
are,
we
are
readying,
an
API
version.
Five
and
we've
had
some
discussion
about
what
to
include
and
I've
pinged
all
the
people
who
have
proposed
new
api's
since
four
and
they've
all
responded-
and
it
looks
like
pretty
much
everything
that's
marked
experimental
can
probably
go
in,
but
we
should
probably
talk
about
it
at
the
next
meeting.
Yeah.
G
The
implementation
is
progressing,
it's
happening
over
in
the
node.js
click
repository.
The
basic,
quick
server
and
client
is
functioning
and
working
on
filling
out.
The
functionality
now
I'm,
not
focusing
on
any
part
of
the
HP
3
implementation,
yet
is
that
spec
is
still
pretty
early
and
pretty
fluid
right
now.
G
A
E
So
I
don't
have
anything
to
report
other
than
that
I
had
some
discussions
at
the
summit
with
with
some
folks
about
what
might
be
done.
You
and
I
should
probably
chat
at
some
point.
Michael
I
know,
though,
that
I've
been
busy
and
you've
been
busy.
I
honestly
got
other
priorities.
Right
now
is
gonna
try
to
like
push
myself
forward
on
this
stuff
with
the
build
meeting
this
week,
but
it
got
cancelled.
Sorry,
so
you
know
I
don't
want
to
drop
this,
but
it
is
moving
slower.
I
have
not
been
as
as
engaged
as
I'd
hoped.
A
A
A
E
A
Yeah-
okay,
sorry
yep,
so
here
just
copying
it
over.
We
had
a
session
to
discuss
in
the
state
of
the
work
at
the
club
summit.
We
discussed
number
of
items
and
decide.
It
would
make
sense
to
go
forward
with
moving
workers
out
of
experimental
State
after
going
through
the
web
messaging
web
platform
tests
and
making
sure
that
our
implementation
aligns
with
them
and
that's
currently
in
progress
and
adding
worker
working
using
modules
example
worker
pool
at
about
250
K
downloads
per
month
decision.