►
From YouTube: 2020.03.12 Triage Meeting
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).
C
C
A
B
C
With
respect
to
ground,
working
working
with
self
is
doing
so
much
right
now,
I'm,
not
so
active
in
that
mood.
Nice
to
burn
me
in
me
and
vegetative
are
like
continue
in
dork
and
specifically
right
now,
I
have
a
paraffin
that
will
make.
For
example,
if
you
have
an
inline
module
food
and
then
you
have
multiple,
which
is
a
out
like
their
own
flying
module,
and
then
you
have
like
safety
once
in
st.
G
Plus.
C
A
B
C
B
One
thing
that
maybe
I'll
go
drop
this
in
the
in
the
channel,
but
you
know
in
the
project
group
RFC,
that's
currently
under
discussion.
The
procedure
around
closing
out
a
project
group
involved
kind
of
writing
up
a
summary
of
what
work
was
sort
of
a
retrospective,
so
to
speak
of.
Where
was
it?
What
work
was
being
done?
What
what
happened-
and
maybe
that
would
be
a
useful
thing
to
write
up
just
kind
of
what
were
we
trying
to
do
and
how
far
did
you
get?
A
C
B
C
B
C
B
It's
connected
to
various
things
and
the
problem
is
basically:
if
you
have
what
is
it
it's,
especially
for
async
functions
we
wind
up
with
types.
The
compiler
is
kind
of
overly
conservative
about
the
relationships
between
lifetime
parameters,
especially
on
generators
and
types
that
are
not
written
explicitly,
and
you
wind
up
with
kind
of
missing
bounds
that
let
leave
the
trade
checker
to
thinking
that
traits
are
not
implemented.
That
could
have
been
so.
You
have
like
a
generator
type
that
captures
a.
B
If
you
have
a
type
like
this
that
gets
captured
by
a
generator
or
a
cloak
or
closure,
then
we
might
not
know
that
Tippie
out
lose
today
or
we
don't.
You
don't
have
the
where
clause
considered
in
scope,
and
so
then,
when
we
try
to
check
if
it's
implements
end,
it
would
fail.
The
well
form
is
check
code,
something.
B
B
C
B
I,
remember
correctly
and
in
general,
that's
so
another
transition.
That's
been
happening
that
another
pr2
is
changing.
How
are
some
of
our
lifetime
checking
works?
Removing
the
so-called
leak
check
cancel
on,
and
it
also
kind
of
is
somewhat
entangled
with
this.
So
I
will
I
think
from
that
perspective,
it'd
be
good
to
talk
about
the
bigger
question
a
little
bit.
Just
that's
something
that's
on
my
list
for
soon,
because
I
want
to
get
the
leak
check,
work
done
and
that
will
help
with
a
lot
of
things
go.
D
C
B
We
kind
of
expand
them
before
we're
doing
any
well
for
miss
checking
or
other
checking,
which
means,
if
you
have
something
like
this,
you
don't
get
an
error
here,
which
you
might
expect,
and
in
fact
we
even
will
warn
you
I
think
if
you
do
add
the
bound
that
you
might
have
a
sense
it
that
you're
supposed
to
have.
We
linked
to
you
and
say:
don't
do
that
because
we
don't
enforce
those
yep,
which
is.
B
B
B
B
B
Like
if
we
wanted
to
keep
the
current
behavior
I
think
we
could
quote-unquote
rationalize
it
or
something
like
the
current
behavior.
In
other
words,
if
we
did,
we
didn't
want
this
to
be
an
error.
There
is
kind
of
a
way
where
you
you're,
essentially
adding
a
sort
of
where
clause
you
know,
support
where
clauses
on
types
I
don't
or
do
we
I,
don't
know,
probably
okay,
you're,
essentially
adding
a
where
clause
like
this.
B
That's
what
we're
doing
sort
of
de
facto
and
we
could
possibly
just
sort
of
make
that
the
rule
that
type
declarations
have
a
kind
of
work
cause
that
their
right
hand
side
is
well-formed.
It
would
be
backwards
compatible
ones,
but
we
can't
really
do
that
today.
I'm,
not
sure
exactly
what
that
would
change.
Well.
B
C
C
B
C
C
F
F
B
C
B
D
B
C
B
C
C
C
B
Matter,
I,
guess,
if
it
let's
say
this,
lets
upon
size
of
the
only
one
scenario
you
would
ever
want
this.
Just
the
arm
example-
and
it
was
only
allowed
on
arm
like
it
still
seems
useful
cases
and
it's
if
it's
easy
enough
to
do
which
I'm
assuming
is
kind
of
like
you
propagate
along
and
maybe
it's
even
a
hint
I,
don't
know.
C
B
B
C
F
F
A
B
Feeling
like
I
guess,
my
real
question
is
okay.
I
might
be
good,
especially
if
it's
coming
from
the
embedded
I,
don't
know
that
feels
I
would
like
if
there's
any
purpose
to
having
commune
working
groups.
Part
of
it
is
to
help
us
prioritize
things
that
we
wouldn't
be
aware
of.
You
know
they
need.
This
government
fits
that
category.
For
me,
I
was
not
be
aware
of
it,
so
I
would
like
to
listen
when
they
ask
for
requests.
B
One
the
reason
I
brought
up
the
hint
question
was
partly
thinking
about
other
backends
and
it's
like
probably
really
easy
for
all
of
them
for
us
to
do
this,
maybe
supposed
to
do
this
LEM,
but
maybe
it's
harder
on
crane,
lift
I,
don't
think
that's
actually
personally,
I
would
be
okay
with
well.
The
crane
lift
back-end
errors
out.
If
you
use
this,
you
know.
C
A
A
B
B
C
B
C
C
So
yeah,
so
that's
a
good
point.
If
you,
if
you
disable
the
follow
unused
unsafe
links,
then
you
are
missing
out
on
useful
things.
That's
probably
doing
too
much
cops.
We
should
split
the
net
linden
or
like
make
the
lender
group
one
which
includes
two:
it's
one
for
the
functioning
and
one
flow
for
the
other
things,
and
then
you
can
disable
their
functioning
things
specifically.
C
C
B
B
I
mean
assuming
they
enclose
an
unsafe
operation
also,
and
then
there
would
be
a
second
lint
that
is
like
unsafe
block
in
function,
unsafe
block
in
an
unsafe
function,
and
that
would
be
allowed
by
default.
Anyway,
you
could
turn
it
to
to
warn
I.
D
B
I
can't
agree
that
I
wanted
to
sort
of
delegate
out
this
detailed
look,
although
I
do
think,
there's
yeah.
Thank
you.
I,
don't
know
is
that
I
think
we
are
sort
of
trying
to
change
the
intent.
So
if
you
had
a
warrant
unused
unsafe
that
you
had
already
there
you
like,
we,
you
might
expect
that
you
now
start
to
get
warnings
you
didn't
get
before,
because
that
was
the
point.
We're
nudging
people
releasing
the
new
additions.
B
B
C
B
E
D
C
I
would
personally
like
to
see
underscore
as
an
expression,
meaning
as
a
violent
expression,
whether
or
not
that
should
work
in
statics
or
Const
I
I
would
personally
do
that.
I,
don't
know
how
people
are
thinking
about
the
ones
like
my
friends
and
stuff,
but
I
I
wouldn't
want
to
have
not
a
specific
thing
for
race,
I.
C
C
B
C
C
C
B
C
Well,
I,
don't
think
you
can
work
for
for
aesthetics
either
if
you
can
try
to
do
it.
The
way
to
do
it
for
for,
let's,
which
is
to
underscore
as
an
expression,
expression,
form
and
then
like
how
type
infants
work
I,
don't
think
that
actually
works.
We
have
to
make
it
work
by
the
human
tation
work
be
done.
If
you
try
to
implement
the
way
you
do
for
left
bindings,
that
I,
don't
think,
will
work
I'm.
B
B
B
C
B
B
Reference
other
static
as
long
as
our
types
are
fully
known,
so
you
can
cut
the
cycle
by
getting
a
full
annotation,
yeah
level,
birth
and
I.
Think
I
would
do
the
same
with
let's
the
only
difference
is
in
the
case
of
let's
they
create
the
variables
in
the
ambient
type-checking
context.
I,
don't
quite
know
right
when
you
said
I
doesn't
work
today,
I,
don't
really
know
what
that
means.
I
understand
how
underscore
even
appears
there
today
it
seems
like
a
bust,
so.
C
The
way
I
implemented
this
in
my
experiment
was
to
create
and
an
inference
where
I
go
for
the
value
1,
and
that
value
has
a
difference
valuable
for
the
prototype,
but
because
anonymous
constants,
which
the
other
score
would
be,
are
in
a
difference
influenced
context
than
than
the
the
ambient
thing
for
the
function
that
doesn't
work
out.
The
station.
C
12
minutes
to
go
calling
methods
on
generic
trousers
of
course
depends
well.
We
use
that
sort
of
implemented.
Part,
that's
a
little
tighter,
so
we're
explained
using
with
it.
So
the
question
is
what
we
should
do
with
the
RFC.
No
I
think
we
should
create
a
project
group.
It
seems
pretty
clear.
This
is
like
exactly
what
they
exist,
for
it.
Isn't
that
don't
we
already
have
a
project
which
is
called
like
I
mean.