►
From YouTube: 2021-03-17 Backlog Bonanza
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).
A
Oh
anyway,
okay
welcome
to
this
backlog,
bonanza
review
meeting
yeah,
so
my
plan
was
to
go
back
over
this
paper
dock
and
figure
out
what
happened
and
what
didn't
happen
and
what
needs
to
happen.
B
B
I
can
write
down
action
items
I
guess,
because
I'm
assuming
that
is
going
to
be
a
thing
that
we
need
to
do.
That
would
be
useful.
That's
a
good
point.
Thank
you.
C
A
C
Oh
quick
procedural
note:
the
top
of
the
dot
currently
says
people
to
assign
work
to
and
lists
six
people.
One
of
those
people
is
not
currently
on
the
team,
so
if
we
have
anything
currently
assigned
to
them,
we
may
wish
to
reassign
it
if
it's
not
already
done,
yep.
A
Good
point:
okay,
I'm
just
taking
that
section
out
because
we're
going
to
replace
it
with
action
items
anyway
sounds
good
rfc,
debuggable
macro
expressions,
see
where
this
landed
looks
like
we
moved
this
also.
I
think
this
was
another
one.
We
yeah
good.
We
had
some
coverage
and
we
moved
to
mcp
perfect
over
constraining
and
omitting
unsafe.
C
Looks
like
this
got
assigned
to
me.
I
would
be
happy
to
write
this
up
and
take
the
action
item
for
it.
D
Oh,
I
think
a
notification
popped
up
while
I
was
talking
great,
I'm
still
planning
on
writing
an
rfc
here,
but
it's
been,
I
kind
of
finished
my
edition
stuff
before
that.
D
D
A
method
to
be
unsafe
and
what
does
that
mean
about
preconditions?
What
is
the
difference
between
unsafe
traits
and
unsafe
functions
in
trades,
which
I
think
we.
D
A
A
Scott,
do
you
think
you
could
open
prior
to
writing
an
rfc?
You
could
add
some
like
design
notes
or
something
that
just
captures
in
bullet
form
a
few,
maybe
like
a
link
to
the.
A
We
could
even
potentially
mentor
that
out
at
that
point.
That
would
be,
I
guess,
but
but
that
would
be
great
to
have
just
some
record.
We
can
point
to
cool
and
then
maybe
link
it
from
the.
D
G
A
Correct
it's
just
defining
what
unsafe
means
at
all
points,
I'm
wondering
if
it
even
needs
to
be
an
rfc
or
like
a
reference
pr,
but
either
way.
Really.
I
I'm
not
like
against
this
idea.
I'm
kind
of
would
like
to
close
the
rfc,
but
not
with
prejudice,
so
to
speak,
sort
of
encourage
that
would
be
okay.
If
the
author
wanted
to
open
a
project,
I
agree
with
that
and
we
talk
about
it
again.
G
I
guess
one
thing
is
that
the
reason
I
ask
is
because
I
think
I
still
want
what
this
rfc
proposes.
A
We
could
accept
it
even
or
try
to.
I
don't
remember
the
details
of
what
there
are
stupid
poses,
though
I
guess
I'm
if
it,
if
it
had
a
champion
in
you,
taylor
that
might
change
my
like
overall
take.
G
A
G
I
I
I
don't
know
like
what
bandwidth
I
have
to
spend
on
it,
but
I
I'm
happy
to
like
do
my
best
to
respond
to
comments.
I
yeah,
I
don't.
I
don't
know
what
what
the
best
like
procedural
step
is
for
that.
I
guess
closing
seems
weird
when
I
actually
like
actively
want
this.
C
I
think
the
part
where
we
had
disagreement
to
the
point
of
wanting
to
defer
this
was
under
what
circumstances
the
compiler
can
guarantee
that
it
knows
you
are
calling
a
specific
implementation
for
which
the
body
doesn't
require
an
unsafe
block
and
therefore
you
don't
have
to
have
an
unsafe
block.
That
was
the.
F
C
G
A
I
don't
see
it
as
very
hard
to
implement
what
is
there
a
concern?
I.
F
A
There
are
other
I'm
trying
to
think
what
other
ways
it's
possible
for
functions.
Functions
in
general
in
an
impul
can
be
less
precise
than
the
trait
or
like
more
more
accepting
than
the
trait.
This
would
be
the
one
case
we're
allowing
like
reflecting
that
right,
like
the
lifetime.
Signature,
for
example,
can
be
different.
C
A
A
C
G
A
It
does
I'm
not.
A
With
what
that
working
group
is
doing,
but
I
recall
it
being
more
at
an
impo
level.
First
of
all,
and
this
feels
more
fine-grained,
potentially.
A
Like
it
seems
like
the
mechanism
we
want
here
is
being
able
to
refine
traits
and
say
like
this
is
a
sub
trait
that
refines
the
signature
of
that
item
in
the
super
trade.
Anyway,
I
don't
know,
I
mean
straw
poll
who
would
who
would
be
okay
with
accepting
the
rfc,
as
is,
I
guess,
is
the
question
I
I
I
would
but
or
maybe
who
would
not,
but
does.
C
Yeah
yeah,
I
think
I
would
be
okay
with
it,
given
two
constraints
that
we
stated
that
the
only
thing
we
guarantee
is
that
if
you
know
the
exact
concrete
type
you
can
do
so
and
not
if
you
know
something
about
the
type
but
not
everything
and
having
some
kind
of
future
extension,
for
we
probably
want
a
general
mechanism
like
this
for
unsafe
and
const
that
allows
I
take
anything
where
this
could
be
declared
as
unsafe
or
not
as
unsafe
or
declared
as
const.
D
D
D
A
A
I
would
I
would
be
okay
if
you
moved
to
accept
personally,
I
wouldn't
guarantee
I
wanna,
I
wanna
like
sit
on
it,
but
I
feel
happy
in
general,
with
the
premise
of
we
try
to
make
your
life
better.
When
we
know
things
that
said
we
used
to
have
we
used
to
have
logic
like
that
and
we
took
it
out,
but
I
think
that's
because
it
was
it
was
it
was
applying
in
the
type
checker
and
it
was
rather
difficult.
It
was
much
better
for
us
to
be
able
to
just
talk
about.
F
G
A
C
But
I
made
a
note
of
what
the
consensus
seems
to
be,
and
so
what's
the
action
name,
the
action
name
is
yeah.
G
I'll
chat
with
scott,
I
think
maybe
that
works
for
you,
scott
and
then
we
get
all
trying
to
follow
up
either
in
a
on
zulu
with
other
folks
or
if
it
seems
like
there's
consensus.
I'll,
leave
a
comment
on
the
rxc.
C
Discussed
this
I
know
other
things
have
come
up
that
would
like
this,
including
things
like
inline
assembly,
where
it'd
be
nice
to
pull
out
more
concept
of.
We
evaluate
this
and
pull
it
out
of
individual
cases.
A
E
I
mean
I
don't
have
the
details
of
this
particular
rfc
in
cash.
I
remember
I
was
critical
early
on,
but
I
you
know
they
incorporated
number
of
feed
pieces
of
feedback.
So
I
don't
know
what
the
current
status
is.
I'm
wondering.
E
A
I
either
want
us
to
say
we're
going
to
commit
to
going
forward
or
we're
like
or
like
not
that
we
have
to
accept
it,
but
that
we
have
the
resources
and
would
be
willing
to
accept
it
or
that
we're
going
to
close
it
kind
of
in
some
form
or
another
and
one
one
thing
that
I
think
is
a
middle
ground
which
I'm
sort
of
okay
with,
is
we're
gonna
close
this
but
encourage
an
mcp
or
encourage
a
project
proposal
so
that
we
can
sort
of
try
to
fit
it
into
a
more
structured
process.
B
B
A
E
E
I
think
it's
changed
so
much
since
then.
I'm
not
sure
it's
meaningful
to
to
take
interpret
their
remarks
at
that
time.
A
Postpone
our
reasoning
is,
I
mean
we're
generally
been
deferring
advanced
macro
machinations
right
yep,
and
I
think
this
fits
in
that
category.
I
would
consider
a
bigger
effort
around
macros
2.0
or
at
some
point.
B
Okay,
who
can
do
that.
A
G
Positive,
just
to
recap,
the
the
actual
rfc
contents
are
like
the
the
sort
of
reference
description
is
actually
just
changing.
The
grammar
that's
accepted
by
attribute
proc
macros,
so
that
you
could
spell
the
bounds,
the
way
that
they
wanted
to
to
allow
which
would
allow
this
to
be
implemented
either
in
the
compiler
or
as
a
third-party
crate.
G
I
don't
feel
particularly
strongly
about
that.
We
had
chatted
a
bunch
about
sort
of
the
proposed
syntax
and
whether
or
not
it
was
like
actually
secretly
good,
I'm
trying
to
I'm
trying
to
pull
this
all
back
into
cash.
Sorry.
H
G
I,
if
I
remember
correctly,
we
were
kind
of
thinking
that,
like
the
exact
you
know,
derived
changes
that
the
rfc
were
suggesting
were
wasn't
what
we
wanted
yeah,
but
but
then
we
generally
kind
of
liked
the
idea
of
oh.
Actually.
This
is
a
different.
A
I
think
I
don't
like
this
version
like
this.
Well,
I
don't
know,
but
this
syntax
seems
less
appealing
whoa.
What's
going
on
like
a
lot
of
variants
of
derived
versus
a
way
to.
G
A
B
I
I
think
it's
also
potentially
worth
noting
that
at
least
some
of
the
improvements
here
could,
in
theory,
come
as
as
part
of
like
a
larger
effort
around
macros.
I
think
yeah
or
you
know
compiler
extension
or
whatever.
B
B
Feedback
there's.
D
G
I
I
actually
think
the
the
rfc
that
I
thought
I
was
talking
about,
I
actually
kind
of
like,
because
the
the
sort
of
first
step
is
just
we
let
macros
have
a
little
bit
of
extra
syntax.
That
would
allow
people
to
start
implementing,
derives
in
in
nicer
ways
right
and
nicer
ways.
G
A
G
A
G
Me
if
someone
knows
how
this
proposal
handles
overlap
between
methods
in
the
trait
and
methods
inherent
impulse
and
the
inherent
ample.
I
don't.
D
Remember
my
vague
recollection
is
that
it
basically
just
adds
an
inherent
method
and
it
behaves
like
another
inherent
dimple
block,
and
if
you
have
two
things
that
can
conflict
even
crossing
here
and
simple
blocks
could
be
getting
there.
D
G
D
A
Who's
sounds
like
taylor
you're,
really
keen
to
to
write
this.
That's
what
I'm
hearing
sure
I
can
write
a
comment.
You
seem
to
have
a
lot
of
ideas
for
interesting
edge
cases
to
explore
too.
All
right.
Yes,
great.
A
Please
note
that
action
item
mark
rfc
delegation,
two,
three,
nine
three,
oh
I
remember
this.
This
is
we're
so
bad.
This
is
terrible,
bad
in
the
sense
that
I
think
we
requested
this
effort
and
then
just
failed
to
follow
up,
and
it's
got
a
lot
of
emojis
of
various
kinds.
A
A
A
B
G
A
I
wasn't
asking
thinking
of
asking
them
to
do
more
work.
What
I
was
thinking
was
trying
to
find
out
if
they
would
be
willing
to
do
more
work.
If
we
were
willing
to
commit,
but
I
do
agree,
we
should
decide
if
we're
not
willing
to
commit.
Then
it's
just
not
an
important
question.
A
E
But
can
someone
give
me
and
like
I
I'm
skimming
this,
I
don't
want
to
be
like.
Have
a
I've
digested
it
at
all,
but,
like
I'm,
just
really
surprised
this
needs
to
have
such
core
syntax.
When
I
would
have
imagined
there'd
be
some
kind
of
procedural
macro
that
could
get
so.
I
was
just
thinking
this
effect.
H
A
A
Probably
in
the
phase
of
like
we're
going
to
add
a
big
thing,
yeah
yeah,
but
we
will
at
some
point
be
thinking
about
what
maybe
it's
time
to
look
at
a
big
thing,
and
I
would
want
to
consider
this
then
I
was.
A
Like
I
don't
want
to
postpone
it
with
zero,
I
like
the
idea
of
giving
at
least
2022
or
some
specific
time
when
we
would
want
to
reconsider,
because
it
feels
better
than
just
general
postpone,
with
no
condition.
B
H
A
I
think
they
are
but
depends
on
your
definition
of
simple.
I
think
what
this
proposed
was
josh
correct
me.
If
I'm
wrong,
when
you
see
this
syntax,
you
resolve
it.
It
basically
gets
kind
of
rewritten
to
this
to
this
notation,
something
like
that.
So
you
resolve
it
with
the
standard
standard
macro
resolver,
taking
these
tokens
and
putting
them
in
front.
A
D
I
just
mean
whether
it
evaluates
inside
the
context
of
the
macro
or
not
is
complicated
in
terms
of
knowing
how
far
back
the
macro
would
go
like.
If,
if
you
made
a
postfix
unsafe
macro
that
wrapped
in
an
unsafe
block,
what
would
that
do.
D
I
thought
it
was
just
basically
the
equivalent
of
putting
it
into
a
temporary
variable
and
passing
the
value
of
that
into
the
macro
to
avoid
those
things.
A
A
D
B
Well,
I
think
it
was.
It
was
both
right.
We,
I
think
we
weren't
sure
about
how
far
back
it
like
you
know,
there's
benefits
and
negatives
to
it.
Going
back.
You
know
to
expression
or
whatever
and
there's
also
the
type
dispatch
problem,
and
you
know
we
could
say
it's
not
a
problem,
but
he
needs
a
decision.
B
A
Well,
I'm
trying
to
decide
how
much
it
interacts
with
hygiene.
I
mean
this
kind
of
is
a
hygiene
question.
What
we're
talking
about
so
there's
some
interaction.
I'd
be
okay
with
postponing
this
until
2022
as
it
to
consider
in
the
next
roadmap.
A
Actually
that,
maybe
should
just
be
what
postponed
means
altogether,
but
like
it
always
means
that
I
do
feel
like
a
lot
of
people
want
this
and
it
would
be
useful,
but
a
lot
of
things
fit
that
definition.
B
A
E
I
know
I
know
we're
trying
to
move
on,
but
I
just
want
to
double
check
my
understanding
of
what
we
can
propose.
The
the
thing
about
that
rewrite
would
let
ten
people
food?
Doesn't
that
like
break
auto,
borrows
and
stuff,
if
you
like,
expand
into
something
that
needs
to
you
know,
do
a
draft
auto
draft
and
whatever
potentially
yeah?
It's
not
it's
not
it's
like
the
right
thing.
That's
all
I'm
trying
to
say
yeah.
A
D
Yeah,
that's
true.
It
reminds
me
of
the
there
was
a
conversation
on
internals
the
other
day
about
macros
that
wrap
a
macro
argument
in
unsafe
and
thus
could
be
used
to
sort
of
unintentionally
smuggle
and
save.
A
Okay,
I
remember
they
said
they
were
gonna
leave
it,
but
hygiene
opt
out
escaping
for
declarative,
macros
2.0.
I
think
this
is
pretty
clearly
a
post
postpone
as
well
for
the
same
reasons.
B
A
I'm
starting
to
feel
like
I'm
getting
a
lot
of
postponements.
I'll,
take.
C
A
C
I
am
fine
with
postponing
that.
I
would
be
sad
if
we
closed
it.
I
am
perfectly
fine
with
us.
Kicking
it
down
the
road.
A
Okay,
great
teach
oh
who's,
going
to
do
this.
Yeah
give
this
one
to
me.
It's
like
the
same
comment
in
fact:
yeah
okay,
teach
concat
to
join
u8
and
byte
strings.
We
we
recommended.
We
move
this
to
the
libs
team.
At
some
point,
did
we
actually
do
that
sort
of.
C
C
A
Okay
and
give
and
report
that
we
are
happy
if
libs
is
happy
all
right
mark,
you
can
have
that.
A
Awesome
type
changing
struct
update,
syntax,
oh
yeah.
I
think
we
should
accept
this
and
I
encourage
them
to
leave
some
of
the
detailed
entrants
right
up.
Oh
I
see
seven
days
ago,
okay,
26,
oh
man,
I'm
so
behind.
A
Yeah,
okay,
I'm
okay!
With
moving
on
this,
it's
not
that
big,
a
change,
and
I
think
I
could
mentor
someone
through
it.
If
folks
are
generally
positive,
I
would
be
good
to
rfc
approve
myself.
Does
everyone
know
what
this
thing
even
does.
A
E
E
But
it's
it's
it's
more
restricted
than
that
right!
It's
like
it's!
Just
like
the
lifetimes
can
change
it's.
My
memory
like
it's.
It's
not
like
arbitrary
changes
are
permitted.
C
More
comfortable,
you
can
also
do
generics,
so
you're
not
allowed
to
do
arbitrary,
struct
a
to
struct
b.
What
you
can
do
is
changing
from
struct
a
with
parameter
t
to
struct
a
with
parameter.
U.
C
C
A
A
A
I
guess
an
example
would
be
something
like
something
like
struct
of
a
where
a
is
option.
A
c
is
u32
and
you
do
foo
of
a
none.
C
A
Yeah,
I'm
okay
to
work
with
them.
I
mean,
I
think
I
think
I
know
the
algorithm
I
want.
I
don't
know
how
I
think
it's
hard
to
tell
whether
these
are
like
theoretical
inference,
failure,
concerns
or
more
serious
it
might
imply.
We
want
to
do
it
over
an
addition,
for
example,
which
means
we
gotta
act,
but.
H
C
For
that
matter,
we
could
optionally,
if
it
turns
out
that
there
is
an
inference
issue,
we
could
always
defer
this
to
an
addition
and
that
I
don't
think
that
stops
us
from
accepting
the
rfc.
We
just
may
want
to
add
a
note
in
the
rfc
stating
that,
if
implementation
considerations
require
us
to,
we
may
defer
this
to
an
additional
very
for
inference
reasons.
D
One
one
thing
I'd
like
to
bring
up
here:
this
is
very
much
doubling,
so
I
have
a
comment
back
in
august
about
this.
This
is
very
much
doubling
down
on
the
like
current
de-sugaring
of
fru.
D
D
D
This
works
because
it's
copying
over
fields
which
means
that
it's
sort
of
stuck
with
the
privacy
concerns
a
lot
of
people
expect
fru
to
work
as
just
like
move,
move
or
copy.
That's
where
it
gets.
That's
why
that's
where
it
gets
complicated
and
then
update
those
fields
that
it
said
on
the
type
which
cannot
be
type
changing,
but
would
work
for
something
like
non-exhaustive
if
the
fields
are.
C
Public
there's
nothing
stopping
us
from
so
I
don't
think
that
anything
in
this
prevents
us
from
fixing
the
idea
of
non-exhaustive
fru.
We
wouldn't
necessarily
be
able
to
do
it
as
a
mim
move,
but
we
could
always
assign
field
by
field
every
field
that
is
not
currently
overwritten.
The
concern
is
this:.
A
Like
this,
and
then
you
add
this
field
c,
which
has
the
same
type
as
a
you've,
now
broken
this
user.
C
A
A
D
Cool
and
esteban
has
a
rfc
that
he's
planning
on
writing.
I
don't
know
if
people
are
aware
of
that,
one
about
default
field,
values
that
it
will
start
taking
over
from
some
of
the
fru
things
here
anyway.
So
we'll
see.
A
All
right,
152,
oh
we're
doing
we're
not
going
to
get
through,
but
we're
doing
we're
doing
pretty.
Okay,
hidden
trait
implementations,
two
five,
two
nine
there
was
this
was
postponed
proposed
to
postpone,
looks
like
that.
A
A
C
C
We
now
have
min
cons
to
generics,
which
means
that
the
time
to
that
for
that
follow-up
is
conveniently
now,
so
I
think
that
the
appropriate
response
now
would
be
we'd
like
to
see
this.
This
needs
updating
to
take
into
account
that
we
now
have
mint
cots
generics.
C
Rather
than
trying
to
revise
this
on
this
thread,
we
should
close
postpone.
We
should
rfc
bot,
postpone
and
request
an
updated
version
that
takes
const
generics
into.
C
Rfc
people
think
that
the
right
answer
is
well.
It
depends
if
we
think
the
right
answer
is
just
revise
this
to
take
const
generics
into
account.
Then
we
can
post
that
response
and
then
just
wait
or,
alternatively,
we
could
postpone
and
request
an
updated
rfc
which
do
we
think
is
going
to
be
more
productive
here.
C
Okay,
then,
I
would
be
happy
to
post
that
and
state,
like
generally
in
favor,
want
to
see
this
written
on
top
of
the
newly
available
men
const
generics.
Please
start
a
new
rfc
and
a
new
threat.
D
Oh,
wait
is
sorry,
is
this
also
about
const
or
about
lats.
A
I
think
this
is
about
both
but
they're
somewhat
different.
Let
is
probably
much.
D
D
A
Agreed
all
right,
that's
fine!
We
can
tackle
those
when
the
arts
get
reopened
or
project.
I
might
recommend
a
project
proposal,
but
okay
got
five
more
minutes
associated
type
lifetime.
Illusion
this
said
that
associated
types
are
output,
so
they
should
link
to
inputs.
However,
there
was
some
concern
about
confusion.
A
It's
been
closed,
wonderful,
it's
too
bad
because
I
liked
it,
but
then
I
think
I
decided
I
didn't
like
it
that
much.
I
think
we
merged
this
didn't
we.
H
A
E
D
A
Enum
variant
types:
this
is
something
I
definitely
want
as
a
user.
It
feels
like
a
big
thing.
I
think
postponed
till
next
roadmap
is
the
appropriate.
C
C
A
B
A
A
Something
like
that
yeah,
I
don't
know
if
it's
that
quite
that,
like
it
might
come
up,
it
might
even
be
like.
I
don't
know
exactly
how
it
works,
but
that's
kind
of
what
I'm
saying.
Yes,
even
if
it
was
just
like
once
a
quarter,
we're
going
to
talk
about
big
ideas
and
see
if
we
have
any
bandwidth
but.
A
Okay
taking
it
because
I
think
I'm
gonna
copy
and
paste
my
message
from
another
place,
basically
saying
we
agree,
this
is
a
real
problem.
We
don't
think
we
have
the
bandwidth,
take
it
on
right
now.
We
would
like
to
encourage
you
to
like
when
we
are
talking
about
roadmap
ideas,
we'd
like
to
encourage
it
to
be
brought
up
again,.
A
B
B
A
Yeah,
I
agree
that
we
should
wish
to
see
25.80
fully
implemented
and
have
time
to
gain.
Experience
would
consider
for
a
roadmap
item
after
that.