►
From YouTube: 2021-09-20 Cross Team Collaboration Fun Times (CTCFT)
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
Okay,
welcome
to
the
september
russ
ctcft.
Today
we
got
a
different
kind
of
agenda
in
prior
meetings.
We've
done
a
lot
of
more
prepared
slides,
but
this
time
I
was
talking
with
forest
and
we
thought
we
wanted
to
do
something
to
do
with
interest
groups,
and
we
thought
we
might
do
something.
A
little
more
interactive,
a
little
more
of
a
roundtable
discussion
with
various
people.
A
Who've
been
involved
in
running
interest
groups
and
with
the
folks
who
are
here
from
the
rest
teams
to
talk
about
kind
of
how
interest
groups
and
the
rust
oregon
stuff
can't
interact.
A
A
I
guess
I'll
kick
off
and
say:
you've
probably
seen
it,
but
I
I've
been
working
on
this.
We
call
the
rust
station
principles
and
I'm
really
eager
to
talk
to
people
about
them
who
have
thoughts
about
them.
So
let
me
know
if
you
do,
and
we
can,
we
can
discuss
you'll
find
them
at
well,
you'll
find
them
on
my
blog.
I
don't
know
what
the
url
is
station
dash
principles,
dot,
something
or
other
netlify
or
something
anyone
else
have
an
announcements
they'd
like
to
make.
A
B
Like
interested
user,
but
at
the
same
time
I
come
more
from
like
the
practitioner
side
and
I
feel,
like
formal
methods,
is
very
much
like
an
academic
discipline.
I'm
very
excited
about
people
who
are
trying
to
like
build
that
bridge
between
the
two
of
them.
So
oh
yeah.
C
Hey
everyone
hi
sorry,
I
have
to
whisper
a
bit
because
it's
2am
here
in
london
and
yeah.
I
really
wanted
to
make
this
panel.
So
my
involvement
in
the
rust
formal
interest
group
and
the
rust
cryptography
interest
group
is
they
are
sister
interest
groups
kind
of
like
running
in
parallel
and
we're
just
exploring
the
synergies
between
the
two
and
I
am
helping.
C
I
co-lead
the
rust
cryptography
interest
group
with
tony
so
hey
tony
and
I'm
helping
support
the
rust,
formal
interest
group,
the
frost,
formal
methods,
interest,
group,
kind
of
exist,
just
facilitating
and
being
like
a
supporting
partner
for
the
rust
formal
methods,
interest
group-
and
we
had
the
inaugural
call
today
well
earlier
today
in
in
in
like
european
time-
and
it
was
like
the
first
ever
meeting
and
it
was
brilliant.
It
went
really
well,
you
were
there
nico.
C
There
were
a
bunch
of
other
people
there,
it's
being
led
by
xavier
dennis
and
bass
spitters,
and
it
was
a
lot
of
fun.
Yeah.
A
That's
exactly
it,
I
thought
it
might
be
nice
for
people
to
know.
We
launched
a
new
interest
group
today,
even
even
even
as
we're
talking
about
them
all
right,
any
last
announcements.
People
want
to
make
otherwise.
I
have
another
brief
one.
A
Okay,
so
I
just
wanted
to
say
one
thing:
is
you
know
this
is
the
third,
yes
or
fourth,
I
don't
know
anyway.
A
We've
had
a
number
of
these
these
meetings,
and
you
know
they
started
out
as
a
kind
of
experiment
for
like
just
doing
something
a
little
different
than
we
had
been
doing
and
trying
to
bring
together
the
community
in
a
different
way
than
just
interest
boards,
and
so
on,
trying
to
figure
out
if
we
could
get
people
to
have
some
synchronous
discussion,
while
not
a
lot
of
times
when
we
try
to
do
that,
we
end
up
leaving
people
out
so
trying
to
make
different
structures
that
include
different
time.
A
Zones
include
more
people
if
we
could
have
new
ways
to
share
experiences
and
ideas,
you
know
if
we
could
give
it
ridiculous
names
and
have
people
still
show
up.
I
know
you
all
love
it,
and
so
it
seems
like
it's
working.
It
seems
like
we're
having
it's
created,
a
kind
of
a
new
venue
for
the
community.
A
So
what
I
want
to
do,
the
forest
that
I've
been
talking
about
is
trying
to
create
a
group
along
with
some
others
around
it
to
help
it
get
organized
right,
and
so,
if
that's
something
you'd
be
interested
in,
then
you
should
contact
forest
on
discord
or
zulip,
and
we
can
talk
about
it.
What
that
is
how
that's
gonna
work
going
forward,
but
so,
apart
from
that,
I'm
gonna
hand
it
off
now
oops,
that's
the
minutes,
where's,
the
slides.
A
I
don't
know
they're
here
somewhere
there
we
go,
I'm
gonna
hand
it
off
for
the
round
table
and
we'll
see
where
it
goes.
D
Perfect
well,
thank
you
very
much
nico
and
so
nico
reached
out
to
me
about
this
a
while
ago
and
I've
been
super
excited
to
be
able
to
sit
down
and
chat
with
these
other
working
and
interest
group
leaders
and
sort
of
be
able
to
address
a
lot
of
the
core
rust
teams
on
a
lot
of
potentially
problems
in
our
our
working
groups
or
areas
where
there
could
be
improvement
or
just
kind
of
give
some
information
about
how
we
might
be
doing
things.
D
And
so
I
I
want
to
be
able
to
highlight
some
of
the
areas
that
we're
doing
with
maybe
community
growth,
maybe
with,
as
I
mentioned,
like
problems
in
our
working
groups.
But
I
do
want
to
focus
on
items
that
are
more
core
to.
D
These
sort
of
like
discussions
now,
will
be
to
inform
different
teams
about
what
kind
of
stuff
each
one
is
working
on,
but
then
also
promote
this
sort
of
communication
and
collaboration
between
teams
that
aren't
really
part
of
like
the
inner
sanctum
of
like
the
core
or
the
laying
teams,
or
anything
like
that.
D
And
so,
as
I've
been
talking
with
several
of
these
other
team
leaders
over
the
past
week
or
so,
I've
definitely
been
pretty
interested
to
see
that
a
lot
of
my
problems
aren't
necessarily
the
same
as
what
they're
finding
and
so
we're
gonna
have
an
opportunity
to
compare
and
contrast
a
lot
of
our
different
ideas
and
opinions
and
hopefully
be
able
to
bring
some
valuable
information
to
the
people
who
can
make
very
structural
changes
to
the
language.
D
And
so,
when
it
comes
to
the
format
of
this
round
table,
I'm
trying
to
think
of
it
sort
of
like
a
podcast
episode
in
a
little
bit
of
a
way
where
we'll
have
several
questions
that
we'll
come
in
with.
And
hopefully
this
will
give
us
enough
to
discuss
and
we've
kind
of
like
structured
it
in
a
way
that
we
think
this
is
like
the
most
valuable
stuff
we
want
to
go
over.
D
But
the
audience
that
is
here
is
very
involved
with
a
lot
of
different
things
all
across
the
rust
language.
So
I
think
it's
very
important
that
if
the
audience
has
particular
questions
that
might
pertain
to
their
specific
working
groups,
then
it
would
be
a
really
great
idea
to
involve
some
of
these
questions
in
like
the
discussions,
and
so
I
did
feel
free
to
put
this
put
it
in
the
chat,
I'm
going
to
be
keeping
an
eye
on
that
throughout
the
panel,
and
hopefully
we
can
get
some
interesting
discussions
going
on.
D
So
for
the
presenters
we
have
five
or
four
different
people
from
four
different
working
groups.
A
D
Perfect
yeah,
I
think
that
would
be
good.
I
yes,
I
suppose
I'll
keep
an
eye
on
chat
and
then
I'll
call
people
up
if
they
want
to
ask
a
particular
question
or
if
you
just
want
to
put
in
shot.
I
can
also
keep
an
eye
on
that
and
so
for
the
separate
presenters
that
we
have
today
and
you
could
just
go
to
the
next
slide
here.
D
So
we
have
four
different
working
group
leaders
across
four
different
yeah,
four
different
working
groups,
so
we
have
game
development,
embedded,
cryptography
and
machine
learning,
and
so
we're
going
to
go
through
some
introductions
right
now.
So
for
myself
my
name
is
forrest
or
my
handle
angel
on
fire.
D
D
I
help
edit
the
rust
gamedub
newsletter,
and
so
this
is
an
opportunity
to
sort
of
gather
what
the
community's
been
working
on
over
the
past
month
and
categorize
it
into
some
sections
that
you
can
digest
pretty
nicely
and
then
I
also
run
a
monthly
meetup
on
the
second
saturday
of
each
month,
and
so
this
is
just
another
opportunity
for
people
to
kind
of
have
a
platform
to
show
off
what
they've
been
working
on,
and
so
next
up
I'll
pass
it
over
to
adam
grieg,
who
is
part
of
the
embedded
working
group.
E
E
E
Most
of
the
work
we
do
these
days
is
in
maintaining
a
sort
of
set
of
core
crates,
so
we
have
some
crates
that
are
used
across
the
board
for
specific
architectures
and
we
have
some
crates
where
we
have
some
sort
of
abstractions
and
traits
that
lots
of
different
people
influence
in
different
ways
and
it's
useful
to
have
a
sort
of
central
location
to
collaborate
on.
We
also
help
coordinate
between
lots
of
other
projects
that
are
somewhat
related.
E
So,
for
example,
was
the
stm32
project
I
run
and
there's
a
similar
project
for
different
manufacturers
embedded
devices
and
architectures,
and
we
provide
a
sort
of
central
place
for
people
to
collaborate
and
talk
to
each
other
and
coordinate,
trying
to
make
things
fit
together
and
in
principle.
We
also
help
coordinate
the
needs
of
embedded
users
with
the
rustling
core
team
and
I
feel
like
in
the
last
year,
we've
not
really
done
a
great
job
of
that.
So
that's,
maybe
something
we'd
like
to
try
and
improve
on,
and
I
guess.
E
D
Perfect,
thank
you
adam.
So
next
up
is
tony
archery.
I
apologize
for
pronouncing
wrong,
which
is
who
is
part
of
the
cryptography
working
group.
Yeah.
B
So
it's
actually
our
siri
it's
impossible
to
pronounce
the
italian
sartri
or
something
like
that,
but
I
work
on
a
few
different
things
throughout
their
rust
ecosystem.
B
I'm
one
of
the
co-founders
of
the
rest
cryptography
interest
group.
Along
with
earnest.
I
also
worked
on
a
project
called
res
crypto,
not
the
ancient
crate
that
hasn't
been
updated
in
like
six
years,
but
the
github
organization
of
the
same
name.
I
know
it's
kind
of
confusing,
so
at
least
I
would
say
with
the
well
I
should
also
say
I
also
work
on
russ
sack,
which
is
a
vulnerability
advisory
database
for
rust,
and
everybody
should
run
cargo
on
their
projects,
so
within
kind
of
the
preview
of
ross
crypto.
B
I
mainly
work
on
the
really
boring
stuff
that
everybody
uses,
so
it's
kind
of
like
the
nuts
and
bolts
of
things
like
symmetric,
cryptography,
elliptic
curves,
that
kind
of
thing
so,
like
the
aes
crate,
the
asgcm
crate,
if
you've
ever
used
either
those
the
digest,
create
the
shot
to
create
things
like
that.
B
So
just
kind
of
these
like
standard
algorithms,
that
everybody
has
to
use
because
they're
used
everywhere,
that's
the
kind
of
stuff
we
work
on.
So
the
rest,
cryptography
interest
group,
there's
just
so
much
cool
rust,
cryptography,
stuff
happening
in
the
ecosystem,
and
we
felt
it
was
all
kind
of
like
happening
in
silos.
B
People
weren't
really
talking
and
sharing
finding
ways
to
collaborate,
and
that
kind
of
thing.
So
I
think
really
that
was
the
impetus
of
the
working
group
to
kind
of
like
solve
some
cross-functional
problems
in
the
rest,
cryptography
ecosystem
and
maybe
just
a
little
bit
of
trying
to
brainstorm
how
we
could
improve
the
language
for
cryptography.
D
Perfect,
thank
you,
tony,
and
so
next
up
we
have
ricky
hosfelt,
who
is
part
of
the
machine
learning
interest
group
hi
everyone,
I'm
ricky.
B
I
basically
saw
the
the
need
for
a
machine
learning
interest
group
just
because
I
was
working
in
the
time
at
a
university
research
center
that
you
know
everyone
knew
I
liked
rust
and
it
was
basically
just
writing
a
bunch
of
models
for
some
professors,
and
everybody
asked
me:
why
aren't
you
using
rust
for
it?
And
I
was
like
good
question,
let
me
go
look,
and
that
was
because
the
ecosystem
was
pretty
fractured
at
the
time
and
there
was
hardly
anybody
in
it.
B
So
I
figured
let's
try
to
stand
one
up
and
see
what
happens
and
see
other
people
we
get.
So
that's
pretty
much
the
start
of
it
and
it's
kind
of
gone
pretty.
Well,
since
then,
we've
had
some
some
great
people
standing
on
the
shoulders
of
giants
as
they
say,
and
we've
paved
the
way
with
with
luca
and
everyone
doing
the
lymphocrate.
So
the
the
schumann
interest
group
has
taken
over
that
as
well
as
basically
like
some
of
the
other
groups
mentioned.
Writing
the
book
on.
B
You
know
kind
of
how
to
use
rust
for
machine
learning.
We
also
maintain.
Are
we
learning
yet.com
and
the
answer
is
still
no,
but
we
we
are
gathering
crates
and
we
are
working
towards
finding
the
areas
of
machine
learning
that
we
need
to
actually
kind
of
lay
the
groundwork
in
order
to
make
sure
that
people
can
use
machine
learning
and
rust.
D
Perfect,
thank
you
ricky,
so
I
first
want
to
dive
in
and
talk
a
little
bit
about
our
communities
and
maybe
how
maybe
we
view
them
as
part
of
where
we
all
meet
and
sort
of
are
able
to
talk
together
or
where,
instead,
people
can
go
and
discuss
different
issues
that
need
to
be
worked
on
and
so
sort
of
just
like
to
start.
This
out,
I
want
to
like
propose
a
question
like
what
what
does
that
look
like
for
your
working
group
and
so
in
game
development?
D
In
particular,
we
focus
a
lot
on
different
discord,
servers
and
so
especially
because
a
lot
of
game
development
happens
within
the
game
engines,
or
at
least
the
development
of
them.
Then
there
will
be
a
lot
of
different
communities
are
focused
on,
for
example,
the
the
bevvy
engine
or
the
amethyst
engine,
and
so
they
will
each
have
their
own
communities
and
then
a
lot
of
smaller
communities
will
arise
from
different,
maybe
crates
or
maybe
games
that
are
being
worked
on
and
so,
for
example,
with
velouring.
D
We
have
a
quite
a
large
discord
server
as
well,
but
then
we
also
have
a
central
hub,
which
is
a
just
well
a
discord
organization
where
we
are
able
to
sort
of
discuss
a
lot
of
the
overarching
issues
that
we
have
or
that
we
want
to
work
towards
improving,
as
in
maybe
we
need
some
like
lower
level
graphic
stuff
to
be
able
to
target
different
platforms
or
just
as
general
stuff
in
the
ecosystem
that
needs
to
be
worked
on
so
for
for
us,
it's
mostly
a
lot
of
game
development.
D
Oh
sorry,
this
is
a
lot
of
github
and
discord
and
then
I'll
just
go
in
reverse
order
here
so
ricky,
can
you
tell
us
a
little
bit
about
the
machine
learning
community.
B
Sure
so
we
organized
mainly
on
zulip.
We
picked
that
just
because
it
seemed
like
the
least
resistance
path.
It
worked
out
pretty.
Well,
we
can
have
you
know
several
channels
we
can.
I
can
sub
people
to.
B
Basically
you
know
a
certain
group
of
channels
when
they
first
join
and
just
kind
of
let
any
number
of
people
kind
of
create
channels
or
restrict
that,
so
it
worked
out
pretty
well
when
I
was
looking
at
the
differences
between
different
mediums,
usually
once
every
other
week,
we'd
meet
up
on
zoom
around
two
o'clock
eastern
time.
That
seemed
to
work
pretty
well,
because
a
lot
of
us
are
span
the
coast
of
the
us
and
are
in
western
europe,
so
it
did
seem
to
work
out
pretty
well.
B
I
think
we
started
at
five
o'clock
and
then
I
kind
of
held
a
you
know
a
democratic
vote
in
the
zulu
channel.
We
only
got
about
10
responses,
but
that
was
overwhelmingly
around
early
afternoon.
U.S
time
seemed
to
work
for
us.
B
The
zulu
channel
was
basically
a
place
to
either
meet
and
talk
and
kind
of
ask
questions
and
introduce
yourself
as
well
as
that's
where
we
coordinate
the
meetings
we
post,
we
post
the
meetings
in
a
specific
meeting
channel
that
everyone's
involved
in,
and
we
just
kind
of
either
kind
of
throw
ideas
at
each
other.
One
of
the
big
one
of
the
biggest
things
is
because
we're
spread
across
such
a
time
zone
that
kind
of
like
either
I'm
in
bed
and
everyone's
awake
or
the
opposite.
B
We
will
post
crates
that
you
know
we
find
on
craigslist
io
or
a
certain
article
and
just
kind
of
looks
at
hey.
Do
we
know
about
this
is
on?
Are
we
learning
yet.com?
Is
it
worth
putting
on
there?
Does
it
only
have
one
download
that
just
exists
overnight?
Do
we
think
it's
gonna
be
useful
stuff
like
that?
So,
and
you
know
it
works
out
pretty
well
to
kind
of
chat
a
certainness
like
this.
B
That
way,
we
can
get
that
kind
of
work
done
before
we
actually
meet
for
only
an
hour
out
of
the
day
and
everything.
So
is
there
anything
else
that
I
can
add
to
that
that
I
may
have
skimmed
over.
D
Well,
I
think
one
thing
that
I
find
interesting
sort
of
about
the
way
that
you're
describing
communicating
with
the
the
community
is
that
a
lot
of
it
is
sort
of
to
gain
knowledge
together
and
so
with
the
end
goal
of
moving
stuff
to
are
we
learning
yet?
And
so
would
you
say
that
like
are,
we
learning
is
really
the
best
way
for
newcomers
to
be
able
to
get
into
the
working
group?
Maybe
I
would.
B
Say
that's
a
good
way
to
capture
what
the
environment
looks
like
and
rest
now
for
machine
learning,
people
they'll
see
a
lot
of
things
that
they
recognize.
You
know
all
the
terms
they
they
like
to
look
for
and
as
well
as
like
a
very
nice
gridded
pattern
on.
B
Does
it
exist
in
rust
yet
and
if
it
does,
if
it's
not
on
the
website,
if
it's
not
there
there's
a
blank
box,
they
know
that,
oh
okay,
maybe
it's
not
there
organization,
wise
zoolop's,
probably
the
best
way
as
long
as
we
also
have
a
github
group
as
well.
D
Yeah-
and
I
think
I
really
like
your
point
about
sort
of
finding
times
for
the
weekly
meetings,
because
I
think,
depending
on
the
size
like
with
game
development,
we
a
while
ago,
we
did
have
these
sort
of
weekly
meetings,
but
they've
sort
of
like
tapered
on
and
off
and
then
fragmented
into
different
periodic
meetings.
Maybe
like
the
russian
community,
will
I
think
they
have
their
own
type
of
meetup
and
stuff
like
that
and
so
yeah.
D
I
think
it's
interesting
that,
with
like
the
smaller
interest
group,
is
you're
able
to
sort
of
find
a
time
and
dive
in
on
that
yeah
all
right.
So
tony
a
little
bit
about,
maybe
the
the
cryptography
community.
B
There's
our
interest
group,
the
rest
cryptography
interest
group,
which
is
kind
of
fairly
new,
so
we've
been
doing
synchronous
meetings
on
kind
of
like
a
monthly
cadence
for
about
the
past
five
months,
so
I
think
that's
kind
of
new
in
terms
of
these
groups,
just
in
terms
of
the
wider
rest
community,
I
think
rust
was
kind
of
this
crazy
sleeper
language
that
came
in
and
like
dominated
the
entire
cryptography
space
and
people
weren't,
really
you
know
getting
together
and
talking
about
that
specifically.
B
But
just
personally
for
me,
I
go
to
like
a
lot
of
cryptography
conferences
and
conferences
on
things
like
zero
knowledge,
proofs
and
that
kind
of
thing
I
would
say,
over
the
course
of
the
past
five
years
or
so
there's
been
this
whole
cohort
of
people
doing
crazy
things
with
cryptography.
I
think,
wouldn't
be
possible
without
language
like
roster,
but
cryptography
people.
Maybe
they
go
and
kind
of
do
things
in
silos
and
they
do
this
really
cool
work,
but
they
don't
necessarily
like
get
together
and
talk
about
advancing
the
general
cause
of
rest
cryptography.
B
So
much
as
like
cartography
as
a
field,
so
yeah
aries,
you
have
something
to
say
about
that.
C
Yeah
yeah,
no
thank
you
yeah.
I
can
jump
in
on
that
and
it's
it's
a
it's
really
interesting
part
of
the
rust
ecosystem
and
looking
from
the
outside
in
it
felt
like
again,
there
was
a
level
of
fragmentation
and
people
working
in
silos,
so
you
know
coming
in
and
kind
of
being
the
calvary
to
kind
of
get
people
talking,
get
people
cross-pollinating
and
helping
each
other.
That's
really
the
mandate
that
we
set
for
ourselves
when
me
and
tony
got
together
to
form
this
this
interest
group.
C
It
was
about
kind
of
like
making
sure
that
there
was
an
apex
point
that
we're
all
working
towards
where
people
come
in
and
leave
their
agendas
at
the
door,
and
we
work
towards
really
kind
of
forming
out
this
important
part
of
the
rust
ecosystem,
which
which
was
a
bit
swiss
cheese.
C
If
you
know
if
that
makes
sense,
and
that's
where
the
sync
up
calls
come
in,
we
style
our
monthly
calls
as
sync
up
calls
where
we're
looking
at
blue
sky
thinking
for
the
ecosystem.
First
of
all,
where
cryptography
is
concerned-
and
things
like
what
do
we
want
to
see
in
cryptography
what
we
want
to
see
done
and
how
can
we
all
work
together
to
make
that
happen?
C
So
we
we,
we
have
a
lot
of
kind
of
like
mandates
and
to
do's
and
a
lot
of
things
that
kind
of
lead
us
towards
a
point
where
we
look
at
other
interest
groups
and
other
working
groups
to
kind
of
aspire
to
be
like
them
and
be
like
hey.
You
know
what
the
embedded
group
are
really
great.
At
governance
like
we
need
an
awesome,
rust,
cryptography
website
as
well,
where
we
can
showcase
all
of
the
different
crates
that
are
being
used
for
cryptography,
because
we
don't
have
that
yet.
So
what
does
that
look
like?
C
How
can
we
work
towards
that,
regardless
of
our
own
specific
agendas?
And
that's
where
you
know
the
power
of
an
interest
group
really
comes
into
play
where
we
really
able
to
get
people
into
the
same
room,
to
kind
of
move
towards
an
apex
point
to
get
things
done.
You
know
also
having
nico
come
in
as
a
special
guest
to
help
us
give
us
ideas
around
vision,
documents
and
things
like
that.
What
does
a
vision
document
for
the
cryptography
space
look
like
and
so
on?
C
D
I
think
it's
very
interesting
to
hear
about
the
agility
that
sort
of
comes
with
an
interest
group
in
sort
of,
since
it's
like
a
little
bit
newer
and
you're
able
to
put
a
lot
of
work
into
start
it
yourself,
then
you
really
have
the
ability
to
make
these
changes
that
you
want
to
see,
because
you
want
to
like
push
the
the
agenda
of
getting
your
sort
of
well
yeah
your
interest
as
as
part
of
something
that
can
be
used
in
rust,
and
so
I
think,
maybe
an
interesting
distinction
where,
when
it
when
it
comes
to
the
game,
development
working
group
is
that,
since
it's
so
large
and
in
a
way
more
democratic
in
a
way
sort
of,
I
think
I'm
sorry
not
not
that
it's
less
democratic
for
a
smaller
one,
but
rather
when
there's
so
many
separate
parties
coming
into
a
working
group
that
are,
for
example,
creating
game
engines
or
doing
graphics
or
doing
some
something
completely
different.
D
Working
on
color
theory
and
stuff.
Like
that,
I
think
then
there
is
the
the
common
the
common
interest,
but
it
is
definitely
more
difficult
to
move
in
a
large
forward
direction.
I
think
so
yeah
it's
kind
of
cool,
to
hear
what
the
what
the
the
pivot
of
the
the
interest
group
looks
like
all
right
and
then
adam.
E
Sure
yeah,
it's
interesting
to
me,
saying
just
now
about
the
slightly
larger
game,
dev
group
having
some
sort
of
maybe
less
agility
or
maybe
harder
to
bring
everyone
together.
I
feel
like
we
have
a
very
similar
thing
and
embedded
where
it's
a
reasonably
large
group,
there's
maybe
14
members.
Apparently
I
don't
really
have
depends
how
you
counted.
E
I
guess
and
a
lot
of
people
doing
various
embedded
things
but
there's
it
feels
like
there's
much
less
agility
and
much
less
ability
to
quickly
move
around
and
change
things
just
because
I
guess
you're
right
again:
it's
a
sort
of
democracy.
It's
not
really
the
word,
but
a
lot
of
different
people
putting
in
different
directions.
E
So
in
the
embedded
working
group
to
round
out
the
sort
of
collection
of
chat
systems,
we
use
matrix,
really
almost
exclusively
all
different
communities
on
matrix
where
we
have
one
main
embedded
room,
and
I
guess
it's
a
bit
like
an
ioc
channel
or
something
right.
A
home
where
we
come
in
people
chat
all
the
time
we
have
a
weekly
meeting
at
8
p.m:
berlin
time
for
some
historic
reason.
Every
week
since
the
start
of
2018,
I
think,
but
it's
still
all
in
chat.
E
So
while
it's
synchronous,
we
don't
really
have
video
or
voice
calls
it's
pretty
much.
Just
anyone
who
wants
to
drop
by
you
can
chat
at
first.
It
was
a
more
structured
meeting
where
we
very
much
deliberately
setting
out
at
one
point
we
were
writing
a
book
and
the
meeting
each
week
was
okay,
we're
going
to
get
these
chapters
done
on
this
book.
Who
can
do
this
chapter?
Who
can
do
this
chapter
these
days?
E
It
tends
to
be
a
bit
more
open-ended,
so
something
will
come
up
for
some
issue
on
a
crate
somewhere,
where
we
need
some
discussion
or
something
to
be
nominated
for
discussion
this
week.
So
we
have
a
label
across
all
of
our
github
repos,
to
nominate
specific
issues
for
discussion
or
just
you
know,
there's
nothing
in
particular
when
you
talk
about
that
week,
so
a
few
people
turn
up
and
we
just
chat
about
flip
is
interesting
and
embedded,
and
that
seems
to
work
out
reasonably
well
for
us
now.
It
does
seem
like,
as.
D
E
Scope
has
gotten
bigger
as
the
number
of
people
doing,
embedded
or
not
has
gotten
bigger
in
some
ways
the
specific
responsibilities
of
a
working
group
have
got
smaller
in
that
we're
maintaining
our
crates,
but
a
lot
of
concerns
now
backwards.
Compatibility
and
stability
of
these
things
that
more
external
users
have
come
to
rely
on
and
we're
feeling
perhaps
less
pressure
to
go
out
and
solve
specific
problems,
more
trying
to
coordinate
people
and
keep
things
slowly
ticking.
E
I
was
just
trying
to
say
yeah,
I
guess
that's
the
gist
of
how
our
community
organizes.
Of
course,
we
have
a
github
organization
and
we
have
a
lot
of
teams
inside
that
maintaining
different
sets
of
crates.
I
guess
that's
pretty
similar
for
everyone.
E
At
various
points
we
tried
to
do
so
yeah
one
other
thing
I
guess
to
push.
The
development
forward
was:
have
sort
of
teams
create,
wish
lists
and
goals
for
each
year.
So
at
the
start
of
the
year,
we'd
have
a
bigger
meeting
and
say:
okay.
Actually,
what
do
these
things
want
to
accomplish?
Looking
forward,
you
know
what
are
important
things.
We
want
to
try
and
get
done
or
have
other
embedded
users
come
to
us
and
say
we're
really
missing
this.
E
You
know
someone
says
I'd
love
to
have
a
real-time
operating
system,
invest
and
say:
well,
can
we
write
one
for
you?
So
if
you
are
maybe
not,
but
in
the
last
couple
of
years
that's
also
slowed
down
a
bit.
It
seems
like
a
lot
of
the
novel
development
effort
is
now
happening
outside
of
a
working
group,
sometimes
with
working
group
members.
Sometimes
people
who
don't
happen
to
be
in
the
working
group
but
they're
pushing
forward.
E
We
embedded
stuff
they're
looking
at
in
a
separate
space,
and
they
just
come
to
the
working
group
to
chat
and
coordinate
with
everybody
else,
and
that
seems
to
be.
I
don't
know
if
there's
a
lot
of
projects
flourishing,
which
we
don't
have
to
directly
write
ourselves
right.
We
just
help
people
accordingly,.
D
Yeah
a
lot
of
really
great
points
there
and
I
just
want
to
cover
adam
and
my
back
by
saying
that
we're
not
trying
to
say
that
our
groups
are
not
agile
or
that
there's
like
too
much
overhead,
and
rather
I
think
that
when
people
come
up
with
new
initiatives
and
want
to,
for
example,
run
these
types
of
weekly
meetups
or
get
something
going,
it's
very
possible
to
do
that.
But
I'm
just
sort
of
inspired
by
ernest
sort
of
talking
about
the
ability
to
do
stuff,
yeah
just
so
freely.
D
Pretty
cool,
yeah,
okay,
perfect,
and
so
next
up.
D
I
want
to
dive
into
the
problems
that
our
working
groups
sort
of
experience,
and
so
this
was
an
interesting
one,
just
because
I
I
think
what
I
assume
the
problems
within
game
development
are,
I
I
sort
of
just
assume
that
that's
very
broad
across
all
of
russ,
but
when
I
was
sort
of
telling
nico
that
one
of
our
largest
problems
was
compile
time,
niko's
a
little
bit
more
surprised
by
that,
and
so
I
think
that
we've
seen
this
a
lot
more
in
more
in
modern
or
rather
like
the
more
recent
game
engines
in
rust
is
that
they
will
focus
very
heavily
on
compile
time
and
so,
for
example,
with
bevy.
D
It
has
a
specific
set
of
instructions
in
the
tutorial
that
will
teach
you
how
to
or
not
really
teach
you
but
show
you
some
ways
to
get
faster,
compile
times
and
then
bevy
also
uses
some
techniques
to
reduce
compile
times
because
of
how
the
engine
is
built,
and
so
within
a
lot
of
rust
game.
We're
sorry
within
a
lot
of
game
development.
D
Iteration
time
is
incredibly
important,
and
so
this
is
either
solved
by
being
able
to
do
more
while
you
are
in
game
or
if
you
need
to
reboot
the
game
many
times
making
sure
that
compile
time
is
low,
and
so
a
potential
solution
is
like
the
idea
of
hot
reloading,
and
so
you
can
keep
the
game
running,
but
you
can
be
changing
design
values
which
will
affect.
Maybe,
yes,
some
part
of
the
game,
maybe
more
enemies
or
something
looks
a
little
bit
different.
D
But
then
a
a
large
as
your
game
gets
a
lot
larger.
You
are
going
to
have
to
start
caring
a
lot
more
about
the
fact
that
it's
going
to
take
half
a
half
a
minute
to
compile,
or
anything
like
that.
So
I
kind
of
want
to
pass
this
around
and
have
this
open
for
any
of
the.
The
other
leads
that
kind
of
want
to
hop
in
on
maybe
discussing
some
of
the
the
issues
that
they
find
relative
to
their
particular
working
or
interest
groups.
D
Or,
if
not
I'll,
I'll
click
on
pick
on
some
of
them,
I
or
tony
okay.
B
Okay,
yeah!
Well,
I
can
hop
on
here
I
mean
so,
oh
preface
this
by
saying
we
created
an
interest
group
instead
of
a
working
group,
because
we
think
rust
is
already
such
a
great
language
for
cryptography,
like
you,
don't
need
to
make
specific
language
changes
and
there
are
a
handful
of
ones.
We
think
would
be
helpful,
but
they're
such
like
crazy
monster
problems
that
we
don't
really
expect
them
to
get
solved
anytime
soon
and
yeah.
At
the
same
time,
the
more
I
like
explore
any
one
of
these
given
problems.
B
It's
just
we're,
so
I
should
also
say
we're.
Gonna
have
a
call
this
friday
to
talk
about
some
of
them.
They're,
really
like
very
specific
things
for
cryptography.
I've
been
trying
to
boil
it
down
to
like
what
is
the
minimum
viable
primitive
for
having
constant
time
code
in
rust?
What
would
that
look
like?
What
does
that
need
to
touch?
B
It
touches
everything
it
touches
like
code
gen,
it
touches
llvm,
I'm
talking
to
ovm
experts
and
we're
like
how
do
we
get
these
guarantees
out
lvm?
And
it's
like
we
need
this
really
crazy
monster
change
to
lobm.
That
probably
isn't
gonna
happen,
so
otherwise,
I
would
say,
like
you
know,
rust
as
it
is,
is
fantastic
for
cryptography,
even
if
we
can't
get
like
guaranteed
solutions
to
these
specific
kind
of
like
monster
problems
or
just
really
hard
to
solve
in
any
language,
but
that
that,
like
that
would
be
one
of
them.
B
How
do
we
get
this
on
intel
cps?
It's
called
c
move.
It
would
be
this
one
instruction
to
do.
Branching
and
constant
time
effectively.
Just
turns
out
to
be
very,
very
tricky
and
yeah
I'll,
we'll
we'll
see
if
it
ever
gets
solved
and
anyone
going
shout
out.
E
Happy
to
jump
in
with
some
embedded
stuff,
if
you
like
forest,
so
I
don't
know
we
have
totally
different
problems.
I
guess
the
compulsion
is
almost
never
an
issue.
Our
crates
tend
to
be
a
lot
smaller,
I
suppose,
and
usually
only
on
a
device
with
you
know
some
hundreds
of
kilobytes
of
ram
and
flash.
You
just
can't
write
how
much
code
the
executable
wouldn't
fit,
but
instead
I
don't
know.
I
think
we
have
the
biggest
rust
related
problems.
We
have.
E
We
keep
hitting
stupid
corner
cases
in
memory
models,
especially
we'll
discover
that
the
sort
of
fundamental
abstraction
that
we
used
across
the
whole
embedded
ecosystem
is
unsound
and
we
need
to
rewrite
everything.
That's
happened
twice
now.
I
guess
it's
like
oh
whoops,
it
turns
out.
Llvm
can
just
insert
a
read
this
volatile
memory
whenever
it
wants
that's
bad
or
oh,
we
need
to
initialize
memory
at
startup.
E
You
know
we
don't
have
a
loader,
but
we
do
so
by
just
writing
to
a
bunch
of
memory
and
in
the
process
create
aliasing,
mutables
and
now
it's
unsound,
and
we
need
to
swap
that
entire
chunk
of
the
crate
to
assembly
or
whatever
so
keep
hitting
these
sort
of
tricky
memory
problems,
and
I
guess
especially
they're
tricky
because
they
require
so
much
domain
expertise
and
it's
hard
to
find
anyone
to
talk
to
even
to
work
out
whether
what
we're
doing
is
is
still
sound
and
then
the
other
half
important
we
have
is
that
there's
a
lot
of
different
types
of
embedded
users.
E
We
have
a
lot
of
people
coming
to
a
working
group.
Who
are,
I
guess,
hobbyists,
essentially,
who
don't
have
very
much
embedded
experience.
They're,
not
professionals
doing
this
for
their
job,
but
they're
people
who
really
want
to
use
trust
in
their
hobbyist
projects
and
need
something.
That's
relatively
quick
to
get
going
and
they
can
understand.
E
You
know
what's
going
on
and
we
also
have
users
coming
from
companies
who
are
saying.
Well,
I
want
to
use
rust
in
my
safety
critical
thing.
I
needed
to
be
certified
10
different
ways.
I
need
to
get
every
last
bit
of
performance
out
of
this
really
niche
architecture.
I'm
using
and
helping
both
of
those
people
use
share
the
same
traits
and
resources
and
stuff
has
been
more
challenging.
I
guess
trying
to
work
out
what
abstraction
levels
work
for
different
people
has
been.
E
It's
been
one
of
the
big
issues
yeah
and
I
guess
the
other
big
problem
we
have
is
just
there's
a
million
types
of
embedded
hardware.
In
some
sense,
it's
a
big
problem.
You
develop
some
support
for
one
ship
and
you're
like
well,
that's
great,
but
this
manufacturer
has
200
ships
in
this
family
alone
and
you've
supported
one
of
them
and
there's
also
50
manufacturers,
and
it
becomes
a
hopeless
endeavor,
so
really
finding
a
structure
that
lets
you
build
support
for
lots
and
lots
and
lots
of
different
hardware.
D
Yeah,
that's
super
interesting
to
hear.
I
never
really
thought
much
about
memory
layouts
on
hardware,
and
so
I
think,
that's
a
very
cool
kind
of
very
domain
specific
thing
to
hear
about
ricky.
I
want
to
open
up
to
you
if
you
like
to
maybe
chat
about
some
of
the
potential
issues
in
machine
learning,
yeah
absolutely.
B
I
would
say
so
far
we
found
that
you
know
overall,
rust
is
is
a
great
language
for
it.
We've
added
a
couple
things
to
the
compiler:
we've
done
the
be
able
to
build
and
enable
llvm
plugins,
which
is
really
great,
because
we're
able
to
then
kind
of
allow
enzyme
in
some
of
our
builds,
which
is
the
automatic
differentiation
tool
that
was
created
at
mit,
and
that's
been
helpful
to
some
of
the
stuff
that
we've
been
able
to
look
at
once.
B
Llvm,
like
writes
our
code,
we
can
do
a
little
bit
faster
things
with
that
and
on
the
machine
learning
level.
Really
it's
been.
It's
been
a
little
bit
more
about.
How
do
we
organize
the
group
into
people
who
know
machine
learning
and
people
who
know
rust
and
get
them
to
kind
of
talk
to
each
other
on
ways
that
they
both
understand
each
other
and
really
the
group's
been
mainly
about
that
so
far,
so
we
haven't
hit
too
many
rust
issues,
but
especially
compile
time
wise.
B
I
think
people
who
use
and
do
machine
learning
for
a
living
are
kind
of
used
to
training
models
so
compile
times
really
aren't
that
bad
for
them,
as
opposed
to
waiting
several
days
for
a
gpu
to
crank
on
a
problem
right,
at
least
so
far.
We
haven't
seen
anything
like
that
yet
and
we're
starting
to
get
our
footing
in
the
door
on
those
kind
of
things.
B
But
first
we
really
need
to
kind
of
lay
the
groundwork
for
people
to
be
able
to
use
it,
and
I'd
say
our
biggest
problem
so
far
is
just
kind
of
figuring
out
where
the
holes
are
for
people
to
actually
get
started,
and
that's
why
we're
writing
the
book.
First.
Is
that
you
know
we
want
to.
We
want
to
make
sure
that
there
are
some
examples
on
how
to
do
some
of
the
basic
things
that
people
like
to
do
and
and
some
of
the
toy
projects
that
are
the
hello
world
of
the
machine
learning
community.
D
Yeah
yeah
and
so
yeah,
it's
very
cool
to
hear
about
these.
These
separate
sides:
okay,.
B
This
is
something
I
think
actually
a
bunch
of
people
might
want,
and
that
would
be
a
first-class
way
to
ensure
code
is
panic,
free
and,
as
part
of
like
this
formal
methods
like
working
group
interest
group
like
we
saw
a
presentation
today
on
a
tool
that
can
do
this
called
prosky.
B
Look
at
it
on
like
the
unit
of
an
entire
crate
or
on
individual
functions,
but
you
know,
I
think,
ideally
I'd
personally
like
to
see
some
like
first
class
functionality
in
the
rust
compiler.
To
say
this
entire
crate
is
panic
free
when
compiled
is
a
release
build
or
with
these
certain
feature
sets
or
panic
free
like
that
kind
of
thing,
I
think,
would
be
very
helpful.
D
Yeah,
I'm
very
excited
to
have
heard
about
the
formal
interest
or
working
group
today,
just
because
yeah
there's
so
much
actually
nico.
If
you
want
to
add
something
here,.
A
I
heard
a
number
of
you
mentioned
like
well,
for
example,
in
the
machine
learning
interest
group
kind
of
collecting
projects
for
are
we
learning
yet
I'm
wondering
whether
you've
had
problems
or
challenges
trying
to
give
people
a
kind
of
coherent,
getting
started
experience
like?
A
Is
it
just
a
matter
of
cataloging
all
the
things,
but
then
that
presents
probably
a
lot
of
choices,
and
maybe
some
things
work
well
together
and
some
things
don't
have
you
thought
about
this
problem
of
like
having
someone
come
in
and
get
started
easily,
with
a
collection
of
crates
that
work
well
together,
or
something
like
that
and
how
to
manage
that.
B
Yeah,
I
we've
definitely
tried
and
the
problem
is
that
machine
learning
it
evolves
fast
too.
There's
always
a
new
paper
out
that
everyone
wants
to
try
and
that's
really.
The
hard
part
is
that
everyone
wants
to
build
that
new
paper
that
came
out.
So
everyone
wants
the
newest
and
latest
things.
B
So
it's
really
hard
as
a
group
who's
trying
to
just
set
up
the
basics,
be
on
that
forefront
as
well,
and
that's
at
least
what
we've
run
into
is
that
no
one's
really
excited
about
building
the
basics,
to
get
everything
working
everyone's
excited
to
come
in,
and
I
I
just
read
about
this
paper.
B
If
you
guys
heard
about
this,
we're
like
yeah,
it's
not
gonna,
really
work
for
us
right
now,
though,
and
that's
that's
when
they
kind
of
lose
their
interest
in
like
staying
with
the
group
and
helping
make
sure
that
make
that
possible.
So
that's
that's
kind
of
what
I've
seen.
Definitely
in
the
past
couple
months
that
we've
existed.
D
D
I
think
that's
a
really
great
way
to
drive
people
to
be
excited
about
these
types
of
ideas
or
to
yeah
just
just
offer
this
boon
of
things
that
you
couldn't
have
before,
and
so
this
is
what
drives
a
lot
of
people
to
try
out
a
lot
of
the
the
really
cool
engines,
but
then,
on
the
other
hand,
there's
a
lot
of
projects
out
there
already
that
sort
of
bring
people
into
a
very
small
section
of
of
game
development,
and
so
one
of
the
big
ones
that
that
is
is
currently
being
developed.
C
D
The
the
crate
that
it
uses
is
specifically
built
for
creating
roguelikes
in
the
terminal,
and
so
this
gives
people
like
an
avenue
to
sort
of
yeah
follow
to
to
be
able
to
complete
or
at
least
get
used
to
to
learning,
and
so
I
do
wonder
what
that
might
translate
over
to
when
it
comes
to
machine
learning
and
like
what
are
those
like.
You
guys
see
this
really
cool
thing
and
then
follow
like
a
few
steps,
whether
it
be
from
a
community
blog
where
someone's
like.
D
Oh,
this
is
what
I
was
able
to
create
create
with
this
crate
or
on
the
other
hand,
if
it's
the
interest
group
itself,
writing
these
tutorials
and
sort
of
trying
to
or
as
the
interest
group
you
know
what
can
be
interesting
and
what
can
be
fun
and
so
you're
encouraging
this
in
others.
And
so
I
do
wonder
what
it
could
look
like
for
machine
learning,
but
that's
more
of
a
comment
than
anything.
I
don't
necessarily
need
an
answer.
B
I'd
say
for
cryptography
like
it's
something
we're
trying
to
feel
out
and
solve
like.
I
know
it's
definitely
been
a
big
issue
like
going
to
meetups
and
stuff
I've
heard
people
be
like
what
do
I
use
for
us
cryptography
like
and
there's
not
really
a
good
resource,
so
we
put
together
this
like
cryptography,
rs,
dot,
rs
like
giant
list
of
crates,
I'm
not
sure.
That's
really
helpful
for
anybody,
especially
like
kind
of
an
app
developer
who
has
like
a
tenuous
grasp
of
to
cryptography
in
general,
like
which
crate
do.
B
I
pick,
which
I
use
like
that
kind
of
thing
I
think
ring
has
been
like
fairly
popular.
It's
like
probably
the
right
answer,
but
like
other
than
that,
like
you
know,
I
think
it's
generally
an
unsolved
problem,
although
one
we're
interested
in
solving,
I
would
be
curious,
like
how
other
working
groups
kind
of
put
together
this,
like
you,
know,
you're
new
to
us,
you're
working
in
this
field,
like
how
do
you
get
started
kind
of
guide
things.
D
I
adam
I'd
be
interested
in
hearing
sort
of
from
the
perspective
of
embedded,
because
I've
had
several
times
where
I've
been
like.
I
really
want
to
try
rust
for
embedded
and
I
go
to
the
website,
and
I
see
like
some
hardware
that
I
need
to
pick
up.
But
what
do
you
think
the
onboarding
looks
like
for
or
like
what
are
the
onboarding
goals
for
new
developers?
D
E
We
a
long
time
ago
wrote
this
book
called
the
discovery
book
or
actually
japaric.
Who
was
one
of
the
founders
of
a
working
group
back
in
2017,
maybe
wrote
it
and
it's
been
maintained
since,
where
we
basically
say
look,
there's
a
million
bits
of
hardware
you
could
have,
and
we
really
can't
help
you
with
all
of
it
as
a
beginner.
E
But
if
you
buy
this
one
board,
which
you
can
pick
up
for
ten
dollars
from
you
know
most
countries
there
are
websites
where
you
can
buy
it,
and
we
then
have
this
basically
a
long
tutorial
that
introduces
a
lot
of
embedded
concepts.
So
it
doesn't
assume
you
have
any
embedded
experience
and
introduces
all
the
rest
concepts
as
well
and
the
sort
of
the
goal
there
is.
E
We
have
now
is
that
it's
getting
kind
of
older,
but
obviously
in
the
years
gone
by
people,
have
invented
a
lot
of
clever
new
ways
of
doing
things
being
new
crates,
even
the
supporting
stuff
when
the
book
was
written,
the
only
way
to
load
your
code
was
using
this
more
legacy:
project
open
ocd,
which
is
tricky
to
use
and
also
to
config
files.
E
They
didn't
have
a
release
for
three
years,
so
people
would
be
using
pre-world
versions
of
the
software,
and
these
days
we
have
loads
of
really
lovely
solutions
and
tooling
and
rust,
but
none
of
that's
mentioned
in
this
discovery
book.
Yet
because
no
one's
gone
through
all
of
these
chapters
to
update
all
the
instructions
to
say
actually
use
these
new
tools
to
use
this
new
library.
So
on
the
one
hand,
it's
been
a
really
popular
and
boarding
thing,
and
I
think
a
lot
of
people
have
followed
it
through.
E
What
next
is
maybe
a
more
challenging
problem
still
for
us
to
solve,
and
I
guess,
as
with
the
other
groups,
the
everything
is
evolving
over
time.
People
are
still
coming
up
with
new
ideas.
You
know
we're
now
getting
more
and
more
async
stuff
and
embedded
rust,
which
is
exciting,
but
we
have
basically
no
tutorials
about
that
whatsoever.
E
I
suspect
the
solution
is
going
to
be
more
updates
to
these
books
right
now.
There's
an
effort
to
really
update
this
discovery
book
using
a
new
hardware,
which
is
now
more
available
and
easier
to
get
and
using
more
modern,
rust,
tooling
and
everything
else.
But
it's
a
big
effort
to
rewrite
this
big
tutorial,
and
then
it
probably
will
have
the
same
problem
again
in
a
few
years
time.
So
how
you
keep
on
top
of
it-
I
don't
know,
but
certainly
for
justin
boarding.
D
Yeah
yeah,
I
think
those
are
some
pretty
good
points.
I
want
to
open
up
the
floor
for
anybody
either
in
chat
to
ask
any
questions
or
if
you
want
to
come
forward,
you
can
feel
free
to
just
turn
on
your
camera
and
chat
a
little
bit
and
then,
if
we
or
yeah
I'll
just
give
a
moment
for
anybody
who
wants
to
potentially
hop
in.
D
And
if
there's
nothing
too
much,
I
will
just
add
on
a
point
that
I
was
sort
of
considering.
Well.
Adam
was
talking
about
the
you,
create
a
book
and
then
it
becomes
outdated
because
there's
new
technologies
and
new
ideas
within
the
like
rust
game,
dev,
not
that
I've
checked
out
all
of
the
the
different
books
that
are
out
right
now.
D
But
I
was
reviewing
one
by
the
author
of
the
the
rltk,
the
the
roguelike
toolkit
and
I'm
sort
of
interested
in
this
idea
that
a
lot
of
the
technology
that
was
used
at
that
point
might
not
be
as
actively
maintained
as
some
of
the
cooler
newer
engines
and
since
the
pace
of
everything
is
moving
very
quickly.
D
It
does
become
very
difficult
to
lock
down
at
a
certain
point,
or
rather,
if
you're,
locking
down
a
two-year-old
version
of
rust,
then
you
could
be
potentially
losing
a
lot
of
the
more
modern
paradigms
that
are
relative
to
or
relevant
to
the
particular
field.
And
so
it
is
interesting
that
in
this
game,
development
textbook
as
it's
describing
to
how
like
how
to
make
it
how
to
how
to
make
a
roguelike.
D
It
does
tell
you
to
go
into
like
your
your
cargo
tomo
file
and
add
dependencies
at
a
locked
version,
but
I
think
that
the
over
arching
principle
is
that
or
it
is
still
there
and
it
still
shines
through
to
be
able
to
get.
You
started
with
these
ideas,
and
so,
whereas
in
game,
development
is
cool
to
build
a
roguelike
or
it's
cool
to
build
a
3d
game
with
some
engine
and
rust
or
then.
D
D
These
means
is
only
a
way
for
like
you
to
get
your
foot
in
the
door
after
that
point,
there's
a
lot
more
that
you
can
do
to
branch
out
a
lot
further,
I
think,
and
so
when
it
comes
to
maybe
some
of
the
the
smaller
interest
groups,
cryptography
and
machine
learning
and
other
ones.
D
I
I
I
think
there
is
a
lot
of
potential
for
finding
these
cool
questions
that
need
answering
and
finding
what
does
russ
do
significantly
better
than
other,
like
attempts
at
solving
these
problems
and
so
yeah
when
it
comes
to
cryptography.
Maybe
at
the
end
of
the
day,
what
is
needed
is
just
crates
that
can
do
like
encrypt
stuff,
really
nice
and
quickly
and
efficiently,
or
on
the
other
hand.
D
Maybe
it
can
be
a
really
great
way
to
learn
some
of
these
more
primitive
concepts
about
what
is
happening
behind
cryptography
and
so
yeah.
I
think
definitely,
as
a
lot
of
these
groups
grow
and
expand
a
lot
more
of
these
cool
use.
Cases
will
come
in,
and
I
remember
several
years
ago
trying
to
figure
out
how
to
how
to
do
some
of
my
own
cryptography
and
rust,
and
just
some
simple
rsa
stuff
and
then
running
into
some
walls
and
so
yeah
tony.
I
see
you
turn
your
camera
on.
B
B
You
know
I
mostly
work
on
the
nuts
and
bolts,
but
at
the
same
time
I'm
constantly
talking
to
people
who
are
writing
like
crazy
next
generation
cryptography.
So
that's
like
a
sometimes
a
difficult
like
gap
to
span
in
terms
of
like
you
know,
even
though
I'm
writing
the
most
boring
of
the
boring
stuff
like
I,
by
interest.
My
personal
interest
is
always
in
these
like
new
cutting
edge
things.
B
I
know
like
like
one
of
the
questions
you
had
for
us
as
panelists
was
kind
of
like
what
are
the
boss
crates
and
right
now
I
would
I'd
actually
say
it's
like
a
complete
toss-up
right
at
this
point
like
there
are
like
so
many
crates,
so
I
would
say,
as
an
interest
group
right
now,
we're
not
like
blessing.
Anything
like
I
named
ring,
but
like
sodium
oxide,
is
a
good
choice
too.
B
I
personally
work
on
a
bunch
of
like
pure
rust
crates,
so
you
know,
I
think,
there's
a
lot
of
very
cool
stuff
happening
right
now
and
if
there's
any
gaps
you
see
in
the
rest
cryptography
this
system,
please
let
me
know,
I
can
probably
point
you
at
something.
D
Yeah-
and
I
I
think
this
is
a
great
point
to
to
close
it
off,
and
I
do
want
to
mention
that
one
of
the
biggest
parts
that
I'm
I've
taken
from
this
is
just
this
being
being
able
to
have
this
conversation
with
these
other
leaders
and
really
seeing
how
much
of
a
contrast
there
is
between
our
separate
interests
and
working
groups,
and
so
I'm
definitely
super
happy
to
keep
this
type
of
communication
going.
D
And
I
think
we
were
also
discussing
earlier
that
it
might
be
cool
to
have
a
collaboration
in
terms
of
a
newsletter
that
really
focuses
on
all
of
the
interest
groups,
sort
of
amalgamated
together
or
sorry,
not
not
them
putting
together,
but
a
newsletter
that
focuses
on
all
the
different
sections.
That
can
give
you
an
overview
of
these
interest
group
ecosystems
within
rust,
so
yeah.
Thank
you
so
much
to
all
of
the
panelists
for
taking
the
time.
I've
really
appreciated
all
of
the
insights
and
I'll
hand
it
back
over
to
you
nico
great.
A
Thanks
thanks,
boris
thanks
everybody
thanks
adam
tony
ricky,
ernest
who's
dropped
off,
so
that's
kind
of
the
end
of
our
ctcft.
Today,
the
next
meeting
is
going
to
be
october
18th
and
it's
at
the
other
time.
A
You
can
click
here
if
you
like
17
o'clock
utc,
and
for
now
I'm
going
to
hand
off
to
jane
to
the
social
hour
where
we're
going
to
make
little
breakout
rooms,
I'm
not
sure
jane.
If
you
have
a
set
you
want
to
make,
but
I'll
make
you
host
and
we'll
pick
up
from
there.
I
guess
we
should
take
a
short
break.
Five
minutes.