►
From YouTube: QIRG Interim Meeting, 2020-04-08
Description
QIRG Interim Meeting, 2020-04-08
A
B
B
Right,
Tech,
why
don't
you
and
I
introduce
ourselves
just
so
people
make
sure
they're
that
where
they
who
is
who
I'm
a
university
and
I'm
one
of
the
co-chairs
I'm
going
to
mostly
keep
my
camera
off
to
save
on
bandwidth
and
whatnot.
But
that's
me,
say:
hi
right,
Android
Tech
is
our
new
co-chair
and
he's
actually
going
to
be
doing
most
of
the
running
of
the
meeting
boy,
tech.
A
If
you
want
to
ask
a
question
after
a
talk
plus
Q
in
the
chat
and
minus
key,
if
you
want
to
remove
yourself
and
wear
one,
the
chair,
that's
currently
running
anything
should
be
monitoring
a
chat
to
do
that.
The
jabber
is
on
and
I
have
one
request
to
everybody.
Somebody
willing
to
take
their
minutes
for
this
meeting.
A
G
A
A
A
A
A
Me
and
rod
we'll
reevaluate
that
over
the
coming
few
weeks
and
might
consult
the
mailing
list
as
to
how
best
to
actually
meet
as
a
research
group
and
so
far
we've
been
kind
of
doing
the
IETF
meetings.
Quite
well,
this
all
those
four
enough
it
for
me
rod
do
you
have
anything
to
add
before
we
move
on
to
presentations.
B
Only
the
administrative
bit
of
there
are
now
56
people
in
the
the
WebEx
conference,
and
only
28
people
have
actually
signed
in
on
the
virtual
blue
sheet.
For
those
of
you
who
came
in
a
couple
of
minutes
late
on
white
text
via
slide
to
the
chairs
very
first
slide,
there's
a
link
to
the
ether
pad.
Please
sign
in
there
and
put
put
your
name
in
the
in
the
blue
sheet
so
that
we
have
a
list
of
people
who
are
attendees.
That's
it.
A
Yeah
so
the
links
now
in
the
WebEx
chat
and
apparently,
if
you're
kind
of
restrictive
firewall,
you
can
remove
that
909
port
number.
Okay,
so
first
speaker
up
is
Axl.
I
was
going
to
start
with
the
first
simulator
that
order
of
the
simulators
is
based
on
release
date
FYI.
So
we
start
with
the
ones
that
were
released
earliest.
It
will
progress
to
the
ones
that
will
release
or
will
be
released
latest
Excel.
Are
you
ready
to
share
your
screen
and
present.
A
H
H
So
you
can
of
course
find
in
there
in
the
documentation
and,
for
example,
other
talks.
I
gave
a
talk
at
first
time
this
year,
which
you
can
also
still
look
at
where
I
gave
a
demo
of
how
to
use
it,
but
maybe
to
get
going
so
to
kind
of
set
the
stage
a
little
bit
of
what
simulacrum
is.
I
would
like
to
kind
of
zoom
out
a
little
bit
and
explain
why
we
why
we
develop
simulacra
know
where
we're
coming
from
from
this.
H
H
Essentially
what's
the
interface
between
the
application
and
Q
no
choice,
so
we
had
a
proposal
for
such
an
interface.
We
call
it
CQC
it
for
closed
classical
quantum
combiner,
which
we
developed
kind
of
in
in
parallel
with
simulacrum
during
the
over
the
years.
We
kind
of
realized
a
lot
of
things
and
learned
a
lot
and
realize
that
maybe
how
we
initially
designed
this
interface
is
maybe
not
ideal
for
what
we
actually
want
to
do.
H
E
E
E
A
H
H
H
Okay,
let
me
try
to
I,
can
show
you
briefly
the
slides
that
I
intended
to
show
here's
kind
of
upscaling
to
a
quantum
network.
We
want
to
bridge
the
gap
gap
between
applications
and
hardware,
and
we
have
this
need
for
an
interface
between
applications
and
QoS,
and
let
me
maybe
double
check
if
this
is
working
now.
H
All
right
thanks
so
yeah
as
I
said,
we
realized
that
maybe
CQC
is
not
exactly
what
we
want.
So
we
worked
on
a
new
version
of
this
and
took
some
inspiration
from
Qasim,
which
is
a
quantum
assembly
language
or
their
multiple
flavors
of
it's
mostly
used
for
quantum
computers,
and
we
can
of
this
signed
or
started
designing
something
using
ideas
from
this
for
for-4
networks
and
now
there's
some
fancy
animations,
where
this
new
interface,
which
we
call
net
chasm,
comes
in
and
kicks
out
CQC.
H
So
net
chasm
is
a
essentially
a
quantum
assembly
language
for
for
networks
and
again
so
the
idea
is
that
net
custom
essentially
the
way
how
you
would
interface
if
you
write
your
own
application
to
the
actual
network,
that
we're
building
and
and
also
later
expanding
in
the
future
and
net
chasm,
is
something
that
we're
working
on
at
the
moment
and
I
won't
have
time
to
go
into
the
details
of
this.
But
it's
essentially
a
lower
level
instructions
that
architecture
for
quantum
maintenance
applications-
and
this
is
something
that
well
hopefully
in
the
near
future.
H
We'll
can
say
more
about
this
and
kind
of
announce
exactly
what
it
is,
but
this
is
something
that
we're
working
on
at
the
moment,
and
so
this
is
kind
of
the
the
end
of
the
introduction
of
where
simulacrum
fits
in.
So
when
we
had
the
CQC
interface
and
we'd
talk
about
how
we,
how
an
application
connects
to
the
quantum
hardware,
it's
kind
of
at
the
moment,
we
don't
have
such
awesome
hardware,
but
you
might
already
now
want
to
write
an
application.
H
So
this
is
exactly
where
simulacrum
comes
in,
where
we
had
essentially
a
library,
for
example
in
Python,
where
you
can
easily
write
an
application.
That's
communicate
with
similar
Ekron
over
this
CGC
interface,
but,
as
I
mentioned,
we're
now
kind
of
changing
this
to
this
new
net
custom
interface
and
in
CQC
we
also
had
multiple
libraries
in
different
languages.
For
example,
we
had
in
rust,
which
Wojtek
made
we
had
in
C
Java.
H
There
was
somewhat
also
working
go,
which
is
not
fully
finished,
but
you
could
write
your
application
in
any
of
these
languages
and
it
would
communicate
with
with
simulacrum,
but
not
the
only
simulacrum.
We
also
want
to
be
able
to
communicate
with
other
simulators.
For
example,
we
have
a
kind
of
a
beta
version
of
this
with
net
squid,
which
will
present
a
bit
later
today,
but
of
course
also
in
the
future
that
you
will
be
able
to
use
these
applications.
H
H
You
just
pay
per
install
simulacrum
and
then
you
can
start
it
where
you
have
essentially
a
daemon
running
on
your
computer,
which
you
can
communicate
to
using
cqc
that
simulates,
essentially
quantum
hardware,
and
if
you
want
to
know
more,
you
can
go
to
our
website.
We
can
find
the
documentation
and
try
it
out
for
yourself.
A
J
J
Is
it
simulates
the
network
and
application
level
quantum
network?
So
therefore,
it's
also
looking
into
the
routing
algorithms
and
stuff
like
this.
It
simulates
some
a
multi
node
communication
network
or
there
even
regional
processes
classical
and
quantum
quantum
information.
So
this
means
that
the
host
can
transmit
classical
messages.
J
So
they
have
some
memory
which
can
be
addressed
and
this
kind
of
allows
them
to
hold
on
to
information
if
they
need
it,
and
then
protocols
can
be
designed
such
that
one
knows
that
the
information
that
they
previously
sent
is
stored
at
the
other
side
yeah.
So
the
host
can
act
as
two
types
of
hosts
where
one
can
be
an
end,
node
or
relaying
node,
and
that
node
is
one
that
runs
an
application.
J
J
Run
a
custom
protocol,
so
you
can
write
a
Python
function
and
then
assign
it
to
a
host,
and
then
the
host
runs
this
function
asynchronously
or
can
also
run
it
synchronously,
and
it
is
kind
of
a
small
small
picture
of
what
the
host
looks
like.
In
my
mind.
It
has
a
month
two
processors,
one
for
a
classical
one
for
quantum,
and
it
has
two
storage
devices,
one
for
classical
morrowind.
J
So
some
of
the
built-in
protocols
that
come
with
these
hosts
are
basically
content.
Teleportation
super
dense
coding,
EPR
generation,
PhD
generation
you
just
reach
what
this
means
is
that
we
have
a
host
object
in
queue,
netsim
and
one
and
the
methods
of
the
host
are,
for
example,
I,
can
create
a
qubit
and
then
I
can
say
host
teleport
to
qubit
to
receiver,
and
these
kind
of
tasks
are
all
built
into
the
host
object
so
that
we
can
recycle
them
and
build
more
complex
protocols.
J
J
Sport
layer,
so
the
transport
layer
is
responsible
for
taking
the
application
data
packet
izing
it
and
then
puts
it
into
the
network.
But
it
also
is
responsible
for
making
sure
that
if
the
protocol
requires
an
EPR
pair
that
this
PPR
pair
is
generated
so
that
if
you
want
to
set
it
teleportation,
but
there's
no
EPR
peridot
generate
the
CPR
pair
ahead
of
time
or
one
can
get
warned
that
there's
no,
if
you're
a
parent
and
it's
not
possible
to
run
such
a
protocol.
F
J
J
The
network
is
responsible
for
connecting
host
through
multi
no
drugs,
so
one
can
construct
a
complex
network
and
I
can
take
a
host
and
say
send
a
teleportation
to
know
that
is
maybe
not
directly
connected,
but
the
network
will
be
responsible
for
moving
this
packet
in
the
network,
just
like
any
other
network
in
the
classical
Internet,
and
it
can
be
programmed
so
that
one
can
design
routing
algorithms
on
top
of
it.
So
maybe
there's
an
approach
to
routing
for
quantum
networks
that
requires
something
novel,
not
just
shortest
path.
J
For
example,
one
can
test
these
things
in
this
network,
and
then
we
can
cut
the
parameters
to
the
network
so
that
they
have
noise.
They
can
apply
heirs
to
the
qubits
that
are
being
transmitted
and
they
can
randomly
drop
packets.
Also,
when
we
want
to
establish
entanglement
between
distant
nodes,
we
should
have
a
we
have
a
in
place
of
multi
hub,
entanglement,
swapping
procedure
that
happens.
So
we
have
this
kind
of
picture
where
alice
is
sending
her.
J
J
So
the
pros
and
cons
of
using
simulacrum
is
that
sorry,
miss
emission
unit
sim
is
that
it's
very
high
level,
the
programming,
the
way
we've
kind
of
structured
the
code
is
it's
build
upon
classical
internet
stack
and
all
of
the
functionality
is
very
kind
of
straightforward
to
use.
I
think
we
can
program
many
Nets
in
areas
that
we
can
quickly
change
the
network
setting
we
can
adding
drop
connections.
J
We
can,
you
know,
put
noise
in
the
network,
you
could
take
noise
out,
we
can
have
memory,
we
can
have
all
kinds
of
configurations
for
this
network
and
we
can
use
it
to
test
routing
algorithms.
That's
an
interesting
feature,
I
think
and
there's
lots
of
vlogging
messages
so
that
if
I'm
wanting
to
write
a
protocol
and
I'm
not
working
I
can
see
step
by
step
what's
happening
in
the
network,
and
since
we
had
this
layer
of
structure,
we
can
kind
of
see
what's
happening,
so
we
can
more
easily
debug.
J
J
When
we
are
increasing
the
scale
of
the
network,
things
tend
to
get
a
bit
slow,
and
it
also
is
assuming
that
the
quantum
Internet
is
based
on
a
packet
style
where
there's
this
transport
layer
in
between
and
things
like
this
or
maybe.
This
is
not
something
that
will
happen
right
away,
so
the
people
I
think
should
use
cue.
Netsim
are
beginners,
who
are
just
learning
about
quantum
networking,
because
this
high
level
simulation
framework
structure
and
stuff
like
this.
J
It's
easy
to
kind
of
go
from
classical
setting
to
a
quantum
setting,
and
therefore
instructors
can
also
use
this
to
teach,
because
it's
very
analogous
to
some.
So
the
classical
networks
and
researchers
can
use
it
in
a
way
that
they
will
test
their
develop
applications
for
roboticist
robustness
and
correctness
as
a
first
stage.
So
maybe
they
can't
benchmark
if
they're
part
of
works
under
all
kinds
of
scenarios
with
some
kind
of
rates,
but
they
can
test
that
at
least
it
works
in
some
kind
of
basic
scenarios
as
a
first
stage.
J
So
just
if
I
have
time
show
a
quick
example
of
how
that
works,
we
would
define
some
protocol
and
this
particular
protocol.
The
sender
is
just
sending
five
EPR
pairs
to
review.
So
I
write
this
functional
protocol,
one
takes
the
host
and
it
takes
a
receiver.
Then
I
run
a
loop
five
times.
I
send
an
EPR
pair
on
line
12.
What
happens
is
the
because
I
set
this
flag
here
to
a
white
act?
J
True,
this
line
of
code
will
block
until
an
ACK
is
received
by
the
by
the
person
sending
it,
and
then
we
can
perform
operations
depending
if
the
ACK
arrived
or
not.
Then
we
have
our
key
bits
to
have
an
ID
and
we
can
kind
of
perform
operations
on
this
qubit
by
fetching
it
out
of
the
memory
or
on
line
17.
We
see
that
the
host
is
fetching
the
qubit
by.
D
J
See
that
the
this
function
makes
the
sender
ID
takes
basically
waits
for
five.
If
you
are
pairs
to
come
now,
we
see
for
this
loop
here
and
we
set
this
flag
here,
wait
for
five,
and
this
means
that
the
host
is
sitting
idle
waiting
for
the
corpse
to
prod
and
then
just
performs
a
simple
measurement
to
determine
that
the
car
was
legitimate,
a
pure
parallel
and
then
what
we
do
is
we
construct
a
network.
J
We
have
this
network
singleton
object,
we
define
the
nodes
in
the
network,
we
define
these
the
node
IDs
and
then
we
have
these
host
objects.
These
can
be,
we
can
add,
connections
to
the
hosts,
and
then
we
put
all
the
hosts
in
the
network,
and
then
you
see
these
two
lines
at
the
bottom
or
you
can
run
our
two
protocols
that
we
defined
above
I
think
I'll
skip
the
next
example,
but
it's
basically
I
would.
Can
we
just
wrap
up
now?
Steven
I
think
this.
For
last
night,
yeah.
A
B
We
have
just,
as
you
have
seen
on
the
mailing
list,
announced
that
Quisp
is
available
open
source,
there's
a
link
to
that
there
Quisp
is
the
quantum
simulation
package.
Here's
a
partial
screenshot
of
one
of
the
larger
networks
that
we're
in
the
process
of
simulating.
If
any
of
you
have
worked
with
on
net
plus
plus
you'll
recognize
the
style
of
icons
and
whatnot.
B
So
we're
designing
Quisp
to
answer
a
series
of
research
questions
and
the
first
one
is
we're
looking
for
emergent
behavior.
What's
the
the
equivalent
of
quantum
question,
the
quantum
equivalent
of
congestion
collapse,
for
example,
we're
also
working
on
protocol
design.
The
basic
architecture
we
have
laid
out
is
what
we
call
the
rule
sets
that
are
based
on
a
condition
clause
and
an
action
clause
like
Sdn
match
action
seems
to
be
sort
of
the
more
common
terminology.
B
So
far,
that's
pretty
similar
to
what
the
other
simulators
are
working
on
a
belief,
but
one
of
the
things
that
we're
working
on
this
rather
different
is
that
we're
studying
not
just
first
generation
purify
and
swap
networks,
but
also
second
and
third
generation
networks
and
hopefully
studying
logical
and
physical
heterogeneity,
including
also
dynamic
behavior
such
as
link,
state
changes
and
traffic
pattern
changes
in
whatnot
so
very
quickly.
There
are
three
generations
of
repeaters
that
have
been
defined.
B
The
first
one
which
I've
already
alluded
to
is
uses
quantum
purification,
which
is
essentially
error,
detection
and
uses
entanglement
swapping
to
build
the
end.
End
connections.
Second
generation
uses
quantum
error
correction
instead,
which
means
that
the
states
that
are
used
are
actually
larger
and
third
generation
is
similar
to
second
generation.
The
difference
between
second
and
third
is
that
in
the
third
generation
done,
reception
is
very
high
probability
and
therefore
you
can
go
with
a
more
of
a
store
and
forward
kind
of
approach,
rather
than
the
the
end
to
end
entanglement
swapping.
B
That
has
to
be
done,
and
these
were
defined
by
the
anjaani's
group
and
there's
a
link
there.
If
you're
interested
so
I
listed
here,
I'm
not
going
to
try
to
go
through
this
entire
slide.
But
I
thought
it
would
be
useful
for
people
in
the
group
to
who
are
not
as
familiar
with
the
quantum
computing
world
to
understand
a
little
bit.
What
about
how
the
different
possible
ways
of
removing
the
unwanted
students
are.
A
B
Right
thanks
a
lot
of
the
most
simpler.
The
the
first
thing
you
learn
when
you're
studying
quantum
computing
is
the
basic
state
vector
approach,
usually
using
what's
called
directs
getting
a
direct
ket
notation,
you
can
see
the
scaling
of
it
there
we're
actually
working
in,
what's
called
the
the
error
basis.
The
last
line
there
at
the
end
of
the
the
slide
in
which
we're
only
tracking
internally,
whether
or
not
the
state
corresponds
to
what
we're
actually
trying
to
build
so
we're
tracking.
We
have,
in
the
simplest
form
your
bitwise
flags.
B
B
But
if
you
want
to
even
just
take
two
of
those
and
combine
them
all
of
a
sudden,
you
are
all
the
way
up
into
a
petabyte
scale
sizes
for
memory,
so
it's
simply
not
feasible
to
simulate
a
large-scale
quantum
state.
That's
entangled
across
multiple
nodes,
but
you're
involved
in
quantum
error,
correction
and
whatnot.
So
all
of
this
led
us
to
working
in
the
error
basis.
Instead,
so
strengths
of
what
we've
got
scalability
is
the
ultimate
target.
B
Our
goal
is
to
simulate
a
quantum
Internet,
not
just
a
network,
but
a
network
of
networks
consisting
a
hundred
No
hundred
networks
of
a
hundred
nodes,
each
and
probably
one
hundred
qubits
per
node,
so
on
the
order
of
a
million
cubits,
and
all
this
is
made
possible
by
working
in
the
error
basis
as
a
company
as
axles
already
live
into
in
his
system
as
well.
One
of
the
things
we're
very
interested
in
is
the
development
of
software
for
the
the
repeaters
and
the
routers
themselves,
and
so
you'll
see
a
diagram
there.
B
That
shows
our
internal
structure
for
how
the
the
modules
fit
together.
There
is
absolutely
endless
configurability
in
this
building
off
of
commnets
capabilities
and
it
provides
animations
and
an
inspector
for
the
state
and
lots
and
lots
of
logging
and
tremendous
capabilities
all
built
into
an
ID
if
you've
used
that
it's
actually
incredibly
powerful.
B
Simulator
I,
just
tossed
in
a
very
brief
summary
of
what
I
think
the
basic
goal
of
the
six
different
simulators
that
I
know
about
here-
and
this
is
not
intended
to
be
sort
of
definitive
but
hopefully
to
give
people
sort
of
a
basis
for
discussion
later,
which
might
actually
be
a
topic
we
might
want
to
take
up
on.
The
mailing
list
would
be
surveyed
comparison.
B
We
might
actually
be
interested
in
building
a
document
for
doing
all
of
this,
and
there
are
the
links
come
join
us
clone
away
in
the
the
software's
on
github
join
us
on
the
slack
if
you're
interested
and
the
best
resource
for
learning
about
it
is
Takaaki,
matzos,
master's
thesis
and
there's
a
link
to
that
there
Wojtek.
That's
it
thanks.
A
B
A
F
Yeah
I'm,
Rob
and
I'll
be
presenting
the
netcode
simulator
on
the
house
to
take
yeah.
So
that's
good
is
an
acronym
for
network
simulator
for
quantum
information
using
discrete
events
and
it's
a
project.
That's
been
under
development
at
GTX
2017
and
the
collaboration
between
Latino
and
IDO,
which
is
the
partner
to
Nikita,
and
it's
been
actively
used
with
amputate
Marie
for
that
time
by
the
groups
of
the
safety
ladder
and
edifice
for
the
theory
groups,
and
also
already
a
bit
by
some
of
the
experimental
routes
in
the
Kings
roadmap.
F
So
that's
yeah
past
three
years
has
been
essentially
private
use
but
yeah.
Essentially,
this
months
were
actually
last
week
we
were
planning
a
public
beta
release,
it's
being
delayed,
unfortunately,
by
hopefully
only
a
week,
because
we're
still
trying
to
finalize
the
user
license
so
very
soon.
We're
hoping
to
provide
this
publicly
so
a
beta
anyway.
So
watch
this
website
from
or
information
to
sort
of
explain
what
net
squid
is
and
how
it
differs.
I
think
Rodney
already
explained
that
from
that
his
last
slide.
F
So
if
we
consider
the
Contin
Internet,
it's
essentially
an
infrastructure
for
sharing
quantum
information,
and
it's
made
out
of
all
this
physical
hardware
and
quantum
channels,
fibers
of
free
space
and
quantum
memories,
and
this
is
one
way
to
move
this
quantum
information
is
really
the
key
resources
in
the
continent.
Ernest,
you
could
say,
quantum
entanglement,
which
is
acting
as
a
virtual
quantum
channel
between
these
nodes,
and
this
resource
has
two
properties
it
as
a
sort
of
the
race
negative
and
the
fidelity
or
its
quality
in
any
realistic
Network.
F
Both
these
properties
are
affected
by
non-ideal,
high
velocity
or
noisy
quantum
Channel
and
quantum
operations
at
the
nodes,
and
so
one
way
to
mitigate
this
when
designing
or
building
a
quantum
Internet,
especially
is
Rodney
mentioned
in
the
first
generation
of
a
quantum
Internet
is
to
use
methods
such
as
Tanglin
purification.
We
have
noise
or
quantum
repeaters
to
be
lost.
The
very
briefly
in
Feynman
purification
means
that
make
multiple
copies
of
this
entanglement
between
two
nodes.
F
You
can
use
the
best
protocols,
such
as
distillation,
to
increase
the
fidelity
of
this
of
these
low
fidelity
links
up
it,
and
you
can
miss
this
process
to
get
a
higher
fidelity
and
if
we
deal
with
noise
that
way
and
to
deal
with
the
loss
exponential
attenuation
across
your
constant
channels,
you
can
start
inserting
nodes
to
shorten
these
links
between
your
notes,
so
called
quantum
repeaters
and
these
cancers
to
establish
entanglement,
and
then
you
can
swap
that
entanglement
at
the
repeater
node.
Now
the
key
elements
in
these
protocol
for
mitigating
the
effect
is
timing.
F
So
these
purification
protocols
are
generally
stochastic
in
nature
and,
for
example,
if
you're
trying
to
perform
this
repeating
it
may
be
that
one
arm
will
succeed
sooner
than
the
other
and
we'll
have
to
wait
for
the
other
one,
and
in
that
waiting
time,
these
qubits
are
waiting
on
memories
and
they
don't
start
speaker
here.
So
this
high
fidelity,
you
thought
you
had
from
your
purification,
starts
to
decrease
and
by
the
time
you
swap
the
fidelity
may
be
not
optimal
and
you
want
to
design
a
better
protocol.
F
So
the
key
takeaway
here
is
to
design
a
quantum
Internet.
You
need
to
solve
complex
timing
dependencies,
and
that
was
the
inspiration
for
developing
next
words.
So
that's
good!
In
a
nutshell!
Oh
it's
just
packaged
here.
It's
a
icing
package
available
is
a
Python
packaged,
but
under
the
hood,
it's
using
C,
C++
and
sizing
code
where
needed
to
optimize
it.
If
you
look
at
this
stack
of
sub
Texas
I'm
just
going
to
check
the
other
window,
then
here
log
beeps,
there's
nothing.
F
If
you
consider
a
speck
of
sub-packages,
then
it's
key
packages
built
on
is
this
discrete
event
simulation.
So,
if
screw
time,
discretely
what's
the
dense
and
then
it
will
update
all
the
quantum
states
depending
on
how
much
time
has
passed,
and
that
way
we
accurately
track
the
time
rather
than
running
in
real
time
or
in
a
distributed
session
such
as
some
of
the
other
simulators
I
should
add.
F
This
is
the
same
as
on
this
plus
plus
we
have
a
specialized
quantum
computation
library
for
the
quantum
network,
so
just
like
we
say
cubed
centric,
so
sort
of
hides
the
details
of
the
quantum
state
to
the
user.
So
you
can
really
just
focus
on
operating
on
the
qubits
locally
at
nodes
and
optimized
for
repeated
sampling,
a
lot
of
the
mystical
simulators.
We
do
involve
sampling
to
learn
something
about
the
performance
of
your
network
or
your
protocol,
and
we
also
offer
different.
F
Formulism
is
Rodney
and
axel
also
mentioned
the
ket
states
and
foreigners
in
Dubinsky
majors
formalism.
Also,
the
stabilizer
formalism,
which
does
scale
well
for
a
large
number
of
qubits,
with
the
trade-off
that
it
doesn't
have
a
universal
gauge
set,
has
Clifford
gate
for
a
lot
of
modern
network
applications.
Those
are
good
enough
and
then
finally,
another
key
feature
of
nit
squib
is
its
modularity.
So
we
offer
a
library
of
base
class
components
that
compose
with
different
physical
models
and
then
be
composed
of
each
other.
F
F
So
how
does
classical
communication
affect
the
performance
by
introducing
delayed
all
the
way
up
to
use
their
applications,
which
actual
showed
there's
an
effort
to
integrate
MIT's
grid
with
this
net
Kevin
language?
We
all
spoke
all
the
way
up
to
the
application
level
and
it
could
act
as
a
simulator
and
also
for
future
hardware
setups.
F
Also
in
this
in
this
thing,
so
there's
a
few
examples
where
it's
been
using
these
for
ease
things
already:
studied
performance
of
a
quantum
link
layer
protocol
and
also
in
a
European
project
which
designing
a
blueprint
for
a
European
continent
with
heterogeneous
Hardware
aerotrain
use
the
parameter,
optimization
and
benchmarking
of
these
different
fibers
and
persons.
The
V
study
of
repetitions
were
check
how
many
minutes
daily
at.
F
Almost
done
so,
how
do
you
get
started
with
next
good?
So
the
first
thing
you
can
do
or
should
do
is
register
at
the
forum
that
will
give
you
a
username
and
a
password,
and
you
can
use
that
both
to
participate
on
the
forum,
but
also
to
access
the
online
documentation
and
very
soon,
once
the
license
is
ready
to
install
network
using
the
standard
five
integrity
installer.
So
here
you
see
how
you
can
do
that.
It's
a
matter
of
just
adding
the
extra
bison
package
index
for
net
squid
with
your
farm
credentials.
F
F
But
if
you
look
at
the
documentation,
there's
a
lot
more
detail
if
you're
interested
and
then
finally,
the
net
squid
is
really
intended
as
a
base
package
for
people
develop
their
simulations
on
and
what
we
hope
to
stimulate
is
for
users
to
contribute
to
nets
good
with
what
we
call
net
squid
snippets,
the
Python
packages
that
are
created
and
maintained
and
shared
by
users,
and
here
we
really
encourage
that
people
share
this
open
source
and
we
already
have
a
number
of
these
packages
available
on
out
on
a
package
server.
So
this
also
does
dependency
management.
F
If
we
put
we
host
these
snippets
on
the
server,
you
can
say,
I
want
to
install
this
snippet
and
it
should
pull
in
any
other
snippets.
A
snippet
depends
on
examples
of
snippets.
We
already
has
available
modeling
snippets,
such
as
general
components
for
physical
layer,
specific
modeling
for
indecency
devices
and
atomic
ensembles
to
things
like
quantum
memory
manager,
constant
program
manager
for
bonus
higher
up
in
the
stress
and
to
get
started
with
the
snippets.
There's
a
template
repository
that
will
generate
a
snippet
template
for
you
and
there's
more
information
available
at
the
website
listed.
L
Great
ok,
I'm
presenting
right
so
full
screen.
My
page,
ok,
hi.
Everyone
just
share
with
you.
Some
of
this
applications
for
information
network
and
some
of
the
promising
use
cases
there.
You
know
I'm
from
additional.
You
know,
I
think
that
profile
implications
that
we
might
have
this.
My
hair
phone
there's
intern
applications.
Ok,
so
you
know
one
of
the
things
that
struck
me
was
the
similarity
between
you
know:
lasers
and
quantum
information
and
work
right.
The
inventors
of
the
laser.
L
It
probably
didn't,
anticipate
its
many
use
right-
and
this
is
article
from
the
Smithsonian
magazine
right,
but
today,
lasers,
I,
used
to
read,
see
these
barcodes.
You
know
missiles
and
everything.
So
what
is
really
exciting
for
me
personally,
is
you
know
this
can
be
this
this?
We
can
see
parallels
being
drawn
right
to
what
we
are
discussing
in
this
this
forum,
and
you
know
what
we
are
starting
today:
the
use
cases
the
applications
might
look
very
different
from,
but
we
will
see
next
time
right.
You
know
under
once
right.
Ok,
so
just
slide.
L
Ok,
this
is
just
in
fact
for
myself.
So
some
of
this
promising
use
cases
then
I
have
just
all
nine
in
maybe
a
high
in
the
sky,
but
I
believe
that
scale-up
quantum
computers
will
be
one
of
the
realities.
You
know
right
now,
many
people
we
just
talked
about.
We
are
looking
at
scale-up,
you
know
quantum
computers
and
you
see
the
whole
internet
revolution
and
how
computing
has
reach
where
we
are
right
now
in
the
classical
sense
right.
L
You
know
we
have
distributed
clusters
of
servers
and
I
believe
that
you
know,
but
now,
with
quantum
computers
we
might
be.
You
know
going
back
to
the
big
mainframe
systems
by
in
the
future.
We
might
see
still
up
quantum
computers,
you
know
leveraging
you
know
fundamentally
stupid
algorithms
I
mean
quantum
algorithms
are
being
worked
on
and
and
I
believe
that
you
know
distributed.
Algorithms,
you
know
would
be
distributed.
Quantum
algorithms
would
be
something
that
would
be
a
reality
sometime
right
and
you
know
to
perform
many
many
truss.
L
Perhaps
in
that
even
move
around
and
one
of
the
things
that
you
know
again
my
seem
like
pie
in
the
sky
right
now,
similars
similar
to
service
computing.
You
know
what
we
see
today:
micro
services
and
all
perhaps
one
day
we
will
see
you
know
a
data
center
less.
You
know
Everman
right
where
you
know
you
have
multiple
quantum
computers
in
the
world,
I
know
driving
Auto,
Bob,
Lutz
right
again,
pie
in
the
sky,
but
I
believe
that
this
is
something
that
that
might
happen
right.
L
It
will
drag
the
cost
of
classical
computing
to
negligible
amount
due
to
pumping
power
and
of
course,
you
know,
we
need
to
hook
up
this
quantum
computers
together
right
and
finally,
perhaps
zero
latency
networks
right.
This
is
something
that
we
often
take
for
granted
but,
for
example,
between
the
clocks
of
us
that
I,
currently,
you
know
I
see
some
latency.
You
know
I'm
25
millisecond,
you
know
from
the
east
coast
to
the
west
coast
of
the
US
right.
L
But
again,
perhaps
you
know
in
the
near
future
we
could
sees
you
really
didn't
see,
networks
right
being
anybody,
alright,
so
I
want
to
talk
about
the
profile
implications
on
existing
Internet
applications,
quantum
sensing-
something
that
really
excites
me.
You
know
when
when
when,
when,
when
all
of
these
things
that
we
discussed,
they
become
a
reality,
I
believe
that
many
of
these
applications
that
we
use
today,
they
will
be
fused
with,
with
this
led
right
to
sense.
L
You
know,
and
to
react
accordingly
right
and
applications
that
leverage
quantum
competition
is
net
would
be
able
to
respond
so
quickly.
You
know
there
are
no
classical
computers
unable
to
do
so,
and
what
is
button
is,
as
we
see
more
internet
applications
right,
moving
towards
event-driven
architecture,
right,
aw
and
I'm
dying
or
the
surplus
functions,
but
UVA
right.
You
know
I,
believe
that
this
would
lay
the
foundation
for
the
quantum
future
that
we
might
see.
You
know
quantum
sensing
in
turn
applications.
L
So
this
is
one
of
the
implications
that
we
have
right,
so
some
popping
thoughts
that
I
have
right.
We
we
cannot
solve
this
problem
through
the
same
thinking
that
we
use
right,
so
I
thought
that
this
is
really
a
new
frontier,
where
we
start
a
new
canvas
right
without
the
constraints
and
limitations
of
classical
computers
and
all
the
networking
standing
OSI
reference
model
back
and,
of
course,
you
know
in
the
process
of
trying
to
come
up
with
the
future.
L
You
know
we
do
provide
some
provides
improvise
and
and
and
then
from
now
to
the
future,
that
we
see
I
believe
that
outcome
and
all
these
use
cases
that
we
discuss
right
would
be
very
different
than
some
of
this
will
even
the
unexpected
so
yeah.
You
know
in
the
industry
that
I'm
working
in
the
financial
services
sector
right
quantum
computing
is
applications.
We
believe
we
have
performing
effect
on
some
fundamentals.
You
know
underlying
data
usage,
protection
right
and
Maurice
modeling
right
and
specifically
with
quantum
sensing.
You
know
we
could
record
and
measure.
L
You
know
some
of
these
fluctuations
in
both
time
and
space,
and
this
could
be
further
incorporating
the
risk,
modeling
and
assessment.
So
you
know
I'm
very
exciting
if,
in
the
future,
where
we
have
applications
are
able
to
copy
these
streams
of
you
know,
data
is
flowing
in
right.
You
know
there
will
be
video
XIV,
so
many
things
that
I
might
share
now,
you
know,
might
seem
like
a
pie
in
the
sky,
but
you
know
again
this
which
this
is
something
that
need
you
know
will
be
covered
within
our
lifetimes.
So
that's
for
me.
A
L
D
B
A
You
see
my
screen,
yes,
so
I'll
be
presenting
the
work
that
has
happened
since
the
last
meeting
on
architectural
principles.
Draft
in
short,
we're
kind
of
moving
now,
hopefully
towards
a
close
on
it
and
there's
been
quite
a
few
updates
since
the
last
meeting.
So
first
a
recap
and
the
first
version
of
the
draft
was
prepared
and
presented
at
the
ATF
104
in
Prague
on
26th
of
March
last
year.
A
A
So
it
was
adopted
by
ATF
104
and
continued
in
a
series
of
web
call
discussions
and
September
October
November.
Last
year
before
IETF
106
I
received
a
lots
of
feedback,
a
lot
of
people
contributed
and
basically
almost
the
entire
document
has
been
reworked.
It's
much
more
comprehensive
now
and
accessible
than
it
was
at
first
and.
H
A
Thanks
for
all
that
feedback
and
I
do
feel
the
document
is
much
better.
Now
it's
maintained
on
github
there's
a
link
I
shared
that
link,
often
on
the
mailing
list.
It's
a
public
repository.
It's
just
a
convenient
way
to
share
updates
so
that
I
don't
have
to
update
the
data
tracker
for
every
type
poll
etc,
and
that
kind
of
basically
there
can
be
in
between
meetings,
live
version.
I
do
update
the
data
tracker
for
the
meetings,
though,
but
I
don't
do
any
of
the
fancy
github
stuff.
It's
just.
A
There
is
a
repository
so
overview
of
the
changes
since
the
last
meeting
and
I
have
two
new
authors
who
have
contributed
with
the
section
on
elementary
linked
generation
which
I'll
go
over
and
slider
to
Bruno
has
contributed
a
comparison
of
classical
networking
which
I'll
also
go
over.
It's
just
a
very
interesting
thought:
experiment,
kind
of
outlines
the
necessary
components
of
in
network
architecture.
There's
some
progress
with
shelter
that
contrasts
entanglement,
swapping
in
Costco
forwarding
and
some
other
minor
updates
and
modifications.
A
So
what
was
this
section
about?
Elementary
link
generation?
So,
as
you
know,
ultimately,
a
we
will
want
to
perform
and
tangle
man
swapping
by
we
to
perform
a
kind
of
swapping.
We
need
to
have
entangled
pairs
on
the
individual
links.
First,
we
refer
to
these
often
as
elementary
links,
and
this
is
actually
a
very
physical
process
and
it
involves
sending
flying
qubits,
which
are
usually
implemented
as
photons
on
a
fiber
and
the
additional
section
kind
of
goes
over
briefly
as
to
how
you
would
achieve
that
there
are
actually
a
few
ways
you
can
do
it.
A
The
key
point
in
all
of
this
is
that
it
helps
you
understand
how
a
node
receives
an
entangled
pair,
because
I
think
that
actually
was
a
question
once
asked
on
the
web
calls
effectively.
It's
kind
of
you
touched
on
the
fact
that
when,
when
you
untangle
two
nodes
that
are
neighbor
there,
there
will
be
a
heralding
signal.
There
is
a
component
in
the
architecture
that
will
effectively
no
entanglement,
has
happened
and
heralds
it
to
the
notes
and
that's
kind
of
how
you
receive
you.
A
A
There's
also
now
a
new
section
at
the
very
end,
there's
a
quantum
networking
with
classical
networking,
most
notably
essentially
an
NPS
architecture.
It's
not
a
proposal
for
a
network
architecture.
It's
I
find
it
a
very
useful
kind
of
few
paragraphs
that,
for
somebody
who
really
understands
classical
networking,
read
that
section
I
hope
and
get
a
good
grasp
of
what
is
needed.
What
are
the
key
components?
What
will
be
the
key
components
in
a
quantum
network
architecture?
A
How
do
we
move
to
an
actual
network
architecture
and
I
feel
that
section
is
actually
a
very
good
starting
point
not
by
through
very
tiny
nitty-gritty
detail
about
how
you
do
it,
but
actually
kind
of
just
throwing
out
an
idea
that
a
lot
of
people
here
will
be
able
to
digest
very
quickly
and
the
benefit?
Is
it
identifies
a
lot
of
key
components
of
a
potential
narrow
architecture
in
a
very
short
amount
of
space,
and
apparently,
shota
has
submitted
a
pull
request
on
github
and
I've.
A
A
The
point
that
I
was
trying
to
make
is
that
the
entanglement
swapping
way
of
distributing
the
way
will
be
build.
This
first
generation
of
entanglement
based
networks
is
that
we
won't
be
forward
in
quantum
packets
like
we
forward
normal
classical
packets
topic
is
in
many
ways
different
you're
forwarding,
so
we
should
actually
probably
just
drop
this
forwarding
rod.
A
Lee
mentioned
that
once
we
get
to
the
third
generation
where
we
can
do
to
prepare
connection
and
Harry
detection
actually
implement
forwarding
and
that's
not
the
way
we'll
be
doing
it
for
the
fridge
generation
of
networks,
I
hear
some
keyboard
and
the
keyboard
person
sees
an
example.
These
entangled
pairs
are
not
directed.
This
is
it
kind
of
at
one
of
the
points
made.
So
it's
currently
work
in
progress.
It's
some
good
hub
and
not
present
in
the
draft
version
and
I'll,
be
reviewing
that
and
we'll
be
seeing
how
that
evolves.
A
All
discussion
happens
through
the
mailing
list
anyway.
So,
if
you're
on
the
mailing
list,
all
the
links
to
the
github
should
be
present,
and
so
they
can
also,
whenever
there's
an
update
on
github,
there
will
be
some
pink
on
the
mailing
list
that
there's
an
update
so
looking
forward.
What
are
the
next
steps?
Well,
first
I
kind
of
wanted
to
have
him
and
there
was
some
feedback.
I
wanted
to
have
more
coverage
of
security.
I'm,
not
an
expert
on
this
and
I've
had
no
updates
from
anybody
on
this.
A
So
I
was
thinking
that
perhaps
it's
a
complex
topic
that
is
not
really
suitable
for
an
introductory
document
and
one
potential
option
is
to
leave
it
as
it
is
for
now,
because
there
is
some
coverage
very,
very
basic
coverage
of
the
security
considerations,
add
some
references
and,
if
needed,
create
a
new
document
in
the
future
that
will
be
devoted
to
security.
That's
up
for
discussion.
If
people
do
want
security
in
this
document,
then
I
will
need
some
contributions
in
this
section.
Otherwise
I
will
leave
it
as
it
is
looking
forward
to
that.
A
Key
next
step
is
to
rework
section
six
and
which
is
about
the
essentially
the
goals
and
principles,
this
kind
of
an
open-ended
section
of
the
draft,
it's
kind
of
looking
at
what
are
the
goals
of
a
quantum
network?
How
do
we
want
to
build
them
and
more
principles?
A
little
update,
I,
basically
deleted
everything.
I
felt
was
my
opinion
by
science,
but
nevertheless,
I
left
things
that
were
basically
definites.
A
Their
habit,
upgraded
to
their
own
section
and
I
was
going
to
have
a
side
meeting
and
Vancouver
to
discuss
to
have
some
community
input
so
that
in-person
meeting
what
I
will
do
is
either
wait
for
all
the
virtual
interns
to
replace
Itt
407
to
finish
and
then
have
a
virtual
meeting
or
I'll
just
continue.
The
discussion
of
the
mailing
list
I
would
prefer
it
to
be
a
meeting
where
people
can
talk
simply
because
it's
just
faster
and
the
mailing
list
may
supplement
that.
A
Other
modifications,
I
will
need
to
add,
is
add
a
references
there's
a
bit
of
discussion
about.
What's
the
right
amount
of
references
for
the
document
and
the
consensus
seems
to
be
more
than
currently
there
I've
put
in
and
but
not,
let's
not
go
overboard
and
after
all,
that
I
will
tell
us
a
plan.
A
final
editorial.
A
rundown
Rodney
sent
an
email
with
a
bunch
of
people
at
the
beginning
of
March
and
I'll.
Use
that
as
a
starting
point
to
this
finally
share
one
time.
I.
I
Have
a
couple
of
questions,
the
first
one
is
it's
the
second
time
you
mentioned
that
in
upcoming
generations
you
can
have
forewarning.
Do
you
mean
that
you
you
you
would
forward?
You
would
teleport
qubits
over
each
elementary
link,
one
after
the
other,
because
then
it
doesn't
look
like
very
efficient,
and
the
second
question
is
is
related
to
them.
A
Right
so
the
first
question
I'll
actually
so
there's
the
first
generation
is
based
on
tangle,
swapping,
whether
it's
less
so
ultimately,
once
we
have
error
correction,
yes,
I,
don't
know
if
forwarding
teleporting
comp
I
hope
may
be
less
efficient.
It
will
definitely
be
simpler
and
potentially
not
require
establishing
circuits,
because,
if
you're
doing
entanglement
swapping
the
current
harder
requirements
effectively
require
you
to
establish
a
circuit
which
means
resource,
reservation
and
stuff,
so
forwarding
might
be
simpler
in
that
context
and
I'm
not
an
expert
on
anything
beyond
the
first
generation.
B
Second
generation
will
still
work
on
entanglement
swapping,
but
it
operates
at
the
logical
level
on
top
of
doubt,
there's
that
are
created
across
the
link
that
are
encoded
in
in
a
qac,
a
quantum
error
correction
system.
Third
generation
is
the
one
that
gets
you
real
store
and
forward,
but
that
requires
float
on
reception.
Probabilities
of
you
know
greater
than
ninety
percent,
or
something
which
is
way
way
way
way
far
in
the
future.
A
A
So
the
second
question
about
layers
and
stuff
I
intentionally
am
avoiding
layers
and
there
is
a
mention
of
planes.
I
think
the
current
document
I
it
for
a
good
reason,
because
we
had
surface
because
I
there's
currently
a
tendency
for
everybody
to
literally
map
the
classical
Network
stack
to
the
quantum
network.
Stack
and
I
do
also
want
to
avoid
this
principles.
Draft
be
avoid
being
in
architecture
proposal
like
formational
document
and
as
soon
as
I
put
layers
in
I
think
that
becomes
too
close
to
a
proposal
for
my
liking
for
that
doc.
A
Well,
the
draft
is
kind
of
just
open-ended
and
kind
of
just
like
an
introduction,
but
in
principle
you
could
use
any
of
those
simulators
to
simulate
to
basically
you
what
you
should
be
able
to
do.
Anybody
actually
on
this
call
you'll
be
able
to
take
this
draft
and
hopefully
understand
it
and
then
take
one
of
those
four
simulators
and
just
start
playing
around
with
it
and
that's
it
and
you
should
be
able
to
start.
A
B
G
G
I
mean
that
the
process
is
once
the
working
group
is
but
believes
it's
ready.
You
send
it
to
me
and
I
make
sure
it
gets
reviewed
by
the
IRS,
gee
and
Conflict
review
with
the
IETF
work
in
terms
of
time
frame
in
the
the
slide
suggests,
having
it
ready
by
I
what
I
wait,
109,
which
I
guess
is
the
end
of
the
year,
and
that
seems
like
a
perfectly
reasonable
time
frame,
but
err
I
mean
fundamentally
it's
when
the
group
believes
it's
done.
B
B
M
This,
let's
drop,
the
PC
is
about
the
quantum
Internet
applications
and
some
selected
use
cases.
So
we
were
trying
to
focus
on
this
topic,
so
this
draft
Archer
is
stuck,
naturally
was
initiated
back
in
January.
So
since
then,
when
you
exchange
the
draft,
we
have
received
many
good
comments
and
feedback
through
the
Middle,
East
and
I
think
currently,
based
on
that
we
generated
the
current
version.
0
5
will
receive
the
many
valuable
inputs
from
lots
of
people,
at
least
here.
M
M
So
the
the
main
contents
in
this
document
basically
has
a
three
major
sections,
like
the
quantum
Internet
applications,
we
briefly
introduce
a
two
classification
categories:
I
like
the
PI
application
usage
or
by
the
control
and
the
data
plan
for
the
application
usage.
So
basically,
we
say
the
there
are
three
categories,
like
a
quantum
cryptographic
applications,
the
quantum
sensor
applications.
Third,
one
is
the
quantum
computing
applications
in
terms
of
control
and
data
plane
classification,
this
one
we
there
are
a
few
pending
comments.
M
We
need
to
address
so
this
one
is
still
pending
about
its
control
and
in
a
planned
classification
so
based
on
the
classification
of
applications,
and
then
we
select
a
few
quantum
Internet
use
cases
by
details.
For
example,
we
have
the
secure
quantum
communications
and
also
we
have
the
East
tribute
a
quantum
computing,
also,
the
we
call
the
quantum
computing
with
their
privacy
preservation
or
blind
computing.
So
after
that
we
were
trying
to
under
section
dedicated
general
requirements,
but
this
one
is
16
the
still
in
the
early
stage.
M
M
So,
for
the
first
class
of
the
for
the
first
type
of
quantum
applications,
the
coffee
applications,
so
we
think
it's
kind
of
the
the
category
for
using
quantum
information
technology
to
ensure
secure
communications
who
examples
can
be
here
is
secure
communication
setup,
for
example,
to
secure
the
cryptographic
key
distribution
between
two.
Are
you
in
Moore
and
notes
quantum
key
distribution,
which
is
of
the
approaches
for
doing
for
this
purpose?
M
This
could
be
used
for
the
financial
brockton
scenario.
The
second
type
of
the
corner,
kitchen
is
called
the
quantum
sensor.
Applications
means
to
use
the
quantum
information
technology
for
supporting
is
the
ability,
sensors
or
human
IOT
devices
example
will
be
like
the
network
clock
synchronization,
it's
not
about
the
two
nodes
to
synchronize
their
clocks
and
so
no
authority
in
middle
work.
A
worldwide
set
of
atomic
clocks
liquefied
upon
Internet
to
achieve
a
Archer,
a
more
precise
clock,
synchronization,
so
the
third
wiser
about
the
quantum
computing
so
actually
the
first
copies
communication.
M
M
So,
given
those
three
types
of
quantum
emissions
in
terms
of
you
know
the
real
use
case,
we
were
trying
to
come
up
with
those
three
use
cases
inference
that
the
first
use
case
here
we
called
a
secure
communication
set,
have
basically,
for
example,
like
the
two
bags.
They
need
to
have
secure
communications
for
transmitting
their
important.
A
financial
transaction
records
between
the
you
know
the
the
back
number
one
and
a
back
number
two.
We
assume
each
back
has
a
corresponding
one
nodes,
a
and
B.
M
So
for
this
purpose,
the
new
to
their
requirement
is
need
to
securely
exchange,
a
classical
secretive
graphical
fee,
which
could
be
triggered
by
either
back
pack
number
two
or
bag
number,
one
so
figure,
one
here,
which
just
says
show
they
will
be
triggered
by
and
the
user
in
bag
number
one.
So
this
use
case
basically
need
to.
A
M
Near
term,
or
even
be
a
term
what
can
do
have
the
noisy
intermediate
scale,
quantum
computers
and
Ice
Cube?
They
are
going
to
be
distributed
in
different
locations.
Also,
they
are
available
to
be
shared
together.
So
ideal
here
is
trying
to
use
a
disability,
an
ice
cube
computers
to
you.
You
know
to
get
more
powerful
computing
capability.
Are
you
order
to
in
this
car
computing,
a
power?
M
M
There
are
many
small
computers,
computers,
disability
in
different
places,
and
we
want
to
collect
them
together
to
have
a
more
powerful
computing
capability
and,
of
course,
if
you,
if
the
quantum
computer,
a
for
example,
lost
to
say
library
transmitted,
the
quantum
base
states
from
A
to
B,
the
quantum
computation
can
be
utilized.
For
this
purpose
use
case.
Here
we
could
secure
quantum
computing
with
the
privacy
preservation.
M
It
also
can
be
referred
to
call
it
blinds
quantum
computing.
So
here
the
use
case
of
a
figure
based
in
agony.
We
have
a
quantum
terminal
node
if
you
look
at
figure
3,
for
example,
from
gateway,
and
then
they
want
to
be
negate
some
computing
to
a
remote
quantum
node
which
could
be
upon
a
computer
in
the
cloud.
But
here
is
they
don't
want
to
share
the
original
source
data,
so
the
user
will
here
why
we
call
the
quantum
computing
with
the
privacy
preservation,
so
basically
the
quantum
terminal,
node
and
home?
M
Could
citizen,
Conan,
Bates
applause,
some
measurement
instruction
to
the
remote
control
node
after
that
the
remote
computer
node?
Can
you
know,
based
on
the
measurement
and
instruction
to
some
competition
and
O'connor,
and
send
back
results
back
to
the
common
terminal
node.
So
this
case
just
shows
living
usage
of
the
quantum
information
technology,
for
you
know
for
the
computation
dedication
with
the
privacy
preservation.
M
So
those
are
three
use
cases
selected,
and
then
we
were
trying
to
have
a
section
about
the
general
requirements.
Currently,
we
have
a
three
high
level
requirements.
The
first
one
is
a
method
for
facilitating
quantum
applications
to
interact
efficiently
with
the
integument
of
hobbits
are
necessary
after
this
requirement,
I
guess
by
her
into
several
simulator
a
plantation
one
of
the
remedy.
I
think
this
is
also
reflected
in
the
current
simulators
kind
of
API,
between
the
applications
and
in
loader
of
phantom
layers.
M
The
signal
requirement
is
more
like
the
quantum
repeaters
and
at
others,
should
support
robust
and
efficient
the
endowment
distribution,
and
then
we
have
a
third
requirement
here
is
more
like
we
need
to.
You
know
even
the
pump
and
those
they
need
to
support
the
classical
communications
to
serve.
You
know,
even
for
the
Coke
ad
and
the
quantum
teleportation,
the
classical
communications
are
required,
it's
chemistry
for,
but
based
on
the
feedback
from
meninist
we
and
a
to
do
notes
in
the
document.
M
Some
performance
indicators
need
to
be
defined
and
described,
putting
in
which
level
of
those
performs
matrix.
We
would
like
to
cover
in
statement
it's
a
question
for
us,
so
in
terms
of
the
liquid
state
and
as
I
mention
early,
we
have
a
few
pending
comments,
relatively
a
control
plan
and
the
plan
from
may
acquire
and
may
pass
so
welcome
to
address
this
control
player
and
data
plan,
an
application
classification.
We
would
like
to
also
continue
to
collect
a
feedback
from
the
KRG
from
Middle
East
and
trying
to
improve
the
diversity
in
the
row
file.
M
We
also
got
a
question
for
the
co
arg,
the
first
christian
apart
he
performs
matrix
in
the
interim
Sorrentino
requirements.
I
just
briefly
talked
about
that.
The
second
question
is,
since
we
have
received
a
lots
of
interest
feedback
from
the
the
quality,
so
we
were
wondering
and
if
this
idea
is
really
for
research
who
bought
option
or
sometime
later
or
just
trying
to
okay
feedback
from
the
co
r
g.
I
Yes,
yes,
the
idea
was
just
to
answer
the
first
question
of
option,
which
is
how
much
detail
I
think
I
think
we
should
in
this
document
it's
a
user
document,
so
you
should
just
I
mean
specify
the
things
or
one
of
the
things
in
a
way
the
user
would
would
would
need
it,
I
need
something
that
is
available.
That's
with
that
level
of
probability.
You
mean
something
that
is,
and
that
has
a
fidelity.
I
M
B
B
B
A
Think
this
is
a
good
successful
meeting.
Thanks
for
the
tendons,
I'm
very
happy
to
see
all
the
activity
and
the
simulators
and
I
think
it'd
be
great.
If
people
could
slowly
start
people
here
can
slowly
start
picking
them
up
for
their
own
exploration
and
use
cases
I'd
be
great
if
people
just
shared
their
output
on
the
mailing
list
as
well.
B
Ether
pad
as
being
attendees,
so
I
think
we've
probably
got
everybody
if
you
did
not
sign
in
on
the
blue
sheet
on
the
ether
pad.
Please
do
so
that
ether
pad
will
be
closing
well,
we
will
be
using
that
as
our
attendee
list,
and
otherwise
we
will
see
you
all
on
the
mailing
list
and
hopefully
in
person.
Sometime
later
this
year,
Kovac
Willick.