►
From YouTube: IETF103-MILE-20181105-1610
Description
MILE meeting session at IETF103
2018/11/05 1610
https://datatracker.ietf.org/meeting/103/proceedings/
C
D
C
C
C
E
F
G
G
H
B
Welcome
to
the
mile
working
group,
so
we
actually
have
a
good
turnout.
It's
not
bad
you're,
smiling
Roman,
all
right
next
slide.
So
it's
kind
of
the
end
of
the
first
day.
So
I'm
presuming
you've
attended
some
sessions,
so
I
don't
have
to
go
through
the
note.
Well,
but
standard
procedures
apply.
Let's
go
through
the
agenda
next
slide,
so
there
is
still
also
a
c-cert
draft.
B
But
since
the
author
isn't
here,
we
we
took
them
off
the
agenda,
so
there
may
be
a
little
discord
with
what
might
be
posted
in
the
in
the
meeting
materials
so
we'll
go
through
the
status
of
the
working
group.
Thank
you.
Everybody
for
providing
the
feedback
over
the
email
in
the
last
couple
months
to
get
the
retargeting
text
done
so
we'll
go
through
that
real
quick
I
will
provide
an
update
on
the
XMPP
draft
taki.
Will
you
provide
the
update
on
the
JSON
draft
and
then,
if
there's
any
other
business,
we
can
discuss
that
last?
B
I
B
B
B
So
basically
mile
for
those
of
you
who
may
be
new
to
this
group,
it
is
the
focus
is
for
us
to
define
data
representations
for
incidents
effectively
and
also
define
the
mapping
for
how
we
transport
the
incidents.
There
are
four
the
main
data
representation
we've
been
working
on
IO
death,
as
if
the
effective
data
model
there
have
been
some
extension
work
on
it,
as
well
as
interoperability
and
implementation
reports.
So
there's
a
couple
of
drafts
there
that
are
listed
for
the
transport.
B
B
Oh
yeah,
yes,
you're
right!
Sorry,
thank
you
for
that
correction
and
then
there's
been
some
guideline
documentation
and
and
practices
for
implementation
of
I/o
death
as
well,
and
that
work
on
the
guidelines.
As
far
as
we
know,
those
have
been
completed
so
the
outstanding
items,
as
we
know
them
now,
although
they're
moving
further
along,
is
the
JSON
and
satiation
for
I/o
death,
providing
c-cert
reports
through
Roly
and
then
using
an
XMPP
based
fabric.
B
Only
one
blocking
comment
and
one
was
because
we
we
failed
to
get
the
acronym
correct.
Four
sticks.
We
had
the
acronym
correct.
We
just
didn't,
have
them
correct
as
well
as
the
link
for
it.
So
we've
resolved
that
the
next
tell
a
chat
to
approve
that
we
chartering
is
happening.
I
believe
November
21st,
so
hopefully
we'll
get
feedback
from
there.
Okay,
so
reviewing
the
milestones
as
far
as
we
believe.
So.
Thank
you
talking
for
going
through
some
of
the
Shepherd
right
up.
We
completed
the
working
group
last
call
for
XMPP
grid.
K
B
B
B
J
B
J
Yeah
so
I've
already
started
draft,
especially
on
the
the
JSON
based
roli
work,
started
working
on
that
and
have
quite
a
bit
of
text.
There.
I
just
have
some
questions
for
Steven
or
they
whatever
to
help
me
make
sure
it
all
fits
together
and
then
start
working
on
the
sticks.
Pieces
of
that
to
make
sure
that
the
sticks
content
can
flow
through
really
okay,.
B
Make
sense-
oh
good,
our
ad
is
here.
No,
that's
all
right,
that's
all
right!
So
Alexi.
What
I
had
mentioned
is
that
the
recharter
is
going
through
the
review
right
now
and
we
submitted
so.
The
telogen
is
I
think
scheduled
for
the
21st
of
November,
and
so
what
I
was
recommending.
The
bread
is
that
he
doesn't
need
to
wait
for
the
recharger
to
happen
to
get
any
new
work
done.
I.
L
B
M
B
G
B
Thank
you
for
pointing
that
out,
but
once
the
draft
expires
it
kind
of
yeah,
so
I
can
check
to
see
if
maybe
Dave
can
help
with
that
no
takers.
If
you
can
note
that
that'll
jog
my
memory
too,
to
reach
out
to
Dave
altameyer
too,
as
well
as
Steven,
okay,
so
with
that
we've
talked
about
the
mile
status.
So
let
me
take
my
chair
head.
B
Can
you
guys
hear
me
okay,
that
was
really
loud?
Yes,
so
I
didn't
make
any
slides
for
the
XMPP
draft,
because
the
only
changes
to
me
were
more
editorial.
So
from
that
standpoint,
the
difference
between
versions,
7
&
8-
is
that
there
are
no
longer
any
dependencies
on
the
second
draft.
Given
that
it's
effectively
showing
the
mapping
of
how
XMPP
gets
used
to
carry
other
incident
data
like
I/o
death,
there
didn't
seem
to
be
any
reason
to
keep
the
second
terminology
so
that
was
removed.
L
B
L
B
L
N
This
is
mapping
to
see
ball
and
Jason
Abhaya
deck,
so
this
is
currently
version
6.
This
is
the
update.
In
the
last
meeting
in
Montreal,
we
discussed
version
flow
of
the
draft,
so
in
the
version
from
draft
we
adopted
seed
et
al
instead
of
Jason
and
we
already
provided
seed
EDL,
but
in
the
last
meeting
the
seed
et
al
data
model-
I
provided
was
not
perfect.
N
I
was
not
confident,
and
so
in
these
two
or
three
months
we
were
rewriting
that
by
asking
expert
in
Japan
and
we
got
some
feedback
from
the
expert
and
we
thoroughly
revised
a
document
and
updated
now
the
version
is
0
6.
In
this
meeting
we
have
6
0
6
version,
so
it
has
a
refined
version
of
the
data
model.
Hopefully
we
can
initiate
the
Skoll
or
we
could
ask
more
intensive
review
from
people.
N
J
On
the
CD
CD
v
l
front
when
I
was
talking
with
Hank,
apparently
there's
no
tooling
yet
really
for
cvbl,
so
maybe
it's
a
tooling
to
actually
consume
the
CD.
Do
I
do
any
verification
or
use
in
other
programming
languages,
so
he
so
so
yeah.
It's
real
ambitious
that
we
want
to
support
it
and
I.
Think
CDLs
is,
has
a
potential
being
really
cool,
but
we
need
like
if
we
want
people
to
use
our
standards,
they
need
to
be
like
usable
in
programming
languages.
So
but.
N
At
least
we
have
some
tool
provided
by
casting
that
that,
if
real
friends
from
seaboard
raft,
we
are
using
that
tool
as
well.
So
we
have
something
to
rely
on
that.
My
understanding,
but
anyway,
as
you
said,
the
draft
is
still
draft.
Mggg
are
still
a
draft
if
we
have
to
wait
until
they
finish
their
work.
We
just
great.
L
L
B
O
Fracture
Howie
kaku
kaku
explained
the
basecode
reason
why
we
switch
from
Jason
to
see
people,
because
I
am
mr.
the
be
our
previous
discussion,
because
my
the
rhythm
guide
with
this
question,
because
if
because
I
know
this
Maya
working
for
the
threat
intelligence
exchange,
so
it
does
not
need
so
high
efficiency
or
so
scalability
problems.
So
why
we?
Why
we
should
seek
people,
because
it's
no?
No!
It's
for
efficiency
for
the
so
yeah.
N
At
the
beginning,
we
didn't
focus
on
seaboard
at
all.
We
were
focusing
on
Jason.
G
is
not
easy
to
handle
by
computer.
That
was
a
certain
point,
but
we
wanted
to
define
or
chips
syntax
by
using
some
schema,
but
JSON
schema
was
not
useful,
not
defined.
Well,
so
we
decided
to
use
a
GDL,
but
once
we
have
a
cvbl
is
easy
to
describe
everything
in
Siebel.
As
a
result,
we
have
Siebel
and
she
she
bow
and
Jason
mapping
currently
so.
O
N
O
M
L
J
This
is
my
last
comment:
I
promise
and
then
I
will
shut
up,
but
the
whole
purpose
of
defining
these
in
specification
documents
and
standards
is
so
that
we
can
do
verification
of
content
that
we
get
to
make
sure
that
the
content
is
compliant
with
some
sort
of
specs
before
we
try
to
parse
it
and
do
anything
with
it.
Cdl
yeah,
I,
look
I
just
looked
that
up.
There's
this.
You
know
Ruby
kind
of
thing
that
kind
of
works
and
I
know.
J
N
Okay,
so
in
this
side,
I
want
to
summarize
the
changes,
so
data
model
refined
second
here
at
I,
wrote
of
JSON
schema,
was
corrected.
Third
example
document
is
now
available
in
not
only
JSON
and
also
civil
binary,
and
also
we
have
a
title.
That
draft
was
changed.
We
added
Sheba
in
addition
to
Jason
and
some
errata
was
spotted,
so
that
is
the
changes.
So
let
me
just
what's
really.
D
N
From
here,
I
want
to
talk
about
to
errata
I
found
in
the
RFC
7970,
because
you
know
we
are
generating
converter
converting
tool
from
XML
to
JSON.
We
happen
to
find
some
kind
of
errata
that
was
not
compatible
to
Jason.
So
this
is
the
point.
The
first
point
here,
current
schema,
it
described
I
exist,
but
if
we
take
a
look
as
main
body
of
the
I
own
the
question
it
allows.
N
N
N
J
M
N
So
we
found
errata
and
based
on
new
correct
schema.
We
generate
some
kind
of
converter
from
XML
to
JSON
and
also
JSON
to
XML,
so
we
have
some
implementation
already
and
the
document
is
almost
ready
and
indeed
the
data
model
itself
was
already
discussed
before
when
we
were
working
on
Io
tech
person.
But
we
do
not
see
any
issue
to
discuss
here
anymore.
So
if
nobody
object,
I
want
to
ask
you
whether
we
can
proceed
to
the
working
group.
Let's
call
the
question
to
the
audience
and
also
to
the
chair.
N
K
K
This
is
overview.
Yeah
I
mean
to
come
back
from
Jason
I,
add
a
document
from
XML
I
would
eat
up
mint
and
vice
versa,
and
this
is
running
code
or
radius.
Jason
ionic
draft
and
we
are
currently
implementing
new
Khambatta,
including
including
I,
give
hugs
on
dates
in
Bangkok,
but
sorry
and
we
cannot
solve
a
rat's
on
previous
implemented
from
basa
based
on
python.
Json
schema
object
which,
which
is
presented
in
IKEA
Londo.
K
K
K
Okay-
and
this
is
the
current
status
and
future
plan-
we
have
successfully
combated
or
examples,
and
this
iota
trapped
by
Diane
bi-directionally,
and
we
also
combated
all
examples
in
IO
their
usage
guide
as
RFC,
and
we
will
published
that
Khambatta
in
the
talk
public
repository
near
future,
and
we
also
planning
to
increment
conversion
to
see
both.
Okay,
that's
it.
Thank
you.