►
From YouTube: IETF95-ECRIT-20160406-1000
Description
ECRIT meeting session at IETF95
2016/04/06 1000
A
B
C
Good
morning
welcome
to
eco
at
ATF
95
I'm,
Roger
Marshall.
This
is
Mark
lens
Newark
and
let's
go
through
the
the
chair
side
before
we
do,
though,
I'm
going
to
pass
around
the
blue
sheets
and
we
missing
a
presenter,
hopefully
he'll,
show
up
and
noticed
and
in
the
pink
box
all
right
goes
without
saying.
C
That's
right:
we
do
need
a
note
taker
and
a
jabber
scribe
for
today's
meeting.
These
notes.
Don't
have
to
be
elaborate,
just
quick
capturing
the
decisions
made.
Is
anybody
willing
to
take
notes
and
we're
essentially
the
impasse
until
we
get
some
Oh
Christine,
cakes,
okay,
we'll
remind
ourselves
and
others
to
speak
slowly
and
how
about
a
jabber
scribe.
D
A
E
D
D
D
The
agenda
today
is
the
ITF
was
rather
generous
and
in
the
time
slot
it
gave
us
so
we're
rather
generous
and
the
time
slot
we've
given
our
presenters
today.
So
10
minutes
for
agenda
bashing
in
the
chair,
slides
that
may
go
over
a
little
bit.
My
guess
is
the
next
three
20-minute
time
slots.
That's
a
total
of
sixty
minutes,
maybe
more
like
six
minutes,
but
yeah
well,
word
so
will
will
certainly
look
forward
to
letting
everybody
out
early.
D
D
So
document
status,
so
these
are
the
five
one.
Two
three
four
five
six
documents
the
the
workgroup
is
is
is
working
on
right
now
we
have
the
additional
data
related
to
an
emergency
call
draft
that
is
in
the
RSC,
editor
q,
it's
in
auth,
169,
x,
24
or
it's
been
in
all
48
for
169
days
now
and
we're
working
out
late
comments
and
the
authors
will
update.
The
slide
was
made
about
three
minutes
prior
to
the
author's
actually
uploading,
the
new
version,
so
that
bullet
up
there
is
actually
wrong.
D
So
there
is
a
new
version
up
there.
It's
actually
generated
a
little
bit
of
activity
in
the
list
if
you've
followed
the
activity
on
that
draft
on
the
list
you're,
probably
aware
of
what's
going
on,
but
I
suggest
we
can
discuss
it
here.
If
anybody
has
any
problems
with
it
and
want
to
talk
about
it
looks
like
alyssa
is
going
to
try
and
make
an
explanation
of
why
it's
been
why
it's
off
1
69,
x,
24.
I
should
calculate
that
out.
Alyssa.
F
D
D
The
next
two
drafts
next-generation
vehicle
initiated
emergency
calls
a
next-generation,
pan-european
eek.
All
the
authors
feel
are
ready
to
go.
The
iesg
3gpp
is
working
on
requirements
for
a
like
service.
We've
learned
that,
through
both
a
liaison
and
comments
to
the
list,
so
since
we're
ready
to
take
that
280
review,
ITF
last
call
iest
review
and
then
pub
instead
of
stopping
and
waiting
for
3gpp.
As
of
right
now
we're
going
to
go
ahead
and
progress.
D
The
two
drafts
all
the
way
through
up
to
the
request
pub
process
before
Alyssa's,
the
ad
Shepherd
of
that
before
you
get
to
the
button
to
pub
request,
we're
going
to
resync
with
the
3gpp
requirements
and
see
where
they're
at
and
make
a
decision
on
that
time.
Rather
to
publish
it
or
not.
So
that's
what
we're
going
to
do
with
those
two
drafts.
D
G
D
D
Last
call
right
after
this
meetings
over,
we
did
a
working
group
last
call
a
year
ago
on
that
craft
and
I
we're
going
to
hear
a
little
bit
from
the
author
today
he's
going
to
present
on
his
latest
he's
updated
that
you
know
just
a
while
ago
this
here
with
those
updates
are
so
if
we
got
any
issues
or
if
you
haven't
been
keeping
track.
You
made
aware
that
today,
a
routing
request
extension
for
the
held
protocol.
D
D
D
Okay
and
then
a
lost
extension
to
return,
complete
and
similar
info
we've
had
several
several
reviews
on
that
from
the
public
safety
community,
and
it's
in
pretty
good
shape.
We're
going
to
take
that
the
working
group
last
call
next
week
and
we're
also
going
to
hear
about
any
the
latest
changes
in
in
that
draft.
So
that's
our
document
status.
Next,
we'll
take
a
look
at
the
milestones
and,
if
you've
looked
under
milestones
and
paid
attention
to
the
list,
you've
seen
that
they
are
magically
lining
up
with
dates
that
are
pretty
close
to
what
we're
meeting.
D
B
D
B
H
Does
that
look
better?
That
looks
better,
so
data
only
is
a
way
to
carry
as
the
way
to
create
a
session
that
has
only
has
data.
It's
used
for
things
like
alarms
on
where
you
want
to
create
an
emergency
call
we're
at
it
like
an
emergency
call,
handle
it
like
an
emergency
call
market
like
an
emergency
call,
but
it
doesn't
create
an
interactive
media
session.
So
we
use
message
to
do
that
and
we
carry
a
cap
message
in
in
the
body
of
the
of
the
mount
of
the
sip
message.
H
We
had
no
objections
on
the
working
group
last
call,
but
we
did
get
a
comment
after
the
last
call
about
handling
privacy
of
that
to
the
fact
that
privacy
is
subject
to
local
regulation
and
what
works
in
one
country
may
not
work
another.
That's
a
very
reasonable
comment
and
well
I
will
create
a
new
version
and
and
add
that
to
the
to
the
draft
no
problem
arm-
and
there
are
there's
there's
a
waiting
constituency
for
this.
H
E
H
Yeah,
okay,
so
we
can
run
it
again,
but
it
it's
really
pretty
stable,
pretty
simple
and
with
this
privacy
issue,
which
is
just
fine,
we'll
make
I'll
make
that
change.
So
all
I
will
I
will
put
a
new
version
out
with
the
change
that
that
Keith
asked
for,
and
we
can
work
less
call
that
one
right,
yep,
ok,
questions
problems,
comments
on
this:
okay,
this
one's
really
straightforward
and-
and
you
know
other
than
that-
one
issue-
that
race
I,
don't
really
think
there
ought
to
be
any
reason
to
not
make
this.
Take
this
go
ahead.
I
I
H
So
this
is
a
deals
with
a
an
actual
real-world
problem
that
changes
in
the
data
that
is
used
for
routing
the
thing
that's
in
the
lost
server
there
are.
There
are
set
of
changes
that
have
a
schedule.
They
go
into
effect
at
a
known
exact
time.
The
example
that
we
canonically
use
is
that
there
that
that
cities
grow
by
annexation,
so
they
take
over
an
area
that
used
to
be
not
part
of
any
city
but
part
of
an
unincorporated
part
of
the
world,
and
they
on
midnight
at
some
day
in
the
future.
H
The
city
is
now
has
a
different,
a
different
boundary
and
the
calls
the
emergency
calls
that
used
to
route
to
the
larger
community
response
now
now,
instead,
should
respond,
should
go
to
the
city's
emergency
response
center.
That
kind
of
change
that
kind
of
a
problem
which
is
a
planned
in
advance
change,
isn't
catered
for
in
in
lost.
It
doesn't
have
a
way
to
do
that.
H
Sequence
of
making
a
plan
change.
Is
that
on
your
your
validation
information,
the
thing
that
you,
when
that
validation
may
or
may
not
be
correct
in
the
future
cities
often
have
when,
when,
when
a
change
like
that
gets
made,
it's
not
at
all
uncommon
for
the
address
to
change,
because
the
community
name
changes
for
example,
but
it
may
there
may
be
other
changes
that
come
as
a
result
of
the
the
the
city
taking
over
that
area.
H
Very
often,
the
rules
for
numbering
streets
are
very
lacks
in
the
non
City
area
when
the
city
takes
it
over,
it
enforces
rules
that
says
now
now
has
to
work
this
way.
So
they
end
up
re
numbering
streets,
sometimes
lots
of
things
that
happen
so
a
number,
so
a
location
which
was
valid
before
the
changes
may
may
no
longer
be
valid
after
the
change.
The
people
who
run
the
location
information
servers
are
aware
that
these
changes
are
going
to
happen.
H
They
need
to
know
whether
or
not
that
it
will
be
valid
going
forward,
and
you
want
them
to
do
all
their
work
before
the
change
goes
into
effect,
so
that
they
can
change
anything
that
they
need
to.
So
what
this
thing
does
is
it
lets
you
say,
is
something
valid
as
of
a
certain
date.
It
lets
you
make
a
change
to
the
database
and
say
this
will
happen
on
this
date
and
you
can
do
a
validation
that
says
well.
If
I
was
trying
to
validate
it,
have
us
of
this
date.
Would
it
be
valid?
H
It's
a
pretty
straightforward
thing.
It
adds
a
couple
of
parameters
to
the
query
in
the
response
and
we've
again.
If
this
is
this
is
a
practical
problem.
We
have
actually
a
counter
right.
We
listen.
We
encounter
these
issues
regularly
in
in
complex
emergency
calling
systems,
so
we
need
this,
and
several
people
have
spoken
up
and
said:
yeah
yeah.
This
is
a
real
problem.
We
really
need
to
do
this
and
they
number
of
people
have
read
it
and
and
and
have
asked
us
to
move
this
forward.
H
It's
still
a
personal
and
individual
submission,
I'd
like
to
I'd
like
to
adopt
it
and
I.
Think
it's
pretty
much
ready
to
go.
I
we
have
not
had
any
substantive
changes
for
a
while
a
bunch
of
little
editorial
knits
and
things
that
I've
done,
but
I.
It's
a
pretty
straightforward
thing:
it
just
adds
a
new
element
to
find
service
with,
has
a
uri
and
an
as
of
date
or
form
validation
against,
and
then
it
gives
you
I'm.
H
Sorry
I've
forgotten
one
more
thing
that
it
does,
that
that
the
it
also
allows
using
that
it
allows
the
server
the
lost
server
to
let
someone
know
that
validation
may
change
as
of
a
certain
date.
So
it's
a
it,
can
send
you
a
message
at
you.
You
stores
a
URI
against
an
address
and
a
validation
will
change.
As
of
some
data
to
let
you
know
that
that
will
happen
so
that
you
know
what
to
do
and
this.
This
is
something
that
will
be
very
useful
on
the
there's.
H
A
secondary
function
that
this
this
uri
save
with
location
gets
you,
which
is
also
a
practical
problem
which
we'd
like
to
solve,
which
is
the
issue
that
validation
does
change
over
time.
For
other
reasons,
other
than
the
plan
change.
The
line
change
is
the
most
common
case,
but
does
change
for
other
reasons
and
right
now
the
only
thing
we
say
about
revalidation
is:
do
it
periodically
right?
You
know
just
because
it
was
valid,
may
not
stay
valid,
so
we
say
well
every
once
in
a
while,
once
you
check
to
make
sure
validation
is
okay.
H
H
E
H
H
I,
don't
know
if
you've
seen
the
ITF
journal,
but
there's
an
article
in
it.
I
wrote
that
talks
about
how
the
that
there
are
50,000
emergency
calls
a
month
being
routed
with
loss.
Now
it's
it
is
a
success.
We
have
multiple
interoperable
implementations
of
this
thing.
There
are
there
I.
The
last
time,
I
tried,
counting
I
think
there
are
six
full-scale
deployed
implementations,
not
just
the
people
are
making
companies
making
money
selling
these
things
using
them
in
in
in
real
emergency
calling
stuff.
H
This
thing
works
and
people
are
using
it,
and
when
we
do
that,
we
discover
that
oh
gee
things
we
thought
would
be.
Okay,
aren't
like
this
revalidate
periodically
and
it
turns
into
most
of
the
traffic
and
the
server
okay,
so
similar
location.
This
is
a
little
farther
along.
It
is
an
ITF
is
it
is
an
adapted
draft
and
we
have
had
some
risk
traffic
on
this
thing.
H
Basically,
what
it
does
is
it
weighs
it
gives
you
a
way
to
return
location
information
in
the
response
and
use
it
in
two
circumstances.
One
is
when
you
have
an
invalid
location.
It'll
give
you
a
did.
You
mean
possible
replacement
value,
just
a
suggestion,
and
then
you
can
try
at
present
that
to
user
and
they
can
say
well
yeah,
that's
like
actually
I
mistyped
what
I
wrote
or
I.
H
So
that's
one
term,
but
one
place,
and
the
other
is
that
you
can.
You
can
send
in
enough
information
for
us
to
determine
an
absolutely
unique
location,
but
we
would
prefer
that
it
that
you
actually
use
the
full
address.
So,
for
example,
if
there's
only
one,
if
there's
only
one
street
in
an
entire
state,
that
has
a
particular
name-
and
the
only
thing
you
give
me
is
the
name
of
the
street
and
the
state.
You
give
me
no
city
name,
no
municipality
name
at
all,
but
it
turns
out
to
be
unique.
H
We
can
say
well,
okay,
you
know,
that's
unique.
I
actually
could
route
it
that
way.
But
would
you
please
put
the
city
name
in
this
thing
for
me
and
that's
because
that's
what's
stored
in
the
list
right
we're
talking
about
what
do
you
store
in
the
list?
We
want
you
to
store
the
full
address
and
the
list.
So
we
can
return
that
piece
of
information
saying
well,
we
found
this
thing
is
unique,
but
you're
missing
the
the
municipality
name,
here's
what
we
think
the
municipality
name
is
so
once
again,
this
has
been
reviewed
on.
H
H
I
hope
to
have
that
in
a
few
days
which
will
fix
the
other
comments
that
I
have
there's
one
one
person
sent
a
whole
bunch
of
calmness
and
I
got
some
of
them
in,
but
I
didn't
get
all
of
them
in
so
I
need
to.
I
need
to
do
that.
I
need
to
put
a
new
version
out
that
has
that
addresses
the
comments
that
the
technical
comments
that
we've
got
on
on
the
Sun
similar
and
I'm
going
to.
H
H
J
H
H
D
Which
is
the
features
of
the
work
group,
so
what
you've
heard
today
is
all
of
our
workgroup
documents
should
be
done,
we'll
throw
them
over
the
wall
within
the
next
few
weeks.
They
should
be
in
iesg
land.
You
know
in
the
next
month
in
starting
that
progress,
and
we
have
one
individual
draft.
That
Brian
is
asking
we
adopt
as
a
workgroup
item,
which
we
will
do
that
work
on
the
list.
D
So
obviously
with
you
know,
if,
if
just
let's,
let's
pretend
that
that
is
adopted
or
troop
item
that
will
be
basically
by
the
end
of
april
middle
of
may
and
the
mabe
will
have
one
document
we're
working
on.
So
obviously
berlin.
You
know
the
next
meeting.
If
that's,
the
only
document
will
have
to
make
the
decision
about
with
the
e
DS
one.
What
we
do
is
the
worker
so
just
wanted
to
put
that
thought
out
there.
H
Rosen,
so
there
are
at
least
two
things
that
I
know
of
that
need
to
come
in
once
again
their
implementation
experience.
What
I
would
like
to
see
personally
is
that
the
workgroup
stay
in
existence,
but
it
doesn't
have
meetings
because
I
don't
think
our
meetings
are
being
are
very
fruitful,
but
our
list
is
still
useful
having
people
who
really
are
only
occasionally
reviewing,
but
when
they
do,
we
get
some
useful
things
like
what
Keith
did
when
he
read
it.
H
Data
only
and
said:
look
privacy
matter
varies
across
countries,
and
some
of
this
may
not
work
in
some
countries.
You
need
to
say
that,
yes,
thank
you.
I
appreciate
him
doing
that,
so
I
would
like
to
keep
it
going.
I
mean
the
list
can
always
go,
but
I
think
that
having
the
work
group
itself
view
review-
and
things
like
that
is-
is
useful,
but
I
don't
think
we
need
to
meet
I.
Think
meeting
is
meetings.
D
F
Right
well,
I
guess.
The
other
thing
I
was
going
to
say
is
that
we
definitely
even
if
people
felt
that
their
they
just
want
to
get
the
list
and
not
the
group
if
you're,
if
you're
the
chairs
are
overburdened
and
that's
kind
of
the
difference
between
a
group
and
a
list
aesthetic
group
has
chairs,
we
would
keep
the
group
until
the
documents
are
in
the
RFC
editors
q.
Anyway,
that's
kind
of
the
what
we
tend
to
do,
because
in
IHG
evaluation
you
definitely
have
times.
F
We
need
to
go
back
to
the
working
group
so
and
clearly,
in
this
case,
in
in
the
our
theatres
q,
you
can
also
have
time
so
you
need
to
come
back
to
the
working
group.
So
what
as
well
keep
the
group
intact
until
that
time,
at
least,
if
not
longer
but
I'm,
totally
fine,
with
kind
of
keeping
it
in
a
dormant
state?
This
is
one
where
it
seems
like
you
can.
Imagine
things
coming
up
in
the
future
and
already
having
the
group
constituted,
will
be
useful.
I
John
Peterson,
so
Ryan
is
there,
like
other
places
where
those
people
are
who
were
actually
discussing.
This
are
gathering
in
greater
number
than
here.
That
just
is
a
consideration.
I
mean
I
and
I
know
that
there
are
obviously
this
is
different
in
different
countries
and
that
that's
a
dimension
in
this,
but
we
could
also
be
looking
at.
You
know
some
kind
of
a
transition
plan
right,
wear
whatever
procedural
responsibilities.
H
I
mean
Nina
is
the
obvious
place
to
do
it
and
they're
capable
of
doing
standards,
documents
and
we've
gotten
fairly,
decent
it
making
decent
documents.
So
you
know,
could
we
do
it?
Yeah
I'm,
not
sure.
That's
really
the
right
thing
to
do,
and
I
personally
appreciate
the
review,
the
occasional
review
that
we
get
from
people
who
have
wider
expertise
within
the
ITF,
so
I
don't
think.
That's
such
a
great
idea.
I
prefer
to
do
them
here
and
then
just
sort
of
drag
the
Nina
guys
into
the
ITF
process.
I.
I
D
Okay,
good
discussion:
I
just
wanted
to
plant
the
seed
in
everybody's
mind
that
point
out
the
obvious,
but
our
work
is
winding
down
them
in
you
know
if
you
have
any
thoughts
around
that.
Obviously
the
list
is
going
to
stay
open
and
we'll
decide
between
now
in
berlin.
If
we're
going
to
meet
in,
but
that's
not
to
chair
an
ad
discussion
at
why
so
any
other
comments
open
mic
come
up
and
tear
us
apart.
We
have
actually
but
two
hours
left.
If
we
want
to.