►
From YouTube: IETF-ASDF-20230927-1540
Description
ASDF interim meeting session
2023/09/27 1540
https://datatracker.ietf.org/group/asdf/meetings/
A
B
B
A
C
C
C
C
C
A
Yeah,
apparently,
if
you
specify
a
city
when
you
ask
for
a
virtual
interim
now,
it
thinks
you
actually
meant
an
it
in
person,
and
my
impression
from
the
web
interface
is
that
you
need
to
specify
you.
You
are
encouraged
to
specify
the
city
so
that
it
will
pick
a
time
zone
for
you,
but
that's
apparently
not
the
case
okay.
So
we
have
Nicholas
here
nice
to
have
Coster.
D
A
A
A
Okay,
well,
let's
start
and
perhaps
we'll
get
my
costers
top
of
the
hour
or
something
okay.
So
welcome
to
this
virtual
interim
meeting
for
the
ASDF
group
and
I
guess,
I'll
say
that
we
did
ask
for
a
time
slot
for
ietf
118
in
Prague
and
the
end,
and
but
this
meeting
is
really
about
today
about
getting
the
current
document
ready
for
to
be
published.
We
went
through
working
group
last
call,
let
me
I
don't
want
to
get
ahead
of
myself.
A
All
right,
well,
I,
will
show
the
obligatory
note,
well
just
to
remind
you
that
there
are
certain
processes
of
the
ietf,
and
this
is
an
official
meeting.
And
so
the
note
will
applies
particularly
relating
to
Ill
intellectual
property
and
that
this
meeting
is
recorded
and
that
we
ask
you
to
act
professionally
and
be
nice
to
each
other.
And
there
are
a
number
of
BCPS
here.
And
if
you
have
any
questions,
you
can
talk
to
work
in
group
chairs
or
to
the
area
directors.
A
And
that's
about
it.
I
would
suggest
that
you
just
click
on
the
microphone
icon
in
the
upper
left
to
request
microphone
status
and
then
mute
yourself.
The
microphone
in
the
lower
right,
which
is
about
10
times
faster
for
muting,
on
a
muting
and
we're
a
very
small
group
and
I,
don't
think
we
need
to
observe
the
Mike
queuing
line.
For
that
reason,
any
other
questions
or
concerns
I,
probably
turn
my
video
off
just
to
avoid.
A
A
Following
your
slides,
yeah,
okay
well-
and
it
finished
last
weekish
or
10
days
ago,
and
we
got
a
couple
of
reviews
and
I
guess
I'll
pass
it
on
to
Karsten,
it's
all
in
his
slides.
So
let
me
see
here,
stop
screen,
share
and
share
preloaded
slides
share
and
then
I
can
give
him
the
control
by
clicking
on
that.
B
F
F
So
the
dash
15
was
submitted
on
on
September
4th
before
I
went
on
vacation
and
the
working
class
12
went
until
the
20th
and
we
got
one
review,
but
a
very
good
one
by
Crystal
heinberg
who
actually
took
the
position
the
perspective
of
a
new
reader.
So
so
he
kind
of
read
this
forgetting
everything
he
knew
about
SCF
and
and
tried
to
find
out
whether
the
document
is
good
enough
and
he
found
quite
a
few
things
that
that
can
get
better.
F
So
we
set
up
a
number
of
issues
that
are
on
the
GitHub
repository.
If
I
want
to
link
that-
and
we
also
had
a
number
of
pull
requests,
we
still
have
three
issues:
the
blue
ones-
that
that
don't
have
pull
requests
yet,
but
it
should
be
possible
to
get
pull
requests
out
there
during
the
rest
of
the
week.
F
F
F
Only
when
we
talk
about
SDF
objects,
so
that's
an
unfortunate
Collision
at
the
time
we
we
started
using
this
terminology,
we
didn't
have
Json
schema
org
object,
so
this
this
was
still
a
little
bit
less
of
a
problem,
but
now
we
have
the
problem
full
on
and
so
pull
request.
119
tries
to
make
sure
that
we
properly
separate
the
Json
objects
by
calling
them
maps
and
the
SDF
objects
and
have
the
language
in
place.
That
explains
all
this,
so
this
was
109
or
per
request.
F
119
112
for
request
121
was
about
the
same
thing
with
the
word
Thing
versus
SDF
thing,
except
they
are
different
in
this
case.
So
SDF
object
is
the
same
thing
as
an
object,
but
an
SDF
thing
is
not
a
thing,
so
we
were
using
the
term
thing
as
as
a
general
term
for
the
kind
of
system
that
this
specification
is
trying
to
to
help
you
model,
so
we
hi
Michael.
F
F
So
we
are
not
no
longer
saying
that
this
is
for
iot,
but
we
are
just
saying
it's
it's
for
things
and
yeah.
It's
a
bit
unfortunate
that
we
call
the
aggregate
structure
SDF
thing,
but
that's
where
we
are.
We
probably
don't
want
to
change
this
at
this
point
and
we
also
remove
text
that
actually
Scouts
around
the
issue
of
thickness.
F
So
that
was
that,
and
finally,
we
had
this
unfortunate
term
SDF
file
and
we
we
don't
really
want
to
talk
about
file
systems
and
things
like
that.
So
we
we
replaced
this
by
SDF
document
throughout
and
we
we
still
have
one
issue
open,
that
they
have
to
add
this
to
the
terminology
section
as
well,
so
that
that
was
a
significant
set
of
improvements
based
on
Crystal's
comments.
F
We
added
a
little
bit
of
missing
detail
that
that
Crystal
noticed
so
in
the
definition
of
an
action
we
didn't
have
the
the
a
pointer
to
the
idiom
of
returning
an
action
resource
for
for
an
ongoing
action,
so
this
is
now
India
without
actually
binding
it
to
a
co-op
where
it's
the
the
usual
thing
to
do
so.
This
is
described
as
a
more
General
idiom
and
it
also
points
to
the
SDF
type
link
document.
F
As
an
example
of
how
SDF
base
base
ICF
has
to
be
extended
to
be
able
to
fully
describe
these
Action
Resources
so
right
now
we
can
only
say
that
there
should
be
or
that
the
interaction
returns
and
an
action
resource,
but
we
cannot
say
right
now
what
what
that
action
resource
is
going
to
have
for
affordances
and
and
with
something
like
SDF
type
link.
We
can
do
that,
but
this
is
outside
the
base.
Spec.
F
So,
first
of
all,
we
have
a
name
for
the
state
of
evolution
that
that
is
crystallized
in
the
base
RFC.
We
now
just
call
this
base
SDF,
and
then
we
talk
about
extensions
that
can
be
made
either
through
the
extension
points
that
base
SDF
already
provides
in
the
form
of
Registries,
where
you
can
can
Define
additional
functionality
or
by
actually
doing
another
document
that
may
even
use
the
the
feature
quality
from
the
info
block.
F
And
finally,
the
the
definition
of
nepalience
assumes
that
you
have
internalized
that
definition
from
what's
his
name,
Norman,
that
the
guy
who
defined
authorities
in
in
human
computer
interaction-
and
this
is
now
a
little
bit
more
explicit.
So
so
it
will
be
less
opaque
for
people
who
are
not
used
to
this
technology.
F
F
F
We
still
have
two
trials
that
have
to
be
done.
One
is
actually,
there
is
a
pretty
lacks
use
of
the
term
instance,
and
we
have
to
get
rid
of
that
not
of
the
term
necessarily
but
of
the
lexness
in
using
it,
and
time
was
not
sufficient
to
do
that.
That's
actually
a
significant
surgery
that
Tom
has
used
about
20
times
in
in
the
document,
and
we
actually
may
have
to
invent
a
couple
of
mod
terms
to
to
make
this
more
precise.
F
F
This
has
all
remnants
I'm,
not
entirely
sure.
We
won't
occasionally
say
something
like
earlier
versions
of
earlier
drafts
of
this
specification
did
this,
because
it
maybe
is
sometimes
necessary
to
understand
that
we
are.
We
have
chosen
not
to
do
something
that
we
we
did
earlier,
but
I
think
that
the
general
objective
here
should
be
to
get
rid
of
SDF,
1.0
and
1.1
and
I.
Think
that's
something
that
we
as
a
working
group
should
discuss.
Is
that
something
we
want
to
do
now.
A
So
in
a
bunch
of
things
in
one
of
the
pull
requests
that
I
just
happened
to
browse
116,
I
guess
or
it's
a
pull
request
125.,
it
says
a
bunch
of
things
where
it
used
to
say
an
SDF
1.0.
It
now
says
an
SDF
draft
1.0.
It
says
that
repeatedly
and
so
that's
the
kind
of
thing
you're
talking
about
that
would
go
away
or
yes,
so
we
wouldn't
have
a
version
number
right
in
the
future.
We
would
refer
to
it
as
just
SDF
RFC
blah
blah
blah.
F
C
F
E
A
A
C
A
I
think
it'll
be
a
change,
makes
a
lot
of
sense.
I
think
it'll
be
confusing
for
reviewers.
If
you
know
we
talk
about
these
previous
versions
that
they
don't
really
see
anymore,
so
I
think
that's
very
valuable,
probably
need
to
have
a
one
sentence
somewhere
that
says
you
know.
Drafts
talked
about
1.0
1.1,
but
we
do
not
use.
But
from
this
point
on,
it's
simply
called
base
SDF
and
enough
said
right,
but.
F
E
F
E
C
And
technically
the
new
features
quality
should
take
away
the
usual
reasons
to
even
make
new
versions
and
of
course,
one
thing.
What
we
may
want
to
be
careful.
If
we
decide
to
do
a
feature
that
lumps
together
and
a
set
of
other
features,
it
will
be
very
tempting
to
call
that
a
person.
Maybe
at
that
time
you
want
to
use
a
different
word
for
it
to
avoid
that
confusing
or
decide
version,
it's
the
right
word,
but
we
don't
need
to
use
this
classical
versioning
schemes.
No.
F
B
There,
okay,
it
took
a
little
while
to
unmute
yeah
somewhat
jokingly.
We
could.
We
could
call
that
a
profile
but
I
think
what
we
could
explain:
the
the
change
management
philosophy,
but
in
a
few
sentences
to
help
people
understand.
You
know
why
and
what's
happening,
but
a
feature
that
you
know
a
system
that
uses
a
particular
set
of
features
isn't
necessarily
compatible
with
the
system
that
uses
a
different
instead
of
features.
But
the
features
being
clearly
identified
and
enumerated
gives
a
developer
a
chance
to
get
whatever
level
of
compatibility.
B
So
I
guess
yeah
in
the
sense
that
maybe
they
are
sort
of
profile,
but
we're
probably
not
going
to
name
them,
but
it'd
be
nice
to
just
have
it
always
to
be
just
a
list
of
features
but
yeah
I
think
that's
I,
think
that's
what
we
could
do
is
probably
a
few
sentences
to
explain.
Maybe
some
of
the
reasoning
behind
you
know
why
we're
doing
this.
This
way.
F
F
F
Submitter-16
on
that,
and
then
Ari
pointed
out
that
now
that
we
we
have
a
much
more
readable
version
of
this,
maybe
you
actually
want
to
ask
a
few
people
from
the
ecosystems
to
put
in
additional
review.
So
this
this
is
a
little
bit
hard
to
control
with
respect
to
the
the
time
it
needs,
but
it
may
be
worthwhile
to
actually
spend
this
additional
couple
of
weeks
or
so
just
to
to
get
more
of
these
new
reader
perspectives
in
there,
which
turned
out
to
be
so
useful
with
Crystal.
F
E
C
And,
of
course,
we
shouldn't
feel
the
dependence
there
that
like
if
we
don't
get
in
reasonable
guide,
we
just
go
forward,
but
I
think
it
would
be
great
to
get
that.
Of
course,
the
challenge
there
is
the
logistics,
because
many
of
those
don't
have
you
know,
data
tracker
accounts
and
and
are
not
signed.
B
C
E
F
E
E
A
A
E
E
F
F
F
Shouldn't
do
a
submission
on
the
23rd,
because
that
that's
always
hectic
and
error
prone.
So
we
should
do
that
a
little
bit
earlier.
But
so,
if
we
say
second
to
12
is
the
additional
external
review
period,
we
should
try
to
submit
it
by
the
19th
and
yeah
I'm
going
to
be
offline
for
the
next
several
of
days
anyway.
B
F
F
A
B
C
F
A
A
F
B
E
F
A
F
A
Yeah,
that's.
The
idea
is
I
think
to
come
up
to
to
have
some
proposals
for
each
of
the
documents
that
there,
and
so
the
idea
is
that
we
would
come
into
the
ITF
meeting
with
some
actual
proposals
for
what
would
be
in
the
charter
and
what
would
out
not
be
in
the
charter
or
some
actual.
You
know
conflict
of
what
is
going
on
right.
A
C
F
F
E
You
know
for
folks
with
more
experience
with
these
three
Charters
and
so
on.
Would
those
you
know
this
amount
of
work,
be
you
know
too
little
too
much,
not
appropriate
I
mean
how
compared
to
other
researchers.
You
know
it's
obviously
some
work
to
be
done,
but
it's
not
an
infinite
amount,
not
a
huge
amount,
or
can
we
do
it
with
the
current
Charter,
with
some
creative
reading.
A
Of
it,
I
I
would
say
that
it
doesn't
really
matter
that
it's
never
I
mean,
maybe
it
all
fits
in
the
current
Charter,
but
that's
something
we
would
have
to
to
look
at,
but
I
think
it's
at
the
point
where,
where
having
done
this
document,
the
next
question
would
be
that
our
ad
would
ask
us:
okay,
you're
done
you're
shutting
down
so,
regardless
of
whether
it
fits
in
the
current
Charter
or
not.
We
have
to
make
a
decision
that
we're
continuing
okay.
D
A
That's
a
conscious
decision
and
our
80
has
to
agree
with
us
that
we're
continuing
and
if
it
requires
some
minor,
Charter
retardering
text,
then
that's
probably
not
a
big
deal,
although
it
requires
an
approval
process,
so
it
takes
an
extra
month,
but
wow
I
think
it's
the
conversation
about.
What
are
we
doing
if
that
matters.
A
A
F
F
D
F
A
C
Yeah
I'm
not
wondering
what
will
be
the
effective
way
of
doing
it.
I
guess
maybe
they
made
the
most
likely
outcome
might
actually
be
that
they.
You
know
whatever
organization
is
the
old
specifications
but
having
a
way
oh
way
to
have
public
reference
for
those
qualities,
I
mean
it's
very
useful
and
perhaps
an
organization
that
doesn't
that
puts
all
of
their
domains
behind
a
paywall
for
them
by
doing
an
RFC
X,
it
might
be
better
option
and
would
be
sad
if
our
Charter
prevents
that
I
guess.
That's
kind
of
my
thinking
here.
F
Well,
that
would
normally
be
an
independent
submission,
so
if
Oma
wants
to
have
full
change
control
over
those
which
I
think
is
what
they
would
want
to
do,
then
yeah
they
would
submit
this
as
an
RFC.
If
things
go
well,
but
they
may
want
to
use
the
IC
and
not
this
working
group
I
mean
we
sure
would
like
to
discuss
it
in
this
working
group,
but
in
the
end
Oma
would
need
to
have
Change
Control.
C
C
C
A
A
A
You
know
the
whole
hour
shouldn't
20
minutes,
so
you
know
two
or
three
slides
at
most
on
each
but
I
I
think
we
also
need
to
make
sure
that
we
are
figuring
out
what
other
people
we
probably
need
to
have
a
hand
minute
presentation
on
SDF
base
SDF
or
is
it
SDF
based
I,
don't
know
anymore
base
SDF.
We
probably
need
to
have
a
10
minute
presentation
on
that,
but
maybe
we
don't
need
to
do
that
at
the
beginning
of
the
meeting.
Maybe
we
can
do
that.
A
You
know
near
the
end
of
the
meeting
so
that
we
can
get
the
real
business
of
the
chartering
out
over
but
I
mean
I,
don't
know
it
depends
on.
We
expect
people
to
read
the
document
before
they
get
there
a
question
of
how
many
other
interested
new
parties
do
we
get
given
that
we
also
should
probably
make
sure
the
agenda
is
posted
well
in
advance,
so
people
know
what's
going
on
and
then
you
know
what
let
me
ask
the
other
question.