►
From YouTube: IETF-CBOR-20230628-1400
Description
CBOR meeting session at IETF
2023/06/28 1400
https://datatracker.ietf.org/meeting//proceedings/
A
A
Let
me
hit
the
agenda
here.
So
I
guess:
there's
since
I
put
the
time
tag
thing
on
the
agenda
after
starting
to
discuss
it
with
Karsten
we've
sorted
that
out.
It
was
already
decided
a
while
ago
and
it
just
sort
of
hung
so
Carson,
just
posted
a
couple
of
updates
to
to
time
tag
and
I
will
process
it
and
and
send
it
off
to
the
iesg.
A
A
B
Yeah
I
think
we
can
do
this
in
parallel
with
other
things.
We
need
to
do
before
submitting
right,
doing
the
write-up
and
so
on
so
I
think
it's
time
weight
spent
and
I
would
probably
do
it
to
make
it
a
one
week
call
because
no
there's
no
reason
to
have
a
two
weeks.
So
people
can
look
at
the
last
changes.
People
can
look
at
the
status
change,
and
so,
if
we
don't
get
anything
we
we
are
really
clean.
B
A
B
Yeah
pretty
much
what
what
the
agenda
says.
So
this
is
now
well
hung,
as
you
would
say,
about
the
stick
and
I
think
that
we
have
been
waiting
for
implementations
and
we
haven't
got
a
lot
of
feedback
from
from
implementers,
probably
because
there
are
still
things
that
might
change,
so
people
are
not
implementing
it
and
we
have
the
usual
vicious
cycle
here.
B
So
we
still
have
a
couple
of
things
to
do,
and
one
of
those
is
actually
implementing
what
we
said
in
the
discussions
that
we
would
simplify
the
table
set
up
even
more
so
we
have
a
stable
part
of
of
the
document
that
defines
the
the
table
reference
tags
and
we
have
part
we
have
been
fuzzing
around
a
bit
with
the
table
set
up
and
by
by
putting
one
table
set
up
tank
there.
That
makes
the
whole
thing
immediately
usable,
but
not
necessarily
exactly
what
you
want
for
a
specific
application.
B
I
think
we
would
be
doing
the
right
thing
so
I
think
further.
Simplifying
that
to
to
do
parallel,
setup
of
the
the
shared
an
argument
tables
would
would
make
it
easier
for
people
to
write
basic
Packers.
So
I
think
that
that
would
be
a
good
thing
to
do
and
I
hope
to
have
the
the
text
for
that
done
before
the
weekend
and
I
hope.
B
Also,
if
an
implementation
of
that,
so
that
there
is
the
sibo
packed
gem,
which
is
not
the
best
packer
in
the
world,
but
it
can
be
used
to
play
around
with
the
format
so
that
my
hope
would
be
to
to
have
a
document
that
we
can
look
at
in
the
implementation.
We
can
look
at
before
the
the
cutoff
and,
if
there's
any
last
minute
change,
we
can
do
even
that
now.
The
question
is
after
having
modified
the
document
a
lot
since
the
the
12
months
ago,
working
with
Roscoe.
A
A
B
Get
this
done
very
soon
and
then
do
the
other
two
documents,
I
think
mostly
in
sequence,
because
we
can
do
the
control
operators
pretty
independently
from
the
actual
module
structure.
So
I
think
we.
We
have
three
things
we
should
be
doing
and
we
can
do
the
the
Errata
plus
clarifications
stuff
very
quickly
and
the
other
two
probably
could
benefit
from
some
discussions
and
San,
Francisco
and
and
then
in
new
versions
of
the
documents.
B
These
are
two
functions
of
the
city,
AC
tool.
That
actually
can
be
very
helpful
when
writing
implementations,
but
they
don't
need
standardization
in
that
sense,
because
it's
kind
of
obvious
what
these
things
are
doing.
So
that's
another
thing:
I
would
like
to
get
feedback
from,
but
not
necessarily
for
for
changing
a
document.
A
A
C
A
B
Looks
good
to
me
so
on
the
double
question
mark
the
the
idiom
literals
actually
is
the
next
one.
After
effect
on
my
list
so
I
have
a
grammar.
I
still
have
to
translate
this
to
ABN
f,
but
since
I'm,
actually
writing
tools
that
help
me
in
translating
things
to
ABN,
f,
I,
think
that's
that
should
be
doable
pretty
quickly.
I
just
need
to
get
around
to
it,
and
that
will
most
certainly
happen
before
to
get
off.
B
C
I'd
like
to
briefly
bring
up
one
of
my
two
note
items
on
before
from
before
that
comes
back
to
pact,
for
which
I've
read
up
the
notes
and
I
think
it
hasn't
been
mentioned.
We
had
some
discussion
about
Seaboard
tax
standing
in
for
concrete
other
values,
which
packages
and
is
is
the
prime
example
of
but
I
failed
to
find
that
discussion
has
that
concluded?
Is
there
anything
more
that
we
that
we
want
or
need
to
do
on
this,
or
just
continue
discussion
on
this?
B
Yeah,
it's
probably
a
good
idea
to
write
this
up.
I
think
we
have
had
the
discussion
and
I
think
we
have
examined
the
the
space
between
pure
syntactic
validity,
that
that
is
what
what
70
49
did
and
and
totally
semantic
validity,
which
is
really
hard
to
get
interoperable,
and
we
we
have
decided
that
we
we
want
to
go
a
little
bit
beyond
pure
synthetic
validity,
but
keeping
in
mind
that
doing
too
much
of
this
can
can
be
detrimental
for
interability.
B
So
we
you
want
to
have
a
rather
limited
set
of
places
where
this
stand
in
mechanism
actually
is
operating
and
I.
Think
we
need
to
write
that
up.
I.
Think
it's
more
more
like
memo
to
our
future
selves
when
we
design
new
features
that
that
we
we
have
started
using
this
functionality,
I
mean
finishing
pack
kind
of
opens,
the
the
the
can.
C
B
I,
don't
think
we
will
finish
this
before
practice,
so
Peck
just
charges
ahead
and
is
then
the
example
that
we
will
use
in
that
other
documents
to
say
that
that
if
we
do
something
like
this,
it
will
be
infrastructural
specification
and
we
don't
want
every
single
tag,
definition
to
come
ahead
and
say:
hi,
I'm,
I'm,
also
kind
of
like
a
number,
so
I
can
be
used.
Every
a
number
can
be
used.
I
think
these
are
the
two
extremes
we
can
have
syntactic
purely
syntactic
validity
and
totally
opening
the
bundles.