►
From YouTube: CBOR WG Interim Meeting, 2020-06-03
Description
CBOR WG Interim Meeting, 2020-06-03
B
A
A
On
the
agenda
today
we
have
a
status
update
on
the
Seaboard
specification,
which
I
saw
got
an
update
like
30
minutes
ago,
but
I
haven't
had
time
to
see
that
update.
So
if
you
custom
can
walk
us
through
that'll
be
great,
then,
since
last
interim
Kirsten
will
talk
about
the
a
B
and
F
addition
to
see.
Ddl
I
had
the
link
here.
B
A
B
So
I
start
talking,
while
you
set
it
up,
so
we
had
discussed
some
redundancy
in
section
5.2,
which
I
think
addresses
the
issue
that
that's
out
there.
We
had
discussed
section,
seven
point,
one
which
wasn't
quite
in
the
right
tents
in
in
all
cases
and
I
slightly
rearranged.
The
text
there
and
I
use
the
occasion
to
change
the
discussion
on
extension,
points
to
actually
go
into
the
Ino
considerations
and
make
a
minor
change.
So
if
you
can
click
on
sides
changed
we
had
discussed
coming
up.
You
have
a
wide
version
of
that
I
know
we.
B
There
is
a
change
that
makes
the
range
from
256
to
half
of
16
bits
specification
required
and
the
rest
of
the
sixty
mid
range
to
the
32,
which
and
the
64
bit
range
I
continue
to
be
first-come,
first-served,
so
that's
something
that
we
discussed
but
apparently
never
got
around
to
actually
doing
so.
I
hope
this
is
still
something
that
the
runner
wants
to
do.
C
B
B
Tags
that
actually
can
receive
multiple
content
and
complicated
content
and
which
react
to
specific
major
types
in
that
content.
So
you
can
say
all
the
bytes
rings,
and
then
this
tree
actually
need
to
be
base64
encoded
when
converting
this
to
JSON
or
need
to
be
base64,
URL
encoded.
So
that's
for
tag
21
and
22
and
for
some
reason
we
try
to
do
the
same
thing
for
tag
number,
33
and
34.
So
the
the
read
takes
that
that
is
going
away
at
the
top
of
this.
You
know
in
the
middle
of
the
stream.
B
This
was
the
Edition
that
came
with
number
18
and
Lawrence
made
us
aware
that
that
this
actually
would
be
a
change,
larger
changed
and
we
thought
it
would
be
because
it
changes
the
the
domain
of
the
tags,
33
and
34,
and
it
gets
particularly
weird
for
other
things
like
UI
eyes
and
so
on.
So
I'm
not
sure
that
we
actually
thought
this
through
public
property
and
I'm
proposing
to
revert
this.
B
B
B
A
A
B
Yes,
so
I
misread
the
agenda
and
didn't
look
at
the
heading
and
only
at
the
draft,
so
there's
not
really
much
to
report
about
the
notable
tags
document,
so
I
didn't
prepare
much
so
I
have
I
now
have
a
mission
from
one
of
the
office
of
the
tight
specifications
to
include
his
text
there
and
I.
Think
I
will
do
that
and
then
take
another
round
asking
the
retinue
whether
this
is
a
good
way
to
handle
this.
But
again,
I
didn't
manage
to
do
this
in
in
the
last
10
days.
B
So,
on
the
a
B&F
essentially
same
thing,
I
have
one
slide:
I
showed
at
the
end
of
the
meeting
on
April
20,
something
that
shows
the
proposal.
I
haven't
really
received
any
feedback
on
this
proposal
and
one
reason
may
be
that
it
simply
is
pretty
good.
So
it's
hard
to
do
it
better
than
that,
but
it
may
also
be
that
people
haven't
really
looked
at
the
proposal.
B
So
please
do
it's
in
the
CBO
Freeza
document
and
if
you
have
any
any
additional
ideas
how
to
improve
this,
so
I
did
a
little
experiment
in
one
city
idea.
It's
back.
We
need
to
define
what
Korea
is,
which
is
a
form
of.
You
are
I
II
that
that
has
a
namespace
component
and
I
took
the
a
B
and
F
and
and
threw
it
into
an
ABN
f2,
reg,
X
converter
and
the
result
is
pretty
funny.
I
mean
it's
completely
unreadable
useless
as
something
you
would
want
in
a
specification.
B
C
I
sat
down
and
started
reading
this
yesterday.
Yeah
I've
got
two
comments:
I'm,
not
too
sure
where
they
go
comment
number
one.
It
looks
like
the
drafts
to
be
able
to
put
Unicode
characters.
Natively
in
a
B
and
F
has
died.
So
I'm
not
too
sure
that
we
need
to
worry
about
that
particular
issue
as
it's
just
a
problem,
if
you're
doing
a
f8
encoded
to
actually
read
it.
B
So,
that's
really
something
that
that
should
be
discussed
with
the
people
who
are
trying
to
maintain
a
B
and
F,
because
I
think
a
common
reading
of
a
B
and
F
is
that
a
B
and
F
doesn't
tell
you
what
Cosette
is
being
used.
It's
all
about
code
points,
so
you
actually
don't
need
a
document
that
tells
you
how
to
do
Unicode
in
a
BNF,
because
ABF
already
can
do
unicode.
C
B
C
B
B
E
E
B
Yeah,
so
there
is
something
called
a
Julian
date
which
is
unwieldy
because
it
starts
at
noon
and
has
an
epoch
date.
That
is
way
back.
So
at
some
point
people
decided
they
they
no
longer
wanted
to
use
Julian
dates.
They
wanted
to
use
modified
Julian
dates
by
removing
2.4
million
by
subtracting
2.4
million
point
five
from
the
Julian
dates.
Well,
the
0.5
is
gives
you
a
start
at
midnight,
and
the
2.4
million
makes
the
the
whole
thing
easier
to
handle
and
that's
because
they
wanted
to
have
a
round
number.
B
That's
why
they
came
up
with
this.
We
had
November
17th
thing,
I,
think
anybody
who's
doing
anything
in
astronomy,
time
keeping
and
so
on
knows
what
a
modified
Julian
date
is.
So
that
would
be
the
value
for
me
as
an
implementation.
Note,
not
not
as
another
way
of
trying
to
define
it,
but
as
an
implementation
note,
if
you
actually
have
an
mjd,
then
subtracting
40,000
578
from
it
gives
you,
the
eunuch
state.