►
From YouTube: CBOR WG Interim Meeting, 2020-09-16
Description
CBOR WG Interim Meeting, 2020-09-16
A
I
posted
the
code
dmd
for
the
minute
taking
and
for
the
attendee
list,
which
I
should
have
filled
the
agenda
for
today
was
a
book
working
document
status
update,
so
a
gym
of
the
work,
the
changes
to
zebra
bass
at
the
top
of
the
agenda.
I
think
I
had
the
isd
review
comments
at
the
bottom,
but
I
think
we
can
take
that
first
then,
and
go
through
see
what
was
left
to
do
then,
captain,
if
you
have
an
update
on
on
the
tagspoi
document,
and
then
we
can
talk
just
for
information.
A
A
B
Yeah
so
paul
and
I
are
happily
generating
and
acting
on
pull
requests,
so
you
really
should
be
following
what's
going
on
in
in
github
right
now,
we
don't
think
we
touched
anything
contentious
yet,
but
please
do
follow.
What's
going
on
there
and
right
now
we
are
trying
to
address
the
discusses.
Most
of
the
discussions
were
about
our
use
of
c
language
or
pseudocode.
B
Protected
I
thing
that
I
talked
about
in
the
email
and
then
there
is
some
some
bad
text
for
an
example
that
I
think
we
can
simply
fix.
So
I
think
the
only
thing
that
really
needs
to
be
discussed
at
this
point
is
the
tag
35
ish
view
and
the
themes
we
don't
have
to
get
out
of
jp
card
here.
We
need
to
address
this
in
in
some
way
and
yeah.
My
may
was
an
attempt
to
talk
about
the
potential
approaches
we
have.
B
B
So
what
one
proposal
was
to
get
rid
of
it,
but
getting
rid
of
it
means
that
we
suddenly
have
is
often
registration
and
we
we
have
to
find
a
way
to
to
keep
this
registration
alive
without
keeping
1749
alive,
which
is
the
place
where
that
thing
is
defined
now.
One
way
I
didn't
write
up
that
we
could
do
this
is
to
use
the
notable
tags
instrument
as
the
future
reference
to
that.
B
A
B
Actually,
my
preference
would
be
that
we
somehow
get
it
out
of
the
way,
because
I
mean
it
is
registered,
it's
not
going
away
and
we
just
need
to
put
in
the
right
words
to
to
shape
the
expectations.
And
so
even
if
we
delete
it,
then
we
still
need
some
some
language.
That
explains
why
we
deleted
it
and
where
it
is
describe
now.
B
B
A
D
Yeah
they
could
just
say
that.
That's
that's
reason.
It's
more
the
case
of
the
subset.
Can
I
say:
well,
it's
the
common
subset
and
the
answer
may
be.
I
don't
know
what
my
users
are
going
to
do,
but
I
agree
if
someone
says
wants
to
pick
one
or
the
other,
then
they
do
that
and
I
replied
and
said
what,
if
you
make
a
copy
of
the
of
the
web
page
or
pointed
archive.org
of
it
as
a
stable
reference
for.
A
D
I
don't
know
if
there's
benefit
from
I'm
sorry,
I
don't
know
the
ecma
reference
very
well.
I
know
pcre
quite
well.
I
am
a
little
bit
shocked
to
realize
how
poorly
specified
it
is,
and
I
kind
of
like,
oh
okay,
so
I
don't
think
we
can
specify
this.
I.
I
don't
think
that
our
our
our
abyss
can
specify
the
common
subset
easily.
D
D
B
B
Version
from
2011,
I
don't
even
know
what
what
regex
subset
is
in
there.
It's
definitely
a
much
worse
than
later
included
in
ecmascript
2015,
which
is
again
smaller
than
the
one
in
2018,
which
is
almost
good
enough.
That
you
could
stop
using
the
reference
to
pcre
and
say
just
do
an
eggmaster
2018
and
you
will
be
fine.
B
Has
been
changing
all
the
time
as
well,
so
there
is
actually
a
third
column
here
which
is
per
5.8
versus
5.10,
and
I
think
the
the
ecmascript
people
have
just
tried
to
make
ekma
regular
expressions
more
useful.
Oh
there's
one
one
little
twist
here,
which
is
that
ekma.
Regular
expressions
are
only
really
useful
if
you
step
unicode
flag,
but
there
is
no
way
to
set
the
unique
flag
with
1035.
B
B
D
I've
never
encoded.
I've
never
used
tag
45.
I
have
no
idea
when
I
would
use
tag
45,
so
I
don't
really
know
care
if
there's
a
change,
but
I
I
would
prefer
that
the
document
you
know
got
got
approved
without
hassle,
so
so
so
anything
that
makes
that
happen
makes
me
happy,
but
I
don't
know
it
could
be
that
it
could
be
that
the
the
set
of
people
that
that
this
really
affects
is
is
is
empty
and
just
specifying
something
will
make
people
happy
or
I
could
be
wrong.
I
have
no
idea.
A
B
B
A
B
B
B
But
for
the
empty
string
the
notation
is
ambiguous
because,
if
you
just
say
open
parenthesis
underscore
closing
parenthesis,
you
never
say
said
whether
this
is
a
byte
string
or
a
text
string,
so
that
that
was
the
observation,
and
I
made
the
observation
that
we
already
have
a
single
quote.
Single
quote:
underscore
double
quote
underscore
as
notations
for
exactly
these
special
cases,
and
we
might
just
start
using
that.
B
But
yeah
there
are
some
people
who
would
like
to
modify
the
syntax
for
a
parenthesis
underscore
and
allow
some
indication
on
whether
this
is
a
byte
or
a
text
string
there,
for
instance
by
adding
a
b
or
a
t,
and
I'm
a
little
bit
averse
to
that
change.
I'd
rather
keep
things
compatible
as
long
as
you
don't
hit
the
one
case
that
we
don't
cover
properly
at
the
moment.
B
B
A
A
B
A
B
Well
there
there
are
lots
of
comments
from
ben
and-
and
some
are
very
good
at
so
for
instance,
that
he
promises
that
we
don't
day
when
we
convert
json
to
a
subor.
What
rounding
mechanism
for
the
photic
book
numbers-
and
I
mean
yeah,
obviously
and
there's
also
an
obvious
answer-
we
do
it
the
same
way.
Javascript
does
it
and
so
this?
This
is
a
technical
clarification
from
my
point
of
view
and
there
are
half
a
dozen
of
those
and
we
are
right
now
writing
these
up
and
generating
pull
requests.
B
B
B
A
B
A
A
C
B
Yeah,
that's
definitely
something
that
that
I
need
to
pick
up,
so
I
I
need
to
write
up
my
my
proposal.
How
to
do
this
and
brandon
had,
I
think,
was
brandon
had
some
some
brilliant
way
of
doing
it
as
well,
so
the
I
want
to
finish
them.
Definitely
this
at
the
moment-
and
I
I
have
limited
time
and
I
want
to
invest
everything
I
have
into
that,
so
that
will
probably
only
come
after
70
49
as
approved.
B
So
I'm
I'm
diagnosing
this
stage
of
the
suit
working
group
as
they
couldn't
really
wait
for
us
to
finish
packing
anyway,
so
we
we,
we
are
not
in
a
particular
rush
for
finishing
packing.
Of
course
we
don't
want
to
wait
another
seven
years
with
that,
so
yeah,
I
think
doing
70
49
is
first,
is
the
right
thing
to
do.