►
From YouTube: 2020-06-15 meeting
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
C
A
D
C
D
D
C
C
A
A
C
C
E
Mean
I,
don't
really
have
much
to
contribute
to
it,
but
it
does
allow
a
like
renovate
bots,
which
were
using
in
the
Jas
repo
to
maintain
our
dependencies
and
ensure
they're,
up-to-date
and
stuff
like
that
yep,
where
previously
we
couldn't
use
bots
to
maintain
any
of
the
source,
so
that
was
I.
Think
I
was
the
one
that
was
really
pushing
for
this.
C
G
C
All
right
next
topic
might
be
long.
It
might
be
short.
Ga
planning,
I
have
an
update
on
this,
so
I
attended
the
metrics
thing
this
week
and
chatted
with
most
the
people.
There
I'm
sure
many
people
here
on
I
call.
It
seems
like
other
than
the
views
specification.
The
metrics
spec
is
pretty
much
done
and
they're
not
expecting
the
SIG's
to
be
blocked
by
other
thing
other
than
just
going
and
actually
implementing.
C
What's
the
spec
now
so
we
are
already,
if
not-
or
we
were
extremely
close,
if
not
already
unblocked,
for
most
SDK
work
to
proceed
so
that
people
can
start
the
the
implementations
of
the
features
that
are
required
for
GA
I
also
asked
if
there
any
additional,
like
sort
of
features,
functionality
implying
planned
for
GA,
and
the
answer
is
generally
no
I'll.
Add
a
link
here
to
the
actual
me
notes
where
we
went
into
more
detail
on
on
what
needs
to
be
done,
but
in
terms
of
SDKs
having
specs
work.
Blocking
them.
C
G
So
the
metric
specification
I
think
is
really
well
formed
from
the
API
level.
I
think
a
lot
of
work
has
gone
into
defining
instruments
and
making
sure
those
are
really
well
structured.
Right
now,
we've
identified
the
fact
that
there's
nothing
to
use
API,
that's
something
we
wanted
to
get
built
for
the
GA
release,
but
then,
after
that,
in
the
SDK
term,
meaning
like
a
language
sake,
I
think
that's
a
pretty
good
opportunity
for
us
to
implement
the
API,
the
SDK,
the
open,
so
interesting
is
below.
The
API,
though,
is
so
something
that
were
structuring,
I.
G
Think
some
of
the
optimal
pipelining
that
needs
to
get
put
into
place
in
I
think
a
more
specified
manner,
and
so
that
part
is
still
something
that's
a
work
in
progress.
We've
identified
quite
a
few
things
in
the
metric
specification
we
wanted
to
address
in
the
actual
SDK
design
around
configurability,
about
probability
of
additional
aggregators
in
the
specifier
in
the
SDK
itself,
and
then
transmission
of
particular
attributes
of
the
metrics
themselves
through
the
SDK
down
to
the
exporter,
which
we've
identified
is
really
critical.
G
If
we
want
to
do
things
like
pass
through
or
optimizations
of
the
exporters
or
the
backend,
that
being
said,
like
the
thing
that
Morgan
point
out,
though,
is
that
the
API
on
the
front
end
with
six
different
instruments
with
those
instruments
meaning
and
how
you
want
to
actually
design
those
from
the
API
level
should
be
really
solid.
At
this
point,
I
don't
think,
there's
any
desire
to
move.
G
There
was
a
an
open
issue
around
adding
another
industry,
but
I,
don't
think
that's
really
gonna
change
the
existing
sixth
one
and
I,
don't
I,
don't
that
last
mystery
was
actually
gonna
get
through.
It's
still
a
debatable
issue,
but
I
think
that
from
a
recommendation
flowing
which
maintainer
is
it's
a
to
go
for
making
sure
the
API
is
current
with
what
the
specification
isn't.
V05.
G
So
the
views
work
is
still
a
work
in
progress.
At
the
last
specification
meeting
there
was
a
I
think
waiting
actually
is
gonna
call.
He
would
said
that
he
was
probably
gonna
pick
up
the
work
again.
He
had
put
a
proof-of-concept
in
the
Python
world,
but
the
actual
work
for
the
specification
and
ironing
out
the
details
was
kind
of
put
on
hold
because
we
were
doing
a
lot
of
other
works,
including
the
API
work.
G
So
I
think,
if
that's
something
that's
going
to
be
started
up
again,
but
the
music
yet
is
still
a
work
in
progress.
It's
somewhat
parallel
to
the
metrics
API,
given
the
fact
that's
going
to
sit
on
the
side
and
it's
going
to
be
its
own,
an
API.
So
it's
not
something.
That's
gonna
block
you
from
appointing
the
six
instruments
currently
defined
in
the
specification.
I
know
the
Layton
wants
to
say
anything
more,
but
I.
Imagine
it's
still
just
an
early
stage
form
know.
E
E
C
So,
let's
talk
about
that
now,
so
we've
gone
over
in
the
last
few
weeks,
high-level
GA
requirements,
but
these
are
mostly
saying
like.
Oh,
we
need
to
have
like
a
certain
amount
of
test
coverage
or
SDKs
must
implement.
You
know
the
state
of
the
API.
It's
such
a
touch
time,
so
I
think
today
we
should
start
getting
very
crisp
about
what
we
want.
So
there's
there's
a
few
things.
I
think
that
we
need
to
do.
C
We
need
to
start
setting
actual
like
within
the
specs
groups
like
actual
sort
of
cut-offs
for
like
this
is
the
GA
version
of
the
spec
and
anything
sort
of
done.
In
addition
to
this
is
like
1.1
or
1.01
or
like
it's
sort
of
post
GA,
and
then
we
need
to
push
those
requirements
down
to
the
SDKs.
This
is
very
much
what
we
do
for
beta
right.
We
had
four
beta.
C
We've
started
that
process
for
metrics,
obviously
we're
not
very
crisp.
We
need
to
do
the
same.
For
tracing
which
we
can
do
tomorrow,
but
I
think
I
think
the
way
that
and
please
chime
in
people
if
you
think
this
is
the
wrong
way
to
do
it.
But
what
we
do
is
we
get
the
spec
groups
this
week
to
have
a
very
crisp
like
this
is
what's
in
version
1.0,
which
is
the
GA
version
of
the
spec,
and
then
next
week
we
go
to
all
of
the
actual
SDK
yeah
SDK
trypsin
actually
tell
them
like.
C
J
So
good
question
for
that,
just
like
I
remember
previously
moving
will
come
beta.
We
try
to
split
racing
in
the
metrics
API
were
saying:
oh,
it's,
okay,
to
call
it
beta,
but
for
metrics,
not
so
fortunate
in
the
spec
we
anticipate,
like
everything,
will
have
the
same
schedule
or
we
try
to
split
life.
Logging
is
not
part
of
that.
Yeah.
C
C
C
G
C
So
my
suspicion
is
that
the
second
will
be
more
useful,
or
at
least
more
realistic,
but
I'm
curious
with
everybody
on
those
call
things
right
like
I'm,
not
a
maintainer
of
one
of
these
SDKs
like
like
the
alternative,
is
we
just
set
a
date
and
say
like
come
hell
or
high
water,
we're
releasing
on
this
date?
I.
E
Mean
unless
we
have
some
like
say,
you
know
something
we're
trying
to
get
out
before
or
like
it's
just
some
salad,
a
reason
to
be
out
on
a
specific
date.
I
would
much
rather
go
the
other
way,
and
so
this
is
the
amount
of
work
we
have
to
do.
This
is
how
long
we
think
it
will
take
and
and
pick
a
date
based
on
that,
because
picking.
C
Way
does
work
for
beta
was
we
started
with
the
work
back
plan
where
we
defined
what
was
needed
for
beta
and
within
a
week
or
two
most
SDKs
had
pretty
good
estimates
for
when
they
could
realistically
hit
those
goals,
and
then
we
used
that
to
feed
into
when
we,
when
we
said
to
date,
I
suggest
that
we
set.
The
other
thing
we
can
do
is
like
set
a
target
date
for
like
the
first
release
candidate.
C
C
C
After
meeting
this
main
Aires,
we
will
set
a
target
date
for
our
c1,
because
I
suspect
after
the
first
release
candidate,
there
may
be
additional
work
that
we
haven't
accounted
for,
that
we
discover
it's
possible,
like
certainly
like
work,
that's
not
necessarily
like
done
on
the
core
SDKs,
but
like
testing
and
deployments
and
demo
projects,
and
all
that
stuff
and
I.
Think
frankly,
before
we
actually
get
the
first
release
candidate
out.
If
we're
we're
gonna
struggle
to
estimate
the
scope
of
that,
we
might
overestimate
it
would
make.
We
may
underestimate
it.
C
F
This
is
work
I've
been
doing
in
in
relation
to
the
stack
sick
in
order
to
at
least
categorize
the
issues
in
order
to
find
out
like
what
we,
what
we
got
left
to
do
and
for
the
next
milestone
trying
to
group
things
make
sense
that
way.
We
can
also
have
this
apply
towards
what
we
want
for
GA
I
had
a
pretty
productive
meeting
with
Carlos
and
Bogdan
to
sort
out
the
labeling
of
the
specification
repo,
which
I
think
is
a
step
towards
putting
some
context
around
it.
F
So
we've
settled
on
labels
that
cross
or
cross
one
dimension,
which
is
just
based
on
the
subdirectory
under
the
specification
repo.
So
we
got
contacts
correlation
contacts
logs.
There
is
no
subdirectory
for
logs,
but
that
was
there
in
order
to
categorize
issues.
So
that
way
we
know
what's
in
or
out
of
GA
metrics
resource
trace
and
protocol.
So
I
think
this
is
related.
F
Would
this
looks
like
this
is
firming
up
to
be
like
one
of
the
labels
common
labels
that
I'd
like
to
recommend
towards
the
language
six
to
make
sure
they
attract
the
things
for
implementing
spec
and
track
and
see
whether
they've
achieved
what's
needed
in
a
section
of
the
the
sake
I'd
like
to
run
this
idea
by
the
other
maintainer
x'.
This
is
sound
like
a
good
idea.
Is
it
sound
like
something
that
will
be
helpful?
Am
I
missing
anything
because
it's
just
the
subdirectory.
There
are
some
things.
It's
a
little
bit
left.
F
B
F
F
So
we
grouped
based
on
spec
section
and,
for
example,
under
this
trace,
you
can
see
like
out
of
131
issues.
64
are
tagged
with
trace
so
like
these
need
to
be
sorted.
If
we're
looking
to
like
lock
down
the
trace
part
of
the
specification
for
GA
but
stuff
on
there
logs,
we
can
just
like
yeah.
This
is
post
1.0,
don't
worry
about
it,
and
so
these
sections
was
based
on
specifically
the
sub
directories
under
open
to
entry
specification
pop
all
right
and
the
second
dimension
that
I
wanted
to
point
towards.
F
E
F
F
So
that
way,
it's
like
you
know
the
next
version,
0.6
spec
comes
out
and
there's
changes
to
the
trace
or
some
like
that-
or
this
is
the
one
for
RC
or
whatever
your
eyes
like
make
sure
that
you
are
tracking
the
changes
are
for
the
for
the
trace
part
of
the
spec
that
8
years
of
this
specification,
and
this
will
help
with
the
grouping
on
in
terms
of
has
it
been
implemented
to
the
spec
for
the
1.0.
F
I'm
trying
to
not
be
too
perspective
it
prescriptive
as
to
like
all
the
labels
are
geez.
You
gotta
definitely
have
all
these.
If
it
doesn't
make
sense,
you
know
you
don't
have
to
do
it
reduce
the
administrative
overhead
which
is
having
a
minimum
amount
that
just
makes
sense
to
tie
a
common
thread
across
all
that
the
six.
G
G
F
D
F
F
The
changelog
that
accurately
describes
what's
going
into
would
be
the
most
useful
thing
in
order
to
signal
to
the
rest
of,
in
order
to
pass
on
the
knowledge
to
the
rest
of
the
six
of
what
needs
to
be
implemented,
and
at
least
categorized
under
these
different
areas
that
way
the
language.
They
will
then
know
what
they
would
have
to
target
for
implementation.
F
F
C
I
C
I
I
I
C
C
J
C
A
C
A
A
J
J
C
I
think
that's
it
all
right.
I
will
take
the
action
on
setting
up
the
airline
sig
meetings
at
3
p.m.
and
7
a.m.
Pacific
and
I'll
be
joining
the
spec
meetings
this
week
to
get
them
to
commit
to
a
1.0
version,
commit
to
what
will
be
in
it
and
then
start
just
give
us
a
rough
schedule
from
what
they
think
they
can
hit.
That
I.