►
From YouTube: 2021-08-30 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
B
B
A
Yeah
it
sounds
like
there's.
There
was
some
some
noise
coming
from
your
mic,
maybe
rubbing
against
a
shirt
or
something.
B
All
right,
let's
get
started
with
three
minutes
into
it.
Folks
could,
just
you
know,
add
in
to
the
agenda
doc
of
their
attendance
and
I'll
just
get
started
through
my
items,
but
I
think
today
it's
really
more
of
a
kickoff
of
anything
of
like
hey.
We
have
this
instrumentation
sig
going.
We
need
to
make
sure
that
we're
getting
the
right
owners,
we
have
the
right
coverage,
we're
creating
the
right
backlog,
and
so
first
thing
that
I
just
want
to
bring
to
everyone's
attention.
B
Is
the
existing
hotel,
instrumentation
dock
that
we,
I
think,
ted
you
had
created?
So
we
could
definitely
reference
that
and
how
to
propel
this
going
forward.
B
Second,
is
we
had
the
semantic
conventions
process
proposal
doc
that
folks
had
contributed
to,
but
I
think
today
the
main
thing
is
what
we
really
need
to
get
towards
is
we
we
do
know,
there's
an
existing
like
otep
for
http,
and
there
are
some
folks
contributing
messaging
has
johannes.
You
know
leading
that
effort.
We
still
don't
have
coverage
in
other
areas,
and
we
want
this
to
be
the
kind
of
main
form
of
like
hey
for
these
general
topics.
B
How
can
we
do
the
q
a
and
really
progress
kind
of
going
forward?
And
so
you
know
ted
I'll
kind
of
look
to
you
and
other
community
leaders
how
things
have
happened
in
the
past?
But
essentially
I
just
want
to
try
to
help
push
things
forward,
make
sure
we
have
the
proper
road
map
and
we're
able
to
get
that
progression
that
we
want.
You
know
for
the
rest
of
the
year.
A
Awesome
and
you
all
have
been
released
michael,
I
know
at
least
some
people
have
been
meeting
already.
It
would
be
great
to
get
maybe
brought
up
to
speed
about
where
those
conversations
been
going.
B
Yeah
absolutely
so,
let
me
first
start
with
the
process.
Doc
make
sure
I
get
this
bigger.
B
Okay
for
us.
A
lot
of
the
conversations
really
was
around
this
document
in
terms
of
what
our
objectives
and
what
the
themes
are
going
to
be,
and
we
really
started
to
look
at
all
right.
We
know
we
want
to
start
with
specific
categories
here.
First
in
terms
of
trying
to
get
the
semantic
convention
stable
within
certain
areas.
Most
of
it
has
been
around
messaging
because
that's
where
johannes
has
started-
and
we
do
have
http
discussions
going
on
too,
but
that
hasn't
been
in
a
broad
audience.
B
It's
mostly
been
around
this
stock
messaging,
and
how
can
we
identify
the
right
people
and
so
right
now
there
has
been
some
interest,
but
we
haven't
had
enough.
So
I
would
say
most
of
the
discussions
have
been
pretty
generic.
Nothing
very
specific.
On
the
messaging
side,
though,
I
think
johannes
could
dig
in
deeper
on
how
those
conversations
internally
have
been
going
with
alolita
and
folks
and
where
kind
of
next
steps
resides
there.
D
Yes,
I
mean
with
messaging:
there
have
only
yet
been
like
microsoft,
internal
discussions
about
what
we
want
to
have
or
what
we
would
need
from
the
messaging
spec
and
the
other.
Next
steps
will
be
on
thursday,
the
first
basically
official,
open
telemetry
meeting
for
the
message
messaging
stuff
and
how
I
intend
to
go
about
it
like
I
prepared,
like
an
old
tap
just
defining
the
scenarios
in
the
roadmap
for
the
messaging
semantic
conventions,
because
yeah,
I
think
that
is
true
for
messaging
and
also
for
or
other
semantic
conventions.
Actually
is.
D
We
need
like
a
definition
of
what
does
table
mean
for
us?
What
scenarios
do
we
want
to
cover?
I,
I
think
there
is
no.
There
is
no
real.
Like
agreement,
I
wouldn't
even
say
agreement.
I
think
if
there's
no,
no
definition
around
that,
so
I
think
before
we
actually
start
diving
into
the
technical
work,
we
need
kind
of
to
agree.
What
is
the
goal?
D
What
are
those
scenarios
we
want
to
cover
and
that
will
basically
be
the
scope
of
this
first
out
tap
that
we
just
kind
of
define
like
the
scenarios
that
we
need
to
cover
and
then,
when
we
cover
those
scenarios
we
can
actually
say:
okay,
we
are
stable
now
and,
and
we
can,
we
can
move
on.
D
So
that
is
that
is
basically
a
first
step
working
on
this
old
tab,
which
is
less
a
technical
discussion,
but
more
like
the
working
out
the
scenarios
and
requirements,
and
then,
after
that,
the
idea
is
to
work
on
a
second
old
tap
in
the
group
that
basically
dania
defines
the
technical
details
and
then
and
then
we
will
go
into
the
actual
kind
of
spec
work.
D
A
D
And
what
we
basically
also
need
for
that
I
mean
for
getting
started.
I
put
it
here
as
a
last
point
on
the
list.
Also.
I
think
that
is
important
for
the
definition
of
done.
There's
in
this
version,
stability
document
for
the
semantic
conventions
there's
still
like
a
to-do
placeholder
in
there,
so
we,
I
think
we
need
to
we
need
to.
We
need
to
work
that
out
and
I
think
actually,
this
group
here
would
be
kind
of
the
ideal
forum
I
think,
to
work
on
that.
D
A
So
there's
really
not
much
there
currently,
except
you
know
to
say
that
we
don't
have
any.
We
haven't
defined
it.
So
that's,
basically
all
all
it
says
in
the
spec.
Currently,
however,
tigran
did
create
a
proposal
for
versioning
semantic
attributes.
A
I'm
not
sure
where
this
is
landed
in
the
spec.
Currently
josh.
Do
you
happen
to
know
whether
this
is
being
utilized
yet
you're?
Talking
about
the
schema
version,
yeah
stuff.
E
Yeah
yeah
so
basically
right
now
there
you're
only
allowed
to
rename
attributes
and
we,
you
have
to
fill
out
a
schema
url.
E
We
haven't
allowed
any
non-additive
changes
to
the
schema
so
far
outside
of
that
one
broken
release,
so
it
hasn't
actually
gotten
used
outside
of
just
denoting
that
there
are
no
renames
so
far
in
semantic
conventions.
So
it's
being
used.
The
plan
is
to
enforce
it
anytime.
There
is
an
actual
change
and
I
believe
it's
in
it's
in
the
it's
in
the
schema
url
section
of
the
spec,
so
it
is
in
the
spec.
But
it's
not
the
convention.
There's
really
a
I'll
find
a
link.
A
So
I
think,
yeah
part
part
of.
I
think
this
is
the
main
aspect,
johannes
about
figuring
out
what
stability
means.
A
So
the
first
I
think
the
first
pass
there
would
be
to
have
a
look
at
at
the
schema
versioning,
that's
already
put
in
and
try
to
put
it
through
its
paces.
I
I
suspect
we
are
going
to
come
up
with
a
number
of
edge
cases,
for
example,
end
users
needing
end
users
having
instrumentation
of
various
ages,
so
they're
producing
multiple
different
semantic
conventions,
for
example,
and
then
having
needs
having
to
define.
A
Their
dashboards
probably
being
skewed
across
several
different
versions
of
semantic
conventions
in
different
places.
I'm
not
sure
if
we've
addressed
that
right
now,
it
seems
like
we
just
have
a
single
schema
version.
That
kind
of
perhaps
assumes
all
of
the
data
is
at
one
schema
version
and
the
end
user
is
consuming
that
data
at
another
single
schema
version.
That's
probably
not
going
to
reflect
the
complexity
that
reality
is
gonna
have.
E
This,
the
schema
version
can
be
encoded
at
more
than
one
level,
so
schema
version
schema
url
is
the
only
thing.
That's
been
encoded
through
the
throughout
the
api,
but
it's
basically
tied
with
instrumentation
library.
So
at
the
instrumentation
library
level
of
otlp,
you
should
have
a
schema
url,
which
denotes
all
the
schema
underneath
that
component.
E
So
for
a
specific
tracer
for
specific
meter
or
specific,
I
guess
logger,
I
don't
know
what
it
is
in
login
yet,
but
whatever
the
equivalent
is
there,
you
should
have
a
url
that
you
can
use
to
validate
your
semantic
conventions.
It
looks
like
the
definition
of
the
schema
file
is
not
in
the
specification.
Just
the
fact
that
it
exists
is
validated.
A
Well,
it
should
get
added
to
the
specification
yeah
and
there's
like
a
practical
matter
of
actually
building
out
the
an
implementation
of
this
in
the
collector.
I
don't
believe,
there's
any
implementation
in
the
collector
as
of
now
so
that's
kind
of,
and
that's
where
we
will
have
to
figure
out.
I
think
some
of
the
details
that
are
not
in
in
this
otep
around
the
configuration
language
for
this
stuff.
A
So
how
do
you
actually
write?
How
do
you
actually
generate
rewrite
rules
and
then
actually
getting
getting
the
schema,
versioning
stuff
baked
into
a
bunch
of
existing
instrumentation,
and
that
would
get
us
to
kind
of
like
the
first
version
of
okay?
We
now
have
versionable
conventions.
We
have
instrumentation
that
is
generating,
including
a
schema
version
with
it,
and
then
we
have,
in
the
collector
a
schema
version
processor
that
can
make
choices
around
that.
So
it's
like
a
practical
matter.
A
Getting
those
two
pieces
built
are
probably
the
next
concrete
steps
towards
getting
stability
in
general
into
the
semantic
conventions.
D
E
D
I
mean,
I
think,
especially
for
messaging.
I
think
we
they
are
so
all
kind
of.
We
also
need
to
consider
changes
in
the
structure
of
how
spans
are
created,
especially
for
messaging
there's
like
links
put
together.
I
think
kato
needs
to
be
a
that
needs
to
be
then
somehow
voted
separately.
There.
F
A
That's
that's
a
good
point.
It's
not
necessarily
just
changes,
it
could
be
changes
in
which
attributes
show
up
changes
in
the
values
of
those
attributes
and
it
could
be
changes
in
the
actual
structure
of
the
data
so
which
is
a
bit
trickier
to
handle
that
last
last
requirement.
E
Yeah
the
quick
question
here
I
opened
two
oteps
around
metrics
and
semantic
convention
stability
just
because
well
I
think
traces
are
a
much
simpler
problem.
Yeah
and
metrics,
unfortunately,
are
not
in
all
sorts
of
fun,
painful
ways.
E
So
there's
so
you
know,
there's
a
there's,
there's
two
bugs
and
sorry
two
others
and
a
bug
which
I
think
outlined
kind
of
the
problem
that
we
will
have
with
semantic
invention
and
metrics
right,
specifically
time
series,
identity
and
metrics
changes
when
you
change
things,
and
so,
if
you
have
a,
if
you
have
someone
reporting
a
metric
in
a
previous
semantic
convention,
schema
version
and
then
someone
else
is
a
new
one.
E
It
gets
really
confusing
when
and
how
you
merge
those
things
together
and
whether
or
not
underlying
systems
get
to
see
it.
So
I
think
there's
a
if
I
can
make
a
recommendation
in
terms
of
not
letting
perfect
be
the
enemy
of
good
fragmenting.
The
problem
might
be
worthwhile
a
little
bit
here
right.
So,
let's
not.
E
It
might
not
be
I'm
just
suggesting
that
as
an
option,
if,
if
we
run
into
issues
because
whatever
we
do
around
stability,
I
think
people
expect
across
the
board
for
everything,
and
I
think,
if
you
start
to
account
for
metrics,
it
gets
really
confusing.
E
So
I
I
don't
know
if
it's
possible
to
fragment
that
off
as
a
separate
effort,
but
I'm
happy
to
help
when
it
comes
to
the
metrics
side.
Of
that
I
don't
think
I
have
time
to
commit
to
anything
else,
but
I'm
more
than
happy
to
help.
D
A
A
The
only
thing
I
would
caution
there
is,
if
we're
gonna
be
generating
metrics
out
of
spans,
then
metric
requirements
to
some
degree
might
come
to
bear
on
on
trace
requirements
as
far
as
what
kind
of
what
kind
of
changes
are
allowed
or
how
you
handle
them.
A
So
I
would
just
throw
that
out
as
like
a
an
addendum
there.
I
I
feel
a
little
bit
like
I
wouldn't
want
to
finalize
what
we're
doing
with
tracing
and
resources
until
we've
investigated
metrics
further,
because
I
have
a
feeling
they're
going
to
interact
with
each
other,
but
we
can
definitely
get
started
with
tracing
resources
without
tackling
all
of
that
other
stuff,
because
there's
some
more
basic
issues
that
we
can
flush
out
without
having
to
worry
about
some
of
that
stuff.
B
Okay,
that
sounds
pretty
good.
I
think,
thanks
for
adding
notes
to
ted,
I
think
what
I'm
I
don't
know
if
you
guys
agree
here.
Clearly,
I
think
next
steps
is
really.
B
We
need
to
focus
on
defining
what
stability
means
as
we
discussed
here,
so
that
could
be
the
start
of
what
needs
to
happen
with
this
sig
and
coming
back
week
over
week,
but
for
the
other,
more
specific
things
that
we
need
to
figure
out
too,
is
how
can
we
get
more
folks
to
commit
to
helping
in
each
one
of
these
categories
so
I'll
generate
at
least
more
conversation
again,
because
there
was
some
interest
in
slack,
but
I
think
we
have
a
challenge
with
time
zones,
because
some
of
the
folks
were
like
in
new
zealand
or
australia
yeah,
but
I'll.
B
I
can
take
that
action
item
to
try
to
get
more
folks
to
to
potentially
lead
a
category.
A
Yeah-
and
I
should
mention
they-
they
haven't
really
joined,
open,
telemetry
yet,
but
atlassian
is
very
interested
in
contributing
to
this
effort
specifically,
and
they
are
in
general,
like
in
australia
like
apac
time
zones.
B
We
have
the
tuesday
four
o'clock
one,
but
I
think
this
is
in
every
other
week.
I
don't
actually
think
I
see
the
calendar
for
tomorrow,
so
we
may
need
to
change
that,
but
maybe
let's
start
with
the
slack
channel,
unless
I'm
seeing
this
wrong.
I
thought
this
was
an
every
other
day.
A
See
it
on
the
calendar
and
it's
been
renamed
from
the
like
apex,
spec
meeting
to
the
apec
instrumentation
meeting.
I
do
wonder
whether,
like
you
know,
one
meeting
in
that
time
zone
is
is
enough,
but
but
we
can
definitely
start
with
that.
A
I
think
one
question
I
have
is
so
for
the
getting
to
stability
in
terms
of
like
getting
the
mechanics,
correct
and
then
adding
as
the
final
bit
of
that,
adding
back
to
the
spec,
the
just
the
rules
for
changing
semantic
conventions
like
what
are
the
rules
what's
allowed?
What's
not,
we
can
handle
that.
I
would
say
that
falls
on
like
the
open,
telemetry
expert
contributors,
side
of
the
fence
for
the
conventions
themselves.
A
I
know
one
of
the
goals
was
to
get
subject
matter.
Experts
involved,
johannes
mentioned
there
were
at
least
some
from
microsoft
on
messaging,
but
possibly
as
far
as
planning
out
when
and
how
we're
gonna
tackle
these
various
conventions.
B
Yes,
so
at
least
for
messaging,
I
think
johannes
can
help
on
the
scheduling.
Well,
we
do
have
some
folks
also
as
smes
for
http
we're
still
trying
to
define
how
they're
going
to
contribute.
There
were
some
changes
for
those
two
categories.
B
I
have
some
confidence
in
having
some
smes,
the
other
ones,
I'm
hoping
that
the
community
itself
or
other
you
know
I
know
alolita
was
looking
for
some
folks.
Let's
see
what
they
come
back,
what
she
comes
back
with
or
josh-
I
don't
know
if
you
can
ask
around
a
bit
if
there's
any
folks
who
might
want
to
contribute
some
of
these
other
categories,
but
that
would
be
super
helpful
in
terms
of
finding
some
smes.
A
Because
sequel
is
the
other
biggie,
I
would
say
that
that
people
really
really
care
about.
E
Yeah
and
the
we
are
contributing
the
sql
commenter,
at
least
that
I
don't
think,
that's
quite
official,
but
that
with
that
work
we
can
make
sure
we
have
one
or
two
smes
to
help
with
basically
databasing
in
general.
A
A
It
sounds
like
to
to
be
holding
these
meetings
in
those
times,
but
I
imagine
we
also
need
to
have
a
european
friendly
time
that
we're
talking
about
this
stuff
so
that
that
seems
like
that'll,
be
a
bit
of
a
juggling
for
the
the
pm's
on
this
project
to
handle
the
fact
that
there
might
be
two
parallel
conversations
going
along
about
this
stuff.
A
Yeah,
but
hopefully
I
guess
maybe
the
first
follow-up
is
like.
Is
this
meeting
time
a
good
time
to
be
talking
about
semantic
conventions
and
if
so,
where
should
we
get
started
there
with
http,
potentially
because
johannes,
I
know
you
have
a
we've
created
a
separate
meeting
for
for
messaging
in
a
european
friendly
time.
So,
presumably
that's
that's.
Working
with
the
subject
matter.
Experts
you're
going
to
bring,
I
mean.
D
A
And
that
kind
of
leads
this
meeting.
If
we
can
get,
is
this
a
useful
meeting
time?
Do
we
need
to
have
this
meeting,
so
I
would
wonder
about
that.
B
Yeah,
I
agree
with
you.
I
mean
if
we're
gonna
leverage
the
tuesday
meetings.
We
may
not
need
this
one,
let's
see
if
atlassian
and
some
of
the
folks
in
the
I
think,
as
in
japan
as
well,
if
they're
able
to
consistently
join
and
participate,
then
we
made
it
just
be
able
to
leverage
that
time.
Slot.
A
Leave
I'm
happy
to
come
to
this
meeting
too.
I
just
we
might
everybody.
B
Yeah,
I
have
no
preference,
I
think
it's
more
for
the
rest
of
the
community
and
who's
actually
doing
the
majority
of
the
contributions
to
the
specifications
and
whatnot.
A
Okay,
well,
maybe,
as
a
next
step
can
for
josh
and
michael,
and
we
can
ping
alolita
about
this.
A
Do
you
want
to
start
in
the
tuesday
afternoon
meetings
trying
to
tackle
http
and
kind
of
get
that
one
done
first,
while
we
kind
of
look
around
for
sql
smes
and
then
so
I
guess
the
action
item
would
be
to
like
try
to
get
our
http
smes
to
come
to
that
meeting
and,
if
and
flag,
whether
or
not
that's
a
bad
time
zone
for
the
people.
We
want
to
participate.
E
A
Okay,
yeah
and
if,
if
not,
then
I
mean,
I
think,
it's
fine
to
schedule
a
morning
meeting
as
well
and
if
we
need
I,
I
personally
think
it
is
fine
to
have
two
meetings
in
parallel.
If
we're
going
to
talk
about
http
and
messaging.
At
the
same
time
we
have
subject
matter.
Experts
like
it
would
be
nice
not
to
pile
them
all
into
the
same
meeting,
and
I
think
that
was
one
of
the
rules
we
set
up.
Is
we
don't
want
to
waste
the
time
of
the
subject
matter?
Experts?
A
B
A
A
A
B
A
B
Good
I'll
go
ahead
and
get.
A
B
Started
I
do
need
to
drop
for
the
next
meeting,
but
I
think
this
was
productive
as
a
first
first
kind
of
meeting.
I
appreciate
everybody
coming
and
being
kind
to
me.