►
From YouTube: 2021-07-22 Governance Committee private 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
Good
good,
it
looks
like
are
folks
joining
or
yes,
morgan.
A
Good
good,
I
was
going
to
get
some
time
on
your
calendar
just
needed
to
catch
up
on
things.
D
Very
observant,
I'm
I'm
in
mine,
I'm
actually
in
my
native
home
in
western
massachusetts
right
now,
cool!
Oh.
D
C
C
A
A
B
A
A
E
So
elise
suggested
this
marketing
thingy,
so
maybe
agenda.
C
No,
we
just
got
started,
and
I
think
the
first
topic
is
yours:
marketing
guidance
on
tweeting
about
products
that
support
hotel
without
necessarily
endorsing
yeah.
F
So
we've
had
this
situation
before
that.
I
think
we
navigated
well
with
with
tweeting
about
the
hotel.
This
roof,
which
was
you
know,
saying
you,
know
we're
not
endorsing
cloudwatch,
specifically
we're
saying
you
know
hey
by
the
way
this.
This
is
a
thing
that
supports
any
hotels,
combatable
fighter
right.
So
I
think
there
are
some
questions
about.
How
do
we
make
sure
that
we
are
fair
in
terms
of
not
like
endorsing
stackdriver
trace
or
endorsing
honeycomb
or
endorsing
lifestyle
right
like
instead
saying,
like
you
know,
we're
highlighting
these
things
that
are
compatible.
A
F
Right
exactly,
and
I
think
the
same
thing
is
true
of
the
right
like
it
was.
It
was
not
100
clear
from
the
blog
post
whether
the
new
google
cloud
agent
is
like
you
know,
built
on
hotel,
but
only
sends
data
to
stackdriver
right
or
whether
it's
a
thing
that
is
built
on
hotel
by
default,
sensitive
stacker.
F
We
can
send
data
anywhere
right,
like
I
think
that
is
important
when
we
kind
of
crystallize
the
mark
right,
like
companies,
are
free
to
say
whatever
they
like
right
on
their
own
blogs
as
long
as
it
conforms
to
our
kind
of
third-party
marketing
guidelines,
but
I
think
our
first
part
of
our
opinion
guidelines.
We
really
need
to
crisp
up
yeah
yeah,
okay,
so
I
think
you
know
maybe
like
amelia,
and
I
should
spend
some
time-
writing
something
and
putting
it
together
for
next
week.
F
But
I
did
want
to
flag
that
in
the
meanwhile,
like,
let's
figure
out
some
language
to
get
to
get
google
and
blocked
with
regard
to
promoting.
You
know.
B
F
E
That's
why
I
propose
this
tweet
and
I
wasn't
sure,
was
it's
okay
to
send
it
like?
Not,
okay,
there's
no
need
to
sign
it
like
it.
Just
I
mean
we
definitely
want
to
highlight
end
users
like
if
we
like
github
was
using
open
telemetry.
We
definitely
talking
about
that,
even
though
they
may
use
specific
technology
yeah,
but
for
vendors
I
agree
I
I
have
no
objections
here.
E
I
I
was
on
on
on
the
fence
whether
I
need
to
send
it
or
not,
because
on
one
hat
it's
highlights
that
open
ceremony
being
used
wider
and
wider
like
more
and
more
people
using
it.
Another,
like
I
mean,
let's
say,
new
relic,
will
already
announce
about
open
telemetry,
say
like
switching
to
open
celebrities
and
basing
hard
work.
Do
you
want
to
tweet
about
it
or
it's
new,
relic
specific,
like
it's?
It's
very
con.
I
mean.
F
A
F
Have
something
and
then
that'll
make
it
easier,
because
if
they
send
something
that
clearly
conforms,
we
can
just
immediately
post
it
as
written,
rather
than
having
to
figure
out
how
to
reward
it
cool
okay.
So
I
will
I'll
I'll
work
with
emilia
on
that
over
the
next
week
and
we'll
come
back
with
something
to
approve,
but
in
the
meanwhile
we'll
figure
out
something
we
can
tweet
like
about.
The
google
announcement
right
away.
E
So
for
google,
specifically,
this
agent
doesn't
support
other
backends
at
the
moment,
so
I
don't
think
like
we
can.
I
mean
we
can
try
to
craft
something,
but
I
don't
think
we
have
a.
I
need
to
post
about
it
from
open,
cm3,
twitter.
F
Aside
from
that,
it's
built
with
the
right
like
that,
it's
built
on
the
open,
telemetry
code,
okay,
cool.
I
will.
I
will
think
on
that.
That's
that's
helpful
information
to
know
that
it
is
currently
only
sending
information
to
stackdriver
and
that
it's
not
configurable.
Okay,
thanks
thanks
yeah
and
it's.
E
Also,
I
mean
I
don't
know
how
much
I
want
to
tell
about
it,
but
it's
like
it
currently
combines
fluent,
beat
and
open
telemetry
collector
because
of
like
we
are
not
that
g8.
Our.
E
Yeah
and
like
I
don't
know
how
much
we
want
to
talk
about
it
like
it's
just
what
it
is
at
the
moment,
so
it's
just
possible.
C
A
Yeah
yeah,
I
mean
I
mean
it's
work
in
progress,
but
the
the
other
thing
that
sega
just
wanted
to
request
is
that
I
mean
I've
commented
on
your
blog
post
that
you
shared
from,
which
is
posted
on
google's
blog,
that
the
open
telemetry
link
is
just
going
back
to
a
google
page.
It's
not
actually
going
back
to
the
project,
so
yeah.
A
E
So
liz
some
time
ago,
I
started
this
document
and
you
can
use
it
as
a
as
a
foundation.
I
I'll
send
I'll
post
it
yeah
just
send
to
my
email,
yeah.
Thank.
F
C
Cool
all
right
topic:
number
two
was
alita
sequel
commentary.
A
Yes,
so
very
excited
sequel:
commenter
has,
you
know,
started
its
journey
into
open,
telemetry
and
then
sega
posted
the
request
already
so
looking
forward
to
you
know
having
that
being
evaluated
by
the
dc.
Obviously,
as
a
next
step,
is
there
anything
else
that
sega
needs
to
be
done
or
morgan
guys.
C
B
A
To
that,
and-
and
I
also
wanted
to
kind
of
just
flag
for
visibility
that
we're
just
you
know-
we're
working
with
the
flo
mill
team,
as
well
as
with
datadog,
to
get
a
group
set
up
on
ebpf
so
that
you
know
we
can
at
least
actually
start
working
on
the
and
on
having
a
clear
backlog.
A
For
you
know
some
of
the
work
that
needs
to
be
done
on
the
floor
mill
component
landing
into
hotel
and
then
into
the
collector
and
then
also
having
a
clear
backlog
that
you
know.
Different
contributors
can
get
involved
in
on.
So.
E
So
I
think
the
process
of
accepting
it
is
governance
committee
need
to
ultimately
accept
and
technical
committee
is
doing
due
diligence.
Yes,
we
may
skip
due
diligence
if
you
want
to
as
a
government
committee,
but
I
know
right
now.
The
problem
is
that
bogdan,
who
volunteered
to
do
diligence
kind
of
do
it
for
a
couple
of
weeks
and
only
like
we
already
like
had
some
internal
communications
delays.
That's
why
we
posted
on
just
now
and
now,
like
another
delay,
will
be
from
blog
them.
So
should
we
get.
A
E
B
I
kind
of
want
to
ask
more
of
a
philosophical
question:
where
are
we
going
to
draw
the
line
with
respect
to
what
is
in
scope
for
open
telemetry,
because,
technically,
like
you
could
think
anything
still
empty
related
could
be
in
scope,
but
that
dilutes
the
meaning
of
the
project,
because,
like
these
projects,
could
just
as
well
be
standalone
project,
they
don't
have
to
be
under
open
to
limit
the
umbrella,
necessarily.
C
D
A
little
problematic
for
you,
because
I
mean
it
just
extends
that
definition,
which
I
think
is
like
a
definite
it's
a
like.
That
describes
an
important
category
of
software
that
needs
to
get
written
that
might
even
want
sort
of
open,
telemetry,
branding
or
something.
But
the
main
thing
that
I
struggle
with
is
that
we
still,
as
far
as
I
can
tell,
have
a
massive
maintain
maintainer
shortage,
and
we
cannot.
D
The
we
cannot
increase
the
corpus
of
software.
The
major
changes
are
responsible
for,
in
my
mind,
without
like
especially
categorically
I
mean
I
I
think
I
you
know
ta-da
the
donation
that
and
maybe,
as
I
think,
I'm
as
excited
about
that
as
anything,
it's
definitely
valuable
data
and
I'm
actually
happy
for
it
to
have
open
femice's
name
on
it.
But
I
think
there's
you
know.
We've
talked
about
this
a
number
of
times,
but
there
needs
to
be
some
kind
of
place
where
it's
like.
D
This
is
like
a
relatively
trusted
project
and
has
some
bona
fides
or
whatever.
So
you
can
say
it's
open
celebratory,
but
it's
in
contrib,
where
it's
like
open,
telemetry,
affiliated
but
not
open,
telemetry
maintained,
depends
conversion
in
the
same
way
that
the
sdks
would
be,
or
something
like
that.
I
think
that's
the
distinction.
You
know
yeah.
A
And
then
I
think
that,
again
to
your
point,
then
we
have
discussed
you
know
and
yuri
has
also
brought
this
up
earlier-
that
we
do
need
to
have
some
kind
of
an
you
know,
plugins
or
externally
maintained
umbrella
of
projects
that
we
can
support,
which
are
not
necessarily
core
maintainer.
You
know
supported,
but
also
come
with
their
support
structure,
but
are
under
the
open,
telemetry
umbrella.
The
reason
you
know
that's
important
is
because
the
sql
commenter
sits
in
you
know
google's
reapers,
for
example.
Just
for
the
sake
of
example.
F
F
That
does
not
necessarily
mean
that
it
has
to
be
under
the
open,
telemetry
umbrella
right,
and
therefore,
I
think
there
is
a
valid
question
of
you
know
should
something
be
a
separate
cncf
sandbox
project
versus,
should
it
be
incorporated
in
hotel
and
I
think
the
limit
test
for
me
there
is
like
you
know
how
core
is
it
to
people's
needs?
F
How
how
much
is
it
something
that
is
kind
of
tied
to
the
core
of
you
know
this
crop
context
propagation
right
like
framework
right?
This
is
kind
of
how
we
set
made
the
case.
For
you
know
this
is
why
metrics
and
tracing
and
all
these
things
go
together
because
they
use
the
same,
connects
propagation.
The
same
tagging.
D
Maybe
one
way
to
think
about
liz
would
be
there
can
be
only
one
you
know
like
it
doesn't
make
sense
to
have
two
different
context:
propagation
mechanisms
for
sure.
Nor
does
it
make
sense
to
have
two
different
ways
to
say
when
a
span
starts
and
ends
or
when
a
metric
counter
increments
right
there
can
be
only
one,
whereas
the
database,
I
mean
again,
I'm
excited
about
the
project,
but
it's
not
as
though
to
me
that
there
can
only
be
one
project
that
does
sequel
annotations
or
something
like
that
like.
D
F
Maybe
but
then
the
question
is
for
things
in
our
contrib
repos
right
like
where
there
is
a
official
semi-officially
supported
right
like
hotel.
You
know
javascript
express
instrument,
auto
instrumentation
right.
Someone
could
also
write
their
own.
The
kind
of
the
de
facto
center
of
gravity
is
right,
like
this
is
something
that
is
community
maintained.
You
know,
at
least
until
the
point
that
express
writes
in
hotel
support
into
their
libraries
right.
C
F
C
C
F
C
And
I
think
to
ben's
point:
there's
also
you
know:
do
we
give
these
things
the
same
guarantee
of
quality
that
we
give
for
the
sdk
and
the
collector
and
the
apis?
I
I
would
pause
it,
probably
not
because
they're
less
right.
C
A
I
think
I
think
you
know
the
the
clear
guidelines
again
need
to
be
spelled
out.
I
mean
right
now
there
is
a
lot
of
greyness
in
the
in
our
interpretation
and
and
we
do
need
to
spell
that
out
clearly
that
what
does
this
mean?
If
it
is
when
core
maintainer,
you
know,
maintained
component
versus
a
contrib
maintain
maintained
component?
A
What
are
the
guarantees
that
are
coming
with
it,
that
the
project
is
willing
to
support
and,
and
then
there
are,
you
know
completely
external
add-ons
which
are
which
have
to
also
be
certified
if
they
want
to
be
open,
telemetry
compatible
right
so
again,
there
are
clearly
distinct
categories
of
components
which,
which
should
have
their
criteria
defined
from
the
project.
D
Would
it
be
helpful
that
it's
not
a
reform
question
to
actually
take
a
stab
at
defining
the
the
different
equivalence
classes
or
something
of
code
in
the
project
and
what
the
differences
are
in
terms
of
both?
I
don't
know
you
can
call
it
branding.
B
D
D
Yeah
yeah
yeah
kind
of
I
mean
I
I
I
like
the
I
don't
I
don't
feel
like.
I
don't
love
the
idea
of
telling
people
who,
especially
if
the
code
is
coming
from
the
community,
where
the
alternative
in
github
is
just
to
have
it
attached
to
someone's
username.
I,
like
the
idea
of
saying,
there's
like
a
open
telemetry
branded
place
for
those
contributions.
D
I
just
think
it
should
be
clear,
they're
contributed
and
that
they're,
not,
for
instance,
being
like
audited
or
maintained
by
the
same
people
that,
like
make
sure
spec
changes
or
kosher
or
whatever.
So
again,
I
I
like
the
idea
of
this
sim
or
a
contrib
thing
that
has
a
conference's
name
on
it,
for
people
who
would
rather
donate
than
keep
it
affiliated
with
their
own
corporate
or
individual
identity.
A
B
F
To
add
glue
and
additional
code
into,
for
instance,
the
collector
repo
or
individual
language,
sigs
yep.
A
Exactly
I
mean
we
are
following
that
pretty
much.
You
know
those
two
guidelines
that
the
project
comes
with
its
own
maintainers
and
you
know
it's
distributed.
D
But
if
inflammatory
gets
where
I
think
it
will
be
in
five
years,
this
is
the
ultimate
supply
chain
attack.
I
can't
believe
it's
being
recorded,
but
it's
like.
Oh
it's
just
telemetry
or
whatever,
and
then
you
bring
in
this
thing.
It's
like!
Oh
no,
actually,
it's
really
not,
and
it's
in
fact
like
a
remote
control
device
or
something
like
that.
So
I
mean
I
think
that
it
would
be
helpful
for
us
to
clarify.
B
D
Something
is
being
maintained
or
not
by
open
flemish
before,
especially
as
there's
and.
F
It
has
some
of
its
own
code
bases,
there's,
no
reason
that
necessarily
has
to
be
coupled
to
the
core
language
sdks
right,
and
I
think
that
the
same
way
that
you
know
serverless
and
lambda
is
one
is
one
important
thing
right.
I
feel
that
the
database
constituency
is
another
constituency
where,
like
you
know,
I
think
both
of
these
should
reach
roughly
the
same
level
of
support
wherever
we
decide
that
that
oval
is
because
you
know
it
is
not
quite
as
foundational
as
the
core
sdks.
B
My
concern
is
not
just
with
the
maintainer
load:
it's
it's
bigger!
It's
the
branding,
concern
right
because,
like
I
don't
want
to
end
up
in
a
situation
where
anything
that
theoretically
can
produce.
Telemetry
needs
to
be
under
open,
telemetry
brand,
and
if
we
explain
to
a
person
like
what,
as
then
said
like
when
they're
getting
open,
telemetry
what
they're
getting
and
also
just
to
explain
to
newcomer
what
is
open
telemetry.
C
B
F
E
Thank
you
for
yeah
for
sql
comments.
Specifically,
it
has
like
three
components:
it
has
components
that
extracts
contacts
from
incoming
requests
like
like
big
context
and
like
like
express
url
or
something
that
may
be
integrated
with
whatever
we
have
in
contrib
right
now.
Another
part
is
sql
driver
itself,
and
this
probably
also
can
benefit
like
you
end
user
will
benefit
from.
It
have
been
the
same
component
that
collect
traces
as
well
as
injects
the
context
so
right.
F
And
I
think
that
goes
to
the
you
know
to
answer
ben's
question
of:
can
there
only
be
one
well?
The
answer
is,
I
think
yes,
because
it
relates
to
context
propagation
right.
It
relates
to
how
do
you
propagate
context
across
to
the
database
layer,
and
how
do
you
read
it
back
afterwards,
like
I
feel
like
you
have
to
have
either
either
a
w3c
standard
around
it
or
you
wind
up
having
just
de
facto
implementation
by
one
one
code
base.
F
A
What's
the
next
step
here,
should
we
actually
embark
on
an
on
an
at
least
a
documentation
of
criteria
where
we
designate
you
know
some
criteria
associated
or
guaranteed
associated
with
different
types
of
contributions.
D
Yeah,
I
think
it's
a
good
idea.
We
have
the
contributing
of
the
style.
I
don't
think
it
needs
to
be
some
giant
thing,
but
just
right
the
main
gap.
I
think-
and
I
we
I
don't
get.
I
don't
know
if
we've
achieved
consensus
or
or
whatever
we've
voted
or
whatever
on
this,
but
but
do
we
want
to
have
a
distinction
from
a
branding
standpoint
about
open,
telemetry
code
that
is
and
is
not
like
audited
by
the
maintainers
in
some
capacity.
That's,
I
think
the
open
question
in
my
mind.
D
Otherwise
I
don't
know
if
it's
worth
adding
the
additional
stuff,
but
if
so,
I
think
we
need
to
have
some
like
a
different
png
that
you
include
in
your
project,
just
like
the
telemetry
contributed
or
something,
as
opposed
to
like
official
open
telemetry
like
as
liz
is
saying
it
sounds
like
the
sql
commenter
thing:
it's
not
just
commenting
it's.
Actually,
it's
like
sequel
propagation.
C
C
I
believe
the
intent
of
the
project
within
google
is
for
customers
to
to
eventually
be
able
to
get
some
back-end
correlations
within
the
sql
server
itself
long
term,
but
but
it
is
really
effectively
just
way
better
instrumentation
for,
like
mysql
and
and
various
other
databases,
the
commenting
part
of
the
name
unless
I've
missed
something.
C
F
D
A
D
D
C
A
C
E
A
C
B
B
There
is
really
no
updates,
because
we
we
have
not
resolved
the
circular
dependency
currently
that
exists
and
it
kind
of
needs
to
be
resolved.
I
think,
as
far
from
what
I
remember
in
the
collector
itself,
to
sort
of
separate
the
plugins
more
clearly
until
then,
we
we
can't
really
move.
Like
we've
tried
several
two,
I
think
two
different
approaches
with
integrating
collector
and
we
we
keep
running
into
the
circular
dependencies,
so
we're
not
doing
it
at
the
moment.
B
I
mean
we're
important,
actually
otlp
directly
again
by
importing
collector
transformation
code
so
that
that
sort
of
integration
is
happening,
but
not
like
the
architectural
change
to
be
based
on
collector.
I.