►
From YouTube: 2022-01-14 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
A
B
B
B
C
Yeah,
no,
that's
a
valid
reason.
No
yeah.
I
was
a
huge
fan
of
intellij
during
my
my
java
days
and
but
I
think
I
have
to
try
go
land
again
because
it
wasn't
that
mature
when
I
tried
out,
but
I
I've
been
hearing
quite
a
lot
of
good
things.
D
About
it,
it
does
way
better
defectively
like
if
I
change
a
name
or
something
visual
studio
code
never
worked
for
me
when
I
had
to
rename
something.
That's
true.
I
had
to
do
a
lot
of
that
work.
Like
said
something
like
a
lot
of
them,
but
in
in
golan
no
once
it
builds
the
index
and
stuff,
so
you
just
simply
move
things
like
in
java.
Remember
the
good
java
days
when
yeah
something
does
not.
It
applies
all
the
changes.
C
C
I
heard
they
have
also
very
good
support
for
multi-modal
modules
projects
as
well.
Yeah.
C
Yeah,
so
for
contrib,
what
I
do
with
visual
studio
code
is,
I
just
open
individual
modules.
C
Yeah,
I
don't
know
what,
otherwise
it
just
complains
yeah,
I
I
I
it
I
don't
know.
Perhaps
it
is
a
set
setup
issue,
but
whenever
I
try
to
work
on
on
something
like
multi-mode,
it
would
just
complain
about.
You
know
things
up
in
the
right
places
and
compilation
issues
and
so
on
cool
okay.
So
I
have
to
try
it
out
because
much
module
support
is
really
necessary
for
construction.
E
F
I
got
these
there's
like
little
toy
like
spherical
magnets
for
christmas,
that
are
like
this
incredible
distraction
I
mean
well,
I
I'm
actually
not
distracted
mentally,
but
I
keep
on
like
looking
down
at
these
things
and
playing
with
them.
This
is.
H
F
These
things
are
amazing:
I've
made
all
sorts
of
beautiful
objects.
You
know
it's
yeah,
it's
an
incredible
product,
incredible
product,
apparently
also
incredibly
deadly.
If
you
swallow
more
than
one
of
those,
I.
H
F
I
think
I
really
wanted
to
ask
is
going
to
be
difficult
without
a
lolita
here
I
was
kind
of
like
came
up
in
a
kind
of
customer
context,
just
an
understanding
of
like
which
managed
cloud
services
are
actually
supported
by
otel,
and
I
mean
I
think
I
sort
of
might
be
able
to
figure
out
the
answer
that,
but
none
of
it's
in
the
registry-
and
this
is
more
just
you
know-
ignoring
my
particular
need
like
from
a
you
know,
outside
hotel
end
user
perspective,
it's
fairly
difficult
to
get
a
clean
answer
about
which
things
are
supported
by
hotel.
F
I
think
yeah
at
this
point.
We
have
no
one
from
a
cloud
vendor
on
the
gc
anymore,
since
morgan
has
changed
teams
whatever,
but
I
was
just
kind
of
curious
how
we
want
to
address
that.
It
does
seem
like
a
good
thing
to
have
a
central
answer.
To
I
don't
know
I'd.
Imagine
people
wanting
to
understand
if
hotel
supports,
dynamodb
or
not
and
stuff
like
that,
but.
I
F
I
F
What
I
mean
is
that
there's
instrumentation
in
the
amazon,
client,
libraries
for
service
x,
y
and
z,
and
I
believe
there
actually
is
in
the
sense
that
there's
an
interceptor
api
for
basically
all
well
by
saying
basically,
I
mean
I
actually
don't
know,
but
I
think
there's
an
interceptor
api
for
all
of
the
popular
amazon
managed
services
and
there's
an
hotel
adapter
for
that
interceptor.
So
in
that
sense
you
can
get
coverage
of
these
things.
F
But
if
you're
like
naive
about
hotel-
and
you
come
into
the
project,
there's
really
no
way
to
figure
that
out,
except
for
reading
the
code
and
actually
reading
quite
a
bit
of
code,
which
is
not
good.
So
the
the
idea,
in
my
mind,
was
just
to
come
up
with
a
hopefully
maintainable
list
of
cloud
dependencies
that
are
covered
by
hotel
instrumentation,
that's
available,
ideally
by
auto
instrumentation.
But
even
if
it
wasn't,
I
would
take
it.
Does
that
make
sense
as
just
a
you
know,
problem
to
solve
or
something.
H
Yeah
I
just
my
only
question
is
I
feel
like
the
like.
I
guess
to
me
the
the
issues.
Maybe
it's.
The
registry
has
incomplete
information,
then.
F
F
Next,
with
this
I've
I've,
I've
felt
for
a
while
that
the
registry
is
something
that's
of
no
value
really
to
contributors
and
implementers,
but
of
huge
value
to
end
users
and.
F
Our
project
is
so
it's
so
focused
on
the
vendors
and
contributors
and
so
not
frankly,
that
focus
on
end
users
has
been
de-prioritized,
but
just
that
the
maintainability
of
the
registry
is
a
little
questionable.
F
I
still
like
the
idea
of
moving
and
we
can
continue
to
have
what
we
have
now
as
an
and
but
like
some
kind
of
tagging
scheme
in
github,
where
you
know
you
can
tag
something,
as
you
know,
being
there's
like
github
labels,
you
can
add
to
repos
and
to
have
some
manifest
file.
That's
in
that
repository
that
would
you
know
list
list.
F
F
Don't
you
don't
require
that
people
make
updates
to
the
registry
when
they
add
new
when
they
add
new
support
and
it's
just
something
they
do
within
their
own
repo
and
then
hotel
does
the
work
of
having
some
kind
of
periodic
job
to
pull
that
stuff
in
so
anyway,
I
mean
that's,
that's
what
I
that's
where,
where
I'm
going
with
this,
but
the
problem
of
trying
to
like
the
immediate
problem
is
just
like
discoverability
of
hotel
support
and
I
just
don't
think
the
registry
model
is,
I
don't
think
it's
bad,
but
I
don't
think
it's
sufficient.
A
I
can
definitely
the
comms
meeting
is
right
after
this
one,
and
I'm
going
to
add
this
to
the
agenda
there.
I
think
it's
a
great
idea
of
making
it
easier
to
add
one
like
side
issue
here
is
the
relation
to
hotel
not
currently
having
a
big
control,
maintainer
community.
That's
a
thing
we're
looking
at
trying
to
to
grow
in
the
information
instrumentation
sig
like
like
just
in
general,
providing
more
tooling
that
makes
it
easier
to
to
maintain
instrumentation
in
general.
D
There
was
a
discussion,
if
I
remember
correctly,
probably
a
year
ago,
about
what
is
an
official
supported
integration
or
instrumentation
versus
what
is
not
officially
supported.
What
are
the
guarantees
that
we
give
to
this?
So
if
we
start,
if
we
start
advertising
this
as
supported
by
open
telemetry,
what
does
it
mean
that
is
supported
by
open
telemetry?
I
think
I
think
there
was
a
long
discussion
back
then,
but
we
should
we
should
have
somewhere
documented.
D
F
I
think
that's
fair.
I
I'm
in
terms
of
baby
steps
or
something
we
we
you're
right.
We
had
that
discussion
and
I
think
we
came
up
with
some
pretty
reasonable
ideas
about
more
sophisticated
things.
We
could
do
to
track
that
and
then
the
reality
is
that
we're
a
year
four.
We
have
still
have
nothing.
So
I'm
just
trying
to
be
like.
Can
we
just
have
like
a
version?
That's
just
saying:
look
caveat
emptor.
Wait!
Yes,
caviar
emperor!
That's
the
right
latin
phrase!
Right!
Buyer!
Beware!
F
No
idea
if
this
thing
even
works,
but
someone
says
that
this
thing
has
dynamodb
support
for
hotel.
I
mean
that
that's
the
level
of
support
guarantee
I'm
talking
about
here.
I
just
think
that
there's
actually
quite
a
bit
out
there
and
it's
not
that
discoverable
right
now,
because
I
think
the
registry
has
been
just
it's
difficult,
yeah
it
it
you
have
to
if
it
could
be
integrated
via
only
local
changes.
I
think
it
would
be
much
easier
to
get
people
to
to
go
in
that
direction,
especially
when
they
complicate
stuff.
So.
F
Basically,
yeah,
I
mean
that's
what
the
registry
is
supposed
to
be,
ideally
yeah,
I'm
really
particularly
thinking
about
you
know
third-party
instrumentation
for
open
telemetry,
of
which
there
actually
is
quite
a
bit
now
and
there'll,
be
a
lot
more
in
a
year,
and
I
just
don't.
I
don't
think
it's
discoverable
and
I
actually
think
it's
an
important
part
of
the
project
like
if
you
go
to
the
hotel
io
home
page,
and
we
had
something.
F
D
Yeah
definitely
agree,
as
I
said,
my
my
two
cents.
There
were
more
more
related
to
the
fact
that
we
need
to
put
that
caveat
or
or
whatever
we
call
it
for
users
to
know
that
we
are
not
responsible
to
fixing
all
this
because
yeah
we
are
100
people
and
hence
we
cannot
fix
every
everything
in
the
world.
A
There's
like
a
very
basic
version
of
that
which
is
easy
to
implement,
which
is
just
is
this
repository
posted
under
the
open,
telemetry
organization
or
not?
If
it's
hosted
by
us,
then
one
way
that
we
are
on
the
hook
for
making
it
work.
F
F
Okay,
I
I'm
very
happy
to
like
I
don't
know,
take
some
sort
of
action
to
move
this
forward.
That's
me
it's
true,
but
I'm
not
sure
exactly
what
we
think
that
is,
it
could
be
like
trying
to
just
you
know,
come
up
with
a
one-page
proposal
for
how
we
could
make
something.
That's
you
know
a
way
to
get
things
into
the
registry.
It
doesn't
require
editing,
registry
or
whatever,
and
what
that
might
look
like,
but
I
I
don't
know
what
people
how
tactical
we
want
to
get
at
this
point,
but.
H
Okay,
no
go
ahead
like
I
know
that
austin
had
done
that
work
a
while
back
to
automatically
start
pulling
in
documentation
into
open
telemetry.io,
which
seems
pretty
well,
I
mean.
A
H
A
H
It'd
be
probably
critical,
I
think,
just
to
list
list
like
the
instrumentation
and
also
list
what
it
works
for,
just
because,
like
like
ben
you
had
this.
You
mentioned
this
in
the
context
of
of
cloud
services,
but
it's
it's
effectively
the
same
problem
for
software
packages
that
people
deploy
and
so
like.
If
we
say
we
support
mysql,
that
can
mean
many
different
things
right.
H
C
Yeah,
so
one
one
question
that
I
do
have
is
like:
are
we
also
when
we
say
manage
cloud
services?
Do
we
mean
also
running
open
telemetry
like
collector
on
those
services
as
well?
So
are
we
interested
in
listing
them
like
open,
telemetry.
H
H
D
Cloud
providers:
there
is
deeper
integration
with
open,
telemetry
things
like
they.
They
pass
our
headers
through
to
their
system
they
produce
telemetry
on
that
is
compatible
with
open,
telemetry
and
stuff.
Like
that,
that's
that's
another
huge
thing
that
cloud
providers
can
do,
and
maybe
maybe
we
do
do
that
sorry
welcome.
We
were
discussing
about
about
the
topic
that
ben
brought
about
how
to
to
advertise
more
our
integrations
how
to
advertise
more,
how
to
use
things
in
different
clouds
and
stuff.
J
Yeah
exactly
and
then
I
think
you
know
again,
we
definitely
you
know,
leverage
and
and
hotel.
You
know
wherever
we
can
advertise
it,
we
we
do
right,
so
we
do
use
the
user
agent
headers
from
the
collector,
the
specific
installs,
the
specific
types
of
os's.
You
know
everything
that
has
been
conveyed
in
the
headers
themselves.
Right
so
again,
I
think
that
I
mean
ted.
You
were
mentioning
using
these
website
or
the
registry.
Is
that
how
you
want
to
do
it
or
it
should
be
there?
Should
there
be
some
other
mechanism?
There's.
A
I
think
it's
only
a
partial
solution
because
you
still
have
to
know
what,
to
even
like,
say,
look
up
in
the
registry
yeah.
I
would
I
have
ideas
for
like
other
things
we
can
do,
but
of
course
they
all
involve
engineering
efforts.
Right
like
it
would
be
great
to
be
able
to
consume
people's
manifest
files
like
there's
various
forms
of
standard
dependency
lists
and
manifest
files.
It
would
be
great
for
someone
to
just
be
able
to
copy
paste
that
in
to
the
website
and
then
get
something
spat
out.
A
That
says,
here's
all
the
stuff
that
support
it.
But
you
know
that's
the.
A
F
H
F
Am
an
open,
telemetry,
instrumentation
plugin
and
then
have
some
sort
of
yaml
file
or
something
in
in
a
designated
place
in
that
repo.
That
will
have
all
the
details.
So
then
we
can
do
a
github
query
to
find
all
these
repos
once
a
night
pull
down
the
yaml
files
and
grab
everything.
So
it's
a
local
change
exactly.
F
Think
that's
exactly
what
I'm
suggesting
and
the
motivation
for
this.
To
also
be.
I
have
a
cure
for
your
take
on
this.
Like
my
like,
I
actually
tried
to
dig
into,
and
I
wasn't
even
exactly
sure
like
to
the
best
of
my
knowledge,
and
this
is
where
you
can
correct
me,
for
instance,
for
amazon
managed
services
that
are
like.
You
know
well
well
used
things
like
dynamodb
stuff,
like
that.
My
understanding
is
that
there
is
a
way
to
connect
hotel
to
the
interceptors
to
get
open,
telemetry
spans
for
those
client
calls.
I
think.
F
Great
and
I'm
I
kind
of
knew
that
was
the
case,
but
if
you
go
to
the
registry
right
now
and
you
search
for
dyno
id,
it
literally
comes
up
with
nothing.
And
so
my
concern
is
that
this
came
from
a
customer
conversation
where
there
was
like
some
confusion
about
whether
or
not
various
amazon
aws
services
were
covered
by
hotel
instrumentation
that
was
available
somewhere
in
the
internet
like,
and
they
just
couldn't
figure
it
out.
F
F
J
And
then
it's
a
typical
customer,
you
know
request
all
the
time
coming
to
us.
You
know
as
to
what
exactly
is
supported
by
hotel
and
where
is
it
right?
So
so
we
do
make
it
available
on
the
aws
side.
You
know
as
mappings
that
we
make
available
to
you
know
on
github
and
we
update
the
documentation.
J
You
know
pretty
regularly
but
again
if
there
is
a
way
that
can
be
easily
used,
such
as
even
updating
and
you
know,
json
markdown
or
a
markdown
file
to
on
the
github
side.
We
can,
you
know,
happy
to
maintain
it
and
update
it
regularly
and-
and
I
think
that
I
mean
my
conversations
with
google
as
well
as
azure-
have
been
they're
very
interested
in
doing
the
same.
So
I
don't.
F
C
So
I
think
we
can.
We
can
start
with
the
contrib
collector
contrib
repository,
because
there
are
a
lot
of
integrations
in
there
yeah.
So
we
can
clean
up
those
from
the
registry
and
let
them
be
managed
by
automatically
by
some
script
within
the
contrib,
because
we
know
the
structure
of
that
repo
and
we
could
have
like
an
action.
A
github
action
that
runs
periodically
like
nightly
run
that
just
generates
the
json
files
that
the
registry
expects
compares
with
what
is
there
and
opens
apr
if
there
are
changes?
C
C
Or
something
because
that
sounds
like
a
entry
of
spam
into
the
website
right,
so
I
can
just
create
plugins
that
are
not
plugins
and
have
information
published
on
openstructure.io,
so
links
to
my
website
and
you
know
just
to
improve
my
seo
and
so
on
so
forth.
So
that's
that's
something
I
wouldn't
do
at
least
not
not
in
the
beginning.
J
You
know,
repos
is
the
first
step
right
and
then
going
from
there
in
terms
of
figuring
out
whether
it's
a
push
or
a
pull
to
the
registry
or
wherever
else
you
want
to.
You
know
just
collate
this
together,
because
I
think
one
of
the
things
against
ted,
as
you
mentioned,
is
discovery
on
the
registry,
and
you
know
just
somebody
just
browsing
through
it
and
it
being
integrated
again.
J
A
J
A
Yeah,
I
I
think
I
just
I
do
want
to
emphasize
that
right
now.
It
is
very
easy
to
make
a
pr
to
the
open,
telemetry
io
registry
and
add
anything
we
currently
have
there.
So
there's
actually,
no,
in
other
words,
for
if
there's
stuff
that
we're
seeing
today
like
it's,
it's
just
as
much
effort
to
to
make
a
pr
to
that
website
is
just
to
add
a
bunch
of
tags
and
json
files
to
a
repo
somewhere.
A
So
there's
there's
nothing
stopping
us
from
getting
that
up
to
date
with
all
of
our
own
stuff
right
now
and
as
far
as
like.
Is
it
something
that
automatically
pulls
things
in
which
makes
it
easier
for
contributors
but
creates
a
lot
of
spam,
or
maybe
some
middle
ground
which
is
like
you
know
you
put
your
json
file
in
your
repo
and
then
you
just
need
to
make
a
one-time
pr
to
the
the
website
or
otherwise
make
a
submission,
but
it's
still
gatekeeped.
K
Yeah,
I
feel
like
the
letter
is
better
like
the
open
to
limit
open
tracing
model.
Essentially
because
that
way
you
can
have
pr
run
by
sort
of
some
ci
validator,
whereas
if
you
pull
files
from
all
kinds
of
repos,
you're
gonna
get
like
not
only
spam,
but
it's
just
like
mel
form
files
and
all
of
that
and
there's
no
way
for
people
to
validate
their
files.
Whereas
opening
api
is
that
difficult.
A
Yeah,
but,
but
certainly
right
now,
it
is
like
very
easy
to
to
just
add
an
entry.
You
know
you
can
you
can
do
it's,
you
can
do
it
through
the
github
web
ui.
You
know
you're,
just
you're,
just
pasting
a
little
bit
of
text
into
a
file
and
then
hitting
the
submit
pr
button.
So
yeah.
J
So
yeah
I
mean
I
think,
as
long
as
there's
a
documented
process,
everybody
can
all
the
maintainer,
I
mean
again,
don't
expect
the
core
maintainers
to
have
to
do
this.
I
think
it's
a
you
know
it's
a
great
project
for
an
contributor
to
actually
work
on
and
I'm
happy
to
help
their.
You
know
again,
let's,
let's
just
have
a
plan
done
what
we
want
to
do
and
then
make
it
happen.
Great.
J
But
jurassic,
I
think
we
leverage
you
as
a
reviewer.
I
know
if
you
have
spare
time,
I
I
have
other
projects
for
you.
So.
C
J
A
I
mean
so
I
put
it
on
the
agenda
for
the
com
sig
to
look
at
how
we
could
improve,
making
it
easier
to
to
add
things
to
the
registry.
Like
is
there
a
way
to
automate
this,
but
if
we
want
a
community
issue
for
just
like
hey
everyone
add
their.
A
J
Yeah
or
or
just
bring
it
up
in
the
maintainer's
sake
I
mean
just
just
so,
for
awareness.
A
H
F
Get
that
maybe
more
than
addresses
yeah
thanks.
F
H
F
Needs
to
be
there,
which
I
totally
get
as
well.
Alita
needs
to
be
there,
and
even
if
that's
it,
you
can't
there's
no
times
that
were
proposed.
So
london,
when
are
you
going
on
pto?
You
said
it
was
the.
G
Week
after
that,
no
so
I'm
going
on
pto
the
week
after
I'm
going
skiing,
and
I
don't
want
to
deal
with
these
things.
F
Why
not
yeah
so
that
that's
fine,
I
I
think
the
choices
we
have
are
to
find
some
super
weird
time
like
evening
or
something
like
that
or
to
push
it
like
three
weeks,
which
feels
like
kind
of
a
long
time.
To
be
honest,
I
don't
yeah.
J
Yeah
I
I
would
rather
address
it
sooner.
It's
not.
You
know
again,
it's
totally
addressable
and
and
it's
something
that
I
mean
again,
we
can
meet
in
the
evening
that,
if
that's
easier,
you
know
anything
past
six
is
always
okay.
Well
I'll
make
time
for
it.
I.
G
Mean
the
good
part
is
jurassic
is
at
least
in
brazil,
so
yeah
this
time.
F
Yeah
totally,
we
can
listen
to
some
samba
music
and,
in
the
background,
to
keep
people's
tempers
down.
No,
that
no,
that
sounds
great,
I
I'll
send
out
another
iteration.
I.
J
Mean
why
don't
you
just
ben
just
suggest
three
times
and
then
people
are
well.
G
G
F
F
And
that's
anyone
on
this
call
feels
that
there's
a
must.
I'm
totally
happy
to
to
include
you
as
such
I
mean
genuinely.
I
just
let
me
know,
because
it's
it's
been
harder
than
I
thought
to
schedule,
something.
J
All
right-
and
I
just
wanted
to
you-
know
before
we
end
again
just
to
give
an
update
to
the
gc
bogdan
and
I,
as
well
as
morgan
and
riley,
have
been
working
through
the
timeline,
for
you
know:
metric
stability
on
the
different
components.
Obviously
it's
a
phased
approach.
People
are
very
anxious.
You
know
customers,
as
well
as
other
stakeholders,
for
some
communication
from
the
project
on
what
our
approach
is
metric
stability
and
and
that's
something
we're
working
on
right.
J
Now
we
have
a
dock
that
we're
working
together
on
obviously
identifying
key
areas
which
are
still
missing.
There's
some
parts
that
we've
established,
which
are
missing
on
the
spec
as
well.
As
you
know,
a
kind
of
blocking
implementation
on
the
sdks.
J
The
good
news
is
that
the
collector,
at
least
core
is
in
pretty
good
shape
to
support
otlp
and
prometheus
interoperability.
So
that's
great
news,
but
again
there's
other
work,
that's
ongoing,
so
we
should
communicate
that
at
least
right
I
mean
we
can
do
this
in
a
phased
way
and
and
again
we'll
continue
to
communicate
that
to
the
gc
and
a
regular
basis,
because
I
think
it's
very
important
for
the
project
at
this
point
in
time
I
have
had
heard
a
lot
of
frustration
from
customers.
J
You
know
on
not
understanding
what
the
landing
point
for
metrics
is
and
we
have
been
delayed
at
least
six
months
now,
if
not
more.
This
is
the
project
in
its
third
year.
We
cannot
afford
this
so
again.
Super
interested
in
you
know
kind
of
getting
everybody's
thoughts
on
this
and
and
help,
but
we
need
to
communicate
this
regularly.
I
was
just
thinking,
I'm
gonna
start
drafting
and
what's
new,
at
least
on
aws's
blogs,
to
communicate
what's
the
status
of
hotel,
because
I
can
move
that
faster.
J
Sometimes
then,
then
hotel,
you
know
process,
but
just
I'm
just
suggesting
just
bringing
it
up.