►
From YouTube: 2021-01-06 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
Okay,
whoever
is
on
the
call-
please
add
your
name
to
that
in
this
list.
I
think
we
can
start
now.
We
don't
have
a
lot
of
agenda
so
we'll
just
go
ahead
and
start.
B
A
Okay,
so
so,
as
most
of
you
are
aware,
we
did
not
have
meetings
in
the
last
two
weeks
of
december,
so
this
is
essentially
a
first
meeting
of
this
year
so
happy
new
year,
everyone
and
if
you
are,
if
you
were
taking
vacation,
welcome
back.
I
was
personally
on
vacation,
so
just
joined
back
work
monday.
A
I
have
mentioned
in
jitter
that,
like
we
are
still
going
through
the
list
of
open
issues
which
were
open
in
the
last
two
weeks.
Interestingly,
we
have
a
lot
of
pull
requests
which
would
happen
in
the
last
15
20
days.
So
I
think
I
like
merged
like
three
or
four
of
them
in
last
two
days,
I'm
still
working
through
them,
so
please
bear
with
us,
while
we
clear
the
backlog
in
the
next
few
days
that
I
said,
let
me
go
get
to
the
agenda
items.
It's
very
small.
C
So
before
we
go
kenny
introduced
himself
on
chat,
can
you
make
feel
free
to
jump
in
and
introduce
yourself.
D
Hey
guys,
I'm
kenny
long,
I
work
at
st
jude
children's
research
hospital
and
we're
we're
starting
to
implement
open
telemetry
with
our
we
got
a
couple
of
hundred
micro
services,
and
so
I
wanted
to
get
involved.
So
I
appreciate
you,
let
me
hang
out.
E
D
D
Yeah,
I
am
it's
kind
of
an
architect
position,
but
it's
more
of
a
hands-on
architecture,
type
position.
We
have
about
50
java
microservices
and
about
150.net
core
microservices,
and
we,
when
we're
rolling
out
a
new
api,
we
have
a
template
and
we
everybody's
been
using
the
template.
So
we
have
this
kind
of
nice
uniform
set
of
services
and
we
want
to
add
open
telemetry
to
that
and
get
that
rolled
out
to
all
our
services.
We
have
instantan
in
place
right
now,
but
they
do
a
pretty
poor
job
with
the.net
core
stuff.
D
They
do
great
with
java
and
that's
kind
of
what
we're
seeing
with
a
lot
of
the
off-the-shelf
commercial
stuff
and
so
instanta
is
implementing
open,
telemetry
and
and
so,
if
they
don't
do
a
good
job,
we
have
a
lot
of
options,
for
you
know
visualization
tools.
So
so
we
feel
like
it's
a
good
move
for
us.
D
Right
now
in
stannah,
so
we'll
see
how
that
goes.
If
they,
if
they're,
if
their
collector
is,
is
good
or
you
know,
their
visualization
part
is
good.
If
not
we
we
have
jaeger
that
we
use
for
some
other
things.
So,
okay,
we'll
probably
use
jaeger.
A
Oh
yeah,
thank
you.
We
do.
I
probably
responded
to
you
then,
like
I
did
really
said
it
was
the
same
person.
You
sure
did
I'm
kenneth
along.
Oh
okay,
yeah.
D
A
Hi,
well,
I
don't
hear
anyone,
so
let's
get
back
to
the
next
item
so
like
just
to
like
bring
back
this
topic
with
the
questioning
and
stability.
We
started
with
a
draft
pr
before
I
went
on
vacation,
which
was
december
mid
in
response
to
the
open,
telemetry
otep,
which
was
about
versioning,
so
that
whatever
got
merged
and
now
there's
a
spec
which
is
made
out
of
it,
it's
still
being
discussed.
So
I
encourage
everyone
to
go
and
look
at
the
specification.
A
It's
it's
already
got
like
tons
of
comments.
So
please
take
a
look.
The
there
were
like
two
action
items
on
me
based
on
like
the
versioning
strategy.
I
haven't
completed
it.
I
am
still
like
literally
working
through
that,
so
I
will
not
have
it
today,
so
I'll
be
sharing
it
over
the
course
of
next
couple
of
days
and
definitely
before
the
next
sig
meeting.
It's
mostly
about
like
listing
down
step
by
step.
A
What
exactly
would
it
involve
because
I
listed
two
approaches
on
how
do
we
deal
with
questioning
one
is
to
continue
with
the
current
approach
of
using
metrics
as
a
absolute,
but
with
the
same
package.
The
alternate
was
to
extract
matrix
into
its
own
package,
so
we
have
like
some
logistic
challenges
to
deal
with
if
you
want
to
go
with
separate
packages
specifically,
how
do
we
ship
different
packages
with
different
versions
from
the
same
repo?
But
apart
from
the
logistics,
we
do
have
like
some
issues
to
still
iron
out.
A
Some
of
them
are
dot
net.
Specific
others
could
be
like
valid
across
other
languages,
because
I
have
seen
like
people
commenting
similar
issues
in
this
pr
as
well.
I'm
mostly
concerned
about
how
do
we,
like
flip
the
switch
once
we
once
we
are
ready
with
the
matrix?
So
what
would
be
the
processing
world?
A
So
those
are
the
things
I
am
still
like
writing
down
the
process
as
soon
as
it's
more
ready,
I'll
update
the
pr
so
that
I
can
get
more
comments
because
we
did
bring
up
like
several
new
points
on
during
the
last
meeting.
So
I
just
need
to
like
organize
them
into
proper
thought
process
then
list
it
down
in
the
pr
itself,
so
everyone
can
get
a
chance
to
comment
on
it.
A
Yeah
I'll
I
mean.
The
second
action
item
was
about
the
logging.
As
we
discussed
last
week,
there
was
some
confusion
that
knight
was
introducing
a
new
open,
telemetry
logging
api.
I
wanted
to
clarify
it
in
some
talk,
or
at
least
an
issue.
We
are
not
introducing
or
shipping
a
new
logging
api.
We
are
just
providing
a
I
logger
adapter,
which
is
like
there
is
no
new
api.
The
api
is
from
logger,
which
is
already
a
stable
package
for
several
years
and
we'll
be
just
adding
capability
to
correlate
that
locks
with
tracing.
A
I
I'll
I
list
out
this
issue
in
detail,
because
there
were
some
confusion
that
we
were
releasing
a
logging
api
before
the
specs
are
even
stable,
so
I'll
create
an
issue
just
to
like
explain.
What
exactly
is
that
we
are
doing
to
avoid
any
confusion
and
potentially
code
locks
us
also
experimental.
If
there
is
a
need
to
do
so,
but
so
far
based
on
what
we
discussed
last
week,
there
is
no
need
to
do
so,
but
we'll
discuss
it
one
when
I
open
the
issue.
A
So
this
is
just
an
update,
saying
that
I
do
not
have
any
update
about
the
action
items
from
last
week,
but
I
will
be
having
something
before
next
week.
A
Last
thing,
which
I
put
here
is
it's
like
dot
net
five
has
already
shipped,
so
the
next
major
version
from
dotnet
is
dot
net
six.
So
unless
there
is
a
critical
bug
fix
any
feature
request
which
we
want
from
dot
net
is
going
to
come
in
dot
net
six
time
frame,
which
is
around
20
21
number
this
year.
But
the
coding
is
almost
complete
by
mid
this
year
like
june
july
time
frame.
A
So
if
there
are
asks
based
on
open
telemetry
needs,
now
is
the
right
time
to
bring
it
up
to
the
runtime
repo
or,
if
it's
a
specific
issue
to
sp
net
core
in
the
sp
net
core
issue
in
the
repo.
So
I
just
like
shared
like
couple
of
sample
things
which
I
have
opened
myself.
A
I
will
be
opening
more,
but
if
you
do
have
a
need
which
is
not
captured,
please
bring
it
up,
because
we
are
like
almost
at
the
edge
where
the
planning
cycle
internet
is
about
to
end
and
they'll.
Consider
issues
only
if
you
bring
it
up
in
time.
So,
like
couple
of,
I
can
just
quickly
show
what
I
have
like.
We
are
asking
to
expose
a
propagators
api
which
would
be
compatible
with
the
open,
telemetry
one.
So
right
now
the
propagation
is
like
hard
coded
into
asp.net
core
and
http
client
libraries.
A
So
this
is
one
issue,
and
I
do
have
like
to
be
open
for
status
status
was
not
spec
wise.
It
was
not
ready
when
dotnet
file
shipped,
but
now
it
seems
steady.
I
actually
meant
span
dot
status,
but
in
dot
net
it
could
be
activity.
A
and
we
also
want
to
get
some
package
improvement.
Some
work
is
already
in
progress
in
dotnet
report
to
support
the
new
header.
The
header
name
was
changed
from
correlation
context
to
baggage,
so
dotnet
is
already
doing
some
work
to
respond
to
that,
and
then
there
are
few
issues
which
is
open
in
the
open,
elementary
repo,
which
is
the
repo
which
we
control,
but
it
should
be
like
brought
into
the
dot
net
team.
A
A
This
another
us
the
spec
for
this
quote
merged
sometime
last
month,
so
there's
an
ask
to
support
customized,
trace
id
and
spanner
degeneration,
and
this
is
something
which
has
to
come
from
runtime.
So
we
need
to
open
an
issue
in
the
runtime
to
discuss
this
and
how
the
ap
would
look
like
things
like
that.
So
far,
it's
I
mean
there
is
a
aws
requirement
that
there
isn't
a
trace
id
should
follow
a
particular
format,
so
they're
using
a
somewhat
hacky
workaround.
A
A
A
That
brings
and
oh
one
final
thing
I
missed.
If
there
is
a
bug
which
is
like
really
serious,
we
don't
have
like
dot
net
six
in
mind
like
it
can
be.
I
mean
if
it's
serious
enough,
it
can
be
released
as
a
hotfix
in
the
5.0
release.
I
think
one
or
two
bucks
were
already
addressed,
which
was
raised
by
open
elementary
community.
A
I
I
will
set
this
status
should
be
released
this
week,
yeah
it's
not
yet,
so
it
should
be
released
like
5.0.01,
which
contains
like
one
or
two
bug
fixes
which
was
raised
by
this
community.
So
please
keep
reporting
bugs
if
it
is
really
severe.
We'll
do
I
mean
when
I
say
b
I
mean
the
dot
net
team
will
do
it
as
a
batch
release
in
5.0
for
anything
major.
We
have
to
wait
until
night,
six.
A
This
is
yeah.
This
is
something
which
I
already
opened
that
ends
the
topics
which
I
wanted
to
share.
If
there
are
any
other
topics
which
needs
discussion,
please
erase
it
now.
B
Hey
cj,
this
is
bashan.
B
I've
been
away
from
the
sick
meeting
for
quite
some
time
now
so
yeah.
I
was
kind
of
wondering
like
what's
the
plan
for
the
package
releases
from
the
contrib
repo,
because
I
remember
last
time
we
talked
we
kind
of
like
discussed
some
of
the
ownership
of
the
contrib
repo
and
how
we're
going
to
plan
the
maintainers
and
the
release
process.
Do
you
have
any
update.
A
This
is
something
which
we
briefly
discussed
in
the
last
meeting
as
well,
so
it
looks
like
other
reports
like
java
and
other
languages
are
doing
a
different
approach
than
what
tornado
was
doing,
because
I
believe,
like
python,
including
python,
like
everyone
moved
all
the
instrumentations
also
into
the
country.
Repo
and
the
main
ripper
would
only
contain
the
things
which
are
required
by
the
spec,
which
would
basically
be
the
api
and
sdk
and
the
three
exporters,
three
or
maybe
four
exporters,
officially
required.
Everything
else
was
moved.
A
I
don't
know
whether
it
is
a
formal
direction
from
the
spec.
This
is
a
I
mean
there
was
an
open
pr
which
I
linked
the
last
week
about
like
what
goes
into
contrary
propose.
So
I
haven't
really
followed
up
into
the
official
guidance,
but
basically
I'm
what
I'm
saying
is
you'll
just
follow
whatever
other
languages
are
doing
and
whatever
is
the
direction
from
overall
open
elementary
community.
It
did
not
have
the
answer
today.
A
I
can
give
a
better
estimate
next
week
because
I'm
just
back
from
vacation,
so
I
would
be
looking
at
the
specific.
There
is
an
issue
already.
D
A
In
the
spec
crypto
about
like
specifically
asking
about
clarification
on
who
owns
or
what
should
go
into
the
main
repo,
I.
F
Think
I
think
siri
just
to
add
to
your
you
know
comments.
I'd
say
that
in
the
maintainers
meetings
at
least
you
know
it
has
been
discussed.
But
again
you
know,
after
people
are
back
from
vacation
this
week
could
probably
come
up
again.
I
think
the
general
agreement
has
been
that
the
contrib
you
know
again,
the
contrib
repos
would
have
the
you
know
anything,
that's
not
non-core,
not
spec,
specific
right
and,
as
you
outlined,
the
only
other.
F
I
think
comment
was
that
you
know
how
do
we
make
sure
that
we
can
also
participate
in
maintaining
you
know
some
of
those
components
that
we
are
contributing,
as
well
as
other
components
and
what's
the
easy
path
there
and
that's
more,
you
know
really
figuring
out
how
to
provide
approver
and
you
know,
merge.
D
F
Rights
for
for
the
components
and
then
to
guarantee
the
ability
to
actually
have
releases
that
we
can
test
with
right
and
artifacts
that
we
can
have.
F
A
F
A
F
Yes,
that's
correct,
and-
and
you
know
we
are
absolutely
committed
to
doing
that.
So
you
know
that's
why
I
think
we
need
to
figure
out
together
what
what's
the
best.
You
know
permissions
model
and.
A
A
Like
can
you
share
like
what
is
happening
in
other
languages?
In
is
the
same
model
being
followed
like
in
all
other
stakes?.
F
Yeah,
I
think
the
discussions
again
were
happening
in
the
maintainer
meetings
and
the
spec
sig
meetings.
You
know
last
month
and
and
and
the
idea
really
has
been
the
java
sig,
the
javascript
go
python,
you
know
and
of
course
dotnet.
You
know
all
of
you
have
participated
and
those
are
the
most
active
participants,
and
that
was
you
know
again
with
the
collector
idea
that
you
know
the
contrib
reaper
could
easily
host
all
the
aws
or
vendor-specific
components
as
well
as
other.
F
You
know,
non-spec
core
components.
A
F
Think
I
think
the
collector
has
done
that
to
a
certain
extent
and-
and
some
of
the
other
reapers
have
also
done
that
like,
for
example
java.
But
again
I
think
it's
because
we
really
haven't
written
down
a
you
know:
permission
sharing
model,
okay
or
you
know-
maybe
an
otep
is
in
order
here.
Maybe
we
could
jointly
work
on
it
together,
but
kind
of
put
it
down,
and
then
you
know
propose
it
to
the
community
and
then
have
that
discussion
finalized.
A
Yeah
and
like
are
there
like
any
special
naming
for
packages
coming
from
country
because
dotnet
we
initially
decided
we'll,
have
a
like
hyphen,
contrib
prefix,
to
the
name
of
the
packages
to
indicate
that
it
is
coming
from
a
separate
repo,
not
maintained
by
the
maintainers
of
the
main
ripple.
But
if
you
were
to
follow
what
other
languages
are
doing,
I
think
we
should
remove
that
special
marker,
like
all
packages.
F
Right
right
I
mean
exactly
so.
I
think
that
you
know
there's
some.
I
think
we
should
just
write
down
the
assumptions
that
you
know
have
been
kind
of
captured
over
different
sig
discussions
and
and
otep
and
other
issues
and
then
kind
of
just
write
it
out.
Yeah.
F
A
Okay,
let
me
add
my
comments
to
it:
okay,
I'll
have
to
find
it.
It
was
definitely
there
because
I.
A
Yeah,
so
just
to
write
down
like
what
specifically,
we
would
need
to
happen
in
dotnet.
We
need
to
get
rid
of
the.
A
A
We
identified
that
we
don't
really
need
like
sql,
probably
and
grpc,
and
stack
exchange
reduce
client.
So
those
things
are
right
now
in
the
main
ripple,
we
should
be
able
to
move
it
off
to
the
contrary.
F
And
and
then
maybe
define
you
know
like
who
can
who
can
trigger
or
when
do
we
trigger
builds
for
contrib.
A
Get
released
yeah,
so
that
is
something
which
we
have
to
still
figure
out,
because
the
way
currently
the
versioning,
this
has
come
up
in
the
version
right.
A
It's
the
same
like
it's
a
git
tag.
So
when
you
push
a
new
tag,
the
entire
project,
like
every
project
from
the
repo,
gets
a
new
build
with
the
same
version.
So
there
is
no
real
ability
to
like
individually
version
so
far,
but
that
is
something
which
we
may
need
to
solve
like,
irrespective
of
the
contrib
status
right.
F
Because
I
mean
on
the
project
it's
it
would
be
useful
to
actually
have
developer.
You
know,
I
mean
versions,
interim
versions
which
are
not
necessarily
stable,
but
are
artifacts
that
you
know
developers
can
test
with
right
and-
and
I
mean
that's
pretty
standard,
so
we
need
to
figure
out
is,
as
you
said,
you
know.
How
does
that
versioning
work
right.
A
Yeah,
I
mean
that's
something
which
we
are
still
figuring
out:
the
respective
of
the
contrib,
because
sure
to
even
extract
matrix
out.
A
G
C
Yeah,
this
is
something
that
we
need
to
clarify
because
originally,
when
we
discussed
it,
we
said
that
open
telemetry.com
trip
will
be.
My
country
will
be
a
significant
marker
that
is
not
coming
from
open,
telemetry
core,
meaning
that
it's
not
it
has
less
support
and
less
important.
So
like
every
vendor
has
a
incentive
to
not
be
in
in-country.
You
can
develop
their
own
like
vendor
specific
packages,
starting
with
vendor
namespace
yeah.
F
So
I
mean
that's
a
good
point
and,
and-
and
I
think
that
again
I
think
there
are
two
layers
to
that
right.
One
is
that
which
are
considered.
You
know
to
have
a
unified
user
experience.
That
is
if
a
customer
is
coming
in.
F
You
know
where,
what's
the
easiest
way
of
ensuring
that
the
core
packages
are
available
and
found,
which
are
you
know,
right
from
hotel
itself
and
then,
of
course,
there
are
the
vendor-specific
extensions
that
could
be,
you
know
always
can
be
developed,
but
with
guidelines
you
know
that
are
compliant
with
open,
telemetry.
E
C
If
the
incentive
is
to
be
like
more
user-friendly
is
to
be
in
this
repository,
it
may
be
a
wrong
consensus
incentive.
So
maybe
we
can.
I
mean
if
you
do,
that
we
probably
need
to
have
a
process
by
for
people
to
claim
namespaces
like
open
ceremony
dot
and
like
get
access
to
those
yeah,
which
is
a
little
bit
complicated.
I
don't
want
to.
F
Get
it
it
is
complicated
today,
so
I
mean
again:
if
you
do
propose
that,
then
we
actually
should
have
a
larger.
You
know
discussion
on
how
we
get
that,
because
I
mean
if
the
project
can
support
that
or
at
least
document
the
process,
so
that
any
vendor
can
come
in
and
actually
do
that.
You
know
set
that
up
if
they're
committed
long
term
to
the
project
and
also
you
know
contributing
to
the
project,
then
that's
useful
right.
C
F
A
So
sega,
I
missed
the
part
about
like
what
were
you
saying
about
the
name
prefix
like,
for
example,
if
we
I
mean
right
now,
we
have
like,
for
example,
an
aws
component
in
the
country
repo.
So
when
we
publish
to
nuget
it
will
still
be
like
prefixed.
It
will
not
have
any
aws
prefix,
so
it
will
be
just
open
elementary
dot
like
some
component
dot
aws,
it
will
be
owned
by
the
elementary
itself
right
when.
C
We
started
open
telemetry
country
prepository.
We,
for
instance,
asked
new
relic
to
rename
their
exporter
to
new,
relic
dot,
open
telemetry
exporter.
So.
D
C
Is
and
if
we
will
start
naming
all
other
vendor
experts
with
open
telemetry
prefix,
it
will
put
a
new
relevant
disadvantage
because
it
may
be
harder
to
to
discover
right,
so
people
type
in
open
telemetry
doesn't
see
new
relic,
but
those
do
see
like
fubar
vendor.
A
So
namespace,
I
think,
like
not
just
open
elementary
dot
like
it,
should
be
like
something
which
you
just
described,
like
open,
elementary
dot,
vendor
name
dot
whatever.
That
should
be
theirs
right,
yeah,
okay,
but
is
that
the
one
which
you
said
is
complex?
Because
I
don't
know
how
do
we
do
that?
Yet
it's
something
which
we
have
to
manually
contact.
Then
you
get.
I
mean
yeah.
C
A
I
see
so
it
would
be
on
us
to
first
figure
out
whether
there
is
an
option
to
do
like
sub
name
spacing
and
give
ownership
to
different
companies
so
that
open
elementary
dot
aws
can
be
owned
by
aws,
open
elementary
dot.
New
really
can
be
owned
by
neutrality,
but
the
overall
open,
elementary
prefix
is
still
retained
by
the
cmcf
account.
F
A
A
No,
I
mean
if
I
log
in
as
myself
I
mean
I
have
access
but
like
to
reserve
prefixes
it's
a
manual
process.
You
actually
may
be.
Oh.
A
C
Reserve
for
open
city
organization.
F
A
Okay,
so
whatever
we
do
with
contrib
repo,
this
is
going
to
be
an
issue
because
we
don't
want
to
give
like
any
vendor
like
special
advantage
by.
A
Okay,
so
that
is
something
which
we
have
to
solve
before
we
start
like
I
mean.
F
A
Okay,
yeah,
okay,
yeah,
and
that
I
said
like
there
is
a
separate
issue
in
control
report,
because
right
now
it
has
not
been
updated
to
the
latest
version.
It's
mostly
broken
because
we
un,
I
mean
unexposed
certain
apis
from
public
to
internal.
It's
it's
still
on
my
list.
To
do.
A
Eddie
did
some
pr
to
make
it
to
the
newest,
but
it
has
not
been
finished,
so
we
will
be
getting
to
it
like
little
later,
because
first,
we'll
clean
up
all
the
open
pr's
in
the
main
repo
and
next
step
would
be
to
actually
do
a
release
from
oh
sorry,
first
cps
actually
update
the
country
report
to
the
latest
version
of
the
sdk
and
then
do
a
release.
A
If
is
there
like
any
component,
which
you
are
specifically
looking
for
in
the
country
like,
I
know
that
there
is
an
awsp
are
waiting,
but
is
there
like
any
like
urgent
need
for
us
to
get
this
released
sooner
than
others,
because
everything
else,
we
probably
can
add
some
weight,
but
I
know
that
you
were
wanting
to
get
the
aws
sdk
part
done,
but
I
think,
like
prashant
did
mention
that
this
was
eventually
going
to
be
owned
by
in
a
separate
repo.
A
But
I
don't
know
what
is
the
current
status
like?
Is
it
still
planned.
B
A
A
Yeah
I
mean
updating
the
version
of
the
api
or
sdk
from
the
main
repo
should
be
like
super
straight
forward.
A
I
mean
the
current
state
is
very
specific,
because
all
these
all
these
I
mean
the
aws
one
probably
is
an
exception,
but
the
elastic
search
and
mass
transit.
They
depended
on
some
internal
classes
on
previously
public
classes
which
were
made
internal,
so
it
requires,
like
some
non-trivial
amount
of
work,
to
make
this
updated.
That's
the
reason
why
we
are
not
able
to
ship
it
from
this
repo,
but
that
is
only
a
temporary
issue.
Once
we
like
fix
that,
like
going
forward.
A
A
Don't
expect
like
any
any
those
kind
of
changes
once
we
do
it
one
dot
or
release,
because
then
we
have
more
guarantees
on
the
epas.
Yeah,
okay,
but.
F
Specifically,
the
x-ray
updates:
what
are
the
dependencies?
You
have.
A
B
I
don't
have,
I
think
this
should
be
okay.
The
propagator
is
broken
because
the
interface
has
changed,
which
I
think
is
this
minor
fix.
A
B
A
Yeah,
I
would
still
focus
on
like
getting
the
like
main
repo
to
like
a
more
manageable
state,
because
right
now
we
have
at
least
30
pr's.
It's
it's
too
much.
We
usually
have
less
than
10,
so
I
will
use
most
of
the
energy
to
like
bring
this
down
to
a
more
manageable
number
and
also
finish
the
questioning
part.
So
we
are
like
in
a
good
shape
to
at
least
1.0
I
mean
I
can.
A
A
Most
likely
that
because
of
the
timing
issues,
because
we
want
to
get
the
apa
and
sdk
out
as
soon
as
possible,
that's
what
people
are
asking
most
people
are
okay
to
have
like
that's
what
I
understood
from
previous
meetings
as
well.
It's
okay
to
release
like
an
instrumentation
for
elasticsearch
or
any
framework
a
week
or
two
or
even
month
after
the
actual
sdk
itself
is
released.
A
So
I'm
trying
to
focus
just
on
the
releasing
of
api
and
sdk
and
figuring
out
the
versioning
part
and
then
come
back
because
it's
not
trivial.
All
the
updates
are
not
trivial
because
we
have
to
like,
I
think,
eddie
started
the
work,
but
for
some
reason
it
didn't
work.
So
we
still
need
to
work
with.
A
This
is
specifically
for
mass
transit,
but
then,
of
course,
after
that,
we
have
to
look
at
other
elastic,
search
and
elastic
search
and
what
is
other
thing
entity
framework
to
see
whether
we
can
just
replicate
whatever
we
learn
or
we
just
need
like
something
else
or
in
the
extreme
scenario.
We
will
bring
back
those
classes
as
public
because
it
was
initially
public.
Then
we
decided
there
is
no.
A
It's
not
solving
any
core
open
elementary
problem,
so
we
just
made
it
internal
so
or
it's
quite
possible
that
if
you
just
copy
paste
all
their
classes
here,
like
literally
copying
that
classes
it
might
so.
F
That's
not
a
great
idea.
Right
short
run,
it'll
work,
but
I
mean
maintaining
the
classes
in
two
separate
places,
probably.
A
I
mean
some
of
the
helpers
are
like
it's
very
scary,
to
look
at
the
code.
It
works,
but
I'm
not
very
confident
of
exposing
them
as
apis.
I
think
they're
not
really
required,
but
it's
it's
like
high
performance
way
of
enumerating
our
activity,
it's
something
which
is
not
really
like
any.
There
is
no
open
element
record
here.
It's
just
like
pure.net
code.
So
that's
the
reason
why
it
was
made
internet.
A
So
yeah,
like
I
said,
like
it's
just
a
timing
issue
we
need
to
like
first
bring
the
main
repo
and
ship,
the
ap
and
sdk,
and
as
soon
as
that
is
done,
we
can
get
back
to
the
problem
of
updating
like
the
those
instrumentations
which
fail.
That
should
unblock
like
everything
else
in
this
report,
because
right
now,
you're
pretty
much
stuck
like
you
cannot
just
release
aws
yeah.
E
A
B
So
the
next
word
next
version
is
going
to
be
one
dot,
the
ga
for
for
the
api
and
the
sdk
right.
A
I
think
we
will
do
like
c2
ourselves,
because
we
we
have
like
a
lot
of
code
change
since
rc
one,
so
we
would
definitely
be
doing
one
more
release:
rc2
yeah,
it's
almost
a
month
since
more
than
a
month
since
we
did
the
last
release
so
so.
F
Cj
is
the
plan.
At
least
you
know.
Ballpark
is
to
have
a
release
every
month
like,
even
if
you
do
an
rc
on.net
or
or
is
it
just
based
on
stability.
A
Yeah,
it's
mostly
based
on
stability
and
the
need
like
so
right.
Now,
it's
bit
of
manual
process,
but
last
time
I
did
I
I
was
able
to
finish
the
whole
thing
in
like
less
than
20
minutes,
like
my
like
physically,
I
was
spending
like
less
time,
20
minutes,
but
it
is
easily
automatable.
A
F
A
A
A
E
So
I
was
gonna
ask
like
if
you
you
so
you're
talking
about
focusing
on
releasing
the
core
repo.
You
would
need
to
like
before
releasing
the
ga
on
the
core
repo
solve
this
contrib
problem
like
if
sql
instrumentation
is
going
into
contrib.
You
would
need
to
solve
that
before
you
do
the
core
release
right.
A
So
that's
what
I
what
I
learned
from
like
the
spec
is
the
first
thing
which
should
go.
Ga
is
the
ap
and
sdk
alone
without
any
of
the
instrumentations,
because
there
was
an
issue
open
like
two
day,
maybe
a
week
back
about
the
semantics
conventions,
since
they
are
not
declared
as
like,
stable,
stable
and
the
instrumentations
are,
depending
on
that,
we
should
not
be
like
calling
the
instrumentations
as
like,
1.08
oh,
but
this
is
something
which
I
learned
like
like
yesterday
in
the
maintenance
meeting.
A
So
I
I
mean
if
we
are
going
with
that
approach,
the
number
one
problem
which
we
will
solve
is:
how
do
we
assign
different
version
numbers
to
each
package
just
because,
right
now,
it's
just
single.
So
that's
the
problem
which
we
had
like.
A
That's
the
reason
why
I
said
it's
a
problem
which
we
have
to
solve,
irrespective
of
the
contract
status
just
because
apa
and
sdk
has
to
go
like
first,
we
have
to
probably
wait
for
instrumentations
for
the
specs
to
say
that
this
semantic
conventions
are
like
ready,
because
the
instrumentations
are
taking
a
dependency
on
that
semantic
conventions.
If
it
is
not
stable
kind
of
pointless
to
say
that
the
instrumentation
is
stable,
it's
a
relatively.
D
A
Issue
for
me,
I
don't
know
like
whether
someone
else
was
paying
attention.
It
was
the
daily
new
issue
yeah,
which
was
brought
up.
At
least
I
learned
about
it
yesterday
in
the
maintenance
meeting
so
based
on
that
we
may
have
to
do
like
step
by
step
release.
First,
we
do
like
api
and
sdk
then
followed
by
the
export.
I
mean
api,
sdk
and
exporters,
and
then
in
instrumentations
we
would
be
doing
a
separate
releases.
D
A
F
A
Okay,
so
it
looks
like
the
collector
has
the
highest
number
of
individual
components
like
every
vendor
has
some
component
there.
A
Yeah,
okay,
so
this
is
like
most
likely
like
once.
The
official
version
comes
out
like
this
is
how
the
contour
paper
for
dot
net
is
also
going
to
look
like,
like
we'll,
have
vendor
specific
things
with
okay.
A
I
think
this
would
be
like
a
logistic
challenge
just
to
get
it
done
like
first
place,
but
after
that
initial
thing
it
should
be.
A
Okay,
are
there
any
other
topics
to
discuss
by
the
way?
There's
a
general
ask
like
just
me
asking:
there
are
a
lot
of
open
issues
and
food
records,
so
if
anyone
has
free
cycles,
please
help
with
these
things.
I
will
be
getting
to
it
as
soon
as
possible,
but
except
like
few
things
which
we
were.
I
think
I
reviewed
this
one
today.
This
one
is
still
waiting
for
changes
so,
but
there
are
things
which
occurred
like
in
the
last
15
days,
which
are
just
waiting
for
reviewers.
A
A
All
right,
I
do
not
have
any
other
topic,
so
we
can
end
a
bit
earlier
and
next
week
I
will
come
back
with
the
or
even
earlier
I'll
have
the
more
solid
proposals
for
washing
and
stabilities
all
right
thanks.
Everyone
see
you
next
week
sounds.