►
From YouTube: 2021-01-20 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).
B
Yeah
chili
for
sure
was
your
say:
capture
flag,
the
cft.
What
is
your
hat?
Oh,
it's
ef,
yeah
yeah,
an
angle.
It
looked
like.
I
don't
know
it
looked
weird.
E
A
E
A
F
A
A
A
C
B
I
I
lurked
for
a
bunch
of
it,
although
we
had
our
holiday
party
in
the
middle
of
it,
so
I
I
don't
know,
I
don't
think
I
don't
think
well.
I
was
not
interested
in
any
of
the
topics.
B
B
And
there's
a
new
plan
put
forward
by
maybe
josh
surth
from
google.
Anyway,
I
don't
know
I
didn't
find
any
of
it
very,
very
illuminating
or
useful.
E
B
Well
the
previous.
The
open
metrics
folks
tried
to
convince
us
that
we
need
to
rewrite
our
apis
to
look
more
like
open
like
prometheus,
and
I'm.
B
B
It
doesn't
work
very
well
with
statsd
so,
and
I
think
datadog
is,
would
not
be
happy
with
that.
E
E
Are
they
going
to
contribute
it
and
can
go
off
and
continue
developing
it
separately.
B
So
yeah,
I
I
don't
know
what's
going
on
but
yeah,
it
sounds
like
the
the
first
they're,
basically
breaking
metrics
up
into
two
waves.
The
first
wave
is
get
the
collector
to
be
able
to
properly
actually
collect
prometheus,
open,
metrics
and
expose
them
and
then
worry
more
about.
C
E
If
you
don't,
if
you
don't
care,
we
should
go
with
the
easy
route
of
the
the
thing
that's
already.
B
A
C
E
E
Going
some
no,
I
was
gonna,
say
going
nowhere
fast,
but
we'll
get
there
yeah.
We
need.
C
B
C
C
B
B
E
A
C
About
this,
since
last
week
I
mean,
I
think
the
date
is
achievable
for
us.
I
just
had
never
heard
of
it
and
it's
frustrating
again
to
hear
that
none
of
you
guys
have
heard
of
it.
B
I
think
it's
achievable
for
us.
If
they've
we
freaking
stabilize
the
spec.
B
B
B
Right
and
the
default
guns
default
resource
hasn't
been
merged.
C
B
So
anyway,
I
think
it's
achievable
if
the
spec
gets
stabilized
within
this,
like
I
would
say,
it's
got
to
be
this
week.
B
Otherwise,
we're
just
I
mean
it's
already
the
20th
tomorrow
right
so
yeah.
We
want
to
actually
get
all
the
changes.
We
were
released
candidate
and
then
yeah
anyway.
C
E
I
question
what
you
meant
by
this
comment:
the
field
adding
this
to
proto
would
be
clearest.
C
Maybe
just
the
word
is
too
similar,
which
is
why
you
didn't
notice
the
difference.
One
is
instrumented.
C
C
C
B
E
A
C
E
B
Oh,
I
think
so
too.
I
think
that
just
the
the
spec
folks
are
like,
let's
actually
write
it
down,
so
that
we
everyone
agrees
and
we
actually
have
documented
what
what
it
means
to
be
breaking.
E
E
The
propagation
there
you
go,
there's
the
yes
the.
What
do
you
call
it.
C
E
C
E
B
E
It
might
mean,
like
the
library
you
know,
we
would.
B
E
Deal
with
different
versions,
because
we
don't
we
don't
want
to
major
bump
the
library
like
instrumentations,
and
I
think
I
I
would
be
okay
with.
I
think
it
would
be.
Okay
to,
I
don't
know,
break
a
span
name
in
a
library
instrumentation
without
a
major
bump.
C
E
C
Yeah
well,
anyways,
that's
still
a
discussion,
that's
happening,
but
span
name
seems
like
the
hardest
one
actually,
because
we
have
had
it
like
servlet
or
something
we've
updated.
The
span
names
to
be
better
before
right.
I
just
remembered
that
it's
like,
if
that
happens
after
one
or
whatever
we
do.
E
Yeah,
it's
also
happened
when,
like
the
spec
gets
around
to
defining
a
span
name
for
something
like
the
message
messaging
and
then
we
sync
up
to
how
they
spec
what
the
spec
define.
C
E
C
B
Anyway-
and
the
merging
of
this
doesn't
matter
to
me
either
way,
I
don't
know.
E
C
E
I
think
maybe
you
missed
the
last
two
words.
E
Yeah,
I
read
it
as
I
mean
he
didn't,
approve
it
right,
it's
a
four-line!
It's
a
four
line,
change
that
he
didn't
want.
C
A
E
C
A
E
So
did
you
see,
did
you
see
the
the
the
let's
see
yeah
the
the
previous
discussion
about
why
we
removed?
Because
we
had
the
version
number
in
there
originally
and
we
took
it
out.
E
No,
so
the
so,
let's
go
to
john's
pr
clarifies.
E
E
Part
is
not
controversial.
Okay,
it's
the
and
I
don't
think
it's
super
controversial,
but.
C
D
C
A
C
E
The
implementation.
C
Regardless
of
our
naming
like
that
field,
has
it's
come
up
a
few
times?
Lately
people
do
get
confused
how
to
even
fill
it
in
like?
Is
it
supposed
to
be
the
version
like
which
we're
like
I've
seen
that
come
up
before
also
and
people
just
get
confused,
what's
instrumentation
versus
instrumented,
so
if
they
just
had
both
fields,
they're
self-documenting
in
some
way.
E
Yeah,
I
mean
it
makes
a
lot
of
sense.
It
also
matches
the
spec
glossary
and
I
know
there's
been
a
bunch
of
kind
of
discussion
on
that.
On
the
the
specific
turns
of
you
know:
instrumented
and
instrumentation
instrumental.
A
E
Spec
pr
contradictory
statement,
this
is
kind
of
a
you
know,
more
confusion
over
instrumentation
library
versus
instrument,
yep
instrumented.
E
E
Anything
in
the
anything
exciting
going
on
in
the
java
repo-
oh
you
all
saw
the
the
exciting
thing
in
the
java
instrumentation.
Repo,
of
course,
is
the
auto
configure
using
your
new,
auto
configure
module.
C
I
guess
still
not
sure
what
the
best
direction
is
as
far
as
those
properties
is
didn't
mind.
This
approach
that
much
where
we
copy
system
properties
and
one
caveat
as
I
am
expecting
that
we'll
have
a
class
which
exposes
publicly
all
the
attributes
that
are
supported,
because
we
were
thinking
about
having
constants
for
those
anyways.
E
E
E
C
E
Reason
not
to
do
that,
it
doesn't
feel
good.
It
doesn't
feel
great.
It
feels,
like
I
don't
know
like
first.
I
can't
think
of
why
I'm
resisting
polluting
the
system,
properties
and
even
calling.
E
So
I
don't
know,
I
guess
doesn't
seem
like
something
we
have
to
change.
Yeah.
C
B
Approach,
the
only
thing
I
wanted
to
talk
about
is:
let's
get
some
stuff
merged
over
in
java
land.
B
B
Yeah
yeah,
I
so
that
one
I
didn't
know
if
anyone
wanted
to
take
a
look
and
make
sure
that
I
deprecated
properly
since
I
had
to
do
it
all
by
hand.
Oh,
I
couldn't
figure
out
a
way
to
auto
generate
them.
B
D
B
C
C
C
E
Is
there
a
global
for
the
sdk,
the
open,
telemetry
sdk.
E
Yeah,
I
think
if
there
was
a
global
sdk,
then
I'd
be
more.
E
Yeah
but
yeah.
C
E
That's
a
yeah,
no
that
that
I
feel
like
that,
mitigates
that,
because
this
is
this
was
a
concern
I
had
initially
also
for.
B
This
reminds
me
of
something
I
was
going
to
ask
you
all
about
based
on
jason
jason,
and
I
were
talking
about
what
would
you?
How
would
you
feel,
or
maybe
there's
something
already
in
place
for
this-
about
instrumenting,
the
sdk
in
the
user's
class
loader
and
logging
errors,
if
they
try
to
do
do
stuff
with
it,
when
the
agent
is
running.
C
C
C
A
E
E
C
A
A
A
Yeah-
and
I
don't
know
in
my
in
my
brain,
I'm
like
the
way
the
way
to
maybe
handle
it
or
address
it
is
like
unless
you're
developing,
open,
telemetry
don't
use.
This
like
there
needs
to
be
like
these
really
clear
posters.
That
say
this
is
not
for
you
or
unless
you're
building
an
extent
like
I
don't
even
want
to
say
extension
like
there's
got
to
be
some
other
sort
of
word
that
says,
unless
you're
building
an
agent.
E
C
C
C
A
Yeah
so
I
got
lost
today
and
didn't
circle
back,
but
I
think
I
think
maybe
just
having
it
like
some
additional
words
in
the
readme
in
the
examples
which
is
where
I
was
calling
from
or
pulling
from
like
just
some
readme
notes
in
there
might
be
helpful
and
I
wanted
to
do
some
of
that,
but
got
lost
it's
gotten
busy.
A
C
A
E
It
certainly
hasn't
slowed
down
in
terms
of
volume.
It
definitely
it's
slowed
down
in
terms
of
like
math
like
breaking
changes,
but
you
guys
are
still
churning.
B
Yeah,
it's
actually
it's
worth
clicking
on
you'll
note:
there's
a
1.0
milestone
in
here
now
I
don't
know.
If
anyone
has
seen,
I
created
a
1.0
that
I
think
is
all
the
things
that
we
need
to
get
done
before
we
get
released.
B
Well,
this
is
still
a
moving
target
in
the
spec,
so
I
would
be
super
happy
to
move
all
of
the
shim.
The
open
tracing,
open
census,
stuff
out
of
1.0,
I'm
being
super
unclear
at
the
moment.
C
E
Sense
that
now
that
ga,
like
it
sounds,
seems
like
it's
a
ga
thing
like
the
idea
of
replacing
ot
and
oc,
but
not
a
1.00
thing,
which
1.0
is
just
about
api
stability.
C
B
D
B
B
C
C
A
D
B
Although
I
think
it
like,
I
said
it's
really
hard
to
do,
yeah,
that
specific
class
is
very
hard
to
actually
figure
out
how
to
effectively
benchmark,
but
judah
said
he
would
do
it
so
anyway,
but
it
doesn't
sound
like
he's
done
any
he's,
I
haven't
heard
him
actively
working
on
it,
so
I
would
be
happy
to
pull
that
out
as
well
that
big,
the
big
one
is
going
to
be
those
synth
tags.
B
E
B
E
E
A
Is
that
synth
tag
just
going
to
be
at
the
class
level,
hopefully
yep,
yeah,
okay,
so.
A
Then
then
you
have
a
real
maintenance
challenge.
A
E
Although
he
did,
he
did
approve
the
auto
configure
merge
into
the
java.
B
Agent,
cool
cool
yeah,
I'm
I
was-
I
was
struggling
to
figure
out
what
to
work
on
next
today,
just
because
I
didn't
want
to
start,
I
didn't
want
to
disrupt
anything
so
anyway
that
pru
commented
on
honorag
about
jaeger.
C
B
C
C
B
B
B
D
B
There
was
yes,
there
was,
and
in
that
meeting
there
was
discussion
about
the
zipkin,
not
protocol,
but
the
the
encoding
like.
E
B
Protobuffer
json,
because
there's
a
pr,
that's
basically
saying
that
there
needs
to
be
support
for
a
new
environment,
variable
that
will
specify
which
of
the
which
of
the
encodings
you
want
to
use
and
that
oh,
not
otlp
protobuf
should
be
the
default,
and
I'm
just
like.
We
can't
be
changing
stuff
like
this
at
this
late
date.
We
want
to
be
releasing
in
a
couple
weeks
like
we
can't
go
and
just
start
flinging
crap
at
the
wall.
C
B
C
B
B
C
C
B
B
B
Anyway,
that
service
name
mapping
minus
there
is
that
pr
that
I,
that
you
took
a
look
at
already.