►
From YouTube: 2022-09-27 meeting
Description
cncf-opentelemetry meeting-2's Personal Meeting Room
A
B
A
The
I
feel
like
the
Genesis
of
ruby
gems,
has
made
me
understand
a
lot
about
like
software
packaging
I,
don't
know
a
lot
of
the
details
but
like
it
was
just
people
like
uploading
stuff
to
like
being
like
here's
some
code,
that
I
use
that
you
might
find
helpful.
And
then
you
like
fast
forward
like
25
years
and
like
900
million
downloads
of
like
rails
or
whatever
half
of
the.
A
B
C
B
C
To
figure
out
how
to
get
a
critter
out
of
my
fireplace
this
is
a
rigged
like
Zoo
entrapment,
with
a
little
Plexiglas
in
Casey.
Just
has
to
come
down
raccoon.
C
C
D
C
C
C
D
D
A
Yeah
I
I'm
I'm
inclined
to
skip
skip
that
without
without
without
the
reader
digest
version
exactly,
but
I
don't
have
that
kind
of
mental
equipment.
To
make
that
synthesis
happen.
A
D
I
know
that
one
of
our
long-standing
agenda
items
is:
do
we
need
to
release?
The
contributory
vote?
Is
that
the
case
this
week.
A
What,
let's
see
I,
don't
know
the
latest
commit
on
contrib?
Is
a
release,
commit
so
I
think
we're.
A
C
Rack
Lobster
is
now
in
a
different
gem
or
elsewhere
in
gems
that
broke
Honeycombs
instrumentation
of
it
too.
What
is
rack
Lobster,
rack
Lobster,
because
Ruby
people.
D
No
I
mean
I
get
the
joke,
actually
that's
kind
of
good.
It's
it's
actually
quite
quite
clever.
Yeah
I,
like
it
actually
I'm,
actually
just
wondering
what
the
heck
is.
It
yeah.
C
B
D
B
A
B
F
I've
been
looking
at
that
one
I
gotta
leave
a
comment
on
it:
you're
talking
about
like
adding
the
route
as
the
Spanish
ampr
right,
the
action,
Factor
yeah,
the
like
the
history
on
that
one
I
think
just
to
share
here,
was
I,
actually
had
the
pr
to
do
that.
Originally
Once
Upon,
a
Time
I,
don't
know
if
my
findings
were
incorrect
or
out
of
date
now,
but
like
it
was
kind
of
heavy-handed.
There
was
a.
F
It
was
a
lot
of
work
to
pull
that
out
on
every
single
request,
whereas
just
getting
the
controller
name
in
action
was
very
trivial
to
grab.
So
that's
why
the
recognized
route,
adding
it
as
an
attribute
was
a
configurable
option
is
saying
that
not
everyone
uses
this
instrumentation
should
have
to
pay
the
cost
of
recalculating
the
entire
route
and
doing
that
whole
shebang.
There's
a
lot
of
code
behind
it.
F
Maybe
it's
super
fast
I
need
to
maybe
Benchmark
it,
but
that
was
the
reason
for
it
not
using
that
and
that's
I
think
something
that
needs
to
come
up
on
the
PRN.
But
I'll
leave
that
as
a
comment
and
say
that,
like
it's
okay,
we
can
add
this,
but
we
need
to
put
it
behind
a
configuration
flag.
Not
everyone
should
pay
that
tax.
D
I
would
be
curious
how
heavy
the
tax
is
because
I
mean
rails.
Obviously,
does
this
for
every
request
that
comes
in
right,
like
it
runs
recognized
route
and
does
all
of
that
I?
So
I
can't
imagine
that
it's
extraordinarily
heavy,
but
I
would
be
curious
to
know
like
how
heavy
it
is
and
if
we
can
cash
it.
Maybe
you
know
yeah
yeah
and
that's
yeah.
That's.
F
The
thing
is
looking
at
how
much
it
actually
costs
and
it's
rails
has
to
do
it
for
every
single
requests,
but
do
we
want
it
to
have
to
do
it
twice
for
every
single
request,
right.
B
F
I
think,
like
the
defaults,
that's
it
is
a
good
question.
I
think
the
default
probably
should
be
something
closer
to
the
semantic
convention.
So
I
think
there's
like
that's
a
reasonable
thing
to
error
to
I.
Don't
know
I
mean
technically
that
becomes
a
breaking
change,
but
all
of
the
instrumentation
is
experimental
and
subject
to
Breaking,
I
guess
so,
I
don't
know,
I
I
think
it's
I
think
it's
a
reasonable
thing
to
do,
because
it
makes
it
consistent
across
other
instrumentation
libraries
and
I.
Think
that
is
a
decent
guiding
principle.
B
Makes
sense
we
could
also
so
poc
in
the
past
is
like
leveraging
if
we
do
choose
to
go
1.0
with
Rack
or
rail,
you
know,
or
the
action
pack
stuff
or
any
rail.
So
it's
like
using
1.0
as
an
opportunity
to
introduce
these
things
that
we
know
we
want
to
do
and
just
let's
be
ex.
B
You
know
in
this
way
it's
very
explicit
that,
like
yep,
it's
breaking
but
we're
doing
it
on
a
major
release
so
that
there's
no
excuses
for
why
you
didn't
see
it
and
you
know
basically
bundling
up
a
bunch
of
small
breaking
things
we
want
to
do
until
I
want
to
you
know
in
a
way
kind
of
delaying
doing
those
until
we
do
the
1-0
but
I
don't
know,
also
not
trying
to
make
a
mountain
out
of
a
molehill.
F
F
This
particular
instrumentation
library
is
kind
of
feels
strange
because
it's
like
what
semantic
conventions
is
it
following.
It's
like
this
little
add-on
that
modifies
a
rackspan
and
rails
right.
You
know
if
it
was
like
rap,
instrumentation,
I
think
it'd
be
a
little
bit
different
or
even
like
any
of
the
HTTP
libraries
that
have
proper
like
stability
in
there
I'm
just
saying
that
it
that
one,
you
know,
could
not
come
for
a
very
long
time.
B
B
E
I
opened
up
a
couple
of
PR's
one
of
the
most
ad,
like
the
metrics
reporter
to
Zipkin,
so
that
I
could
finally
get
that
hey,
you
could
add,
a
custom
mess
recruit
reporter
through
the
configurator
that
I
opened
three
months
ago.
Thank
you
very
much
to
Chris,
who
did
the
Jaeger
one
Chris
Holmes,
so
shout
out
to
Chris
Holmes
to
unblock
that
portion.
C
E
Chance
to
take
a
look
at
that
one:
it's
about
renaming.
The
HTTP
connects
fan
to
something
else.
I
didn't
have
a
good
name
for
it,
but
I
wanted
to
disambiguate
that
it's
not
the
HTTP
connect
protocol,
but
rather
it's
a
span
measuring
the
time
it
takes
to
open
up
an
SSL
connection
and
start
an
HTTP
session.
So
I
want
to
disagree
with
that,
because
I
confuse
some
folks
at
GitHub
when
they
were
like
wait,
a
minute
you're
not
connecting
to
a
proxy
server.
Why
is
this
span
showing
up
as
HTTP
connect?
E
So
that's
that's
what
we
wanted
to
that's?
What
that
that
PR
for
but
again
I
picked
HTTP
session,
not
a
great
name,
I,
don't
know
what
else
I
think
about.
Was
it
established
connection
I?
Don't
it
was
weird
I,
don't
know
what
to
do.
B
I
I've
not
looked
at
these
PRS,
but
thank
you
for
servicing
them.
A
Yeah
I
looked
at
that
PR
and
I
was
like
I,
don't
I,
guess
I,
don't
know
how
HTTP
actually
works,
but
it
does
seem
if
it's
not
doing
the
HTTP
connect,
actual
thing
and
I
definitely
support
renaming
it.
A
A
A
F
E
C
F
I,
don't
I
don't
have
any
the
naming
I
chose.
Obviously
it
wasn't
meant
to
follow
any
convention.
It
was
just
mirroring
the
name
of
the
method
that
I
instrumented.
So
if
anyone
is
wondering
what
the
shallow
reasoning
was
that
that's
quite
literally
it
I'll
have
a
look
at
this
one
get
a
sense
of
what
the
name
session
might
make
sense:
I,
don't
I'm,
not
that
familiar
with
the
Affinity
gritties,
so
I'll
throw
it
in
front
of
Francis
as
well,
because,
like
all
things,
he
probably
has
an
opinion.
E
I
think
HIV
session
literally
is
a
straw
man
because
there's
no
such
thing
really
in
the
protocol,
so
it's
kind
of
like
it's
all,
misleading
in
a
sense
right
where
it's
like.
Oh
there's,
another
method,
HIV
method,
it
is
not
so
that's
just
an
attempt
at
like
getting
somebody
to
like
spin
on
it
or
poopoo
or
whatever.
C
Yeah
somebody
wrong
on
the
internet
method
of
provoking
a.
F
Think
we're
doing
this
or
this
this
similar
span
exists
in
other
instrumentation
libraries
as
well
for
just
the
other
one,
not
the
look,
so
we
change
it
here.
We
should
also
change
in
the
other
places
once
we
settle
in
the
name.
A
I
noticed
there's
a
couple
PR's
that
are
approved
that
aren't
merged.
There's
a
race
car
instrumentation
and
a
ad
Force
flush
option
to
rescue
I.
A
Sadly,
don't
have
a
lot
of
context
on
either
of
those,
but
I
guess
is
anyone
prepared
to
you
know,
merge
and
release
them
or
do
we
need
to
try
to
prioritize
getting
another
set
of
eyes
on
those.
E
B
E
Think
we're
waiting
yeah,
that's
definitely
my
fault
for
keeping
y'all
waiting,
because
I
think
also
Francis
had
some
commentary
in
there
too.
So
I'm.
B
E
E
C
E
As
it's
just
one
F-bomb
and
then-
and
it
is-
and
it
is
not
like
sexual
in
nature-
then
you
get
PG-13
right,
yeah,
all
right,
whatever
digression
I
accidentally
undid
your
review,
Eric
I'm,
sorry
for.
B
A
I
mean
the
race
car,
it's
like
new
instrumentation,
so
I
don't
feel
like
you're,
not
necessarily
gonna,
break
anyone's
stuff
anyway.
I
should
just
review
it
because
we
don't
want
to
have
broken
software,
but
mistakes
don't
seem
super
high.
E
Shoot
the
fourth
flush
one
right
wasn't.
D
A
B
A
B
B
E
Think
so
all
right
I
do
have
two
questions,
then,
if
if
people
want
to
talk
a
bit
shop
over
here,
so
what
is
the
story
with
the
latest?
One
I
just
saw
that
they
merged
instrumentation
scope,
attributes
right
and
that's
merged
into
the
spec
and
I
guess
do
like
I
guess.
How
do
we
want
to
handle
things
that
can
merge
into
the
spec?
Do
we
open
up
issues
and
then
just
put
them
like
on
a
this-
is
what
we
gotta
do
kind
of
thing.
E
D
It
does
I
I
have
some
thoughts
on
that
specifically.
The
first
is
that
we
provided
some
feedback
to
various
important
people
in
open
Telemetry
recently
and
I,
don't
remember
whom
or
for
what
purpose.
But
one
of
the
things
we
said
was
that
it
was
really
nice
when
changes
in
the
spec
were
proactively
pushed
out
to
the
sigs,
because
we're
not
all
watching
the
spec
cigs
in
in
that
repo
closely
at
least
one
request
has
so
far
been
pushed
to
us
that
way,
and
that
was
nice.
D
The
other
thing
about
it
was
that
I've
been
a
little
bit
lacks
in
Ruby
contributions.
The
past
couple
of
weeks
for
various
reasons,
but
that
was
one
of
the
things
I
was
just
kind
of
tackling
on
the
side,
was
paying
attention
to
when
changes
merged
in
the
spec
repo
and
then
opening
issues
on
our
repo
and
tracking
them
on
a
project
board,
so
I'm
on
my
phone
and
can't
link
very
easily.
D
But
if
you
go
to
like
the
hotel,
Ruby
projects
classic
tab,
there's
a
spec
compliance
project
and
there's
a
similar
one
on
the
contributory,
though
that's
tracking
everything
that
I
know
about
right.
Now
that
needs
to
be
done,
so
I
am
more
than
happy
to
continue
doing
that
on
behalf
of
the
group
since
I
know
it's
kind
of
work
that
is
not
necessarily
fun
or
easy
to
remember.
To
do
so,
I,
don't
mind
and
I
need
to
catch
up
for
the
113
release.
That
just
happened.
D
So
those
were
my
thoughts
on
it.
Don't
know
if
that
Sparks
any
more
discussion
or
thoughts
or
anything.
E
All
right
so
there's
like
the
proactive
pushing
out
to
the
communities.
I
guess
Eric
said
that
this
is
kind
of
like
on
all
of
us
to
do
sort
of
like
project
management,
kind
of
things
where
we're
like
to
inspect
compliance
reviews
or
was
there
anybody
like
that?
Essentially,
a
physical
role
like
I,
said
compliance
review.
I,
know
Andrew.
You
were
kind
of
doing
that
stuff.
C
A
D
Didn't
mind
it
wasn't
very
difficult.
A
lot
of
it
is
just
like
watching
for
a
new
release
to
be
cut
Upstream
on
the
spec
repo
and
then
kind
of
just
going
through
it
with
a
fine
tooth
comb
and
seeing
like
do
we
comply
right,
I,
don't
mind
continuing
honestly,
if
that's
fine
with
people
but
like
I'm,
also
more
than
happy
to
like
say
someone
else
can
do
it
too,
like
I,
have
no
real
ownership
feelings
of
it.
If
that
makes
sense,
no.
E
A
E
Figuring
out
some
sort
of
a
project
management
process,
that's
like
here's,
all,
the
oh,
here's,
all
the
repos
and
here's
their
Matrix
and
you
know
we've
released
whatever
Otep
got
merged.
Is
this
Otep
included
or
whatever
you
know?
That's.
D
An
interesting
thought
like
I
wonder
the
the
pushing
changes
out
proactively
was
something
they
were
receptive
to
and
were
happy
to
hear
about.
They
might
also
be
receptive
to
saying,
like
hey
as
a
condition
for
changing
the
spec
when
you
merge
a
spec
change,
PR,
there's
a
checklist
where
you
go
through
all
of
the
sdks
and
be
like
hey.
D
E
D
Yeah
because
at
least
then,
you
have
the
list
of
things
that
need
to
be
there.
It
eliminates
the
like
manual
review
step
that
I
was
doing
of
like
just
going
through
the
pr
and
being
like
okay,
what
were
all
the
things
that
actually
changed?
Some
of
them
not
always
reflected
in
the
change
along
yeah.
That
could
be
a
nice
improvement
too.
Actually
it
saves
some
manual
steps.
Definitely.
D
A
A
A
I
really
thought
there
was
anyway:
oh
J,
Mac
D
J
macd
opened
up
hey
an
issue
for
partial
success
responses.
D
A
I
feel
like
I'm,
getting
a
window
into
the
GitHub,
the
GitHub
team
and
Andrew
and
Arielle
talk.
You
know
what.
B
B
E
That
could
be
up
y'all
all
right,
then,
if
it
ain't
nothing
else,
you
know
Robert
welcome
back
from
vacation
and
I
hope
you
moved
into
your
new
house
and
it's
nice.
F
F
F
F
A
F
D
Well,
on
that
note,
I
have
to
drop
off,
but
I
will
talk
to
all
of
you.
Wonderful
people
later.