►
From YouTube: 2020-05-27 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).
C
B
B
A
B
A
B
B
A
D
D
B
That
tends
to
be
the
way
things
work
in
open-source
software.
Is,
you
know
no
one's
paying
anyone
to
be
here,
so
it
tends
to
be.
The
people
show
up
when
they
can
and
do
what
they
can,
and
it
tends
to
be
more
about
just
doing
things
like
the
best
way
to
get
involved.
Isn't
wait
for
the
group,
have
lunch
meetings
and
do
something:
it's
just.
You
know,
jump
out
there
and
do
the
work
and
like
just
look
for
stuff
to
do.
I
always
like
finding
a
just
my
own
personal
style.
B
Is
you
know,
look
for
that
either.
Look
for
a
low
hanging
issue
that
looks
like
like
a
simple
small
bug
fix
and
try
to
get
involved.
That
way,
it's
a
really
good
way
to
gauge
how
engaged
someone's
gonna
be
via
github,
but
it's
important
just
you
see
something
broken
or
you
see
something
you'd
like
implement
something
and
throw
it
out
there.
It's
a
much
more
much
less
corporate,
much
more
culture
that
respects
doing,
as
opposed
to.
C
Talking
about
doing
yeah
totally
I
do
appreciate
how
much
especially
so,
we've
taken
a
peek
at
a
few
other
repos
köppen,
telemetry
ones,
so
the
C
and
C++
one
yeah
Genji
Ohana,
says
a
pretty
good
understanding
on
absolutely
who
participates
quite
a
bit
and
the
Ruslan
house,
like
some
good
pending
issues,
but
I
was
extremely
impressed
with
the
PHP
one
and
yeah.
C
Just
a
touch
I
think
when
we
there's
some
some
piece
of
work
a
while
back
that
we
reused
as
it
Kim
core
yeah
justjust
as
an
evaluator
like
a
cross
cross
agent,
cross
project,
evaluator
I
think
was
the
but
yeah
I'm,
not
too
sure
it's
not
really
relevant,
but
yeah.
Some
kind
of
cursory
introduction
to
it.
Yeah
because.
B
I
think
the
again
this
is
Beit
I'm,
an
expert
because
I
looked
at
the
repo
once,
but
my
memory
of
it
is
that
the
exporter
they
have
right
now
for
PHP
is
as
if
can
one
and
the
sample
stuff
they
have
in.
The
repo
is
for
Zipkin,
so
being
familiar
with
ads
would
be
a
big
help
in
getting
started
on
the
PHP
repository.
B
B
B
A
C
A
PHP
for
20
years,
I'm
kind
of
curious
what
they're
doing?
Oh,
no
totally
open
I've,
been
really
glad
to
see
like
so
much
cross-pollination
in
those
fields,
son
yeah,
good,
I'm,
glad
you
are
reasonable
people
who
don't
feel
threatened
by
unexpected
people
showing
up
it's
great
yeah.
Let
us
know
if
anyone
hops
on
and
yeah
I
think
moving
forward
will
what
kind
of
maybe
this
week
yeah
some
some
of
the
Gator
stuff
and
just
yeah?
Let
folks
know
that
there
will
be
attendees
if
they
have
this
thing
meeting
got
it
cool
thanks.
A
E
D
B
E
E
B
Don't
judge
me
too
hard
for
that
and
I'm
just
here
to
learn
about
what
y'all
doing
and
find
ways
to
get
involved
in
a
reasonable
way
that
won't
Rick
your
stuff
and
I
think
Kyle
just
joined
the
call
he
is
also
from
here
relic
but
currently
on
the
PHP
team.
So
he
might
be
interested
in
introducing
himself
hello
guy,
oh
hey,.
E
E
C
E
E
E
That
we
started
is
like
30
lines
of
plumbing
that
I
started
doing
for
the
API,
so
the
nice
part
about
that
is.
If
any
of
y'all
want
to,
is
that's
your
subject
matter,
expertise
or
even
if
that's
an
interest
area
for
you,
it's
a
really
great
way
to
start
with
greenfield
that
has
an
open,
spec
and
other
languages
that
I've
already
implemented,
but
you
can,
but
you
can
implement
it
in
the
way
that
you
see
prudent.
C
Yeah
totally
noted
I
also
really
appreciate
how
the
repo
is
organized
and
yeah
one
of
my
team
members,
so
my
team
at
New
Relic
handles
both
the
PHP
agent,
but
also
our
C
SDK,
and
so
one
of
our
team
members
is
working
pretty
heavily
on
the
C++
open
to
mature
libraries.
So
we've
had
a
lot
of
insight
into
at
that
process,
but
yeah
out
of
all
the
repos
we
looked
at
the
C
or
PHP.
C
One
has
been
really
the
most
welcoming
like
in
terms
of
like
just
just
github
issues,
and
it
was
really
easy
to
dive
in
and
start
getting
a
look
at
stuff
yeah.
My
team.
We
are
yeah
very
much
like
have
time
allocated,
release
to
like
starting
in
the
coming
weeks
to
them
yeah,
to
devote
engineer
time
to
this
project,
so
I
think
it's
going
to
take
a
little
bit
of
ramped
up
time,
but
excited
to
figure
out
what
that
looks
like
to
my
team
and
what
that
looks
like
I'm,
working
with
the
rest
of
the
community.
E
That's
like
unbelievably
wonderful
music
to
my
ears,
like
it's
myself.
Another
man
named
Benjamin
with
Benjamin,
with
battle,
J
and
and
Levi
Morrison
from
from
data,
doc
and
I
feel
like
we've
been
slogging
through
this
by
ourselves,
and
it's
really
nice,
because
I
think
the
outcome
will
be
unbelievably
fruitful
to.
F
Give
a
perspective
from
a
brand-new
contributor
literally
a
week
and
a
half
ago
I
was
googling.
What
is
a
span,
and
this
morning
I
pushed
out
my
first
PR
towards
the
PHP
open
till
entry
library,
so
getting
started
is
pretty
pretty
straightforward.
You
know
the
first
step
was
just
get
it
familiarizing
myself
with
a
spec
which
I
was
not
familiar
with
to
begin
with,
and
if
you
already
are
and
you've
got
that
leg
up
and
so
yeah
just
diving
into
the
deep
end.
There's
there's
a
lot
of
low
hanging
fruit.
It.
E
E
So
there
are
a
couple
of
really
nice
and
tactical
things
about
a
repo,
the
API
and
SDK
or
innocence
their
own
separate
folders,
which
is
clear
and
concise,
we're
working
on
making
the
testing
it
all
being
one
like
very
centralized
located
test
folder
that
has
SDK
and
contributor
parts
of
the
repository
with
both
unit
and
integration
tests
for
each
the
makefile.
For
this
is
real.
It's
pretty
clean
and
concise.
It
gives
you
the
ability
to
install
the
dependencies
you
need
into
a
vendor
folder
and
and
execute
both
our
PHP
unit
files
and
our
our
fan.
E
Our
fan,
dependency
checks
and
some
example.
Some
example
use
cases
of
the
open
flow
entry,
library
and
again
we're
very
welcome
to
changing
on,
if
there's
an
improvement
that
you'd
like
to
make.
But
may
you
can
do
make?
Oh
and
the
PHP
CS
linting
is
also
has
a
really
great
main
processor,
like
I,
have
a
chain
in
my
in
my
terminal
that
I
use
all
the
time
it's
make
style
and
make
test
and
make
fan
and
make,
and
that's
just
like
a
really
great
way
to
be
able
to
test
all
these
individual
things
together.
So.
B
E
Are
in
the
alpha
stage,
I
would
not
recommend
it
yet
and
for
production
use
it
has
not
been
load
tested,
it
has
not
been
battle
tested,
so
I
think
that
there
are
we're
getting
close
to
where
we
could
probably
do
a
pretty
decent
load
test
in
the
tracing
shot
like
the
metric
side.
That
is
like
I,
think
I
mentioned.
It
probably
has
30
lines
of
code,
maybe,
but
the
tracing
part
I
think
is
getting
close
to
a
park
lows
to
a
point
where
we'll
be
able
to
do
some
of
our
tests.
E
E
Okay,
my
seat
you,
so
maybe
that
maybe
I
should
turn
this
on
the
on
the
three
of
you.
What
can
we
do
as
maintain
airs
and
contributes
to
help
you
all
ramp
up,
get
up
to
speed
and
make
you
like
enable
you
I
guess
would
be
the
right
thing?
Is
it
rolling
through
the
repo?
Is
it
answering
questions
that
you
have
currently
what
would
be
the
most
self?
What
would
be
the
most
helpful
thing
for
you.
B
Well,
I'll
start
just
throwing
it
out
there
and
talking
again
so
the
repo
layout
that
we've
organized
it
and
everything
is
great
and
I.
Think
a
couple
months
back,
you
actually
helped
me
out
with
a
small
tra
made
to
kind
of
make
some
of
those
dogs
clearer.
One
challenge
I
faced,
though,
is
when
I
kind
of
looked
at
the
issues
and
I
thought.
What
do
I
want
to
do.
B
A
lot
of
them
are
very
broad
in
high
level
and
say
like
go
implement
this
part
of
the
spec
and
so
they're
kind
of
hard
for
somebody
not
familiar
with
the
project
to
get
started
with
so
I'm
wondering
if
there
is
like
a
just
thinking
out
loud.
Is
there
a
particular
smaller
issue
that
might
be
useful
for
newcomers?
Is
there
a
way,
some
of
those
larger
implement?
This
part
of
suspecting
might
be
maybe
better
broken
out
into
smaller
parts
or
planned
out?
B
It's
basically
that
classic
thing
of
you
know
we
all
don't
have
a
whole
lot
of
time
to
spend
on
the
open
source
projects
and
kind
of
facing
this
wall
of
open,
telemetry
stuff
that
even
coming
from
New
Relic
and
being
familiar
with
the
domain.
It's
sort
of
like
oh
there's,
a
lot
of
things
that
are
different
here.
E
It
that's
it.
That
is
a
great
question.
That
is
something
that
I
have
considered,
that
I'd
recognized
that
we
probably
need
to
do
a
better
job
of,
but
there
wasn't
any
real
to
be
completely
Frank.
There
hasn't
been
any
real
demand
to
work
on
the
project
and
all
the
past
couple
months,
so
I've
had
a
difficult
time
like
justifying
breaking
down
some
of
these
large
tasks
into
subtasks,
but
I
would
be
more
than
happy
to
break
down
some
of
these
larger.
Like
epic
milestones,
if
you
only
use
the
jurist,
the
JIRA
speak
into
smaller
tasks.
E
If,
if
that
would
be
helpful
for
you,
I
think
that
the
writing.
The
other
right
answer
to
that
question.
Ellen
is
I'm,
not
certain
the
right
way.
I
am
I,
am
still
a
new
like
I'm,
not
writing,
I'm,
not
involved
in
the
language
specification.
You
know
as
much
as
I'd
like
to
be
just
because
of
time
constraints
and
we're
constraints,
but
I
am
more
than
happy
to
break
these
down
into
individual
tasks.
Is
there?
E
Would
you
like
me
to
do
that
and
then,
like
add
a
specific
label
to
the
task,
to
make
sure
that
there's
we
have
I
have
a
bunch
that
are
labeled
those
good
first
issues,
but
maybe
they
aren't
really
good.
First
issues
you
know
like
these
are
some
of
these
might
be
like
good
first
issues,
for
you
know,
people
that
are
interested
in
working
on
developing
a
very
large
chunk
of
the
spec,
but
like
maybe
like
bite-sized
label,
or
something
like
something
like
that.
That
would
be
advantageous
for
y'all.
B
I
think
there
would
be
advantageous
for
new
folks
I
wonder
if
we
maybe
get
the
opinion
from
the
two
people
actually
on
the
PHP
agent
PHP
team
at
New
Relic.
You
might
have
time
to
work
on
this
since
I'm,
here
kind
of
a
sconce
work
and
just
to
learn
what
about
you
two
do
you
think,
would
that
help
you
all
or
do
you
have
other
ways
things
that
Bob
and
the
rest
can
kind
of
help
you
get
started
with
the
project.
C
Yeah
so
yeah
when
Bob,
but
urgently
asked
a
question.
I
didn't
have
a
good
answer,
because
I
felt
like
I'm.
It
would
take
a
little
bit
like
so
we're
very
much
right,
just
getting
started,
getting
ready
in
the
process
of
participating.
So
there's
a
lot
that
we
haven't
experienced
firsthand
yet,
and
so
it's
gonna
originally
say
yeah.
E
E
E
The
context
issue,
because
that
one
is
small,
yet
I
think
it
can
be
probably
be
segmented
pretty
well,
and
that
might
be
a
really
great
place
for
us
that'd.
Be
that
the
context
part
of
the
spec
is
ever
is
going
to
ever
become
more
and
more
important,
because
there
are
a
lot
of
languages
that
have
a
built
in
context.
The
idea
of
a
built
in
context,
idiom
PHP,
does
not
at
least
not
one
that
I'm
aware
of,
and
they
it
has
been
started.
E
They've
started
to
for
lack
of
a
better
word
interweave
that
lived
within
both
the
metric,
both
tracing
and
the
metrics,
and
now
the
vlogging
too.
So
I
think
that
getting
that
right
is
gonna,
be
of
paramount
importance
for
us
and
make
sure
that
it's
reusable
in
different
wouldn't
be
reusable
in
different
contexts,
but
I
think
that
that
that
probably
is
going
to
be
a
good
first
effort
for
us
also
like
the
sound.
E
The
open,
telemetry
specification
room
so
like
if
you
start
reading
through
on
these
specs,
you
know
like
WTF.
Why
did
they
think
about
this
in
this
particular
way?
They're
they're,
unbelievably
receptive
and
I've,
seen
them
change
things
in
the
spec
like
mid-flight,
because
somebody
comes
in
and
asks
a
question
in
that
room.
So
that's
a
really
great
place
to
go
and
ask
questions
if
you're
considering.
Oh
man,
I
wonder
what
what's
missing
here.
Excetera.
B
E
I
think
as
especially
as
we
continue
to
move
along
and
the
metrics
and
marks
metrics
is
still
not
a
completely
defined
spec
they're,
still
like
it's
still
very
much
in
beta
they're
still
making
changes
to
it.
That's
why
I
haven't
been
like
unbelievably
pushing
on
it,
because,
with
the
limited
resources
we
have
on
the
PHP
side
as
of
right
now,
it's
like.
Well,
we
don't
want
to
like
implement
there
spec
and
then
implement
the
implementation
and
then
like
go
back
and
try
and
find
things.
E
B
E
B
If
you've
done
much
thinking
about
how
the
open
tracing
projects
going
to
behave
and
live
inside
of
this
kind
of
new
wave
of
asynchronous
PHP
frameworks
are
coming
coming,
I'm
thinking
of
things
specifically
like,
is
it
school,
swoll
I'm,
not
even
sure
how
to
pronounce
it?
I
just
know
that
PHP
age
is
kind
of
targeting
a
lot
of
the
performance
improvements
for
these
async
frameworks,
and
that
there's
a
lot
of
folks
who
think
the
next
couple
of
PHP
years
in
PHP
is
gonna,
be
all
about
async
I'm
wondering
if
anyone
is
try.
E
Makes
sense
the
question
makes
sense
the
answer's?
No,
we
have
not
yet
considered
I
think
better.
That
was
that
would
definitely
be
a
fruitful
endeavor,
but
I
think
that
that
was
an
art
of
our
immediate.
Our
media
concern
is
like
catching
up
with
the
language
spec
and
making
sure
that
we
can
implement
a
clear
tracing
powder
and
a
clear
metrics
pattern
and
potentially
a
clear
logging
pattern
and
then
adding
the
syntactical
sugar
of
oh.
E
We
have
asynchronous
things
and
oh,
we
have
like
another
one
that
when
my
co-workers
brought
up
was
like,
if
right
now
in
any
of
these
in
any
of
these
packages,
open
census,
open
tracing
or
open
telemetry.
If
you
open
up
a
a
PHP
fork,
it
just
doesn't
work.
It
actually
drops
fans
without
like
silently
without
telling
anybody.
So
there
are
other,
like
key
implementation
factors
that
we
still
have
to
nail
down
and
then
I
think
Alan
that's
up
there.
E
But
I
do
think
that
that
is
something
that
we
should
be
very,
very
thought,
mindful
of
as
we're
implementing,
if
possible,
I
probably
need
to
read
a
little
bit
more
about
those
asynchronous
features
because,
like
I've
been
sort
of
lying
to
them,
so
I
think
that.
But
thank
you
for
calling
that
out.
I
think
that's
really
important.
Yeah.
B
B
E
It
yeah
that's
a
big
puzzle.
The
other
thing
that
I
learned
recently
that's
kind
of
scary
and
daunting
is
a
lot
of
those
asynchronous
things
or
written
into
the
PHP
language
by
one
person
and
they
don't
like
they
haven't
really
shared
them
and
done
like
a
very
big
knowledge
transfer
of
that
to
anyone
there's
an
a
very
important
job
of
documenting.
It
I
think
it's
one
of
those
things
like
I'm
sure
it's
like
that
classic
dog.
E
This
is
fine
kind
of
thing,
but
I
think
that
it
hasn't
like
really
been
evangelized
or
documented
over
well
yeah,
and
so,
like
it's
kind
of
daunting
and
scary,
to
use
those
features
if
you're
not
if
like.
If
the
troubleshooting
method
is
going
and
reading
underlying
C
code,
you
feel
like
you're.
Probably
you
know
up
the
creek
without
a
paddle
very
frequently
and
a.
B
C
That
does
actually
lead
into
a
point
that
that's
kind
of
relevant
yeah.
Our
implementation
of
any
of
the
PHP
instrumentation
we
do
so
far
is
pretty
much
like
99%
sunday
pio,
like
internals,
like
everything's,
primarily
written
in
C,
but
yeah.
We
do
have
like
a
very,
very,
very
working
knowledge
of
PHP
but
I.
Think
there's
a
a
bit
of
like
the
feature
set
that
maybe
we
don't
understand
the
best
for
Mike
an
end
user
perspective.
C
So
I
do
see
that
it's
like
a
little
bit
of
a
challenge
for
some
of
the
folks,
but
very
much
myself
included
on
the
team
and
just
yeah
I
think
we
might
need
there
yeah.
There
might
just
be
your
conversations
about
like
API,
like
urgh
UNAM
mix
and
all
that
that
we
work
on
and
just
verifying
that
it's
done
in
a
way
that
you
know
the
broader
community
would
find
really
really
pleasant
to
work
with
and
not
clearly
a
C
C
API.
That.
E
Has
been
that's
been
a
I,
don't
wanna
use
the
word
struggle,
because
it's
not
a
Charlotte's
been
a
real
eye-opener.
I
open
in
working
with
a
couple
of
these
other
people
that
work
in
PHP
lenses
or
MailChimp
has
like
its
own
abstract,
abstract,
bespoke
framework
that
we
use
on
top
of
PHP,
so,
like
things,
become
very
different,
very
quickly.
They're,
like
the
abstractions
that
we've
added
to
this
particular
library,
are
the
manifestation
of
a
couple.
Different
companies.
E
Implementation
strategies
like
the
using
the
API,
is
an
interfaces,
and
SDKs
is
a
new
new
paradigm
to
me
and
like
implementing
some
of
these
other
strategies,
or
also
have
also
been
just
like
a
high
bind
of
a
couple
places
which
I
think
is
great.
It's
helped
me
to
learn
more
about
how
about
PHP
gets
used
to
other
companies.
That
was
one
of
my
core
goals
of
being
the
main
one.
E
We
maintain
errs
of
this
project
and
I
think
that
that's
something
that
we
are
very
happy
to
like
be
involved
in
and
we're
very
but
like
this
project
is
very
like
unbelievably
open
to
change
like
if
there's
something
that
you
don't
like
open.
A
PR
you're
probably
getting
accepted
as
long
as
the
passes.
All
the
unit
tests
but
I
mean
obviously
that's
very
hand
wavy,
but
you
get
the
general
sentiment
of
what
I'm
trying
to
say.
F
On
the
subject
of
threads,
there
async,
though
Allen
I,
think
it
was
last
Friday
I
deemed
Bob
and
was
like
I'm
working
on
this
span.
Processor,
stuff
and
man
do
I,
miss
P
threads,
so
I
was
comparing
I
was
cross-referencing,
some
of
the
work
in
the
go
and
Python
SDKs
and
they
were
using
native
language.
Implementations
of
either
go
routines
or
pythons
built.
Then
they
sync
stuff
and
I
was
looking
at.
You
know.
How
would
we
process
this
bulk
dispatch,
export,
stuff
and
I
was
like?
F
Oh,
it's
gotta,
be
in
in
process
in
line
with
the
call
crap
there's
just
not
really
any
like
good
ways
around
that
today,
but
looking
forward
that
doesn't
have
to
stay
the
same,
but
thinking
about
maintaining
language
version
compatibility
so
that
we
don't
force
users
to
be
on
eight
dot,
whatever
yeah.
So
you
know,
there's
there's
options
there
for
optimistically
upgrading
behavior
when
that
becomes
mainline.
B
D
B
Amber
off
sorry
did
begin
with
a
neighbor
look
at
the
age.
Yeah
I'm,
sorry
amber
why
you
know
we
hire
for
you
know.
C
programming
is
one
of
the
big
skills
on
that
team,
because
our
whole
the
whole
thing
it's
an
extension,
but
where
is
it
going
with
that
I
don't
know,
I
was
just
talking
about
my
behind
in
leading
mode,
yes
exciting
times.
C
Good
stuff,
yeah,
no
I
get
the
no
pun
intended,
but
the
thread
you
were
going
on
and
and
the
the
answer
is
kind
of
yeah.
Whenever
we
hit
kind
of
roadblocks
like
that,
it's
just
it's
been
trivial
for
well,
not
trivial,
because
it's
C,
but
it's
been,
you
know,
there's
never
those
concerns
about
like
the
limitations
of
PHP,
because
we
can
just
do
whatever
we
want,
C
and
so
I
think
there
is
going
to
be
yeah
like
a
little
interesting
challenges.
C
E
Just
changed
the
version
from
7.2
to
7.3
and
the
reason
for
that
is
with
each.
Our
time
is
a
new
monotonic
timer
that
is
only
available
in
PHP,
7,
3
and
above
and
PHP
7
2
only
had
like
has
what,
like
it,
I
think
it
has
like
six
months
or
a
year
left
of
security
support,
so
we're
like
trying
to
kind
of
follow
that
pattern,
but
we,
but
we're
bumping
it
up
in
the
case
that
we
have
and
like
in
the
case
that
we
have
like
non-negotiable.
E
What
we
feel
are
non-negotiable
features
in
the
language
that
are
that
we
need
to
implement
like
the
mana
time
timer.
We
felt
because
the
precision
that
they're
asking
for
for
some
of
the
tracing
thing
is
like
micros
or
microseconds
or
nanoseconds
or
whatever,
and
we
couldn't
you
can't
feasibly
get
that
time
reliably
from
the
timestamp
implementation.
I
have
not
yet
woven
in
I
had
a
PR
at
last
week.
That
was
it
was
this.
Is
that
this
tells
you
how
much
improvement
we've
made?
E
We
started
off
with
strings
for
the
tracing
implementation
for
keeping
the
time
stamps
to
last.
Last
week
we
changed
those
to
infer
to
like
begins
when
we
needed
to,
and
now
we're
gonna
weave
in
be
the
monotonic
timer
bits
and
pieces.
But
I
think
that
that,
like
that
is
pg7,
3
is
the
lowest
we'll
target
for
now
and
I.
I
think
that
the
cross-section
of
people
that
would
want
to
use
this
library
and
that
are
going
to
use
in
a
very
old
version
of
PHP.
By
the
time
we
release
it,
will
be
0
at.
C
E
C
But
that
doesn't
ring
a
bell
when
you're
talking
about
the
time
and
there
I'll
look
into
this
for
next
meeting.
But
there's
we
have
this
concept
called
logs
and
context
at
New
Relic,
where
essentially,
we
just
add
some
metadata
to
your
spans
and
traces,
and
we
get
to
basically
just
correlate
that
with
log
entries
by
just
having
your
log,
something
so.