►
From YouTube: 2021-04-23 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
By
the
way,
yes
sorry
go
ahead,
yeah
I
just
I
was
mentioning
to
the
dc
members
that
they
should
appear
here.
So
I
don't
know,
I
told
them
that
they
should
ask
their
managers
to
spend
more
time
helping
with
this.
A
B
A
B
And
I'll
skip
this
one,
because
I
already
like
I
created
this
issue
and
I
already
tagged
it
properly.
B
A
A
B
A
Yeah
this
one,
I
think
we
did
mention
it
in
the
spec
call
it's
about
who
has
higher
priority,
and
I
think
that
it
was
anthony
mirabella,
who
said
that
he
will
be
providing
so
many
scenarios
and
josh
suarez,
probably
will
be
helping
with
the
actual
pr
portion.
A
A
A
I
I
I
I
I
don't
know.
We
may
have
something
like
another
issue
for
this
one,
probably
probably
not
like.
B
It's
more
like
packaging,
and
some
of
the
feature
might
not
be
available,
and
I
think
if
the
user
decided
not
to
install
a
particular
component,
it
might
be.
Okay,
we
just
clarify
this
environment.
Variable
will
be
simply
ignored,
because
we
cannot
expect
the
other
component
to
have
knowledge
about
all
the
other
components.
Yeah.
A
Yeah
yeah,
that's
part
of
the
situation.
The
other
part
is
that
ruby
and
now
javascript
they
want
to
do
this-
is
that
they
would.
They
will
try
to
detect
it
from
the
sdk
like
if
the
package
exists,
try
to
load
it,
which
feels
like
a
bad
thing
to
me.
Honestly,
I
don't
I
I
don't.
You
know
like
it's
like
a
circular
situation.
A
I
could
say
my
impression
is
that
this
specific
variable,
because
of
the
specific
requirement
should
be
mostly
used
by
packages
that
include
that
you
know
that
bundle
a
lot
of
things
like
the
agents.
B
A
The
auto
instrumentation
systems,
so
it's
basically
two
two
things
there.
In
any
case,
we
need
somebody
to
well.
I
was
supposed
to
be
working
on
that.
I
remember
now.
Yes,
I
remember
that
I
was
supposed
to
work
on
this
a
few
weeks
ago.
We
had
a
lot
of
things
to
do
because
I
want,
but
I
wanted
to
do
a
general
clarification
of
this
environment.
Variable.
B
So
this
is
very
important,
sdk
area
and
I
I
think,
although
it
mentioned
expander,
but
it's
not
specific
to
the
xpeller,
it's
just
the
generic
isdk
right
right,
yes,
and
should
I
assign
this
to
you
yeah,
please.
A
Yeah
yeah:
I
think
that
this
issue
well
there's
another
issue
for
generally,
you
know
tracking
this,
but
this
one
is
whether
the
sdks
should
try
to
detect
it,
even
if
they
don't
include
this
package
or
additional
artifact,
whether
they
should
try
to
detect
it
all
right.
Let's
go
to
the
next
one,
yeah.
B
I
I
guess
this
one
we
can
see
the
isdk
can
they
can
try
to
detect
it,
but
it
is
not
required
because
my
voice
like
later
then
the
knowledge
will
spread
among
those
components
and
break
the
isolation.
That's
correct!
Yes,
exactly
and,
for
example,
like
in
c
plus
plus
there's
no
easy
way
to
detect
something
dynamically
and,
for
example,
people
decided
to
statically
link
everything
in
gold
or
c
plus
plus
there's
no
reason
for
people
to
ever
consider
this
right
right.
Yes,
totally
there.
You
are
so
that
that's
the
thing.
B
B
B
B
B
Instead
of
trying
to
clarify
everything,
because
on
metrics
we
we
have
the
same
problem
and
people
come
here
and
ask
me
like
what
do
you
mean
by
counter
and
I
I
like
they
were
referring
to,
for
example,
the
radiation
in
the
ocean,
and
they
can
use
that
like
some
like
conquer
that
congo
radiation
value
and
they
believe,
that's
the
content,
but
we're
telling
them
it's
not
a
counter.
It's
a
gauge.
B
So
that's
where
the
computer
was
coming
from,
but
most
of
the
people
have.
If
they
have
the
the
essential
experience
working
on
metrics,
they
would
immediately
know
what
the
counter
is.
Because
premise
is
micrometer,
almost
all
the
matrix
api
have
counter
and
and
and
they
know
that
radiation
counter
is
a
gauge
not
a
counter.
B
A
Yeah,
so
I
think
in
this
case,
based
on
the
fact
that
tristan
is
a
maintainer,
I
think
we
should
do
a
clarification
on
the
specification
for
exotic
languages.
I
would
say.
A
That's
yeah,
I
think
that's
a
good
question,
I'm
pondering
initially
to
initially
to
me.
The
feeling
was
that
we
have
to
clarify,
because
this
is
something
that
for
airline
can
be
specific
and
it's
a
clarification,
but
now
I'm
I'm
starting
to
feel
that
we
should
just
we
shouldn't.
I
don't
know:
okay
yeah.
In
any
case,
I
think
yeah.
We
can
leave
the
issue
open
and
ask
for
opinions,
and
if
it's
not
too
important,
we
can
just
close
it
yeah.
This.
B
Could
be
like,
if
you
try
to
clarify
it,
could
it
could
become
even
worse,
like
people
ask
hey
what
about
the
propagation
between
the
user
mode
and
the
kernel
mode
call,
I
think
we
call
the
operating
system.
You
want
to
propagate
that
from
your.
It
is
within
the
same
process,
but
your
you're,
probably
getting
from
the
user,
address
space
to
the
kernel
address
space
and
are
we
saying
this
is
an
rpc?
B
If
it's
not,
then
why
can't
we
just
use
contacts
and
because
contacts
by
default
won't
flow
across
user
and
kernel
mode,
there's
a
protection
level
right
and
for
language
for
languages
that
have
co-routine
or
they
have
explicit
threading
model.
Might
there
might
be
many
different
stories?
We
need
to
clarify
yeah,
yeah,
yeah.
A
B
Yeah,
I
I
I
think
this
will
probably
because
the
the
cluster
has
been
running
fine
for
a
while.
I'm
I'm
not
seeing
so
many
questions
asking
for
examples.
So
there
should
be
a
balance.
If
we
keep
adding
examples
to
the
glossary,
it
might
turn
out
to
be
too
big
and
also
I
I
think
I
think
example
will
always
face
the
challenge.
At
least
I
I've
seen
that
in
the
metric
space,
people
are
saying:
oh,
like
you're,
very
biased.
You
put
an
example
for
the
object-oriented
language,
but
right
scripting
language.
B
I
end
up
having
to
write
examples
for
python
for
java
and
for
c-sharp
three
languages,
which
is
very
painful
and,
and
that
makes
that
like,
even
if
something
like
very
simple,
you
can
explain
in
one
sentence
with
all
the
examples
it
turned
out
to
be
a
very
long
document,
so
yeah
for
this
one.
I
wonder
if
we
can
split
that,
for
example,
if
the
glossary
like
we
can
put
the
example
in
the
spike,
if
someone
is
willing
to
do
that
and
and
have
a
link
yeah
looking
at
in
the
spec.
A
Yeah,
so
my
impression
is
that
yeah
we
shouldn't
have
examples
for
every
term
there
for
sure
and
if
there's
a
specific
item
that
is
not
clear,
then
we
can
add
an
example,
but
this
shouldn't
be
like
a
requirement.
You
know,
I
don't
want
this
to
be
a
requirement
because
then
it's
gonna
be
a
nightmare.
B
B
B
B
A
B
A
A
B
Yeah,
you
can
see
also
christian
mentioned.
I
consider
this
as
a
design
problem
for
the
w3c
transparent,
and
so
I
I
here
is.
We
see
whether
this
is
a
common
pattern
used
by
many
folks
that
worst
clarification
here
or
is
just
a
special
case.
Like
99
percent
of
the
folks
won't
care,
then
it
probably
won't
make
sense
to
load
the
open
climate
project
and
if
it
turned
out
to
be
a
common
practice,
we
might
want
to
see
if
it
makes
sense
to
put
this
as
a
w3c
transparent,
like
the
spike
issue.
A
Yeah
yeah
yeah,
I
feel
like
we
should
bring
it
to
the
to
the
trace
context,
meeting
and
try.
It
would
be
very
nice
if
we
can
actually
manage
to
put
that
that
part
there
instead
and
probably
provide
that
workaround
on
our
site
for
now,
but
yeah
it
could
be
good.
A
B
B
B
A
B
Yeah
and
I'm
not
sure
when
people
try
to
identify
synthetic
requests,
do
they
want
to
like?
Let
me
ask
you:
would
you
prefer
to
park
this
in
the
in
the
baggage
or
something
like
just
make
it
like
a
protocol,
so
we
can
like
in
this
way,
regardless
of
the
rpc
mechanism.
If
you
ever
implement
that,
like
grpc
or
http,
they
will
be
having
a
consistent
way
to
identify
this,
or
we
can
leave
this
as
hey
like
individual
http
or
like
instrumentation
library.
They
will
need
to
do
something.
A
Yeah,
so
I
don't
know
whether
bagash
is
a
good
place
for
that,
but
we
should
probably
semantic
conventions
would
work
just
fine
for
this.
B
A
A
B
Here,
like
we
believe,
hey
like
if
you
have
hdb
sewer-
and
you
want
to
do
some
synthetic
testing,
it's
the
application
developers
responsibility
to
know
from
the
server
side
whether
this
is
a
synthetic
test,
we're
not
in
the
place
to
standardize
this,
but
once
it
is,
we
expect
the
instrumentation
library
to
use
semantic
convention
to
capture
this
behavior,
correct,
yeah,
yeah.
That
makes.
B
B
B
A
Yeah,
actually
I
don't
know
there
because
you
know
there's
a
special
case
that
he
mentions,
like
special
sampling
rules,.
B
Details,
if
you
think
this
makes
makes
sense,
we
think
I
may
not
want
to
define
a
problem
because
I
guess
inside
hex
is
a
complex
topic
and
yes-
and
it's
not
just
about
telemetry
and
given
there's
so
many
existing
stuff.
If
we
try
to
define
something
it,
it
will
blow
to
what
about
migration
story
and
what
about
security
it.
A
B
Effort
yeah,
it's
it's
like
I
I
could
also
say
like,
for
example,
people
later
could
say:
hey.
Can
we
have
a
mechanism
to
identify
different
requests
from
multiple
tenants
and
and
that
I
agree,
that's
a
valid
requirement
and
most
of
the
folks
would
probably
care
about
it
at
some
time.
But
handling
attendance
is
not
a
like.
It's
not
an
open,
telemetry
mission.
B
B
B
It
like
that,
okay,
it's
done
anything
else.
I
think
that's
all
for
now.
Okay,
so
I
have
a
small
update
on
the
on
the
slack
server
yeah.
I
think
the
api
part
is
it's
99
down
and
there's
one
remaining
task:
it's
not
a
concern.
It's
a
task
that
we
we
want
matrix
api
to
provide
hinting
mechanisms.
So
people
like
who
write
the
library,
for
example,
if
you
own
a
http,
instrumentation
library
and
you
instrument
the
library
using
the
matrix
api
and
because
you
don't
know
what
the
user
would
need
for
specific
cases.
B
B
For
example,
if
you
have
metrics
that
covers
10
dimensions,
probably
nobody
would
take
all
the
dimensions,
otherwise
they
got
to
pay
for
the
memory
cost
of
the
time
series.
If
you
multiply
all
the
combinations,
it
will
be
enormous,
so
the
http
library
might
give
people
a
hint
saying
hey
in
this
way.
I
expect
most
of
the
users
by
default
will
take
the
http
status
code,
the
http
request,
duration
and
the
http
verb.
B
We
think
is
good
enough,
but
we
also
provide
the
http
url
and
I
think
most
of
the
users
shouldn't
take
the
url
as
a
dimension,
because
it's
a
very
high,
criticality
and
and
that
api
will
probably
be
very
similar
to
the
sdk
view
api,
because
if
you
have
the
like
the
raw
matrix
api
people
describe
the
signal
in
ick.
We
want
to
give
you
the
flexibility,
so
they
can
config.
B
A
Yeah,
thank
you
for
that.
Yeah
josh
mentioned
something
very
general
yesterday
to
me
about
this
yeah
very
good
there
for
two
teams:
yeah.
B
Yeah
and
and
so
for
the
matrix
part,
I
I
think
the
next
step
is
to
like,
like
almost
done
with
the
api
part,
and
to
focus
on
the
idk
part,
where
I'm
I'm
a
little
bit
worried
about
the
the
funding
currently
like.
I
I
think
it's
like
I've,
seen
four
people
who
are
quite
active
in
this
space,
and
I
was
originally
hoped
that,
once
the
data
model
part
is
done,
folks
would
be
freed
up,
so
they
can
help
on
the
api
sdk
spike.
But
it
looks
like
there
was
a
delay
in
the
data
model.
B
Part
like
like
josh,
decided
to
delay,
so
the
the
josh
knight
from
google.
He
decided
to
delay
that
to
end
of
april,
which
I
I
think
is
a
fair
date.
Otherwise
it's
just
mission
impossible,
but
still
I
I
think
it's
busy
with
many
other
things
like
java
and
c
plus
parts.
So
if
you
know
anyone
that
can
help
like
mean
it'll,
not
just
be
me
sending
all
the
prs,
we
can
break
that
down.
B
I
I
can
lay
out
the
high
level
skeleton
of
the
sdk
spike,
but
will
need
some
help
like
if
you
have
some
energy
or
or
I
can
raise
the
ass.
I
already
asked
several
times
in
the
in
the
api
and
I
stayed
specific
meeting
and
like
I
I
think
I
got
some
help
from
reviewing
the
pr
and
putting
comments,
but
for
the
ick
I
I
I
think
we
need
some
help
from
people
working
on
separate
pr's
and
and
without
that
I
I
think
I
would
call
it
the
end
of
may
timeline
might
delay.
A
Yeah,
okay,
so
this
is
like
just
to
be
a
little
bit
clearer.
Is
this
like
for
the
specification
portion,
or
you
also
want
people
who
can
write
codes
to
prototype
things?
It's.
B
Just
for
it's
just
for
the
respect
in
respect,
there
are
many
things
that
we
need
to
clarify.
So
so
one
one
thing
I
I
need
help.
Is
I'm
not
a
native
english
speaker?
Sometimes
it's
it's
like
it.
I
I
know
how
to
write
spag,
because
spike
has
a
limited
set
of
like
languages,
but
when
it
comes
to
describe
a
scenario
or
could
be
a
story,
and
this
is
what
a
lot
of
folks
are
demanding.
B
When
they
see
full
time
on
matrix
or
his
video
with
other
stuff,
it
seems
like
when
I
ping
him.
Sometimes
it
seems
he's
very
busy,
but
he
always
replied,
but
but
it
seems,
he's
quite
occupied
by
other
stuff.
A
Yes,
he
is
okay,
but
that's
a
good
call,
because
I
can
yeah
okay,
I
can
probably
ask
like
try
to
poke
my
manager
to
help
him
freeze.
You
know
to
have
some
more
cycles
for
these
other
ways.
That
would
be
the.
B
Last,
okay,
my
biggest
worry,
is
not
about
the
like
me
as
the
only
one
who's
sending
all
the
pr's.
My
worry
is
number
one.
We
might
delay
the
date
because
I
I
have
this
amount
of
energy
I'll
do
the
best
I
can,
but
if
the
work
is
more
than
what
I'm
capable
of
doing
then
will
spread
things
and
take
longer
time,
and
my
biggest
worry
is.
If
we
don't
have
enough
attention
on
this
space,
it
will
be
like
four
of
us
working
on
that
spec.
We
believe.
B
Oh,
it's
perfect
and
I
probably
can
reach
out
to
like
10
other
folks
to
see
whether
they're
okay,
so
that
in
the
end
we
got
14
people
who
are
okay
with
this
bag.
But
later
we
have
more
people
trying
to
implement
that
we
start
to
discover
new
problems
and
if
those
problems
can
be
fixed
in
the
additive
manner,
I
think
it's
fine.
My
biggest
worry
is
we
realize.
Oh,
it
won't
work
at
all
and
we're
going
to
start
from
scratch.
A
Okay,
I
see
that
sounds
yeah,
that
that
could
be
bad
but
but
okay
yeah.
Thank
you.
Thank
you
for
the
thank
you
for
the
for
the
feedback.
For
that.
Yes,
I
will
totally
try
to
talk
to
jim
magnus
and
me
and
our
manager,
so
we
can
probably
discuss
options
here
and
the
risk
here
yeah.
I
think
so.
A
B
Let
you
know
how
things
go
yeah
and
in
the
next
tuesday
meeting
I
I
think
the
the
my
last
matrix
api,
big
pr
will
get
merged
and
then
there
will
be
some
minor
clarification.
So
I
I'll
bring
this
topic
briefly
in
that
meeting
and
say:
hey
the
matrix
api
part,
we
believe
is
almost
done
and
we
give
people
a
month
by
end
of
may.
B
Please
review
the
spike
and
if
you
want
to
do
some
experiment
you
you
can
do
that,
but
you
should
expect
that
spike
is
not
currently
recommended
for
sdk
or
client
to
implement
until
end
of
may
so
before
that
time.
If
you
see
any
blocking
issue,
please
please
call
out,
and
after
after
me,
starting
from
june,
we
don't
expect
to
make
any
big
changes
or
like
increase
the
scope
of
that
spec,
and
I
hope
in
that
way
I
can
buy
some
attention.
B
A
Yeah,
perfect
yeah,
I
think
that's
good
yeah
yeah.
That
makes
sense.
So
that's
all
from
my
side,
yeah
thanks.
So
much
for
the
update,
yeah
yeah,
nothing
else
from
my
side
just
trying
to
to
get
more
dc
time.
You
know
yeah
same
mention,
so
that's
yeah.
I
don't
know
I'm
trying
to
yeah
yeah.
I
don't
know.
I
think
that
one
of
the
problems
with
tc
is
that
I
think
my
impression
is
my
impression.
Is
that
impression
they
have
is
that
the
specification
is
the
one
that
is
like
the
specifications.
A
B
B
We
have
some
people
in
the
tc
that
I
I
totally
understand.
I
mean
like
similar.
Similar
thing
happened
for
me.
Even
I'm
not
part
of
the
tc,
like
I'm,
probably
doing
better
better
progress
on
the
metrics
api
part,
but
that
means
I'm
doing
a
very
bad
job
on
open
time,
3c
plus
class.
I
almost
dropped
the
ball
to
one
developer
in
my
team
and
put
all
the
pressure
on
him.
A
Yeah,
this
is
kind
of
the
situation
that
I
also
mentioned
that
you
see
that
trust
fair
aspina
from
the
java
world
did,
and
he
said
that
probably
people
in
this
kind
of
big
project
that
affect
all
the
languages
we
shouldn't
be
writing
code
or
so
much
code.
You
know
and
is
something
valid
because
it's
it's
hard
to
balance
things
out.
So
in
any
case,
I
hope
we
can
get
some
more
cycles
from
the
tc
or
or
something
you
know
from
people
working
there,
because
yeah.
B
B
A
B
A
Yes,
yeah,
but
I
have
to
say
that
not
only
for
them,
I
have
to
say
that
I
saw
a
lot
of
people
after
1.0
going
back
to
work
on
their
stuff
and
kind
of
feeling
issues
in
this
specification,
but
not
writing,
prs
or
nor
anything
else.
So
yeah
this,
I
don't
know
I
think
it's
part
of
the
process,
but
we
have
to
mention
that,
like
we
know
that
this
is
what
you
guys
are
mostly
focused
on,
but
we
are,
it
would
be
still
nice
to
have
some
cycles
from
you.