►
From YouTube: 2022-03-22 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
A
C
C
A
Okay,
so
there
is
a
wee,
are
curious
about
the
agenda
you
mentioned.
You
want
to
talk
about
http
scope.
D
Yes,
we
can
do
it
just
give
me
a
second.
D
D
All
right
so
can
you
see
it.
D
Yeah
great
so
here
we
have
several
items
that
we
put
to
the
scope
of
v1
previously,
so
error
status
is
something
which
is
which
already
was
discussed
and
the
the
corresponding
pull
request
was
just
closed.
So
the
intention
is
to
keep
everything
as
is
right
now,
basically
to
do
nothing.
D
So
we
do
this
required
attribute,
sets
item
and
it
will
be
good
to
also
discuss
it
today.
So
we
do
have
this
corresponding
issues
issue
opened
by
the
milla
in
november,
so
this
this
one
can
we
can,
we
can
discuss
for,
tries
and
redirects.
We
do
have
two
things
here,
so
the
first
one
is
already
was
already
emerged,
and
basically
this
this
pull
request
is
is
about
how
we
want
to
model
retries
in
the
direction
when
it
comes
to
spence
and
their
structure.
D
But
linking
parts
is
not
there,
so
linkin
park
part
was
excluded
and
we
are
currently
talking
about
like
how
we
can
or
should
we
use,
links
to
to
add
this
casuality
or
like
how,
if
it
will
be
links,
how
we
want
to
use
them
which
attributes
we
want
to
add
and
so
and
so
on
and
another
item
we
do
have
here
is
context
propagation,
and
this
one
is
also
kind
of
important
when
it
comes
to
links,
because
that's
that's
really
important
how
we
want
to
propagate
the
information
between
retries,
for
example,
or
between
redirect
even
within
the
same
process
so
and
since
no
one
started
to
work
on
it
yet,
and
it
is,
it
is
the
end
of
march
right
now
my
suggestion
here
just
to
exclude
it
from
the
scope.
D
So
essentially
we
have
two
items
here
and
items.
The
item
related
to
links
for
entries
and
directs
probably
also
can
be
excluded
because
links
themselves
are
not
yet
finalized.
How
how
like
which
scenarios
you
want
to
use
them
and
which
are
exact
kind
of
way
of
using
links,
so
it
it
just
easier.
So
we
can
consider
this
can
be
kind
of
links.
Part
can
be
just
excluded
from
this
code,
and
probably
we
need
to
focus
on
this
required
attribute
set
item,
which
is
only
left
here.
A
A
Then.
Oh
sorry,
sorry,
this
is
the
first,
it's
basically
a
refactoring
of
this
thing
and
picking
a
set
of
attributes
for
a
server,
but
then
the
we
will
need
to
go
and
categorize
the
rest
of
attributes
into
like
not
required
or
optional,
but
a
different
set
of
things.
So
let
me
share
my
screen
for
a
second
okay,
so
I
did
it.
A
A
A
It's
just
an
example
so,
and
here
yeah,
as
I
mentioned,
the
first
thing
would
be
to
split
attributes
into
server
and
client,
because
currently
it's
a
mixture
and
it's
very
hard
to
say,
what's
required
for
server,
what's
required
for
client
right
and
then
there
are
like
questions
how
to
split
them
and
then
basically
apply
this
categorization.
A
So
when
I
like
my
concerns
on
doing
this
work
is
that
unless
we
get
some
sponsorship
from
tc,
it's
just
endless
arguments
right.
So
it
would
be
great
if
we
can
get
some
agreement
that
it's
the
right
direction
before
we
do
all
this
convincing
stuff
right.
A
So
my
question:
if
there
was
a
dc
member
here,
was
to
if
they
want
us
to
do
this
work
at
all,
but
I
don't
know
what
do
you
guys
think
about
it.
D
So
when
it
comes
to
tc
member
looks
like
riley
we'll
be
working
with
us
with
this,
this
particular
sick,
but
he
will
be
working
with
us
starting
from
april.
D
D
Well,
yeah,
that's
a
good
point
and
we
started
this
discussion
for
scope
itself
back
in
october,
probably
last
year,
so
that
time
it
was
different
understanding
how
we
want
to
make
progress.
So
previously
it
was
not
so,
at
least
at
least
my
understanding
was
that
we
don't
require
any
kind
of
dc
member
to
for
us
to
make
progress,
but
at
the
same
time
it's
it.
We
like,
regardless
of
what
we
have
in
scope
and
how
we
want
to
make
a
progress.
It's
another
thing.
D
That
probably
will
be
also
be
interesting,
actually
how
we
want
to
call
or
how
we
want
to
announce
the
spec
stable.
I
believe
we
don't
have
any
kind
of
experience
like
like
this
in
open
telemetry.
Yet
so
all
the
specs
are
currently
experimental.
So
there
is,
there
are
no
stable
specs
and
that's
not
a
question
like
how.
What
is
the
procedure?
How
we
want
to
make
it
make
it
official
how
we
might
how
we
want
to
announce
we
won
and
yeah.
D
That
definitely
requires
requires
some
communication
with
tc
technical
committee
and
again
like
a
scope
discussion,
probably,
and
what
not
so
yeah.
That's
a
very
good
question.
I
don't
have
like
a
clear
answer
to
that.
D
Right
so
yeah
th,
that's
that's
fair,
but
at
least
what
we
can
do
here
like
we
can
prepare
ourselves
and
have
an
agreement
within
this
group
that
we
are
overflying
this
with
the
current
http
semantic
convention
spec
to
be
announced,
as
we
want
so
looks
like
it's
it's
for
it.
It's
it's
our
possibility
here
to
make
some
progress
and
just
like
include,
let's
say
or
ask
dc,
to
participate
at
some
at
some
later
points.
Still
so
when
it
comes
to
required
attribute
sets,
I
believe,
yeah.
D
The
proposal
that
you
have
is
is
different
is
actually
something
that
clarifies
a
lot,
but
in
the
same
time
it
looks
like
we
need
to
restructure
the
whole
document
and
like
a
practically
it
means
we
have.
We
will
be
having
a
lot
of
changes
to
the
to
the
document
to
the
specification
itself,
so
maybe
we
can
have
some
tray,
like
a
have
some
like
a
first
step
here,
just
to
split
these
two
attributes
this.
D
This
split,
these
two
attribute
sets
that
we
have
basically
this
this
kind
of
the
ones
that
we
see
as
a
controversial
like
url
versus
scheme
host
and
target.
So
we
can,
we
can
start
from
that.
So
we
already
have
http
client,
section
and
http
server
session
and
those
attributes
are
basically
copy
pasted
in
both
two
sections.
D
So
at
least
this
part
can
be
clarified
and
we
can
explicitly
say
like
how
it
should
be
done
for
our
client
and
how
it
should
be
done
for
server.
So
I
I
believe
this
part
can
be
done
regardless,
so
yeah
I
will.
I
will
share
my
screen
in
a
second
again
just
to
well
yeah
yeah.
You
can
do
it.
Oh.
D
No
go
ahead,
that's
fine!
So
if
you,
if
you
go
to
the
http
client
section,
you
see,
we
have
this
additional
attributes
requirements
right
so
and
it
says
like
at
least
one
of
the
following
attributes
is
required,
but
based
on
my
understanding,
we
have
on
the
client
side,
we
have
http
dot
url
only
right.
So
there
is
no
case
when
we
can
have
scheme
hosts
and
targets
separately.
D
D
A
Okay,
so
let
me
see,
I
think
we
have
an
issue
like
that.
I
think
trust
created
one
back
in
the
day
millionaire.
If
you
remember,
let
me
see
if
there
is
an
issue
and
if
there
is
some
discussion
there,
if
there
is
any
agreement,
I
can
try
approaching
this
problem
and
sending
it
here.
D
D
We
will
be
not
having
this
controversial
kind
of
things
and
in
two
different
sections,
basically
saying
the
scene,
but
when
it
comes
to
the
overall
kind
of
spec,
it
probably
might
be
also
like
a
better
if
we
just
maybe
eventually
split
this
into
two
documents
instead
of
having
just
you
know
some
some
shared
part
and
like
different
sections,
maybe
it
will
be
better
to
have
it
to
have
two
additional
two
separate
documents?
Maybe
not,
but
this
is
something
that
we
can
discuss
as
a
second
step,
so
that
that's
my
proposition.
A
D
A
Yeah,
I
understand
the
proposal.
I
think
there
will
be
some
controversy
here
right
because
for
some
reason
we
have
three
sets
and
like
still
it's
better
to
discuss
with
some
people
beforehand.
A
D
That's
true,
it
probably
will
require
some
additional
work
for
us
to
do
in
some
additional
cycles
to
be
spent,
and
maybe
some
additional
iterations
but
yeah
I
mean,
even
if
we
agree
with
that.
That's
another
thing
which
is
not
really
clear
for
me.
Even
if,
if
you
have
one
like
a
tc
member
on
our
side,
it
still
might
be
the
same
right.
So
we
can
have
another
c
member
who
is
whoever
like
who
can
have
another
opinion,
and
it's
still
still
something
that
we
need
to
do
like
in
iterations.
D
So
yeah,
it's
the
overall
since
the
overall
process,
not
really
clear
it's
hard
to
predict.
Does
it
make
sense
to
do
something
or
not
like
right
now
or
just
postpone
it,
but
at
least
we
we
can.
We
can
get
an
agreement
within
this
group.
B
I
think
for
required
attributes
we
do
need
to
understanding
of
sort
of
the
state
of
libraries
and
languages
like
I
can
provide
that
for
java,
but
I
don't
know
what
this
data
is
in
python
or
go
and
stuff
like
my
first
reaction
here
is
at
least
in
java
libraries.
We
always
have
the
host,
so
I
would
not
need
these
other
sets.
I
don't
know
what
it's
like
in
other
languages.
D
Yeah
that
make
that
that's
fair,
so
basically
it's
almost
always.
We
do
have
the
scheme
hosting
targets,
but
like
this
peer,
pure
name
and
pure
ip,
is
basically
some
some
kind
of
corner
cases
right.
D
So,
for
example,
if
you
have
just
ap
address-
and
there
is
no
kind
dns
name
then
the
the
only
thing
that
you
can
have
is
just
kind
of
ip
address.
A
Yeah,
I
think
that
the
point
that
we
should
think
about
is
that
what
we
actually
have
everywhere-
yeah
in
java-
we,
if
we
have
this
list
or
like
a
handful
of
languages
right,
we
can
say
that
90
percent
of
instrumentations
have
let's
say
this
set.
Then
probably
we
should
use
this
stat
and
wait
anyway.
So
having
some
data
would
be
useful
and
maybe
a
good
step
forward
would
be
to
collect
this
data
before
we
make
any
decision
at
all.
D
That
makes
sense,
but-
and
I
said
I
don't
think
that
we
can
collect
this
data
and
it
will
be
kind
of
enough
for
all
the
for
all
the
platforms
and
and
languages.
So
we
can.
We
can
do
it
for
javafor.net,
probably
for
python,
but
we
do
have
much
more
so.
A
A
B
Technical
detail,
but
you
mentioned,
we
wouldn't
have
http
host
if
we're
not
using
a
dns
name,
but
at
least
the
interpretation
I've
always
had,
and
I
think
how
implementing
java
is.
The
value
of
host
is
just
the
hostel
sent
by
the
place.
There
would
be
like
the
address
if
they
won't
use
any
dns
name.
B
A
Oh,
I
see
well,
I
think
there
is
some
different.
B
A
Right-
and
I
think
I
also
need
to
refresh
my
understanding
of
the
difference
between
these
things-
yeah,
okay,
yeah.
So
then
I
guess
that's
the
next
step
here,
regardless
of
any
support
from
tc,
would
be
to
actually
formalize
the
proposal
based
on
some
information
that
we
can
collect
about
instrumentations
and
some
clarification
of
difference
between
the
host
names
attributes
here
right
there.
The
rest
is
the
same.
It's
just
the
hostname,
which
has
different
things.
D
That'll
be
awesome,
yeah
and
I
will
create
all
the
other
issues
just
to
indicate
what
what
is
in
scope
and
what
is
not
what
we
want
to
exclude.
So
basically,
that's
the
only
item
we
have
in
in
scope.
From
this
perspective
and
the
sooner
we
get
it
done,
the
better
we
can
announce
or
start
announcing
v1
as
a
stable
or
the
current
specs
table.
D
Yeah
it's
this.
This
one
is
hard
because
we
do
have
different
like
perspectives
on
how
we
want
to
use
links
and
do
we
want
to
use
them
and
for
rich
use
cases.
So.
D
It's
definitely
something
that
we
probably
need
to
just
start
with,
but
yeah
it
was.
It
was
a
great
discussion
last
thursday,
with
bogdan
about
links
and
how
we
like
it
doesn't
make
sense
to
use
links
from
from
the
retrys
and
redirects
for
a
level
perspective.
D
It
was
mentioned
that
we
might
achieve
the
same
when
it
comes
to
kind
of
casuality,
when
we
just
use
attributes
saying
that
if
you
have
just
attribute
describing
the
the
retry
session
kind
of
that's
something
that
we
can
just
add
as
attributes
to
all
the
spans
and
and
then
we
can
group
so
like
on
the
server
side,
we
can
do
some
grouping,
it's
probably
true,
but
in
the
same
time
it's
not
really
obvious
how
to
do
that
and
how
to
propagate
this.
D
This
attribute
this
this
session
id
or
something
which
basically
just
is
the
same
as
propagation
information
about
links,
but
with
links.
We
can
also
have
a
direction
and
we
can
like
add
some
kind
of
ordering
like
which,
which
retry
was
what
which
trial
was
the
first
and
which,
which
is
second,
which
is
not
really
possible
to
do
this.
D
Just
a
separate
attribute
it
will
be,
it
should
be
another
attribute,
then,
to
indicate
the
retry
count,
but
in
our
case
we
still
we
have
it
anyway,
so
these
two
can
be
can
solve
the
problem
but
yeah
when
it
comes,
for
example,
to
to
redirects
it's
not
that
simple,
so
I
will
probably
update
the
pull
request
that
I
already
have.
D
So
I
do
have
this
the
separate
pull
request
to
add
links
for
it,
rights
and
redirects,
and
we'll
just
add
this
information
showing
some
examples
and
basically
just
explaining
what
I
just
said
in
written
form
so
and
I
believe
the
next
discussion
about
links
this
book
done
will
be
happening
in
on
tuesday
on
thursday.
Sorry,
so
in
today's
8
a.m,
third
day
am
and
that
that's
we
can.
We
can
use
this
time
to
to
make
some
progress
on
links.
A
Where,
maybe
I
have
one
question,
I'm
sorry
josephine
to
put
you
on
spot,
we
haven't
met,
we
can
introduce
ourselves
and
it
would
be
great
to
hear
from
you
sure
yeah,
I'm
I'll
go
ahead.
Sorry,
no
you're,
good
yeah!
I
just
wanted
to
introduce
myself
I'm
with
neil.
I
work
at
microsoft
on
azure
sdks.
I
spent
some
time
on
the
tracing
and
semantic
conventions
efforts.
F
Yeah
I
work
in
a
startup
called
the
mode
analytics
and
we
are
doing
a
lot
of
tracing,
and
so
our
guiding
into
the
open,
telemetry,
github
library
and
fund
is
fascinating.
So
I
decided
to
join
some
of
these
six
to
the
mo.
A
So
like
there
are
these
meetings
where
we
just
talk
about
mostly
random
stuff,
with
some
proposals,
there
are
some
different
big
work
streams
people
participate
in.
You
can
also
find
us
on
slack
if
you
want,
probably
you
know
about
it.
Otherwise,
you
can
find
everything
in
the
community
repo.
They
have
community
repo
link
right.
F
A
A
So
I
can
there's
like
channels
here
yeah.
If
you,
let
me
paste
the
link
here.
So
this
is
a
slack
channel.
You
can
find
open,
telemetry,
instrumentation.
A
Okay,
so
look
around.
There
are
multiple
channels.
There
find
the
area
you
would
be
interested
in
the
most
and
feel
free
to
join
any
meetings
or
participate
in
any
discussions.
You're
very
welcome.
Thank
you.
D
D
A
A
Okay,
very
quiet.
A
I
see
okay,
it
sounds
like
the
the
action
item,
so
we
don't
have.
We
don't
yet
know
what
should
be
the
required
attributes
for
in
scope.
4V1
we're
just
figuring
out
some
stuff
to
understand
right.
A
D
And
another
item
is
proposal.
D
Yeah,
it
was
something
something
else.
My
idea
was
to
mostly
kind
of
describe
why
it's
essential
or
why
we
require
links
for
advice
and
redirects,
but
it's
most
it's
not
about
attributes.
Attributes
can
be
the
part
of
it.