►
From YouTube: 2021-04-14 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
A
A
A
A
Right
at
the
minute,
it's
11
degrees,
but
it's
quite
warm
since
it
hasn't
been
out
for
a
while.
It.
B
B
B
A
Not
much
really
kind
of
just
in
between
a
few
things
closing
down
a
date
center
at
the
minute,
so
lots
of
migrations.
Oh,
are
you
going
you're
going
to
what
does.
A
B
B
A
Not
really
kind
of
just
as
it
goes
like,
I
am
trying
to
put
more
time
to
this,
but
I'm
not
I'm
only
getting
like
a
few
hours
a
week.
Hence
why
there's
sporadic
commits,
but
I
have
I've
brought
in
the
open,
telemetry
profiles.
A
I
thought
so
that's
getting
there.
I
now,
since
I
brought
those
in
mainly
because
the
get
the
github
actions
were
failing
for
everything,
so
that
got
that
far
now
it's
basically
now
going
through,
I
think
php
fan
and
some
other
stuff
to
clean
up
my
code
to
get
them
compatible.
I
did
have,
I
guess
one
quick
question
on
bringing
the
pro
proto
stuff
that
make
style
kind
of
complained.
A
C
B
B
C
B
B
A
Yeah,
I'm
not
not
entirely
sure
it'd,
be
I
think
it.
I
was
thinking
that
my
other
sdks
look
like
they
use
sub
modules
to
kind
of,
at
least
so.
You
know
that
whatever
proto
you've
got,
you
know
what
version
you're
going
to
work
into,
because
in
this
there's
no
reference
to
what
version
I
generated
right,
those
protofiles
at
which
is
isn't
great.
B
A
B
I
think
I
think
I'm
gonna
work
yeah,
I'm
gonna
open
a
low
priority
ticket
to
to
evaluate
that
eventually.
But
I'll
do
that
now,
I'm
thinking
about.
C
A
Yeah
other
than
that,
I'm
kind
of
I
think,
making
a
reasonable,
reasonable
progress.
I've
moved
a
few
few
bits
around
in
more
recent
commits
time
is
still
being
a
bit
complicated,
because
what
I
realized,
comparing
the
the
go
open,
telemetry
thing
at
least
time
within
go,
you
can
get
kind
of
wall
time
and
a
monotone
atomic
clock
kind
of
in
one,
whereas
the
way
that
we're
doing
it
they're
two
separate
entities,
which
is
what
I
was
having
issues
with
yeah.
So
yes,
yeah.
A
Yeah,
because
otl
or
tlp
is,
after
both
unix
time
in
nanoseconds,
so
it's
a
bit
harder
for
it
to
yeah
the
timing
tying
those
two
together,
I've
kind
of
fudged
it
in
my
test
at
the
minute,
we're
basically
specifying
basically
three
time
stamps,
I
think
in
effect,
okay.
So
at
least
it's
validating
that.
I
think
I
put
this
into
my
commit
at
least
it's
validating
that
the
duration
is
calculated
correctly,
but
other
than
that.
It's
not.
I
don't
think
it's
a
great
fix.
B
For
that
yeah
the
whole
that
time
package
we
spent
a
lot
of
time
like
coming
up
with
the
right
way
to
generate
a
real-time
clock
and
a
monotonic
clock,
and
they
are
it's
just
sort
of
like
it's
annoying,
because
you
know
you
want
to
use
both
at
specific
places
yeah.
But
it's
like
so
difficult
to
distinguish
between
the
two
of
them
and
know
which
one's
the
right
one
to
use
at
which
place.
B
A
B
So
we
we
were
just
go
walking
through
sean's
pr
and
talking
about
the
open-ended
things
that
he
has
rolling
on
is
there
anything
you
want
to
discuss
today.
Follow
me.
C
No
just
one
pr
that
I
have,
I
know
you've
looked
at
it,
but
I
think
we're
waiting
for
one
more
review
on
it,
but
yeah.
That's
that's
all.
I
have
really.
B
Okay,
I
will
I
I
pinged
all
the
contributors
for
that,
but
maybe
I'll
just
dm
amber
and
ask
her
just
give
it
the
thumbs
up.
So
I
can
merge
it.
I'm
trying
to
be
cognizant
of
getting
two
reviews
for
every
pr,
but
I'm
also
trying
to
be
cognizant
of
like
just
get
done
so
that
we
can
start
moving
on
to
the
next
thing,
rather
than
wait
a
week
for
somebody
else
to
review
it
when
the
review
is
going
to
be
inconsequential.
B
C
Yeah
yeah,
I
know
it's
a
tough
balance.
I
can
ping
amber
as
well.
Okay,
cool
yeah.
I
do
see
here
on
a
daily
basis.
B
That's,
I
guess,
that's
what
happens
when
you
work
at
the
same
company,
sometimes,
but
all
right,
so
we've
got
that
sean
we've
got
your
hlp
thing,
we've
got
the
udp
and
thrift
is
coming
along.
I
am
working
with
nebuci
on
the
event
tagging.
I
spent
a
long
time
rolling
through
the
events
api
today
and
I
realized
that
we
had
one
mismatched
signature
for
the
events
and
that's
why
events
fans
didn't
have
attributes
tied
to
them.
B
We
were
discussing
the
right
way
to
do
it
and
so
like
there
are
three
things
in
a
function:
signature
for
an
event.
You
have
a
time
stamp,
a
string
and
a
list
of
attributes
that
get
passed
to
in
that
function,
signature
and
in
the
so
we
have
an
event
api
and
events,
like
plural
and
the
api
has
the
right
signature
for
both
the
sdk
has
the
right
signature
for
event,
but
not
events.
B
So
that's
why
they're
not
getting
passed
in
properly
they.
I
guess
the
original
implementation
had
a
time
stamp
and
the
attribute
backwards,
and
then
both
sifkin
and
yeager
die
on
those
without
saying
anything.
So
they
just
don't
post
the
attribute.
If
it's,
if
it's
created
incorrectly,
so
we're
I'm
planning
on
fixing
that
with
him,
so
we
should
start
seeing
attributes
in
the
choices
for
both
in
zipkin
and
yeager.
For
our
examples
where
previously
we
were
not
so
look
for
a
pr
for
that
sooner
rather
than
later.
B
B
I
think
that
that's,
I
think,
that's
like
a
pretty
good.
Stop
like
a
pretty
good.
Next
start,
stopping
point
and
yeah.
That's
I
think
that's
all
I
have
for
today
did
anybody
else
have
anything
they
wanted
to
bring
out
fine.
Do
you
have
something
to
work
on
next?
After
you,
since
that's
w3c
services,
getting
closer
closer
to
done
dawn.
C
Yeah,
I
have
a
another
pr
I'm
working
on,
but
I
haven't
had
time
this
week.
It's
just
fixing
some
of
the
the
test
failures
in
there,
but
also
I
do
want
to
close
out
the
the
w3c
trace
context
issue,
because
I
mean
that
one's
been
it's
kind
of
been
done
for
a
while.
C
I
was
waiting
to
close
it
out
once
I
added
the
other
pr
to
fix
the
test
failures,
but
I
think
it
might
just
be
better
to
close
that
one
out
and
then
I'll
create
a
few
issues
for
the
other
smaller
test
failures.
C
Just
so,
I
think
someone
looking
in
can
see
that
we
do
have
the
trace
contact
stuff
in
there
already
and
then
there's
just
a
few
other
small
bug
fixes
really
with
them.
So
at
least
when
we
do
the
release
that
could
be
part
of
it
that
we
know
that
there
is
a
trace
context
stuff
in
there
all.
C
Forget
about
it:
okay,
cool
yeah,
I'll,
create
a
few
more
tickets
issues
relating
to
the
other
test
failures,
which
should
be
fixed
with
the
pr
that
I
put
up
sweet.
C
B
C
No,
that's
that's
separate
as
well
yeah,
and
I
know
I've
been
lingering
on
that,
so
I'll
get
to
that
soon,
no
sweat
at
all.