►
From YouTube: Monitor:APM Weekly Meeting
Description
Discuss 12.5, and new issues
A
A
B
Before
joining
github
I
was
working
as
a
Ruby
on
Rails
developer
for
unum
Oh
for
over
the
fears.
I
started
there
as
a
junior
developer
and
walked
up
to
the
senior
and
team
leader
role.
The
UNA
provided
the
software
service
platform
for
the
tracking
search
and
the
search
engine
presence
and
the
social
media
presence.
So
basically
we're
crawling
Google
being
Yandex
and
provide
the
insights
for
our
customers.
B
Back
then
I
was
reading
the
back-end
team,
which
were
responsible
for
maintaining
and
improving
the
reporting
up,
which
was
built
on
Ruby
on
Rails
fifth
version
and
right
now,
I
joined
in
developer.
For
me,
a
PM
team
and
I'm
happy
to
be
here
and
to
happy
to
collaborate
with
all
amazing
people,
I
already
met
and
I'm
looking
forward
to
get
to
know
each
of
each
of
TD
club
member
great.
Thank
you.
B
A
All
right,
I
have
a
couple
topics
to
kick
us
off
today.
First,
one
is
around
the
kind
of
the
scope
of
12.5.
We
have
oh,
it
went
up
now.
We
have
48
issues
and
that
changed
in
the
last
hour
since
I
wrote
that
so
we
have
48
issues
currently
assigned
to
12.5,
which
seems
like
too
many
by
a
considerable
amount.
A
So
I
just
wanted
to
kind
of
talk
about
process
a
little
bit,
and
my
next
point
kind
of
gets
to
part
of
this,
but
also
part
of
the
problem,
is
that
we've
moved
like
I
counted
14
that
have
the
two
scheduled
label,
which
just
means
that
we
move
them
from
12
for
12
5,
and
we
need
to
figure
out
what
to
do
with
them.
Now.
Yeah.
C
C
Usually
the
ones
on
the
button
which
represent
like
lower
priority
will
move
out
so
I
assume
that
14
issues
will
be
removed
from
the
court
milestone.
Probably
although
it
depends
on
their
for
destination,
but
still
it's
like
a
remote.
You
don't
try
to
do,
but
still
even
without
14
issues.
It's
the
four
which
I
think
a
bit
higher
than
what
we
planned
before.
We
know
positions
right.
A
Yeah
one
thing
is:
we
had
a
few
new
issues
actually
moved
a
couple
of
them
out
already,
but
there's
new
issues
that
were
kind
of
created
and
added
to
12
5
in
the
last
well
this
week,
so
I
listed
out
some
of
them
there
and
I'm
not
saying
I,
don't
need
to
like
dig
into
each
one
of
them.
I'm.
Just
saying
like
that's
another
thing
contributing
to
the
number
going
up
is
that,
as
we
add
new
issues,
sometimes
we
just
assign
them
to
the
current
milestone
and
I.
D
One
of
those
ones
issues
when
I
mentioned
down
there,
34
784
I
just
created
an
mr
that
when
I
stumbled
on
this
bug
earlier
this
week
and
the
reviewer
asked
me
today,
could
you
make
an
issue
for
that
which
I
hadn't
planned?
Don't
do
them,
but
I'll
do
it
from
now
on
fsor,
if
that's
the,
if
we,
if
we
need
an
issue
to
go
along
with
it,
but
you
know
that
one
should
be
it's
already
in
review.
E
We
shouldn't
we
don't
necessarily
need
an
issue,
one
issue
per
mr,
I
mean
I'm
not
sure
if
there
was
maybe
there
was
some
reason
for
it.
I
don't
know,
but
in
general
I
don't
think
we
need
to
so
you
just
mentioned.
If
you
have
to
do
it,
I
don't
know
that
we
I,
don't
know
that.
That's
a
thing,
I,
don't
think
we
have.
D
D
A
I
know
a
couple
of
them
that
I
moved
out
already.
There
were
just
things
that
and
a
similar
they
were
small.
They
came
up
during
an
MRI
review,
so
this
in
my
case
it
was
Adrian
had
an
mr
that
was
during
the
review.
It
created
an
issue
to
address
something
and
it
was,
but
he
wasn't
quite
sure
where
to
assign
it
to
make
sure
I
got
prioritized,
so
he
just
put
it
in
the
current
milestone.
A
I
moved
it
to
twelve
six,
the
next
milestone,
but
I,
don't
know
if
that's
the
right
thing
to
do
either
if
there's
a
better
way
to
to
schedule
that
so
that's
kind
of
what
I
was
trying
to
get
at
is.
Do
we
have
a
good
way
that,
as
new
things
come
in
to
kind
of
like
them
and
yeah
I
feel
like
it's
similar
to
the
to
schedule,
label
that
maybe
we
could
just
use
reuse
that
same
thing,
but.
G
G
And
we
maybe
put
it
in
the
backlog
or
something
when
p.m.
gets
an
opportunity,
they
can
just
go
through
the
top
of
the
backlog
and
see
if
there's
something
really
important,
and
that
goes
in
the
next
white
straw.
Instead
of
pushing
everything
into
the
next
milestone
ourselves,
you
just
put
everything
in
the
backlog.
If
it
looks
like
something
that
should
go
into
this
or
next
release,
then
we
pull
them
as
amended
wire
to
the
milestones.
E
Do
that
we
did
that
at
one
point
and
then,
when
I
guess
I
said
I
think
it
would
just
depend
as
far
as
like.
What's
most
helpful
for
youdo
I
mean
we
can
help
support,
you
know,
whatever
workflow
you
think
is
the
best
for
Ken
for
Kenny.
That
meant
like
taking
like
not
putting
a
milestone
on
it
now,
putting
back
on
just
leaving
it
empty
because
then
it
would
get
surfaced
and
they
get
lab
triage
issue
every
week.
E
C
Know
I'm
just
thinking
out
loud,
so
one
thing
about
putting
things
on
the
backlog
that
I
feel
for
is
that
they
will
end
up.
You
know,
get
lost
in
an
enormous
amount
of
issues.
You
know
it
can
go
to
more
than
hundreds
issue
under
the
issue
and
it
will
get
lost.
I
can
go
over
them
every
now
and
then,
but
you
know
going
over
a
hundred
of
issues,
not
scalable,
so
we
can
put
them
in
without
without
any
milestones,
so
they
will
surface
out,
and
then
we
can
try
to
understand
where
we
need
to
fit.
C
C
I
mean
there
is
some
sort
of
like
a
grooming
meetings
that
people
are
doing
for
issues
I
so
like
the
plan
is
driving
them
together
with
with
engineers
and
I,
was
thinking
you
know,
maybe,
which
we
have
time
we
can
carve
like
10
15
minutes
and
go
over
the
new
issues
for
the
last
week,
and
then
you
know
we
can
talk
about
it.
We
could
put
some
comment
and
then
we
can
assign
them
to
the
right
milestone.
E
E
C
C
Of
course
it
depends
on
the
amount
of
issues,
so
we
can
give
it
a
try
for
you
know
few
iterations
like
we
are
doing
with
the
weights
and
see
if
it
works,
and
if
not,
we
will
try
something
else,
so
it
doesn't
so
basically
doesn't
matter
if
you
put
them
on
blank
backlog
or
whatever,
let's
make
sure,
to
take
10-15
minutes,
30
minutes
from
the
weekly
meeting.
If
you
have
time
to
go
over
the
new
issues,
then
you
should
go
and
make
sure
to
label
them,
and
you
know
asylum
to
the
like
milestone.
F
E
A
A
E
A
A
I
C
I
D
A
C
This
is
something
that
that
I
open
based
on
Ashla
conversation,
what
happen
in
the
in
the
product
channel
and
so
right.
Now,
when
we
are
customizing,
we
asked
our
customer
to
customize
the
magic
dashboard.
You
need
to
go
to
the
root
directory,
folder,
locate
the
file,
update
the
file
and
then
go
back
to
the
dashboard
and
apply
to
chain
and
see
the
changes,
and
the
proposal
is
that,
once
you
will
click,
there
is
something
similar
in
that.
C
Once
you
will
click
on
the
on
the
panel
on
the
tie,
you
can
immediately
either
it
will
redirect
you
to
the
file
or
it
will
reveal
the
file
kind
of
fits
into
the
file,
create
some
sort
of
a
visualization
that
you
see
the
Yammer
file
and
then
you
can
amuse
it
immediately,
change
it
updated
and
maybe
to
like
a
major
quest
or
or
commit
update.
The
file
immediately
see
and
see
the
changes
made.
D
F
E
The
queries
in
that
llamo
file,
although
this
is
a
really
interesting
proposal-
I
wonder
whether
we
wouldn't
be
better
served
to
kind
of
pursue
this
other
path
here,
because,
if
we're
essentially
having
like
a
link
out
to
the
web,
IDE
I
don't
know.
This
is
interesting,
because
if
this
is
like
the
longer
vision
for
where
we're
going,
maybe
we
need
to
rethink
yeah.
C
C
Is
it
true,
Yaman
file,
or
is
it
true,
you
know
updating
it
on
the
UI
I
think
you
know
it's
easier
and
nicer
to
do
it.
You
know
to
the
UI
without
updating
the
ml
file,
but
do
we
think
get
with
that
suggestion
that
I've
made
we
have
like
more
options
because,
usually
when
you
are
limited
to
to
the
UI
you
have
you
have
some
limitation.
C
D
G
D
You
know
where
you
edit
your
thing
and
and
then,
when
you
try
to
go
away
from
it,
it's
like.
Don't
you
want
to
save
your
your
changes
and
it
does
change
and
save
that
to
some
kind
of
Jaison
representation.
So
the
only
thing
that's
missing
from
that
you
know
interaction
is
a
is
creating
a
merger
quests,
but
I
mean
they.
They
do
something
along.
The
lines
of
you
know,
track
changes
for
for
each
of
those
things
so
I
mean
we
can
start
from.
We
can
start
from
that.
C
D
E
I
mean
it's
because
we
we
already
have
all
that
interface
built
out,
that
whole
thing
is
already
done
for
us,
even
if
we
could
like
get
that
into
a
modal
or
something
I,
don't
know
how
reusable
it
is.
Maybe
we
kick
them
out
to
the
web
IDE
for
now
and
then
worry
about
how
to
integrate
it,
but
it
would
be
really
cool
to
kind
of
see
like
you
make
a
change
to
the
yamo,
and
then
it
like
updates
your
dashboard
and
you're
like
previewing.
What
changes
you're
gonna
make
and
then
you
can
commit
them.
F
A
C
C
J
C
C
Basically,
when
we
tell
our
customers
to
change,
even
the
default
metric
dashboard
and
not
like
you
know,
using
some
sort
of
version
control,
we
tell
them
to
start
the
file
from
scratch
and
I
know
that
we
have
Modiano
file
in
our
source
code.
So
why
not
expose
it?
So
you
know
customer
at
least
will
have
some
sort
of
template,
so
they
can
play
around,
but
this
boat
it
again
would
file.
C
They
can
do
like
one
on
one
plus
one
say:
okay,
that's
that's
the
ML
file
that
I
need
to
then
I
need
to
update
and
I.
Think
it's
like
an
easier
experience
than
starting
the
MFR
from
scratch,
because
it's
very
sensitive
you
know
within
the
Dacian
and
everything
and
it's
just
a
burial
for
experience,
but
it's
similar
to
to
the
previous
one.
It's
kind
of
related,
so
sure
we
can
skip
the
discussion.
I
guess
for
now.
G
E
A
A
Sure
yeah,
so
now
this
one
doesn't
have
a
milestone.
Should
we
go
over
few
one?
Yes,
I'm
wondering
what
the
what
we
wanted
to
hit,
how
we
have
want
to
handle
these
now.
Do
we
you
wanna,
leave
it
without
a
milestone?
It's
you
want
to
take
this
time
right
now
to
try
to
put
it
where
you
think
it
belongs.
This
one
might
be
a
weird
case
because
there
might
be
a
duplicate
of
it,
but
no
I
mean
we
should
general.
C
It
depends
I
mean
it
really
depends.
If
we
think
it's
critical
bug
you
can
assign
it
to
the
code
milestone
if
we
think
we
are
missing
some
I
mean.
The
reason
for
this
discussion
is
to
better
understand
the
the
issue
is
clear
and
we
have
enough
detail
and
how
critical
it
is,
and
if
you
know
it
has
to
right,
labeling
and
everything
okay.
So
if
this
is
a
duplication,
then
it's
better,
you
know,
put
up
duplication
here.
Yeah
then
we
can
assign
the
right
milestone.
G
D
G
G
G
A
C
A
D
A
D
C
C
A
D
D
H
A
E
This,
so
is
this
something
that
we
want
to
do
from
the
front
end
like
this
I
mean
I,
think
that
was
part
of
the
discussion
that
Miguel
and
I
were
having.
Whether
we
want
to
essentially
pass
the
entire
UML
file
to
the
front
end,
and
then
we
rearrange
it
and
then
pass
it
back
or
whether
we'd
rather
I
mean
I,
don't
know,
what's
easiest
as
far
as
he.
D
D
F
E
Of
the
other
things
that
we
talked
about
was
potentially
I
mean
eventually
right
now,
all
we're
interested
in
is
rearranging
right
for
this
particular
issue.
We're
just
interested
in
rearranging
the
existing
queries,
but
eventually
the
goal
is
we're
gonna,
be
able
to
add
and
remove
them
as
well.
So
when
we
think
about
it
like
from
that
perspective
and
doing
all
those
things,
I'm
not
sure
what
type
of
communication
between
front
end
and
back
and
makes
most
sense.
E
Do
you
have
a
preference
thinking,
I
mean
I
mean
we
haven't.
We
don't
we'd
have
to
investigate
I'm,
sure
I'm
not
don't
mean
to
put
you
on
the
spot,
but
I
guess
what
I'm
saying
is
I'm,
just
not
sure
yeah
that
whether
we
should
proceed
like
this,
with
the
Jason
approach
and
or
whether
we
should
do
something
else.
That
makes
sense
when
we
think
about
you
know
right
now
we're
just
interested
in
rearranging.
E
C
E
D
You
know
so
the
things
we
would
need,
if
we
just
did,
that
would
be
some
kind
of
facility
similar
to
the
CI
lent,
that
we
have
for
checking
your
syntax
or
whatever,
making
sure
it's
it's
valid.
We
might
meet
my
needed
a
dashboard,
a
linter,
but
that
would
be
pretty
I
mean
that
would
be
just
it's
Santi
a
Jason
looks
good
or
something
along
those
lines,
but
but
that
would
be
a
hundred
sent
back
in
thumbs-up
or
thumbs-down.
With
some
error
message,
you.
E
E
E
Yeah
I
mean
I
think
we
have
the
design
like
from
the
front-end
perspective.
We
have
the
designs
that
I
think
Amelia
gave
us,
which
was
you
know,
creating
that
modal,
that
pops
up
and
then
oh
yeah,
it's
right
there
like
on
the
on
that
page
there,
which
was
this,
was
the
design
and
it
makes
sense
it's
just
I'm,
not
sure,
from
an
implementation
standpoint,
whether
we
should
continue
down
this
path
or
whether
we
need
to
rethink
it.
Now
that
we
know
kind
of
the
longer
term.
E
E
Okay,
I
don't
know,
maybe
this
maybe
this
needs
an
epic
and
we
can
kind
of
tie
these
issues
together,
but
this
whole
effort,
I,
think,
is
you
know
something
we'll
have
to
kind
of
strategize
and
I
know
that
Miguel
is
dedicating
some
time
to
working
on
this.
So
before
we
get
too
far
down
that
path,
I
think
we
should
have
had
that
discussion
and
you
know
figure
out
what
what
approach
we
want
to
ultimately
take
here.
Just
so
we're
not
if
we
do
need
to
pivot.
K
A
A
A
C
I
C
E
E
C
I
think
you're,
right
and,
and
the
question
is:
should
we
start
building
it
from
the
ml
first
with
the
web
ID?
Should
we
go
straight
to
the
UI
I
prefer
to
start
building
it?
You
know
bottom
bottom
up
and
I
think
like
whether
the
first
and
only
then,
the
UI,
that's
my
personal
preferences,
but
I
mean
it's
up
to
you
to
basically
decide
what's
better.
What
do
you
think
is
better
I
think
you
know,
building
is
from
bottom
up
makes
more
sense,
because
the
UI
is
something
that
we
can.
Ideally
we
can.
We
can
change.
J
C
E
J
D
I
Little
nervous,
if
we
were
like
getting
I
mean
I
know,
maybe
we
can
work
towards
it,
but
if
we're
requiring
users
be
able
to
look
at
it,
the
ml
file
or
dashboards
like
it's
kind
of
it's
making
it
pretty
complicated
and
I.
Don't
know
that
I
know
that
you
know
we're
talking
about
using
dashboards
for
lots
of
different
things
outside
of
this,
and
that,
like
maybe
non-technical
people
are
gonna,
be
using
it,
and
in
that
case
they
like
are
they
gonna,
be
able
to
edit
a
dashboards
like
longer
term?
I
C
C
D
Got
an
idea
for
how
we
deal
with
these
dashboards
I.
Think
I
could
give
you
30.
Second
blip
is
I
think
we
ought
to
be
able
to
do
it
like
we
do
edit
issues
now.
I
should
be
able
to
say
you
know,
pound,
sign,
dashboard,
name
and
select
it
and
have
it
show
up
in
a
wiki
page
and
either
be
live
or
you
know
either
be
live
or
or
an
image
from
a
point
in
time,
and
that
gives
and
that
way
everybody
dashboards
get
created.
Often
you
know
and
I
think
we
should
start
with.
D
D
In
Cortana
dashboards
start
naming
our
dashboards
a
particular
way
so
that
we
can
reference
with
a
you
know,
a
special
character.
I
know:
we've
used
almost
all
the
special
characters
in
the
wiki
now,
but
you
know:
I
use
a
special
character,
dashboard
name
to
pull
that
dashboard
into
a
just
a
general
wiki
page,
and
that
would
give
us
the
ability
to
make
gitlab
style
dashboards,
that
anybody
could
use
start
with
crow
fauna
end
up
with
using
that
dashboards,
that
we've
decided
we've
defined
and.
E
E
You
know
with
who
and
I
think
the
problem
here
that
I'm
trying
to
avoid
as
front-end
is
out
a
little
bit
too
far
ahead,
like
we're
already
working
on
this
and
I
think
we
need
to
slow
that
down
a
little
bit
until
we
get
the
actual
like
back-end
piece
in
place
and
then
we'll
know
a
little
better
how
to
in
integrate
this,
because
otherwise
I
guess
that's.
My
concern
is
I.
Think
we're
just
out
a
little
too
far
ahead.
E
C
A
C
It's
multiple
iteration
I
think
you
know
at
least,
let's
provide
our
users
with
some
sort
of
something
that
is
easy
to
use.
I
mean
I'm,
not
saying
it's,
it's
not
usable
today,
but
it's
hard
time
as
a
new
user
that
try
to
update
dashboard.
It
took
me
a
while
until
I
was
able
to
do
some
and
if
I
would
have
liked
a
link
to
the
file
and
no
one
have
some
sort
of
a
template.
Okay,
this
is
the
name.
Is
it
change?
It
will
make
my
life
video.
A
C
H
A
G
A
E
D
The
alerts
when
the
alerts
were
sent
over
to
to
Prometheus
they're
sent
over
as
a
as
a
background
process
and
there's
zero
reporting
back
goes
to
if
they
were
valid
or
not.
That's
how
I
discovered
this
equals
and
not
equals
and
double
equals
issues.
I
was
looking
through
broad
logs
figure,
not
wama
alerts
were
being
said.
E
E
C
A
F
C
C
F
A
C
E
E
H
C
No,
maybe
just
leave
it
the
way
it
is
if
I
mean
let's
continue
doing
this
for
the
next
couple
of
meetings
and
see
if
it
makes
sense,
I
think
that
some
of
the
issues
that
were
open
this
week
was
reviewed
because
of
like
we're,
like
in
the
middle
of
I'm,
just
thinking
about
finalizing
some
issues
and
opening
new
one.
Maybe
next
week
will
have
even
less,
but
let's
not
see
the
trend.
Let's
see
know
how
many
issues
we'll
picked
up
last
week,
yeah.
C
Yeah
one
Wireless
comment
from
from
one
last
comment
from
a
just
sharing
some
feedback
that
I
got
from
who
get
labels
on
the
latest
a
list.
They
were
really
happy
with
some
of
the
features
that
we
launched,
formerly
the
one
that
we
can
select
like
multiple
codes
in
multiple
environments.
I
got
like
really
good
feedback
and
I
just
wanted
to
share
it
with
the
team.
This
is
really
valuable,
valuable
features
that
we
are,
we
can
actually
be
using
in
production.
So
thank
you
all
for
for
the
hard
work.