►
From YouTube: Weekly Meeting 08-05-2019
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
Okay,
okay,
so
reviewing
that
12.3
milestone
the
first
two
issues
on
their
den
and
I
added
the
deliverable
tags
last
Friday,
and
these
are
two
issues
that
came
from
the
hacker
one
report.
I
won't
go
into
details
since
we're
recording
this
now.
But
if
you
haven't
taken
a
look
at
those
issues,
you
take
a
look
at
them.
They
live
the
top
two
issues
in
the
twelve
three
milestone
during.
A
The
next
one's,
the
actually
I
think
Gigi's
been
making
really
good
progress
on
this
one
that
delete
only
selected
tags
from
the
container
industry,
and
this
one
is
about
there's
this
doctor
limitation,
that
when
you
delete
one
tag,
it
deletes
all
tags
with
the
same
image,
ID
or
the
same
tag
ID
rather,
and
it's
one
of
our
most
popular
issues.
People
been
complaining
about
it
for
a
long
time.
A
Nick
mates
in
front
end
changes
to
help
accommodate
that,
but
really
didn't
solve
the
core
problem,
and
this
should
address
the
core
problem
and
the
implementation
that
we're
working
through
is
by
creating
when
a
user
tries
to
delete
a
tag
that
will
create
a
dummy
tag,
that'll
move
it
and
then
it'll
delete
the
moved.
One
I
believe
is
how
it's
working.
So
it's
a
little
little
trickery
that
we
we
took
on
from
how
our
users
were
doing
this
and
Gigi
just
coded
it
into
how
how
it
works
on
our
platform.
A
A
So
we
went
through
and
we've
identified.
What
are
all
the
different
data
points
for
the
container
registry
sort
of
generic
get
lab
data
that
we
might
want
to
surface
to
the
user
and
then
as
well
as
we
thought
through
the
at
least
briefly
NPM,
maven
and
Konan.
In
terms
of
what
metadata
we
would
want
to
show-
and
so
now
it's
like
an
information
architecture
question:
how
do
we
display
that
data?
How
does
it
make
sense,
and
how
do
we
turn
that
into
a
user
interface?
A
That's
useful
for
our
users,
so
I
think
this
one
and
I
are
almost
done
on
this.
It's
I
think
we
just
need
to
keep
filling
some
of
the
definitions
and
then
share
it
with
the
team
and
get
some
feedback
on
what
everybody
thinks
and
then
probably
go
through
and
see
what's
available,
what's
not
available,
you
know
and
what
what
engineering
works
and
needs
to
happen
to
make
that
data
available.
A
So
it's
kind
of
exciting,
it's
probably
a
big
can
of
worms
when
we
actually
start
to
dig
into
it,
but
I'm
excited
to
think
that
we
that
we
could
start
to
surface
this
information
for
the
user
and
Ian.
You
had
some
ideas
about
doing
like
a
card
sword
and
some
other
UX
research
type
of
things
right,
yeah,.
C
A
Cool
cool,
the
next
one's
kind
of
nebulous
and
Nick
actually
opened
this
issue.
It's
improved
the
container
registry
table
UI
and
he
had
some
ideas
in
there
and
I
kind
of
clout
muddied
the
issue
by
throwing
in
some
competitors
UI's
and
things
we
would
like
to
see
and
that's
where
the
idea
for
the
data
model
came
I'd
like
to
see
what
we
can
talk
to
Dan
about
this
on
Friday
we'd
like
to
see.
A
Maybe
we
can
make
some
improvement
to
the
you
container
UI
for
this
iteration,
but
we're
not
I,
don't
know,
I
think
we
need
to
keep
iterating
on
this
issue,
and
maybe
we
could
I
had
Nick
had
some
really
good
ideas
and-
and
maybe
we
could
just
implement
a
couple
of
them
and
that
would
still
be
useful
and
if
we
could
have
it
in
a
better
state,
it
doesn't
have
to
be
Oh.
Tim
said
you
know,
look
at
code,
freshers,
UI
element.
All
of
that
on
day.
A
C
A
A
B
B
A
A
Cool
the
next
one
on
there
is
for
the
data
collection,
so
we
found
out
in
12:2.
We
had
this
issue
originally
scheduled
and
turns
out
that
we
are
collecting
pageviews
and
I.
Do
I
talked
to
the
data
team
and
I
got
some
a
self-help
guide
on
how
to
put
together
the
periscope
dashboard
for
some
data,
so
I'm
gonna
work
on
that,
but
separately
it
it's
time
for
us
to
start
tracking
some
of
the
other
data.
So
we
can
make
more
informed
decisions.
So
I
wanted
to
pull
this
one
and
prioritize
it
for
the
team.
A
It
seems
again
like
a
another
messy
one,
because
there's
a
bunch
of
metrics
in
there.
We
could
work
through
the
p2s
in
this
case
for
this
iteration,
which
are
mainly
let's
see
and
then
looked
at
this
issue
today,
like
number
of
build
or
push
events,
if
possible,
number
of
images
deleted
number
of
garbage
collection
commands
executed.
If
we're
able
to
tell
those
so
yeah
I
would
start
with
the
P
twos
for
the
container
registry
ID
leak,
since
that's
where
we're
making
the
biggest
investment
over
the
next
couple
of
months.
A
Okay,
next
one
is
a
follow-up
to
an
issue
we
had
in
in
12-2,
which
was
the
group,
an
API
endpoint,
to
list
images
and
tags
at
the
group
level.
This
next
one
is
for
listing
the
packages
at
a
group
level,
so
our
Maybin
and
NPM
packages
at
the
group
level
I
like
this
one,
because
I
think
it
will
enable
us
to
think
through
what
does
a
group
level
user
interface
look
like
once
we
know
once
we
think
about.
A
A
B
So
the
on
that
one,
it
seems
that
there
might
not
really
be
much
of
anything
to
do,
because
there
really
isn't
any
heavy
earnest
usage
of
this
moment
and
anywhere
in
the
package
code.
So
we
might
be
kind
of
good
at
this
point
where
you
know,
we've
looked
around
and
haven't
found
anything
so
they
might
be
able
to
call
it
good
for
now.
Oh
sweet.
A
B
B
A
A
Does
that
make
sense
for
preferred
publicly
to
only
talk
about
the
primary
issues
that
word
we're
delivering
okay
separately
either
week
I,
just
you
and
I
talked
to
Nazz
customer
on
on
a
Friday,
and
we
were
talking
to
them
about
how
they're
running
garbage
collection
and
removing
tags,
and
it's
funny
because
I've
been
talking
to
users
for
over
a
month
about
this
now
and
going
back
and
forth
with
you.
There's
an
issue
and
I.
A
Think
I've
always
been
anchor
bias
to
this
idea
that
we
can't
do
garbage
collection
and
we
can't
do
retention
policies
because
gitlab
itself
can't
do
it
because
we
can't
have
downtime,
but
in
fact
our
customers
can
have
downtime
and
they're
they're
doing
this
now
they're
running
garbage
collection,
now
they're
running
the
API
to
on
to
programmatically
untag
images.
Is
there
anything
stopping
us
from
from
doing
a
retention
and
exploration
policy
and
automated
garbage
collection
for
self
manage
instances.
B
A
A
B
B
It's
mostly
just
kind
of
the
investigation
to
make
sure
that
that
idea
is
true
where
you
know,
if
you
are
on
a
self-managed
instance,
and
you
have
the
ability
to
have
some
down
time
and
deal
with
these
things.
Would
the
policies
and
all
those
changes
work
as
expected,
or
is
there
something
we're
missing.
A
Okay,
yeah,
that
sounds
cool
and
then
I
think
we
have
a
lot
to
go
off
of
from
our
user
research
to
of
what
the
how
people
would
want
to
do.
Garbage
automated
garbage
collection
and
and
and
expiration
and
retention
policy,
so
I
think
we
could
prior
if
it
works
with.
That
would
be
an
awesome
feature
to
prioritize
him
in
12
for.