►
From YouTube: Implementation Sync: 2021-06-10
Description
Meeting notes: https://bit.ly/38pal2Z
B
B
I
guess
let's
get
started
with
status
updates.
I
can
share
a
few
quick
updates.
I
just
merged
a
couple
pr's
to
the
life
cycle,
to
add
some
of
the
features
that
we
want
for
the
next
release.
So
that's
good
news.
We
have
a
couple
other
prs
that
are
in
flight.
I
think
jesse's
validate
stack
is
in
need
of
a
re-review
and
asset
packages.
B
C
A
I
did
merge
one
subteam
specific
rfc,
the
one
regarding
lifecycle
documentation
into
the
rfc's
repo.
I
created
an
issue
sort
of
suggesting
that
we
document
that
process
somewhere
in
the
lifecycle
repo
like
release
md,
even
though
that
wasn't
called
out
specifically
in
the
rfc.
If
we
wanted
to
sort
of
like
capture
that
outcome
differently,.
B
B
B
B
I
don't
think
we
can
count
on
that
exactly
so.
I
guess
it
might
be
worth
pinging
marty.
I
think
just
to
ask
what
it,
what
what
he's
been
able
to
to
do
with
it.
If
he's
had
a
chance
to
look
at
it
really,
maybe
when
we're-
I
don't
know,
maybe
six
two
and
ten
hey,
maybe
it's
okay
to
to
ask
at
this
point.
B
B
C
C
C
B
Yeah
like
this,
so
this
is
like
pr
ready
for
review.
This
is
like
pr
almost
finished.
Pr
is
up.
I
don't
think
we
can
do
anything
about
this.
I
made
a
comment
about
it,
but,
like
probably
not
a
lot
of
work
here,
this
is
the
image
util
pr
is
already
up.
So
it's,
like
you
know
a
little
bit
of
extra
work
to
add
it
to
the
life
cycle,
but
most
of
the
work
has
been
done
already,
so
I
feel
good.
I
feel
like
maybe
even
this
week
for
these
prs.
C
B
C
B
So
I
think
what
we
establish
is
that
everything
that's
in
phase
one
will
not
be
thrown
away
right.
If
we
don't
do
stack
packs,
it's
all
still
has
value
right.
So
this
is
something
that
will
need
to
and-
and
my
understanding
is,
there's
no
there's-
no
guarantee
that
we're
changing
direction
here
right,
but
it's
just
an
idea
has
been
put
forth
right,
but
it
it
should
be.
I
guess
clarified
before
we
start
working
on
phase
two,
because
that's
the
one
that's
really
going
to
right.
It's
gonna
matter
right.
B
A
A
Is
it
still
still
the
right
path
to
move
forward
on,
and
I
don't
know
exactly
what
the
answer
to
that
is.
But
I
think
it
is
a
good
question,
given
that
you
know,
despite
all
the
work
we've
put
into
it,
there's
still
a
lot
of
work
between
here
and
done,
especially
considering
all
the
features
that
were
laid
out
in
that
rfc
and
before
we,
you
know
dedicate
the
next
several
months
to
continuing
down
that
path.
I
think
we
all
need
to
feel
that
is
the
best
way
forward.
C
B
Oh,
it's
just
that
you
know
thus
far
right.
Obviously
this
has
been
a
conversation
that
was
not
happening
in
the
open,
but
there
is
you
know,
efforts
underway
to
bring
that
out
into
a
more
public
forum
so
that
people
can
weigh
in
so
just
I
guess
you
know
stay
tuned,
but
that's
something
that
I
think
is
very
important
to
happen
right.
A
I
don't
think
that
there
was
a
reason
like
that
it
was
purposefully.
You
know
kept
secret.
I
think
some
of
these
conversations
happened
when
the
rfc
was
originally
approved,
so
they're
definitely
conversations
we're
willing
to
have
in
the
open
the
reason
this
conversation
started
again
in
a
forum
that
wasn't
public
was
because
one
of
the
motivations
for
not
like
taking
a
recording
of
the
summit
and
posting
it
was
to
create
a
safe
space
to
ask
hard
questions
or
have
conflict
if
that
needed
to
happen,
which
can
be
hard
when
something's
public.
C
B
So
I
guess
we'll
all
just
kind
of
look
forward
to
seeing
more
conversations
happening
and
hopefully
also
you
know,
stepping
in
to
add
our
opinions,
because
I
think
that
is
my
impression
is
that
the
the
thoughts
and
feelings
of
the
implementation
team
are
very
important
to
this
question.
Right
and-
and
you
know
our
like
our
knowledge
of
what
it
will
be
like
to
implement
this
stuff
is
going
to
be
something.
That's
you
know
factors
in
to
how
we
consider
each
option
or
whatever
options
there
are.
A
A
A
B
B
B
B
I
think
I
saw
a
note
about
it
in
in
a
meeting
doc.
I'm
just
wondering
like.
Should
we
close
this
like?
Is
there
anything
else
we
can
do?
Just
I
guess
context
is
not
super
obvious.
This
is
like
when
you're
configuring,
your
secret
for
your
registry,
and
you
don't
put
a
trailing
slash
behind
v1
you
might.
I
think
this
is
actually
you
will.
I
had
some
weirdness
in
my
previous
environment,
but
you
will
get
like
you
know.
B
You
will
not
be
able
to
authenticate
with
the
registry
because
we're
doing
not
we
ggcr
is
doing
an
exact
string
match
here
right,
and
this
is
like
ggcr
is
the
one
that's
reading
the
docker,
config
or
whatever
it
is
like.
You
know
I'm
a
little
bit
fuzzy
on
exactly
how
this
happens
in
a
kubernetes
environment,
but,
like
I,
my
takeaway
from
reading
the
code
was
that
we
don't
have
much
control
over
how
this
happens.
B
C
A
Is
about
like
when
you're
matching
right
in
your
in
the
keychain
when
the
keychain
is
saying
like?
Is
this
secret
for
this
registry?
It
seems
like
the
absence
or
presence
of
a
trailing.
Slash
should
not
be
fundamental
to
the
answer
there,
because
in
a
url,
the
absence
or
presence
of
the
trailing
slash
doesn't
matter
right
in
general.
A
Is
that
you're,
basically
creating
this
docker
config
json,
the
user
sort
of
taking
responsibility
for
the
contents
of
it,
whereas
when
you're
using
pac
or
something
on
the
local
file
system,
you're
doing
a
docker,
login
and
then
dockers
taking
responsibility
for
it.
So
when
docker
writes
it,
it
can
be
perfectly
consistent
about
whether
it
includes
that
trailing
slash
or
not.
A
And
then,
when
docker
reads
that
value,
it
doesn't
have
to
ask
the
question,
because
it
knows
that
it
wrote
it.
It
generally
is
the
one
that
wrote
the
docker
config
json
file,
but
in
a
situation
where
users
are
writing
it.
I
think
it
would
make
sense
to
be
a
bit
more
flexible
because
it
bothers
everyone.
All.
The
time.
B
A
B
B
B
C
B
That
was
another
one
that
scared
me
like.
I
I
we
were
talking
about
complexity
right
like
sorry,
if
I
just
click
through
to
this
one
nope,
there's
that
there's
this
thing
right,
like
the
nope,
not
this
one,
this
one
like
the
matrix
of
compatibility
when
you
have
new
and
old,
build
packs,
and
also
new
and
old
platforms,
and
just
making
them
all
work
together.
B
Yeah,
I
think
we
sort
of
said,
like
the
benefits
are
worth
it,
but
the
road
to
this
like
end
goal
of
having
you
know,
a
perfectly
organized
file
system
is
going
to
be
bumpy.
So
that's
all
just
I
I
don't
know.
If
there's
anything,
we
can
do
about
it
right
now
or
ever,
but
it's
like
how
do
we
manage
all
this
complexity
that
we
already
have
in
the
life
cycle
like
it's,
it's
already
hard
to
do.
B
A
C
B
I
remember
we
put
that
and
the
only
real
downside
which
you
know
it
is
a
downside
but
remember
we
said
like
that,
will
exclude
certain
layer
names
right.
You
will
never
be
able
to
have
a
layer
named
like
cnb
or
config
or
whatever
we
decide
to
call
it,
and
that
is
a
downside.
But
I
do
wonder
if
it's
a
downside
worth
taking
in
favor
of
something.