►
From YouTube: Platform Sync: 2020-11-04
Description
* Standing: Status Updates
* Standing: Release Planning
* pack search on registries
* Extending pack-orb to add additional commands
* Moving Tekton template maintenance to lifecycle
* New heroku builder (to be trusted)
A
A
D
A
Moving
on
to
release
planning
this,
we
definitely
have
something
around
david,
as
I
guess,
one
of
the
things
I
want
to
mention
is
david
will
be
doing
the
release
management
for
this
upcoming
release
and
so
I'll.
Let
you
david
speak
to
most
of
that.
B
Sure
we
are
scheduled
to
release
version
pack
version
0015
in
a
week
from
today.
So
I'm
going
to
cut
the
release
branch,
there's
still
kind,
there's
one
pr
that
I'm
waiting
to
talk
through.
But
after
that,
I'm
happy
to
create
the
release
branch
and
then
we're
gonna.
Do
that
with
an
rc
to
start
creating
rc
releases
as
well.
A
Cool
I
did
hear
through
the
grapevine
that
windows
is
anticipating
this
release,
pretty
heavily.
It
seems
like
they're,
starting
to
do
a
little
bit
of
a
promotion
around
it
and
some
talks
with
different
user
groups
about
windows,
support,
and
so
they'd
like
to
see
this.
You
know
integration
with
pac
available
for
people
to
be
able
to
download
it
as
soon
as
possible.
So
I
think
that
coordination
is
going
pretty
well
and
we
should
be
able
to.
You
know,
satisfy
their
request.
A
So
I
mean
this:
the
people
that
we're
talking
about
right
now
are
via
more
specific
team
members
but
yeah.
Overall,
it's
the
people
involved
in
the
windows,
cnb
project.
B
B
E
Okay
sounds
good.
I
don't
really
have
anything
to
add
either
just
been
some
public
conversations
going
on
in
slack
with
regards
to
the
dashboard
for
doing
the
registry
search
stuff,
so
you
can
all
follow
along
and
see
what
has
been
discussed
there.
I
know
javier
has
threw
up
some
pretty
cool
mocks
of
what
it
could
look
like.
So
that's
exciting
to
kind
of
see
it
come
to
life
a
little
more
and
as
we
visualize
where
we
might
head
so,
but
yeah,
nothing
really
new.
Besides
that,
for
me,
yeah.
A
Partially
on
topic,
so
I
know
one
of
the
discussions
we
had
just
you
know
before
this
meeting
was
about
some
of
the
search
functionality
right
and
how
that's
going
to
be
portrayed
on
the
website
and
how
we
may
or
may
not
be
able
to
use
the
same
search
functionality
within
pack.
I
know
that
you
know.
Pack,
you
know
search
is
something
that
you
intend
to
kind
of
introduce
or
propose
in
the
future,
so
that
may
be
worth
having
a
discussion
around
in
the
near
future.
A
E
E
So,
like
the
question
is
like
ideally,
it
would
be
nice
to
leverage
the
the
same
underlying
logic
that
performs
the
search
you
know
on
both
pac
and
on
the
api
itself.
Pac
is
go
right.
Now
the
the
api
is
written
in
node
I
mean
we
could
rewrite
it
and
go
and
maybe
share
the
library
between
the
two
but
yeah.
It's
just
something
we're
gonna
we're
thinking
about.
We
brought
it
up
this
morning
when
we
were
talking
in
a
one-on-one.
E
A
Yeah,
I
think,
just
to
verbalize
what
we've
already
talked
about
a
little
bit
more.
I
think
the
idea
is
a
lot
like
homebrew
and
app
get
right
where
I
think
we,
since
we
already
have
the
cache
locally
instead
of
making
api
calls
to
any
you
know
central
service
and
then
having
to
formalize
that
interface,
it
would
be
best
to
just
kind
of
use
the
local
cache
and
then
because
of
that,
I
think.
E
E
I
was
wondering
if,
if,
if
you've
been
giving
given
or
hearing
about
any
other
requests
for
that
kind
of
a
feature
or
if
it's
just
kind
of
a
random
question
that
came
out
of
left
field
from
this
particular
person,.
B
E
That's
good
to
know
yeah,
I
think
I
mentioned.
I
don't
know
if
you
if
you
did
create
an
issue
or
not.
If
not
don't
worry
about
it,
that's
fine,
but
I
was
just
saying
like
let's:
let's
keep
track
of
that
work
and
if
it
makes
sense
we
should
go
ahead
and
do
it.
I
think
that's
that
seems
reasonable
to
me.
B
Yeah
looks
like
she
did,
make
a
issue
for
it.
She
posted
it
in
that
thread.
E
Yeah,
as
I
mentioned
like,
we,
don't
really
have
any
hard
like.
Oh,
this
is
a
road
map
for
where
we're
taking
it.
So
it's
pretty
much
just
being
community
driven.
So
if
anyone
has
any
feedback
or
thoughts
just
express
it
and
if
it
makes
sense
we'll
go
ahead
and
do
it.
C
A
And
if
I'm
not
mistaken,
heroku
is
using
the
pac
orb
right,
like
dog
fooding
it
in
a
sense.
D
Yeah
I
saw
the
that
reminds
me.
I
forgot
tecton
is
kind
of
this
group
right.
I
think
the
the
tecton
examples
I
think
someone
flagged
it
as
problematic
in
the
latest
life
cycle
releases
right
because
of
some
execution
like
user
stuff.
I
think
the
tech
top
examples
don't
run
as
the
cmd
user,
like
setting
the
security
context,
so
I
don't
know
where
we
stand
with
that,
like
you
said
we're
not
dog
fitting
that
either
so
I'm
on
the
heroku
side.
A
A
In
part
of
me,
like
wonders
whether
more
of
the
knowledge
is
actually
on
the
implementations,
team
side
versus
the
platform
team
right
like,
and
I
don't
necessarily
know
where
the
dividing
line
is
but
like
yeah
like
the
tecton,
is
a
platform
but
again
because
a
lot
of
these
issues
that
have
come
up
are
like
very
ingrained
with
the
implementation,
implementation
right
and
so
part
of
me
is
like
well.
If
you
have
the
knowledge,
why
do
I
have
to
try
to
figure
this
out
right
and.
C
A
D
D
Yeah,
I
think
it'll
become
more
true
with
like
extending
like
stack
packs
and
stuff
right,
like
that's,
gonna,
be
very
specific
and
pretty
impossible
to
do
with
the
current
template.
I
think
around
yeah
techton,
for
instance,
but
maybe.
D
If
both,
if
both
the
cato
and
heroku
start
adding
stack
packs,
then
that
will
quickly
devalue
that
template,
we
add
those
to
our
default
builders,
for
whatever
reason
for
like
after
packages
right.
A
So
so
probably
nothing
actionable
right
now,
but
something
to
keep
our
ears
to
the
ground
about
and
be
cognizant
of,
as
we
add
more
complexity
to
the
life
cycle
usage,
I
know
from
what
I
hear
like
natalie
right,
like
I
think,
she's
expressed
some
concerns
about
not
necessarily
being
in
line
with
what
they,
the
users
are
close
enough
to
the
users
right.
I
guess
you
could
say,
and
I
wonder
if
you
know
as
by
example,
if
supporting
tecton,
that
would
bring
them
a
little
bit
closer
to
like
you
know,.
B
B
One
like
we
may
want
to
consider.
I
have
a
pr
adding
some
circle,
ci
and
techton
like
stuff
to
the
samples
repo,
and
so
I
would.
I
would
ask
for
someone
to
set
up
circle
ci,
so
we
can
at
least
dog
food
it
and
you
know
our
samples
repo
and
show
that
it
showed
that
it
works.
We
may
also
want
to
have
an
active
like
just
you
know,
test
pipeline
of
tacton
as
well.
B
A
B
Already
use
that
for
our
for
the
our
demo,
like
we
made
a
bunch
of
clusters
there,
I
just
I
don't
know
long
term
how
expensing
works
in
terms
of
like
whether
it's
charged
by
the
long
running,
I
don't
know
how,
how
long
a
a
cluster
runs.
I
I
would
have
to
look
more
into
like
what
gke
pricing
is
like
yeah,
just.
B
A
I
do
okay,
this
might
be
less
ideal,
but
I
guess
another
thing
we
could
do
is
like
set
up
ephemeral,
tecton
instances
right.
So
I
know,
for
instance,
like
on
in
kind
or
using
kind.
You
could
set
up
a
or
even
what
what's
the
other
one,
the
kind
alternative.
C
A
I
don't
see
why
not
as
long
as
you
have
access
to
docker
container
d
on
a
linux
environment
worth
exploring,
I
mean
if
we
want
to
really
step
away
from
the
gke
right,
but
I
think
an
action
item
for
me
from
this
unless
you
want
to
take
it
on
david,
is
looking
up
how
that
who's
paying
for
it
and
how
much
we
have.
A
D
We'll
probably
have
a
minor
pr
coming
in
at
some
point
to
add:
heroku
build
packs
20
as
a
trusted
builder,
at
some
point
in
the
near
future,
we're
kind
of
working
on
getting
a
new
builder
off
of
bionic
onto
the
the
newest
one.
So
whenever
that
lands
just
so
that
no
one's
surprised
but
we'll
probably
want
to
have
the
trusted
one
just
so
that
we
don't
have
to
have
users
pay
for
the
cost
of
you
know,
building
and
pulling
down
a
life
cycle
and
everything.
B
Is
it
all
is,
are
you
going
to
deprecate
18
at
that
point.
D
We
yeah
we
just
follow
lts
roughly,
like
I
think.
For
instance,
hurricane
just
deprecated
builds
on
cedar
14
like
last
week.
Officially
so
like
you
know,
but
it's
a
little
different
with
docker,
because
you
can't
do
like
the
extended
stuff
from
images
right
that
people
build
themselves.
So
I
think
it
may
be
slightly
more
aggressive
for
18
than
it
was
for
cedar
14,
but
no.
E
E
D
B
Bass,
there's
they
they
point
to
three
different
stacks.
They
point
to
tiny,
which
is
like
a
very
small
based
off
of
digitalis.
Then
they
point
to
base
which
is
based
off
of
18,
and
then
they
point
to
full,
which
is
the
equivalent
of
like
the
cloud.
Foundry
stack,
cf
linux,
fs3.
D
A
D
E
A
She's
well,
she
was
in
the
spring
team
for
a
bit.
I
don't
know
if
she's
still
there,
but
her.