►
From YouTube: 2020-01-21 Crossplane Community 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
Wasn't
it
a
thumbs
up
from
in
alright,
so
yeah,
so
that
we're
not
on
Friday
one
of
the
interesting
things
to
note
there
there
were
full
release.
Notes
are
available
on
github,
so
you
can
see
every
all
the
features
and
there,
those
included
in
there.
But
one
thing
of
interest
was
that
template
stacks
experience,
for
that
is
now
available.
It's
an
experimental
still,
it's
an
early
early
effort
on
it,
but
you
can
now
try
out
templates
tax
with
the
0.7
release.
A
One
question
that
I
thought
of
as
I
was
adding
things
to
the
agenda
doc
here
today
is
that
we
probably
need
to
do
a
release
or
also
push
at
least
build
and
publish
the
images
for
the
minimal
stacks
that
we
have
as
well.
We
have
an
agenda
item
later
on
about
how
we
want
to
do
that,
but
I
think
that's
something
that
we
probably
want
to
do
so
that
they
are
available
and
semantically
versioned
as
well.
A
So
let
me
know
if
there's
any,
if
anybody
has
big
objections
to
that,
but
we
could
talk
in
more
detail
later
on
in
the
agenda
and
then
we
also
want
to
do
a
blog
post.
That
was
something
that
was
not
part
of
the
Friday
release
there
dan.
Do
you
wanna,
like
Dan
Ramona.
You
think
you
all
have
a
latest
update
on
that
yeah.
A
A
Right
cool
episode,
0.7
I
I,
feel
like
every
month.
The
this
process
is,
you
know,
fairly
consistent,
it's
not
a
high
cost
to
it
as
well,
so
we
have
regular
cadence
of
features
and
fixes
going
out
with
a
you
know,
scheduled
expected
mid
month,
type
of
release.
So
that's
been
really
effective,
I
think
and
so
the
next
the
next
release
around
0.8
would
be
expected
to
be
the
you
know,
mid-february
timeframe,
and
you
know
the
week
of
february:
17th
is
where
we
can
target
there.
We
can
talk
more
about.
A
If
we're
you
know
a
code
freeze-
or
you
know
specific
dates
around
that,
but
the
week
of
monday's,
the
February
17th
looks,
looks
pretty
good
from
a
regular
cadence
perspective.
One
thing
we
might
want
to
consider
also
is
that
we
have
been
doing
releases
on
Fridays,
but
I
think
that's
not
always
the
best
idea.
A
You
know
obviously
there's
some
well-known
wisdom
around
not
pushing
to
production
on
a
Friday
afternoon
in
Cross
Plains,
not
really
running
a
specific
live
service,
or
so
it's
a
bit
of
a
different
scenario,
but
you
know
having
having
a
release
running
on
Friday
evening.
Well,
you
know
some
people
are
trying
to
get
to
other
engagements
and
such
is
it's
not
always
ideal
either.
So
we've
considered
doing
it
like
on
a
Wednesday
or
earlier
in
the
week
than
doing
it
on
a
Friday
I.
Think
the
roadmap
now
that
we're
going
into
0.8
here
needs
updating.
A
So
if
we,
you
know,
I
think
that
we
had
it
updated
through
0.7,
and
then
we
have
got
a
lot
of
nice
releases
here,
a
while
yeah.
We
have
a
whole
bunch
of
items
here
that
are
in
the
roadmap
order
or
not
scheduled
or
assigned
to
particular
release.
So
we
need
to
follow
up
and
update
the
roadmap
for
specific
items
that
we're
targeting
and
planning
in
the
0.8
release.
Let's
take
an
action
item
for
that.
It.
B
Had
one
that
was
down
there,
it's
in
PR,
which
I
guess
it's
actually
an
issue.
I've
I
opened
it
this
morning,
Steven
brought
it
up
that
the
RC
tag
doesn't
get
picked
up
on
master
for
either
cross
playing
are
the
stacks
until
we
merge
another
commit
and
that's
part
because
the
promotion
happens
on
master
is
part
of
the
build
pipeline.
So
just
adding
a
new
tag
doesn't
actually
result
in
promotion
of
that.
A
So
just
running
running
the
master
pipeline,
build
normal
build
pipeline.
Would
with
me
near
that
immediately
right,
yeah,
that's
a
good
point.
We
should
just
probably
do
that
so
that
that
could
be
part
of
the
release
process
if
you
want
to
to
actually
tag
and
master
to
go
ahead
and
just
run
the
build
pipeline
and
have
it
out
there
yeah.
B
A
B
A
So
we
have
a
set
of
stacks
that
you
know
easy
stack,
minimal
stack
that
make
it's
very
simple:
to
set
up
a
particular
environment
with
the
right
networking
and
cloud
artifacts
there
to
be
able
to
easily
configure
applications
within
that
configured
infrastructure
and
they
have
a
home
in
the
crossplane
org
that
the
stacks
and
those
repos
there,
but
they
do
not
have
you
know,
build
pipelines
and
they're,
not
part
of
the
release
process.
So
it
seems
like
similarly
to
the
other,
the
infrastructure
stack.
A
So
we
have
the
cloud
provider
of
structure
stacks,
such
as
GC
and
Azure
and
AWS
that
it
would
make
sense
to
have
these
stacks
here.
Be
part
of
the
you
know:
version
release,
process
with
the
automatic
build
pipeline
and
all
that
sort
of
stuff.
So
they
just
kind
of
go,
go
with
the
flow
of
the
other
releases.
A
A
A
It
is
Susskind
on
the
call,
it
doesn't,
doesn't
look
like
it
but
yeah
this.
So
this
is
an
item
that
I
wanted
to
bring
up
and
I
know
Daniel,
it's
related
to
the
work
he
was
doing
and
he
had
F
and
opinions
on
it.
So
we
I
think
we
okay
open
an
issue
for
is
well,
so
perhaps
the
discussion
can
take
place
there
instead,
but
just
to
bring
up
the
issue
is
that
you
know
we
have
what
we
consider
a
definition
of
done
in
the
cross.
A
Plain
organization
where
you
know
features
you
know,
need
to
have
a
certain
level
of
design
documents
or
architecture.
Thinking
needs
a
certain
amount
of
test
coverage
need
to
have
user
guides
for
how
to
use
them
and
such
so
we've
defined
that
process.
But
we
don't
really
have
a
good
process
in
place
for
features
or
work
that
we
consider
more
experimental.
A
That
we're
not
quite
sure
you
know
it's
like
a
really
a
preview
type
of
thing,
we're
not
quite
sure
what
the
feature
is
going
to
look
like
long
term
or
what
the
lifecycle
of
it
is
going
to
be,
and
we
want
to
get
it
out
to
be
able
to
get
some
some
people
using
it
in
the
community
getting
some
feedback
from
folks-
and
you
know
it's-
it's
pasta
quite
possible
that
features
that
are
considered
more
experimental.
We
want
to
have
maybe
a
different
process
around
them,
like
maybe
test
coverage
doesn't
make
as
much
sense.
A
If
we're
not
even
sure
if
the
feature
is
going
to
live
in
the
same
state
that
it
does,
you
know
you
might
want
to
hide
them
behind
feature
flags
so
that
they're
not
enabled
by
default
through
people
as
well.
There
needs
to
be
more
opt-in,
so
we
need
to
kind
of
think
through
that-
and
you
know,
figure
out
kind
of
a
framework
for
that
or
some
process
around
what
we
want
to
do
with
more
experimental
features
or,
if
that's
something
that
we
really
want
to.
Even
you
know
accept
in
the
first
place.
A
C
C
B
Yeah,
but
to
do
to
update
the
developer
guide,
pretty
bringing
that
up
to
me
Marcus
another
thing
that
might
be
worth
keeping
in
mind.
That
is
also
in
line
with
this
kind
of
point
we're
talking
about
is
for
this
change
in
cross
flow
in
runtime.
We
opted
to
just
go
ahead
and
kind
of
like
switch
over
to
the
new
package
structure,
since
we
were
doing
a
new
release
that
you
know
indicated
breaking
changes
and
that
sort
of
thing
in
the
future.
B
It
might
be
worth
especially
at
a
more
mature
phase
of
the
project
and
of
the
library.
Considering
keeping
you
know
the
old
structure
as
a
deprecated
way
to
do
it,
but
still
allowing
users
to
upgrade
and
continue
usage,
especially
since
there
were
no
feature
changes
as
part
of
the
reorganization.
So
that
might
be
something
to
just
consider
when
we're
writing
up
like
a
policy
for
feature
flagging
and
deprecations,
and
that
sort
of
thing.
A
Yeah
and
then
I
agree
to
that
they're,
not
quite
the
same
issue.
No
experimental
features
like
new
things
that
are
coming
out,
that
you
know
that
we're
not
quite
sure
the
future
future
of
them
is
you
know
the
contractor
on
those
is?
Is
you
know
something
a
little
bit
different
than
existing
things?
We
have
like
across
main
runtime
features
where
we
want
to
make
sure
that
the
updates
and
changes
there
are
done
in
a
reasonable.
C
A
C
A
B
And
we
probably
won't
reach
any
sort
of
conclusion
here,
but
just
something
to
have
on
people's
mind.
So
we've
talked
a
little
bit
and
some
past
community
meetings
about
implementing
integration
test.
So
we've
done
that
we're
now
actually
running
them
with
stack
GCP
on
a
periodic
schedule
and
I've
been
tracking
those,
since
it
was
merged
into
master
and
they've,
been
pretty
successful.
There
hasn't
been
a
lot
of
churn
on
the
code
while
they've
been
running,
but
it
is
nice
just
to
see
every
night
that
were
able
to
provision
instruction.
B
It
comes
up
as
expected
and
that
sort
of
thing-
and
it's
done
in
an
automated
fashion,
so
I'm
really
liking
that
we
have
this
feature
set
I'd
like
to
continue
roll
out
to
other
stacks
and
potentially
some
kind
of
like
full
end-to-end
stuff
involving
multiple
stacks
just
to
continue
to
test
our
the
quality
of
our
software
as
it
evolves,
but
definitely
would
love
some
input
in
terms
of
how
frequently
people
think
these
test
needs
to
be
run
is
a
nightly
basis.
Good.
Is
that
too
frequent?
What
kind
of
things
need
to
be
included
as
test?
B
A
B
Instead,
GCP
so
they've
run
four
times
since
being
merged
to
master
very
consistent,
all
taking
about
sixteen
and
a
half
minutes
each
time,
they're,
spinning
up
new
clusters,
making
sure
things
come
available
and
then
and
that
sort
of
thing,
and
they
were
all
green
yep,
all
green,
no
flakes
or
failure.
Nice
and
very
good.
C
One
thing
it
was
an
act
and
I've
seen
the
pattern
were
when
you're
opening
your
pull
request.
It
come
with
automatically
detects
if
it's
touching
some
patterns
or
some
pieces
of
code
and
then
base,
and
that
triggers
puts
the
integration
test,
one
of
the
conditions
to
pass
the
BlueCross.
So
we
could
look
into
that.
But
I
am
NOT.
You
know,
on
top
of
my
head,
I,
don't
know
anything
any
patterning
golang,
so
probably
some
search
which
results,
I
think
yeah.
B
Yeah
I
know
right
now
like
with
like
prowl
or
something
like
that.
They
have
the
bot
that
basically
checks
different
directories.
If
there's
code
changes
in
them
that
basically
run
tests.
Based
on
that
at
this
point
with
like
the
rapid
evolvement
of
the
project
right
now,
I
think
it
might
be
difficult
to
implement
that
sort
of
thing.
I
will
say
we
are
able
to
run
integration
tests
on
any
specific
branches
manually.
B
So
you
know,
since
we
do
have
a
pretty
narrow
scope
of
like
reviewers
and
approvers
right
now,
then
we
can
do
that
when
we
feel
it's
necessary,
but
I
do
think
long
term,
potentially
implementing
some
sort
of
feature
set
into
the
bot
that
currently
runs.
Our
tests
might
be
useful
in
terms
of
automatically
triggering
integration
tests
right
now,
though,
it
running
it
on
a
nightly
basis,
pretty
much
catches,
everything
because
it's
it's
unusual-
that
we
have
more
than
a
few
PRS
on
a
given
repo
and
at
night.
A
Yeah,
that's
that's
one
approach
to
is
that
you
know,
while
we're
kind
of
vetting
that
process
it's
a
new
process,
you
know
and
seeing
like
what
the
stability
of
it
is,
like
you
know,
the
consistency
of
it.
It's
you
know
having
it
like
in
a
time-based
periodic
approach,
is,
is
really
pretty
reasonable
because
we
can
just
kind
of
keep
an
eye
on
and
see
how
it's
going.
A
Alright,
the
host
aware
stack
manager
designed
from
Sun
he
had
provided
the
design
right
up
and
the
implementation
for
it
is
well
I,
know
Marcus,
we
got
to
take
a
look
at
it.
I
think
maybe
a
couple
others,
but
this
is
something
that
we
haven't
merged
yet
and
it
had
been,
you
know
had
been
needing
a
little
bit
more
attention
recently.
So
that's
something!
That's
still.
A
You
know
it's
on
my
list
to
look
at
soon,
but
so
what
do
you
feel
there
any
other
eyes
that
had
gotten
on
it
and
you
know
had
added
approvals
to
it
and
such
so.
C
A
A
A
A
C
A
A
A
A
A
Okay,
that's
I'll,
just
put
my
name
on
it
right
now,
because
I
know
you
got
some
of
the
things
to
do
today.
Dan
I
might
find
some
time
while
in
other
other
means.