►
From YouTube: Learning Sync: 2021-07-09
Description
Meeting notes: https://bit.ly/38pal2Z
A
A
A
And
then
it
basically
became
this
like
conversation
about
newer
versus
older,
like
capabilities
of
slack,
because
I
think
you
know
they
have
this
like
like
new
way
that
you
could
share
workspaces
right,
yeah,
actually
embed
him
or
whatever,
and
I
think
that
was
more
or
less
what
we
were
aiming
to
do
to
like
keep
our
own
set
of
information.
And
you
know,
structure
for
everything,
but
but
ultimately,
that
kind
of
just
fell
apart
and
didn't
go
anywhere.
B
Because
I
know
the
the
ghetto
slack,
the
six
store
slack
as
well
as
the
cyclone
dx
slack.
They
all
have
it
they're
all
paid
you're,
saying.
A
I
guess
I
would
think,
like
I
know,
for
sure
the
pocato
one
it's
a
little
bit
easier,
because
it's
part
of
the
cff
and
vmware
or
pivotal
really
influenced
the
cff
right
and
they
have
a
lot
of
buy-in
power
there.
A
A
Dies
off
I
mean
it's
probably
worth
bringing
up
again
right
just
to
see
if
there's
anything,
we
want
to
do,
I'm
not
completely
against
moving
to
the
cncf
slack
myself,
just
because
there
are
a
lot
of
other
projects
that
are
there
and
I
feel
like
being
a
little
bit
closer
to
them
is
actually
like
a
good
thing
or
positive,
but
I
do
see
the
downsides
of
essentially
putting
everything
that
we've
gotten
so
far
and
slack
into
one
little
tiny
channel
right,
like
that
seems
very
painful.
A
B
B
B
I'm
trying
to
look
for
channels
learning.
Do
you
just
wanna?
Should
I
just
act,
cure
the
learning
team
channel
just
in
case
or.
A
A
B
B
A
Cool
yeah,
that
sounds
good,
so
we
want
to
go
ahead
and
add
those
to
our
thing.
B
A
A
Yeah
so
so
on
the
docs.
He
contributed
a
update
where
we
got
the
contributor
stocks
up
right,
so
I
went
ahead
and
merged
that
I
I
want
to
give
it
a
polish
look,
but
I
didn't
want
to
like
hold
off
depending
on
how
long
it
took
me
to
get
it
back
to
it,
because
he's
been
waiting
for
about
a
week
seven
days
or
something
like
that.
So
go
ahead
and
merge
that
in
I'm
thinking
about
adding
tabs
or
something
like
that,
so
that
it'll
be
more
condensed
but
yeah.
A
B
Apart
from
that,
I
recently
worked
on
the
reworked
concepts,
page
mostly
based
on
feedback.
I
received
internally
about
how
hard
it
was
to
navigate
concepts
related
to
bulk
banks.
So
the
current,
like
concept
starting
page
assumes
you
have
zero
knowledge
and
tries
to
introduce
terms
progressively,
rather
than
like
an
alphabetical
list
of
concepts
that
we
currently
have.
B
So
people
know
where
to
start
with
and
then
build
upon
that
the
only
thing
missing
there
is
platform,
which
is
the
last
part
that
I
need
to
add.
But
I
think
all
the
other
terms
have
mentioned
also
try
to
keep
like
concepts
detail
based
on
how
much
user
facing
they
are
so
you'll
see
like
build
packs,
have
a
huge
section.
B
Builders,
a
smaller
section
and
life
cycle
is
just
a
very
small
section
yeah
in
there,
because
when
people
first
see
buildbacks
like
when
they
see
life
cycle
and
stuff,
they
just
get
confused
like
what
it
is
and
most
of
the
times,
it's
just
a
magic
binary
that
they
need
to
put
in
a
certain
place,
and
sometimes
they
don't
even
have
to
do
that,
like
black
automatically.
Does
it
for
you.
A
No,
that
makes
sense.
I
did
look
over
this
and
obviously
it
looks
great
the
the
platform
being
integrated
into
that.
Are
there
any?
A
B
I'm
thinking
so
it's
like
there's
that
builder
diagram
right
where
the
build
process
is
just
three
arrows
right
now
I
was
hoping
like
put
the
platform
in
there
and
tell
people
that
hey.
This
is
what
like
the
platform's
responsible
for
taking
like
actually
executing
all
of
this
and
providing
the
builder
with
any
additional
resources
it
might
be
like
volumes
or
whatever
else.
A
So,
are
you
stealing
stealing
right?
Are
you
stealing
all
these
images
from
one
of
the
presentations
that
we
had?
Is
that.
B
Yeah,
like
I
think,
presentations
I
made
the
builder
one
myself:
okay,
because
the
the
other
one,
I
think,
is
the
one
that's
actually
used
in
the
builder
concepts
page,
which
felt
a
bit
complicated
to
me
because
it
described
the
builder
normal
and
like
buildings,
drawn
images
without
trying
to
like
tell
you
what
it
does
so,
instead
of
starting
from
the
left,
I
started
from
the
right
where
you
already
have
the
builder.
B
So
yeah,
like
the
others,
are
taken
from
like
our
past
presentations,
except
for
the
builder
one.
A
B
A
A
Yeah
no
yeah,
maybe
I'll
I'll,
find
it
and
send
it
to
you
just
to
see
whether
or
not
that's
something
you've
seen
you
looked
at
cool
yeah
this.
This
is
great.
I
guess,
is
there
a
way
that
you
could
think
of
that
we
could
track
the
the
impact
or
benefit
of
this
page.
A
B
B
A
A
But
yeah
because
to
me
that
seems
like
a
really
good
thing,
which
we
still
have
to
get
funnels
working
but
like,
if
you
think
of
the
quote-unquote
funnel
right
like
when
they
first
hop
into
our
website,
they
go
from
there
to
the
docs
right
then
from
the
docs.
You
want
them
to
either
check
out
the
videos,
but
also
go
into
the
concepts
page
and
then
the
concepts
page.
A
B
Yeah,
I
think
the
way
I
imagine
it
is
you
go
to
the
docs
page.
B
B
So
there's
this
your
first
app
sort
of
thing
we've
talked
about
this
in
the
past.
What
I
feel
the
impact
of
buildbacks
would
be
like
would
be
increased
manifold
if
we
switch
out
the
samples
to
one
of
the
actual
builders
like
if
people
could
take
that
and
run
it
against.
B
B
Okay,
so
we
discussed,
I
think
in
the
past,
having
like
vendor
builders
or
something
as
like,
something
like
people
could
try
out.
I
don't
know
what
came
of
that.
A
A
A
B
So
the
way
I
look
at
this,
as
is
hey,
you
know
at
least
other
documentation,
websites
where
you
have
like
you,
can
have
a
drop
down
and
it
renders
the
page
based
on
your
drop
down,
selection
or,
like
you,
have
tabs,
and
you
can
pick
between
the
tabs
or
whatever
you
want
be
a
possible
workflow
where,
like
there's
a
default,
drop
down
selection,
you
view
the
page
there.
If
you
want
to
switch
you
just
click
that
yeah.
A
So
I
get
the
implementation
part
of
it.
That's
not
really.
I
guess
my
my
question
or
concern
my
concern
is
the
the
natural
flow
right
like
of
this
you're
going
through
this
you're
like
breezing
through
it
right
everything
you've
done
so
far.
Right
like
it,
has
taken
very
little
thinking,
effort
right,
you're,
just
like
absorbing
all
this
and
then
all
of
a
sudden
you
get
hit
by
this
like
bam.
Here's
10
builders
select
one
and
you're
like
oh,
which
one
right
and.
A
See
which
one
do
we
want
to
buy
us
on?
No,
I'm
just
kidding,
I
heard
the
potato
ones
are
really
good.
A
And
then
you
say
you
know
next
step,
try
it
on
your
code,
but
let's
use
a
production,
ready,
builder
right
and
then
at
that
point
that's
you
know
so
at
least
you've
gotten
the
first
build
and
you
saw
how
easy
and
quick
it
was,
and
the
next
step
is
like
okay.
Now
you
actually
have
to
think
about
what
you
want
to
take
to
production
right.
So
you're
gonna
actually
run
this
on
your
app.
You
actually
want
to
look
at
the
offering
of
builders
and
build
packs.
B
B
So
in
the
past,
we've
discussed
I'm
getting
outside
a
bit
outside
of
learning
territory,
but
in
the
past
you
discussed
how,
even
before
a
builder
is
selected,
we
could
look
at
like
the
registry
like
if,
if
the
build
packs
had
information
about
specific
files
that
they're
interested
in
like
even
if
they
don't,
they
have
more
complicated,
detect
procedures
that
they
could
just
say,
like
I'm
interested
in
this
file
or
whatever,
when
you're
on
pack
build
if
it
could
just
scan
through
those
files
and
see
if
it's
present
in
your
project
and
then
like
it'll,
say
hey
from
this
press
like
enter
like
it's.
B
The
typical
link
like
if
you're
on
an
applicator
you'll
ask
people.
Okay,
confirming
this
press
y
people
typically
just
pressing
to
enter
it's
done,
or
you
could
just
probably
like
pack
build
minus
my
app
name
and
selects
the
first
one
by
default,
gets
it
done,
and
I
would
assume
once
we
have
registry
stats,
we
could
pick
the
default
ordering
based
on
number
of
downloads
or
whatever,
let's
talk
about
so
it
won't
be
biased
to
a
vendor.
B
It
would
be
biased
to
how
many
people
for
a
particular
language
or
project
use
a
specific
pullback
or
like
if
it
comes
with
one
with
the
builder
fine.
But
that's
what
my
idea
was
like
once
registry
has
that
and
we
have
like
some
way
to
get
a
rough
heuristic
of
what
a
particular
buildback
looks
in
a
project.
We
could
like
simplify
this
a
lot.
A
A
I
think
that
that
would
be
extremely
valuable,
so
for
maybe
in
the
more
short
term,
though,
what
if
we
did,
the
workflow
that
I
talked
about
right
like
you've,
went
through
the
the
sample
builder
with
the
sample
app,
then
we
throw
like
choices
at
you
right
and
say
like
here:
are
all
the
builders
select
one
for
your
actual
code
as
the
secondary
step,
but
then
to
satisfy
what
I
think
you
were
trying
to
get
at
in
the
initial
value
is:
let's
say
it
was
even
way
before
all
this
before
they
even
knew
the
concepts
before
they
even
looked
at
the
videos
or
anything
like
that,
what
it-
and
I
think
this
has
really
good
value
for
the
what's
called
the
returning
user
right,
which
is
basically
what
if
on
our
home
page.
B
I
I
think
we
need
like
a
thing
on
the
topic.
People
look
at
all
the
terms
and
buildbacks
get
confused
if
you
could
just
say,
run
this
command
and
on
your
application
and
let
it
do
its
magic.
People
would
then
be
intrigued
at
what's
happening
and
then
they'll
look
at
dogs
and
be
more
interested
in
the
project
yeah.
Currently
they
just
have
to
go
through
like
this
huge
learning
curve
before
they
even
get
to
the
point
where
I
can
build
stuff
with
buildbacks.
B
B
A
No
again,
I
totally
agree
right.
I
think
that
the
biggest
contentious
point
is
just
what
builder
to
use,
and
so,
if
we
were
able
to
assess
or
remedy
that
and
yeah,
we
could
totally
do
do
that.
B
A
Yeah
well,
how
about
this
I'm
gonna
play
devil's
advocate.
I
work
at
vmware,
there's
like
30
000
employees.
A
What
if
we
like
fudge
those
numbers-
and
you
know
all
the
pocato
ones-
are
the
most
popular
ones
or
even
just
naturally
right,
because
vmware
just
chooses
to
use
pocato
and
and
just
by
that
nature
of
magnitude
right
it
just,
and
I
guess
the
worst
part
again,
I'm
playing
devil's
advocate
is
because
it
would
be
the
first
one
off
the
gate.
It
like
continuously
gradually
increases
in
popularity
and
starts
leaving
all
the
other
ones
behind.
B
A
Yeah
and
then
yeah
red
hat
you,
you
saw,
we've
they've
got
new,
build
packs
coming
out.
So.
B
B
B
A
B
I
mean,
I
definitely
think
we
need
some
sort
of
build
packs,
whether
they're
language,
specific
or
not
maintained
by
the
project
so
that
we
dog
food
or
tech.
I
know
like
in
the
past,
we've
had
bugs
in
the
life
cycle
that
we
discovered
later
on
in
like
production
scenarios,
because,
like
no
one
was
using
the
o6
api
and
then,
when
someone
started
to
use
it,
they
were
like.
B
Oh
damn,
caching
doesn't
work,
so
I
I
personally
think
there's
value
in
having
some
minimal
set
of
12
packs
maintained
by
the
project
that
we
use
for
dog
fooding,
and
that
also
drives
us
like
to
think
of
new
use
cases
and
like
to
see
how
our
api
is
actually
like,
leading
like
what
kind
of
user
experience
it
provides.
A
Yeah,
no,
I
totally
agree,
I
think,
for
a
while
there
pack
was
not
using
build
max
to
build
the
pack
image.
I
don't
even
know
if
it
is
still
but.
A
A
Yeah
dark
fooding
for
sure,
okay.
So
I'm
not
sure
that
that
has
an
action
item.
But
I
agree
with
the
sentiment.
I
do
have
two
action
items
which
is
the
copy
and
run
command
on
the
home
page
and
then
kind
of
again
what
I'm
proposing
but
feel
free
to
like
push
back
on,
saying,
like
sample
builder,
with
sample
app
and
then
having
to
select
a
prod
builder
with
your
application
as
the
next
follow-up.
B
B
B
I
think
the
other
thing
you
mentioned
a
while
ago
is
that
the
way
back
currently
gets
like
built
back
metadata
is
through
cloning,
the
entire
registry
and
then
storing
it
locally.
Yes,
so
it
also
sort
of
hampers
the
statistics
that
we
have
on
the
registry
side.
Yes,
what's
happening.
A
Yeah-
and
I
think
that's
part
of
I
think
I
know-
ben
hill's
been
asking
quite
a
lot
for
trying
to
figure
out
how
to
get
stats
out
of
the
book
pack
registry,
but
there
I
know
the
historically,
based
on
the
initial,
like
rfc
of
why
we
went
with
this
implementation,
is
to
remove
the
service
burden
right
like
we
want
to
have
api
that
wraps
around
certain
stuff.
B
I
mean
it
gets
really
hairy
when
you
talk
about
telemetry
big
into
a
cli
too
people
don't
like
that.
I
like
it,
but
I
am
pretty
sure
enterprises
would
be
like
going
anywhere
if
back
had
some
like
telemetry
data,
it
was
sending
back
to
the
registry.
A
I
I
find
that
hard
to
believe
because,
like
g
cloud
scaffold
a
whole
bunch
of
other,
like
I
mean
I'm
sure,
aws
probably
does
too
like
they
all
do,
but
it's
like
all
opt-in
type,
stuff
right,
and
so
I
think
we
follow
that
model
where
it's
like.
Do
you
opt
in,
for
you
know,
analytics
anonymized
analytics.
B
B
Is
so
I
published
a
bunch
of
build
packs
on
ghcr
and
I
see
downloads
for
the
image
there,
which
I
think
are,
I
think,
somehow
bloated
or
there's
something
weird
going
on
there,
because
all
of
my
milk
packs
have
the
same
number
of
downloads
and
they're
all
like,
like
each
of
them
consistently
get
a
thousand
downloads
every
day.
A
B
B
So
I'll
try
to
come
up
with
an
rfc
or
something
for
like
at
least
the
back
parts,
and,
let's
see
if,
like
I
don't
know
for
the
registry
parts
you
mean
in
rfc,
or
will
it
just
be
issues?
I
don't
know
how
that
works
because
I
see
there
are
for
the
registry,
like
even
builders
or
some
other
things
are
just
open
issues.
A
I
think
I
don't
know
how
that
sub
team
is
working,
but
I
think
every
other
sub
team
is
using
subteam
rfcs.
So
that's
what
I
would
propose,
but
I
would
talk
to
them
in
case
it's
an
anti
workflow
for
them,
so
the
rfc
itself
was
just
for
the
builder
list
or
for
something
else.
Sorry
I
feel
like
I
missed
it.
B
A
B
Think
he
wanted
to
create
an
rfc
for
like
some
of
the
things
I
mentioned
like
which
environment
variables,
the
build
pack
is
configured
with
which
files,
and
like
some
other
documentation
around
that
is,
he
is
dan
still
involved
with
the
project.
I
have
not
seen
him.
A
No,
I
think
I
think
he
might
have
taken
a
sabbatical
when,
when
he
left
vmware,
because
I
haven't
seen
him
active
on
github
at
all
so.
B
A
Yeah
yeah,
just
I
guess
for
for
historical
purposes.
Please
still
do
so
because
I
think
it's
valuable
to
revisit,
but
I
did
bring
this
up
like
very
early
on
in
the
project
and
that's
how
we
ended
up
with
like
a
url
like
docs
url,
or
something
like
that.
It
was
because
I
was
looking
for
the
exact
same
thing,
but
through
the
discussions
it
like
got
really
complicated.
So
we're
just
like.
Oh,
let's,
just
throw
a
url
and
they'll
go,
find
the
documentation.
A
Yeah,
no,
I
I
get
it
and
again
we
walk
through
that.
So
I'd
be
happy
to
revisit
that,
because
I
was
a
proponent
for
it.
But
again
there
were
probably
a
lot
of
stuff
up
in
the
air
that
it
maybe
didn't
make
sense
at
that
point
in
time,
but
I
think
at
this
point
we
might
be,
you
know,
mature,
to
be
able
to
do
something
like
that.
A
B
We
did,
we
do
scale,
issues
and
things
this
thing,
no.
A
A
All
that
cool,
so
still,
issues
want
to
take
a
look
at
those.
A
B
A
A
A
B
B
Here,
let's
can
you
do
like
operator
create
a
stack,
should
have
paths
somewhere.
B
A
I
wonder
if
I
could
see
if
micah
could
take
this.
A
A
B
A
Document
sending
images
you
know,
honestly,
I
do
wonder
if
this
is
even
something
relevant
to
the
process.
Yeah.
B
B
A
B
I
think
you
know
what
this
may
be
like
a
good
thing
to
roll
six
store
people
into.
I
don't
know
if
you're
gonna
go
with
cosine
number
three,
but
they're
planning
a
1.0
release.
B
B
Hype
around
it,
I
I
don't
know
if
you
want
to
coordinate
it
with
like
a
601.00,
you
use
it
with
buildblack,
so
it's
like
double
height
for
both
us
and
then.
A
How
much
so
I
guess,
what
are
you
thinking?
Are
you
thinking
that
they
could
write
a
blog
post
on
our?
You
know
on
our
behalf
on
our
thing.
B
Either
way
I
know
down
is
very
receptive
to
these
things.
So,
just
like
he's
on
the
buildback
slack,
you
can
just
ping
him.
Okay,
he
does
a
lot
of
posts
like
he
does
one
every
week
or
something
for
like
emmett
signing
over.
A
Okay,
yeah.
Let
me
I'm
gonna
reach
out
to
joe
see
what
he
thinks
of
the
idea,
because
these
are
our
marketing
one
person
and
then
either
see
if
he
wants
to
reach
out
to
him-
or
I
I
don't
mind
reaching
out,
but
I
want
to
definitely
get
joe's
buy-in
before
I'm
going
to
close
this
for
now
then,
and
then
just
follow
up
with
that.
A
Cool
cool
all
right,
I
think
that
takes
us
to
the
end
of
that.
Do
we
have
any
still
pr's
no
still
prs
and
then
any
unlabeled
issues,
none.