►
From YouTube: WG Platforms Meeting 2023-03-28
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
A
B
A
B
D
B
C
B
Tell
you
the
state
of
things
but
April
10th,
but
let's
talk.
Let's,
let's
go
ahead
and
get
started,
I
guess
it's
1103
people
will
join
us,
they
come,
they
can
watch
the
thing
all
right.
Let's
do
this.
Let
me
paste
the
notes
in
if
people
want
to
follow
along
Okay.
So
first,
let
me
share
with
you
all
where
we're
at
on
the
paper,
as
we
were
just
saying
it
is
done
or
V1
is
done,
it's
merged.
B
So
now
we're
we're
doing
the
work
to
get
it
published,
posted
all
that
kind
of
stuff.
You
can
see
in
this
Milestone
issue.
The
last
couple
things
are
to
get
the
PDF
generated.
Thank
you.
Saeem
for
the
graphic
they've
got
the
dock
they
need,
so
they
just
have
to
generate
it.
The
site
is
actually
up
and
running.
You
know:
yeah
I,
I
think
I'm.
Okay,
it's
a
soft
launch,
so
you
can
actually
see
it
here.
I'll
paste
it
in
yeah.
B
So
but
then
so
that's
great
I'm
excited
about
that.
If
you
see
anything
there,
I
guess,
let
us
know,
and
then
a
blog
post,
I,
I,
just
Abby
and
me
were
working
on
it
and
I
just
put
it
in
as
a
pull
request.
Just
a
draft
pull
request.
For
now
till
we
get
the
PDF
and
it's
in
there
oh
and
yeah,
and
that
website
like
check
out
the
website.
B
B
B
D
I
think
have
this:
have
they
I
think
have
this
running
might
be
a
good
idea.
We
can
see
some
good
responses
like
on
the
final
day,
somebody.
This
is
not
good.
You
have,
while
you're
doing
it,
I
think
I
think
a
one
week
of
review
or
you
we
can
ask
CNC
have
to
read
to
me
or
something
I
think
there
is
some
kind
of
Base
awareness,
yeah.
B
C
E
B
Yeah
somebody
can
copy
it
and
say:
oh,
they
have
to
give
us
attribution,
I,
don't
know
if
it
has
a
license
on
there.
Actually,
the
repo
has
a
license
yeah
and
then,
if
you,
if
you
do,
want
to
change
the
way,
oh
so
two
things
I
should
tell
you
actually.
So
if
you
look
in
the
website,
folder,
which
I
just
put
in
the
notes,
that's
where
we
can
make
changes,
except
that
we
worked
it
out
that
we're
using
Sim
links
to
the
actual
paper.
B
So
now,
if
you
look
in
the
platforms
white
paper
directory,
which
I
can
also
paste
in
here,
you'll
see
there's
two
versions
now
a
V1
and
latest
so
going
forward.
The
latest
is
actually
what's
live
on
this
on
the
site.
V1
will
be
what
goes
into
the
PDF.
So
if
you
want
to
make
changes,
just
submit
a
a
proposal
to
latest
and
that
will
even
be
picked
up
by
the
by
the
website
right
away.
C
B
E
B
F
C
F
So
from
the
Builder
maintainer
perspective,
they'll
be
able
to
directly
through
some
proxy
questions,
answer
the
state
of
things
in
terms
of
culture
and
and
product
and
how
yeah,
essentially
how
mature
the
platform
is
from
the
consumers
is
the
kind
of
other
proxy
way
we
can
see
if
it's
self-serve,
if
people
are
able
to
contribute,
what
does
that
look
like?
Can
they
access
the
services
they
need?
Are
the
guardrails
in
place.
F
So
the
goal
is
to
get
that
feedback
to
probably
start
thinking
about
how
we
could
iterate
on
the
survey
and
launch
it
with
some
sort
of
cadence
to
do
kind
of
a
longitudinal
thing
to
see
how
things
change
in
the
industry
over
time
and
then
for
us
more
specifically
and
more
immediately.
We
want
to
recruit
those
people,
so
we
can
interview
them
one-on-one
and
go
deeper
on
their
experiences
in
their
organizations
and
I
think
Josh.
F
B
F
B
F
Yeah
I'm
I'm.
It's
going
to
be
great
to
see
kind
of
in
aggregate
how
things
Stack
Up
in
terms
of
what
things
look
like
out
there
and
what
problems
folks
are
facing,
but
particularly
it's
going
to
be
great
to
start,
lining
up
interviews
and
then
be
able
to
coordinate
on
our
end
to
kind
of
divide
and
conquer
across
regions
like
interviewing
folks
in
across
all
these
kind
of
things.
B
F
C
B
All
right
next
thing
on
the
agenda
I
wanted
to
share
with
you
all
just
make
sure
you
know
that
we're
scheduling
a
few
Tag
app
delivery
things
at
kubecon.
We
have
the
pre-day
event
from
one
to
five:
we'll
have
a
booth,
Wednesday
and
Thursday
for
from,
like
after
the
Keynotes
till
like
two
three
o'clock,
I
think
the
what
first
days
till
2
30
a
second
days
till
four
would
love
to
come
meet
us
there.
B
I
also
put
up
there's
a
link
in
the
thing
and
I
should
share
it
here,
also
just
in
case,
but
we're
looking
for
to
have
kind
of
like
unconference
unmead
up
lightning
talks
about.
You
know
realize
you're
talking
to
people
that
you
know
spend
their
whole
days
deep
in
this
stuff,
so
you
don't
have
to
start
from
101,
but
we're
looking
for
like
five
to
ten
minute
talks
on
the
300
400
level,
stuff
that
you
know
is
new
to
us,
and
you
know
we
should
know
about,
and
collaborate
on
and
things
like
that.
C
B
G
I
have
a
question
use
of
here.
There
are
three
interest
points
and
tag
ability,
number
one
is
get
Ops.
The
number
one
is
number
two
is
app
delivery
and
the
third
one
is
platform
on
the
cncf
calendar.
I
only
see
the
meter
for
platform,
I
am
more
interested
in
ability
and
GitHub
githubs
as
well.
So
how
do
so?
What
is
the
calendar
for
that.
E
Just
just
to
clarify
when
the
app
delivery
is
actually
the
tag
itself,
while
the
working
group
The
platforms,
work
in
group
which
is
and
the
open
what
used
to
be
the
Gatehouse
work
Group,
which
is
now
kind
of
the
open
get
us
project,
is
underneath
Tag
app
delivery.
So
it's
all
kind
of
in
the
same
domain
so
to
speak.
But
there's
there's
one
meeting
for
Tag
app
delivery,
one
meeting
for
the
getups
working
group
or
open
getups
project,
and
then
this
there's
this
one
and
yeah.
C
B
The
way
it's
it's,
you
know,
cncf
is
kind
of
interesting.
We've
got.
You
know
the
talk,
tags
ouija's,
but
the
tag
kind
of
cope
is
the
you
know,
coordinating
all
the
works
and
Luigi's
do
the
project.
So,
like
we're
writing
this
paper,
we
created
a
Ouija.
You
know
the
operator
could
call
themselves
a
Ouija
if
they
want
it
also
or
they're
working
on
the
paper.
Sometimes
you
know
ouija's
shut
down
if
they're
done
with
their
work.
Sorry
I
keep
calling
it
Ouija
because
I
don't
like
to
have
to
say
the
whole
W.
G
E
There's
also
the
slack
challenge
where
a
lot
of
these
things
get
shared
and
if
there's
anything,
that
kind
of
falls
under
Tag,
app
delivery
or
any
of
the
working
groups,
or
even
projects
that
are
underneath
the
tag
you
just
kind
of
surface,
that
as
a
thing
that
you
want
to
work
on
and
and
people
start
organizing
around
their
ideas.
That's
that's
usually
how
it
works.
B
B
Think
of
the
paper
is
the
attributes
of
the
platform,
which
are
things
like
platform
as
products
self-service
things
like
that,
and
then
the
capabilities
of
a
platform
which
are
things
like
artifact
storage
or
that's
a
really
random
one
observability
identity
anyway,
there's
a
tracking
dock
in
there.
Also,
if
you
go
into
our
GitHub
issues,
there's
a
couple
that
I'm
trying
like
secrets
and
service
bindings,
I'm
kind
of
trying
to
herd
cats
on.
B
Basically,
you
know
what
I'm
doing
in
mind:
I
guess
I'll
share
and
we're
starting
to
maybe
do
this
for
artifact
storage,
which
is
why
I
bring
this
up
is
gather
the
people
that
have
a
stake
like
let's
say
in
secrets.
Of
course,
hashicor
we
need
to
get
involved.
Dapper
has
a
Secrets
management
system.
B
We
talked
to
the
people
contributing
external
Secrets
operator,
a
couple
others
you
know
trying
to
get
them
together.
My
next
step
when
I
have
some
time-
and
we
have
you-
know
the
10
right
people
or
15
20
right.
You
know,
stakeholders
will
be
pulled
them
into
a
meeting
and
talk
about
the
things
we
want
to
achieve.
The
ideas
that
we
have.
You
know
where
the
synergies
might
exist
and
then,
if
that
succeeds,
if
we
can
get
some
momentum,
then
maybe
form
a
Ouija.
B
Another
working
group
in
Tag
app
delivery
kind
of
what
I'm
experimenting
with
so
the
artifact
storage
I
wanted
to
mention
this
to
you
all
the
folks
that
work
on
oras
oras,
which
stands
for
oci
registry
as
storage,
their
goal
is,
to
you
know,
establish
standards
and
patterns
and
practices
for
using
oci
to
bundle
any
kind
of
content,
not
Conti
everything,
not
container
images
they
think
about.
So
you
know
when
you
see
a
health
chart
in
an
oci
image
or
operator
framework
operator
or
an
oci
image
or.
B
C
B
A
G
Yes,
that
is
why
I
asked
in
the
beginning
that
okay
I
see
three
working
groups.
One
was
one
was
get
Ops
and
one
was
this:
I
see
that
okay,
we
had
the
meeting
for
get
Ops
on
on
Thursday
on
on
Wednesday,
14th
I
think,
and
these
are
two
weeks
so
I
was
missing
it
on
the
calendar.
That
is
why
I
wanted
to
ask
that:
okay,
no
meeting
after
two
weeks.
That
is
what
the
calendar
said
but
I
think.
C
E
The
air
there
are
two:
there
are
two
slack
channels
that
could
then
be
interesting
for
you
to
fall,
but
pay
attention
to
is
the
WG
Dash
get
Ops
and
the
open
get
UPS
channels
on
the
cncf
slack.
So
you
know
a
lot
of
those
information
like
like
when
the
meetings
are
and
stuff
like
that
is
also
getting
spammed
there
and
and
discussions
are
happening
there
as
well.
E
G
E
G
B
B
E
A
Sure
I'm,
probably
not
the
best
best
person
to
talk
about
it,
maybe
same,
can
also
add
something.
But
the
main
thing
that
happened
was
that
last
week
there
was
a
bit
of
a
discussion
in
the
issue
about
basically
the
discussion
that
we
also
had
like
hey.
This
is
like
one
example
of
a
platform,
but
how
do
we
make
clear
that
it's
not?
This
is
not
the
the
one
and
only
way
to
do
things
and
yeah,
so
that
was.
A
It
was
just
like
a
lot
of
clarification
and
then
the
big
question
which
I
think
we
didn't
answer
in
this
discussion,
was
who
will
be
maintaining
this
prototype
because
it
will
go
over
into
the
ownership
of
the
cncf
of
this
group.
So
while
we
can
all
say
and
agree
that
maybe
the
group
as
a
whole
maintains
like
the
Prototype
I,
don't
think
that's
I,
don't
know
it
just
seems
like
kind
of
like
you're
inviting
some
trouble
I'm.
E
A
D
Yes,
absolutely
so
George.
Basically,
we
have
a
chat
with
Sala
boy,
Mauricio
solatino
in
the
past,
that
he
wrote
some
kind
of
book,
so
basically
we're
working
on
some
of
the
Prototype
I
think
we
started
some
discussion
around.
What
should
the
projects
look
like,
so
he
spent
some
of
the
time
around
his
book
where
anybody
wrote
it
wrote
his
book
about
it.
D
Can
his
continuous
delivery
with
the
knee
table
so
he's
basically
actually
have
some
opinions
around
using
cross-plane
V
cluster
K
native
to
work
together
to
build
a
platform
that
the
basically
you
consume,
the
API
using
the
cross
plane
and
then
the
V
cluster?
Actually
using
give
you
some
Dev
environments,
and
then
you
have
a
get
Ops
approach.
You
can
create
the
environment
on
the
fly,
so
we
have.
He
has.
This
prototype
is
already
built
up,
but
I
think
we
have
something
on
the
paper
that
is
actually
in
the
past
discussion
like
now.
D
D
Basically,
in
that
document,
so
basically
on
that
documentary
I
will
try
to
share
very
like
it
is
we
actually
document
some
of
the
stuff
like
this
is
the
Prototype
or
platform
should
be,
but
currently
the
demo
is
missing.
Like
example,
code
sample
Mauricio
has
the
word
example
from
one
of
his
talk
at
the
cubecon,
but
we
need
to
figure
it
out
the
structure
of
what
should
be.
We
can
extract
from
his
his
repository
and
make
it
available
on
the
separate
repository
and
show
these
kind
of
projects,
because
those
are
all
cncr
projects.
D
A
A
B
A
A
Was
that
we
we
start
with
one
example,
and
we
add
other
examples
on
with
other
tools
and
other
paradigms
or
approaches
yeah,
which
is
what
was
part
of
this
discussion,
was
I.
Think
right
right
now,
at
the
moment,
I
think
the
biggest
problem
is
just
that.
We're
all
super
busy
with
kubecon,
so
I
don't
expect
any
big
movements
until
after.
B
D
Yeah
so,
basically
like
I
think
Josh.
What
we
can
do
is
like,
basically,
you
know,
remember:
we
have
a
three
layer
like
infrastructure
app
and
basically,
if
they
have
three
layers
to
work
with,
so
there
is
a
content
already
being
the
different
places
from
the
different
contributions
of
the
cncf.
We
can
use
that
and
that
we
can
actually
build
upon,
or
actually
we
can
skip
that
because
the
time
is
running
short.
It's
actually
we're.
Actually,
two,
probably
two
weeks
away
from
the
cube.
B
D
A
G
B
No
I
think
you're
right,
I
think
that's
what
everyone
is
is
saying
that,
like
there's
there's
a
billion
different
possible
platforms.
Where
do
we
start
and
I
think
we're
kind
of
saying,
like
we'll
start
with
this
one,
we'll
Define
it
in
a
certain
way
and
then
we'll
we'll
figure
out
we'll
go
from
there,
I
mean
if
you
want
to
build
an
ml
Ops
platform,
you
know
or
a
minimum
viable
platform.
You
know
we
can
figure
out
where
that
can
go.
C
A
G
No,
no
no
I
am.
This
is
a
prep.
We
have
a
problem
statement
right
that
we
need
to
come
up
with
a
possible
approach
to
a
platform
this
can.
This
is
what
a
platform
looks
like
before.
We
start
actually
start
coding
that
we
need
to
Define
okay.
How
do
we
attack
this
problem?
That
is
why
I'm
saying
that,
okay,
what
is
the
attack
Vector
of
this
particular
problem?
Not
the
threat.
A
E
A
We're
just
trying
to
generalize
it
a
little
bit
so
we
can
like
soak.
We
can
just
create
a
showcase,
it's
not
supposed
to
be
an
actual
solution
per
se.
It's
more
like
this
is
how
it
could
look
like
like
if
you
were
to
do
something
like
that.
But
then
you
have
your
own
specific
problem
and
your
own
specific
attack
vector
and
the
the
Prototype
is
not
going
to
help.
You
find
out
the
exact
way
to
solve
your
particular
problem.
Does
that
make
sense
to.
G
You
before
it
makes
sense
to
me
I
have
a
couple
of
questions
number
one.
You
are
saying
that
okay,
this
the
the
the
Prototype,
is
not
a
an
actual
working
prototype.
So
that
means
we,
since
there
are
a
very
the
the
the.
What
you
are
trying
to
say
is
that
the
problem
space
is
so
vast.
We
can
have
multiple
prototypes.
G
C
G
A
I
think
this
is
what
Josh
was
saying
that
some
documentation
that,
like
just
gives
you
a
high
level.
You
know
this
is
this-
is
how
we're
solving
this,
this
very
big
problem
and
so
we're
solving
it
in
this
particular
way.
I
think
that's
kind
of
the
the
idea
there's
also
a
doc
linked
to
in
the
issue
right
like
a
Google
doc,
where.
A
B
Yeah,
yes,
I
I,
really
like
what
I
mean
I'm
understanding.
What
you're
saying
Yusuf
like
as
we're
as
we're
developing
a
prototype.
We
should
also
Define
what
is
the
theoretical,
hypothetical
problem
that
this
platform
was
meant
to
solve,
that
that's
a
fair
that
that's
a
fair
thing
like
yeah
Mauricio
built
this.
What
was
I
guess
with
demo,
where
it's
kind
of
which
came
first
yeah?
What
was
he
thinking
to
solve
with
this
and
make
sure
we're
clear
on
that?.
D
So,
basically,
Joseph,
actually
I
shared
his
link
to
his
repository
he's,
giving
a
talk
with
the
cube
day.
Job
on
so
basically
I
think.
The
idea
I'll
give
you
some
little
little
bitty
of
detail
on
this
platform.
So
basically,
currently
the
model
is
in
the
devops.
Word
is:
whenever
you
actually
have
some
report
some
code,
then
you
need
a
separate
environment
for
the
testing
and
then
you
have
a
CI
CD
pipeline
from
the
development
testing
to
the
production
and
it
can
cause
a
lot
of
time.
D
You
can
build
your
developed
environment,
then
you
file
a
ticket
to
your
devops
team.
Give
me
a
separate
environment,
so
I
can
test
my
change
it's
low
in
that
environment
when
everything
is
actually
working
correctly,
then
I
move
my
changes
into
the
production
environment,
so
basically
a
g
r,
ticketing
process
startup.
So
what
his
talk
is
what
the
cube
ditch
talk
he
actually
mentioned.
You
can
use
these
three
projects
like
cross,
plane
and
basically
the
recluster,
and
then
the
K
tops.
Basically
what
you
can
do.
D
Is
you
give
your
infrastructure
as
an
API
for
your
Dev
teams,
who
can
call
this
infrastructure
and
then
can
create
an
environment
like
Development
Department
using
the
V
cluster?
Then
you
can
deploy
your
code
app
on
top
of
it
and
let's
say
when
you
actually
have
a
new
app
new
version
of
that
app
is
deployed,
then
you
can
zoom
in
other
apis
create
another
separate
environment.
When
those
environment
actually
merge.
You
use
a
github's
approach
to
deploy
that
to
the
Productions.
You
actually
shorten
the
loop
from
go
to
the
production.
So
what
is
actually
showcase?
D
G
Understood
understood
so
when,
when
you
are
saying
this,
then
there
is
no
difference
between
the
platforms
they
are
that
are,
or
maybe
a
little
bit
different,
that
are
offering
environment
as
a
service,
for
example,
release
Hub.
Then
there
is
EnV
0
they
are,
they
are
offering
environments
as
a
service.
So
what
they
do,
is
they
completely
spin
up
a
new
environment
with
everything
so
probably
What?
What
the
demo
is
that?
Okay,
how
can
you
build
an
environment
as
a
service
platform
using
kubernetes
I,
think
that
is,
if
assessment,
but.
D
Yes,
absolutely
so,
basically,
that's
the
question
for
the
Josh
as
well
like
if
we
want
to
have
some
documentation
for
the
Prototype,
like
remember,
we
mentioned
in
our
talk
in
in
our
platform
paper,
reducing
the
cognitive
low
on
the
platform
games.
So
actually
what
this
repository
is
actually
showcasing
that
how
you
can
introducing
the
cognitively,
because
developers
end
up
doing
the
repetitive
task.
You.
D
A
service
a
let's
say,
monitoring
system,
so
they
have
some
kind
of
repetitive
task.
So
this
approach
that
the
salad
bar
is
actually
mentioning
can
actually
reduce
those
repetitive
tasks.
So
what
we
can
include
that,
because
this
is
already
have
some
content
available,
we
can
refactor
it
make
a
new
document,
and
then
we
can
link
into
the
repository
to
that
platform
prototype,
but
that's
actually
showcasing
the
reduction
of
the
cognitive
workloads
from
the
platform
teams.
Yes,.
G
Yes,
from
the
platform
teams
or
from
from
the
because
platform
team
will
be
the
one
that
would
end
up
maintaining
that
particular
platform,
but
from
from
the
actual
developers
who
are
actually
solving
the
business
problem,
we
don't
want.
Actually
the
the
point
point
of
the
platform
as
I
understand
is
I
can
be
wrong.
Is
that
okay,
we
reduce
the
cognitive
load.
We
don't
want
to
worry
the
developer,
who
is
actually
solving
the
business
problem
to
worry
about
the
infrastructure?
G
G
B
B
A
Yeah
I
mean,
if
you
have,
if
you
feel
up
for
it,
not
all
the
things
that
you
just
said,
or
the
substance
that
that
you
want
to
write
down
like
add
them
as
comments
to
the
issue.
So
we
can
capture
that
and
we
don't.
It
doesn't
get
lost.
You
know,
and
that
would
be
super
helpful
or
to
the
to
the
dock.
If
you
feel
like
it's
more
fitting
in
there
either
way,
if.