►
From YouTube: Keptn Community Meeting - Feb 24, 2022
Description
Keptn 0.13.x status, project demos and Keptn in Google Summer of Code 2022 discussion.
Meeting notes: https://docs.google.com/document/d/1y7a6uaN8fwFJ7IRnvtxSfgz-OGFq6u7bKN6F7NDxKPg/edit
Learn more: https://keptn.sh
Get started with tutorials: https://tutorials.keptn.sh
Join us in Slack: https://slack.keptn.sh
Star us on Github: https://github.com/keptn/keptn
Follow us on Twitter: https://twitter.com/keptnProject
Sign up to our newsletter: https://bit.ly/KeptnNews
A
Good
morning,
everyone
yeah
just
quick
updates
and
everyone
joining.
We
moved
our
developer
meetings
to
9am
utc,
so
as
it
was
announced
in
captain
slack
now,
it
will
be
a
regular
time
for
the
time
slots.
So
there
is
announcement
in
a
general
channel
on
slack
with
the
summary
of
how
the
meetings
are
going
to
happen,
going
forward
and
very
updated,
as
well
as
captain
calendar.
So
just
take
a
look
if
you
want
to
participate
in
the
meetings.
As
always,
everyone
is
welcome
and
I'm
not
sharing
my
screen
right.
A
No
well,
but
that's
right.
Okay,
so
yeah!
If
you
want
the
announcement
is
here
so
it's
in
the
general
channel
and
the
captain
calendar
and
community
meetings.
Everything
is
up
to
date.
So
take
a
look,
then
another
heads
up
that
next
week
on
wednesday,
we
will
have
a
presentation
about
my
supermoon
and
captain
integration.
So
we
have
integration
service
and
I
will
announce
the
meeting
today,
but
it
will
happen
same
time.
My
name
you
see
on
wednesday,
so
developers
are
welcome
to
join.
A
B
Yes,
so,
unfortunately,
we
discover
a
small
regression
on
the
quality
gates
use
case
and
we
fix
it,
and
now
we
are
in
the
process
of
doing
the
release.
It
will
come
later
today
or
tomorrow
morning
and
also
contains
some
other
small
bug,
fixes
that
were
discovered,
especially
with
the
open,
idc
implementation.
B
A
So
well
humber
deployed
013.
I
guess
they
had
an
opportunity
to
revert
if
we
depicted
them.
So
I
think
it's
okay.
We
got
some
comments
in
the
slack
channel,
but
everything
is
contained
by
now.
So
thanks
to
everyone
who
is
working
on
these
issues,.
A
C
A
Yeah-
and
maybe
only
one
person
cannot
share
the
screen-
yes
sure,
okay,
now,
then
I'll
go
ahead
and.
C
C
Okay
yeah,
so
my
sprint
was
mostly
fixing
small
issues
in
the
pipelines,
so
I'm
gonna
get
right
into
it.
The
first
pr
is
about
fixing
our
oci
levels,
labels
in
in
all
docker
images.
C
We
actually
had
the
the
labels
before,
but
I
think
giovanni
noticed
that
if
you
press
doc
like
if
you
call
that
inspect
on
one
of
our
images,
the
labels
are
actually
all
meshed
together
into
one
label,
and
that
was
yeah
actually
a
pretty
minor
issue.
C
C
Then
next
up
we're
using
go
test
some
now
for
all
go
based
services.
That's
basically
a
command
line
wrapper
for
the
go
test
command,
which
gives
you
some
some
nice
lock
output
and
all
that
stuff.
C
So
now,
if
we
check
the
unit
tests
for
any,
let's
go
to
the
secret
service,
whatever
you
would
actually
get
a
nice,
a
nice
summary
of
all
the
different
unit
tests
that
were
run
with
a
nice,
green
or
red
or
yellow
tag
if
they
passed
or
failed.
So
it's
it's
much
easier
now
to
actually
find
failing
tests
in
in
a
whole,
in
a
whole
lot
of
log
messages
in
the
ci
pipeline
yeah,
so
that
was
introduced
for
all
the
different
go
based
services.
C
What
else?
Oh
yeah?
We
had
some
labor
cleanup
in
our
hemp
chart,
so
we
actually
had
already
some
some
helper
functions
to
to
set
up
common
labels
for
different
resources
that
we
have
in
our
hemp
chart.
But
actually
they
were
not
used
so
much
so,
especially
in
the
rbac
file.
We
we
have
our
like
service
accounts
and
role
bindings
and
stuff.
Basically,
all
the
different
resources
had
all
the
labels
duplicated.
C
So
basically,
I
replaced
all
those
with
the
common
set
of
labels
and
just
included
them
from
the
helper
functions
yeah,
which
in
the
end,
doesn't
actually
change
much
in
the
hampshire
itself
in
the
outcome,
but
it
reduces
our
lines
of
code
a
little
bit.
It
makes
to
make
them
more
readable.
C
And
then
one
of
the
bigger
changes
I
did.
We
have
a
new
captain,
dev
docker
hub
organization
now,
and
that's
going
to
be
used
to
push
all
our
development
images.
So
everything
that
comes
from
any
ci
runs,
except
for
release,
builds,
is
going
to
be
pushed
to
the
captain
death
registry
instead
of
the
captain
registry,
and
this
has
the
result
that
the
normal
captain
registry
should
actually
be
clean
and
only
has
our
released
images
in
one
place.
A
One
practical
question:
so
captain
dev
is
generally
helpful.
It's
nice
to
have
this
repository
for
as
a
playground,
but
at
the
same
time
it
might
make
sense
to
actually
use
github
packages
going
forward,
because
in
this
case
you
can
set
up
our
identification
for
particular
repository
and
also
it's
much
more
easier
from
developer
experience.
You
can
connect
with
have
actions
to
deployments.
C
Because
for
the
captain,
death
docker
hub
organization,
not
really.
D
D
C
A
Yeah,
because
if
we
create
captain
dev,
I
guess
we
would
need
to
apply
to
docker
for
another
sponsorship,
so
that
anonymous
users
don't
experience
download
limits.
A
So
if
we
want
to
interace
any
ci
flow
that
push
pulls
images
from
docker
hub,
which
might
become
a
problem
we
passed
through
that
in
jenkins
six
months
ago,
because
we
also
had
jenkins
foreign
organization,
where
we
forgot
to
request
sponsorship
and
our
pipelines
became
quite
unstable.
After
that.
D
C
Yeah
so,
as
far
as
I
know,
we
are
not
sponsored
anyways,
yeah
or
like
open
source
tag.
Basically
on
docker
hub
we
just
have
a
free
organization
and
actually
in
the
pipelines,
we
always
use
our
captain
bot
user.
So
we
are
authenticated
and
we
don't
run
into
limits.
So
that
should
be
fine,
but
yeah
users,
of
course,
could
still
run
into
the
limits.
That's
true
yeah,
so
maybe
guitar
would
actually
be
a
good
option.
There.
C
It's
not
planned
yet,
but
it
makes
sense,
of
course,
yeah
to
also
push
integrations
there.
Yeah.
A
That's
a
good
point
so
well,
maybe
at
that
point
we
should
actually
think
about
the
github
register
for
sure,
because
permission
model,
et
cetera,
is
going
to
become
a
pain
so,
for
example,
in
junkies
for
a
while,
we
just
say
that
we
don't
guarantee
any
security
of
these
images,
because
everyone
on
our
ci
instances
can
push
to
any
image
and
yeah.
It
comes
with
disclaimers,
but
it's
let's
say
sub-optimal.
C
Yeah
yeah
for
us
it's
actually
restricted
anyways,
so
it's
only
the
captain
parties
that
I
can
push
there
for
either
of
the
organizations
on
underground
understood.
Thank
you
cool.
If
there's
no
other
questions,
I
will
go
to
my
last
pull
request,
which
is
that
I
fixed
the
cleanup
of
our
static
kubernetes,
not
the
static
ones,
but
the
integration
test,
resources
that
we
are
using
on
our
static
clusters.
C
C
So
so
now
all
those
resources
get
depending
on
on
what
run
it
was,
and
they
either
get
a
three
day
time
to
live
or
a
six
hour
time
to
live
or
they're
actually
deleted
all
right
depending
on.
If
it
was
a
scheduled
integration
test,
one
a
manual
one
or
what
was
the
last
one,
if
the
run
actually
succeeded
all
the
way,
then
the
stuff
is
deleted
right
away,
yeah-
and
here
also,
I
fixed
a
small
issue
with
this
read
array-
function
that
I
use
here.
C
Basically,
if
you
don't
get
anything
out
of
there,
it
will
still
put
out
some
white
space
and-
and
I
couldn't
actually
check
the
array
to
be
just
empty
because
it
always
returned
that
it's
not
empty,
even
though
there
was
only
white
space
inside
so
now.
I
fix
that
by
still
iterating
over
it
and
then
having
every
single
item
in
there
replace
the
white
spaces
and
then
check
if
it's
there
or
not
yeah,
and
that's
it
from
me.
G
A
And
from
maybe
it
makes
sense
to
try
reconnecting
because
I'm
sure
what's
going
on.
G
Good
so,
hopefully
you're
receiving
the
screen-
yes
yeah
great
yeah,
so
my
screen
was
basically
about
or
mainly
about
two
features.
G
D
G
Assume
that
all
users
know
this
this
type
of
authentication
authentication
from
classical
repositories.
I
have
a
simple
example:
how
to
use
it
yeah.
Until
now
we
have
supported
only
usage
with
token
and
the
https
protocol.
G
You
can
see
an
example
command
here.
Also
when
creating
a
project
user
will
will
add
the
shipyard,
the
username
git
token
and
git
remote
url.
Now,
from
now
and
from
the
next
version,
we
will
support
also
the
private
public
key
authentication.
G
So
when
creating
a
project,
the
user
will
add
user
name,
then
a
private
key,
the
git
remote
url
with
the
ssh
protocol
and
voluntary
also
get
privately
private
key
passphrase,
as
the
private
public
private
key
can
be
encrypted,
also
with
the
fastphrase
yeah
other
than
that
user
needs
to
upload
his
public
key
to
the
remote
git
depository
and
therefore
the
private,
the
private
key
will
be
used
as
an
authentication
method
to
the
remote
kit
repository.
C
I
have
one
question
to
that:
yes,
just
out
of
curates
curiosity,
what
happens
if
you
provide
private
key
and
password
and
also
a
git
token,
there.
C
G
H
G
As
in
the
same
manner,
it
is
base64
encoded.
Okay,
it's
not.
D
A
A
I
think
what
we
actually
need
to
do
is
for
pull
request
template
at
the
checklist.
What
needs
to
be
done
so
that
basically
definitional
done
so
common
things
like
if
the
change
needs
test
automation,
it's
done
if
the
change
needs
recommendation
and
then,
if
it's
done,
if
there
is
needed
in
tutorial
updates,
it's
done
because
at
least
well
downstream,
pull
requests
are
created.
A
So
basically,
ideally,
when
we
merge
changes,
there
should
be
already
stitched
changes
to
documentation,
etc,
so
that
we
are
sure
that
everything
gets
delivered
as
one
chunk.
So
if
the
team
agrees
with
that,
we
can
actually
try
it
out.
A
A
Well,
not
much
everything
at
the
same,
but
at
least
have
everything
prepared.
So,
for
example,
we
have
a
change
towards
the
upcoming
release
and
before
imagining
this
change,
we
set
expectations
that
there
is
a
documentation,
pull
request
which
is
already
stationed
so
that
we
don't
have
it
in
the
backlog
for
some
time
later
in
the
cycle
to
create
it
and
sometimes
as
we
all
do
but
yeah
we
could
just
set
basically
a
checklist
of
what
commonly
needs
to
be
updated
in
other
repositories.
B
Yeah,
the
problem
with
this
specific
case,
though,
is
that
the
resource
service
is
not
yet
so
it's
not
for
the
configuration
service,
which
is
the
base
service
which
is
shipped
with
captain.
This
is
for
the
resource
service,
which
is
in
beta,
so
it's
not
yet
stable
and
we
will
make
the
documentation
when
we
will
mark
this
as
a
stable
service.
I
agree.
A
A
A
So
it's
completely
off
the
scope
of
for
this
particular
pull
request,
but
maybe
we
could
try
putting
documentation
inside
the
same
repository
same
request
and
then
we
figure
out
when
to
move
it
over.
D
B
D
D
A
A
A
This
problem
offline,
so
it's
definitely
a
chunk
of
work
to
do.
I'm
might
not
just
suggest
an
update
to
pull
request
template
and
if
you
agree
that
we
for
pre-release
services,
we
could
just
start
the
staging
called
the
recommendation
in
the
readme
for
these
pre-release
services.
I
think
it
would
be
a
convenient
I
think
for
everyone,
because
for
valiation
we
can
just
refer
them
to
this
documentation.
A
A
D
A
G
So
probably,
I
can
continue
yeah.
The
second
feature
I
was
working
on
was
supporting
the
git
coming
id
flag
when
triggering
the
evaluation.
G
This
is
also
possible
only
in
the
resource
service,
as
the
configuration
service
or
when
kept
in
using
the
resource
service
of
the
configuration
service
yeah.
So
really
short,
example.
Actually,
the
aim
of
this
feature
is
to
make
con
consistent
resources
through
the
long
evaluations
runs.
G
G
It
will
point
to
the
specific
state
of
the
git
repository
defined
by
the
git
commit
id.
So
the
previous
usage
is
without
the
git
commit
id.
As
you
know
it,
the
new
feature
adds
the
git
commit
id
with
a
specified
id
and
yeah.
Actually,
the
evolutions
is
run
in
this
kind
of
state
of
the
kit
repository
until
the
end.
G
This
the
evolution
can
be
used
also
without
the
git
commit
id.
It
is
voluntary,
but
it
is
actually
highly
recommended.
G
All
types
of
reasons-
yeah,
that's
kind
of.
B
B
G
The
git
commit
id
would
not
be
found,
so
probably
we
will
get
an
error
that
the
git
repository
cannot
check
out.
The
specified
git
commit
id.
F
G
No,
it's
not
ever
available.
We
faced
some
problems
with
the
home
service
where
fetching
the
resources
or
helm
charts
which
were
uploaded
as
terribles
to
the
git
remote
repository
was
not
actually
possible
using
the
git
commit
id.
So
this
functionality
this
functionality
we
do
not
have
yet.
G
No
captain
trigger
deployment
does
not
work
with
git
compute
id.
J
Another
question:
can
I
also
use
the
the
shortened
version
of
the
id,
or
do
we
always
have
to.
F
E
E
Okay,
so
for
me
it's
a
free
pr's
this
week,
so
the
first
one
is
fix
for
the
distributor.
E
So
therefore,
the
the
default
timeout
has
been
increased
to
30
seconds,
but,
of
course,
for
these
kinds
of
values,
it's
hard
to
find
the
one
size
fits
all
solution.
So
that's
why
I
also
added
this
environment
variable
here
that
can
be
set
in
the
distributor.
E
C
C
Yeah
is
this
also.
E
So
when
installing
captain
we
are
using
yeah
default
security
contacts,
that's
used
for
most
of
the
services
and
there
are
two
special
cases
like
the
bridge
and
the
api
gateway
which
needs
slightly
different
configurations
than
the
other
ones
and
the
problem
we
had
here
that,
due
to
these
being
hard
coded
in
the
helm,
chart,
for
example,
on
more
restrictive
environments
like
openshift,
with
security
construct
constraints,
enabled
this
led
to
some
incompatibilities
and
that's
why
we
added
the
possibility
to
define
the
security
context
both
for
the
port
security
context
and
the
container
security
context
via
their
values
yaml.
E
So
these
are
the
default,
so
they
have
been
added
for
the
for
the
bridge
and
the
api
gateway
and
the
common
security
context,
and
if
there
are
no
further
specified
and
these
default
values
are
used.
But
now
users
will
have
the
possibility
to
overwrite
them
by
providing
these
values
during
the
helm,
upgrade
or
helm
install
command.
E
E
This
has
or
will
have
two
big
advantages,
so
the
first
one
is
that
this
new
data
model
will
support
threat
safety,
which
means
that
this
will
allow
us
to
run
the
shipyard
controller
with
multiple
replicas
without
each
of
the
replicas
interfering
with
each
other's
during
the
execution
of
a
sequence.
E
So
that's
the
first
benefit
we
expect
from
that,
and
the
second
one
is
also
quite
nice,
so
it
will
make
the
shipyard
controller
and
its
logic
actually
a
lot
more
easy,
because
during
this
pull
request,
I
noticed
that
I
can
get
rid
of
quite
a
number
of
collections
that
were
previously
used
and
kind
of
made.
The
logic
of
the
shipyard
controller
unnecessarily
complex,
and
this
has
been
simplified,
so
that
will
hopefully
also
contribute
to
the
shipyard
controller
being
easier
to
maintain
going
forward
all
right.
E
F
So
the
first
thing
I
worked,
this
sprint
was
a
simple,
but
we
have
with
integration
test.
Basically,
because
we
changed
the
name
of
the
file
to
something
which
is
not
actually
the
real
name
of
the
file.
One
of
the
tasks
could
not
run,
and
since
I
was
there,
I
also
removed
some
folder
dependency.
We
had
among
the
tests
so
now
locally.
We
can
run
all
of
the
tests
in
any
order.
F
Then
I
wanted
to
show
something
that
maybe
can
be
interesting
for
the
community
that,
thanks
to
moritz
having
now
a
dev
repo
for
our
helm,
charts,
we
can
now
deploy
any
pr
or
the
master
pr
with
the
helm
in
a
very
easy
way.
So
now,
in
the
documentation
under
install
master,
anybody
can
copy
the
write
commands
to
deploy
on
a
cluster,
their
current
images
and
then
the
final
thing.
I
worked
this
time,
even
though
it's
still
waiting
for
the
last
integration
test
is
adding
leader
election
to
ship
your
controller.
F
The
problem
here
is
that
dispatching
of
sequence
and
events
is
something
that
happens
periodically
and
that
cannot
be
done
by
multiple
instances
of
shipyard.
So
there
was
a
proof
of
concept
from
florian
that
was
trying
to
use
the
kubernetes
api
to
solve
this.
I
implemented
some
changes
and
tried
to
test
whether
or
not
this
worked
on
openshift.
F
It
can
work
on
openshift,
but
the
library
that
we
would
need
to
use
is
0.0.0
version
of
kubernetes,
which
doesn't
sound
very
safe.
So
instead
of
that,
I've
added
a
simple
disabled
leader
election
in
the
deployment,
so
that
we
can
avoid
having
multiple
replicas
of
shipyards
in
case
of
things
such
as
mini.
D
F
There
is
another
observation
I
did
some
extensive
tests
for
what
concerns
the
concurrency
problem
with
the
graceful
shutdown.
Basically,
we
have
around
30
60
second
delay
introduced
by
the
graceful
shutdown
at
shutdown
of
the
service.
So
if
this
happened
with
the
shipyard
controller
that
in
that
moment
is
in
charge,
what
will
happen
is
that
for
around
30
seconds
there
will
be
no
replica
of
shipyard
that
is
handling
the
sequences.
Nothing
will
be
lost
because
of
nuts.
Now
everything
is
fine,
but
there
may
be
this
delay
in
handling
of
the
sequences
yeah.
J
Okay,
so
I'm
currently
still
in
progress,
so
this
issue
is
still
in
progress,
so
pr
review
is
basically
in
progress.
The
new
feature
for
the
bridge
to
trigger
sequence.
So
I
would
just
shortly
demo
what
this
is
about.
J
So
it's
not
possible
really
from
the
bridge
to
directly
trigger
a
sequence
either
from
the
sequencer
screen.
There's
a
button
here
I
can
click
it
or
from
the
environment
directly.
So
it
will
look
like
this
in
the
environment
screen.
There's
a
button
trigger
new
sequence,
then
the
form
opens
it's
possible
to
trigger
a
delivery
and
evaluation
or
a
custom
sequence.
J
J
Also
values,
if
you
put
in
something
weird,
it
has
to
be
valid
json,
so
this
is
validated
and
if
I
click
now
on
trigger
sequence,
it's
sent
to
api
and
you
can
see
it's
already
started
here
in
the
sequence.
Screen
same
goes
for
evaluation,
so
service
stage,
and
I
will
so
you
can
now
choose
if
you
want
to
use
the
time
frame
with
hours
minutes
seconds
milliseconds
in
my
nanoseconds
also
you
can,
for
example,
I
would
put
it
for
five
for
the
last
five
minutes.
J
Please
start
the
12th,
so
this
is
parsed
into
the
iso
format
directly.
So
do
not
wonder
if
the
time
is
not
that
what
you
put
in
here
and
yeah
that's
triggered
yeah,
it's
not
waiting
because
delivery
is
not
finished
yet
you
can
also.
I
will
just
show
this
shortly
instead
of
the
time
frame.
You
start
end
time,
so
it's
basically
the
same.
J
Also,
if
I
put
in
here
the
start
date
for
24th
and
the
end
date
for
23rd,
you
will
get
an
error
that
this
start
date
has
to
be
before
the
end
date.
Basically,
so
we'll
just
do
it
like
that.
J
J
A
J
A
J
A
So
one
of
the
questions
would
be
whether
when
you
retrigger
the
sequence,
would
you
have
information
to
actually
what
parameters
are,
for
example,
somebody
will
eventually
implement
parameters
like
to
which
environment
to
deploy
etc,
and
when
you
retrigger
the
sequence,
you
might
make
a
mistake
and
retrigger
for
proper
and
improper
environment
so,
for
example,
in
jenkins,
if
you
have
ever
seen
a
rebuild
plugin
when
you,
which
is
basically
the
same
as
retriever
sequence,
to
some
extent,
when
you
open
it,
you
will
see
all
the
parameters
in
read-only
mode,
so
maybe
from
user
experience
standpoint
it's
something
that
one
would
expect
for
a
trigger
sequence
too.
A
It's
not
a
blocker
for
this
pull
request.
It's
rather.
B
A
B
J
K
Which
screw
is
the
right
one,
this
one?
Oh
okay,
so
let's
start
with
the
next
one.
Next
topics
in
the
bridge,
this
one
are
still
in
river
and
aren't
merged,
but
yeah
they
are
pretty
done.
So,
let's
go
to
the
first
one,
the
secret
selection,
in
a
bridge
before
we
just
had
a
static.
K
We
had
static
atoms
here,
which
is
just
these
three
and
they
are
hard
coded
in
the
bridge,
and
now
we
changed
it
that
the
scopes
here
are
fetched
from
the
back
end.
So
if
I
just
modify
our
config,
we
have
for
the
scopes
and
maybe
add
here
another
scope,
just
a
custom
scope
or
something.
K
Then
this
data
should
be
a
new
scope
should
be
shown
here.
This
will
take
some
time
as
I
realized,
so
I
guess
yeah.
Okay,
it's
already
here,
so
my
scope
and
yeah.
This
is
then
added
and
can
be
used
for
whatever.
K
K
If
there
are
questions,
I
will
go
to
the
next
one
for
the
open
id
log
in
and
log
out
after
bendy
flow.
We
had
some
problems
with
the
latest
release
because
we
added
some
content
security
policies.
K
The
logout
did
not
work
anymore.
If
there
was
an
n
session
endpoint
provided
by
the
ssu
provider,
because
we
execute
a
form,
a
post
to
the
end
session
endpoint,
and
this
one
was
not
possible
because
the
security
was
said
that
only
the
website
itself
can
be
used
for
such
rules,
because
they're
only
referring
to
self
but
then
yeah,
the
sso
provider,
endpoint
was
blocked.
K
K
So
if
the
company
provides
some
and
session
endpoint,
then
redirects
to
let's
say
microsoft,
this
will
not
be
possible
because
it's
a
different
one
than
the
and
session
endpoint,
and
this
can
is
just
fixed
in
a
way
that
you
can
add
another
environment
variable
where
you
determine
which
redirect
ui
should
be
should
be
allowed.
So,
let's
say
just
https
login.microsoft.com
and
that
is
then
allowed
and
the
logout
is
working
again
because
everything
is
now
allowed
for
the
form
action
here.
L
Thanks
yeah,
I'm
going
to
show
a
bug
fix
that
I
made
for
captain's
bridge
it's
regarding
the
score
that
is
displayed
in
the
heat
map.
So
let's
have
a
look
at
the
issue
here
yeah.
It
might
happen
that
the
score
is
89.68,
for
example,
and
in
this
case,
for
example,
it
would
yeah
be
a
warning
state.
L
But
since
we
have
used
rounded
values
here
in
the
heat
map
or
evaluation
result
site
yeah,
it
was
showing
90
is
smaller
than
90
and
results
warning
which
was
confusing,
and
so
now
we
change
this
behavior
so
that
we
round
the
result
or
the
score
if
there
is
if
there
are
decimals
to
two
decimal
points,
so
that
should
not
be
confusing
anymore.
Quick.
L
K
Is
it
up
rounded
or
rounded,
and
not
yeah.
L
Oh
yeah,
true,
we'll
have
to
check
that
one
that
special
case
yeah
could
be
that
it's
them
rounded
to
yeah
we'll
check.
Actually,
that
should
then
be
displayed
as
99.99.
A
I
think
oleg
yeah
just
have
a
quick
update
justice.
Second.
A
Yeah,
so
I
have
a
quick
update
about
google
summoner
code,
so,
first
of
all
we
applied,
I
announced
it
on
social
media.
Yesterday
I
will
put
my
information
channels
and
how
it
currently
looks
like.
So
we
have
this
my
screen.
A
Okay,
better
to
check
this
time
yeah,
so
we
created
a
landing
page
which
basically
provides
a
basic
guidelines
about
what
is
gsoc,
how
to
apply
its
contributor,
how
to
apply
the
mentor
and
summary
of
the
expectations.
So
what
you
are
currently
looking
on
for
is
more
project
ideas,
potential
mentors
and
potential
contributors.
You
already
had
a
few
potential
mentees
reaching
out
and
we
have
several
project
ideas
posted.
So,
for
example,
githubs
for
captain
is
a
scope
for
captain
and
guitars
cap,
67
and
69
for
captain
operator.
A
I
believe
then
improvement
captain
plugin
for
backstage,
so
the
project
which
currently
started
by
mauro
sala
and
it
was
discussed
with
smithy
18
and
a
few
other
contributors
at
meetings
and
improving
github
integrations
in
captain,
including
about
identification
service,
but
integrations
with
deployments
and
checks
apis,
so
that
captain,
who
pushes
statuses
right
inside
the
github
interfaces,
which
is
really
good
for
detox
and
other
things
and
also
integrations
with
p6
for
performance
testing
and
junk's
pipeline
executor
service.
A
Another
idea
we
have
in
mind
so
basically
what
we
discussed
multiple
times
the
work
of
the
documentation
side.
So,
as
we
discussed
with
mac,
haven't
documentation
site
engine
as
one
of
the
project
ideas,
so
I'm
yet
to
push
it
as
pull
request,
but
we
have
a
consensus
that
you
want
to
have
this
as
project
idea.
A
So
this
what
we
have
by
now,
what
we
need
is
more
potential
mentors
and
also,
if
you
see
any
potential
project
idea,
you
would
like
to
mentor
as
a
team.
It
may
be
everything
web
ui
with
javascript
angular.
It
might
be
low-level
logic,
it
might
be
automation,
including
hub
actions,
so
everything
which
is
called
it
could
be
a
project
idea
for
a
google
summer
of
code
and
everything
that
is
not
a
code
could
be
a
project
idea
for
lfx
mentorship,
for
example.
A
So
if
you're
potentially
interested
to
mentor
someone,
it's
a
good
opportunity
to
join
and
participate,
and
it's
also
great
for
team
building
and
for
community
collaboration.
So
everyone
would
be
welcome.
A
Well,
we
are
not
yet
part
of
that
we
have
just
applied
so
the
deadline
was
on
monday.
We
submitted
everything
the
next
step,
the
results,
a
description,
application
draft
which
describes
what
exactly
a
client
means.
So
if
you
go
captain,
we
have
started
the
gsoc
channel
again.
Everyone
is
welcome
to
join.
We
also
have
logger
in
drafts,
etc.
A
But
what
what's
important?
We
have
captain
application,
which
sets
some
expectation
how
the
project
would
operate.
I'm
happy
to
discuss
more
the
community
meeting
next
week,
so
I
see
that
we
are
running
out
of
time,
but
yeah
we
are
there
and
on
march
6th
we
will
know
which
mentoring
organizations
are
accepted,
to
increase
our
chances
of
being
accepted.
Of
course,
having
more
project
ideas
and
potential
mentors
would
be
something
beneficial
for
them,
so
call
for
action.
A
D
H
A
If
we
accepted,
we
are
guaranteed
to
have
at
least
one
project
if
we
select
amenti
for
that,
but
the
number
of
projects
we
would
get
depends
on
our
application.
So
it's
late
in
the
stages
in
april
after
we
receive
applications
from
potential
mean
amenities.
We
will
be
able
to
request
the
number
of
project
slots
we
want
to
get
and
then
google
will
make
a
decision
how
many
slots
we
actually
get.
But,
as
I
said
already
in
the
chat,
we
have
an
opportunity
to
also
run
projects
through
lfx
mentorship.
A
So
even
if
we
let's
say
get
two
slots
and
have
four
awesome,
maintenance
appliance
different
projects
and
we
have
mentorship
capacity,
which
is
also
essential
to
have
something
running.
Then
we
can
have
four
projects
because
yeah
shaking
some
trees
and
finding
the
budget
is
the
easiest
part
of
g.
So
the
most
complicated
part
is
for
forming
mentorship
teams
and
that's
why,
if
you're
interested,
you
would
be
more
than
welcome
to
participate.
D
D
Meeting
that
they
should
be
starting
to
come
to,
I
know
I
know
at
least
one
person
that's
out
on
our
slack
looking
to
chisok.
A
So
basically,
I
recommended
to
the
same
community
and
developer
meetings
right
now.
I
don't
see
a
point
in
creating
special
office
hours
for
google
summer
of
code.
We
can
create
that
if
there
is
enough
traction,
if
there
is
enough
questions,
but
right
now,
I
would
rather
prefer
potential
candidates
to
participate
in
the
same
community
meetings.
We
already
have
so
like
this
one.
H
A
A
B
G
A
I
Okay,
should
I
ask
my
question.
I
Yes,
so
I
think
I
have
two
questions.
My
first
question
is
apart
from
getting
more
contributors
to
the
cabin
project.
Do
we
have
any
objectives
outside
that
by
doing
g-shock.
A
Yeah
so
doing
jsoc
actually
has
multiple
purposes
and
it's
probably
not
the
most
popular
opinion,
but
in
my
opinion,
having
more
contributors
is
not
the
first
goal.
For
me,
one
of
the
biggest
goals
is
actually
community
bonding,
because
when
you
run
the
gsoc
project,
you
just
have
people
participating
it's
fun.
A
I
A
If
you
expect
the
goal
to
have
something
delivered,
let's
it's
nice
to
have,
we
had
in
jenkins
I've
been
a
mentor
since
2016.,
so
we
had
awesome
projects
delivered
there,
including
things
like
gitlab,
multi-branch,
source
integration,
with
github
checks,
etc.
All
of
these
projects
are
really
great
and
widely
adopted
by
users
by
now,
but
it's
not
guaranteed
but
having
great
experience
and
learning
opportunities
for
mentors
and
the
mentors
is
what
we
definitely
want
to
have.
A
I
Second
question
is:
are
these
project
ideas
limited
to
gsoc,
based
on
what
you
said,
it
seems
to
me
that
we're
looking
at
a
much
much
wider,
a
bigger
goal,
basically
just
not
cheese
up
so.
A
It
doesn't
make
sense
to
create
a
isolated
project,
sometimes
it's
nice,
but
well,
it's
basically
like
common
education
project.
The
main
goal
of
google
summer,
of
course,
is
showing
mentees
how
to
compete,
participate
in
open
source,
and
for
that
I
think
that
it's
important
to
actually
get
them
involved
in
the
community
as
much
as
possible.
So
what
we
want
to
have
projects
is
later.
I
Right,
I
think
I
think
the
idea
is
what
you're
saying
sounds
really
interesting.
First
of
all,
and
the
second
thing
is
I
I
feel
like
this
might
be
useful
outside
the
only
problem
is,
how
do
you
get
the
same
kind
of
structure
as
g-shock,
because
you
get
quality
candidates
in
g-shock
and
you
get
to
spend
a
lot
of
quality
time
with
them.
A
Yellow
commercial
show
you
a
few
things
so
first
of
all,
gsoc
is
far
from
being
the
only
program
in
the
world
in
terms
of
mentorship
outreach.
There
are
other
programs,
let's
say
outreach,
mozilla
hosts
somewhere.
Of
course,
alibaba
was
somehow
called
and
the
results
a
project
by
the
linux
foundation
for
lfx
mentorship.
A
So
maybe
you
have
heard
about
community
bridge
before
so
coming.
Rich,
actually,
even
captain
participated
there
a
few
times,
so
we
had
at
least
two
students
participating.
So
here
you
can
find
captain
and
you
can
see
familiar
faces
here
and
maybe
mac
remember
some
kid,
because
he
also
participated
in
the
jinx
community
but
yeah.
He
participated
in
lfx
for
captain.
So
we
have
an
opportunity
to
run
this
program
if
needed.
Basically,
lfx
mentorship
is
a
platform
as
a
service
and
all
the
guidelines
we
create
for
gsoc
all
the
practices.
I
I
I
do
not
have
the
idea
of
lfx
mentorship,
but
can
anyone
apply
to
lfx
mentorship,
even
if
they
are
not
in
education.
A
Yes
and
one
thing
to
keep
in
mind
that
this
year,
the
google
summer
of
program
also
changed
so
before
that
it
was
only
for
students
for
you're,
a
bachelor
master
or
phd
student.
But
now
things
changed,
so
you
can
see
that
it's
not
about
student
anymore.
A
They
use
new
term
gsoc
contributor
and
actually
it's
important
because
yeah
now
you
can
apply
not
as
a
student,
so
they
made
first
steps
during
coronavirus
because
not
everyone
are
able
to
attend
because
of
all
the
changes
and
they
went
even
forward
to
make
a
much
wider
audience
eligible.
So
I'm
just
looking
mentors
with
contributors.
D
Okay,
yeah
also
change,
there's
some
flexibility
in
the
duration
of
projects
this
year
that
there
was
not
before
there's.
A
A
So
these
are
key
changes,
so
we
expanded
eligibility.
A
Sizes,
so
basically
it
was
a
case
even
in
the
previous
year,
so
you
had
an
option
to
choose
and
now
well
last
year
they
just
reduced
everything
to
175
some
mentees
were
not
high
ips.
Some
organizations
were
not
happy
so
now
they
introduced
two
options
so
pretty
much
like
they
do
in
google
season
of
dogs
at
the
moment
and
the
same
flexibility
of
timing.
It
can
also
be
adjusted
but
yeah.
I
think
that
we
will
still
want
to
have
something
sizeable
so
extending
the
project
to
22
weeks
yeah.
A
A
Well,
it's
possible
the
problem
that,
even
for
three
months,
it's
really
difficult
to
keep
the
project
together,
because
some
life
events
always
get
into
picture
for
students.
D
A
A
The
fact
that,
at
least
in
jenkins,
we
saw
students
staying
maintaining
their
projects
for
a
longer
time,
so
maybe
one
of
the
approaches
would
be
to
have
basically
22
weeks,
but
not
of
equal
workload,
start
many
things
in
the
beginning
and
then
basically
extend
the
period
so
that
mentees
get
more
experience
with
maintenance
with
community
direction,
etc
already
delivered
code,
because
it's
also
a
great
practice-
and
this
is
what
google
summer,
of
course,
honestly
was
lacking
before
so.
I
I
I
think,
I
think
what
you
said
sounds
really
interesting,
and
the
second
thing
is,
as
as
a
first
as
an
individual
contributed
a
captain.
This
sounds
really
exciting
to
me.
Basically,
because.
I
To
work
with
someone
from
the
core
team
and
understanding
things
just
really
increases
the
pace
at
which
you
can
contribute
to
captain,
and
that
is
why,
in
my
head,
it
was
like.
Maybe
the
the
quicker
you
know,
expanded
beyond
gsa,
yeah.
A
So
whatever
we
do
here
is
portable
and
we
probably
should
expand,
but
google
summer
of
code
is
really
a
good
starting
point,
because
the
program
is
17
years
old.
By
now.
There
is
a
lot
of
documentation,
a
lot
of
experiences.
A
They
continuously
adjust
the
program.
They
collect.
Feedback
from
orcad
means
from
mentors
every
year,
other
retrospectives
at
the
mentor
summit.
So
as
framework
gsoc
is
really
great,
and
if
you
could
follow
this
framework
in
the
beginning,
I
think
it
would
be
nice.
Then
we
can
introduce
some
deviations
right.
I
Oh
yeah,
so
so
sorry.
A
I
I
don't
think
I'm
eligible
to
be
honest:
I've
never
really
applied
to
g-soc,
because
in
my
head
it
was
always
like
it's
for
students,
I'm
not
a
student
anymore.
I
have
a
full-time
job
and
I
contribute
a
captain
as
a
as
a
free
time
thing.
Basically,
so
what
I'll
be
interested
in
is
a
middle
ground,
basically
as
an
individual
contributor,
and
I'm
sure
there
are
many
people
like
me.
I
I
would
like
to
contribute
to
captain
I'm
just
thinking
if
the
the
the
learning
curve
of
contributing
to
captain
can
maybe
softened
if
we
have
something
similar
to
this
in
our
community.
The
only
problem
that
I
see
at
this
point
of
time
is
we
would
need
mentors
and
they
would
obviously
have
many
people
interested
in
contributing,
but
we
we
don't
know
how
much
time
do
we
spend,
because
many
of
them
might
not
contribute
at
all.
So
there's.
A
Yeah,
so
I
can
provide
some
insights
for
google
summer
of
code
from
previous
years.
First
of
all,
it
depends
on
the
house
how
stupid
the
student
is,
but
if
we
take
a
healthy
project
where
student
is
active,
where
the
student
is
engaging,
initially,
google
summer
of
code
was
a
full-time
job
for
students
for
several
months,
so
they
will
expect
to
spend
30
to
40
hours
per
week
and
in
order
to
make
it
possible
significant
membership
capacity
was
needed
so
up
to
10
hours
per
week.
A
But
the
good
point
that
this
time
can
be
split
across
multiple
mentors,
and
for
me
it's
a
must.
So
I
wouldn't
even
consider
running
a
gsoc
program
where
we
have
only
one
mentor
in
jenkins.
Our
recommendation
is
to
have
three
mentors
because
well,
usually
you
end
up
with
two
oldest
mentors
anyway,
because
some
people
drop
out
availability,
changes,
job
changes,
interests,
sometimes
change,
so
this
load
can
be
balanced,
but
still
it's
significant
time,
investment
from
contributors-
and
we
always
have
to
be
explicit.
I
How
much
time
are
we
looking
at
spending.
A
A
Sometimes
there
might
be
spikes,
for
example,
during
the
revolution,
phases,
etc.
But
the
workload
on
mentors
is
also
reduced
with
the
new
format.
I
Okay,
okay,
okay,
so
what
I
am
personally
interested
in
is
some
kind
of
a
mix,
because
at
this
point
of
time
I'm
looking
at
kubernetes
operators
and
that's
I'm
working
on
web
books
and
stuff
like
that
at
work,
and
that
that
is
something
which
interests
me.
So
I
was
actually
looking
at
the
captain
and
get
ups
issue,
and
I
was
like
looking
to
wrap
up
the
data
docking
and
maybe
start
looking
at
the
captain
and
get
ups
or
there's
one
more
integration.
I
A
Yeah
so
yeah,
it's
a
really
a
good
intent
and
yeah
if
you're
interested
to
participate.
You're
welcome
tour.
Still
it's
quite
a
big
commitment.
So
there
is
no
blame
if
someone
decides
not
to
not
at
all,
but
if
you
are
potentially
interested
it's
probably
time
to
at
least
maybe
proposed
project
id
or
just
join
one
of
pending
projects,
because
even
now,
if
you
add
yourself
as
a
potential
mentor,
you
make
no
commitment,
it's
just
initial
interest.
So
then
you
can
work
with
applicants.
A
Students
see
whether
you
get
excited
whether
you
want
to
work
with
each
other
and
then
sometime
in
april
during
project
selection.
Then
it's
time
when
you
would
have
to
make
a
final
decision
and
to
make
a
time
conference
before
that.
It's
basically
just
talking.
I
Okay,
all
right,
I
think
I
think
this
sounds
interesting.
I'm
just
worried
that
it
won't
cut
into
my
coding
time,
because
I
want
to
do
both
three
to
four
hours
per
week
sounds
reasonable
to
me
so
yeah.
I
think
this
sounds
interesting,
being
a
mentor
something
it
sounds
interesting
to
me.
I'm
just
worried
that
it
might
cut
into
the
coding
time
that
I
get
so
my
this.
A
It's
also
learning
opportunity
for
mentors.
You
generally
do
not
do
consulting
process,
so
one
of
the
first
things
would
be
to
ensure
that
the
student
bemendi
gets
introduced
to
the
community
because
usually
they
are
not
experienced
in
contributing
to
open
source
so
helping
them
how
to
understand
how
open
source
works,
how
community
works
getting
them
introduced
to
whatever
processes
best
practices.
This
would
be
one
of
the
first
things
and
then
you
can
organize
your
work
as
you
wish.
A
If
you
want
to
let
student
hack
and
review
pull
requests,
it's
also
fine.
You
just
need
to
agree
as
a
team,
how
you
press
it
it's
an
expectation
in
the
g-shock
that
the
most
of
code
is
produced
by
the
student,
but
it
doesn't
have
to
be
a
solo
work.
The
only
requirement
that
you
can
basically
evaluate
the
work
by
the
student.
So
if
you
want
to
have
a
joint
project,
etc,
it's
perfectly
fine.
A
Well,
we've
seen
a
lot
of
our
projects
and
whatever
happens
there
will
be
orca
means
already
to
help
with
advice
et
cetera,
so
you
won't
be
handling
that
alone
and
I've
been
a
g
supporter.
I've
been
for
five
years.
This
is
my
sixth
year
seventh
year
in
google
summer
of
code,
I
am
getting
gold
so
yeah.
I
did
see
a
lot
of
successful
projects,
a
lot
of
unsuccessful
projects
and
usually
I
spend
more
time
on
helping
other
teams
than
on
mentoring.
My
own
project.
A
Well,
this
is
also
what
keeps
me
excited
actually,
so
it's
not
like
I
grumble,
so
I
will
be
here
to
help
in
in
captain.
I
I
am
working
on
an
I'm
working
on
an
integration
meg,
so
I
was
thinking
yeah,
so
I
was
thinking
of
taking
up
so
one
of
the
project,
ideas,
the
captain
and
get
ups
was
something
I
was
thinking
of
looking
into
basically
after
I'm
wrapped
up
for
the
integration-
and
this
was
benson
mentioned
in
gsong-
and
the
entire
idea
sounded
really
interesting
to
me
and
I
was
like-
maybe
I
could
contribute
to
this
thing,
while
you
know
being
a
part
of
the
pisa
community
from
the
captains
and
yeah.
A
I
Right,
what
do
I
have
to
do?
I
I
have
two
three
questions.
What
do
I
have
to
do
if
I
have
to
apply
as
a
mentor?
It's
my
first
question.
A
So
after
resolution
it
gets
much
worse
on
github,
so
here
we
have
g-shock
landing.
I
will
post
a
message
and
there
is
joining
us
mentor.
Basically,
I
will
extend
it
based
on
the
feedback,
but
what
you
need
to
do
at
the
moment,
firstly,
is
stepping
up
in
2022
project
ideas.
So
there
is
a
list.
If
you
like
any
of
these
project
ids,
you
can
just
join
as
potential
mentor
or
if
you
want
to
propose
your
own
project
idea,
for
example,
extending
a
data
docs
integration
service.
I
A
Yeah,
well,
it's
nice!
So
basically
you
can
just
write
up
some
summaries.
So
what
this
project
idea
about
some
potential
use
cases,
deliverables
and
summary,
or
technologies
to
improve,
etc.
So
I'm
still
not
confident
about
this
format.
We
use
another
one
in
jenkins,
but
well
I
created
this
for
over
the
years,
but
it's
maybe
a
bit
too
heavy
for
captain
at
the
moment
so
jenkins,
for
example,
it
looks
like
that,
so
there
is
id
and
each
id
has
wide
description
links
mandatory,
quick
start,
mentoring,
beer
friendly
issues.
A
But
yeah
it
creates
a
lot
of
time
and
well.
We
have
extensive
guidelines
etc,
created
over
six
years
by
now.
So
I
didn't
want
to
create
such
a
complex
framework
in
captain
right
away,
because
I
think
that
it's
too
complex
even
for
jenkins
at
the
moment,
but
I'm
definitely
taking
some
best
practices.
A
So
what
I
suggest
at
the
moment,
you
just
have
a
kind
of
summary
that
describes
what
to
do.
A
If
you
have
ideas
about
good,
first
issues,
getting
started,
etc,
feel
free
to
also
put
it,
but
we
will
be
putting
everything
incrementally
and
yeah.
We
can
improve
the
things
together,
so
we
don't
have
to
write
a
massive
application
right
away.
A
I
Right,
do
we
have
the
link
for
this.
A
Yeah
I'll
put
the
links
to
the
chat
just
after
this
call.
I
A
It's
active
phase,
but
actually
so
g-stock
has
already
started
for
administ
for
many
organizations,
because
many
students
actually
reach
out
early
before
the
official
application
phase
starts
and
even
in
captain
we
had
a
few
students
reaching
out
to
mentees.
Sorry,
I
still
keep
sending
students
but
yeah.
Now
it's
mantis
and
yeah.
So
before
these
12
weeks
of
coding
start
there
is
application
phase
when
potential
mentees
explore
project
ideas
and
come
up
with
a
project
proposals.
A
So
again
we
don't
have
we
don't
write
project
descriptions.
We
just
write
project
ideas
which
describes
what
we
would
like
to
see
in
principle
and
then
potential
mentees
should
explore
these
ideas
and
create
a
proposal
which
is
much
more
to
the
ground,
including
deliverables,
scopes,
time,
matching
timeline,
etc.
I
Okay,
and,
and
when
is
it
starting,
is
it
starting
in
me.
A
I
think
well
generally,
it
would
start
soon
after
you
submit
the
project
idea,
because
then
somebody
may
discover
that
and
reach
out
in
channels,
but
it
would
be
something
that's
important.
It
would
be
not
much
time
consuming,
but
yeah.
Basically,
it
happens
like
in
open
source
potential
contributors
may
reach
out
at
any
moment,
so
the
same
would
be
with
jsok.
I
A
No,
you
get
actually
them
before
so
active
students
start
exploring
google
summer
of
code
organizations
in
december,
sometimes
even
earlier
so
long
before
g-stock
applications,
officially
open
and
long
before
organizations
honest
because,
for
example,
if
you
want
to
participate
in
jenkins,
you
know
that
this
is
established
organization
with
good
track
of
participation.
A
So
it's
most
likely
to
be
accepted
to
google
summer
of
code.
So
you
can
invest
some
time
and
explore
in
advance
to
have
advantage
against
other
applications
and
the
same
for
captain.
So
somebody
may
explore
it
right
now
and
once
the
organizations
announced
on
march
6,
if
captain
is
accepted,
then
they
would
be
research
in
people
exploring
and
probably
contributing.
I
Okay,
so
so
we
would
have
students
were
accepted
in
gsoc
coming
to
captain
whoever's.
Students
which
are
interest
who
are
interested
in
captain
come
to
captain
around
end
of
ma
april
is
is
what
is
happening?
No,
it
would
be
early
much,
oh
end,
of
course,
yeah.
A
But
it
would
be
a
rather
relaxed
face,
so
somebody
asks
some
questions.
Maybe
it
tries
to
submitting
a
newcomer
friendly
issue,
pull
requests,
maybe
asking
for
some
feedback
in
their
proposal
drafts,
but
they're
active
in
and
time
consuming
phase
would
start
in
only
in
may
when
there
is
when
the
project
is
announced
before
that
it
will
be
very
relaxed.
A
I
I
think
that
answers
all
of
my
questions.
I
will
update
the
the
markdown
on
the
link
that
you
shared
with
me
and
yeah.
Just
please
take
it
as
a
tentative
thing,
because
I
don't
know
what's
going
to
happen,
if
you
love
me
well
I'll
just
add
my.
A
I
A
We
live
in
a
real
world.
Nobody
knows
what's
going
to
happen,
especially
with
all
the
news
happening
last
week
and
yeah
in
2020
when
the
the
world
was
nuked
by
coverage.
Restrictions
looked
down.
Well,
I
appreciated
everyone
who
participated.
Despite
all
these
events,
so
everything
was
a
mess
because
applications
was
ms.
Nobody
knew
how
the
things
are
going
on.
All
travel
was
cancelled.
A
Some
of
the
students
were
unable
to
participate
because
the
study
schedule
changed.
Some
students
were
unable
to
participate
because
so
they
were
no
longer
able
to
go
to
their
country
where
they
study
so
yeah.
It
was
a
lot
of
logistics,
a
lot
of
edge
cases
to
focus
on,
and
I
appreciate
a
lot
of
help.
We
got
from
the
resolve.
Organizers
at
google,
open
source
team,
augmented
circuit
means
students.
A
A
Yeah
so
yeah
what
I
will
probably
do.
I
will
split
this
video
to
two
parts.
So
one
part
will
be
community
meeting
up
to
demos.
Another
one
will
be
just
introduction
to
google
summer,
of
course,
so
I
will
post
it
as
a
separate
video
and
announce
it
in
channels
so
that
everyone
can
watch
if
they
want,
and
if
you
want
to
put
some
notes,
your
experiences
to
documentation.
Please
go
ahead
because,
okay,
I
just
created
something
really
simple
for
now
and
we'll
be
expanding
over
time.