►
From YouTube: GSoC 2023 Office Hours - March 9, 2023
Description
Special session about:
- Building Android Apps with Jenkins
- Docker-based Jenkins quickstart examples
- Jenkins Configuration as Code (JCasC) drift detector
A
That
we
are
okay
good
afternoon,
good
morning,
good
night
to
everybody
around
the
world.
This
is
March
9th
Google
summer
of
Code
office
hour,
so
welcome
everybody.
Today,
I
have
some
a
general
announcements
to
do
and
we're
going
to
discuss
and
explain
three
projects
more
in
detail.
So
there
will
be
presentation
and
we'll
also
answer
some
questions,
so
the
projects
will
be
building
Android
apps
with
Jenkins
a
Docker
based
Jenkins,
quick
start
example,
and
then
Jenkins
configuration
is
code.
The
drift
deduct
detector.
A
So,
first
of
all,
we
with
well
I
wanted
to
say
with
spring,
but
with
this
period
around
the
world,
the
world's
makes
a
big
hiccup
because
of
the
various
time
changes
in
parts
of
the
world.
So
some
parts,
starting
with
the
the
west
coast
of
the
United
States
or
even
the
whole
United
States,
is
changing
time
and
moving
to
Summertime.
A
From
next
week
on
to
adapt
to
that
situation,
we're
going
to
start
one
hour
earlier
everywhere.
So
this
means
that
for
India
and
China,
as
there
is
no
time
change
there,
we'll
start
one
hour
earlier,
which
is
good
because
it's
late
for
you
this
time
so
meeting
time
will
be
15
hours
UTC,
so
3,
P.M,
UTC
I,
didn't
make
the
mess
for
India
in
China
at
that
time.
So
be
aware
of
that.
A
The
other
thing
that
I
wanted
to
share
remind
everybody
is
that,
or
students
or
g-so
contribute
contributors.
It's
now
plenty
time
to
submit
your
drafts
of
your
proposals
were
getting
quite
a
couple
now.
I
think
we're
running
about
six
six
or
seven
proposals
now
they're
being
reviewed,
continue
working
on
that
it's
important
give
some
time
for
the
mentors
to
review
the
documents
so
be
patient,
but.
A
A
You
will
still
be
able
to
change
and
update
this
proposal
until
the
deadline,
but
at
least
you
have
already
one
proposal
recorded,
so
you
can
start
whenever
you,
you
feel
ready.
The
door
closed.
Definitely
April,
4th
April
4th
is
the
deadline,
everybody
pulls
their
debt
and
nothing
moves
anymore.
No.
A
Yes,
indeed,
404.
so
deadline
very
important
for
all
the
students,
energy,
so
contributors,
then
we
will
have
one
month
to
review
and
rank
all
the
proposals
set
up
our
proposals.
So
we
will
then
start
we
mentors
we'll
then
start
our
homework.
Focus
on
April
4th
is
the
final
date
where
you
can
submit
your
proposals.
A
A
A
C
A
So,
currently,
there's
no
ranking
in
the
review.
If
I
understood
your
question
correctly,
this
is
a
review
where
we
help
to
improve
your
your
proposal
and
it's
not
only
mentors
mentors
I
chase
them,
because
I
want
every
month
mentor
to
to
review
them
the
the
best
they
can,
but
there's
no
judgment
at
this
stage,
so
we
really
tell
hey
this
is
unclear.
So
this
you
should
rework.
A
A
Is
this
particular
candidate
or
able
to
achieve
successfully
the
goals
of
the
project
during
the
summer?
Will
he
achieve
with
the
mentors
what
he
promised
that
he
will
that
he
will
do
without
having
the
mentor
having
to
spend
20
20
hours
a
week
or
40
hours
a
week
to
help
or
to
to
mentor
to
Mentor
him?
I,
don't
know
if
this
is
clear
enough
for
you.
The
criterias
are
explained
in
various
documents.
I
can
point
you
to
that.
C
A
B
Yeah
Android
Med,
why
not
the
guy,
who
has
to
build
Android
apps
with
various
CI
CD
tools
and
the
latest
being
Jenkins,
of
course,
so
I
don't
know
if
you
still
have
somewhere
Mark
that
description
of
the
project.
Of
course
I
know
it,
but
it's
better
if
I
have
a
support
somewhere.
So
do
you
plan
on
sharing
your
screen
or
should
I
do
with
mine?
Can
you
do
it
with
yours?
I'll
try
to
if
I
ever
can
find
the
ride
link.
What
do
I
do,
though,.
C
B
B
Okay,
so
I
can't
see
you
anymore,
but
I
guess
people
can
see
my
screen.
Yes,
that's
cool,
so
building
Android
apps
with
Jenkins.
Why
not?
The
thing
is
when
you
search
on
the
Jenkins
I
o
website
about
Android
and
Jenkins.
You
don't
find
much
if
any,
and
that
doesn't
mean
it's
not
possible.
That
means
that
not
that
many
people
have
contributed
to
the
documentation
or
to
proof
of
concept,
to
show
that
yes,
it's
possible
to
do
with
Jenkins.
B
So
the
thing
is
when
you
want
to
build
a
mobile
app
with
CI,
it's
a
different
kind
of
CI
CD
that
what
you
have
for
other
stacks,
for
example,
for
the
web
or
for
embedded
boards
or
things
like
that.
It's
different
you
have
to
have
the
cicd
mindset,
but
you
also
have
to
have
the
mobile
developer
mindset
and
the
two
of
them
sometimes
Collide.
They
don't
merge
easily.
B
So,
most
of
the
time,
what
I've
seen
in
my
previous
jobs
is
that
mobile
developers
do
their
CI
more
or
less
by
themselves,
and
it
takes
them
10
to
20
percent
of
the
time
and
they're
not
really
efficient
at
that.
On
the
other
hand,
when
people
who
do
only
CI
try
to
do
mobile,
CI
CD,
it's
kind
of
difficult
for
them,
because
they
don't
really
know
the
ecosystems
which
are
tightly
constrained
by
Google
and
Apple
for
iOS.
B
So
it's
difficult
to
have
something
efficient
and
when
I
came
to
Jenkins
I
made
a
proof
of
concept
with
Jenkins
which
allows
to
build
and
release
Android
apps
easily,
with
Jenkins
on
lots
of
platforms
on
Windows
Linux
vagrant
and
on
various
CPU
architectures.
So
it's
something
that's
worth,
but
it's
only
a
proof
of
concept
and
maybe
it's
still
too
complicated
for
people
to
use
it
in
their
production.
You
know
so
I
was
wondering
if
I
could
get
some
help
for
the
Google
summer
of
code
in
order
to
propose
something
that
would
work
better.
B
That
would
Target
more
massive
audience,
not
only
people
who
are
familiar
with
Docker
compose
rails,
CPU
architectures,
but
yeah
I
would
like
that.
Someone
who
already
has
done
something
with
Android
development
and
somebody
who
is
also
interested
in
CI
CD
and
who
knows
some
not
everything
but
some
of
the
particularies,
the
tips
and
tricks
that
you
need
to
build
and
release
to
build
tests
and
release
an
Android
application
with
a
CI
CD.
So
if
you
don't
know
Jenkins,
yet
that's
not
really
a
problem.
B
We
still
have
some
time
to
learn
but
I'd
like
that.
The
people
who
want
to
tackle
with
this
project
know
already
about
Android,
because
we
won't
have
time
to
learn
Android
Development
before
starting
with
cicd
with
Jenkins,
so
so
yeah.
What
I'd
like
in
the
end
is
that
we
have
some
documentation
on
jenkins.io,
not
a
blog
post,
real
documentation
on
how
to
start
Android
development
with
Jenkins
and
also
like
that.
We
have
a
repo
somewhere
with
a
working,
not
proof
of
concept,
something
that
could
work
even
in
production.
So
that's
maybe
a
little
bit
ambitious.
B
So,
yes,
first
of
all
documentation
so
that
when
people
will
search
for
Andrew
Jenkins
boom,
they
will
get
some
links
on
jenkins.io
website
and
links
to
things
tutorials
how
to
whatever
that
do,
work
for
them
that
they
can
experiment
at
home
and
then
more
detailed
documentation
that
will
help
people
to
get
that
into
a
production,
Jenkins
controller,
at
least
that's
a
goal.
So,
of
course
you
will
have
to
study
and
improve
the
Jenkins
technical
architecture.
B
If
you
don't
know
it
yet
that's
not
a
problem
for
the
time
being,
but
you
have
to
know
you
have
to
have
the
basics
of
Android
application
development,
because
it's
very
very
specific,
at
least
to
me
it's
very
very
specific
when
you
want
to
make
a
release
on
Google,
Play
Store,
for
example,
it's
kind
of
difficult
and
if
you
never
ever
have
done
it.
That
will
take
some
time
to
learn
and
I'm,
not
so
sure
that
we
will
have
the
time
to
study
that.
B
So
if
you
think
that
you
could
do
that
before
it
starts
for
real,
that's
okay,
but
if
you
never
ever
have
touched
Android
development,
maybe
you
should
find
another
project,
I
hope
I'm,
not
too
rude
or
gross.
When
things
happen,
that's
the
truth.
It's
kind
of
it
will
take
some
time.
You
know
to
get
accustomed
with
Android
development,
so
yeah.
Another
thing.
Of
course
you
have
to
know.
B
Even
if
you
don't
know
Jenkins,
you
have
to
know
basics
of
CI
principle
and
practice,
and
it
would
be
nice
if
you
already
knew
the
basics
of
Docker,
because
that's
the
way
I
Envision
it
or
even
if
we
find
something
else,
another
way
of
touring
and
trade
build
with
Jenkins
without
Docker.
The
thing
is
proof
of
concept
have
made
is
using
Docker.
So
as
this
will
be
our
starting
point,
I'm
afraid
you
will
have
to
know
things
about
Docker
and
Docker
compose
I,
think
I'll,
say
everything.
A
A
A
A
So
there
we
are
okay,
so
I
think
all
of
you
experienced
that
very
frustrating
feeling
and
say:
I've
been
told
that
Jenkins
is
a
super
duper
tool
and
I
want
to
try.
It
I
want
to
experiment
it
and
see
how
it
works
and
then
start
the
difficulties,
because
it's
either
you
have
all
the
machines
that
you
that
you
want.
A
You
have
Linux
machines
or
or
whatever,
there's
not
a
simple
way
in
a
couple
of
minutes
to
start
a
Jenkins
environment
so
that
you
can
start
playing
around
with
it
and
having
a
good
feel
of
its
major
features
and
that
you
have
a
good
demo
environment
is.
This
is
what
quick
Stars
Squid
quick
starts
are
for,
meaning
that
you
have
a
setup.
A
A
Not
requiring
complicated
knowledge,
creating
SSL
keys
or
or
these
kind
of
things
everything
should
be
in
and
then
probably
properly
document
it
and
automatic.
The
updating
should
be
automated
should
be
automated
in
the
sense
that
when
somebody
wants
to
try
the
Quick
Start,
he
always
has
the
latest
version
available
and
running
so
that
we
have
some
kind
of
CI
of
this
CI
environment
in
that
it
that
we
work.
A
So
this
is
a
very
important
project,
because
this
is
something
that's
missing
and
I.
Think
that
probably
everybody
here
around
the
table
knows
that
it's
frustrating
the
first
experience
is
frustrating
to
get
something
running
and
but
you're
proud
afterwards,
when
you
have
done
it,
but
this
is
unnecessary
energy.
This
is
our
showcase
and
we
should
improve
that.
B
Yeah
I
can
remember
the
frustration
when
starting
with
Jenkins
and
Docker,
and
yes,
of
course,
I
was
kind
of
proud
once
I
got
it
working,
but
I
didn't
know
why
it
worked,
and
that
was
even
more
more
frustration
in
the
official
documentation
you
have
installing
Jenkins,
starting
with
Docker
and
so
on.
So
you
have
a
commands
that
you
can
replicate
so
that
it
works
on
your
computer,
but
it's
unnecessarily
complicated
and
it
makes
you
do
something,
create
a
Docker
file
and
for
no
real
reasonable
reason,
in
fact,
so
that's
kind
of
disturbing.
B
And
yes,
we
have
to
change
that
and
the
idea
was
to
create
a
set
of
different
Docker
Compass
file
for
different
need,
for
example,
the
things
I
just
shown
with
Android,
for
example,
there
could
be
an
example
for
Android
with
a
a
drinking
controller,
an
Android
agent,
a
known
Android
agent.
We
could
have
some
examples
to
build
some
I,
don't
know,
embedded
software.
We
could
have
something
to
build
some
Java
code.
B
We
could
have
something
that
produces
a
website
and
so
on
and
so
on
and
so
on,
and
the
magic
in
that
would
be
that
every
Docker
Compass
file
would
be
then
tested
regularly
on
ci.genkins.io,
so
that
anybody
wanted
to
start
is
assured
that
it
will
work.
It
has
been
tested
and
every
time
we
change
something.
We
know
that
it
works
and
that's
wonderful
and
in
the
end
it's
written.
It
could
also
run
thanks
to
gitbot.
So
I
don't
know
if
everybody
is.
Oh
sorry
go
ahead,
remote,
no.
A
No
you're
you're,
saying
the
magic
word:
I
was
keeping
it
for
the
end.
I
I
was
go
ahead,
but
git
pod
is,
or
these
kind
of
environments
are,
are
very
promising
and
give
the
necessary
a
standardized
environment
or
or
infrastructure
agnostic
environment.
So
my
proposal
for
for
that
would
be
to
look
into
gitpod,
which
has
Docker
compose
abilities
enough
memory
enough
CPU
and
and
to
use
that
for
a
demo.
B
A
Spaces
has
it
or
it
it's
part
of
the
complete
package
that
you
can
do
and
thank
you
for
Bruno
for
raising
it.
I
was
keeping
it
for
for
the
end
or
I.
Sorry,
I
didn't
want
to
throw
because
well
one
of
the
problems
that
needs
to
be
thought
about
is
where
gitpod
is.
Currently,
there
is
a
free
tier
in
it,
so
it's
very
open
source
oriented
what
happens
if
they
need
to
make
money
and
say:
well,
we
don't
give
it
for
free
anymore.
B
You
introduced
me
to
gitbull
I
think
so.
That's
okay!
The
thing
is
yes,
we
may
have
vendor
lock
two
years
from
now
we
never
know
and
even
same
with
GitHub
called
spaces.
You
never
know
what
could
happen
two
years
from
now
and
GitHub
and
Jenkins
already
intertwined.
If
ever
we
would
have
to
leave
GitHub
wow.
That
would
be
difficult
because
it
helps
us
immensely
every
day
for
with
GitHub
action,
for
example,
but
anyhow,
that's
not
the
subject.
B
The
thing
that
we
have
to
remember
is
that
we
would
like
a
set
of
well
working
examples
for
people
to
be
able
to
start
easily,
with
Jenkins
and
Docker
on
various
platforms
on
their
laptops,
with
Windows
Mac
OS,
even
on
Mac
M1
for
the
most
fortunate
ones.
Why
not
off?
With
Raspberry
Pi,
for
example?
That's
also
possible,
that's
one
of
the
aims,
so
a
lot,
not
all
of
them,
but
lots
of
the
examples
should
be
able
to
work
just
about
everywhere,
so
that
anybody
could
start
with
Jenkins
easily.
That's
a
goal.
A
A
Okay,
the
last
subject:
Jenkins
configuration
is
called
Drift
detector,
so
gcask
for
those
who've
unfamiliar
with.
That
is
a
mechanism
where
you
can
describe
the
configuration
of
a
Jenkins
environment
as
code,
meaning
you
have
a
text
file
where
you
define
the
various
parameters,
the
various
configuration
for
plugins
for
the
security,
it's
rich
and
it's
a
well
it's
evolving.
This
is
really
a
a
great
great
feature.
A
The
configuration
is
to
do
it
through
the
code
and
I've
seen
people
pushing
that
very
very
far.
They
even
disabled
the
administrative
consoles
and
did
not
allow
people
to
to
go
through
it.
Now.
This
is
a
little
bit
extreme,
because
in
certain
cases
you
need
to
be
able
to
experiment
or
to
fix
a
production
incident.
A
So
if
you
you
configure
it
once
and
then
you
revisit
this
configuration,
so
jcast
configuration
two
months
afterwards,
there
can
be
have
been
a
lot
of
things
changed
that
you're
not
even
aware
because
there
may
be
another
system
and
they
change
these
parameters.
How
do
you
know?
How
can
you
control
it?
How
can
you
monitor
that
incorrectly?
A
It
can
be
done
manually.
So
you
need
to
export
the
configuration
and
start
to
compare
by
hand.
It
doesn't
come
in
the
right
order
and
there
are
there
a
lot
of
problems,
hurdles
potholes
to
go
that
way,
but
this
is
a
feature
that
many
people
told
me
I
say:
hey.
That
would
be
handy
so
that
I
can
quickly
verify.
A
Did
my
configuration
diverge
drift
from
what
was
charted?
What
is
written
in
the
document?
What
has
changed?
Oh
this
is
the
detail.
I
know
there
was
a
temporary
fix.
I
can
reapply
the
configuration.
So
this
is
the
idea
about
this
drift
detector.
It
is
writing
or
creating
a
tool
that
you
can
run
that
has
some
command
command
line
parameters
that
will
spit
out
and
tell
you
either
your
configurations
are
equal
what's
running
and
what
you
defined.
A
Oh
no,
you
have
this
and
this
and
this
that's
different
and
the
different
ways
to
to
to
present
it.
So
you
can
read
in
the
project
proposal.
There's
no,
no
additional
value
for
me
to
to
read
that
I
try
to
explain
as
clearly
as
possible
what
it
is
why
it's
important
the
tool
can
be
written
in
whatever
you
want
does
not
need
to
be
Java,
because
it
would
work
with
the
command
line
interface
with
the
available
interfaces.
A
There's
no
limit
the
sky
is
the
limit.
The
way
you
can
start
going
into
the
internals
or
you
can
use
existing
application
interfaces
to
read,
you
can
do
it
in
Java.
You
can
use
a
modern
language
like
golang
or
or
even
rust.
If
you
want,
you
just
need
to
have
enough
people
that
know
that
can
help
you
improve
your
tool
and
and
maintain
it.
So
there
are
no
limit
there.
A
Other
doubts
or
questions
about
this
project
yeah
there
is
one
thing
I
need
to
share
with
you-
is
that
seeing
the
number
of
proposals
seen
my
personal
workload,
I
will
not
be
able
to
Mentor
this
project.
So
it's
it's
you're
not
going
to
be
disappointed
by
that.
Although
I
like
this
feature,
I
will
not
be
able
to
commit
and
we
might
run
into
issues
finding
good
mentors
for
that
that
project.
D
A
A
So
I've
seen
there
were
questions
also
on
on
guitar
and
so
I
I
was
not
able
to
answer
them
correctly.
There's
some
exploration
to
be
done
there
and
there's
no
simple
path
or
or
EV
or
obvious
path
to
solve
it.
I
had
some
ideas
in
the
time.
I
thought
writing
that
myself,
but
so
first
tip
learn
Jenkins
configuration
as
code
all
the
details.
Do
it
do
it
second,
then
learn
to
do
it
with
the
command
line
interface,
not
with
the
UI,
and
once
you
can
reproduce
that
start
thinking.
A
D
D
C
D
A
So
what
we
strongly
advise,
G-Shock
contributors
or
candidates,
is
that
they
get
acquainted
with
the
way
the
community
works,
the
principle
of
pull
requests
and
that
you
do
actual
work,
actual
contributions
to
Jenkins
in
any
form
and
you
can
use
or
you
should
use
these
contribution
and
reference
them
in
your
a
proposal
and
saying
I
know
what
a
pull
request
is
I
know
how-
and
this
is
what
needs
to
be
demonstrated
that
you
know
how
the
Jenkins
project
handles
pull
requests.
What
is
the
review
process?
A
What
are
the
communication
channels
used
to
discuss
them
that
you
demonstrate
also
that
you
have
enough
understanding
of
the
of
the
the
the
applications
architecture
that
you
know?
What
are
the
different
components?
What's
an
agent
controller
that
that
you
know
all
that
this
will
all
help
you
to
demonstrate
to
us
when
we
do
the
ranking
and
say
well,
okay,
this
particular
candidate
has
good
experience.
We
will
not
have
to
spend
time
to
teach
him
these
basic
elements.
A
We
don't
need
to
teach
him
how
git
works,
for
instance,
or
how
to
do
a
pull
request
on
GitHub,
because
this
is
extra
cost
that
we
will
have
to
take
on
on
ourselves
and
and
time
that
will
not
be
spent
on
actually
doing
the
project
is.
Is
that
the
answer
to
the
question
you
had
or
did
I
answer?
Another
question.
A
Okay
and
by
experience
by
doing
two
pull
requests,
you
get
already
enough
terrain
experience
to
know
what
you're
talking
about
and
so,
but
we're
we're
not
going
to
reject
or
not
read
your
proposal.
If
you
didn't
do
it,
but
the
more
you
do,
the
better
we,
you
know
the
better,
your
integrated
in
the
community,
the
better
your
chances
are.
A
A
So,
while
you
think
about
other
questions
so
start
working
on
your
Drive
proposal,
draft
now
early
it
would
the
community
will
be
able
to
guide
you
and
tell
you.
Oh,
this
information
is
missing
or
we
don't
understand
that
part
of
your
reasoning.
It's
part
of
the
learning
process,
so
it
open
source
is
all
about
communicating
and
we
want
to
help
you
to
communicate
your
ideas
who
you
are.
We
want
to
help
you
to
communicate
the
best
available
manner,
and
this
is
the
review
process.
A
People
sections
that
we
don't
understand
and
and
are
not
to
the
point
and
you've
wasted
a
lot
of
energy
with
that,
so
submit
them
as
early
as
you
can-
and
it's
not
a
judgment
so
be
to
remember
that
we're
not
judging
your
proposal
and
and
old
proposals
are
good.
Everybody
is
good.
We
just
at
a
certain
point,
need
to
order
them
to
rank
them.
A
A
A
E
A
A
It's
not
because
you're
you're
working
on
Jenkins-
and
we
most
welcome
you
in
on
the
Jenkins
project.
You
can
submit
ideas
on
other
projects,
so
other
g-soc
organization
is
the
correct
word
for
that,
but
just
remember
a
whole.
Listen
to
an
old
man,
the
more
you
spread,
your
efforts,
the
weaker.
The
results
will
be.
A
A
A
A
A
A
On
that
another
question
that
I
see
so
first
project,
so
I
guess
this
is
the
first
g-soc
organization,
you're
you're
approaching
so
I
think
it's
a
very
good
idea
to
start
with
Jenkins
we're
very,
very
friendly,
and
this
is
a
very
useful
product
to
learn
about
their
other
organizations
available
on
Google
summer
of
code.
But
I
need
to
be
dishonest
here
and
say
that
Jenkins
is
the
best.
A
A
My
hunches
no
more
than
six.
If
we
go
over
six,
that
would
be
a
miracle.
So,
but
we
it's
it's,
a
handful
of
candidates
will
make.
Last
year
we
had
four
projects
running,
so
it's
very
competitive,
but
you
can't
prepare,
you
can
learn
and-
and
you
can
do
it
for
a
long-
a
long
run
long
period.
You
can
already
start
preparing
for
2024..
A
C
A
A
B
A
A
There's
no
easy
help.
The
only
thing
you
can
you
can
do
to
learn
that
is.
Take
your
helmet.
Take
your
your
rucksack
and
get
into
the
jungle
and
start
learning
starting.
This
is
why
we
recommend
doing
pull
requests
in
in
starting
to
dig
into
the
the
core
of
the
code
and
start
learning
they're.
The
usual
tips
are
book
as
the
tests
run,
the
tests
with
a
debugger
step
through
the
code
understand
the
the
various
components.
So
these
tips
are
General.
A
A
So
they're
I
have
a
question
about
the
length
of
the
docker
based
Jenkins,
quick
start
example:
project.
Now
we're
we're
expecting
no
I,
don't
don't
have
the
calculator.
My
screen
here
is
locked
in
a
weird
state,
so
the
project
for
the
docker
based
Jenkins
quick
start
is
1727
hours
project
in
there.
There
is
time
for
writing
to
have
meetings.
There
is
time
to
write
the
documentation.
There
is
time
to
do
the
testing.
A
A
A
A
What
could
go
wrong
if
this
goes
wrong,
then
I'll
do
that
or
these
kind
of
things,
but
we
can
discuss
that
in
other
office
hours.
If
you
want
to
come
back
next
week,
don't
forget
that
we're
going
to
start
one
hour
earlier
in
some
parts
of
the
world
we
we
can
discuss
these
strategies
and
and
we're
here
it's
open.
We
can
take
a
virtual
cup
of
coffee
together
and
discuss
your
doubts
or
your
questions
and
and
people
are
there
to
help
you.
A
Okay,
we're
reaching
the
top
of
the
hour,
so
I
didn't
think
that
we
would
fill
an
hour,
but
I
managed
to
talk
so
much
with
the
help
of
Elisa
and
Bruno.
Talk
that
much
to
fill
the
complete
hours.
Don't
hesitate
to
ask
your
question
lie
in
it.
It's
an
adventure,
don't
be
disappointed
if
you
don't
make
it
this
year,
at
least
in
the
process.
You
will
learn
something
useful.