►
From YouTube: Cells: Demo and Office hours 2023-09-07
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
Okay,
so
yeah
for
today,
I
have
some
demo,
which
is
basically
I,
want
to
demo
the
work
that
I
have
been
doing
regarding
the
GDK,
with
with
users
being
able
to
log
in
on
two
cells.
Let
me
start
because
it's
gonna
take
some
time
to
to
set
up
okay.
Let
me
share
my
screen.
A
Okay,
so
basically
I'm
gonna,
now,
like
I,
have
one
GDK
running
I,
hope
that
the
front
is
big
enough.
I
have
only
one
GTA
running
I
I
already
like
have
the
branches
that,
of
course
they
haven't
been
merged.
Yet
I
will
follow
the
same
steps
that
Tonk
has
done
before
undocumented
regarding
sitting
sitting
up
the
second
GDK
and
we'll
see
how
how
this
works
and
I
only
tested
it
once
I
hope
it's
gonna
work
again
and
so
yeah
the
the
same
steps
now.
A
Okay,
that
was
first
and
then
this
one
yeah
and
then
like
running
the
script
that
I
I
already
modified.
Okay,
now
I
did
something
wrong.
I
should
have
been
on.
The
I
should
have
been
on
the
on
the
root
of
GDK.
Sorry.
A
And
then
this
environment
variable
here
and
then
running
the
script
which
I
mentioned,
like
I,
modified
to
set
up
to
modify
gdk2,
and
this
is
going
to
take
some
time
to
it's.
Gonna
take
some
time
to
to
run
so
in
the
meanwhile.
I
will
go
over
the
details.
A
little
bit
I
commented
on
that
before,
but
maybe,
while
it's
running
I
will
take
some
time
to
just
explain
what
I
have
been
doing.
I
promised
before
that
I
wanted
to
do
diagrams
for
the
sales
project
and
put
them
on
the
blueprint.
A
So
I
was
practicing
a
little
bit
with
the
with
the
work
on
the
GDK
deal.
Hopefully
we
can
take
that
and
modify
it
later
for
the
how
it's
going
to
look
on
production.
Basically,
that's
that's
a
like
that's.
Basically,
the
idea
like
gdk1
is
running
two
databases:
okay,
CR
database
and
Main
database,
the
main
database.
A
We
are
studying
the
tables
into
gitlab,
Main
and
gitlab
Main
cluster
White
I
mean,
of
course,
it's
going
to
be
called
gitlab
main
cells,
but
we
are
still
classifying
the
tables,
the
challenge
with
users
being
like
logged
in
into
GDK
2
automatically.
A
Once
we
log
in
in
gdk1,
that
was
to
make
the
rails
application
pointing
to
the
user
session
user
sessions
on
gdk1-
and
this
is
the
approach
that
we
are
going
to
take
on
on
production
as
well,
like,
maybe
in
the
beginning,
like
just
pointing
to
the
same
radius
or
maybe
later
taking
this
to
this
and
putting
it
somewhere
else.
That
is
that's
we
that
we
use
for
sessions
so
that
it's
not
part
of
the
cell
anymore
and
for
GDK.
We
use
the
same.
A
We
use
just
one
with
the
server,
but
we
are
just
splitting
the
different
use
cases
of
readers
into
different
databases
within
different
than
production
production.
We
have
different
cluster
for
each
use
case
and
there
was
three
things
that
needed
to
be
done.
A
Basically,
we
needed
to
tell
rails
and
gdk2
that,
where
the
session
said
this
of
gdk1
and
the
other
thing
was
modify
copying
the
key
which
is
used
to
encrypt
the
sessions
and
the
other
thing
also
the
key
on
which
we
store
the
sessions
on
gdk1-
and
this
is
like
that
was
a
like.
It's
three
things
need
to
be
changed
and
I
already
like
submitted.
A
One
merger
quiz-
that's
almost
going
to
be
emerged,
hopefully
today
by
manud,
to
actually
like
configure
GDK
tool
to
like
to
choose
the
the
right
to
so
that
we
can
change
the
sessions
key
because
it
was
fixed
like
GDK,
was
all
the
time
like
just
or
GitHub
version
anything
unique
key
based
on
the
root
on
GDK
and
yeah.
That's
basically
it
the
script
is
I
hope
that
I
clarified
a
little
bit.
I
hope
the
diagram
helps
we
are
still.
This
is
going
to
take
some
time.
A
B
A
B
Thanks
so
much
we'll
come
back
to
you
after
after
this
I
think
so
what
I've
been
doing
for
the
last
week
is
finding
cross
database
transactions
between
users
and
members
and
similar
things
like
users
and
projects,
for
example.
So
you
can
have
a
look
and
correspondingly
I've
opened
in
an
epic
I
should
probably
link
the
Epic,
an
epic
of
issues
to
fix
this
cost
database
transaction.
So
this
much
request
just
identifies
and
and
marks
them
as
allow
for
now
yeah.
B
So
I'll
link
the
Epic
later.
But
the
main
thing
is:
it's
more
of
a
complaint
is
active
record
callbacks,
a
lot
of
fun,
the
implicity
of
a
transaction,
and
it's
it's
not
great
to
try
and
untangled
app.
The
second
thing
is:
we
have
this
cross
database,
ignore
tables
can't
remember
who
created
it
was
it
Omar?
Did
you
create
that
yeah?
The.
B
Oh
cool,
do
you
know,
do
you
know
it
will
get
ignored
for
GitHub,
CI
schema
as
well,
so
I
know
we.
We
definitely
do
want
to
ignore
for
ster,
white
and
Main.
But
let's
just
say
we
ignore
user
and
as
a
transaction
that
has
the
user
stable
and
a
CI
table,
will
they
get
ignored.
A
I
mean
it's
gonna
ignore
any
tip.
I
mean
it
has
nothing
to
do
with
me
like
it's
going
to
ignore
any
tables
that
you
specified.
So
if
you
specify
you
I,
think
I
think
in
the
testing
that
I
was
doing
yeah
no
I
did
it
on
on
gitlab
Min
gitlab
main
cluster
white,
but
it's
the
same.
B
Right
one:
is
it
plus,
is
it
possible
to
only
have
it
ignore
only
between
Gila,
Master,
white
and
Gila
main
cell,
but
have
it
trigger
foreign.
B
A
C
B
That's
true,
that's
true,
that's
true,
because
it's
it's
the
tables
that
you
ignore
on
the
other
side.
Okay,
that's
fine,
then
I
think
in
that
specific
enough,
then
yeah
yeah
I
was
worried.
There's
some
more
more
about
future
code.
Coming
up
that
we
miss,
but
that's
all
good
cool
is
yours.
Done
is
the
compiling
done.
Let.
A
Me
see
I'm
just
like
writing
down
my
answer:
yeah,
okay,
so
it's
it's
just
finished
running.
Let
me
share
my
screen
again
now.
It's
just
the
moment
of
truth.
A
Okay,
can
you
see
my
skin
again
cool,
so
this
is
gdk1.
3000
I
haven't
modified
anything
I
promise
I
just
run
the
skip
so
yeah
did
you
kill
one
take
some
time?
Okay,
so
I
I
can't?
Okay,
let
me
see,
let
me
see,
ddk
2
is
also
here.
So
no,
this
is
I
need
to
modify
the
URL
I
added
this
bookmark
today.
A
A
Bit
slow
in
the
beginning
when
it
starts,
maybe
because
of
compiling
the
JavaScript
assets
and
I
haven't
switched
to
the
new
solution
that
we
have
added
recently
forgot.
The
name
byte
I
think.
A
C
A
So
I
signed
in
on
the
first
GDK,
let's
see
now
on
the
second
jdk.
Okay,
that
works
cool.
So,
as
you
have
seen,
I
just
signed
in
on
gtk1
and
I
was
automatically
signed
in
on
GDK
2,
and
we
can
verify
that
by
we
can.
We
can
click
on
inspect
application
cookie.
So
we
can
see
that
the
cookie
name
on
jdk1
is
the
same
as.
A
In
gdk2,
it's
the
same
thing
like
there's
some
unique
key
here:
I
I
did
that
in
a
way
that
also
that
so
I
made
jdk
to
copy
the
same
unique
key
objective
one
so
that
we
don't
pick
any
like
if,
like
users
already
signed
in
on
GDK,
they
are
not
going
to
to
be
signed
out,
I
mean,
maybe
it's
not
a
big
deal
for
GDK
people
can
sign
in
again
it's
a
development
development,
environment
yeah.
A
So
let's
try
like
okay,
now,
jdk1
I
can
sign
out
or
let
me
try
to
sign
out
from
gdk2
instead,
so
here
sign
out:
okay,
there's
a
H:
this
is
the
we
cannot.
We
should
not
sign
out
in
the
demo,
maybe
because
this
is
the
problem
of
the
namespaces
that
we
talked
about
before
that
when
the
user
is
modified.
So
that's
something
we
need
to
take
into
consideration.
A
Okay,
but
are
we
signed
out?
Okay?
No,
so
we
cannot
send
out
from
cell
to
let's
sign
up
from
cell
one
okay
and
let
me
get
back
to
cell
two
okay
and
signed
up.
So
that's
problem.
If
you
want
to
sign
out
from
session
two
or
maybe
also,
if
you
try
to
log
in
from
cell
from
cell
2,
we
might
get
the
same
error
because
of
the
name
spaces
so
yeah
we
I
mean.
Maybe
we
solved
the
problem,
but
yeah,
that's
something
we
need
to
discuss.
A
A
C
C
Sorry,
it
wasn't
mute
so
I've,
linked
to
a
proposal
for
the
iteration
board.
Demo
I
think
for
iteration
three,
we're
quite
clear
that
we
want
to
achieve
being
able
to
create
a
project
on
a
cell
so
very
similar
to
the
group's
work.
C
I've
discussed
this
with
Camille
a
bit
yesterday
that
we're
creating
quite
a
large
backlog
now
of
issues
that
need
to
be
fixed
either
by
us
or
other
teams
for
cross
joints
transactions
and
so
on,
and
so
on
that
we
will
focus
up
until
and
including
iteration
four
to
just
get
things
working.
So
we
only
fix
the
items
that
are
absolutely
necessary
to
get
the
demos
out,
so
we're
kind
of
broadening
the
scope
and
then
iteration.
Four.
C
We
were
discussing
that
we
want
to
have
the
ability
to
create
an
organization
on
a
cell
and
then
in
that
organization
to
create
a
group
in
a
project.
Once
we
have
achieved
that
we
would
go
back
and
revisit
all
these
crotch
joint
issues
and
coordinate
with
the
other
teams.
C
So
that's
more
just
from
the
planning
strategy
side.
What
I
would
need
your
help
with
from
the
engineering
side
is
to
go
in
check.
The
demo
plan
see,
if
that's
reasonable
and
then
work
backwards
with
me
to
decide
what
issues
we
really
need
to
create
to
achieve
what's
described
for
the
demo,
so
you
will
see
in
the
blueprint
that
I've
linked
to
Below
in
the
iteration
plan
that
we
have
a
couple
of
work
streams
that
we
defined
would
be
necessary
for
iteration,
four
but
I
think
to
scope
it
to
something
concrete.
C
We
Orient
against
the
demo.
So
that's
the
goal,
so
anything
we
need
to
get
routing
working.
So
we
can
use
a
single
to
domain
to
interact
with
cells
like
those
are
the
issues
we
need
to
create
and
then,
from
there
on,
we
fill
the
gaps
moving
forward.
Does
that
make
sense
or
do
you
have
any
questions.
B
Oh
okay,
I
see
yeah
I
was
wondering
what
this
iteration
for,
but
Christina
you
you
linked
to
the
the
blueprint
iteration
plan
and
I
see
that
a
lot
of
essential
workflows
have
been
pushed
to
iteration,
five
and
and
Beyond.
So.
B
Yeah
but
yeah,
maybe
we
want
to
update
the.
C
C
A
C
I'll
move
those
out
good
catch
I'll,
make
it
a
bit
smaller,
so
I'll
move
those
to
the
later
iteration
epic,
and
then
we
can
focus
on
just
that.
I
think
that
gives
also
quite
a
clear
guideline
for
the
organization.
So
that's
good,
and
then
we
can
give
something
to
other
teams
to
test
with
and
to
play
around
with
cells.
So
that
would
be
the
main
goal.
C
As
for
the
iteration
2
demo,
where
are
we?
How
far
are
we
I
think
I've
seen
this
like
two
issues
that
are
still
open
blocking
the
demo
and
they
were
quite
big,
so
I'm
just
wondering
if
we
will
manage
to
get
this
done
by
this
Milestone
or
if
we
expect
it
to
be
completed
in
the
next
one?
Also
for
planning,
because
I
owe
you
a
proposal
today
or
tomorrow.
A
I
think
one
of
the
two
issues
that
you
mentioned
is
the
one
I'm
working
on
the
shared
session,
so
there's
still
another
Mr
that
I
need
to
to
do
for
GDK,
so
I
mean
it's
working.
We
have
seen
it's
working,
but
maybe
there
will
be
some
concerns
about
the
implementation,
but
because
it's
luxury
Decay,
hopefully,
if
there's
something
blocking,
we
can
still
like
make
a
delayed
issue
for
it
and
get
that
merged.
B
C
All
right,
I
think
for
16.5
I'll,
keep
it
light
on
the
cross
join,
so
we
can
move
into
creating
a
project
on
a
cell,
since
that
seems
overseeable
I
think
we'll
probably
create
another
bunch
of
epics
around
cross
joints
from
there,
but
we
can
do
those
later
if
they're
not
necessary
and
then
take
this
off.
Hopefully
so
that's
the
plan.