►
Description
This virtual meetup from 2020/12/15 features an AMA featuring four GitLab team members from the GitLab CI team. They also offer a preview of our pipeline editor plans including an early look at the visual pipeline builder.
The panel features:
Thao Yeager, Senior Product Manager, Continuous Integration
Veethika Mishra, Senior Product Designer, Continuous Integration
Dov Hershkovitch, Senior Product Manager, Pipeline Authoring
Nadia Sotnikova, Product Designer, Pipeline Authoring
You can find additional information from the meetup and join the discussion via this GitLab issue: https://gitlab.com/gitlab-org/gitlab/-/issues/293913
-----
We would love for you to contribute to our virtual meetups.
You can find upcoming virtual meetups and resources to plan your own here: https://about.gitlab.com/community/virtual-meetups/
We want to hear your ideas for topics. Remote? Security? Contributing to OS? Email suggestions to evangelists@gitlab.com.
We also love when community members present at GitLab meetups. Send an email to evangelists@gitlab.com if you’re interested in being the featured speaker at a future remote meetup.
A
Hi
everyone,
I'm
john
coglin,
I'm
on
the
community
relations
team
at
get
lab,
and
I'm
really
excited
to
be
here
today
for
this
meetup.
This
is
one
of
the
biggest
virtual
meetups
as
far
as
registrations
go
that
we've
had
in
a
while,
and
I
think
it
speaks
to
the
kind
of
awesome
topic
and
group
of
speakers
that
we've
pulled
together
for
this
one.
So
thank
you
all
for
being
here
today.
A
If
you're
not
comfortable
verbalizing
your
question,
you
can
put
them
in
the
chat
and
I
will
verbalize
them
on
your
behalf,
so
yeah,
so
that's
kind
of
what
we're
going
to
be
doing
today,
but
before
we
get
into
that,
I
just
want
to
remind
folks,
you
know:
git
lab
one
of
our
values
is
diversity,
inclusion
and
belonging,
and
so
in
order
to
ensure
our
community
is
welcoming
and
inclusive.
We
ask
that
everybody
abide
by
our
community
code
of
conduct.
I
will
be
pasting
a
link
to
that
in
the
zoom
chat
right
now.
A
A
A
And
I
will
be
liberal
and
muting
people
who
interpret.
We
have
thousands
of
features
we
release
every
month
on
the
22nd,
so
there's
always
new
stuff
coming
into
the
product,
and
I
think
you'd
be
really
amazed.
If
you
dig
in
and
see
what
gitlab
offers,
just
everything
that
we've
included
and
and
really
the
product
continues
to
improve
every
month
is
the
best
release
ever
so
it's
a
great
time
to
get
started,
exploring
gitlab
and
using
it
for
your
projects
or
for
your
team.
A
Gitlab
is
an
open
core
company,
so
everybody
can
contribute,
that's
the
mission
of
the
company
and
we
have
an
open
source
version
with
3
000
folks
who
that
have
contributed
to
that.
And
so
it's
really
you
know
we
have
a
really
great
community.
Around
gitlab
other
cool
things
about
gitlab
were
before
kobit.
We
were
the
largest
all-remote
company
in
the
world.
A
I
think
we're
you
know,
still
the
largest
kind
of
all-remote
company
that
hasn't
been
forced
to
become
remote
by
a
pandemic,
and
so,
if
you're
interested
in
remote
work,
best
practices,
just
google
get
label
remote
you'll
find
lots
of
great
information.
That's
been
put
together
by
our
head
of
remote
daramurf
and
his
team
and
yeah
we're
also
really
transparent.
So
our
road
maps
and
strategy,
which
you'll
be
you
know,
hearing
a
little
bit
about
today
in
regards
to
our
cei
kind
of
plans,
are
all
public
everything
that
we
do
to
run.
A
You
can
watch
our
meetings
on
youtube
and
follow
all
the
work
that
we're
doing
on
our
issue
trackers,
so
we
try
to
be
public
by
default
and
so
there's
tons
of
information
about
how
we
work
and
what
we're
working
on
and
what
we're
planning
to
do
out
there.
You
know,
in
our
handbook
and
in
our
public
projects,
so
definitely
take
a
look
at
that
yeah
and,
as
I
mentioned
today,
we're
going
to
be
having
our
ci
team.
A
Do
this
presentation
on
their
kind
of
strategy
and
plans
and
then
followed
by
an
ama
so
really
excited
about
that
and
a
couple
of
requests
before
I
headed
over
to
dove,
as
I
mentioned
earlier,
or
if
you're
just
joining
us,
please
keep
your
cameras
on.
It
makes
it
a
lot
more
fun
when
we
can
see
folks
during
the
meet
up,
but
we
are
recording
this.
So
if
you
don't
want
to
be
on
youtube,
keep
your
camera
off.
Please
stay
muted.
A
A
If
you
could
share
where
you're
joining
from,
I
think
we
have
pretty
good
geographic
diversity
just
from
the
panelists
alone,
but
it'd
be
really
cool
to
see
like
where
the
53
people
here
are
joining
us
from
I'm
guessing
like
at
least
I'll
say
the
over
is
15
countries,
that's
my
guess,
and
if
you're
going
to
tweet
about
it,
use
hashtag,
getlab,
meetup
and
lastly,
just
close
tabs.
You
know
close
all
your
tabs
mute
your
slack
and
just
relax
and
enjoy
be
present.
A
C
A
C
Folks
here,
but
I
want
to
introduce
nadia
sutnikova,
which
is
the
product
designer,
and
we
are
a
part
of
the
pipeline
authoring
team.
The
pipeline
authoring
team
is
a
new
team
and
before
we'll
touch
a
little
bit
about
how
is
it
related
to
ci?
I
would
like
to
invite
you
all
to
go
into
our
category
direction,
page
where
you
can
read
in
detail
all
about
our
mission,
our
goal,
our
vision,
what
we
are
planning
to
work
on
next
and
why.
D
Hi
everyone,
my
name,
is
tao
yeager,
I'm
the
product
manager
for
verify,
continuous
integration
and
also
joining
me
is
my
awesome
designer
vitica
misra.
She
and
I
are
responsible
for
both
the
vision
and
the
issues
that
we
work
on
for
continuous
integration.
Again,
as
dove
mentioned,
we
are.
We
also
have
a
direction
page
for
ci
and
I'll
share
it
in
the
zoom
chat
as
well.
So
we
invite
you
to
go.
Take
a
look
there
it'll
give
you
updates
on
what
we're
working
on
next
and
the
things
that
we're
focused
on.
D
So
one
thing
that
we
wanted
to
give
you
a
quick
overview
about
the
the
separate
but
very
related
to
product
groups.
I
know
you're
joining
an
ama
to
speak
with
the
gitlab
ci
team,
but
within
ci
we
actually
have
two
distinct
groups
that
collaborate
together
and
own
very
distinct
areas
of
the
interaction
of
the
user
with
the
pipeline.
You
can
see
there
in
this
line
that
we're
illustrating
that
for
the
ci
group,
which
vitica
and
I
are
part
of,
it-
represents
a
specific
interaction.
C
C
Okay,
so
we
talked
a
little
bit
about
the
pipeline
authoring,
but
I
want
to
emphasize
our
team
goal,
because
this
is
a
rather
new
team
and
I
want
to
talk
about
the
goal,
because
the
goal
is
what
defined
the
scope
of
work
for
the
team
and
actually
defined
us
as
a
team.
So
public
authoring
goal-
and
this
is
straightforward
from
our
handbook-
is
to
make
the
ci
authoring
experience
as
easy
as
possible
for
both
novice
and
advanced
user
alike.
C
This
means
that
this
team
will
this
team
will
have
two
continuous
themes
that
will
continue
to
work
on
in
one
hand,
we
want
to
continue
to
be
a
leading
vendor
in
ci
and
support
the
advanced
ci
use
cases
with
our
advanced
features
like
a
parent-child
pipeline
multi-project
pipeline,
our
robust
syntax
and
syntax
capability.
Basically,
basically
give
you
all
the
possibility
to
do
whatever
you
want
with
your
pipeline,
using
our
advanced
syntax
capabilities.
C
But
with
that
said,
we
want
to
make
sure
you
will
be
supported
by
a
superior
user
experience,
because
we
do
understand
that,
although
some
of
you
may
be
maybe
very
technical
and
knowledgeable
around
ci
there.
First
of
all,
there
are
parts
in
your
organization
that
may
be
less
familiar
with,
and
you
want
to
offload
some
of
the
work
to
them,
so
you
want
to
increase
your
ci
adoption
and
also,
if
you
are
new
to
ci
or
if
you
want
to
have
a
smooth,
a
smooth
move
from
source
control
management
to
ci.
C
We
want
to
support
you
support
without,
with
with
that
move,
so
a
guided
experience
that
will
hand
hold
you
across
your
experience.
As
you
define
and
run
your
pipeline
is
something
that
will
also
keep
in
mind,
keep
close
to
our
heart.
So
there
is
a
gap
between
having
a
sophisticated
capabilities
and
a
superior
ui,
so
the
team
goal
is
actually
making
sure
we
can
bridge
this
gap
and
also
bootstrap
some
of
those
capabilities
using
our
ui.
C
So
today,
when
you
come
to
write
your
pipeline
in
in
gitlab,
you
realize
that
it
is
no
different
than
editing
any
yaml
file
in
your
project.
Repository
now,
although
this
is
this
can
be,
this
could
be
a
huge
benefit
if
you're
familiar
with
the
ammos.
You
know
that
there
is
a
lot
of
capabilities
that
you
can
do
with
that.
But
then
again
you
need
to
know.
Yammers
and
also
yammers
can
be
very
complex
and
there
is
a
limit
to
what
you
can
do
with
that.
C
In
addition
to
that,
we
don't
have
any
unified
flow
that
will
support
you
from
creating
a
pipeline
to
running
a
pipeline
and
maybe
debugging
a
pipeline.
Sometimes
you
don't
know
where,
where
to
go,
and
we
don't
provide
you
any
hints
on
support
in
the
ui
and
also
a
part
of
the
experience,
and
I'm
only
talking
about
the
gitlab
ui
experience,
it
is
very
fragmented
and
when
you
start
writing
your
pipeline,
you
realize
that
you
can
do
it
through
our
web
ide
web
ide.
You
can
do
it
to
our
single
file.
Editor.
C
Do
you
know
where
to
go
first
and
why?
I'm
not
sure
we
have
an
enter
with
linting
capability,
which
is
which
is
great,
but
this
is
why
I
actually
added
this
this
slide.
It
is,
it
is
actually
not
positioned
pretty
well,
it
is
I
like
to
say
it
is
well
hidden
somewhere.
You
know
pipeline
pages,
which
is
a
shame,
and
then
you
have
the
pipeline
view
where
you
can
see
your
python.
E
Yeah
so
hi
everyone.
This
is
nadia,
as
dove
already
mentioned
when
he
talked
a
little
bit
about
the
problems
with
the
existing
authoring
experience.
We
realized
that,
in
order
for
us
to
address
those
pain
points
and
to
solve
some
of
those
problems,
we
need
to
create
one
single
place
for
the
pipeline
authoring
experience.
E
So
here
in
this
little
demo
recording
you
see
just
a
quick
preview
of
the
vision
that
we're
exploring
for
the
new
pipeline
editor
and
some
of
the
things
that
we
want
to
offer
in
the
new
authoring.
Experience
is
allowing
you
to
create
pipelines
visually
so
we're
currently
working
the
designs
for
the
pipeline,
editor
visual
builder,
which
would
allow
us
to
abstract
some
of
the
yaml
syntax,
make
it
easier
to
create
your
pipelines,
especially
for
those
new
to
cicd.
E
E
E
Users
who
are
more
advanced
and
who
have
very
complex,
ci
cd
configurations-
and
perhaps
you
do
prefer-
writing
your
configuration
using
code
using
yaml,
so
you
would
still
be
able
to
do
that.
You
would
be
able
to
take
advantage
of
the
visual
builder
and
still
write
your
configuration
using
the
editor
if
you
prefer
to
do
so.
E
I
just
want
to
note
that
the
preview
you
see
here
this
is
still
in
discovery,
so
we're
still
working
on
polishing
that
vision
and
figuring
out
exactly
what
it
should
look
like.
So
also,
if
you
have
any
questions
or
ideas
or
suggestions,
we
would
be
really
happy
to
hear
what
you
have
to
say
about
the
vision
that
we're
currently
exploring
and
next
dove
will
tell
you
more
a
little
bit
about
the
validation
that
we
want
to
offer
in
the
pipeline.
Editor.
C
Thank
you
nadia.
So,
as
I
mentioned,
as
you
start
writing
your
pipeline
in
yaml,
we
want
to
provide
you
some
support,
a
little
bit
of
handholding,
so
syntax,
highlighting
making
sure
we
highlight
the
the
different
keywords,
syntax
validation,
making
sure
we
validate
on
the
fly,
as
you
type
your
your
configuration
and
let
you
know
if
there
are
some
errors
and
and
warnings
on
your
pipeline
configuration
and
autocomplete.
C
All
of
that
is
something
that
we
are
working
or
actively
working
on
making
sure
it
will
be
available
for
you
in
the
pipeline
editor
and
lastly
or
lastly,
another
feature
that
we
want
to
bring
closer
is
our
linking.
I
mentioned
our
linkedin
capability,
which
is
hosted
in
in
in
a
separate
area
in
a
separate
page
in
in
gitlab,
so
we
want
to
make
to
bring
the
link
the
lintel
closer
as
possible
to
your
to
your
code.
C
E
Yeah-
and
I
would
like
to
finish
our
little
preview
of
the
pipeline-
editor
experience
with
a
few
words
about
the
use
of
templates
cicd
templates
in
gitlab.
We
currently
already
have
dozens
of
ci
cd
templates
that
are
available
in
gitlab
editors
in
the
single
file
editor
and
in
the
web
id,
but
those
templates
are
not
very
discoverable.
E
E
E
So,
currently,
the
only
way
to
use
a
template
is
to
either
open
it
and
copy
the
whole
template
and
paste
it
into
the
editor
of
your
choice,
whatever
you're,
using
whether
it's
yeah,
maybe
it's
vs
code
or
maybe
you're
using
web
id,
or
something
like
that
or
you
can
apply
the
template
which
will
override
your
whole
configuration.
So
that
is
not
the
ideal
experience,
so
we're
looking
into
ways
to
allow
you
to
use
those
templates
in
a
more
granular
way.
E
So
you
would
be
able
to
copy
the
include
syntax
for
this
template,
which
would
allow
you
to
sort
of
plug
that
template
into
your
configuration.
So
you
would
be
able
to
assemble
your
pipeline
from
many
different
templates,
which
opens
up
a
whole
world
of
possibilities
in
how
you
can
configure
your
pipeline
this
way
so
yeah.
If
you
have
any
questions
or
any
feedback
around
this
experience.
Also
let
us
know-
and
now
I
will
hand
it
over
to
tao
and
vitica,
to
talk
more
about
what's
happening
on
their
team.
D
D
There
are
pipeline
details
that
we
can
do
better
at
surfacing
and
that's
our
focus
listed
there
is.
We
want
to
give
you
access
to
information
about,
what's
happening
with
a
writing
pipeline
everything
from
job
duration,
warnings,
even
understanding
the
queue
time
that
will
help
you
to
be
able
to
react
in
monitoring
a
pipeline
and
then
going
beyond
that,
giving
you
some
analytical
detail
about
historical
pipelines.
D
So
next
slide,
please.
We
know
right
now
the
existing
pipeline
overview
experience
can
be
even
better.
What
you
see
there
is
a
snapshot
of
when
you're
looking
at
the
pipeline
overview
and
looking
at
the
jobs
within
the
stage.
As
the
pipeline's
running
you
don't
have
a
way
to
see
the
job
queue
time
or
job
duration
for
jobs
that
have
completed.
That
certainly
would
be
an
improved
pipeline
overview
experience.
D
You
also
are
missing
a
quick
count
of
how
many
jobs
are
running
or
the
ones
that
have
passed
and
failed,
and
we
certainly
want
to
address
that.
One
other
thing
is
having
visibility
into
the
number
of
jobs
triggered
by
a
single
job
would
be
helpful.
That's
something
that
is
missing,
that
we
want
to
focus
on
delivering
and
then
also
in
the
absence
of
the
previous
three
bullets
as
well.
You're
missing
out
on
some
actionable
insights
of
where
the
bottlenecks
are
and
where
you
can
optimize
your
pipeline
to
make
it
run
faster.
D
G
Hi
everyone,
I'm
vittika,
and
I
just
mentioned
the
key
problems
that
we
are
facing,
that
we
identified
in
our
existing
design
that
we
realized,
after
speaking,
with
our
users,
about
their
experience
when
they
are
trying
to
understand
the
health
of
the
pipeline,
identify
potential
bottlenecks
which
could
provide
an
opportunity
for
optimization
and
even
when
it
comes
to
debugging
their
jobs.
So,
based
on
all
of
those
feedbacks
and
suggestions,
we
made
a
bunch
of
changes
to
our
design.
G
So
what
you're,
seeing
in
front
of
you
is
the
new
layout
that
we're
working
on
and
some
major
changes
that
we
have
done
is
we
have
explored
rearranging
the
information
so
to
give
importance
to
items
that
our
users
deem
as
more
important?
We
played
around
with
the
hierarchy
of
the
different
elements
in
the
layout.
The
next
thing
we
did
was
I
mean
we
do.
G
We
don't
want
our
users
to
spend
a
lot
of
time
on
this
page
when
all
they
are
looking
for
is
just
a
quick
status
of
the
pipeline,
and
that's
the
reason
that
you'll
see
a
small
donut
chart
on
placed
on
the
top
left
corner.
What
it
does
is
upon
hover.
It
gives
a
very
quick
glimpse
of
what's
going
on
in
your
pipeline,
like
how
many
pipelines,
how
many
jobs
are
running,
how
many
have
failed
or
succeeded,
or
even
pending
that
kind
of
information.
G
Then
moving
to
the
next
point
is
we
have
placed
a
trigger
to
the
most
important
action,
but
in
the
context
of
the
present
status
of
the
pipeline,
so
you'll
see
a
primary
action
button
placed
on
the
top
right
corner
and
that's
an
action.
That's
going
to
change
with
the
current
context
like
what
what
starters
the
what's
the
status
of
the
pipeline
currently
based
on
that
it's
going
to
change
every
time
it's
a
dynamic
trigger,
and
the
next
thing
is
of
course
the
graph.
G
The
gray
part
of
that
denotes
the
time
for
it,
which
it
has
been
queued
like.
How
long
has
this
job
been
queued,
then
the
colored
part?
It
shows
the
execution,
duration
and
the
color
itself
denotes
the
status
of
the
job,
and,
apart
from
that,
you'll
also
notice
that
there's
this
tiny
warning
icon
on
some
of
the
jobs
and
what
they
do.
G
That
means
those
are
the
ones
that
would
that
could
really
use
some
optimization,
and
so
we
are
highlighting
those
bottlenecks
on
the
gantt
chart
itself
and
that
apart,
if
you
see
like
when
in
the
gif
when
you,
when
there's
a
hover
over
a
job
pill,
then
the
corresponding
job
in
the
pipeline
graph
below
it's
also,
it
also
gets
highlighted.
So
we
are
trying
to
make
that
correlation.
G
The
next
slide.
Please
now
that
wasn't
enough.
What
we
also
wanted
to
look
at
was
so
today
when
users
they
want
to
just
go
and
have
look,
have
a
look
at
the
logs
for
a
job.
They
have.
They
eventually
end
up
planning
on
a
whole
different
page
and
besides,
like
taking
away
the
larger
context,
it
also
kind
of
makes
this
task
longer
and
slower.
G
So
we
really
wanted
to
do
something
about
it,
and
so
we
thought
why
not
bring
the
job
locks
to
the
same
page
as
the
pipeline
overview
and
that's
what
we
are
doing
here
so
each
job
after
we
after
being
executed,
I
mean
each
part
of
this
job
after
being
executed.
What
happens,
is
it
gets
collapsed
into
sections?
So
this
is
our
effort
to
make
the
logs
more
readable
and
easy
to
understand.
G
Now
it
doesn't
matter
if
it
has
succeeded
or
failed.
It
gets
collapsed,
but
if
it
has
failed,
then
besides
just
showing
the
title
of
the
part
that
has
failed,
it
also
kind
of
gives
some
additional
information
in
that
section
that
could
quickly
help
you
identify
like
what's
wrong.
What
what
what's
wrong?
With
my
job,
I
mean:
where
could
I?
G
How
could
I
debug
it
better
so
without
like
going
through
the
whole
entire
log,
you
might
be
able
to
get
some
information
just
from
a
glimpse
and
that's
what
we
are
aiming
at
and
that's
not
all.
So
when
the
side
panel
or
the
drawer
is
open,
you
can
quickly
switch
between
jobs
on
this
graph
like
if,
if
I'm
clicking
on
a
certain
job,
but
I'm
more
interested
in
what's
happening
in
the
next
one.
G
Suddenly
I
could
just
click
and
information
of
that
new
selected
job
is
going
to
appear
on
the
side
panel.
So
this
is
our
goal.
I
mean
this
is
what
we
are
aiming
at
achieving
in
our
next
year,
and
I
hope
you
have
a
lot
of
feedback
around
it,
because
I'm
going
to
share
the
link
for
the
epic
and
also
figma
prototype,
if
you
just
want
to
play
around
with
the
new
design
yeah
that
that's
all
from
my.
A
A
All
right,
so,
is
that
the
is
that
the
end
of
kind
of
what
everyone
wanted
to
show
today?
Yes
great,
so
we
have
received
a
ton
of
questions
so
far,
which
is
awesome.
Thank
you.
So
much
to
all
the
folks
who've
been
submitting
questions.
I
think
bjorn
and
I
apologize
if
I
mispronounced
any
of
these
names,
but
bjorn
had
the
first
question:
if
bjorn,
if
you
want
to
verbalize
that.
H
We
are
using
kaneko
at
the
moment
because
we
need
we
need
docker
containers
without
any
root
access
and
it
doesn't
work
so
well
and
I
thought
okay,
I
cannot
be
the
only
one
which
was
having
that
problem
and
what
is
what
is
gitlab
doing
if
they're
building
docker
containers
inside
the
cluster
or
or
are
are
they
doing
it
and
I'm
doing
the
trust
wrong
and
use
another
normal
docker
host
for
that?
So
that
was
the
question.
Is
there
is
a
good
advice?
H
D
Honest
ewan,
I
actually
don't
have
an
answer
right
now
for
your
question,
I'm
reaching
out
actually
to
a
couple
of
our
product
managers.
That
would
be
the
best
to
answer
your
question.
As
far
as
with
the
configuration
you
mentioned,
using
gitlab's
kubernetes
renders
and
building
docker
containers
might
be
something
that
I
can
follow
up
with
after
this
ama
or
maybe
at
the
end.
If
I
get
an
answer
back
from
from
my
fellow
pams.
I
H
H
E
A
K
K
If
we
put
jobs
into
stage
that
then
have
dependencies
on
jobs
from
previous
stages,
it
will
be
a
similar
concept
of
using
that.
The
problem
is
that
from
an
implementation
perspective
today,
the
dog
needs
the
a
job
to
be
specified
in
a
specific
stage.
Just
because
we
are
not
allowing
a
job
from
do
specify
needs
from
jobs
within
the
same
stage
and
that
can
create
some
cyclical
dependencies
and
that
we
need
to
overcome.
J
E
Yeah,
so
you
were
asking
if
that
feature
set
that
where
we
were
talking
about
for
the
pipeline
editor
if
it
would
be
possible
for
the
web
ide.
So
currently
it's
not
something
that
we're
planning
we're
we're
we're
really
focused
on
creating
this
dedicated
pipeline
authoring
experience,
because
we
want
to
make
that
editor
really
tailored
to
your
needs
as
you're
configuring
your
pipeline.
But
that
being
said,
I'm
curious
to
learn
more
about
why
you
prefer
using
web
id.
Is
there
some
functionality
that
you
really
like
in
web
id?
J
E
Actually,
allowing
you
to
edit
multiple
files
in
the
pipeline
editor
is
something
that
we're
thinking
about.
We
don't
have
a
timeline
for
that
yet,
but
eventually
you
should
be
able
to
edit
multiple
files
in
the
pipeline
editor
as
well
yeah
dope,
if
you
want
to
add
something,
feel
free
to
weigh
in
as
well.
J
F
Okay,
maybe
I
can
complement
a
little
bit
to
the
answer
from
the
technical
side
actually
underline
under
the
editor
that
we're
making
and
the
web
ide
there
are.
We
have
some
shared
components,
so
it's
very
likely
that
we
will
see
improvement
in
both
in
both
places,
especially
regarding
some
syntax,
highlighting
like
the
one
we
saw.
This
is
probably
going
to
be
contributed
to
both
places.
So
we
expect
to
see
some
improvements,
we're
working
closely
with
the
editor
team,
that
are
words
that
are
building
the
ide,
but
they
are
separate
teams.
L
A
L
L
I
was
wondering
if
you
guys
could
provide
an
easier
way,
maybe
a
link
on
the
original
gitlab
ci,
because
based
on
what
they
include
you
could
you
could
derive
what
what
what
what
the
url
is
to
get
to
that
file
right.
So
I
was
hoping,
if
there's
an
easier
way
for
us
to
just
go
from
the
original
github
ci
open
that
file
just
through
click.
It's
there,
you
go
what
I'm
saying.
C
C
C
L
Would
be
something
very
useful
that
would
be
very
useful
because
when
we
troubleshoot,
if
I
could
just
click
up
with
a
click
of
a
button-
and
I
see
exactly
which
job
they're
running
that
that's
bombing
out
and
I
see
what
what
the
commands
running
or
the
script
tag
that
it's
using.
It's
very
easy,
much
easier
for
me
to
go
just
to
do
that,
instead
of
just
drill
down
into
the
necessary
depth
yeah.
That
would
be
very
useful
to
me.
L
A
C
L
L
So
the
current
question
is
one
of
the
most
frequent
question
I
get
from
user.
Is
my
job
works
two
days
ago
and
now
it
doesn't
work
right
and
what
I
want
to
do
is
I
go.
I
want
to
go,
see
all
occurrences
of
that
job
and
see,
okay,
what
the
differences
are
and
what
the
output
are
and
from
the
pipeline
page
I
could
filter
pipeline,
but
I
can't
filter
by
a
job
name
and
when,
on
the
job
page,
you
can't
filter.
A
D
Yeah,
so
that's
that's
a
that's.
A
good
feature
request
on
what
I've
added
is
a
placeholder
issue
for
this,
which
is
simply,
I
haven't,
filled
out
the
template
for
that
issue,
but
to
to
have
a
placeholder
to
add
filters
on
the
job
page,
similar
to
what
we
have
or
well.
Let
me
ask
you
this:
what
specific
attribute
would
you
want
to
be
searching
on
when
you're
looking
at
a
jobs
page
with
filters
for.
D
L
A
Sure,
yeah
yeah,
what
I
can
do
is
I'll
take
we're
working
off
a
dock
where
we're
kind
of
tracking
the
questions
and
answers
so
I'll
I'll
share
a
version
of
that
with
everyone.
I
just
didn't
want
60
people
that
weren't,
you
know
familiar
with
how
we
kind
of
managed
meeting
agendas
at
gitlab
in
the
dock
at
the
same
time
trying
to
figure
it
out,
because
I
thought
it
might
get
a
little
confusing
for
folks.
But
I.
L
A
M
Yeah
sure
yeah
from
where
we
currently
work
out,
there's
a
lot
of
co-workers
that
have
trouble
working
with
the
pipeline
editor
because
they're
not
really
into
devops
kind
of
way
or
the
ci
or
pipeline
way.
So
the
first
sneak
preview
has
given
me
a
lot
of
detail
that
I
can
share
with
my
colleagues
and
I'm
just
excited
for
them
to
see
the
new
editor.
So
I
was
wondering
when
there
would
be
an
estimation.
C
C
I
So
this
question
is
answered
for
me,
but
I
have
two
more
questions.
One
is
how
do
you
handle
in
general
if
you
have
a
lot
of
repositories
and
you
have
one
change
in
one
repository
and
you
want
to
build
multiple
child's
repos
in
one
go,
but
you
have
general
the
problem
that
you
have
a
lot
of
these
repositories,
so
it
takes
hours
and
hours
to
process
them.
Is
there
any
general
mechanism
I
can
use
to
handle
such
so?
I
don't
want
to
have
it
on
every
push,
but
something
on
the
schedule
basis.
I
Suggestion
how
I
can
do
this
because
we
had
a
lot
of
we
have
a
core
product
and
a
lot
of
extensions
and
separated
repositories,
and
for
us
it's
really
hard
to
to
to
have
an
overview
which
repository
is
in
state
with
the
development
state
and
which
repository
we
have.
We
have,
as
each
extension
has
a
stable
state
build
and
the
development
state
built,
but
the
problem
is
we
have
so
many
builds
going
on
that
every
time
when
we
push
something
to
the
main
repository,
we
had
a
lot
of
builds.
I
I
20
repositories,
so
we
have
one
major
repository
and
when
the
developer
is
pushing
to
the
develop
branch,
for
example,
I
wanna
know
if
every
of
this
child
repositories
is
built
with
the
new
development
state.
But
the
problem
is
it's
a
it's
a
gigantic
mass
of
of
of
builds.
So
we
couldn't
do
this
on
every
push.
I
K
K
K
So
there
is
with
the
rules,
job
rules
you
can
specify
and
when
basically
specific,
when
some
specific
files
change
and
that
job
or
that
pipeline
will
be
triggered
only
when
those
jobs,
those
file,
changes
and
yeah.
It's
it's
more
set
of
rules
that
you
have
to
define
in
the
gitlab
ciamel
on
each
repository
for
in
order
to
decide
which
jobs
should
run
when
something
is
triggered.
I
Yeah,
of
course,
of
course,
so
is
it
possible
to
to
detect
file
changes
in
another
repository
with
gitlabs
yeah,
so,
for
example,
because
it's
either
a
huge
project,
so
every
project
builds
two
three
hours.
It's
a
gigantic
sea
pass
those
copies
and
figuring
out
which
parts
affect
which
one
would
be
really
cool.
But
the
problem
is
only
the
child
project
know
which
things
are
they
depend
on
so.
K
Yeah,
I
think
this
depends
on
the
case
the
case
scenario,
but
one
idea
would
be
that
you
will
trigger
all
the
pipelines
for
all
the
sub
projects,
but
some
pipelines
will
simply
do
nothing
because
they
have
no
changes
related
to
it
so
we'll
simply
either
you
have
a
pipeline
that
passes
immediately
because
there's
nothing
to
do
and
where
there
are
actually
changes.
You
might
run
the
jobs
for
those
changes,
so
each
sub-repository
is
responsible
for
knowing
what
changes
have
occurred.
K
I
Yeah
so
right
now
I'm
doing
a
mix
of
rest
api
and
all
this
kind
of
stuff
right.
It's
it's
not
that
easy.
I
know
that
it's
not
a
general
general
problem,
so
it's
a
special
case,
and
but
I
only
want
to
hear
your
suggestions,
because
maybe
I
missed
something
or
you
have
a
better.
You
have
more
insight
than
I
have
so.
K
L
D
I
will
make
a
note
of
that
and
possibly
create
an
issue
that
would
probably
fall
under
our
existing
epic
just
to
help
you
troubleshoot
by
the
way.
Your
two
questions
today
alone
really
helps
confirm
to
us
that
that
jobs
page
is
actually
being
used
by
folks
in
the
community
so
absolutely
and
and
you're
talking
about
the
the
page,
you
would
go
to
navigate
from
ci
cd
from
the
left
navigation
and
then
selecting.
L
D
L
D
D
L
L
D
G
A
K
O
K
O
A
Okay,
I'll
read
this
out
loud
for
foster
not
strictly
related
to
today's
topic,
but
it
was
interesting
in
built-in
cold
quality,
but
was
interested
in
built-in
code
quality
checks.
They
require
docker
and
docker,
which
has
security
implications.
What
is
the
reason
for
this
requirement?
Why
can't
it
be
executed
in
plain
docker
in
non-privileged
mode.
D
A
Yeah
foster
I'll
I'll
give
a
to-do
on
the
dock.
Where
we've
been
keeping
notes
to
the
pm
who
owns
this
and
when
I
share
this
with
the
team,
hopefully
you'll
be
able
to
find
an
answer
from
james
in
there.
I
think
that'll
be
the
best
way
to
to
solve
for
changes
not
being
here
right
now
and
then
the
last
question
from
gabriel.
B
Yeah
sure
so
hi
everybody.
I
just
want
to
ask
about
like
super
models,
because
we
in
the
company
like
use
a
lot
of
them,
and
it
has
not
been
like
very
easy
for
the
devs
to
set
it
up
on
the
ci
and
sometimes
like
there's
a
lot
of
errors.
So
I
don't
know
if
you
have
any
road
maps
for
supermodels
and
adding
it
into
this
like
new
graphics
tool
and
adding
more
logs
to
it,
because
sometimes
when
it
fails
because
of
that's
a
model,
there
is
no
like
locks
attached
to
it.
D
B
Yeah,
so
basically,
it's
that
when
we
run
a
ci
job
on
a
repo
that
has
a
module,
that
also
needs
to
be
like
run,
and
so
we
like
configure
it
in
into
the
jumbo
file.
But
not
every
time
runs
properly.
Sometimes
it
has
some
troubles.
It
can't
pull
the
project
successfully
to
build
it
into
the
runner
and
and
like
currently
there's
no
way
to
see
any
logs
on
it
like
it
only
say
like
it
fail
pulling
the
supermodel
and
that's
it
and
so
like.
E
I
also
just
wanted
to
quickly
say
that
we
do
want
to
show
more
information
about
the
running
pipeline
in
the
pipeline
editor.
Eventually,
it's
not
in
the
roadmap,
we're
just
starting
the
discovery
around
that,
but,
for
instance,
as
you
run
your
pipeline,
you
alter
your
pipeline.
You
run
your
pipeline.
We
want
to
show
more
information
regarding
the
status
of
the
pipeline,
including
the
logs
potentially
in
the
pipeline
editor,
so
we'll
be
looking
into
what
that
should
look
like
later
on
so
yeah.
A
D
By
the
way,
we're
so
glad
all
of
you
were
able
to
join
today,
if
you
didn't
already
realize
it,
you
are
the
first
to
see
some
of
the
designs
that
we
are
planning
for
future
development.
It's
actually
the
first
time
we're
showcasing
it
publicly.
You'll
eventually
see
it
in
other
social
media
mentions,
but
you
definitely
got
a
advanced
sneak.
Peek.
A
Thanks
yeah,
I'm
gonna
ask
the
last
question
and
then
this
is
gonna
be
the
last
question
from
foster.
Foster's
mic
is
broken,
so
I'll
verbalize
foster
also
has
some
thoughts
on
get
lab.
Ci
token.
Are
there
plans
to
grant
get
lbci
token
right
privileges?
In
addition
to
reading
it
would
be
nice
to
see
a
possibility
to
run
jobs
without
such
a
token
for
untrusted
pipelines.
C
K
D
D
D
So
are
there
plans
to
grant
get
lab?
Ci
tokens
write
permissions.
We
do
have
an
issue
for
this
foster
the
I
can't
it.
The
name
of
that
issue
has
changed
a
couple
times,
but
it's
probably
relevant
to
what
you're
asking
as
far
as
the
plans
it's
not
currently
scheduled
in
any
specific
upcoming
milestone,
but
we're
aware
that
this
is
something
that
is
a
popular
feature,
request
I'll,
find
that
issue
and
link
it
in
the
this
document.
John.
A
Awesome,
thank
you
and
thank
you
to
all
of
our
panelists.
What
we
always
do
at
the
end
of
our
gitlab
virtual
meetups
is
unmute
our
microphones
to
give
a
round
of
applause
to
the
speakers,
and
I
feel,
like
this
group
in
particular,
really
earned
it
today.
Answering
so
many
great
questions.
So
if
you
don't
mind
just
unmuting
and
joining
me
and
giving
the
speakers
a
round
of
applause,.
A
That
was
a
great
session.
Thank
you
so
much
to
everyone
who
joined
us
today.
Thank
you
so
much
to
everyone
who
asked
questions
and,
of
course,
to
our
panelists.
Thank
you.
So
much
for
your
time
and
preparing
to
present
to
the
group
and
and
all
the
insightful
answers
that
you
shared
today
yeah.
I
guess
you
know.
Lastly,
I
just
wanted
to
let
folks
know
you
know
we're
we're
always
looking
to
kind
of
plan
new
virtual
meetups.
A
We
think
it's
a
great
opportunity
to
bring
our
global
community
together
and,
if
you're,
interested
in
speaking
or
hosting
or
just
have
topics
that
you
want
to
hear
about.
Please
email,
evangelist,
gitlab.com
to
get
more
involved
in
the
community
and,
if
not,
I
hope,
to
see
you
at
our
next
virtual
meetup.
I
know
I
think
everyone
here
is
part
of
that
virtual
meetups
group
on
meetup.com
now,
so
we
hope
to
see
you
at
the
next
event
and
thank
you
so
much
for
your
time
and
attention
today
so
take
care
and
happy
holidays.