►
From YouTube: 2023-04-25 GitLab Hackathon AMA / Live Coding
Description
Day 2 of the Hackathon, we answer a few questions and attack some live coding!
https://about.gitlab.com/community/hackathon/
A
A
So
thanks
for
joining
so
far,
we
have
15
participants
and
we
are
one
and
a
half
days
in
there
are
X
number
of
merge
requests,
which
is
what's
going
to
be
the
the
work
that
we
work
on
later
in
this
call
to
to
add
that
number
to
this
leaderboard.
So
we
can
actually
see
how
many
merge
requests
we've
got
at
the
moment.
The
best
place
to
find
that
is
on
the
the
hack
form
page
itself.
So
we
have
41
merge
requests.
A
B
A
C
C
A
A
Any
point
to
interrupt
me
and
say:
I
know
a
better
way
of
doing
that.
In
this
instance,
you
may
already
have
the
project
owned
locally
I.
Don't
the
www
project
is
quite
large,
so
I
will
just
make
changing
the
the
web
IDE
so
in
here
somewhere,
it
will
say
#hack
phone.
Yes,
it
does
so
it
will
open
this
page
in
the
web
ID.
If
you
don't
already
have
access
to
the
GitHub
community
group
yeah,
and
here
it
tells
you
for
everyone
how
to
get
access.
A
C
C
C
B
Not
everyone
who
may
review
the
change
might
understand
the
reasonings
behind
it.
So
mentioning
that
in
the
mode
request,
template
or
at
least
putting
as
much
information
as
possible
in
the
mode
request,
helps
the
reviewers
to
figure
out
exactly
what
you're
trying
to
achieve
here
and
why
you're
trying
to
achieve
it.
A
I
did
do
some.
What
should
I
call
it?
Empathetic
merge
requests
at
the
weekend
using
a
non-team
member
account
and
going
through
the
standard
process
using
the
bot
to
assign
it
so
doing
a
bit
of
both
I'm.
Just
gonna
wait
for
the
there
you
go.
The
change
has
gone
through.
I'm
gonna
do
self
review,
which
I
always
encourage,
because
in
a
lot
of
times,
I
made
a
change,
pushed
it
up
and
then
reread
it
and
it's
something
that
I've
made
a
mistake.
A
D
A
E
A
D
C
A
I've
jumped
ahead.
I
do
need
to
go
back
to
the
one
that
we
were
talking
about
a
minute
ago,
but
this
one
I
want
to
just
have
a
look,
what
the
right!
So
the
first
thing
is
it.
This
extends
the
community
process
of
Base
Class
and
the
community
process
that
yeah.
He
checks
that
it's
a
wider
Community
contribution.
So
this
spot
isn't
going
to
listen
to
me
because
I'm,
not
a
member
of
The,
Wider,
Community,
I'm,
a
team
member,
but.
B
A
B
That
the
the
community
contribution
label
doesn't
exist
on
the
on
the
mode
request,
so
the
there
is.
There
is
another
process
that
happens
that
adds
the
Community
contribution
label
once
that
Community
contribution
label
is
there,
then
many
of
the
other
bot
actions
will
will
fire
on
that
merge,
request,
yeah.
A
A
The
more
that
we
can
do
to
to
emulate
the
better.
So
this
is
the
account
that
we're
using
at
the
weekend-
and
here
we
can
see
this
Mr-
that
I
raised
and
it
it
did
the
usual
hey,
eventually,
contribution
and
I
use
the
ready
command
and
it's
assigned
to
someone
somewhat
randomly
off
the
back
of
that.
So
maybe
do
I
want
to
so
I've
already
raised
this
change.
C
E
A
E
A
And
there
is
some
debugging
that
team
members
can
do
troubleshooting
that
team
members
can
do
from
the
logs
in
gcp,
but
even
that's
quite
limited,
definitely
to
the
level
of
being
able
to
specifically
see
each
kind
of
what
I
call
it
criteria
and
and
why?
What
evaluated
and
why
it
did
or
didn't
get
triggered
would
be
really
helpful,
because
at
the
moment,
what
I
tend
to
have
to
do
is
something
that's
not
working
a
reverse
engineer
it.
A
The
way
we
just
did
there
I
have
a
look
at
the
rules
and
see
why
I
think
it
may
or
may
not
have
worked,
and
the
best
I
could
then
do
would
be
to
run
it
locally.
So
pull
down
the
the
triage
Ops
project
and
run
it
locally,
and
you
know,
try
and
emulate
and
and
debug
it
that
way,
but
yeah.
It
would
be
nice
to
to.
E
A
C
C
A
E
A
D
A
Triage
Ops
merge
requests
in
the
triage
Ops
projects
do
count,
but
we're
not
counting
anything
outside
of
merge
requests
at
the
moment.
It's
really
interesting.
You
should
mention
that,
because
it
came
up
both
on
a
call
yesterday
and
I
I
have
a
a
one-to-one
with
with
Raymond
earlier
today
and
something
that
we've
been
wanting
to
do
for
a
long
time
and
and
hopefully
we'll
get
to
do
eventually
in
the
future
is
start
to
pull
in
more
gitlab
data.
So
issues
comments,
reactions
things
like
that,
but
also
data
from
wherever
a
field.
A
So
you
know
the
the
help
that
you
provide
in
Discord
is
valuable.
The
help
that
is
provided
in
the
Forum
on
Reddit
on
you
know
other
even
social
media
Etc.
So
there's
there's
definitely
a
desire
to
be
able
to
pull
more
data
from
more
sources
going
forwards
and
I
would
love
that
to
be
a
Target
we've
kind
of
set
some
of
our
objectives
for
the
following:
the
quarter,
two,
which
for
us
is
May
June
July.
A
So
so,
for
both
to
be
honest,
I
think
that
at
the
moment
we
even
for
like
the
MVP,
whilst
we
consider
somebody's
contributions
as
a
whole,
the
primary
driver
is
based
on
merge
requests.
So
that's
for
the
the
MVP
decision
process,
this
leaderboard
here
the
hackathon
scoring
the
there's
some
bit
Azure
dashboards.
There's
lots
of
everything
really
is
driven
off
of
merge
requests
and
merge
requests.
Alone
and
all
of
those
I
would
like
to
extend
going
forwards
to
be
to
consider
more
of
those
data
sources.
A
E
A
So,
there's
there's
a
couple
of
a
couple
of
questions
that
I'm
going
to
ask
myself,
but
I'll
also
ask
you
and
I'm
Keen
to
understand
your
view
on
it,
so
the
the
way
that
this,
in
fact
I'll
go
back
to
it.
Now
the
the
way
that
this
leaderboard
gets
its
data
is
through
graphql,
so
it
makes
queries
to
the
the
gitlab
org
and
the
gitlab
com,
Brew
or
merge
requests
which
were
created
within
the
hackathon
window
and
it
also
checks
for
the
hackathon
label
as
well.
A
A
So
it
looks
like
we've
got
a
count
here
which
helps
a
little
bit,
but
it's
still
still
doesn't
help
a
huge
amount,
because
we
still
have
to
execute
this
graphql
query
for
the
GitHub
org
and
the
gitlab
com
group,
and
then
we
would
have
to
add
those
numbers
together.
So
it's
one
way
we
could
do
it
so
the
other
way
and
the
way
that
the
the
hackathon
this
page
is
actually
doing
it.
We
can
even
do
some.
C
A
C
A
So
this
tracking
issue
here,
every
time
a
merge
request
is
opened,
the
bot
references,
the
hackathon.
So
it
says
that
the
merge
Professor
will
be
counted
as
part
of
the
hackathon
and
it
links
to
this
issue.
So
by
doing
that,
we
end
up
with
this
this
issue,
which
tracks
all
of
the
merge
requests-
and
this
is
really
interesting
actually
because
I
can
immediately
see
here
as
well,
that
we've
got
one
two
three
four,
but.
A
Per
page
X
total
there
you
go
so
the
the
response
header
from
a
single
API
called
gets
us.
The
total
number
of
merger
requests
so
yeah
we've
we've
got
two
options.
We
could
add
that
API
call
to
that
endpoint
to
get
that
number
or
we've
actually
already
got
the
data
from
this,
but
we
need
to
spend
a
little
bit
more
time
kind
of
collating
and
manipulating
that
number
do
you
do
you
have
an
immediate
thought
on
your
preferred.
E
Well,
I
would
prefer
to
have
this
query
documented
somewhere
from
comparison,
but
the
hack
with
the
issue
and
Counting
marriage
request
in
the
issue.
It
sounds
more
logical
for
hackathons,
okay,
so
because
well
in
the
long
term,
you
still
need
to
count
the
monthly
contributions
the
merge
requests
from
community
so
that
that
will
be
useful
and
bitter
glia.
Does
this
already
I
believe,
but
but
I
it
probably
uses
some
kind
of
database
dump
dump
I'm,
not
sure
how
bitter
glare
works.
It
probably
doesn't
use
like
API
directly,
because.
A
A
Great
before
okay,
okay
I
thought,
the
number
seems
quite
high,
but
it's
because
this
isn't
merged
merge
requests.
This
is
only
created,
merge
requests
and
I've
actually
changed
it
to
not
be
able
to
filter
on
that.
Okay,
fine,
because
I
know
we
get
roughly
we
get
closer
to
100
and
between
100
and
120
unique
authors
each
month,
but
that's
based
on
the
number
of
merged.
Merge
requests,
not
the
number
of
creative
badge
requests
so.
E
E
How
many
requests
are
not
merged,
like
these
statistics?
Yeah
I
already
feel
like
this
is
repeating
the
the
turquoia
dashboard,
but
the
turquoit
dashboard
is
interface
is
too
overwhelmed
with
the
with
with
digits.
It's
not
really
a
catches,
and
you
people
probably
will
have
hard
time
trying
to
figure
out
what
what.
A
A
E
Yeah
well
another
open
source
index,
one
of
the
Outsourcing
companies.
They
build
the
open
source
contributions
index
and
they
used
commit
emails
to
see
which
are
the
company
contributions
yeah
and
they
usually
like
the
same
dashboard.
Basically
like
companies
and
rated
by
the
amount
of
comments
made
I
think
it's
we're
like
this
dashboard
is
replicating
some
of
the
universal
Solutions,
but
maybe
Universal
Solutions
are
not
good,
because
here
you
have
to
to
make
like
keep
it
simple
and
make
comments
and
that's
the
spirit
of
the
hackathon.
E
A
So
what
we're
gonna
I
would
love
someone
who's
good
with
ux,
UI
and
Design
By.
The
way
to
to
you
know
these
text
boxes
I'll
give
us
how-
and
you
know
this
and
but
I'm,
definitely
not
going
to
say
that's
one
of
my
stronger
points
at
the
moment,
so
I'm
happy
for
someone
else
to
to
contribute
that,
as
of
when
so,
let's
just
slip
in
a
a
box
somewhere.
So
if
loading
else
complete
so.
E
C
A
A
C
A
E
A
Yeah
yeah
I
I
come
from
a
different,
slightly
different
world
C
sharp
world
where
you
have
to
compile
and
build
and
wait,
and
but
this
is
awesome,
yeah
new
view
or
probably
even
older
view,
apps
that
they
have
hot
or
live
reload
whatever
you
want
to
call
it
so
as
soon
as
you
make
a
change,
I
think
that
it
uses
websockets
again
I'm
not
overly
qualified
when
it
comes
to
that.
But
it's.
A
Again,
I,
don't
know
the
specifics.
What
do
we
used
to
use?
There
was
the
old
view.
Cli
I
was
I
think
it
was
just
called
UCLA
that
we
used
that
we
may
be
using
the
gitlab
project
or
that
I've
used
in
the
past
to
build
a
view
apps.
But
more
recently,
this
this
vital
byte
I
think
is
a
the
off-the-shelf,
the
recommended,
well-documented
solution
and
yeah.
This
I
mean
it's
just
an
npm
command,
but
every
time
I.
C
A
A
C
C
A
C
C
A
E
E
C
C
C
A
E
A
C
C
A
E
C
E
A
E
E
A
C
A
A
E
C
A
C
A
E
A
A
A
So
I
did
think
about
having
columns
if
that's
kind
of
what
you
refer.
So
you
know
we
could
show
how
many
merge
requests
each
person
had
erased
as
well,
but
I
I
was
really
struggling
from
a
UI
ux,
a
design
point
of
view.
It
started
to
look
quite
horrible.
We've
already
got
this
big
number
and
this
big
number,
if
I
put
another
big
number,
it
starts
to
look
really
horrible
so
and
that's.
E
E
A
E
A
C
A
A
A
C
A
E
E
E
A
E
A
C
A
They
both
appear
Dash
to
me
so
I
think
it
zooms
compression,
that's
maybe
nice.
Actually,
it
looks
like
that.
Yeah
awesome
really
appreciate
you
jumping
on
and
as
Raymond
said
great
to
actually
see
your
your
face
for
the
first
time,
I
think
so.
Yeah
I
hope
we'll
be
seeing
you
over
the
next
couple
of
days.