►
Description
In this issue Jean mumbles to himself while he refines the issues relating to converting the GitLab handbook from Middleman to Frontman.
A
A
We
became
aware
of
the
frontman
project
a
few
months
ago,
things
to
romantically
or
bring
us
onto
it
and
back
then
it
wasn't
public.
Yet
it
was
developed
by
the
documentation
team
at
algolia
who,
similar
to
us,
was
using
middlemen
as
they
static
site
generator
and
random,
similar
to
us
into
performance
issues
and
went
on
a
journey
of
creating
their
own
stack
site
generator,
and
so
we,
I
had
a
look
at
their
video.
I
was
reminiscing
put
us
in
touch
long
story
short.
A
We
got
in
touch
they've
recently
released
an
alpha
version
of
it
and
we
are
hoping
to
migrate
for
the
handbook
specifically
now
that
we've
split
the
marketing
and
web
handbook
website
into
their
own
sites,
primarily
in
the
mono
request
structure.
There's
still
some
shared
assets
that
we're
going
to
work
to
move
away
from,
but
overall
the
risk
of
trying
to
at
least
migrate
to
frontline
is
low.
And
from
that
point
of
view-
and
you
know-
hopefully
it
can
bring
with
it
a
lot
of
performance
benefits.
A
A
So
while
it's
not
exactly
a
dropping
replacement
for
middlemen,
it's
it's
close
enough
that
you
know
the
algolia
team
converted
their
whole
site
to
frontman
in
three
days
and
they've.
They've
got
a
large
site
with
thousands
of
pages,
and
you
know
my
handbook
site
is
now
8
000
pages
chad,
that's
an
initial
kind
of
like
exploration,
and
one
of
the
first
things
that
we
need
to
do
is
move
away
from
custom
extensions.
That
middleman
allows
you
to
write
in
favor
of
using
helpers.
A
Frontman
does
not
support
custom
extensions,
but
they
do
have
helpers
and
based
on
initial
kind
of,
like
exploration,
doesn't
seem
like
this.
Anything
we
can
do
is
by
moving
to
helpers
instead
of
custom
extensions.
So
there's
a
problem
with
a
bit
of
a
discussion
about
it
and
so
there's
obviously,
some
prerequisites
that
we
need
to
do
in
terms
of
inverting.
The
existing
code
base
to
be
ready
to
even
attempt
the
conversion
to
the
front
brake.
A
A
A
One
thing
that
I
really
shared
previously
is,
if
we
go
into
the
extensions
we
currently
have
here,
have
a
bunch
of
them
that
are
specific
to
the
monorepo,
as
well
as
the
assets
that
we
that
we
have
shared
between
kind
of
like
the
the
handbook
in
the
marketing
site.
You
know
the
partial
build
that
we're
using
in
the
handbook
anymore.
A
A
A
A
A
A
A
A
A
A
A
A
A
I
will
say
that
something
all
risky
in
a
sense
of
you
know
we're
doing
a
bunch
of
work
to
and
then
we
might
not
end
up
migrating
to
frontline
chad
that
initial
kind
of
flight
wanted
validation
and
you
know
everything
seemed
feasible,
and
so
the
risk
we're
taking
is
a
measured
risk.
So
that
should
be
fine.
Let's
talk
about
him
from
library
to
reference.
A
A
A
A
A
A
Now,
considering
all
the
other
work
on
our
that
we've
planned
for
13.4,
it's
unlikely
that
I
will
get
more
than
the
migration
of
the
helpers
and
monitor
express
for
intel
front
fan.
What
I
do
have
looked
in
is
leave
this
as
13.4
for
now,
but
it's
likely
this
one.
We
read
move
over
to
35.
I
just
want
to
see
if
there's
anything
we
wanted
to
migrate
around.
A
A
A
So
currently
we
we
did
have
some
shared
assets
in
this
in
the
sense
of
images,
styling
scripts
and
also
even
partial
templates,
that
is
shared
between
the
handbook
and
the
marketing
site
and
considering
the
journey
that
the
marketing
team
is
on,
as
well
as
the
journey
we're
on
it's
going
to
be
important
that
we
break
that
up,
even
though
it's
loosely
defined,
I
guess
at
this
stage
we
need
to
break
that
going
forward.
It's
just
going
to
cause
more
trouble
than
to
work,
even
if
it
means
duplication
in
the
thought
all
right.
A
A
A
A
A
A
A
A
A
A
A
A
So
essentially,
what
we'll
end
up
with
is
most
of
the
shared
assets
is
probably
related
to
the
marketing
site.
It's
going
to
be
corrected
on
that
point,
but
I
feel
fairly
confident
in
making
that
statement,
and
so
I
think
it
makes
sense
to
just
say:
look
remove
everything
there
and
from
there
you
can
do
a
clean
up.
A
A
A
A
A
A
A
But
considering
the
context
and
where
we're
heading,
I'm
sure
he'll
come
around
to
see
my
way
on
this
one.
This
is
the
hammer
approach
rather
than
the
surgeon
approach,
but
if
we
try
and
comb
for
everything
now-
and
it
just
feels
like
a
a
hard
problem
to
solve
that,
we
could
rather
do
this
okay
right.
So
if
we
go
back
to
our
epic
here,
refresh
make
sure
it's
everything
we
need
to
completely
decouple
the
handbook
from
market
size
car.
A
So
all
of
these
decoupling
converting
as
well
as
so
these
four
things
here
can
literally
happen
in
parallel,
so
we
can
have
any
team
who
picked
this
up.
This
one
is
obviously
dependent
on
these
getting
into
place.
So
that's
one
of
the
nice
things
that
we
can
share
the
load
here
and
then
once
that's
all
in
place
and
chat
can
run
with
this
one,
I'm
very
happy
with
what
we
have
here
and
I
think
yeah.
This
seems.