►
From YouTube: Complex Drupal 7 to Backdrop CMS Upgrade Case Study with Stanford University and Aten Design Group
Description
This session is from Backdrop LIVE July 15, 2022. Eric Toupin from Aten Design Group shares his experience upgrading a complex Drupal 7 site to Backdrop CMS. Eric estimates that using Backdrop was 20% of the cost of upgrading the site to Drupal 9.
A
Cool,
so
let's
see,
first
of
all
thanks
everyone,
so
much
for
taking
the
time
to
be
in
this
one
today,
of
course,
give
me
one
second,
and
let
me
see
if
I
can
share
my
screen
and
get
us
started.
A
Sweet,
so
my
name
is
eric
coopen.
I
have
worked
with
at
design
group,
I'm
a
senior
developer.
There
been
there
for
more
than
15
years.
Also
the
lead
developer
for
two
projects.
I'm
going
to
talk
about
today,
solo
and
ssf,
that's
stanford
on
and
off
campus
learning
opportunities
and
stanford
seed
funding.
A
A
So
today,
what
I'm
going
to
talk
about
is
triple
seven
to
backdrop:
cms
upgrades
specifically
for
complicated
applications
and
a
little
bit
about
what
those
are
like
what
I
mean
by
complicated
and
then
how
we
approach
it
from
estimation
and
execution.
The
reason
this
might
be
interesting
to
some
folks
is
because,
when
we
approached
these
projects,
it
was
myself
and
joel
steitle,
the
director
of
engineering
at
design
group
that
actually
did
the
bulk
of
the
work.
A
A
Okay,
so
I
threw
a
slide
in
here
with
the
the
disclaimer
on
it,
and
basically
all
I
mean
by
this
is
that
this
is
purely
experience
sharing.
I
would
certainly
not
call
this
a
prescriptive
approach
as
much
as
here's
what
we
did.
It
went
pretty
well
for
us,
and
hopefully
that's
useful
for
somebody.
A
So
the
things
that
I'm
planning
on
talking
through
today,
first,
what
do
we
mean
by
a
complex
application
and
I'll
talk
a
little
bit
about
solo
and
ssf
and
and
try
to
qualify
what
we
mean
by
that
and
then
specifically
how
we
were
able,
how
we
came
to
choose
backdrop
cms
and
again.
I
think
this
is
interesting
because,
well,
I
wouldn't
say
at
in
design
group
is
a
drupal
shop.
We
do
tons
of
drupal
work.
A
A
Probably
a
lot
of
folks
are
familiar
with
that,
so
I'm
not
going
to
spend
much
time
there
and
then
some
information
about
estimation
and
execution.
This
is
going
to
be
extremely
high
level,
not
diving
into
the
specifics
of
execution
stuff.
Just
talking
about
general
strategies,
so
this
is
stanford
on
and
off
campus
learning
opportunities.
A
So
if
you
visit
those
sites,
solo.stanford.edu
and
seedfunding.stanford.edu,
you
can
click
around.
You
can
see
some
opportunities
and
about
page,
but
you
can't
really
see
much
if
you're,
not
an
authenticated
user,
and
you
can't
really
authenticate
unless
you're
stanford.
So
what
I'd
like
to
do
is
just
show
some
slides
of
the
back
end
and
talk
a
little
bit
about
some
of
the
features
of
these
websites
to
really
qualify.
A
What
we
mean
by
by
complicated
the
screens
that
we're
going
to
walk
through
will
show
a
user
called
a
program
officer
that
would
be
creating
an
opportunity
and
an
application
form
to
apply
for
that
opportunity,
a
custom
application
form
and
then
collecting
applications
and
managing
those
applications
and
eventually
trying
to
award
some
folks
the
opportunity.
An
opportunity
on
this
website.
The
stanford
on
and
off
campus
learning
opportunities
website
would
be
an
educational
opportunity
opportunity.
So
go
work
at
a
hospital
in
china.
A
Go
work
in
a
manufacturing
plant
in
taiwan
or
go
intern
somewhere
like
learn
about
whatever
it
is
that
you're
that
you're
studying
as
well
as
other
like
educational
experiences?
Besides
internships,
the
other
site,
seed
funding
would
be
more
about
faculty
members
looking
for
research
funding,
and
so
these
would
be
opportunities
to
have
your
research
funded
and
they
will
be
applying
for
funding,
but
we're
going
to
stick
with
just
talking
about
solo
today,
just
because
there's
not
much
time
for
both
of
these.
A
So
I'm
going
to
turn
this
on
auto
play
real,
quick
and
just
go
through
some
of
these
slides.
While
I
talk
about
some
of
these
features,
okay,
so
this
should
just
kind
of
run.
I
think
the
most
important
takeaway
is
that
we
started
these
projects
in
2015
and
they
have
been
under
active
development
since
and
so
every
month
or
two
months
at
the
longest,
except
for
the
break
to
upgrade
the
site,
we
have
been
releasing
brand
new
features
to
our
audiences
they're
still
under
active
development.
A
I
release
new
features
this
week
on
wednesday
to
both
sites,
so
we've
been
building
these
for
quite
a
long
time.
These
two
websites
facilitate
complex
application
management,
workflows
that
were
developed
specifically
for
stanford
faculty
staff
and
students
they're
highly
customized.
We
use
more
than
60
contrib
modules,
more
than
30
custom
modules
that
we
wrote.
We
have
two
custom
api
integrations.
A
So
what
we
mean
there
is
just
as
program
officers
collect
dozens
or
hundreds
of
applications
and
there's
you
know
they
create
these
custom
forms,
there's
10
different
form
or
field
types
you
can
use
in
those
forms
and,
as
folks
fill
those
out,
you
can
then
download
those
as
kind
of
on
the
fly
rendered
pdfs
and
you
can
use
an
automated
download
manager
to
package
those.
So
you
have
you
know
you
download
10
at
a
time
and
send
them
off
to
one
reviewer
or
20
and
send
them
to
another.
A
Both
of
these
sites
use
five
distinct
user
roles,
there's
actually
a
little
more,
but
the
the
major
user
roles
are
program
officers
who
create
the
opportunities
applicants
who
apply
to
them
reviewers.
So
you
can
create,
create
custom
review,
rubrics
to
tell
people
how
to
score
your
applications
and
then
invite
people
to
review
them
and
score
them.
A
Student
program
officers
program
officers
can
recruit
students
to
help
them
with
management
tasks
recommenders,
who
would
like
upload
recommendation,
letters
or
other
assets
to
recommend
applicants
and
all
of
those
have
unique
experiences
and
unique
permission
sets,
etc.
We
also
have
a
customizable
notification
system.
We
send
a
couple
of
dozen
different
notifications
in
designed
emails
and
in
dashboard
notifications
on
the
actual
platform.
A
Saying
things
like
you
know,
you
start
an
application.
You
didn't
finish
it
your
application's.
Due
in
a
week
your
application
is
under
review.
You've
been
rewarded.
This
person
started
an
application
on
your
opportunity,
tons
of
stuff
that
just
kind
of
pushes
people
along
the
workflows.
I'm
going
to
pause
this
real,
quick,
the
slideshow
yeah.
A
We
have
more
than
18
000
stanford
users
on
these
sites
and
we
have
advertised
more
than
a
thousand
opportunities
since
the
beginning
of
of
the
projects
back
in
2015,
which
represents
tons
of
student
experiences,
educational
experiences
and
also
incredible
amounts
of
research
funding
monthly.
We
receive
as
many
as
20
000
visits
on
these
and
serve
as
many
as
around
275
000
pages.
A
Okay.
So
the
reason
that
I
walk
through
that
whole
thing
is
really
just
to
talk
about
what
we
were
tackling
here
in
terms
of
an
upgrade
and
really
the
focus.
There
is,
is
years
and
years,
seven
years
of
development,
active
development,
active
user
base.
And
what
do
you
do
in
a
case
like
that,
when
all
of
a
sudden
drupal
7
is
at
risk
of
end
of
lifeing?
A
So
now
I'm
going
to
talk
a
little
bit
about
why
we
chose
backdrop,
as
I
mentioned
before,
we
do
tons
of
drupal
work
and
definitely
our
kind
of
go-to
plan
would
have
been
upgrading
to
drupal
9.
and
it's
one
of
the
first
things
we
started
talking
about.
There's
a
couple
of
interesting
factors
here:
one
is
the
on
the
stanford
side,
the
reedy
team,
the
director
of
that
team
is
zach.
A
Chandler
and
zach
was
familiar
with
backdrop
and
familiar
with
some
of
the
folks
involved
with
backdrop
and
so
right
from
the
beginning
and
and
he's
also
a
tech,
savvy
tech
savvy
guy
and
knew
what
we'd
been
building
and
knew
that
rebuilding
and
drupal
9
was
going
to
be
a
huge
huge
project
and
so
right
from
the
beginning,
he
just
kind
of
threw
out
the
idea
like
hey.
Why
don't
we
look
in
the
backdrop
and
see
if
that's
something
we
could
do,
and
he
thought
maybe
we
could
save.
A
You
know
80
of
the
cost
of
the
project,
and
so
his
reaction
to
to
that
was,
let's
talk
about
it
as
a
team,
let's
figure
out.
If
this
is
something
that
we
can
actually
do,
something
that's
viable.
A
So
then
it
came
to
me
to
kind
of
do
that
investigation
part
and
see
if
it's
viable
and-
and
actually
I
had
been
like-
I
said
earlier
kind
of
piddling
around
with
backdrop
for
a
couple
of
months.
I
built
some
personal
sites
in
it,
helped
a
friend
build
a
small
site
with
it,
mostly
because
I
was
interested
myself
and
what
I
was
going
to
do
with
drupal
7
sites
and
also
just
because
I'm
really
comfortable
in
a
lot
of
the
the
existing
drupal
7
apis
that
I've
been
building
in
for
so
long.
A
Also.
For
me,
I
really
do
love
coding
and
spend
much
of
my
day
every
day
coding
really
enjoy
that
part
of
my
job,
but
also
really
more
than
anything.
Programming
is
a
means
to
an
end.
I
like
to
deliver
value
to
clients.
I
like
clients
to
be
excited
about
where
we're
going
with
technology
and
so
for
me
saving
that
seven
years
of
coding
work
seemed,
of
course,
like
an
excellent,
an
excellent
opportunity.
A
Another
thing
here
is
that
we
had
over
the
years
begun
building
tests
back
then
they
were
in
selenium
now
they're
in
cyprus
to
cover
a
lot
of
these
features,
because
it's
a
real
small
team.
So
it's
a
five-person
team
in
total
managing
this
product.
With
you
know
thousands
of
users
and
over
the
years
it's
really
hard
to
keep
track
of
all
those
features,
and
so
we
have
these
tests.
A
I
haven't
seen
that
feature
in
like
two
years
and
with
that
experience
it
worries
me
quite
a
bit
to
rebuild
a
platform
completely
in
drupal
9,
because
anything
that
you
don't
have
documented
or
you
don't
have
really
clear
context
on
how
it's
working
could
really
easily
slip
through
the
cracks,
and
that
was
really
worrisome
for
me.
A
Okay,
so
then,
a
little
bit
of
why
you
would
upgrade
to
backdrop
instead
of
rebuilding
in
a
different
platform
or
in
the
newest
version
of
drupal-
and
I
think
some
of
this
stuff
is-
is
pretty
obvious
to
everyone
here.
Maybe
not,
but
I
I
did
do
a
couple
of
talks
on
this
previously
specifically
about
why
upgrading
to
backdrop,
cms
is
a
really
good
idea
versus
drupal.
In
some
cases,
if
you
google,
at
in
webinar
backdrop,
you
can
find
my
talk
on
that.
A
If
you
are
not
yet
convinced-
and
I
go
over
a
lot
of
the
specifics
in
that
one,
but
on
this
one,
I'm
just
going
to
like
extremely
high
level
summarize,
what
kind
of
our
decision
making
process
was
there.
A
So
the
first
thing
is
that
the
cost
is
dramatically
different.
We
estimated
between
4
500
and
5
000
hours
to
rebuild
these
platforms
in
drupal
9.,
and
I
think
that
was
a
really.
A
That
was
a
really
optimistic
estimation
and
at
least
in
my
opinion,
when
we
actually
finished
the
project
in
backdrop,
there
was
a
lot
of
risk
going
in
because
we
didn't,
we
didn't
know
what
we
were
getting
into,
but
at
the
end
of
it
we
ended
up
spending
550
hours
to
complete
everything,
and
so
you
can
see
that
the
you
know
the
cost
difference
is
just
absolutely
huge.
A
Another
thing
is
lower
risk.
I
mentioned
this
a
second
ago,
but
when
you,
when
you
upgrade
to
backdrop,
you
don't
lose
your
code
base
and
you
don't
lose
your
content,
you
don't
lose
your
data
structures,
so
normally
with
the
rebuild
you'd
have
to
migrate.
Your
users
migrate,
your
nodes,
migrate
terms,
re-establish
all
of
that.
Those
data
structures,
the
connections
between
all
of
those
and
with
an
upgrade
to
backdrop.
You
don't
have
to
do
that,
and
so
a
lot
of
those
structures
are
preserved.
A
So
then
timeline
is
pretty
obvious.
If
you
think
about
cost,
but
timeline
here
is
going
to
be
extremely
shorter.
Compared
to
the
timeline
for
a
rebuild.
We
were
able
to
use
far
few
developers
and
far
less
time
than
we
would
have
used
for
a
full
rebuild
and
then
some
ongoing
costs
and
this
one
doesn't
apply
if
you're
thinking
about
upgrading
somebody
else's
site
that
you
didn't
build.
A
But
if
it's
your
own
site,
if
you've
been
building
in
drupal
7
or
your
team's,
been
building
in
drupal
7,
and
then
you
upgrade
to
backdrop
all
of
the
infrastructure
that
you
built
is
there
and
any
like
any
helper
functions,
any
anything
that
you've
built
is
kind
of
put
together
for
you.
A
A
Yeah,
so
if
you're
building
in
in
a
code
base
that
you've
been
working
in
drupal
7,
and
you
upgrade
basically
all
of
the
things
that
you
have
built
for
yourself,
all
the
little
helpers
that
you've
built
for
yourself
and
your
approaches
are
still
there
and
so
what
we
would
call
you
know
velocity
and
kind
of
the
the
agile
speak
of
the
more
you
work
on
your
code,
the
faster
you
work
on
your
code.
That
stuff
survives.
A
Okay,
so
let's
talk
a
little
bit
about
about
estimation-
and
I
want
to
kind
of
forewarning
here
that
this
is
a
little
bit
of
like
black
magic,
as
estimation
often
is
at
least
to
me.
One
big
thing
here
is
the
estimation
is
extremely
unique
per
project
right,
and
so
this
is
going
to
change
quite
a
bit
from
from
one
website
to
another,
from
one
application
to
another.
A
There's
some
knowns
and
I
think
that's
where
we
start,
but
there's
also
tons
of
unknowns
and
that
that's
true
in
your
own
codebase,
if
it's,
if
you've
been
building
it
for
years,
and
it's
doubly
true
in
somebody
else's
code
base
that
you're
now
taking
on.
I
wouldn't
quite
say
it's
a
shot
in
the
dark
to
estimate
on
projects,
but
it's
extremely
extremely
difficult
to
get
an
accurate
number,
a
precise
number
and
the
difficulty
is
that
you
can
say
those
things
to
clients.
A
A
A
The
more
precise
your
estimation
is,
but
you
know
that's
a
that's
an
interesting
dynamic,
because,
if
you're,
if
you're
bidding
on
a
project,
you
can't
look
at
the
code
for
two
weeks,
because
that's
time
lost
you're
not
getting
paid
for
that
time
and
even
if
you
aren't
bidding
on
a
product
even
if
it's
a
current
client
that
you're
in
a
you
know
high
confidence
relationship
with
and
they're
like
hey,
I'm
really
interested
in
learning
more
about
backdrop,
can
you
estimate
that
for
me
it
might,
even
in
that
case,
be
a
hard
sell
to
say,
yeah.
A
A
There
are
lots
of
unknowns,
but
there
are
also
plenty
of
knowns
and
I
think
the
really
obvious
knowns
are
contrib
modules,
custom
modules
and
theme,
and
then
I'd
go
beyond
that
and
I'll
spend
a
little
more
time
on
these
in
a
second,
but
also
testing
testing
is
extremely
important,
and
you
know
we'll
talk
about
more
about
that
in
a
second
actually
and
then
also
time
for
for
dry
runs
so
at
some
point,
you're
going
to
need
to
deploy
the
site,
and
we
need
to
practice
that
and
again
I'll
return
to
that
in
a
second.
A
So
for
these
first
couple
of
things
for
contrib
modules,
specifically
there's
a
module
called
backdrop
upgrade
status.
That
is
provided
primarily
by
the
backdrop.
Folks-
and
this
is
I
mean
you
want
to
use
this
right.
This
gives
you
an
absolutely
easy
overview
of
what
your
work
is
going
to
be.
So
you
install
this
module
the
backdrop
upgrade
status,
module
on
your
drupal
7
site,
and
this
is
a
local
copy
of
my
of
one
of
these
sites.
Actually,
the
old
drupal
7
version
that
I'm
looking
at.
A
I
can't
actually
click
any
links,
because
I
think
I
I
think
my
virtual
machine
is
no
longer
running,
but
so,
if
you
run
this,
if
you
install
this
module-
and
you
click
like
you
know-
check
the
status
of
my
site-
real,
easy
interface
and
it
goes
through
a
little
progress
bar
and
then
it
spits
out
this
screen,
and
this
screen
shows
you.
Okay,
admin
views
is
now
in
core
and
off
cache.
That's
not
yet
available
and
automatic
node
tiles.
Yeah,
that's
a
module,
that's
available!
Being
that's
in
core!
A
So
you've
got
all
this
useful
information
about.
You
know
what
you
need
to
do
with
your
contrib
modules.
So
if
you
start
with
something
like
this,
and
then
we
moved
all
that
information
into
this
extremely
messy
spreadsheet,
which
I'm
a
tiny
bit
embarrassed
to
share.
But
you
can
see
that
you
know
we
have
this
module
column
and
we
throw
all
of
our
modules
in
there
and
then
we
move
over
some
of
the
information
that
we
get
from
this
backdrop
upgrade
status.
A
I
will
mention
that
even
if
something
says
not
yet
available
in
backdrop,
I
would
go
ahead
and
just
google
backdrop
cms
name
of
module,
you
might
find
there's
a
repo
out
there
of
somebody
working
on
it,
something
unofficial,
which
is
a
great
starting
point
right.
So
then
we
have
all
these
modules
and
we
have
their
status
you're
going
to
want
to
ignore
most
of
these
columns,
at
least
for
now.
A
But
basically
this
column
is,
does
it
exist
in
backdrop,
and
you
can
see
yeah
there's
plenty
in
core
got
lots
of
project
pages,
and
then
we
get
down
a
little
bit
farther
and
we
get
a
couple
that
aren't
now.
This
is
a
copy
of
a
spreadsheet
that
we
actually
used
last
year.
So
it
may
very
well
be
outdated.
If
you
see
any
module
in
here
that
now
exists.
Don't
get
hung
up
on
that.
A
This
is
kind
of
an
old
old
spreadsheet,
but
you
can
see
that
we
have
a
bunch
like
encrypt
or
colectomy
or
offenses
role.
Reference
send
grid
that
do
not
yet
exist
in
backdrop.
So
what
we
did
to
try
to
get
a
sense
for
the
size
of
the
work
ahead
of
us.
This
will
be
a
bit
of
a
rabbit
trail,
but
on
the
on
the
solo
and
ssf
project,
we
use
story
points
to
estimate
almost
everything.
The
story
points
is
like
this
agile
practice
thing.
I'm
not
sure
how
familiar
people
are
with
that.
A
A
Do
this
different
ways,
but
the
way
that
we
do
it
for
this
particular
project
is
we
stick
to
the
fibonacci
sequence
for
the
series
of
complexities
right,
so
it's
1,
2,
3,
5,
8,
13,
etc,
and
I
think
what
that's
supposed
to
portray
is
the
more
complex
something
gets
the
more
you
don't
know
how
complicated
it
is,
and
so
it
really
gets
like
like
real,
complicated,
real
fast
and
an
easy
way
to
think
about
this.
An
easy
way
that
I
think
about
this
anyways
is
like
playing
chess.
A
If
you're
playing
chess
with
somebody
and
you
study
the
board,
it
becomes
pretty
obvious
what
your
next
move
is,
and
so
it's
pretty
easy
that
one's
pretty
easy
to
guess.
But
if
you
want
to
guess
like
two
moves
in
advance,
what's
your
next
move
going
to
be
well,
a
lot
of
variables
are
introduced
and
you
become
real
uncertain
about
what
your
next
move
might
be
and
three
moves
ahead.
Like
maybe
chess
masters
think
three
moves
ahead.
I
certainly
can't,
and
so
I
think
that's
what
we're
kind
of
talking
about
as
those
numbers
get
higher.
A
So
what
we
did
is
joel,
who
at
this
time
had
never
touched
backdrop
was
going
to
do
our
first
module
conversion
and
we
chose
sendgrid
because
we
needed
the
module
and
we
we
thought
that
it
was
going
to
be
pretty
complex
to
convert
just
kind
of
picking
it
out
of
the
crowd,
so
joel
assumed
from
the
beginning
that
that
was
going
to
be
a
level
five
in
complexity
which
for
us
and
with
our
the
way
that
we've
historically
used
points
on
this
project
represents
an
extremely
complex
task
like
up
to
a
week
of
work
or
something.
A
A
He
finished
the
sendgrid
integration
conversion,
at
least
to
a
standard
that
was
working
for
all
of
our
needs
and
then
looking
back
on
it
in
retrospect,
you
can
see
over
here
in
the
complexity
column,
he
was
like
you
know.
Actually
in
retrospect,
that's
closer
to
like
a
three
which
for
us
would
be
like
a
day
to
a
couple
of
days.
So
then
he
had
a
little
backdrop
experience.
I
had
a
little
backdrop
experience.
A
We
went
through
each
one
of
these
contrib
modules
and
we
just
kind
of
browse
through
the
code
we
looked
at
includes,
and
we
tried
to
get
a
general
sense
of
like
what
do
we
think
this
is
going
to
be.
Is
it
a
one,
a
two,
a
three
or
a
five,
and
we
marked
every
one
of
these
that
needed
to
be
converted
with
a
number.
This
was
a
process
of
hours.
A
Okay.
So
then,
if
we
go
down
to
the
bottom
of
this
column
and
again,
this
is
not
like
really
a
presentable
spreadsheet.
But
if
we
go
down
here
and
add
all
this
up,
you
can
see
that
all
of
our
complexity
points
add
up
to
54
points,
and
then
we
can
also
see
that
sendgrid.
A
We
called
it
a
complexity
of
three
and
it
took
joel
about
a
day
and
a
half
to
get
things
all
working
and
all
in
place,
and
so
we
round
that
up
a
little
bit,
we
call
it
12
hours
and
we
get.
Three
points
is
12
hours,
which
means
one
point
is
four
hours.
We
do
a
little
bit
of
math
and
we
come
out
with
216
hours
for
converting
all
these
contrib
modules.
A
So
then
you
can
see
you
know
in
our
parentheses
back
on
this
on
these
slides
that
we
have
kind
of
hours,
estimates
next
to
everything
and
that's
how
we
came
to
the
contrib
modules
estimate.
I'm
absolutely
aware
that
this
is
a
little
bit
of
black
magic
and
I
think
that
the
goal
is
is
again
not
to
be
precise,
but
to
try
to
efficiently
get
to
a
number
that
it's
in
the
right
ballpark.
A
So
then
we
look
at
custom
modules
and
I
think
that
if
you
were
upgrading
a
site,
that's
not
yours.
I
would
use
a
similar
process.
So
now
we've
been
on
a
couple
of
other
backdrop
conversion
sites,
and
this
is
the
process
that
I
use,
but
with
their
custom
modules
as
well.
A
In
in
my
case,
I
had
written
the
majority
of
the
custom
modules
for
this
platform,
and
so
I
felt
pretty
confident
that
I
could
spend
just
a
couple
of
hours
in
each
and
and
be
done
in
about
two
weeks.
A
So
then
we
come
to
the
theme
and
with
the
theme
we
used
a
really
similar
process.
So
joel
did
the
first
theme
conversion
here
and
all
we
did
is
we
picked
one
page,
an
opportunity
page
from
that
seed,
funding,
site
and
said:
okay?
What's
it
gonna
take
to
get
one
opportunity
page
to
look
right
on
backdrop
and
we're
not
talking
about
back
end
here.
A
So
when
we
look
at
like
the
node
template
everywhere,
where
it
says
print,
title
or
print
content,
we're
taking
all
that
stuff
out
and
just
putting
in,
like
you
know,
typing
in
content.
We
just
want
to
see
if
the
template
looks
right
and
getting
that
template
to
look
right
for
an
opportunity.
Page
just
took
a
couple
of
hours
for
joel,
and
so
he
looked
at
the
rest
of
the
theme
and
said:
hey,
you
know.
I
think
I
can
do
this
in
about
a
week,
so
we
have
40
hours
there.
A
A
Okay,
so
then
testing,
we
had
a
large
suite
of
selenium
tests
that
we
had
built
and
we
ended
up
during
this
pro
project,
actually
converting
them
to
cypress.
But
that
wasn't
the
plan
from
the
beginning
and
the
thing
that
those
sony
m
tests
gave
us
was
a
really
great
framework
for
for
testing.
A
Some
things
had
to
change,
because
the
markup
changed
between
drupal
and
backdrop,
and
so
the
tests
that
we
use
do
target
specific
markup
structures,
but
because
we
had
all
those
tests
already
in
place,
getting
them
to
run
on
the
new
platform
and
then
actually
running
through
them
was
not
extremely
time
intensive.
A
Okay
and
then
the
last
thing
that
I'll
mention
here
is
time
for
for
try
runs
so
this
this
process
took
us
three
months.
It
was
actually
10
weeks
because
we
took
some
vacation
in
december,
but
it
took
about
10
weeks,
and
you
know
the
culmination
of
all
of
your
work
is
that
you
have
your
backdrop
project
ready
and
then
you
drop
in
your
drupal
7
database
and
you
should
be
able
to
just
run
the
upgrade
process
like
through
the
point-and-click
interface.
A
The
reality,
for
us
at
least,
was
that
there
was
a
few
other
steps
we
had
to
disable.
This
module
was
throwing
errors
until
we
get
this
module
to
work
and
then
enable
it
again,
etc.
So
there's
some
practice
that
we
had
to
go
through
to
get
that
point
and
click
process
to
really
work.
The
way
we
wanted
it
to
and
for
us
we
needed
to
to
move
from
drupal
7
to
backdrop.
A
In
a
matter
of
hours,
we
were
going
to
have
four
hours
to
move
these
projects
from
live
on,
drupal
7,
to
live
on
backdrop
at
some
point
and
so
for
me
practicing.
These
dry
ones
was
extremely
important
and
if
you're
in
a
similar
position
where
you
need
to
deploy
quickly
at
the
end
of
this,
do
the
practice
and
just
feel
really
confident
with
that,
so
that
you
don't
feel
too
under
the
gun
when
the
time
comes
to
you
know,
do
it
quick?
A
Okay,
so
all
these
numbers
together,
you
do
a
little
math
here
and
we
come
up
with
a
total
of
416
hours.
I
mentioned
earlier.
We
actually
took
550
hours,
that's
a
little
bit
bigger
than
416..
A
We
did
some
other
stuff
in
there.
That's
not
backdrop
related.
We
decided
to
move
to
pantheon.
We
had
to
do
some
code
changes
there.
Like
I
mentioned
earlier,
we
converted
selenium
tests
to
cypress
tests,
so
there's
some
other
work
kind
of
in
that
that
larger
bucket,
but
we
came
in
in
the
right,
ballpark
right
and
if
I
had
to
do
it
again,
even
though
I
felt
like
those
were
pretty
generous
numbers
to
begin
with,
if
I
had
to
do
it
again,
I'd
I'd
had
more
generously
I'd,
add
another
30
onto
this
whole
process.
A
I
would
say,
if
you're
estimating
pad
as
much
as
you
feel
comfortable
presenting
to
your
client
and
hopefully
that
gets
somewhere
around
30.
A
Okay,
so
now
we'll
talk
a
little
bit
about
the
actual
execution,
what
it
looked
like
to
convert
these
platforms
again,
this
is
going
to
be
super
high
level,
like
I
said
before,
it
was
about
a
10
week
process
and
I'm
certainly
not
going
to
go
through
step
by
step
there
first
thing
real
important
is
again
coming
back
to
this
backdrop,
upgrade
status
module.
This
again
gives
you
a
great
list.
It
just
sets
your
work
out
in
front
of
you.
A
Obviously
this
doesn't
include
the
theme,
but
in
terms
of
backend
upgrade
process,
it
just
tells
you
what
you're
going
to
have
to
do.
We
moved
all
that
stuff
into
a
spreadsheet
that
I
already
mentioned,
and
then
we
tracked
our
own
progress
in
our
own
columns
like
hey,
how's
it
on
our
side.
Is
it
spot
checked?
Is
it
reviewed?
A
Okay,
so
then
also
coder
upgrade
is
a
module
that
you
absolutely
want
to
use.
So
there's
a
lot
of
api
changes
between
drupal
and
backdrop
that
are
formulaic
like
find
this,
replace
it
with
that
and
coder
upgrade
knows
all
of
those,
and
does
it
all
for
you,
and
so
really
I
mean
I
feel
like
you
would
just
be
wasting
your
time
not
to
use
coder
upgrade
like
definitely
use
it
for
many
modules.
A
So
what
we
did
and
I'm
interested
in
feedback
on
this,
but
what
we
did
was
we
set
up
a
backdrop,
sandbox
site,
just
a
fresh
install,
and
then
we
selected
the
modules
that
we
needed
to
convert
that
had
the
fewest
dependencies,
no
dependencies
or
already
built
or
defend
or
upgraded
their
dependencies,
and
we
moved
them
in
to
this
fresh,
install
site
and
tried
to
install
them.
A
What's
going
on,
you
fix
them
once
you
actually
get
it
to
install
effectively,
then
you
try
to
go
to
the
config
screens
and
see
if
those
are
working
and
if
there's
no
errors,
that's
awesome,
and
if
there
are,
you
start
debugging
again
and
figure
it
out
until
you
aren't
throwing
errors
once
you're,
not
throwing
errors
on
the
config
screens
or
the
install
screen
then
start
looking
at
the
actual
feature,
like
you
know,
role
reference.
Did
it
create
a
whole
reference
field?
A
Can
I
make
that
as
a
field
type
and
if
you
see
errors
or
if
it's
not
working
it's
time
to
debug
again,
otherwise
you
could
be
good
to
go
the
other
last.
No
no
two
weeks
ago,
I
converted
web
form
html
field
from
drupal
7
to
backdrop
for
for
the
same
project
actually,
and
it
took-
I
mean
less
than
three
minutes.
It
was,
it
was
absolutely
nothing.
It
was
changing
the
info
file
and
changing
like
one
place
where
it
said.
A
Drupal
and
shoulda
said
backdrop,
so
some
of
these
can
be
incredibly
fast.
Also,
I
spent
like
a
I
don't
know.
I
want
to
say
a
week
on
encrypt
encrypt
is
really
confusing
me
working
on
this
project
back
in
last
year,
so
it
really
varies
quite
a
bit,
there's
a
lot
of
gambling
in
there.
A
So
once
you
get
your
your
modules
working
in
your
kind
of
vanilla
site
and
your
fresh
install
and
then
you
start,
you
keep
adding
in
your
contrib
models
as
you
get
them
working,
adding
them
in
adding
them
in
and
once
you
have
your
whole
list
working,
then
we
move
that
into
our
you
know
our
converted
code
base
for
backdrop
and
try
to
spin
it
up
with
the
with
the
database
and
see
what
happens
if
you
do
convert
your
modules
and
you
are
interested
in
making
them
community
ready.
A
You
can
absolutely
use
code
or
review
to
try
to
help
get
things
back
up
to
standard
after
your
kind
of
potentially
messy
upgrade
process
theme
conversion.
There
are
some
pretty
easy
to
follow
theme
converting
docs,
this
stuff
laid
it
out
for
us
pretty
easily.
We
did
not.
I
mentioned
this
before
in
previous
talks.
We
didn't
take
full
advantage
of
the
layout
system
and
I
want
to
that's
something
that
I'm
hoping
to
do
later
this
year
for
these
platforms.
A
But
originally
our
goal
was
just
to
get
our
templates
working
and
the
information
in
these
documents
definitely
helps
you
get
there
pretty
quickly,
okay
and
then
I'll
just
mention
real
briefly
a
few
of
the
things
that
were
kind
of
big
things
to
overcome.
For
me
at
the
time
we
did
this
stuff
last
year,
I've
since
been
working
with
backdrop,
a
lot
more,
but
in
retrospect
context
was
a
big
one
that
I
didn't
plan
on
if
your
drupal
7
site
makes
extensive
use
of
context,
just
know
that
that's
complexity.
A
A
There's
there's
some
decisions
to
be
made.
The
way
libraries
is
included
or
are
included
rather
has
changed,
and
that
confused
me
a
couple
of
different
times,
and
so,
if
you're,
looking
at
modules,
you're
thinking
about
estimating
upgrade
and
they're,
including
libraries
things
like
whether
it's
javascript
or
php
libraries
we
use
like
in
pdf
library,
for
example,
a
simple
saml
like
a
bunch
of
stuff
and
that
that
was
a
little
confusing
as
well.
So
I
would
say
add
some
complexity.
Add
some
estimation
for
for
that
class
extensions
with
another
another
one.
A
That
kind
of
caught
me
off
guard
is
any
modules
in
drupal,
7,
contrary
or
custom
or
whatever
that
are
extending
classes
or
that
are
extending
some
like
native
drupal
classes.
I
would
pay
special
attention
there.
Some
of
those
classes,
they're
attributes
and
methods,
have
changed
a
little
bit,
and
so
that
can
also
be
kind
of
a
big
thing.
Like
oh
my
gosh,
I
didn't
realize
I
have
to
rewrite
these
class
extensions
for
this
module
and
then
one
final
little
gacho
was
user
roles.
A
Basically,
role
id
in
the
user's
roles
table
has
changed
and
that's
a
real
small
thing.
It's
not
that
big
a
deal
to
fix,
but
if
you're
doing
a
lot
of
evaluating
user
roles,
making
decisions
on
experience
or
permissions
based
on
that,
that's
definitely
something
to
to
pay
attention
to
and
pretty
fixable
with,
like
a
find
and
replace.
A
But
it's
nice
to
to
look
out
for
and
then
finally
the
way
this
actually
looked,
it
was
a
10
week
process
and
I
think
I
spent
about
six
weeks
of
it
just
on
my
couch,
this
couch
pouring
through
code,
just
looking
at
x,
debug
and
trying
to
figure
out
what's
going
on
and
trying
to
get
it
working
and
that
may
sound
tedious
to
a
lot
of
developers.
I
really
enjoyed
it.
A
I
found
that
it
gave
me
quite
a
bit
more
context
on
how
things
actually
work
and
on
how
things
interact
with
core,
and
it
was
you
know,
of
course,
really
irritating
at
times,
but
also
just
gave
me
a
lot
more
kind
of
breadth
of
understanding
of
how
the
the
platform
we
built
the
hooks
we
built
kind
of
interacted
with
the
rest
of
the
code,
so
I
would
say
yeah
plan
on
doing
a
lot
of
debugging,
the
the
short
of
it
for
the
outcome
of
these
projects.
A
They're
doing
great
we're,
building
new
features,
they've
been
on
backdrop
since
january.
The
client
is
absolutely
thrilled,
so
you
know
as
much
of
a
success
as
it
can
be.
It
is.
I
don't
want
to
talk
too
much
about
that,
and
then
let
me
try
to
see
that's
the
end
of
what
I
got
so
I'm
going
to
see
what's
in
the
chat
and
see
if
there's
anything,
I
can
add
some
more
light
on
here.
A
Actually
I'll
switch
you
over
to
this
here
also,
if
you
do
think
of
any
questions
comments
if
you're
interested
in
talking
more
about
this
stuff,
I
love
talking
about
it.
My
email
is
real
easy.
It's
eric
at
aton.io
so
feel
free
to
shoot
me
an
email
anytime,
I'm
going
to
read
through
some
of
the
stuff
in
the
chat,
real,
quick.
A
Okay,
so
this
is
this
is
real.
This
is
going
to
be
hard
to
talk
about
clearly,
but
well
you
you
talked
about
it.
Some.
A
Thanks
thanks
for
saying
that,
I
think
the
estimation
that
we
use
for
backdrop
is
similar
to
something
we
use
for
drupal
9.
One
of
the
biggest
differences
is
that
with
backdrop,
we
were
planning
on
keeping
the
same
theme
right
like
everything
was
going
to
be
exactly
the
same,
and
in
drupal
9
we
would
be
switching
to
the
drupal
9
theming
system,
and
that
would
almost
by
default,
at
least
in
our
agency
include
a
whole
nother
department,
which
is
the
information
architecture,
user
experience
and
design
departments
of
design
group,
and
so
by
default.
A
It
becomes
a
much
larger
project
because
we
would
start
with
things
like
design
discovery
and
design
comps
and
rethinking
the
whole
design
aspect,
and
that's
mostly
because
the
front
end
is
just
engineered
totally
different
in
drupal
9
than
it
is
in
backdrop,
but
other
than
that
extremely
similar
process.
We're
just
going
through
modules
and
going
through
features
and
myself
and
joel
and
a
few
other
developers
kind
of
brainstorming
about
what
we
think
it's
going
to
take
to
build.
It.
A
Use
block
explo
exposed,
filter
block
is
on
the
list.
Coder
upgrade
ports.
It
oh
interesting.
I
don't
have
any
additional
context
on
that.
The
list
is
a
little
bit
old,
that's
from
last
year
sometime,
but
we
whatever's
on
there
is
now
working.
So
I
should
take
a
look
and
see
if
there's
a
version
that
maybe
we
wrote
that
we
should
replace.
A
It
seems
about,
let's
see
that
seems
about
it,
any
questions,
any
other
thoughts
or
anybody
have
anything
else.
You
want
to
talk
about.
C
C
A
Yeah
absolutely,
I
also
wanted
to
call
out
ryan
said.
Another
in
pitfall
is
ctools
thanks
for
pointing
that
out.
I
didn't
put
that
in
my
list
and
that's
just
my
own.
I
forgot
to
absolutely
that
was
a
pain
using
anything
that
uses
c-tools.
Plug-Ins
was
another
thing
that
was
just
like.
Oh
my
gosh.
How
am
I
gonna
do
this,
but
absolutely
solvable
and
still
solvable
in
a
much
smaller
timeline
than
rebuilding.
Obviously,.
D
Yeah
definitely
pad
your
schedule
if
you've,
if
you've
discovered
c-tools
stuff,
even
if
it's
as
basic
as
as
modal
windows,
because
oh
man,
that
the
documentation
on
it
is
fair
and
it's
not
it's
it's
it's
not
easy.
I
will
say
that.
A
A
Cool
well,
if
there's
nothing
else,
any
other
questions,
yeah.
E
Quick
question
in
the
recording
that
you
did
of
for
the
drupal
camp
asheville,
which
I
watched.
You
mentioned,
that
a
lot
of
the
modules
that
you
ported
were
preliminary
ports
as
in
they
were
for
your
purposes
only
and
that
they
weren't
kind
of
ready.
Are
there
any
that
you
sort
of
feel
that
could
be
made.
Contrib
candidates.
A
Yes,
there
are,
I
know
that
send
grid
and
simple
sample.
Php
are
usable.
I
know
that
role
reference
at
least
our
version
is
usable,
although
someone
may
have
may
have
done
that
again,
since
we
did,
I
have
a
usable
version
of
encrypt,
but
man
is
it
not.
I
mean
it's,
it's
ready
for
somebody
else
to
work
on.
It's
not
really
ready
for
the
community.
A
F
Yeah
yeah
would
it
be
possible
to
have
a
look
at
the
encrypt
module,
because
it
is
something
that
I'm
interested
in
with
something
that
I
want
to
do
obviously
encrypt
data,
if
that
makes
sense,
and
I'd
just
like
to
say
thank
you
for
the
the
presentation,
absolutely
fantastic.
F
It's
really
helped
me
bring
a
small
sort
of
one-man
band
so
to
speak,
seeing
other
people's
workflow
and
how
they
do
things,
and
it's
surprisingly,
similar
to
how
I
work
as
well.
So
it's
absolutely
really
good
to
know
that
other
people
are
working
in
a
similar
vein.
So
thank
you.
Yeah.
A
Yeah
thanks
for
saying
that
I
really
appreciate
it
and
I
I
feel
the
same
I
feel
like
if
somebody
had
told
me
before
yeah
I've
done
this.
It's
doable.
It
took
me
10
weeks
like
if
I
just
had
the
confidence
of
somebody.
Somebody
else
saying
that
I've
done
it.
It's
doable.
That's
that's
really
helpful.
Sometimes
at
the
beginning
and
yeah
absolutely
I
can
share
what
I
have
so
far
with
encrypt
and
again,
if
you
just
want
to
shoot
me
an
email,
so
I
have
your
contact.
I
can
do
it
that
way.
A
Great
cool
I'm
going
to
assume
that's
it
thanks,
so
much
everybody
for
taking
the
time
and
if
you
do
have
any
questions
or
want
to
poke
at
me
for
anything
erica
and
io,
and
thanks
again,
you
all
have
a
great
rest
of
your
afternoon.