►
Description
Adam Pidlisecky
A
Okay,
well
I'm,
going
to
actually
pick
up,
probably
on
a
lot
of
the
themes
that
Doug
has
addressed
from
a
slightly
different
different
point
of
view.
So
for
those
of
you
who
give
you
two
weren't
here
yesterday
so
I'm
out
in
purple.
Second,
so
I'm,
the
chief
research
officer
for
Iran's
geo,
which
is
a
geologic
visualization,
modeling
company
and
I'm,
also
an
associate
professor,
the
UFC
John
and
I'm
I
kind
of
been
working
in
the
inverse
field
like
us
for
about
the
last
15
years
and
I've.
A
A
Some
of
them
is
done
a
lot
of
that,
and
so
my
perspective
is
coming
out
from
a
combined
an
academic
industry
perspective
when
we
say
industry,
that's
when
I'm,
when
I'm
saying
that
it's
a
specific
part
of
I've
been
involved
in
a
couple
successful
startups
working
around
the
idea
of
taking
research,
sort
of
cutting-edge
research
and
turning
it
into
software
solutions
for
end-users,
and
so
so
I
kind
of
have
two
different
perspectives
on
it.
So
I
really
want
to
think
about
these
ideas
of
how
do
we
make
things
viable?
A
So
that
means
it
endures,
accessible
people.
So
the
really
solving
problems
for
end-users
and
really
to
me
that's
what
innovation
is
is
when
you,
when
you
do
something
new
and
novel
you
make
it
viable
and
you
make
it
accessible.
Sona
start
this
out
by
just
thinking
about
how
we
actually
in
the
work
that
I've
done
on
the
startup
side,
how
we
approach
software
problems
in
terms
of
a
design
issue,
and
so
this
is.
A
This
is
a
Venn
diagram
that
I
first
saw
it
was
a
from
Tim
Brown
who's,
the
CEO
of
ide
0,
which
is
one
of
the
world's
leading
design
firms,
industrial
design
firms
and,
and
they
were
trying
to
codify
what
are
the
important
things
to
consider
to
create
an
innovative
design
in,
in
particular
industrial
design,
but
it
really
abstracts
to
a
lot
of
a
lot
of
different
applications
and
they
came
with
this
idea
that
you
really
need
to
be
you
know,
innovation
kind
of
lizard
is
sweet
spot
between.
Sometimes
it
does
it.
A
I
rrible,
some
it's
feasible
and
some
hands
viable
and
so
desirable
for
a
lot
of
people
has
been,
and
in
the
software
world
gets
interpreted.
As
what
is
your
user
me?
What's
their
problem?
What's
they're
paying
how
you
know,
what
are
the
things
that
need
to
be
solved?
Feasibility
is
a
technical
issue.
Are
the
tools
there
to
make
this
so
is
possible
to
realize
this
situation
viable
in
business
world?
We're
often
talking
about
the
economic
sir?
Can
we
actually
pull
this
whole
thing
together?
A
Sell
it
somebody
in
a
way
that
we
create
a
solution
that
endures
and
really
the
only
time
you
have
successful
long-term
innovation
is
when
you
hit
all
three
of
these.
You
know
if
you
saw
a
user's
problem
and
there's
some
really
great
technology
that
lets
you
do
it
and
you
don't
have
a
viable
solution.
Then
you
actually
haven't
solved
that
person's
problem.
You
don't
have
innovation
than
nervous,
because
your
company
goes
up
business
I,
think
this
model
really
actually
applies
to
what
we
do
and
we
just
need
to
to
change.
A
Some
interpretation
think
those
things
so
with
that
I'm
going
to
kind
of
walk
through
the
way
I
see
or
my
thinking
has
evolved
around
what
we're
really
looking
to
do
here
in
the
research
side
of
things
I'm
in
a
first
start
around
is
what's
desirable,
because
that's
really
what
you
drive
and
thinking
about
what
it's
desirable.
We
have
to
think
about
who
our
users
that's
what's
important,
so
we're
gonna
put
some
scrabble
chips
up
here
and
talk
about
first
of
all,
we're
in
fact
about
a
consumer.
So
who
is
a
consumer?
A
A
A
So
if
we
take
this
box,
this
is
now
the
solution
to
put
together
and
for
a
consumer
solution.
If
I
consumer
I
do
not
necessarily
mean
this
is
an
app
going
out
somebody.
This
could
be
an
industry
partners,
somebody
at
a
Mayan
company.
He
needs
to
invert
some
Carolina
em
data.
Doug
was
getting
at
this
in
terms
of
the
overhead
that
you
have
to
carry
if
you're
a
researcher
to
sort
of
get
a
piece
of
software
up
to
speed.
This
person
is
doing
a
whole
bunch
of
things
around
visualization
around
database.
A
You
know
how
data
is
going
to
move
back
and
forth
that
ok
for
a
consumer
solution,
no
problem,
these
kind
of
things
you
know
visualization.
The
database
are
things
that,
when
we're
doing
our
academic
research
we
have
to
take
on
because
you
need
it
to
be
able
to
work
with
your
data.
It
has
nothing
to
do
with
what
our
end
goals
are.
A
So
how
does
this
work
in
the
reason
why
I
make
this
distinction
of
say
the
startup
or
the
early
phase
or
the
young
piece
of
industry?
Software
I
want
to
invert
some
airborne
TM
data,
great
yeah,
get
my
algorithm
researcher.
Somebody
who
just
came
out
of
DBC
understand
it's
tem,
stick
them
with
a
developer,
who
has
got
a
really
strong
background
and
understanding
at
least
some
of
the
geophysics
language.
They
work
together
as
a
team.
A
A
What's
really
interesting,
what
you
we
actually
are
developing
the
software
you
want
to
make
it
robust
a
critical
thing
in
our
space
is
developers
created,
doing
things
like
unit
testing,
making
sure
your
code
doesn't
have
bugs
in
it
logical,
bugs
associated
with
algorithm
testing.
Researchers
are
the
ones
who
are
actually
uniquely
qualified
to
create
the
tests
to
make
sure
the
thing
is
doing
the
right
thing,
not
just
that
it's
free
of
sort
of
syntax
errors.
A
A
The
same
developers
here-
and
maybe
I
pair
them
with
a
new
researcher,
comes
from
a
bit
of
a
different
background,
but
very
quickly.
You
can
put
together
a
solution
that
we
go
to
consumer
because
you're
leveraging
everything
that
this
person
is
previously
a
lot
of
visualization,
a
lot
of
that
database
management
and
you're
just
simply
porting
it
over
that's
that's
kind
of
how
I
would
view
early
stage
in
his
street
code
development.
A
A
When
you're,
when
you
got
your
early-stage
here,
you've
got
a
customer.
Maybe
some
early
adopters
early
adopters
happen
to
have
risk
tolerance,
it's
higher
than
you
know
the
entire
market.
So
you
don't
want
to
want
with
this
little
box
breaks.
Maybe
they
don't
get
too
upset,
but
you
start
to
move
out
of
that
knee
start
to
try
to
capture
more
of
the
market.
Well,
you
need
more
developers.
So
now
you
bring
in
somebody
who's.
An
expert
in
code
might
know
nothing
about
the
problem,
totally
fine,
great,
no
problem,
so
we
throw
them
in
this
team.
A
The
person
is
doing.
The
research
is
contributing
less.
This
person,
sort
of
senior
developer
research,
develop
or
they're
still
driving
things,
they're
overlapping,
again
sharing
language
with
this
developer.
This
is
an
upcoming
lots
about
research
and
more
about
providing
a
stable
user
experience.
Making
this
works
on
platforms
handles
the
data
sizes
that
people
need
not
research
anymore.
A
You
move
this
further
along
you
get
into
these
really
mature
products,
and
you
probably
have
something
that
looks
like
this.
Now
I've
got
a
whole
bunch
of
developers.
Bunch
of
them
don't
even
know
this
researcher
ever
existed.
Even
this
developer,
who
who's
got
the
research
background,
it's
starting
to
not
really
play
a
big
role,
and
this
consumer
is
now
having
a
great
experience.
A
They've
got
a
really
stable
platform,
its
robust
and
they're
kind
of
happy
with
that,
and
so
they're
happy
until
about
consumer
come
to
exist,
oh
yeah,
but
you
know
what
I'd
like
it
to
do
this.
Could
you
guys
make
it
do
this,
and
this
is
this
really
mature
product
and
your
consumer
has
just
asked
you
to
change
what
it
does
which,
when
you're
early
stage,
it
was
really
easy
to
do
that.
I,
just
simply
move
that
developer
researcher
over
into
a
new
pod
came
up
with
a
new
solution.
A
Now
one
of
the
now
when
we
get
along
here,
we've
got
this
big
mature
solution,
some
nasty
of
change
things.
This
isn't
a
good
model
to
do
that
either.
So
you
know,
there's
very
much.
I've
seen
a
lot
of
people
talk
about
the
difference
between
industry
and
academic
software,
development
and
I.
Think
what's
what
I'm
trying
to
get
out
here
is.
A
It's
also
really
important
that
there
are
some
good
lessons
to
be
pulled
from
the
way
we
we
do
some
software
development
in
industry,
but
there's
also
some
real
baggage
with
that,
when
you
start
to
get
a
really
mature
product,
it
actually
gets
very
heavy
and
it's
a
very
different
model
than
what
might
help
in
the
academic
world.
However,
I
do
think
there
is
something
to
be
learned
from
that
lightweight
sort
of
startup
model,
and
so
so
so
what
is
the
research
approach?
A
I
see
it
and
I
think
this
is
certainly
I've
injuring
my
career
was,
you
know
doing
a
lot
of
optimum
version.
Work.
I
I
would
have
considered
myself
an
Alberta
researcher,
and
my
consumer
was
other
algorithm
type.
Researchers
is
it
worthy?
They
had
a
shared
language
with
me.
We
were
talking
back
and
forth,
and
that
was
exciting.
You
got
you
got,
publications
of
everybody
seemed
happy,
and
then
one
of
the
things
that
I
started
to
do
is
I
started
to
work
a
lot
with
end
users.
A
A
So
suddenly
the
burden
of
what
you
needed
to
do
here.
I
wasn't
just
talking
to
an
algorithm
person
where
I
could
kind
of
be
lightweight
on
what
the
visualization
look
like
I
needed
to
actually
come
up
with
some
kind
of
consumer
solution
that
the
people
could
really
digest,
and
this
is
totally
not
scalable,
and
this
is
I
think
we're
a
lot
of
us
have
seen.
A
You
start
to
really
top
out
in
what
your
research
can
do,
because
this
now
becomes
this
person
running
around
fixing
nuances,
all
the
time
for
all
the
different
consumers
who
want
to
work
with
their
outputs,
and
it's
just
simply
not
scalable,
because
unless
you
can
duplicate
those
people,
you
really
can't
take
this
anywhere
further.
So
that's
kind
of
where
I
see
a
real
opportunity
in
the
research
side
of
things
is
to
try
to
move
from
a
model
looks
like
this
to
a
model
that
looks
like
this.
A
That
can
scale
so
what
we
have
to
put
in
this
box
so
that
research
stuff
can
scale
and
get
it
out
there
disseminate
it
and
really
let
it
have
the
impact
it
should
so
I
think
the
way
we
do
this
and
you're
I.
Think
of
it
as
I.
Go
back
to
that
that
Venn
diagram
I,
think,
okay,
what's
desirable,
desirable,
is
from
the
point
of
view
of
the
user,
it's
from
the
point
of
view
of
the
person
who's
trying
to
get
something
out
of
this.
So,
let's
think
about
who
those
people
are.
A
So
we've
got
this
researcher
who
works
on
algorithms,
so
we've
got
our
so
by
yesterday,
whenever
he
is
talking,
this
is
sort
of
was
listening
to
the
different
use
cases.
People
were
talking
about
what
do
they
want
to
get
from
their
own
versions,
so
I've
kind
of
come
up
with
a
few
other
sort
of
persona,
as
if
you
will
so
I
think
we
have
a
research
consumer.
A
So
this
is
somebody
who
is
a
researchers
working
on
a
problem
and
once
outputs
and
enough
control
over
the
process
to
be
able
to
ask
some
questions
of
a
model
one
day
been
and
how
some
confidence.
So
the
example
that
spoke
to
me
is
tammana.
You
got
who's
working
on
this.
Looking
at
her
tree
imaging
for
from
a
high
rolla,
a
job
logic.
Point
of
view
wants
to
use
the
kind
of
outputs
geophysical
data
can
can
give
her.
A
We
have
this
other
one
where
we
heard
a
few
people
along
this
idea
of
a
research
users.
So
this
is
maybe
somebody
who's
in
it.
Applied
geophysics
person
just
wants
to
sort
of
Darcy
what
you
were
speaking
about
the
wants
to
work
with,
say
some
mt
data,
but
in
a
way
where
they
are
going
all
the
way
from
field
acquisition
to
interpretation
of
those
data
and
they've
got
that
whole
workflow.
So
this
person
wants
access
to
this
this
inverse
code
and
they
want
to
talk.
A
They
want
to
look
at
all
parts,
but
maybe
they
don't
need
to
dig
too
deep
into
anyone.
One
part
they
go:
they
have
to
be
able
to
see
y'all
all
parts
of
the
car,
but
don't
need
a
little
cylinders
and
then
finally,
there's
somebody
called
the
research
developer.
This
is
somebody
who's
working
on
to
your
physics
for
on
a
geophysics
research
problem
and
develops
because
they
need
to.
A
This
was
everybody
yesterday.
This
was
everybody.
We
literally
heard
a
crossword.
What
are
your
challengers
want
have
to
build
stuff
because
I
have
to
like
we
need
these
pieces,
so
this
is.
This
is
a
persona
that
exists.
This
is
a
persona
we
need
to
get
rid
of,
or
we
need
to
at
least
minimize,
because
the
interesting
thing
here
and
I
done
this
for
a
long
time
and
when
I
actually
left
academia
to
do
my
first
startup.
This
was
where
was
I
was
like.
A
This
team
at
a
start-up
I
was
talking
them
about
parallel
computing
and
we
had
some
folks
from
from
the
mich
MIT
computer
science
department
who
would
come
on
board
the
team
just
from
a
comp
sci
point
of
you
and
I
realized
how
my
idea
of
what
parallel
computing
was,
was
so
an
amateur
compared
to
where
they
were
and
that
actually,
by
putting
this
hat
on,
I
was
really
slowing
things
down.
So
so
I
think
we
need
to
to
get
rid
of
those
all
right.
So
here's
where
our
scrabble
board
starts
to
really
get
a
strike.
A
This
is
nice.
This
is
kind
of
how
I
see
this
spectrum
I
see.
I
know.
If
we,
if
we
think
about
this
idea
of
software,
you
know
I've
got
some
consumer
up
here.
That
I
think
we
would
all
love
our
work
to
get
out
there
and
really
help
end
users
as
much
as
possible
might
not
be.
You
know
this.
This
can
be
much
longer
time
lines
and
also
we
have
to
remember
that
a
ton
of
our
algorithms,
you
know
we
work
through
them.
We
go
oh
yeah.
Actually
that
wasn't
a
good
idea.
A
Some
of
that
doesn't
make
great
way
at
this
end.
I've
got
this
developer,
who
doesn't
need
to
know
anything
about
the
problem,
they're
solving
and
then
there's
the
spectrum
in
between
and
there's
some
overlap
here,
and
so
we
bring
the
I'm
going
to
fado.
Let's
not
talk
about
the
consumer,
because
I
think
it
complicates
things
right
now,
because
that's
not
our
mission.
Our
mission,
I
think,
in
my
opinion,
is
to
identify
what
are
the
desirable
traits
of
this
group
in
here.
A
Actually,
they
the
research
group,
because
that's
that's
what
we're
coming
at
this
as
as
researchers
and
I,
think
one
of
the
things
that
would
be
really
powerful
we
can
get
out
of
out
of
this
this
week
is
really
an
intimate
understanding
of
what
the
different
user
needs
are
because
first
spectrum
like
this
to
actually
succeed
in
terms
of
providing
long-term
solutions.
You
don't
only
need
to
know
what
these
young
people's
desires
are.
Whether
these
cases
are,
we
need
to
start
identifying
this
overlap,
because
the
overlap
is
what's
critical
for
bridging
this
idea.
A
You
talk
brought
about
yesterday.
This
idea
the
valley
of
death,
but
we
talked
about
where
you
have
this
academic
idea
and
then
you
have
possible
industry
application
and
it
never
jumps
over
well.
The
way
you've
got
to
jump
it.
/
is
got
a
bridge
and
a
lot
of
times
it's
the
language
barrier.
So
it's
a
matter
of
having
somebody
go
from
the
research
language,
so
pure
algorithm
researcher
can
speak
today.
A
The
research
user,
the
research
user
can
speak
to
the
research
consumer
and
these
people
can
provide
bridges
and
that's
part
of
what
makes
it
a
college
expand.
So
I
think
we
really
want
to
understand
how
people
want
to
use
this
and
what
language
they
use
around
their
use
cases
and
so
I
think
we
focus
out.
We
can
take
this
box
off
here's
what
I
think
I
it's
so
exciting
these
days
and
I
think
you
know
you
know
Doug
was
pointing
to
this-
is
that
these
are
probably
not
a
new
ideas.
A
I
think
what
we
want
to
do
with
the
idea
of
getting
a
framework
getting
our
our
research
out
there
in
a
more
usable
consumable
format.
I
think
what's
happened
now-
is
that
the
feasibility
of
the
technology-
I'm
not
talking
about
the
technology,
word
research
talking
about
the
technology
that
we're
basing
our
research
on
so
the
software,
the
underlying
hardware.
A
We
are
at
a
unique
time
where
the
feasibility
for
us
to
treat
this
in
the
research
world
much
more
like
we
would
and
say
the
startup
world
is.
Is
there
we
have
the
ability
to
do
things
like
proper
software
practices,
coding
practices,
version
control
with
very
low
overhead,
not
because
we
did
anything
but
because,
in
other
silos
of
computer
science,
you
know
we
have
github
now,
which
is
kind
of
a
no-brainer
didn't
exist
a
few
years
ago,
in
a
way
that
somebody
up
here
could
easily
access
it.
A
That's
really
powerful,
so
there's
a
bunch
of
things
that
actually,
if
we
think
about
it,
have
somewhat
reduced
our
dependency
on
here,
the
number
of
visualization
packages
that
are
open
source
that
we
can
pull
in
and
so
the
opportunity
is
there
I
think
we
just
need
to
avail
ourselves
of
it
and
what's
really
important
to
remember
here
and
again,
this
is
what
the
open
shows.
You
know,
this
move
to
open
source
is
really
helping.
People
do,
is
you
know
the
adage
when
I
they
might
do
a
physics
undergrad
is
you
know?
A
One
person's
noise
is,
you
know,
is
another
person
signal
right?
We
all
think
about
that.
One
person's
consumer
is
another
person's
researcher,
and
so
we
have
to
start
thinking
about
that
in
this
open
source
world.
That
a
lot
of
times
in
the
geophysical
world,
we
had
people
developing
stuff
here
that
we
needed
so
I
needed.
You
know,
Doug.
We
spoke
about
monks,
I
needed
some
kind
of
better
solver.
Well,
that's
that's!
A
We
can
easily
get
the
desirable
pieces,
we
can
get
the
feasible
pieces
where
we
have
the
feasible
pieces
they
presented
themselves.
What
does
it
take
to
make
this
vile,
and
why
would
you
want
it
to
be
viable,
so
Bible's
really
simple
in
some
ways
and
they
on
the
software
side
of
things
or
started
from
that
stage,
startup
side
of
things
you
have
a
value
proposition,
somebody's
going
to
pay
for
it
and
what
they're
willing
to
pay
exceeds
what
it's
gonna
cost
you
to
build:
hey,
that's
viable
great,
perfect.
A
So
what
does
it
mean
to
be
viable
and
in
our
world
and
I?
Think
obligations?
Okay,
that's
our
currency
for
a
lot
of
us
in
the
academic
world.
So
that's
a
driver
and
you
know
if
we
can
actually
make
this
sort
of
thread
much
more
efficient.
Okay,
you
got
more
publications,
we
get
it
up
faster,
that's
sort
of
a
driver,
I
think
what's
really
more
important,
though,
is
how
do
you
actually
make
out?
A
How
do
you
enable
a
lot
and
it
and
it's
about
it's
about
building
a
community,
and
so
that's
the
other
piece
that
I
think
the
theme
that
we
heard
people
coming
out
in
yesterday's
you
know
talking
about
a
desire
to
be
part
of
a
community
here
and
I
think
it's
actually
probably
an
understanding
of
them
that,
but
that's
actually
what
it
takes
to
create
a
viable
scenario.
Is
you
need
a
bunch
of
people
working
together
and
contributing
to
each
other?
A
So
when
one
does
speaking
about
that
transition
from
walking
in
you
know
that
the
the
store
tomorrow,
we'll
talk
about
the
competitive
advantage,
was
that
you
had
this
technological
advantage
when
you
become
part
of
a
community
and
it
suddenly
accelerates
this
whole
line
that
becomes
your
competitive
advantage.
You
actually
give
you
things
much
faster
member
results,
because
you're,
essentially
working
in
parallel
and
I,
mean
truly
parallel.
A
There's
an
interesting
thing
here
in
terms
of
that
that's
going
on
down
here
in
the
developer
world
in
terms
of
a
if
you
want
to
abstract
a
community
of
this
idea
of
you
getting
your
developers
working
in
pairs
or
even
people
used
to
think
of
this
as
I'm
going
to
put
my
new
developer
with
an
older
developer
to
train
and
I'll
get
you
know,
I
won't
lose
too
much
time
from
the
first
developer.
You
know,
and
I'll
get
you
know
bring
this
person
up.
Speed
what's
come
out
of.
A
This
is
actually
to
stick
to
developers
together,
you
get
more
than
2x
out
of
them
because
they
actually
start
to
do
these
feedback
loops
that
are
much
faster
than
if
they
work
on
their
own.
They
stopped
making
mistakes.
So
when
you
create
a
community
around
that,
were
you
same
thing
can
happen
here
you
actually
can
get
applicative
gains
that
are
are
larger
than
associated
with
the
number
of
people
that
you
you
put
in
okay.
So
I
think
we've
got
all
the
pieces
here
to
create
a
solution.
So
what
is
what
is
the
solution?
A
That's
going
to
actually
sit
here
at
leverage.
What's
in
this
box,
I
think
what's
critical
going
forward
for
the
next
couple
days
is
for
people
my
call
or
be
four
triple
to
reflect
on
cydia
we've
actually
been
talking
about
software
whole
time
and
I.
Don't
think
we
should
talk
about
software
because
I
don't
see
this
as
being
a
software
solution.
I
see
software
is
being
involved
in
it,
but
it's
a
framework
solution.
A
If
we
talk
of
software,
we
are
then
talking
I'm
going
to
build
something
in
salsa
tem
problem
I
may
build
it
in
a
different
way,
but
I'm
still
focused
on
the
tem
problem.
What
we
want
to
be
doing
is
we
want
to
be
looking
these
overlaps.
What
are
what
are
the
common
pieces
of
languages
your
language
here,
and
how
can
we
break
this
problem
up
in
a
way
that
it
can
be
repurposed
so
that
we
can
attack
a
whole
bunch
of
problems?
A
Things
like
Python
Julia?
Those
are
ephemeral,
things
that
are,
we
can
use
to
implement
a
framework,
but
the
framework
should
be
bigger
than
not,
and
the
framework
is
something
that
I
think
we
really
need
to
pull
people's
understanding
to
get
right,
and
so
this
goes
back
to
in
I'll
talk
a
little
bit
more
about
this.
This
idea
of
one
we
put
up
there
is
a
straw.
Man
is
this
idea
of
what's
called
sim
peg,
which
is
a
simulation
parameter
estimation
in
geophysics
and
it's
based
in
Python.
A
It's
all
object,
oriented
which,
to
some
research
consumers
ago,
well,
I,
don't
need
to
know,
object,
oriented,
don't
you
need
to
think
about
what
are
the
compartments?
What
are
the
ways
you
break
up
the
problem?
How
do
you
naturally
approach
it?
Let's
abstract,
let's
put
you
know,
object
or
underneath
and
really
think
about
what
are
the
common
pieces.
All
of
us
need
to
play
with
what
are
the
Lego
blocks
die
satisfy
the
research
consumer
and
they
may
just
put
them
together
in
bigger
blocks
themselves,
but
that's
fine.
A
They
can
assemble
them
and
just
leave
that
larger
set
of
locks.
What's
sort
of
the
minimum
point
of
breaking
down
a
problem
that
will
allow
everybody
to
access
it
at
the
level
they
need.
So
you
know
what
we've
got
up
here
is
the
idea.
This
is
sort
of
the
framework
we
think
of
it.
We
think
of
meshes
as
being
their
own
thing
that
are
associated
with
a
forward
model
and
the
inverse
model.
Okay.
So
let's
break
that
out,
somebody
can
talk
about
a
mesh
that
way
they
can
for
the
person's
then
get
a
joint
version.
A
They
can
think
okay,
how
does
mesh
from
one
space
go
to
another
space?
Okay,
so
that's
broke.
That
out
is
its
own
thing
happens
to
be
a
class.
Who
cares
if
it's
a
class?
It's
an
idea,
it's
a
mesh
in
a
forward
simulation.
We
break
it
up
into
thinking
about
okay.
Well,
there's
a
survey
part
of
it.
You
know
the
actual
things
you
did
and
there's
maybe
a
physics
part
of
it.
Okay,
let's
break
that
up,
and
then
that
way.
A
If
somebody
comes
in
as
I
just
want
to
talk
about
the
physics
part
of
it
perfect,
they
can
dig
into
this
framework
and
just
look
at
one
component.
They
don't
have
to
tear
the
whole
package
apart
and
so
on.
So
what
we're
in
the
seven
people,
people
can
and
really
can
really
look
at
look
at
this,
so
I
think
one
of
the
key
things
we
want
to
get
out
of
here
is
is
you
know
this
is
again
an
idea,
we've
thrown
up
and
said
that
again
happens
to
be
a
Python
implementation.
Not
important.
A
What's
important,
I
think,
is
how
we
get
this
framework
to
the
right
level,
but
we
can
all
from
it
sort
of
what
we
need
and
really
start
to
leverage
the
ability
to
pull
in
laterally
from
other
fields
very
quickly.
So
if
I'm
doing
optimization
and
somebody
comes
up
with
it-
you
know
you
know
we
much
it
much
better,
solving
algorithm
in
applied
math
from
somewhere
and
they
published
it
as
an
open
source
code.
How
do
we
make
sure
this
thing
plugs
in?
So
it's
really
about?
A
What's
the
way
to
break
these
pieces
down
and
how
do
these
things
connect
to
each
other,
and
so
that
is
the
piece
that's
been
the
real
thing
to
crack.
I
think
if
we
can
crack
that
I
come
up
with
something
by
support
over
the
community
so
that
it's
viable
we're
actually
going
to
have
a
framework
here.
So
we'll
start
to
call
this
whole
thing
innovative
framework,
that's
really
going
to
service
this
core
group.
Here
we
have
a
very
efficient
way.
It's
gonna
stop
people
from
have
and
always
do
visualizations
it's
going
to.
A
Let
all
of
us
do
the
work
we
want
to
do
and
minimize
what
we
have
to
create
just
to
make
that.
So
that's
my
thoughts,
we're
going
to
do
an
exercise
this
afternoon
to
try
to
get
people
to
take
you
take
a
couple
focus.
Actually,
some
forward
simulation
to
get
everybody's
thoughts
on.
You
know:
hey!
What's
the
right
way
to
break
this
up
and
do
we
think
this
is
the
right
way
yeah?
Actually
we
thought
we're,
based
from
our
point
of
view
that
this
is
the
right
way.