►
From YouTube: SimPEG meeting June 5
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Also,
if
the
other
thing
I
was
thinking
of
this
morning,
is
even
keeping
track
of
like
sort
of
some
standard
emails
that
we
all
can
send
out
like.
If
you
get
a
question
from
somebody,
we
probably
want
to
get
them
on
the
mailing
list.
Asap
cuz,
that
way
everything
is
tracked.
Otherwise,
you're
doing
all
of
this
work
to
get
somebody
onboarding
and
then
that's
for
a
single
person
and
that's
lost.
A
B
C
E
C
Because
I'm
first
and
the
focus
is
really
on
simple
I
thought:
I
would
spend
more
time
on
what
we
are.
What
we're
trying
to
do
we're
trying
to
achieve
and
then
at
the
end,
is
the
revive
the
kibbutz.
They
give
us
a
case
study
that
we
presented
be
CGS
and
then
just
finished
with
how
to
connect
with
us.
You
know
with
this
like
channels,
not
all
there,
so
the
I
I
think
the
audience
is
going
to
be
very
like
non
computer.
C
Oriented
is
gonna,
be
very
it
either
academic
or
applied
so
I
thought
I
would
define
some
of
the
concepts
of
what
is
what
does
it
mean
to
be
open-source?
You
know
in
terms
of
that
downloadable
and
shareable
everybody
is
contributing
to
it,
so
yeah
sort
of
give
an
overview
of
who
we
are
and
we
asked
Joey.
So
we
need
like
a
few
more
contributors
here,
Joe
and
them
yeah.
A
C
And
then,
after
that,
so
what
would
it
mean?
What
is
what
is
same
thing
in
terms
of
the
framework,
so
building
blocks
right,
so
you
don't
need
to
be
an
expert
and
everything.
That's
what
I
all
try
to
emphasize
here
that
you
could
just
be
either
an
end-user
to
wrap
like
a
script
for
for
them,
and
then
they
can.
C
They
can
run
it
or,
if
they're,
an
expert
in
one
specific
field
and
again
they
can
tackle
one
problem
at
a
time
and
then
what
is
it
and
the
first
one
is
simulation
so
for
modeling,
the
ODC
fields
to
be
able
to
write
the
physics
and
math.
You
know
like
write
code
like
yeah
like
like
math
and
then
be
able
to
get
to
use
all
the
the
Python
stack
to
be
able
to
see
things
and
then
the
inversion
them
go
from
data
to
that's
useful.
Thank
you.
C
All
the
different
physics
and
that's
kind
of
a
unique
package
for
this
is
that
everything
is
it's
easily
exchangeable
between
yeah
then
now
dive
into
the
kibbutz
show
the
data
just
with
the
ps4
this
one
and
then
I'll
run
some
inversions
for
some
inversion.
Pictures
talk
about
that
and
then
what's
next
here,
I'm
kind
of
open
I
would
like
to
just
put
some
of
you
what
you
guys
are
working
on.
So
if
you
have
anything,
you
would
like
me
to
go
over.
I'll.
C
Add
like
one
one
picture
for
my
stuff,
but
that's
gonna
have
a
placeholder
right
now
and
then
I
can
introduce
a
lynches
with
the
next
dog
for
Tebow.
It's
going
to
be
the
logical
constraints
and
then
finish
up
with
a
slide
where
all
the
places
that
they
can
reach
us
or
install.
That
would
be
to
talk
how.
C
G
A
A
couple
like
general:
well,
first,
we
need
to
fix
this
lack
the
it
might
be
nice
to
show
the
DC
part
of
kibitz
as
well,
even
if
you
don't
necessarily
go
through
it
like
I,
think
that
was
one
thing
that
people
don't
really
expect
voted.
Pcgs
was
just
being
able
to
see
the
currents
and
use.
This
is
like
for
modeling
survey,
design,
yeah.
D
D
A
A
So
you
like
one
thing
you
can
even
start
with,
and
this
might
be
kind
of
cool
it's
to
sort
of
a
book
candidate
with
skip
it
says.
So,
even
though
you're
not
talking
about
it
all
that
much
it's
like
it's
a
motivation.
It's
a
really
good
like
a
motivating
problem,
because
it
is
like
air
physics.
We
can
tackle
most
the
different
physics
that
we
look
on,
that
there's
some
subtleties
with
the
remnant
magnetization
and
all
these
kinds
of
things.
So
there's
a
lot
of
interesting
question
and.
F
A
F
C
A
E
A
A
C
C
C
C
F
A
F
F
C
A
Even
thinking
like
a
single
problem
like
the
remnant
like
just
meeting
rapid-fire
through
a
lot
of
the
problem,
I
wouldn't
be
surprised
if
some
of
the
audience's
familiar
with
convince'
yeah.
We
don't
really
know
man
but
I
mean
even
if
you
look
generally
speak
to
it,
just
a
whole
slack
of
interesting
research
problems
and
then
save
anyone.
You
know
we.
This
is
some
of
the
things
that
we're
trying
to
be
able
to
investigate
question
fake
here's,
one
problem
with
it:
I
like
being
connected
with
and
talk
to
them
and
bank
a
little
bit.
B
B
B
We
can
have
some
idea
of
the
extent
of
the
bodies,
but
we
definitely
don't
get
this
distinction
between
the
TK
TK
unit
in
the
MV
k,
and
so
the
framework
we
propose
here
is
that
framework,
so
that
in
VII
in
red
here
it's
like
the
usual
Tippin
of
Stein's
version
and
what
we're
gonna
add
here
is
actual
physical
and
the
geological
data.
So
you
see
that
it's
after
each
iteration
of
the
inversion,
we
gonna
add
this
information
back
into
the
the
conversion,
and
for
that
we
use
a
Gaussian
mixture,
presentation
of
our
physical
properties.
B
B
So
you
see
that
was
a
bit
of
physical
constraint
at
each
situation,
with
changing
the
reference
model
to
guide
it
towards
a
more
clustered
solution
that
you
can
use
our
virtual
physical
data.
So
back
to
the
AC,
we
can
add,
stop
resisting
for
the
information
we
have
about
the
petrol
physik
into
the
inversion.
So
now,
I
have
to
steal
the
gravity
and
the
magnetic
data,
but
in
the
IOM
I
also
feed
it.
B
A
true
physical
distribution
for
density
and
for
my
Mac
magnetic
susceptibility
and
that's
a
step
we
we
get
after
data,
so
we
see
we
have
like
a
merchant
way
to
define
body,
but
we
defeat,
and
we
still
definitely
don't
recover.
Well,
like
this
very
high
susceptible
unit.
Here
in
the
in
the
mag
and
the
gravity
we
see,
we
have
some
sort
of
lacquer
hello
things
going
on,
and
the
vol
volume
I'll,
also
don't
don't
match
between
the
twin
version.
So
what's
nice
about
also
about
the
petrol
physic?
Is
that
it's
tightly?
D
B
So,
instead
of
having
two
distribution
separate
for
density
and
Max
accessibility,
we
we
we
ties
them
come
in
a
tooth
in
a
2d
distribution
like
this
one
and
when
to
don't
lean,
goes
to
data
set
and
that's
when
we
get
finally
Modra
Matic
gain
in
the
internal
side.
So
that's
the
model
we
recover
for
the
mag
and
gravity
and
that's
a
can
of
Georgia
coal
mode
or
we
can
infer
from
this
inversion.
So
we,
finally
here
we
finally
able
to
recover
the.
B
The
unit
and
we
we
also
have
like
some
features
such
as
like
this
protuberance
and
in
the
east
here,
and
what
I
want
Steven
to
be
when
DC
things,
but
starting
from
starting
from
this
model,
I
want
like
a
like.
Maybe
a
single
Bohol
here
and
I'm,
able
to
constrain
on
the
little
G's
saying
that
certain
cells
should
belong
to
a
certain
cluster,
so
not
assigning
notice,
not
assigning
like
petrol,
physical
constraint
or
like
bounds,
but
saying
that
oh
is
it
belong
to
that
type
of
fortunate
audience.
B
So
that's
what
I
recall
file
like
we
can
we
gonna
constrain
on
the
little
og
themselves,
you
and
so
a
bit
like
done.
Don't
don't
slide
like
what's
next,
so
one
of
our
goal
here
is
also
to
really
the
community,
not
something
that
we're
gonna
have
inside
a
or
that's.
Only
a
certain
lab
will
have
access.
We
it's
a
gonna,
be
an
open
source.
One
and
we're
gonna
have
use
it
with
example,
and
also
for
we
possibility
to
to
release
a
case
today.
We're
gonna
do
with
this
and
then
also
what's
nice.
B
Is
that
I
think
we've
done
a
pretty
good
job
at
constraining
the
petrol
physique?
And
now
it's
more
about
all
the
geology
side
of
constraint
that
we
can
really
improve
the
classification.
So
it's
just
like
a
few
example
that
I
have
here
where,
like
and
I,
will
see
how
I
organize
that,
but
basically
this
example
like
its
release.
It's
a
it's
a
silly
example
saying
that
I
don't
want
to
resist.
B
You
need
to
touch
the
background,
so
it's
actually
enclose
it
into
conductive
unit,
but
I
can
use
that
also
for
like
here
at
time
domain
p.m.
example.
We
have
a
resistive
layer
between
two
conductive
one,
but
I
can
impose
the
type
of
things.
Oh,
he
should
probably
put
it
bigger,
but
that's
this
one
is
like
it's
like
mold
I
drew
Georgie
example
where
I
actually
constrain
the
classifications.
B
That
I
want
the
dry
stuff
to
be
on
top
the
freshwater
to
be
in
the
middle,
in
the
salt
water
to
be
added
at
the
bottom,
and
I
can
actually
put
that
type
of
constraint
and
that
el
solo
also
a
lot
like
tile.
That's
also
one
way
also,
you
can
get
some
juice
that
istic
to
because
if
you
have,
if
you
have
few
Vida
water
like
you're
auditioning
around
to
get
an
idea
of
the
of
the
depth
such
an
information,
you
can
add
constraint
that
up
roster.
B
B
G
D
G
B
B
G
A
B
B
Like
the
Nikon,
the
next
thing
that
is
like
side
them
is
like
the
open-source
community
side
that
we
want.
This
could
go
out
and
be
usable
by
other
people.
It's
not
something
we're
gonna
keep
in
our
like
something
that
we
want
to
yeah
broadcast
advertise,
and
then
there
is
like
a
research
spot
where
we
we
want
to
improve
how
you
can
like
very
complex
type
of
constraint
for
the
majority
group
to
put
more
after
you,
information,
I.
F
Think
that
words
like
a
really
emphasize
that,
like
ever
before,
simply
talking
about
communities
and
how
it
helped,
because
I
that's
of
a
generic
challenging
memory,
really
integrate
information.
How
are
you
have
you
put
the
NOAA
geological
information
into
your
version,
because
a
pretty
big
topic
and
I
think
that's
what
John's
I
think
there
was
really
comment
at
first
and
then
you
can
comment
the
house
if
it
helped
you,
so
you
do
that
I
think
that's
a
bit
more
powerful,
rather
than
make
starting
from
sit
back.
F
B
D
A
B
A
A
A
B
F
Think
at
the
end,
I
would
make
a
big
deal
about
when
you
like
this.
Do
things
right?
It's
problem
solving.
It's
probably
it's
very
challenging.
Maybe
it's
probably
a
more
than
one
PhD
one
right
like
actually
solving
a
graviton,
a
math
problem
en
problem,
it's
yeah,
but
that
part
is
like
oh
really
done
we're
moving
on.
So
it's
kind
of
gone
so
like
it,
you
can
you.
So
that's
that's
where
you
can
build
upon
and
cuz
like
during
conversion
itself.
It's
also
challenging,
but
you
really
need
all
of
those
pieces
to
do
that.
F
F
B
F
C
D
B
B
C
F
E
I've
got
one
comment:
I
just
wanting
sort
of
at
at
the
beginning,
where
you're
kind
of
introducing
the
challenge
you
know
really
working
with
different
rocks
and
rock
properties.
If
you're
going
to
introduce
tkc
at
the
beginning
where
you
have,
but
if
maybe
there
was,
you
know
a
table,
you
know
basically
saying
you
know
Rock
one
two,
three
four
and
then
you
know
maybe
physical
properties.
E
You
know
density
susceptibility,
first
of
all,
just
with
a
number
or
maybe
a
range
of
numbers
which
is
kind
of
people
would
generally
see
and
then
you'll
kind
of
have
another
image.
That
kind
of
goes
from
say.
Well,
it's
not
really
just
a
number,
a
range
of
numbers,
it's
really
a
distribution.
And
then,
if
you
looked
at
density
of
these
different
rocks,
you'd
have
you
know
some?
E
You
know
function
with
a
whole
bunch
of
peaks
and
to
use
that
as
a
you
know,
also
a
little
bit
of
a
motivator
to
say
you
know
what
I'm
up
against
I
mean.
How
do
I
incorporate
all
of
this
sort
of
probability,
probabilistic
information
about
properties
and
the
different
rocks?
How
do
I
put
that
all
into
you
know
one
package
yeah.
E
B
E
Sort
of
open
that
up
a
little
bit
more
right
at
the
beginning
and
then
just
toss
that
out
as
a
challenge
like
okay,
how
do
I
actually
do
I
do
this?
And
here's
where
you
know
like
okay
I
can
invert
these
datasets
individually,
but
really
other
information
I
want
to
incorporate
that
and
find
something.
A
solution.
That's
compatible
with
that
yeah
a.
E
Yeah,
so
then,
once
there
once
there
there,
then
you
just
have
to
decide
what
gets
to
what
has
to
get
omitted
and
what
basically
just
flashes
up.
As
you
know,
a
complicated
diagram
in
which
you
only
have
you
know
10
or
15
seconds,
to
point
out
one
thing
and
just
illustrate
that:
okay,
it's
a
it's
a
process.
It's
non-trivial,
I'm
not
going
to
work
through
this.
H
Did
you
want
to
chat
about
anything
with
your
poster,
comport
I?
Don't
really
the
draft
had
done?
Okay,
I,
don't
think
I'm
going
to
focus
too
much
on
a
cynic
framework
or
anything
but
I'm
just
going
to
go
through
that
survey,
design
aspects
for
the
underground
environment,
I
chose
some
of
the
challenges
and
how
we're
starting
to
try
to
address
them
and
then
just
kind
of
discuss
how
simp
expen
a
great
tool
to
use
and
do
that
one.
A
A
A
A
A
A
List
they
come
through,
it's
like
all,
but
there
was
a
message
that
was
like
a
very
basic
you're
gonna
answer,
a
little
question
that
was
their
specific
variant,
and
so
yes,
so
if
like
a
few
mostess
were
to
just
make
a
habit
of
checking
that
regularly,
and
even
if
you
don't
know
the
answer,
you
don't
have
the
time
like.
If
you
can,
you
cannot
slack
because.
A
A
A
G
A
Lately,
we've
decided
that
you
can
branch
director
classroom
and
then
we
do
request
some
positive,
so
everything
it's
a
very
short
release
cycle.
But
then
what
is
up
to
you?
So
when
you
generate
a
claw
request
ambassador,
you
should
document
every
review
process
and
that,
like
here's,
a
few
things
that
we
look
forward
generally,
yes,
they
don't
need
sort
of
for
contributors
merch,
because
we
need
to
that
also
like
off
the
version
and
send
out
at
least
and
so
so.
The
best
documents
and
I
know
that
we
are
always.