►
From YouTube: SimPEG meeting November 4, 2020
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
So
we
don't
have
that
listed
there.
There
are
the
notes,
and
so
what
we
can
do
is
we'll
we'll
go
so
we'll
structure
we'll
give
this
a
try
and
then
an
agenda
item
perhaps
can
be
at
the
end
of
the
meeting.
Did
this
meeting
structure
work
and
what
would
we
do
differently?
But
what
we'll
try
and
do
is
just
go
through
go
through
everyone
state
what
you've
been
up
to?
A
If
there's
anything
you
want
to
sort
of
update
us
on
and
then,
if
you
have
longer
discussion
items
just
state
that
I
think
this
is
a
longer
discussion
item
just
want
to
raise
it
for
folks
and
then
we
can
circle
back
once
we've
gone
through
everyone
on
sort
of
what
the
priorities
are
to
discuss,
discuss
for
the
week
yeah.
A
But
yes,
please
feel
free
to
add
your
own
bullet
points
to
the
to
the
meeting.
Notes
there
perfect
dumb.
Would
you
like
to
start
us
off.
C
Yeah,
so
I
continued
work
on
the
das
client.
I
had
a
breakthrough
finally
on
friday
and
managed
to
be
able
to
get
like
really
good
parallelization,
all
the
way
through
the
the
cg
iterations,
with
almost
no
downtime,
so
that
was
was
a
huge
success
because
I've
been
struggling
for
a
while
and
now
I
kind
of
understand
what
what
needs
to
be
done
elsewhere
and
then
I
started
running
some
tests.
You
know
on
like,
in
terms
of
like
number
of
processors,
number
of
tiles,
all
that
stuff.
So
it's
coming
along.
C
I
probably
just
share
something
next
week,
not
the!
I
don't
think
it's
mature
enough.
I'm
still
running
examples
so
we'll
see,
but
I
think
it's
promising.
A
D
Okay,
yeah,
I
reorganized
my
branch
reorganize,
a
code
about
the
drone
teamwork
across
greedy
and
drone
team
on
my
branch,
and
everything
is
last
night,
so
I
scheduled
meeting
with
the
doctor
soon
this
afternoon
and
we
take
we.
We
finally
go
through
all
the
code.
If
everything
is
fine
and
we
are
going
to
submit
the
pr
tonight,
I
think-
and
also
when
I
take
a
look
at
my
code.
E
Yes,
I've
been
mainly
continuing
with
the
automatic
reading-
oh
sorry,
probably
better
now
mainly
with
the
automatic
reading.
I
hope
to
get
in
a
new
release.
This
speak,
which
then
will
make
it
possible
that
you
can
throw
your
model
and
the
survey
file
at
emg
3d,
and
it
will
create
frequency,
dependent
models,
compute
the
forward
models
and
the
gradient.
If
you
want
and
give
it
back,
which
is
sort
of
sort
of
fundamental
for
the
command
line
interface,
you
have
so
that
other
other
inversion
routines
can
interface
with
it.
E
So
I
hope
to
get
that
out
soon,
or
at
least
some
working
version
of
it
yeah
something
had
worked.
A
F
I
guess
one
thing
show
of
hands
who
is
actively
working
on
mt
stuff
in
simpek.
F
Likes
actually
like
making
changes
to
bran,
I
just
wanted
to
see
who
is
actively
working
on
it
right
now.
So
this
okay.
G
Not
sure
I
don't
have
too
much
new
things
to
report,
except
currently
I'm
writing
up
our
manuscript
and
also
on
monday.
We
had
a
meeting
about
the
pr
scenes.
I
think
it's
very
helpful
for
me,
so
I
was
also
motivated
to
set
up
my
own
github
account
and
try
to
make
it
a
little
bit
more
organized
yeah.
So
that's
about
all
from
me
so
far,
excellent!
That's
awesome!.
B
So
here
once
you
sorry
sorry
the
interrupt
once
you
commute
to
your
manuscript,
maybe
you
can,
because
we
designed
quite
a
few
I
mean
similar
but
different
scenarios
where
the
density
models,
the
monoliths
assembly
models
they
have.
They
are
looking
at
different
depth
and
different
densities
values
trying
to
understand
the
recoverability
of
the
physical
properties
from
geophysically
version.
So
maybe
we
can
share
the
the
the
models
and
maybe
the
codes,
but
I
mean
we're
using
the
same
package,
so
nothing
new,
but
the
models
and
the
data
probably
will
be
useful
for
others.
B
No,
no!
No!
What
I'm
saying
that
after
you
submit
your
med
school,
okay
yeah,
we
can
make
our
models
available
to
others,
maybe
by
contributing
one
example,
application
example
in
simpac.
C
Yeah,
I
think
that
would
be
huge
for
for
an
email,
because,
because
I'm
working
on
a
specific
project
right
now
and
then
people
that's
their
their
whole
questions
like
the
depth
right.
They
they're
not,
they
don't
trust
the
depth
of
the
model.
So
I
think
if
we
could
have
an
example
page
on
the
site
that
shows
exactly
this.
C
That
would
be
awesome
yeah
if
we
could
finally
have
something
documented
yeah.
That
would
be
a
great
contribution.
A
Absolutely
I
noted
that
actually,
as
potentially
a
bit
of
a
longer
discussion
item
is
how
do
we
want
to
be
sharing
models
and
and
code?
We've
got
some
stuff
about
sharing
code,
but
data
is
actually
a
bit
of
a
bit
more
challenging
right
now,
so
joe,
would
you
like
to
go
next.
H
Yeah,
that's
fine.
Sometimes
I
started
like
double
checked
or
I
checked
my
signed
conventions
against
the
senpai
stuff
and
a
few
more
backwards
from
what
they
should
have
been.
So
I
have
it
all
properly
set
up
now.
Oh,
it
works
well
secondary
works
and
then
I've
been
pulling
a
bunch
of
pr's
into
the
master
branch.
If
you've
noticed
it's
currently
failing
for
another
reason,
but
I
I
have
it
working
on
the
on
the
per
inch
or
the
or
the
release.
Note
branch.
H
H
F
I
mean
I've.
I've
done
this
sort
of
extensive
validation
against
dcip
octree,
but
also
yeah.
I
could
write
something
new.
I.
H
H
Yeah,
just
that's
all
I
mean
just.
I
want
to
make
sure
that
those
lines
are
hit
so
on
the
dci.
The
dcid1
there's
another
conversation
that
I
had
on
there
about
where
you
want.
The
secondary
solution,
like
the
primary
secondary
formulation
to
live,
just
want
to
come
to
a
consensus
on
that
on
the
discussion
on
github.
So
if
you
guys
can
take
a
look
at
that
and
read
through
it,
yeah.
H
Yeah,
like
I
said,
I
pushed
a
few
changes
on
there
just
to
simplify
the
code
a
bit
if
you
notice,
like
I
didn't,
really
change
anything.
I
just
simplified
a
little
bit:
okay,
yeah,
that's
great,
and
then
the
yeah
and
the
mag
stuff
like
add
tests,
if
you,
if
you
don't
mind,
especially
like
just
all
the
I
o
routines,
that
you
that
were
added
okay.
F
F
To
the
maybe
the
group
for
all
this
is
right
now
in
dcip,
there's
a
there's
just
a
flag
that
says:
do
you
want
is
as
a
primary
secondary
formulation
or
do
you
want
it
as
interpolating
the
sources
to
the
mesh,
but
it
sounds
like.
Usually.
We
would
like
to
make
that
two
different
simulations
instead
of
having
a
flag
for
the
choice
of
how
you
want
your
right
hand,
side
to
be
treated
or
do
you
guys.
H
F
H
H
H
We
can
talk
about
that
later.
I
I'd
like
to
keep
most
of
it
on
github,
so
we
have
a
documentation
which
we'd
like
to
talk
about
in
person
too.
That's
fine,
okay,.
A
That
sounds
good
thanks
joe
do
you
have
any
other.
I
A
Fair
enough
tebow,
would
you
like
to
go
next.
J
Yeah
sure
so
I
made
progress
on
my
on
the
pg
ipr.
I
I
think
I've
actually
bring
the
code
to
the
level.
I
want
it
to
be,
simplifying
a
lot
like
the
way
to
use
it
and
try
and
minimize
the
number
of
different
directives
you
need,
etc.
J
Just
trying
to
have
like
things
as
I
would
I
like
as
much
as
you
can
expect
it
to
work
like
when,
if
you
don't
don't
know
it
in
a
good
shape
code
wise
and
we
can-
and
so
I
think
we
can
probably
start
the
review
process
quite
soon.
Maybe
I
can
give
a
tour
to
people
that
want
to
review
on
that
soon.
I
was
thinking
especially
lindsay
and
joe,
but
if
other
people
want
to
be
reviewers
on
that,
it's
going
to
be
like
a
7
000
line,
good
pr.
J
A
A
I
might
have
cut
out,
or
you
cut
out
just
at
the
the
last
part
of
your
sentence,
but
did
you
have
any
any
other
things
or
questions
you
want
to
raise
with
with
that
pr,
or
is
that
good
to
go.
J
Well,
it's
gonna
be
well,
as
I
say,
I've
wanted
to
say
was
like
I
think
it's.
We
can
start
the
review
process.
The
only
thing
missing
is
extensive
documentation
at
the
moment,
okay,
but
I
brought
it
like
to
a
good
level
code
wise
in
terms
of
functionality,
excellent
and
use
it
and
usability
excellent.
A
Cool
devin.
F
Yeah,
let
me
know
if
you
guys
hear
any
background
noise
or
somebody
with
a
weed
whacker
or
something
yeah.
I
guess
let
me
just
find
myself
so
yeah
as
per
dom's
comment
last
week,
I
I
added
that
same
stability
to
magnetic
and
gravity
gradiometry,
so
that's
been
added
and
I've
done
that
comprehensive
testing
of
the
dc
ip
simulation.
So
comparing
what
we're
doing
to
the
dcip
octree
code,
that
ubc
has-
and
I
added
a
primary
secondary
approach
for
discretizing
the
source
term,
which
actually
works
really
really
well.
F
So
I've
I've
summarized
this
up
in
the
the
validations
there's
notebooks.
We
can
recreate
all
the
results.
So
once
my
the
two
pull
requests,
I
have
dcip
and
potential
fields
are
added.
Then
we
can
put
these
validations
like
it's
all
kind
of
tied
in
together,
but
basically
I
found
that
the
simpeg
dcip
does
a
really
good
job
and
I
got
better
results
by
doing
this
primary
secondary
approach.
I've
tested
it
for
layers
and
steer
in
a
whole
space
and
compared
it
against
the
dcip
auxry
manual.
F
So
there's
a
lot
of
really
good
tests
there
and
the
errors
are
very
reasonable
and
it's
doing
which
what
it's
expected
to
do.
So
that's
really
good
news
and
I've
decided
to
start
working
on
the
mt
stuff.
F
So
I
talked
a
little
bit
with
john
the
other
day
and
just
kind
of
looking
to
to
jump
in
and
and
help
out.
So
I
guess
maybe
I
should
go
and
assess
kind
of
what
everyone's
working
on
and
what
the
issues
are
etc,
but
yeah
I'm
willing
to
throw
some
time
into
into
that.
So
that's
what
I've
been
doing.
A
It
sounds
like
there's
potentially
two
discussion
points
out
of
that
that
so
first,
I
guess
maybe
talking
just
a
bit
about
where
we
want
to
post
these
code
comparisons
or
what
we
want
to
do
with
them.
Now
that
you
you've
done
that,
because
I
think
that's
a
really
valuable
thing
to
have,
I
know
I'm
just
stating
these,
so
we
can.
We
can
circle
back
is
where
do
we
want
to
put
those?
Because
I
think
that's
something
we
might
want
to
have
public
and.
F
I
was
going
to
say
that
we
we
had
sort
of
decided
that
part
of
the
simpeg
grouping
of
repositories.
We
were
going
to
have
a
validations
repo
and
I
think
we
created
it.
I
don't
know
how
we're
going
to
let's
say
like
advertise
it
or
kind
of
put
in
that
link
to
to
show
you
where
it
is
and
show
the
results
of
it.
That's
that's
a
really
good
question,
but
at
least
we
have
a
place
to
house
them.
K
How's
it
going
guys
so
this
past
week,
I
guess
I've
just
been
trying
to
start
playing
around
with
some
gravity
data,
getting
just
some
basic
plots
and
stuff
of
it
loaded
up,
and
I
guess
things
loaded
into
python,
so
I
can
start
messing
with
it.
I
guess
one
thing
I
was
messing
with
yesterday
was
trying
to
get
a
good
good
digital
elevation
model
so
that
I
can
make
my
mesh
and
stuff
like
that.
K
Do
we
already
have
some
utilities
in
simpeg
for
loading
in
digital
elevation
models
like
srtn,
and
I
guess
is
srtm
kind
of
a
good
place
that
people
usually
start
with
digital
elevation
models
or
they
get
them
from
somewhere
else.
It
was
kind
of
a
general
question
that
I
had.
C
I
C
And
no,
I
I
don't
think
you'll
have
any
ios
for
you
have
like
ers
files.
What
format
is
it
giving
you.
K
So
I
mean
some
of
it
was
geotiff
and
then
some
of
it
was
like
a
arc
raster
ascii
yeah.
So
I
was
able
to
read
in
like
I
was
able
to
find
something
online
and
just
kind
of
get
that
that
ascii
file
read
in
so
I
think
I
think
I'm
on
the
right
track.
I
think
I
almost
pretty
much
have
it
into
a
into
a
matrix
that
I
can
start
messing
with,
but
I
was
just
wondering
if
we
already
had
something
like
that:
okay.
E
Maybe
yeah
I'm
not
too
familiar
with
the
raster
file,
but
they
had
they
mentioned
it,
and
maybe
it's
a
good
note
to
take,
because
probably
a
few
poor
people
are
gonna
meet
two
three
leagues.
So
if
you
want
to
be
part
of
that,
let
me
know
okay,
because
I
don't
think
there
are
well
there's
this
gem
chase
gem
gis-
I
don't
know
if
you
heard
of
that.
Okay,
which
is
kind
of
a
jam
pie,
spin
off
that
does
more
of
it
with
a
2d
surface
data.
Okay,
so
they
might
look
into
the
same.
K
I
mean
I'm
not
looking
at
a
huge
area,
maybe
like
100
kilometers
square
or
so
is
maybe
what
I
want
to
plot
up,
but
yeah
I'll
keep
looking,
but
the
stuff
I've
come
across
so
far
in
the
last
day
or
so
wasn't
wasn't
impressive.
So
that's
why
I
was
going
about
just
trying
to
load
load
in
one
of
these
dems
and
kind
of
well
use
it
to
one
make
the
model
and
then
to
to
kind
of
just
get
a
nice
background
for
some
of
the
data
plots.
C
Yeah,
I'm
not
just
interesting
to
pandas
for
because
they
they're
probably
the
most,
they
want
to
support
the
best
gdal
and
you
know
shapely
and
all
these
guys
so
yeah
yeah
and
it's
from
it's
on
converse.
If
you're
on
pc
not
to
worry
about
the
above.
The
super
stuff.
K
Yeah
I
was
looking
at
that
yesterday,
but
I
haven't
gotten
far
enough
through
it
to
have
a
working
example
yet,
but
I'll
keep
playing
with
that.
Thanks
for
the
suggestion.
B
Yeah
so,
as
charlotte
mentioned,
I'm
going
to
work
with
shalom
to
go
over
his
code,
joining
version
cross
screen
code,
one
more
time
clean
it
up
before
we
do
our
pull
request.
I
also
had
a
meeting
with
lindsay
joe
and
keyboard
on
a
monthly
right,
a
monday.
They
workouts
through
the
the
pull
request.
B
Travis,
I
mean
the
version
controller
codes,
all
that
which
is
really
helpful.
So
hopefully
we
can
start.
I
mean
do
each
step
after
we
start
after
after
we
submit
our
first
two
requests.
So
looking
forward
to
work
with
with
you
guys
to
to
to
go
throughout
the
process,
all
the
way
up
to
getting
our
codes
approved
yeah.
So
that's
what
I'm
going
to
work
with
shalom
later
this
afternoon
right,
shannon.
B
L
L
Yeah,
I
got
nothing
too
much
new.
I
put
mt
down
for
a
little
bit
there,
because
that
joint
stuff
was
kind
of
giving
me
a
headache
but
yeah.
I
might
take
a
look
and
just
start
writing
out
again
the
stuff
that
you're
saying
there
lindsay
see
if
I
can
tackle
it
with
you
there
and
yeah
playing
with
the
desk
at
the
das
clients
and
distributed
there.
So
I've
done
like
a
split.
L
I
split
up
a
problem
into
simulations
and
it's
just
like
a
fitting
a
cold
coal
curve,
just
some
stuff
that
we
do
for
the
cocoa
analysis
and
I
got
it
working
and
yeah.
I'm
getting
good,
really
good
parallelization
that
way,
but
yeah
right
now,
I'm
just
trying
to
do
that
with
calculating
the
fields
for
a
whole
bunch
of
sources
as
simulations
for
dc
and
yeah
I'm
having
a
little
bit
of
problems.
I
don't
know
baskets
yelling
at
me,
so
I
just
got
to
figure
that
out,
but
yeah
almost
there
getting
getting
somewhere.
C
E
A
Excellent
thanks
john
yeah,
so
for
my
updates.
I've
also
been
working
on
mt
adjoints
made
some
progress
just
first
off
sort
of
understanding.
What's
there,
the
structure
of
the
code
is
fairly
different
than
how
we
structured
the
derivatives
for
sort
of
the
frequency
domain.
So
first,
I
just
started
by
like
writing
out
stuff
to
make
sure
I
kind
of
understand.
What's
going
on,
I've
made
some
progress.
A
The
adjoints
aren't
quite
working,
but
I
I
feel,
like
I've
got
a
better
understanding
at
this
point,
so
I'm
gonna
hope
to
throw
a
bit
more
time
at
it
this
afternoon
and
and
see
where
we're
at.
I
haven't
pushed
anything
because
I'm
mostly
making
a
mess
right
now
and
so,
but
we'll
see
we'll
see
where
again,
but
one
thing
that
that
sort
of
raised
is,
I
wonder
if
it
would
be
worth.
A
I
don't
know
if
folks
have
time,
either
later
this
week
or
early
next
week,
who
are
interested
specifically
in
mt
to
just
have
a
bit
of
a
smaller
meeting
as
to
how
we
want
to
progress
with
that
code
base,
because
I
think
there's
a
few
things
to
think
through
both
with
how
do
we
want
to
potentially
integrate
with
the
frequency
domain
code
and
like
do
we
actually
want
to
just
get
rid
of
the
nsem
simulation.
I
think
that
would
be
nice,
but
I
think
we
it's
worth
a
worth
of
conversation.
L
A
A
That
sounds
good,
so
what
I
will
do,
as
an
action
item
is
just
send
out
a
potential
time,
would
later
this
week
be
of
interest
to
folks,
yeah
sure
yeah.
K
F
I
would
maybe
prefer
like
friday
at
some
point,
I
have
an
appointment
on
the
thursday
afternoon
night
and
I
also
have
a
meeting
with
with
doug
and
roman
on
that
day.
A
Okay,
you
can
try
and
plan
for
that
cool
other
items.
A
I
guess
so
sort
of
thinking
through
the
well
sorry,
one
other
mt
question
for
atu
we're
giving
a
presentation
sort
of
on
updates
for
mt,
and
I
was
wondering
if
anybody
actually
knew
of
a
simple
data
set
that
would
be
like
field
data
set.
That
would
be
nice
to
nice
to
demo
and
demonstrate
don't
want
anything
huge.
But
if
not
that's
fine,
we
can
stick
in
the
synthetic
world,
but.
F
Yeah,
I
would,
I
have
a
comprehensive
workflow
for
the
e3d
mt
code,
the
uvc
code.
This
is
a
data
set
from
clone
curry
in
australia
and
it's
it's
freely
available
and
I
would
be
able
to
like
if
you
click
the
download
from
the
website,
workflow,
that
I
have
you
just
get
it
already
in
ubc
format
so
or
the
edi
format.
A
What
I
love
to
hear
cool
okay,
so
then,
are
there
other
longer
discussion
items
that
folks
would
like
to
dive
into
we've
got
about
half
an
hour
here.
So
a
couple
things
well
I'll
I'll
stop
talking
and
does
somebody
have
a
discussion
item
you'd
like
to
jump
into.
B
A
Okay,
if
no
one
has
discussion
items,
what
do
folks
think
of
meeting
structure?
How
do
we
want
to
run
these?
Does
this
sort
of
work
and
like
next
time
we
think
about
sort
of
agenda
items?
Do
we
want
to
more
so
rotate
through
and
have
somebody
give
a
bit
of
a
presentation?
What
would
be
still
useful?
A
I
mean,
I
think,
having
some
discussion
time
to
just
give
updates
on
what's
going
on
is
is
helpful,
but
what
else
is
a
worthwhile
use
of
your
time
and
would
help
you
in
your
work
and
help
sort
of
progress
things
forward.
F
I
I
think
maybe
one
thing
we
could
do
in
the
meeting
is
just
identify
the
action
items
and
put
them
somewhere,
because
it
really
would
tell
us
who's
going
to
work
on
things
next.
I
think
it
might
keep
people
sort
of
a
bit
more
on
task
going
forward
and
then
you
sort
of
have
a
bit
of
an
expectation
of
something
something
moving
forward.
I
kind
of
like
that.
F
E
H
You're,
like
now
yeah
I
get
hooked,
it's
just
makes
me
looks
a
little
bit
more
open
and
showing
that
okay.
Well,
these
are
the
things
that
we're
intending
to
work
on
sure
like
at
the
meeting
yeah,
we
could
probably
just
oh
I'm
gonna
work
on
this
product
like
this
item.
It's
fine
as
long
as
like
it's
kind
of
list
of
the
items
somewhere
outside
of
the
meeting
of
the
greek.
G
K
Yeah,
maybe
a
combination
of
this
type
of
meeting
kind
of
where
people
are
sharing
kind
of
general
updates
and
what
they're
doing
and
kind
of
working
on
things
to
go
move
forward.
But
then
also
every
now
and
then
it'd
be
nice
to
have
kind
of
more
in-depth
things
where
somebody
gives
a
little
bit
of
a
like
if
somebody's
heading
off
to
a
conference
or
something
and
wants
to
give
like
a
practice,
talk
or
whatever
for
the
conference
and
give
a
little
more
detail
and
go
a
little
deeper
into
their
specific
project.
A
That
sounds
good.
Maybe
one
thing
we
can
do
with
that.
I
liked
sort
of
the
suggestion
of
agenda
items,
and
so,
even
if
either
posting
in
the
slack
channel
or
we
can
actually,
I
can
be
more
deliberate
about
creating
the
template
for
the
day's
notes
ahead
of
time
and
so,
for
example,
dom
if
you're
ready
to
demo
some
desk
stuff
next
week,
we
can
make
sure
you've
got
20
minutes
to
show
that
off,
which
would
be,
I
think,
extremely
valuable.
C
C
E
A
A
E
I
would
like,
if,
if
one
time
we
could
maybe
discuss
some
possibilities
to
fund,
maybe
slack
or
something
else
I
don't
know
it
happened
to
me
several
times
recently
that
I
was
looking
for
a
message
we
talked
about
and
it's
blocked
behind
the
paywall,
because
we
used
the
free
version
so
moving
forward.
I
think
it
would
be
good
to
somehow
think
or
estimate
we
could
even
ask
max
their
estimate.
How
much
would
it
cost
and
where
could
we
get
funding
for
it?.
A
Yeah,
that's
that's
a
great
point
and
I
think
that
that
leads
into.
I
think,
a
bigger
topic
that
we've
mentioned
a
couple
of
times,
and
I
chatted
with
joe
a
little
bit
about
this
this
week
is
what
sorts
of
things
do
we
want
to
place
with
sort
of
simpag
governance
and
roles
and
titles,
and
things
like
that.
I
mean
now
that
it's
a
bit
of
a
bigger
project.
A
Yeah
those
things
are
important,
and
so
maybe
the
question
to
ask
here
is
not
sort
of
dive
into
the
governance
discussion,
but
it's
actually,
where
do
we
want
to
have
that
discussion?
Is
that
on
the
I
think,
we've
got
a
simple
community
repo.
G
H
I'm
just
trying
like
I
wanted
that
to
be
something:
that's
this
facilitates
us
talking
about
it
and
chatting
about
it.
I
mean
slack
slack.
The
history
will
go
away.
Eventually,
it
yeah,
I
guess
it'd,
probably
be
nice
on
the
synthetic
community
if
it
weren't.
If
you
want
it
on
github,
if
not
maybe
on
the
discourse,
I'm
not
sure,
if
that's
an
appropriate
place
for
it
or
not,
but
it
could
be
a
good.
H
D
E
A
Yeah
well
peter,
maybe
one
thing
we
could
follow
up
on
and
if
you're
willing
to
help
here,
I'd
appreciate
because
you're
in
touch
with
matt
pretty
regularly.
I
think
that
there's
a
way
to
apply
for
open
source
and
or
education
slack,
I
think
that
we
would
probably
qualify
for
the
education.
Certainly
I
don't
think
we're
going
to
qualify
for
the
non-profit
slack,
because
we're
not
a
non-profit,
but
I
think
we
probably
could
apply
through
ubc,
as
this
is
a
education
slack.
A
I
don't
know
if
you
need
to
because
they
have
like
their
own
sort
of
like
with
fernando's
research
group
they
just
applied,
because
we
have
emails
that
are
associated
with
the
university.
Now
I
don't
know
if
that'll
yeah
and
I
think
that
there's
not
a
problem.
If
people
have
other
emails
as
well,
I
don't
think
that
restricts
it.
So
yeah.
A
A
Excellent
soggy
did
you
have
any
things
you
wanted
to
share
with
folks
or.
M
Nothing
nothing
particular.
I
really
want
to
look
at
the
joe's
layered
solution,
but
I
was
a
bit
busy,
but
I
actually,
like
the
project,
is
actually
finished
here.
I'm
going
to
get
a
little
bit
more
time
to
work
out.
A
Cool
okay:
well,
does
anyone
else
have
questions
or
things
you'd
like
to.
F
Yeah,
maybe
I'll
bring
up
one
more
thing
since
soggy's
here
I
think
last
time
we
spoke
so
I
was
working
on
the
spontaneous
potential
stuff,
and
so
as
we
have
it
right
now,
we
are
we're
defining
the
potentials
at
the
cell
centers,
which
makes
sense,
but
it
means
that
you,
the
boundary
conditions,
sort
of
make
it
so
that
you
can't
really
do
that
easy
problem.
F
M
M
That
should
be
fine,
okay
theory
so,
but
there
could
be
a
bug,
but
I'm
not
sure.
M
I
bet
that
part
shouldn't
be
a
problem.
I
guess
because
no,
I
think,
yeah
dev,
I
don't
really
see
any
problem.
M
F
M
F
H
F
H
M
M
F
Yes,
but
I
guess
I
mean
I'm
sure
everyone's
working
on
lots
of
stuff,
but
it
sounds
like
playing
around
with
spontaneous
potential
and
with
the
dc
ip
stuff
it
seems
like
there
would
be
some
benefit
to
to
implementing
the.
I
guess,
the
the
other
boundary
conditions
we
sort
of
have
like
an
incomplete
list
of
boundary
conditions
we
can
impose
on
on
tree
meshes.
M
And
if
you're
thinking
about
the
completeness
yeah,
we
should
impose
the
mixed
boundary
condition
to
the
normal
problem
as
well,
so
yeah,
our
cell
centered
problem.
So
you're
right,
I
I
should
have
done
but
because
it's
yeah
it
was
easier.
M
A
Okay,
I
think
that
probably
wraps
up
most
things
for
this
week,
so
we'll
try
we'll
give
this
sort
of
meeting
structure
a
try
with
like
quick
rounds
and
then,
if
folks
have
longer
discussion
items,
if
we
can
discussion
items
or
if
there's
something
you
want
to
present
anything
from
code
to
if
you've
got
a
cool
result,
you'd
like
to
share
with
the
group
if
we
can
just
make
a
habit
of
posting
those
in
the
items
before
the
meeting,
so
that
we
can
plan
our
time
so
cool
thanks.
H
Sagi,
just
so
you
know
before
you
leave
talked
about
having
get
the
primary
secondary
solution
on
that.
H
Put
up
the
pull
requests
that
he
has
a
primary
secondary
solution
in
the
dcip
for
one
of
them.
Everyone
had
a
conversation
on
that
on
on
github,
so
it's
documented
so
we're
moving
it
forward,
just
take
a
look
at
it
and
that
your
two
cents
in
when
appropriate.
H
On
adding
in
like
a
primary
secondary
type
solution
for
the
dcip
code,
if
you
could,
I
just
learned
that
the
conversation
to
make
sure
that
we're
doing
that
in
a
consistent
way
across
senpai
and
it's
what
we
intend
to
do
and
so
that
if
we
have
a
documented
saying
that
okay,
this
is
how
we
intend
to
do
it
moving
forward
for
other
further
sources,
good
for
other
systems,
right,
okay,.
M
Yeah,
so
how
that
that
that
the
primary
secondary
field
approach
works.
Oh.
H
I
know
how
it
works.
I
just
want
to
make
sure
that
we're
doing
it
in
a
way
that
was
and
we're
being
consistent
across
symphony
so
like
on
in
the
in
the
fbm
fdem
code.
It's
part
of
the
source
term,
like
the
source
class,
is
part
of
the
first
class
like
the
magnitude
as
a
primary
approach.
It's
part
of
the
simulation
itself.
It's
not
part
of
the
first.
M
Yeah
there
is
a
it's
like
a
there's,
a
little
bit
of
specialty.
What
roman
is
doing
because
it
takes
account
that
topography.
M
H
I'm
just
saying
like
I
just
want
to
make
sure
that
we're
doing
it
in
a
way
that
okay
well,
this
is
the
way
that
we
intend
to
be
consistent
across
synthetic
that
this
is
we
either
put
it
entirely
on
the
source
class
every
time
or
we
want
this
simulation
to
handle
it
over
right
right.
That's
right!
That's
what
I'm.
M
H
M
H
M
Okay
got
it,
I
thought,
like
my
impression,
was
he
have
he
hasn't
done
that,
but
he
did
you.
Did
it
just
like
this
within
this
last
week?
I
believe
so.
Okay,
nice.