►
From YouTube: SimPEG Meeting, August 9th, 2023
Description
Weekly SimPEG meeting from August 9th, 2023
D
No,
we
were
luy
enough
to
not
get
into
a
bear,
didn't
see
too
many
wildli.
E
D
I
mean
yes,
we
saw
like
deers.
Marmots
are
yeah,
we
actually
saw
we
were
coming
down
one
of
the
the
hikes.
It
was
one
in
one
of
the
ski
centers.
We
were
right
next
to
the
chairs
and
we
saw
a
little
wolf
like
a
puppy
running
into
us
and
when
he
saw
us
he
turned
back
into
the
forest
and
we
kept
walking
and
at
some
point,
I
turned
and
I
saw
him
crossing
the
path
to
get
to
the
other.
D
E
C
A
Right
good
good!
Well,
let's
maybe
start
with
a
quick
recap
of
birs
I
know
we
chatted
a
little
bit
about
that
last
week,
but
you
weren't
here
last
week
so.
B
D
Okay,
yeah
I
was
reading
the
notes,
I'm
really
glad
that
everyone
enjoyed
it
I,
don't
know
I
I
said
it
a
lot
lot
of
times,
but
yeah
I
was
really
grateful.
That
people
was
very
energetic
during
the
worship
both
during
the
sessions
and
both
like
hanging
out
I,
really
like
that
yeah
they
were
really
into
U.
Getting
to
know
each
other
and
start
working
more
together.
I
saw.
D
From
like
students,
early
career
researchers,
let's
say-
or
people
not
heavily
involved
in
the
project
that
really
wanted
to
contribute
and
wanted
more
documentation
in
how
to
do
it
to
learn
how
to
do
it
so
I.
C
D
It's
very
refreshing
to
to
hear
that
and
yeah
I
can
complain
about
the
place.
D
D
C
E
B
Go
well
as
far
as
announcements,
yep
I'm
back
in
Colorado
I,
another
thing,
I
tagged,
20,
Sy,
020
that
should
be
getting
out
and
getting
through
I
think
it's
still
disseminating
the
last
I
checked
on
it.
It
takes
a
while
to
build
all
the
documentation,
but
it's
everything
seemed
to
be
going
well,
it's
all.
It
was
uploaded
and.
C
B
I
think
Santi
probably
saw
that
come
in
where
I
switched
up
to
using
the
the
setup
tools
SCM
Version
Control,
instead
of
we
were,
do
using
bump
version
before
so
it
just
it
changes.
It
reduces
a
step.
Now
you
just
have
to
create
the
tag
on
GitHub
and
it
releases
and
does
everything.
B
Instead
of
using
like
the
command
line
interface
to
bump
the
version
and
then
tag
it
and
then
push
the
tag
and
then
creates
another
commit
and
then
all
this
other
now
we
just
get
to
tag
it
on
GitHub.
It's
the
same
thing
that
we're
doing
for
discretise
and
geano.
A
B
B
D
E
B
It
looks
better
there's
no
errors
on.
It
looks
like
all
of
Devin's
regularization
stuff
made
it
in.
It
looks
pretty
nice.
E
C
E
B
Yeah,
so
it's
out
it's
ready
to
go
well,
I
think
it
still
hasn't
had
cond
of
Forge
yet,
but
that
usually
takes
a
little
a
few
minutes
or
takes
a
little
while,
but
it
should
be.
Okay.
I
know,
there's
now
a
since
there's
a
small
issue
that
gioan
required
UTM,
but
s
also
kind
of
required
it.
But
I
I
took
care
of
that
already
a
while
ago,
so
I'm
gonna
might
have
to
change
the
dependencies
on
the
cond.
C
C
A
All
right,
I.
A
Think
some
of
well
one
of
the
things
that
came
out
of
burs
I
think
be
great
to
Think.
Through
is
also
putting
just
a
bit
of
documentation
down
with
P
requests
and
really
thinking
about
what
we
can
do
to
streamline
that
and
to
be
bringing
in
things
faster,
and
if
we
see
something
and
if
it
is
an
improvement
to
you
know,
start
to
get
stuff
in
and
worry
about.
Iterating
later
is
I
think
you
know
heard
a
bit
of
feedback
from
folks.
A
We've
heard
this
before
too
that
if
you
know
when
that
process,
stalls
or
slows
down
with
lots
of
revision
requests,
it
does
dampen
sort
of
enthusiasm,
and
so,
if
there's
things
we
can
do
to
to
accelerate
that.
We
still,
you
know,
capturing
the
things
that
we
want
to
improve,
but
perhaps
doing
that
in.
C
A
Issue
or
or
something
like
that,
so
that
we
can,
you
know,
fix
things
up
after
the
fact
I
think
yeah.
If
there's
things
we
can
do
to
be
accelerating
that
process,
that's
something
I'd
be
really
excited
to
to
see.
B
Mean
there's
there's
a
difference
of
like
if
they're,
certainly
smaller
polar,
quests,
they're
easy
to
come
in
Faster
rate,
it's
like
it's
just
the
large,
the
ones
that
are
really
large
that
have
a
lot
of
changes
that
have
new
additions
and
stuff
that
we
still
need
to
be
careful
about
bringing
them
in.
We
just
like
bring
them
in
and
then
say,
we'll
fix
it
later,
because
we
generally,
we
won't
fix
it
later.
You
know
what
I
mean.
A
B
I
would
like
to
see
us
make
a
little
bit
more
use
of
it,
at
least
the
the
templates
that
we
already
have
out
there
for
the
PO
request
templates
just
to
keep
track
of
where
people
are
at
on
some
po
request,
because
I
I've
see
people
come
in
and
I
see,
people
like
take
that
and
then
just
delete
all
of
it,
and
it
gives
no
sense
of
where
things
are
so
I
think
I've
seen
that
a
couple
times
with
Devon
on
your
poll
requests
that
just
you
deleted
all
the
template
stuff
that
pops
up
right
away.
B
So
it
doesn't
really
give
a
sense
of
like
where
you're
at
in
the
poll
request.
I
would
like
to
see
us
make
use
of
that
more
often
than
we
do,
especially
for
people
come
like
new
contributors
might
be
able
to
point
them
to
a
few
more
guides
within
that
templates.
But
we
are
still
not
making
very
good
use
of.
C
E
Finally,
got
it
in
well:
I
started
the
PM
yeah
followed
the
template
there,
so
I'm
just
going
to
have
to
make
some
tests
and
yeah
a
few
comments.
Joe
I'll
just
have
to
let
just
to
see
what
you
want
to
do
with
it
so
to
make
it
work.
There
was
a
in
The
Meta
simulation,
there's
something
like
a
line
that
says:
if
persists
or
something
where
was
I,
don't
know
it
was
just.
E
It
was
like
a
setting
looking
for
true
or
false
and
for
some
reason
with
the
frequency
domain
and
the
natural
source
it
was,
it
wasn't
getting
triggered.
So
then
the
model
wouldn't
be
assigned
to
each
sub
simulation.
It
wasn't
doing
the
mapping
but
yeah
I,
I'll
I'll
P
you
in
the
pr
there
and
you
can
take
a
look
at
it.
E
Pretty
pretty
small
PR
and
scope
RS,
just
I
just
added
the
get
J
function,
the
get
jtj
and
I
added
another
function
called
compute
J.
It's
carried
over
from
what
I
had
from
the
D.
E
D
From
my
end,
I
did
a
little
bit
of
work
on
the
Sy
website.
One
example
is
that
there
were
some
link
to
the
contributing
guidelines
and
that
Ling
was
pretty
old
yeah.
He
was
pointing
to
a
Master
branch
which
I
don't
know
how
old.
C
D
And
he
was
also
pointing
to
contributing
to
rstd
file
and
we
are
using
now
Mar
file
for
that.
So
yeah
just
change
the
link,
and
the
other
thing
I
noticed-
was
that
we
currently
have
the
the
s.g.
repo,
which
is
the
actual
content
of
the
we
website,
but
there's
also
a
website
repo.
So
it
was
confusing
which
one
is
actually
the
sources
for
the
website.
So
I
asked.
D
An
is
should
to
archive.
old
one.
The
website
one
and
yeah
Rowan
already
did
that,
but
meanwhile
he
noticed
that
there
is
some
content
from
the
old
blog
in
there.
So
maybe
in
the
future,
we
want
to
recover
that
content
and
publish
it
again,
yeah
trying
to
follow
up
the
little
things
I've
been
doing
in
the
workshop
regarding
Computing
guidelines,
so.
E
C
E
F
I
can
yeah
so
playing
around
with
we're
trying
to
extract
the
tutorials
out
of
the
sneg
code
base
and
then
augment
them,
and
so
we're
going
to
build
that
with.
A
F
Apparently,
sort
of
an
improved
Jupiter
book
type
thing
from
what
I
understand
so
yeah
I've
been
able
to
get
misted
up
and
running,
move
some
of
those
tutorial
python
notebooks
out
and
build
a
test
website
for
that
locally.
So
it's
actually
pretty
straightforward.
You
can
build
it
really.
Similarly,
to
how
you
build
a
Jupiter
book,
you
can
also
make
a
a
talk,
file,
animal
and
you
can
customize
stuff
in
there.
F
So
yeah
that's
seems
to
be
working
pretty
well
and
then
I've
also
been
playing
around
with
this
G
mesh
stuff
to
make
a
fancy
mesh
to
try
basically
em
simulations
on
a
tetraedro
mesh.
So
that
is
a
work
in
progress,
but
I
think
I've
learned
a
lot
about
how
gesh
sets
things
up
and
I
may
be
able
to
turn
my
I
python
notebook
into
sort
of
a
tutorial
and
post
that
s.
So
if
anyone
else
in
the
group
is
trying
to
design
a
mesh,
you
know
discretizing
around
their
their
transmitter.
B
Do
you
working
off
the
the
examples
I
was
showing
you.
A
F
Some
like
subtlety
in
terms
of
what
are
you
actually
doing,
you
know
you're
kind
of
putting
points
in
a
model,
but
then
you
also
have
to
add
them
to
the
mesh
and
there's
just
kind
of
some
overarching
Concepts.
That
I
wasn't
exactly
sure
where
to
it.
Wasn't
like
explicitly
given
to
me:
I
had
to
sort
of
trial
and
error,
but
yeah.
It
makes
sense
for
the
most
part.
Now.
C
B
C
B
Yeah,
like
I
I,
imagine
it
would
be
nice
to
get
the
tutorials
out
of
the
documentation
like
it
just
takes
so
long.
The
other
thing
that
I
was
going
to
I
was
meaning
to
create
a
issue
on
this
and
that
I
I
really
want
to
work
towards
getting
a
lot
of
the
useless
items
that
are
in
the
name
space
as
in
like
useless.
As
far
as
a
user
is
concerned,
I
want
to
move
a
lot
of
them
to
Hidden
variables
like
things
like.
A
User,
oh
sorry,
like
the
physical
property,
matrices.
B
A
A
You
captured
so
Sagi
started
a
like
tutorial
for
developers
at
burs
like
how
to
create
a
new
simulation.
Did
he
link
to
that?
Or
did
we
capture
that
in
some
way,
because.
B
B
F
C
B
C
A
C
B
Yeah
I
got
S,
I
got
the
I,
have
it
using
setup
tools
CM
now
to
make
sure
to
report
the
version
back
on
installation.
C
0.20
nice
yeah.
B
G
Speaking
of
versions,
so
do
we
have
a
list
of
things?
Where
is
your
PR
for
020.
G
G
B
Yeah
so
I
generally,
we
create
create
a
like
a
staging
thing,
where
we
add
release,
notes
and
capture
that
sort
of
information
about
all
everything
that
was
happened
from
the
last
tag,
and
that
goes
into
the
release
notes,
and
then
it's
also
part
on.
So,
if
you
click
on
the
release
on
the
GitHub
website,
like
there's
the
releases
and
there's
ver
version
020
on
the
right
hand,
side
there,
if
you
click
on
that,
you'll
also
see
every
like
all
the
PRS
that
went
into
the
last
one
as.
C
B
But
on
on
the
documentation,
there's
release
notes
right
away:
I
did
not
I
will
fix
that
forgot
to
add
the
020
to
the
to
the
table.
Contents
on
that
thing,
but
I
can
switch
fix.
B
G
G
Isn't
it
a
bit
dangerous,
though,
that
we
like
I,
see
that
you,
you
know
you
added
a
class
in
the
internet
like
we're
doing
stuff?
That
is
not
strictly
related
to.
B
G
B
G
G
Request
cool
well
congrats,
so
20
and
do
we
have
a
plan
for
what's
21.
C
C
D
So
then,
there's
an
issue
of
request
that
I
really
want
to
be
on
particular
release.
I
just
add
that
Milestone
to
that
repo
to
that
issue
or
PR,
but
yeah
we've
been
discussing
about
trying
out
the
projects.
D
But
basically,
mostly
for
trying
to
group
stuff
together
into
different
projects
like
okay
working
on
formal,
so
include
several
issues
related
to
that
project
or
working
with
like
fft
filters
and
Transformations.
That's
another
project
and
issues
to
that.
But
for
versions
I
usually
go
with
milon.
It's
really
easy,
then
to
say:
okay,
I
really
want
to
make
the
release.
What
is
needed
to
be
done
before
that,
and
you
have
a
list
of
those
Stu
that
sounds.
G
We're
breaking
sorry:
how
are
Milestones
used
I,
never
used
milestones
in
GitHub.
So
how
are
you
filtering
with
them
like.
E
E
F
D
D
I
usually
don't
put
a
description
if
you
want
and
then
when
you
go
to
your
issues,
let's
say:
I
want
this
one
to
be
in
the
next
one
you
just
put
like
here
and
select
the
Milestone
you
wanted
and
then,
if
you
go
to
the
oh
sorry,
if
you
go
to
the
issues
then
to
the
whole
list,
you
can
filter
by
Mouse.
So
if
I,
if
I'm
about
to
release
0.6
I,
can
just
click
here
here
and
say:
okay,
what
needed
to
be
done
in
this
case.
C
B
C
A
Milestone
or
you
just
create
a
milestone
for
the
release
and
then
just
tag
tag.
The
issues
that
you
want
to
be
a
part
of
that
is
that.
D
D
E
G
Yeah
and
I
guess
the
advantage
of
project
is
that
we
can
keep
track
of
what's
under
review.
What's
what's
under
Dev
all
that
stuff,
but
then,
as
de
said,
we
have
to
be
diligent
to
be
using
it.
Otherwise
it's
too
much
for
what
we
want
so
I
guess
we
should
have
a
meeting
and
decide
what's
on
on
20
one
right
which
isues
all
the
bugs
picks
bugs
first
new
features
and
then
just
scope
it
out
next
six
months,
type
of
thing:
when
is.
B
Quarter
or
something
for
about
every
quarter
but
I
still
don't
so.
B
G
It
well
I
mean
it's
easy
to
just
assign
issue
existing
issues
to
the
current.
You
know
you
create
all
your
Milestone
like
21,
22
23,
and
then
we
say
every
quarter
and
then
we
just
tag
like
this
is
top
priority
like
move
it
to
you
know,
and
then
we
fill
it
up
like
okay,
that's
enough
items
for
One
release
and
then
the
rest
can
go
down
to
the
next.
B
A
F
A
D
One
thing
we
could
do
is
create
an
issue
like
the
one.
You
mentioned,
something
like
a
meta
issue
with
like
just
ideas
like
broad
ideas,
and
you
can
list
up
in
GitHub
and
if
you
have
this
list
list,
they
create
like
this
button,
convert
to
issue,
that's
cool,
so
you
can
list
stuff
like
very
broadly
and
when
you
want
to
actually
describe
what
is
needed
to
be
done,
just
hit
that
button
cre
isue,
for
it.
C
F
C
G
C
B
G
A
a
ranking,
our
point
system
for
those
but
I
guess
we'll
come
over
time.
It's
always
different,
but
we
can
eyeball.
Maybe
do
that
next
next
meeting,
if
you
guys
agree,
we
we
do
just
a
just
a
a
release,
21
brainstorm
of
like
what
what
needs
to
come
in
what
we
want
for
next.
A
Quarter
that
sounds
good
and
yeah
we
can
come
I
mean
I,
think
aiming
for
about
every
quarter
sounds
good
and
I.
Think
it's
fine,
if
there's
some
that
are
shorter
or
longer.
That's
also
fine
like,
for
example,
I
mean
bringing
in
John
Weiss's
mag
pte
stuff
would
be
great
to
have
that
prioritized
for
near
future
and
if
that's
a
small
release
on
its
own,
like
that's
also,
we
can
kind
of
scope
out
what
feels
appropriate
for
for
release
to
kind
of
balance.
The
timeline
and.
C
A
C
A
That
I've
got
a
couple
updates.
If
anybody
are
we
closed
on
that
topic
for
now,
yeah
I
started
drafting
a
statement,
Santi
and
Rona,
provided
some
input
for
website
analytics
we've
been
using
Google
analytics
in
the
past,
but
there's
a
lot
of
problems
with
that
and
moving
over
to
plausible
seems,
like
a
nice
thing
to
do
so,
we'll
put
in
a
propos
proposal
to
see
if
software
underground
is
willing
to.
Basically
let
us
use
some
slots
for
that.
A
They've
already
allowed
us
to
use
the
top
level
domain
for
S,
which
is
great
but
it'd,
be
nice
to
pull
the
Geo
siiz
sites
over
there.
One
of
the
things
that
it
brought
up
is
that
we
have
a
lot
of
subdomains
with
our
websites
and
each
subdomain
takes
up
a
different
slot
and
some
ways
it'
be
better
to
have
folders.
Is
that
like
it's
all,
just
one
website
and
I
think
it
does?
R
is
saying
too
that's
a
better
approach
for
like
search
ability
and
website,
crawlers
and
stuff
like
that.
A
So
that's
something
we
might
want
to
consider
like
switching
over
to
and
I'm
pretty
sure
it
should
be
straightforward
to
put
up
a
redirect
on
the
subdomains
to
like
map
us
to
the
right
folder.
But
that's
a
thing
we
can
start
to
make
some
issues
on
I
meant
to
do
that,
but
haven't
gotten
to
it.
But
if
that's
something
anybody's
interested
in
in
helping
tackle
it'
be
great.
If
we
can
put.
C
F
A
For
evenings
when
you
want
to
do
some
code,
but
it's
not.
A
Not
selling
pdes
so
yeah.
G
A
B
C
D
Each
one
of
them
is
their
only
head
pages,
and
basically
you
change
like,
like
the
I
mean
you
need
to
assign
the
domain
for
each
one
of
them,
I
think
and
and
that's
it
yeah
or.
C
F
A
D
A
Oh,
it's
on
the
the
bir.
E
C
E
G
A
Yeah,
but
if
you
were
willing
to
provide
you
know
a
few
points
on
burs,
just
kind
of
quick
summary
and
we
can
write
up.
We
have
to.
We
owe
burs
a
report,
but
it
would
be
nice
to
actually
post
that
somewhere
too.
In
the
past
we
have
had
a
simpeg
Blog,
but
we
haven't
used
that
in
quite
a
while.
So
we
can
figure
out
a
place
to
put.