►
From YouTube: SimPEG Meeting: January 22nd 2019
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).
B
B
Three
items
on
the
list
for
now:
we've
got
supporters
and
the
discretized
doc
ones
chat
a
bit
of
burgers
wobbles
or
think
that
they're
a
bit
of
a
student,
a
finder
and
then
also
there
is
you
know,
like
came
around
approach.
How
Felix
in
the
old
cluster
is
going
to
be
seen
and
they're
planning
on
installing
Phil
Austin
is
leading
the
charge
on
that
planning
on
installing
and
you
on
there,
which
is
pretty
cool.
Oh
nice,.
B
C
B
Like
I
guess,
the
connections
that
are
you
trying
to
share
on
my
end
state,
you
guys
all
know
and
sort
of
thinking,
narrator
I'm
gonna,
be
putting
in
a
grant
like
rip,
NGO
or
belated
depend
you
so
I'm
working
quite
closely
with
the
pan
geo
group.
So
ask
questions,
and
things
like
that
come
up
like
I'm
happy
to
serve
as
a
connection.
Point
I
also
think
that
this
is
a
real
opportunity
for
us
to
actually
bring
some
paid
into
the
pinch.
B
You
know
world,
so
they,
when
they
first
enthrall
Pangea
simpang,
probably
won't
be
in
that
image.
But
I
think
that,
like
very
rapidly,
we
should
make
the
case
that
same
page
should
be
installed
an
available
campus
life,
and
that's
where
having
a
handful
of
examples
will
be
quite
useful,
so
some
stuff
that
we
can
showcase.
A
B
B
If
we
get
a
nice
mag
example
like
it
would
be
nice
too,
for
working
with
open
source
data,
so
something
that
we
can
download
from
the
camiĆ³n
government
or
that
we've
already
gotten
permission
to
share
widely
would
be
yeah,
quite
beneficial,
mm-hmm
and
then
there's
something
you
can
think
of
that
like
within
the
department,
would
spark
somebody
else's
interest
and
get
them
excited
about
using
something
like.
Let's
take
note
of
that
and
make
that
happen.
I.
D
B
Don't
you
have
like
you
sent
me
a
couple
of
examples.
Is
that
are
the
examples
that
you
sent?
Are
they
running
off
of
master.
D
B
C
D
E
E
B
B
B
Uh.What,
quite
it's
actually
closer
to
running
on,
like
the
UTC
Jeter
hug.
Okay,
so
like
then
we
will
pre-populated
software
environment.
E
B
I,
don't
know,
but
I
think
that
that's
like
an
important
thing
that
everyone
at
UBC
uh-huh
like
just
keep
your
keep
involved
in
that
conversation,
because,
ideally,
what
they
should
do
is
they
should
create
like
a
public
repository
that
describes
the
deployment
and
then
we
can
make
pull
requests
to
it.
So
that
would
be
like
the
ideal
workflow.
Is
that
Phil
or
whoever
is
administering
this?
They
have
a
public
github
repository
that
describes
the
setup
right.
A
B
E
B
D
You
guys
are
done
yeah,
you
guys
are
not
on
the
channel
away
on
the
timon
cluster
Channel.
No,
you
shall
receive
an
invite
from
failed
to
join
the
the
EOS
and
them
there's
a
private
channel
for
400
I.
B
Yeah
school
like
if
there's
it
sounds,
like
dawn
you're,
probably
the
most
connected
at
this
point
so
but
you
can
sort
of
just
like
help
nudge
them
things
so
that
we
can
be
a
part
of
the
conversation.
There
I
think
that
that
would
be
huge
at
this
stage.
Yeah.
D
C
B
I
gets
to
it
like
you,
it
would
be
worst
asking
how
closely
connected
they
are
to
pend
you
and
how
closely
connected
they
want
to
be,
because
I
can
try
and
help
make
connections
where
needed
and
then
I
think
like
how
open.
Basically,
they
will
be
too
including
packages,
and
things
like
that,
like
synthetic.
B
B
A
A
A
B
A
B
F
B
So
I
think
the
purpose
of
the
binder.
So,
yes,
we
deployed
a
binder
first,
we
want
to
show
weights
what's
impact,
can
you
and
basically
make
the
case
as
to
why
it
should
be
much
more
widely
available
to
people
through
a
platform
like
Pangaea?
So
the
purpose
of
need,
of
course,
is
to
demonstrate
the
workflow
and
start
getting
to
the
hand
of
other
people.
So.
B
B
A
rough
version
of
the
discretized
Docs
to
Google
App,
Engine,
mm-hmm
I'm,
just
gonna,
share
my
screen
here,
but
the
URL
is
in
the
notes.
Okay,
yeah
I'll
see
that
yeah
yeah.
So
what
I've
tried
to
do
is
show
this
briefly
last
time,
but
show
like
have
some
getting
started,
some
motivation
as
to
why
you're
gonna
use
discretize
how
to
install
it
that
sort
of
stuff
have
a
user
guide,
so
the
user
guide
will
be
all
Python
scripts
that
are
then
built
by
like
this
Fink's
gallery.
B
So
everything
should
include
code,
and
this
whole
thing
I
can
download,
is
a
Jupiter
notebook
and
understand
yeah,
and
it
just
talked
to
you
through
your
various
pieces.
So
that's
going
just
talks
about
here's,
some
different
meshes.
Here's
how
you
count
the
things.
Here's!
What
faces
and
edges
are
that
sort
of
thing.
B
B
Then
I
think
I,
don't
remember
if
these
ones,
so
this
one
is
currently
like
a
very
short
rst
page.
This
should
be
translated
over
to
a
Python
script
and
same
thing
with
inner
products,
for
example.
These
things
should
be
transferred
over,
but
that's,
okay,
I
think
that
we
can
sort
of
get
this
going
as
is
and
then
start
making
issues
about
what
kind
of
examples
and
use
cases
we
need
to
demonstrate
in
the
docs
like
to
make
this
useful
mm-hmm.
B
B
Yep,
yes,
so
here
the
API
we're
actually
going
in
and
documenting
all
of
the
classes.
So
if
you
go
in
and
look
at
tensor
mesh,
it
shows
you
what
it
inherits
from
and
then
this
is
so
I'm
working
on
this
with
Franklin
or
been
in
touch
with
Franklin
he's
he's
doing
the
work
I'm
talking
Oh,
so
the
there's,
a
distinction
between
like
the
numpy
style,
documentation,
which
I'll
render
more
like
this
and
then
the
more
custom
like
the
basic
Sphinx
style
which
renders
like
this.
So
eventually
all
of
these
will
be
closer
to
this.
B
So
that
gives
you
all
of
the
properties,
one
of
the
things
that
this
is
like
brought
to
light
for
me
and
I
think
this
is
a
broader
question
that
we
should
think
about
is
like
right
now.
We've
got
multiple
ways
to
access,
identical
information
and
I,
don't
know
if
we
actually
want
to
continue
doing
that.
So,
for
example,
all
of
the
information
that
you
need
for
vector,
number
of
edges,
X
or
sorry,
no,
that's
not
strictly
true.
B
So
like
number
of
faces,
X
can
be
grabbed
from
vector
number
of
faces,
so
you
know
it's
two
different
ways
to
access
the
same
information:
oh
yeah,
do
we
I
mean
because
just
looking
at
this
documentation
like
is
intimidating?
How
much
stuff
is
on
this
object,
and
so
that's,
okay,
I!
Don't
think
we
need
to
solve
that
now,
but
it's
just
it's
something.
To
keep
in
mind.
B
This
is
a
bug
that
will
get
figured
out.
There
should
be
just
a
quick
summary
of
each
of
the
methods
here
and
then
we
can
look
below
at
the
actual
documentation.
So
this
is
still
in
the
same
exact
page.
This
just
scrolled
me
down
to
the
right
place
which
describes
the
docs
for
forget
these
inner
product,
for
example.
B
The
other
thing
that
Leo
implemented,
which
I
think
is
super
cool,
is
so
after
it
describes
all
the
methods,
then
you
can
actually
see
which
examples
use
this
class.
So
if
you
wanted
to
see
examples
of
it
in
action,
you
can
click
on
any
one
of
these
and
it
takes
you
to
a
working
Python
script
that
makes
use
of
that
class.
Thanks.
B
B
D
F
B
So
what
I'm
trying
to
do
right
now
is
get
first
of
all,
get
the
API
reference
like
the
methods
being
described
properly.
So
here
when
we
look
at
the
tensor
mesh
like
this
table
should
be
filled
in
and
in
not
currently
and
I.
Don't
know
why,
so
once
that
is
sorted
and
wants
to
build
roughly
passes,
then
I
think
we
can
merge
it
by
roughly
passes
what
I
mean
is.
This
is
brought
to
light
a
lot
of
links
that
are
not
properly
rendered
in
the
documentation
and
all
of
that
sort
of
stuff.
B
So
what
I
would
like
to
do
is
then
we'll
merge
this
and
then
we'll
create
a
branch
called
docks,
or
something
like
that,
where
we'll
put
more
strict
testing
on
the
documentation
and
just
ask
that
as
people
have
time,
you
know
if
you
spend
half
an
hour
once
a
week,
just
beating
on
the
docks
a
little
bit
we'll
get
them
into
shape,
but
yeah
I
think
we
can
sort
of
get
it
into
a
state
where
it's
good
enough
with
merge.
B
D
B
So
one
thing
that
I
did
have
to
do
is
inside
the
docks.
There
is
a
makefile,
so
this
is
not
the
makefile
on
the
outside
of
the
docks
this
is
inside,
I
did
actually
go
in
and
modify
the
HTML,
so
the
way
that
we
build
is
slightly
different.
So
Dom
was
it
scaling
because
it
was
like
a
windows
makefile
thing
or
was
it
scaling
because
we
had
had
there
were
not
windows
friendly
I
will.
D
B
D
B
B
D
B
Sounds
good
I'm
so
then,
with
that
too
actually
well
I've
got
the
code
up.
I'm
gonna
show
you
just
like
what
the
numpy
style
documentation
looks
like
I
know.
We
showed
it
briefly,
but
I
think
it's
actually
quite
amazing
to
me
how
much
more
readable
the
docstring
is
even
in
the
even
in
the
code.
So
just
having
stuff
like
this,
where
you've
got,
you
know
an
example
and
then
there's
some
code
underneath
and
then
even
looking
at.
B
This
is
very
clean
and
friendly,
so
I'm
I'm,
all
in
for
going
for
this
format,
so
we
can
check
it
out.
I've,
cleaned
up
the
tensor
wrench,
the
inner
products
in
the
base
classes,
but
then
I'll
make
some
issues
as
to
like
what
hasn't
been
done
yet,
and
we
can
sort
of
break
it
up
into
small
chunks
so
that,
even
if
you've
got
ten
minutes
and
want
to
go
in
and
fix
the
docks
for
the
sill
mesh
or
whatever
it
shouldn't
be
a
hard
thing
to
do.
B
It's
sort
of
the
default
style,
whatever
glue
is.
B
B
So
if
I'll
hopefully
get
pull
request
set
up
in
the
next
couple
of
days
and
then
I
think
that
we
should
think
about
if
this
makes
sense
to
everyone
I'm
doing
the
same
thing
over
in
stag,
so
we
can
play
around
with
it
here.
First
think
like
this,
because
it's
a
nice
least
and
these
things
and
then
and
then
we
can
point
it
over.
It's
not
work.
Everyone.
B
B
Franklin
is
so
the
properties
there's
some
other
documentation,
stuff
that
happens
in
the
properties,
library
and
so
there's
a
couple.
People
who
have
asked
is
it
possible
to
get
the
property
stuff,
auto,
documenting
and
just
like
different
styles,
and
so
that's
something
he's
interested
in
in
making
happen.
So
I
will
check
in
with
him
and
just
see
if
there's
stuff,
that
you
know
we
can
do
to
help
make
that
happen.
Okay,.
A
B
D
B
Feel
free
to
make
an
issue
a
fact
that
we
don't
forget
about
it,
but
it
probably
won't
get
prioritized
if
we
don't
need
it
like
right
away.
Yeah.
D
B
The
other
thing
that
I
wanted
to
talk
about
the
meeting
with
Jennifer
and
and
iam
tutorial
for
that,
so
we're
scheduled
like
leave
or
next
day
or
Tuesday,
something
like
that
she's
actually
coming
up
pretty
quick.
So
what
I
was
thinking
and
I
meant
get
to
this
last
year,
did
not
it
to
start
a
to
tour
like
get
a
notebook
together
based
on
bootprint
on.
B
Do
you
one
time
domain,
one
frequency
domain
and
and
chat
through
I
think
just
using
like
the
course
impact
for
now
so
do
the
SIL
mesh
so
that
they
can
go
to
3d
right
away
because
I
think
they're
pretty
interested
in
3d
and
then
we
can.
We
can
introduce
like
song
these
extra
notebooks
for
using
to
make
em
one,
if
that's
that's,
where
they're
more
interested
in
going
so
yeah
I
don't
have
who's
able
to
join
that.
Actually,
let
me
just
look
and
see.
B
B
You
will
do
a
hangout
style
meeting.
People
can
come
and
go
as
they
need
to,
but
it
would
be
nice
if
yeah.
C
B
B
Okay
and
then
the
only
it
was
there
any
other
comments
or
like
thoughts
on
on
those.
B
E
B
E
B
B
Just
get
them
to
download
it
and
run
it
because
they're
gonna
be
using
their
own
tools
for
this
anyways
right.
E
B
B
B
B
Okay,
yes,
you
don't
yeah,
okay,
so
there's
there's
two
flags
that'll
be
important
for
this
to
run,
there's
one:
that's
called
library
can
fake
donkey,
Amal
and
all
it
has
in
it
right
now
is
so
you
have
information
so
which
user,
which
github
user
hosts
the
repository
or
organization
which
repository
it
is
so
right.
Now
we're
looking
at
the
gos
iLabs
repository
I
wanted
to
specify
a
branch
name
so
that,
for
example,
it
sequence
running
a
course.
B
She
wrote
a
notebook
or
one
of
his
grad
students
wrote
a
notebook
yesterday
and
they
need
it
in
their
students
hands
tomorrow.
He
could
actually
switch
this
to
whatever
his
French
is
and
then
there's
a
directory
of
where
the
notebooks
live,
so
that
when
I
search
for
the
notebooks
on
the
envy
library
inside
I
can
search
just
part
of
the
repository
rather
than
the
whole
thing,
then
there's
two
environments
set
up
things,
and
so
one
is
the
in
like
you,
you
don't
actually
have
to
specify
both
of
these,
but
I
think
it's
easiest.
B
If
we
do,
what
this
does
is
it'll
look
for
the
environment
file
that
we
specify
in
the
Geo
site,
labs
repository
and
it's
gonna
download
that
for
you
and
then
because
we've
specified
the
requirements.
It's
also
going
to
do
that
for
you,
so
it's
gonna
just
grab
those
files
so
that,
basically,
at
the
end
of
the
day,
you
end
up
with
a
repository
that
is
find
erisa
below
very
easily.
B
B
B
So
then,
the
next
file
that
you
need
is
this
dot.
Jupiter
include
file.
What
this
is
is
it's
very
similar
to
you.
Everyone
seemed
like
a
git
ignore
file
right,
yeah,
okay,
so
I'll
show
you
the
structure
of
that.
It's
a
very
simple
name:
matching
file
matching
systems,
so
here,
for
example,
for
the
sinks
documentation.
We
ignore
everything.
That's
under
the
docs,
slash
build
like
that's
just
not
going
to
be
going
into
the
the
get
history
right
and
the
way
that
it
keeps
track
of
that
is.
B
It
looks
for
any
file
that
has
this
string
characters
in
it,
and
then
it
says:
okay,
just
ignore
those
files.
Don't
ask
them
to
get
so
that
you
can
include
is
doing
basically
the
same
logic
but
backwards,
and
so,
in
this
case,
I'm
saying
okay,
I
want
this
exact
notebook.
This
exact
notebook,
but
I
could
also
just
specify
like
this.
B
B
Does
that
make
sense
yep
the
other
one.
That's
kind
of
neat
too
is
like
so
in
this
case,
under
electro
magnetics,
let's
say:
I
saw
the
sphere
examples
and
I
want
all
of
those.
So
then
what
this
does?
It
says:
ok,
anything
with
sphere
in
the
filename
I
want
you
to
download
that's
under
electro
magnetics.
B
B
Just
on
perhaps
the
longer
term
on
the
longer
term,
why
we
sort
of
separated
into
two
files,
so
what's
kind
of
cool
that
I
was
talking
to
you
via
boat
and
I.
Think
that
this
would
be
a
fun
thing
to
work
on
is
that
we
could
actually
basically
write
a
Jupiter
server
extension
that
looks
at
the
Jupiter
include
file,
and
then
basically
only
shows
you,
the
notebooks
that
satisfy
that
included,
and
so
it
would
be
really
nice
in
that
sense.
B
Is
that,
instead
of
every
instructor
having
to
create
their
own
courses
for
posit
or
ease
all
that,
they
have
is
right.
The
doc
Jupiter
include
and
then
keep
that
with
the
repository,
and
so
we
there's
no
more
diverging
or
copying
of
notebooks
anymore.
You
can
always
just
basically
have
your
repository
and
then
that
you
better
include
and
choose
what
you're
gonna
expose
the
students.
E
B
So,
ideally,
all
development
happens
inside
of
geocell
ABS,
so
any
of
you
know
roads
right
that
they
want
to
do
they
branch
and
create
a
pull
request.
So
ideally,
hopefully
it
doesn't
just
completely
diverge.
We
want
stuff
feeding
back
into
master,
but
ya
know
what
happens
there
and
then
all
they
do
with
their
like
geosite
course.
Repository
is
then
run
this
MV
librarian
update.
Basically,
and
then
it
should
be
github
repository
and
says:
oh
yeah,
these
ones
changed.
Let
me
grab
them
and
give
them
to
my
course.
No
yeah.
E
B
So
he'll
create
a
juice,
I
you'll
fork,
geosite
courses,
which
basically
is
the
skeleton
of
all
the
command-line
interface
to.
B
Like
it
has
all
the
MD
librarian
commands
that
you
need
in
order
to
set
up
and
fetch
all
the
notebook
c1.
So
let's
say
that
Daquan
wants
one
notebook
from
UC
resistivity,
one
from
him
that
or
not
so
all
he's
gonna
do
is
work.
The
geosite
courses
repository
change,
the
Jupiter
include
run
like
a
command
that
says
grab
me
those
notebooks,
and
then
he
will
send
that
repository
up
to
binder
know
if
he
discovers
a
mistake
or
wants
to
change
something
in
the
notebook
he's.
B
C
Well,
just
with
respect
to
the
last
meal,
so
Kevin
will
be
coming
here
in
the
more
than
a
week
and
yeah.
So
if
there
are
tasks
that
are
kind
of
dupa
like
I
started,
a
document
I
sent
it
to
him
and
as
well
as
to
both
human
and
soggy.
So
that's
just
a
place
that
we
can
kind
of
keep
track
of
things
that
might
benefit
him
as
well
as
ourselves,
and
so
both
from
the
point
of
view
of
inversion
and
tutorials
and
background,
as
well
as
various
documentation
just
to.