►
From YouTube: NuPIC Sprint Planning - May 22, 2014
Description
This is scheduled a day earlier than usual.
A
A
A
A
Okay,
so
nothing
huge.
I
I'm
happy
to
get
the
mac
os
x
and
the
build
mix.
So
that's
that's
good.
Let's
go
to
open
issues
and
new
core
okay.
A
So
here's
something
I
know
nothing
about
this
assembly
thing
I
think,
let's
see
what
the
status
of
this
is
right.
So
it
is,
I
think
it's
basically
ready
for
review.
So
last
I
checked
this
was
on
monday.
I
asked
wilson
k
to
merge
master
in
it.
He
has
so
let
me
mark
this
as
ready.
A
A
There's
the
associated
ticket
austin
took
on
this
nuked
core
he's
he's
out
today,
so
he
was
planning
on
trying
to
get
a
building
on
solaris.
I
don't
know
what
the
status
of
this
is,
but
I
think
we
might
as
well
keep
it
in
case.
He
finds
time
to
do
that.
He
seemed
to
be
very
interested
in
that
helping
the
guys
from
fujitsu
get
new
running
on
their
big
servers,
so
here's
another
ticket
on
starting
input
dimensions
that
is
currently
chaitan's
ticket.
A
So
I
still
have
this
ticket
on
defining
the
public
api.
I
think
there
is
a
wiki
page
for
that.
I'm
not
sure
how
much
it's
been
updated
yeah
here.
It
is
so
not
not
a
whole
lot
of
of
stuff.
Nothing,
nothing
really
down
here,
except
places
to
start
looking,
but
we
wanted.
I
wanted
to
try
and
define
just
in
in
a
document
what
needs
to
be
defined
as
the
api.
So
then
we
can
make
sure
that
those
things
are
documented
and
that
they
have
tests
written
for
them.
A
So
that's
still
work
to
be
done
since
you're
currently
assigned
pseudotype
is
out.
I
don't
know
when
he's
getting
back.
Is
he
back
on
tuesday,
scott.
A
Neither
okay,
I
don't
leave
that
in
his
court,
so
the
api
docs
utensil
has
been
working
on
that.
So
I'm
gonna
keep
that
on.
I
don't
know
if
you
guys
have
seen
it,
but
he's
actually
done
some
work
just
recently.
A
A
Doxygen
than
I
do,
this
is
apparently
more
maintainable
or
sustainable
output
format
that
he
updated
to
for
oxygen
1.8.7,
so
moving
forward.
This
is,
should
be
more
supportable.
So
that's
good.
A
A
Okay,
where
am
I
right
engine
dock?
So
that's
still
going
on.
I
think
you
can
still
still
have
some
things
to
do
there.
The
help
text
for
cmake
options
is
misleading.
A
Let's
see,
okay,
that
seems
like
a
valid
issue
seems
like
an
easy
one
to
fix.
I
don't
know
you
could
probably
just
fix
that
or
somebody
so
we'll
leave
that
up.
That's
that's!
I
got
it.
Help
wanted
a
newbie,
so
it's
an
easy
thing
to
do.
We
will
leave
it
around
and
then
the
unit
tests
thing.
I
got
a
little
bit
of
support
on
this
on
the
mailing
list.
A
Some
people
expressed
their
interest,
but
so
I
was
waiting
on
on
fergal
to
see
if
he
could
get
something
done
on
this
so
furlough
this
is
this
was
in
your
court,
you
volunteered
for
it
I'm
going
to
sign
it
to
you
and
move
it
along
into
the
next
sprint,
so
you're
on
tap.
A
A
Okay,
now
new
pic
all
right,
swarm,
error
messages.
That
is
an
open
pr.
A
A
I
have
been
working
on
the
odd
gem
anomaly
sample,
but
it
uses
the
anomaly
likelihood
algorithm
in
new
pic,
so
I
don't
want
to
merge
that
until
the
other
is
merged
in,
but
there
is
already
a
video,
it's
just
not
published
yet
right
so
more.
I
sorry.
I
forgot
your
name
but
more
from
portland
he's
working
on
this
unit
test
cleanup.
A
He
was
going
to
submit
a
pr
on
this
hasn't
happened
yet
so
I'm
just
going
to
leave
it
around.
Hopefully,
he
works
on
that
he's
on
irc
a
lot,
so
I
can
bug
him.
There
run
swarm
fail
on
ubuntu.
I
do
not
remember
what
the
status
of
this
was.
Let's
see
related
to
888,
oh,
I
think
this
was
this
had
to
do
with
the
misleading
error
and
we
could
not
really
identify
exactly
what
was
wrong.
A
So
hopefully,
once
the
swarm
error
message,
vr
is
merged
in
we'll
get
a
better
idea
of
what's
wrong
here.
So
the
there
is
no
job
result,
so
something
something
is
either
wrong
with
the
input
or
the
swarm
description,
most
likely
so
I'll
leave
that
open.
Hopefully,
I
can
follow
up
on
that
after
we
get
better
error
messages
in.
A
A
Yeah,
so
this
is
also
a
valid
issue
with
swarms.
I'm
going
to
sign
that
to
me.
That's
I'm.
I've
worked
in
that
area,
so
I
should
take
that.
I
know
what
he's
talking
about.
Solaris
is
still
in
austin's
court,
so
this
is
a
ticket
for
new,
because
this
is
the
sister
ticket
for
the
core.
This
error
thing
that's
the
pr
I
referred
to
before.
A
B
The
map,
potential
and
and
dimensional
stuff
that
was
work
that
trayton
did
to
support
arbitrary
numbers
of
dimensions
in
the
spatial
pooler,
but
that
pr
is
is
waiting
on
it,
we'll
see,
plus
plus
one,
so
they
can
be
merged
together,
and
I
think
he
is
tentatively
planning
on
doing
it.
This
next
spring.
A
Okay
sounds
good
to
me
yeah,
so
there's
utensils
saw
some
ideas
about
how
to
improve
the
readme.
Some
of
them
have
been
implemented.
I'm
gonna
leave
it
open
because
there's
still
some
things
that
could
be
done
there.
A
A
Okay,
this
run
swarm,
produces
model
params
with
non-temporal
model
and
also
with
tp
enable
is
true,
that's
just
a
I
believe
again.
This
is
a
confliction
just
in
the
output,
but
I'd
like
to
find
some
time
to
dig
into
it
and
into
this
swarm
process
and
and
see
exactly
what's
going
on
there
if
there
are
just
defaults
that
always
show
up
and
some
override
the
others
I'd
like
to
get
the
the
model
crams
that
come
out
of
this
form
to
be
a
little
less
confusing.
A
So
if
there's
conflicting
details
in
there,
we
should
probably
fix
it,
and
here
is
the
intex
developer,
build
thing,
so
this
is
sort
of
become
I'd
like
to
maybe
change
the
label
of
this.
What
we
need
is
a
feature
for
developers
that
basically
does
what
intex
developer
build
used
to
do
now,
as
we've
discussed
before
the
once,
we
get
new
big
pip
installable.
A
This
shouldn't
be
an
issue
anymore,
but
in
the
meantime,
it's
still
a
major
pain
for
developers
to
make
python
changes
without
rebuilding
all
of
the
seed
stuff
running
cmake.
A
So
there's
a
workaround
with
sim
links
in
the
release
directory
which
I've
I've
been
doing
as
well
now,
but
it's
a
pain
in
the
ass,
so
we
should
leave
this
around
and
hopefully
somebody
will
work
on
it.
Actually,
I'm
going
to
keep
it
open
and
probably
add
some
clarification
here
to
exactly
what
this
means
and
probably
change
the
title
to
be
more
descriptive
about
what
this
is
more
of
a
feature
request
or
a
feature
replacement
of
something
that
was
removed.
A
Okay,
this
problem
running
python
tests
is
that's
still.
I
can't
reproduce
this.
I
don't
know.
What's
going
on
with
it,
but
I
I
I
should
try
a
little
more
oops
I'll
label
signing.
A
I
get
some
weird
weird
failure
in
python
tests
that
I
don't
know
how
to
fix,
but
and
then
the
improved
hot
gem
anomaly
example.
That's
in
progress
waiting
for
vrs
to
merge.
A
A
And
fill
the
old
sprint
and
we're
experimenting
23.,
there's
a
bunch
of
satellite
projects.
I
don't
really
have
a
good
view
for
these
yet,
but
I
do
sort
of
have
a
command
line
tool
that
I
use
so
this
this
is.
These
are
all
of
the
other
projects,
the
tooling
server
darwin
and
linux
64
and
the
wall
board.
These
are
the
only
things
that
have
issues
that
are
currently
in
sprint
23..
A
So
I'm
not
going
to
go
over
all
of
these
and
and
github
ui,
just
because
I'll
be
switching
between
a
bunch
of
tabs,
but
all
of
the
darwin
linux,
64
stuff
is
assigned
to
austin.
None
of
them
are
extremely
important
or
critical,
but
they're
all
stuff.
That
does,
I
believe,
need
to
be
done
at
some
point.
A
This
is
an
interesting
one
for
for
the
tooling
server
anytime,
there's
a
pr
that
merges
a
new
core
to
master.
So
when
you
pick
core
gets
updated,
it
would
be
kind
of
cool
if
automatically
a
pr
was
created
for
new
pic
that
updates
to
the
new
core
shaw
and
just
sits
there,
so
that
the
build
can
run
against
it.
So
we
can
get
an
indication
that
everything
still
works
properly.
A
Then
it's
up
to
our
discretion,
whether
we
want
to
merge
that
and
update,
maybe
record
a
point
to
the
new
or
excuse
me
to
update
nuke
to
point
to
the
new
new
core
shock.
So
I
kind
of
like
that
idea.
So
I'm
going
to
keep
that
around
and
try
and
work
on
it
and
then
the
the
wall
board
thing
there's
a
couple
of
a
couple
of
things
here:
jeff
is
assigned
to
one
which
is
just
a
nice
to
have
thing
that
the
developer
tab
be
its
own
tab
or
its
own
page.
A
Excuse
me,
because
that
makes
a
lot
of
http
calls,
and
sometimes
it's
not
used,
so
we
do
have
a
rate
limit
that
I've
exceeded
occasionally
having
the
wall
board
up
and
then
I
would
like
a
page
on
there
and
I
have.
I
have
the
back
end
functionality
in
place
for
this.
I
just
need
a
ui
to
display
all
the
issues
across
inventory
repos,
because
it's
kind
of
a
pain
to
go
between.
A
I
don't
know
eight
or
ten
different
repositories
that
we
have
just
to
see
what
issues
are
currently
open
in
the
current
sprint,
so
I've
I've
got
a
command
line
tool
that
does
this
and
which,
which
is
nice
for
me,
but
it's
it
would
be
nice
to
have
a
on
our
wall
board
somewhere
and
that's
basically
it.
I
believe
that
was
a
fast
sprint
planning
meeting,
but
I
don't
think
I
have
anything
else.
A
We've
got
more
than
enough
work
on
in
the
next
sprint
in
newport
news
core,
I'm
just
going
to
keep
plugging
along,
as
I
have
been.
Hopefully
you
guys
will
too
does
anybody
else?
Have
any
comments
or
questions.