►
From YouTube: NuPIC Sprint Planning - June 6, 2014
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
Well,
any
big,
it's
just
me
all
by
my
lonesome
today,
a
lot
of
the
guys
are
out
of
the
office,
but
I'm
still
going
to
do
sprint
planning
at
least
it'll
be
recorded
for
everybody
to
watch
later.
So
first
we're
going
to
go
over
the
issues
that
have
been
closed
on
the
spread,
spread,
23
which
ends
today.
This
is
new
pic
or
so.
Some
of
the
things
we
got
done
under
the
core
we've
got
better
documentation
better.
A
Automated
API
Docs
I
closed
this
issue
about
defining
the
public
network
on
the
with
you
called
on
some
unused
and
Scott's
going
to
be
on
in
a
minute.
So
I
closed
this
issue
about
the
network
API
on
the
wiki,
because
I
realized
it
was
already
pretty
well
documented
and
utensil
had
no
problem
started:
kicking
off
getting
the
api
docs
in
place
with
Lopes.
Already
there
kelly
wilson
helped
us
out
fixing
upper
assembly
to
support
64-bit
some
of
the
c
code.
So
thank
you
for
that.
A
We've
got
you
way
and
then
the
offices
an
intern.
He
has
helped
us
update
the
spatial
polar
documentation
which
looks
nicer,
some
diagrams
involved
there.
So
that's
good!
That's
again
and
the
automated
api,
docs
and
amenta
org
slash
docs,
some
more
api,
doc
improvement
and
then
another
assembly
bug
fix
so
that
was
neva
core
or
new.
Pic
a
lot
of
closed
issues.
A
We
got
the
chance
to
another.
One
of
our
interns
jared
worked
on
removing
some
of
this
old
LPF
code,
that
was
legacy
code
and
the
test.
So
that
was
some
cleanup.
I
put
in
place
a
regression
test
architecture,
russian
test
suite
that
is
currently
up
and
running.
There's
one
regression
test
there.
It
doesn't
really
do
a
whole
lot,
but
the
structures
in
place
it
runs
automatically
on
push
to
the
big
bastard
and
then
emails.
The
hackers
mailing
list.
A
See
some
readme
improvements
fixed
some
errors.
The
errors
when
swarming
fails
are
now
better.
The
used
to
be
the
errors
were
kind
of
hidden
within
a
stack
trace
and
now
they're
shoved
up
front.
Another
swarming
fix,
which
I
worked
with
somebody
else
on
that.
So
thank
you
see.
We
add
it
as
a
as
a
part
of
the
hot
Jam
nominally
example.
We
added
the
anomaly
likelihood
code
directly
into
new
pic,
so
that
is
now
a
module
with
a
new
bikini
you
can
use.
This
is
the
this
anomaly
likelihood.
A
A
This
is
another
swarm
issue
that
was
committed
by
the
community
that
helped
clean
up
some
of
the
swarming
direct
like
directory
handling,
more
documentation,
stuff,
David
ragazzi
has
been
working
on
this
pip
install
feature
and
as
a
part
of
that,
he
moved
around.
Some
of
the
directories
so
that
it
would
be
more
fitting
of
a
regular
Python
project
for
the
dist
tools.
A
Let's
see
that's
the
same
thing,
more
documentation,
I
had
a
an
issue
with
Travis
wear
it
wasn't
installing
things
on
the
first
try
so
I
had
to
reach
rest.
Is
enough
not
a
big
deal
all
right.
So
that's
what
we
got
done
so
quite
a
bit
of
stuff
on
you
pick
there
too
so
currently
open
pull,
requests
nipa
core!
A
There
is
one
open
for
quest
that
utensils
been
working
on
to
include
well
to
help
out
with
our
move
towards
having
a
windows
friendly,
build
so
I,
don't
think
this
is
going
to
get
us
all
the
way
there,
but
you
can,
but
it's
going
in
the
right
direction
anyway.
So
that's
the
only
for
us.
We
have
on
new
trick.
A
Let's
see,
excuse
me
a
new
pic
or
so
on
new
pic.
We
have
several
open
for
request.
Old,
most
of
them
are
are
not
an
emergency.
All
of
them,
I
think
are
not
in
a
virtual
state.
So
this
first
one
is
still
an
old
one.
That's
probably
just
going
to
be
removed
as
soon
as
pip
installs,
but
in
place
that
we
don't
need
this
anymore.
A
A
So
this
cleanup
issue
still
needs
some
some
attention,
so
that's
again
and
forgot
your
first
name
mr.
Moore
anyway,
that's
in
your
court
and
then
the
anomaly
modular
again
is.
These
are
all
things
that
the
really
most
responsibility
is
on.
The
creator
of
the
poor
applies
to
get
builds,
passing
and
running,
etc.
So
there's
nothing
that
we
can
really
take
action
on
here
at
the
moment,
but
they're
all
at
least
at
this
point
valid
PRS.
A
So
let's
look
issues.
I
have
got
the
issues
all
in
one
page
here
so
first
about
23.
That's
where
we're
all
right
now,
current
new
pic
core
issues,
let's
just
plug
through
these
real
quick
and
roll
to
click,
disk
fashion
and
then
pop
them
in
to
into
the
experience
we
want
to.
Yes,
definitely,
I
want
the
unit
test
to
generate
X
to
install
reports.
Fergal
is
still
working
on
this.
You
just
notified
me
today
squeeze
it
made
progress
up
but
Fergal.
A
If
your
watch
and
I'd
love
CPR,
even
if
you're-
not
even
if
it's
not
ready,
I'd,
be
more
than
willing
to
test
things
out
for
you,
okay,
so
that's
going
into
next
sprint
fixed
current
build
issues
with
new
pic.
This
is
important.
I've
got
a
lot
of
comments
and
links
in
here
about
this,
but
new,
pic
or
head
of
Master
currently
is
not
compatible
with
new
pic
at
a
master.
A
So
this
is
not
a
critical
issue,
but
it's
something
we
want
to
get
fixed
as
soon
as
we
can
so
I
think
I
sinus
to
David
you
we
chatted
about
this
and
on
another
issue,
and
he
was
going
to
tackle
this
and
once
he
does
get
a
bit
core
fixed,
a
PR
will
will
show
up
and
hopefully
pass
against
me
kick.
So
this
is
an
important
thing,
so
this
definitely
needs
to
go
into
next
friend.
A
Austin
says
he
will
be
working
on
this:
getting
solaris
are
getting
a
big
building
on
solaris
near
the
core
and
away
so
that's
valid,
Jade
and
I'm
going
to
assume
this
is
going
to
sprint
24.
They
even
responded,
I
think
Jayde's
other
office
today,
and
this
C
make
option
does
misleading.
This
is
a
really
a
very
new,
be
friendly
to
ask
if
anybody,
even
if
you
don't
necessarily
know,
see,
make
all
that
well
and
you
want
to
learn.
This
would
be
a
good
test
for
it.
A
A
12345678
issues
and
apparently
sprint
that
were
not
cleaned
up
I'm,
assuming
since
Austin
said,
the
other
issue
for
new
bacoor
would
go
into
next
friend
I'm
going
to
pop
this
one
in
there
too,
this
is
a
new
big
runs
on
solaris,
not
just
me
before
this
run
swarm
failure
on
a
boon
to
I
have
asked
sir
j
to
update
and
see
if
it
gets
any
better
error
zone.
This
was
the
ticket
that
spurred
me
to
read
better
error
reporting
and
during
the
swarming
process.
So
we're
going
to
put
this
one
to
the
next
friend.
A
Hopefully,
you
can
follow
up
and
see
what
type
of
response
to
gets
from
running
the
same
code
now
replace
acronyms
be
more
descriptive.
So
this
is
a
discussion.
That's
going
on
internally,
as
well
as
on
the
new
pic
theory
list,
everyone's
waiting
on
Jeff,
to
comment
about
this,
and
he
will
certain
of
it
so
I'm
just
going
to
leave
this
open
until
we
get
some
type
of
resolution
and
we
may
have
actionable
items
to
take
in
the
code
for
this
or
we
may
not.
A
The
problem
with
this
issue
is
even
if
we
decide
we
have
names
we
want
to
use
and
we
want
to
change
code
in
the
OPF
last
names
like
especially
that's
a
backwards,
incompatible
change
because
of
the
way
our
model
serialization
works.
So
we
might
make
a
decision
about
what
we
want
to
name
certain
things,
but
we
may
not
be
able
to
take
action
until
we
get
to
release
one
point
I'll
release,
but
I
will
be
tracking
this.
A
A
Exposing
the
new
anomaly
likelihood
code
so,
like
I
mentioned
the
nominee
likelihood
code
that
suit
I
introduced
at
our
last
hackathon.
That
we
use
in
Brock
is
now
within
the
OPF.
However,
it
is
kind
of
an
add-on
functionality
to
the
OPF
it's
not
baked
in
to
a
seal
a
model.
So
I
think
what
Mark
was
asking
about
was
incorporating
it
into
the
OPF,
so
you
don't
have
to
do
any
post
processing
of
the
anomaly
scores.
Zubat
and
Scott
have
talked
about
this
in
the
office
about
baking
it
into
a
metric
in
the
metric
manager.
A
Oh
yeah:
well,
there's
an
open
PR
about
this
and
I'm
just
waiting
on
him
to
do
some
work
to
clean
up
that
PR,
so
it
passes,
I
needs
it
needs
to
be
merged.
It
needs
emerged
from
master
and
just
to
track
this
in
case
the
pippin
soluble
thing
does
not
get
merged
and
quickly
I
opened
an
issue
to
make
it
easy
to
make
python
changes
without
recompiling
everything.
A
This
is
up
for
grabs.
If
anybody
wants
to
work
on
this,
everybody
seems
to
want
it.
Nobody
seems
to
want
to
work
on
it,
so
maybe
we're
just
all
waiting
forward
to
install
which
is
fine,
okay,
but
I'm,
going
to
move
this
and
I
think
I
missed
moving
one
of
these
issues,
so
we're
within
our
moment
this
this
tool
that
I'm
using
there
is
kind
of
new
yeah
cleanup
coding
unit
tests
that
needs
to
be
moved,
spread
to
14,
and
then
we
have
a
couple
tooling
issues.
A
Providing
a
status
for
non-emergent
er
s-
this
would
be
nice
because
currently,
if
you
have
a
PR
in
the
list
of
PRS,
it
shows
up
as
good
to
merge,
even
though
when
you
go
to
it,
the
merge
button
is
not
activated
because
your
merge
conflicts
so
I
think
this
is
to
provide
a
failing
status
if
the
PR
is
not
mirja
below
for
reporting
purposes.
Utensil
was
going
to
do
this,
so
more
than
happy
to
have
him.
Do
it.
So
there's
that
and
then
yeah.
A
A
Ok,
wallboard
really
really
quickly.
All
over
this
perfectly
I
know,
Jeff
full
has
been
really
busy,
so
you
can't
do
it.
One
thing
I
noticed
or
I
told
you
was.
We
have
a
regression
test
fold
test
repository
now
called
neuberger
progression
and
it
has
its
own
built-in
Travis,
and
here
it
is
right
here.
A
So
you
look
at
the
build
history.
It
builds
every
time,
there's
a
push
to
new
pic
master,
so
the
past
five
and
then
running
I
would
love
to
have
a
some
area
on
the
wall
board
to
report
this
regression
test
process
on
the
last
ten
built
or
something
at
the
gresham
test
screen
or
read.
Something
like
that.
It's
nice
to
have
I
might
get
to
it.
I
might
not,
but
I'm
going
to
throw
it
in
and
I'm
not
going
to
sign
it.
A
Anyone
if
you're
a
web
developer
and
you're
watching
a
dog
involved
would
be
a
great
way
to
get
involved.
The
wallboard
is
all
JavaScript
HTML,
so
I
create
a
new
page
to
display
all
issues.
That's
what
we're
looking
at
here,
it's
just
very
ugly,
so
the
dish
you
here
is
really
to
kind
of
make
it
look
better
because
it's
it
looks
ugly,
so
I'm
going
to
move
that
over
just
currently
signed,
but
you
guys
may
never
get
to
it.
So
that's
on
a
sign
up
there
right.
Anybody
can
jump
on
that.
I
might!
A
If
I
have
time,
but
I
doubt
it
and
then
break
the
dev
tab
into
its
own
page,
if
you
look
at
the
wall
board
here,
you'll
see
there
is
a
dev
tab
right
here.
I'd
like
this
to
be
its
own
page,
instead
of
a
tab
in
here,
because
it
makes
a
lot
of
HTTP
calls
and
it
reloads
every
60
seconds
or
so
so
there's
that-
and
once
again
I
forgot
to
move
these
this
guy
and
Sprint's
14,
and
that
guys
did
those
two
good,
okay,
Darwin
64
in
line
up
64.
A
These
three
issues
have
been
hanging
around
I've,
just
been
bumping
them
just
from
sprint
to
sprint,
nobody's
really
fussing
about
them
at
all,
but
they
are
things
that
do
need
to
be
done.
I'm
just
going
to
leave
them
I
think
are
all
assigned
to
Austin
currently,
but
nobody
is
crying
about
them,
so
I'm
just
going
to
keep
bumping
them
from
sports
bra.
Yes,
just
go
at
it,
so
spry
24,
verse,
14
and
spread
24
all
right.
We
should
have
p
nothing
left
in
spread
23.
A
A
So
there's
this
this
and
this.
These
are
all
new
tickets
that
popped
up
over
the
past
few
days.
This
is
a
build
failure
that
was
reported
on
the
mailing
list.
I've
got
a
link
to
the
original
message
and
the
log
file.
I
asked
the
the
person
who
created
this
Steve
to
add
some
details
about
his
operating
system
and
stuff
like
that
florin
early
work
on
it,
but
there's
a
look
seems
like
a
legitimate
failure.
A
They're
implementing
a
geospatial,
encoder
Chayton
I
believe
is
going
to
give
some
more
information
to
the
mailing
list
about
this
soon
so
stay
tuned,
and
then
David
created
this
ticket
to
rename
all
the
Python
files
to
follow
the
naming
conventions.
We
have
our
wiki,
so
you
took
a
lot
of
time
to
find
all
of
the
files
that
need
to
be
renamed.
So
this
is
going
to
be
a
big
task
or
a
lot
of
changes
anyway
for
this
PR.
So
that
is
on
hook
for
the
coming
sprint
other
than
that
in
New
picure
we
need
to.
A
A
So
that
is
where
we're
at
I,
don't
think
I
think
there's
anything
else.
We
need
to
talk
about
a
couple
things:
I
guess
if,
if
you're
involved
in
the
season
of
new
pic
make
sure
that
your
you're
making
some
progress
in
contact
with
the
mentor
or
if
your
mentor
get
in
contact
with
your
student
or
group,
one
of
the
season
of
music
projects,
the
epilepsy
seizure
detection
has
turned
into
a
cattle
competition.
So
there's
actually
a
team
of
people
involved
in
that
now.
A
So
you
can
get
details
on
that.
Our
mailing
list,
as
well
other
than
that
nothing
is
really
pressing
going
on
work.
As
usual,
there's
still
a
lot
of
good
stuff
going
on
a
lot
of
good
work
on
plates.
So
let's
get
it
done
and
we
can
move
on
it's
a
regression
test,
its
some
more
testing,
c++
testing
in
place
and
still
moving
towards
a
1
point.
0
release
at
some
point
of
future.
Thanks
for
watching
and
I'll
talk
to
you
on
the
mailing
lists.