►
From YouTube: CHAOSS.Value.April.12.2019
Description
CHAOSS.Value.April.12.2019
B
B
A
B
What
structure
over?
Let's,
let's
do
it?
Okay,
so
we
did.
We
did
talk
about
this
on
on
Tuesday
a
little
bit.
The
idea
of
you
know
kind
of
being
systematic
about
and
figure
out
what
to
target
hello
man,
I
being
systematic
about
trying
to
figure
out
what
type
of
metrics
to
target
and
I
went.
You
know,
based
on
that
I
went
back
and
I
looked
at
the
focus
areas
that
you
put
in
our
issue:
tracker,
Georg
and
there's
there's
five
of
them.
B
C
Approach,
yeah
can
I
ask
a
question
about
the
approaches,
make
sure
I
understand
it.
Yes
is
the
idea,
then
that
so
is
the
idea
we
would
like.
We
define
a
metric
and
then
the
is
it.
The
immediate
step,
then,
is
that
we
work
on
the
tool
and
for
that
metric
that
that's
a
constituent
part
of
the
working
group,
so
we're
not
going
to
talk
about
30
metrics
to
find
them
all
and
then
have
an
implementation,
but
I
I
hear
you
describing
more
of
a
working
piece
of
software
as
opposed
to
code.
C
B
I
guess
you
know,
there's
two
there's
a
couple
of
different
ways.
We
can
think
about
our
work
here
and
one
judgment
either
way.
I
just
wanted
to
make
sure
I
understood
right,
so
that
I
think
there's
a
couple
different
ways.
We
can
think
about
our
work.
One
way
is,
we
could
say:
hey
our
mission
is
to
define
specifications
and
requirements
that
somebody
else
will
implement
mm-hmm.
B
B
B
D
C
I
mean
one
of
the
things
that
drives
me
to
work
on
auger.
Is
that
it's
hard
to
really
know
what
these
metrics
mean
and
whether
or
not
and
how
they're
useful?
If
I
don't
have
a
system
that
shows
me
what
that
metric
looks
like
and
I
can
show
it
to
people
and,
and
then
I
can
see,
who
consumes
which
metric
so
I
mean
it
is
a
software
developer
by
background
that
it
totally
makes
these?
It
makes
it
easier
for
me
to
think
about
what
we're
doing
right.
B
B
One
way
is,
you
know
we
could
like
be.
We
could
ever
like
real
engineering
purity
and
we
can
think
through.
You
know
a
bunch
of
different
alternatives
and
come
up
with
a
way
that
we,
we
think
is
you
know
like
ultimate
best.
You
know,
might
almost
call
that,
like
German
engineering
approach
and
then
there's
another
approach
where
we
could
just
say,
let's
just
take
what
we
got.
You
know
bag
a
bucket
of
parts
and
just
trying
to
just
try
and
assemble
it.
B
You
know
in
in
as
fast
a
way
as
possible
and
I,
like
the
latter
approach,
I'd
like
to
just
look
at
what
we've
got
and
figure
out.
What
is
the
simplest
thing
we
can
do
with
the
tooling
that
we
have,
let's
just
get
that
working
end
to
end
in
a
way,
that's
repeatable
in
a
way
that
you
know
just
stands
up
and
then,
let's
start,
you
know,
let's
start
running
metrics
through
it
and
just
expand
on
that
little
by
little.
B
C
B
B
Oh,
that's,
an
okay
starting
point,
I'd,
really
love
to
just
start
there
and
ask
what
is
the
simplest
thing
we
can
do
so
that,
like
there's,
maybe
a
usable
utility
or
a
website
or
just
whatever,
so
that
we
can
start
collecting.
You
know
real
value,
metrics,
simplest,
possible
thing
and
and
just
build
on
that.
Are
you
proposing,
like
a
sample
context.
D
B
You
know
we
could.
Okay,
if
you
look
at
the
focus
areas,
you
know
the
one
that
kind
of
jumps
out
at
me
is
labor
cost
as
a
value
as
a
focus
area
for
for
value
and
labor
cost
can
be
expressed
in
you
know
so
many
ways,
but
I
think
the
simplest
way
to
express
labor
cost
is
just
you
know,
a
number
of
commits,
and
you
know
a
number
of
commits
per
person,
and
you
could
almost
say
that
a
commit
yeah.
B
We
could
almost
think
of
a
commit
as
like
a
form
of
currency,
and
just
you
know
just
let
that
be.
You
know
the
simplest
you
know
without
in
the
exchange
into
you,
know,
dollars
or
or
euro
or
yen,
or
anything
like
that.
We
could
just
say.
Oh,
the
yen
is
look
that
the
commit
is
like
the
simplest
form
of
currency.
Let's
just
start
tracking
that
you
know
a
number
of
commits
and
then
maybe
number
of
commits
by
person
and
number
of
commits
by
organization.
D
D
At
the
same
time,
I
think
we
should
create
a
pull
request
that
would
add
labor
investment
as
a
focused
area
and
then
build
out
that
focus
area
a
little
bit
with
the
goal
of
that
focus
area
and
the
questions.
And
then,
when
you're
talking
about
these
specific
metrics,
what
questions
those
metrics
are
answering
as
well.
So.
B
I
think
that'd
be
awesome.
You
know,
we've
got
that
five
areas,
I
like
the
idea
of
picking
you
know
just
taking
a
selection
of
one
of
these
and
see
and
saying
what
can
we
do
with
this
area?
And
you
know
let
that
be
sort
of
a
template.
Let's,
let's
try
and
implement
this
in
software.
Let's,
let's
sort
of
you
know
pick
something
that
gives
us
the
biggest
bang
for
the
buck.
Mm-Hmm.
E
B
D
Know
I,
agree
and
I
think
focusing
on
one
focus
area
losing
focus
twice
in
a
sentence,
but
I
think
it's
a
good
idea
too,
especially
if
we're
starting
to
think
about
releasing
the
metrics.
You
know
as
we
not
just
for
this
working
group,
but
as
a
chaos'
project
in
its
entirety.
You
know
we're
gonna
be
releasing
metrics
that
come
from
the
focus
or
from
working
groups,
and
so
really,
at
least
in
this
first
release.
Having
the
value
working
group
really
even
just
have
one
focus
area
that
might
have
a
few
metrics.
A
To
build
on
that
idea
and
say
that,
right
now,
when
you
go
into
the
issues
there
is
a
Google
Doc
associated.
So
what
we
could
do
today
is
decide
on
one
focus
area
to
drill
down
into
and
then
collaboratively
work
on
the
Google
Doc
and
then,
after
the
meeting
we
can
create
the
poll
estimate
the
content
be
created.
Today's
good.
D
B
Okay,
so
I
I
have
some
suggestions
of
a
proposal.
Some
things
that
we
could
affect
commit
count
is
one.
Maybe
a
pull
request.
Count
is
two
and
then
maybe
the
third
one
is
tracking
issues.
Issue
closed
rate.
You
know
pick
here
pick
your
favorite
pick
your
favorite
sort
of
metric
around
issues
and
I.
Think
in
my
take
that'd
be
enough
to
get
started
with
I
mean
there's
a
hell
of
a
lot
we
could
do
with
that,
and
the
data
is.
C
Never
already
I've
already
run
lines
of
code
and
labor
investment
is
just
on
a
bunch
of
stuff
with
labor
estimation,
so
using
this
sec
tool.
That
I
think
I
mentioned
on
a
previous
call.
Suck
integrating
that
up
put
into
augur
would
be
that's
just
an
integration
effort
right.
We
don't
have
to
build
anything
that
doesn't
exist.
We
just
have
to
use
it
to
develop
metrics
and
I.
Think
then
we're
going
to
get
into
the
discussion
about
how
to
parameterize
labor
investment
right,
because
there.
F
F
E
A
I
have
a
thought
on
that.
Well,
we
already
identified
in
case
that
we
will
have
overlapping
use
of
metrics.
The
value
working
group
is
putting
a
different
focus
on
these
than
the
growth,
maturity
and
decline,
or
not
sorry,
evolution
working
group
because
we
are
focusing
on
the
organization
and
what
value
it
provides
to
the
organization
or
what
value
they
put
in
to
talk
about
labour
investment.
So
the
core
of
the
definition
might
be
the
same,
but
we
are
adding
a
different
focus
here.
E
D
D
A
D
B
B
You
know
something
like
a
commit
or
a
pull
request
as
a
as
a
unit
of
currency,
and
you
could
use
that
to
you
know,
do
comparisons
across
organizations
so
be
interesting.
Let's
say
if
you're
running
an
open
source
program
office
to
compare
your
level
of
activity
to
your
peers
and
then
the
only
thing
you
could
do
once
you
have
that
level
of
activity.
B
Potentially
you
could
you
could
plug
in
your
own
valuation
numbers,
for
example,
you
might
I'm
sure
we
are
going
to
be
able
to
find
studies
that
that
show.
You
know
how
much
time
it
takes
to
write
a
line
of
code
for
example,
or
how
much
time
it
takes
to
you
know,
do
a
commit
on
an
average,
and
you
know,
based
on
that,
you
can
back
out
to
an
hourly
rate,
and
then
you
know
you
can
you
can?
You
know,
generate
a
total
economic
value.
B
You
know
just
just
based
on
you
know,
paying
like
an
hourly
consulting
rate
and
different
organizations
are
going
to
have
their
own
numbers.
You
know.
So
if
you
go
into
Facebook,
if
you
go
into
you
know,
Google
or
Comcast's,
or
one
of
these
companies.
Of
course,
you
know,
they've
got
they've
got
their
salary
data,
you
know
they.
They
know
how
what
their
aggregate
salaries
are.
They
know
how
many
lines
per
code.
B
You
know
lines
per
coder
are
produced
and
you
know
they
can
come
up
with
their
own
numbers,
which
might
or
might
not
agree
with
academic
studies.
But
just
having
to
me,
having
like
the
baseline
of
the
number
of
commits
the
number
of
pull
requests,
the
the
number
of
issues,
the
issue
close
rate,
that
sort
of
thing
will
allow
people
to
plug
in.
You
know
their
own
assumptions
about
how
much
each
of
those
things
are
worth
in
them,
and
then
that
would
generate
an
economic
value.
C
Because
I
know
you're
gonna,
no
no
I
was
just
listening.
What
Mandy
was
saying
and
I
I
agree
with
them
completely
I
think
we
also
have
I
mean
in
addition
to
that
commits
like
you
could
also
evaluate
just
the
total
cost
of
build
by
looking
at
the
lines
of
code.
We
could
snapshot
lines
of
code
over
time
and
apply
a
Kokomo
like
models
over
over
it.
C
We
could
also
apply
and
I
think
I
think
if
I'm
listening
correctly
and
understanding
the
same
way,
whatever
what
you're
saying
we
could
do,
labor
estimations
for
the
contributions
as
they
come
in
beginning
on
a
day.
So
not
just
what's
the
value
to
create
this
whole
project,
but
what's
the
value
being
contributed
to
it
on
a
regular
basis
and
I
can
see
some
real
arguments
for
that
actually
being
like?
C
C
B
Me
that'd
be
a
phase
to
you
know
a
phase
one
might
be
just
exporting
like
a
CSV
with
with
number
of
lines
number
of
commits,
and
then
you
could
import
that
into
your
own
spreadsheet
and
and-
and
you
know,
add
your
own
labor
costs
and
you
know,
do
the
math
sort
of
outside
of
the
software
and
then
maybe
maybe
a
phase
two
would
be.
You
know
some
sort
of
a
web
form
where
you
can
plug
in
your
your
assumptions.
Your
labor
costs
your
amount
of
time
to
close
an
issue
and
and.
D
C
I
know
we're
yeah.
This
thing
like
I,
know
how
to
do
that
already,
so
you
know
it's
just
putting
it
in
the
pipe
I've
actually
done
it.
I
just
haven't
put
it
into
augur
yet
right.
So
a
lot
of
times
what
happens
is
somebody
wants
to
see
something
I
do
it,
and
then
we
put
it
on
the
augur
list.
You
know
you
know
one.
C
Doing
with
augur
that
I
think
will
be
nice
as
its
we're
combining
right
now
we
use
facade,
gh
torrent,
yeah
I,
save
all
these
data
places
that
we
get
things
and
we're
actually
gonna,
be
we're
combining
right
now,
our
own,
basically
creating
our
own
data
model
in
Postgres
that
we're
just
gonna
feed
everything
into
and
tag
it
with
its
data
source
of
origin
and
the
tool
that
brought
it
there,
so
that
that
you
can
do
all
and
in
parameterizing
things
like
cost.
You
know
a
lot
of
times.
C
B
D
C
Yeah
in
comparisons
among
that
and
the
comparisons
Andy's
talk
about
these
are
ones
that
would
let
because
I
think
a
lot
of
people
are
making
guesses
about
what
their
labor
costs
are
and
they're,
using
certain
assumptions
and
I
think
seeing
other
people's
assumptions
will
inform
my
assumptions
over
time
and
and,
like
you
said,
we
have
to
think
they
need
it,
of
course,
a
way
to
trust
that
those
contributions
will
be
anonymous.
Otherwise,
you
know
why
would
I
give
you
my
labor
cost
estimates
if
you're
a
potential
competitor
right.
B
So
it
seems
like
we
all
agree.
I
think
so
yeah
seems
like
auger
is
gonna,
give
us
most
of
what
we
want
and
by
the
way
put
putting
it
all
into
a
Postgres
database.
Just
seems
awesome.
You
know
just
just
as
an
architecture
having
things
feed
into
Postgres.
Where
there's
a
you
know,
you
can,
you
can
tag
the
source
and,
and
you
can
extend
with
new
sources
over
time
that
just
sounds
killer.
B
D
D
D
B
C
The
I
would
say
the
just
to
make
a
note
that
the
issue
count
and
the
pull
request
count
are
not
stored
in
the
git
repo.
So
we
would
pull
those
from
whatever
issue
tracker.
The
the
team
used
and
I
guess:
I
guess,
pull
requests
ultimately
do
show
up,
as
merges
in
in
the
git
repo
I
haven't
I
should
I,
don't
know
if
they're
distinguished
from
other
kinds
of
commits
in
the
git
repo
log.
C
Facade
otter
work
is
all
straight
against
the
repo
and
that's
I
just
have
to
look
and
see
if
pull
requests
can
be
made
are
visible
in
some
form
in
the
kit
repository.
Otherwise,
we
just
have
to
know
what
they're
using
for
managing
those
issues
often
do
have.
A
diverse
issues
are
often
not
on
github,
especially
for
projects
that
have
been
around
a
while
and
had
long
had
issue
trackers,
that's
not
a
problem
to
get
them
yeah.
C
A
E
A
D
C
I
mean
I
think
well,
like
I
said
the
things
that
are
awful
to
get
repo
for
sure
or
will
be
implementable
most
quickly,
I
think
I
think
we
have
with
pull
requests
to
do
it
with
github
I.
Don't
is
we
have
that
data
and
can
get
it
and
that's
okay,
difficult
I,
think
where
we
have
issue
tracking
in
other
environments?
That's
just
something
that
we
have
to
have
people
tell
us
about
the
repository
or
we
have
to
know
it
from
somewhere.
Okay,.
D
C
D
B
D
From
that,
I
would
guess
from
what
from
that
effort
each
one
of
these
metrics
pages
for
each
one
of
these
metrics
then
gets
a
detailed
metrics
page
behind
it
mm-hm.
So
there's
another
another
document
below
here,
which
does
anybody
have
one
handy
that
they
can
show
in
the
chat
a
document
below
here,
Ramin
Georg.
Do
you
have
a
the
template,
I'm.
F
C
D
A
D
D
So
Brandi
and
Sean,
and
actually
everybody
on
here,
the
one
that
gets
shared
the
chat,
the
second
one
that
points
to
the
working
group
GMD
mm-hmm,
oh,
that
description
use
case
formula
sample
that
one
so
just
just
again
not
to
do
it
right
now,
but
just
to
put
it
on
your
radar
that
that
text
box-
you
know
it's
gonna,
be
the
pull
request.
That's
gonna
kind
of
define
these
goals.
Question
metrics
right!
Each
of
the
metrics
themselves
will
be
given
a
page
based
on
that
template.
Okay,.
D
So
then
the
idea
is,
is
so
I
mean
in
a
in
a
perfect
world
for
the
release
of
these
metrics.
As
part
of
the
version
1.0
from
chaos,
you
would
have
the
focus
area
which
in
this
case
is
called
labor
investment,
and
you
would
have
this.
That
is
the
name
and
the
question,
and
then
each
one
of
the
names
or
the
metrics
here
would
have
an
Associated
template
page.
D
D
So
that's
it!
So
again,
it's
not
to
be
done
now,
just
to
put
on
your
radar
and
these
template
pages
I'm,
guessing
that
as
Sean
as
you're.
Taking
a
look
at
deploying
the
metrics
and
auger
yeah,
the
these
template
pages
will
kind
of
come
in
super
handy,
yeah
and
they'll
kind
of
revoke
those
auto
populate
based
on
ya.
D
C
C
Tell
us
what
you
really
think
about
that
meeting:
John,
no
it'sit's!
It's
for
students,
I'm
helping
them
get
to
understand
their
semester;
projects
I'm
using
open
source
for
undergraduate
junior
computer
science,
students
and
the
whole
idea
of
them
not
having
an
assignment
handed
to
them
of
all.
The
steps
is
done
a
little
mind,
blowing
for
them
to
last
week's.
A
B
D
C
The
NSF,
if
there's
certainly
SPE,
has
things
that
look
at
organization
and
culture
and
value
I.
Think
if
we
you
know,
this
is
the
kind
of
thing
that
might
be
appealing
to
foundations
that
are
tied
to
corporate
strategies
in
some
way,
or
maybe
even
a
corporate
sponsorship
of
some
kind.
It's
because
it
could
serve
a
value
for
them,
but
but
NSF
SPE,
certainly
social
behavioral
health.
C
D
D
Why
don't
I
am
Andy?
Tear
to
your
question?
I?
Think
I
will
talk
with
my
grants,
folks
here:
they're,
not
people
who
have
money,
they're
people
that
help
find
money,
no
I'm,
just
kind
of
because
this
is
not
a
space
that
I
normally
write
into,
but
they
you
would
be
more
familiar
with
the
landscape
of
you
know
the
agencies
and
the
different
programs
that
they
have
yeah.
A
C
D
So
it
helps
you
know,
kind
of
overcome,
a
hurdle.
If
you
just
allow
everybody
to
submit
all
the
time,
it's
kind
of
streamlines
the
process
downstream,
a
little
bit
for
the
NSF,
and
so
this
one
that
Sean
sent
out.
If
you'll
see
it
has
a
letter
of
intent
on
May
15th,
and
so
then
you
submit
it
and
then
they
kind
of
give
you
a
red
light.
Green
light
with
a
full
proposal
do
looks
like
October.
C
C
C
F
B
B
B
D
D
D
Sure
is
long,
I
mean
it
as
long
as
it
retains
support
for
kind
of
this
broader
work
around
community
health.
Yes,
then,
a
hundred
percent
yeah
so
I
mean
part
of
part
of
writing.
These
grants,
too,
is
kind
of
making
it
right
making
the
appeal
to
fit
with
the
program
and
then
trying
to
fund
the
efforts
of
the
folks
that
are
included
in
the
grant,
while
keeping
everything
cohesive
and
tied
together.
B
D
B
Right
well,
that
sounds
great.
So
so
maybe
one
of
the
things
that
that
we
can
put
up
this
weekend
is
a
spreadsheet
or
a
Google
Doc
to
where
people
can
add
the
grant
opportunities
to
it,
and
then
maybe
part
of
our
maybe
part
of
our
weekly
sort
of
routine
is
going
to
be
that
we
all
sort
of
look
at
at
that
spreadsheet,
together
talk
about
which,
which
things
are
most
interesting,
we
want
to
do.
We
did
we
do
it
in
github.
Would
you
mind
that
I'm
up
for
anything?
Well,
what
would
the
format
be?
D
B
D
A
B
D
C
D
B
D
A
B
The
students
that
you
guys
mentor
it
seems
like
you
know
we
might
actually
learn
something.
I
mean
I'm.
Looking
at
these
NS
NSF
focus
areas,
and
it's
like
okay.
Well,
you
know,
there's
actually
valuable
information.
Justin
Justin
reading
through
this
for
me
mm-hmm,
knowing
knowing
what
funders
are
interested
in
is
is
is
an
interesting
thing.
There's
probably
gonna
be
networking.
You
know,
I
just
see
a
lot
of
benefits
coming
from
it,
whether
or
not
the
dollar
volumes
are
higher
good.