►
From YouTube: CHAOSS Sept 25, 2018 Weekly Hangout
Description
CHAOSS Sept 25, 2018 Weekly Hangout
A
B
Yeah
I
was
oh,
hey,
Jeff,
Kevin
I'm
on
the
call
cool
I
was
just
sending
an
email,
Kevin
I
think
we
can
probably
go
ahead
and
create
the
the
webpage
for
the
conference
and
just
leave
things
like
the
location,
TBD
and
I'm.
Just
gonna
send
you
basically
the
initial
initial
content
to
spin
it
up,
and
then
the
rest
of
us
can
can
edit
it
as
needed.
If
that
sounds
ok
to
everybody
for.
B
C
A
B
B
D
B
B
B
C
A
B
B
B
A
C
There's
one
45
to
50
minute
workshop
at
the
main
conference,
but
I
think
you're
talking
about
a
larger,
more
open
tutorial
the
day
before.
Okay,
that
was
mentioned,
vaguely
I've
made
my
travel
arrangements
to
account
for
it
and
the
last
message
I
got
was
more
details
forthcoming.
So
okay
well
wait
for
them
to
fourth
cotton,
since
that's
that's
December.
C
A
So
but
I'm
thinking,
even
with
respect
to
the
risk
component,
you
know
so
like
last
week
we
started
talking
about
potential
integration
of
some
of
the
scanners
that
are
info,
salah
gee
yeah
and
maybe
I
can
work
with
Matt
here
to
start
getting
that
moving
forward.
So
even
if
they're
kind
of
rough
prototypes
as
to
what
they
look
like
I
actually
saw.
C
There's
a
full
request
that
was
recently
made
that
yeah
there
was
a
reason,
pull
request
made
where
somebody's
written
a
shell
script
with
all
kinds
of
regular
expressions
in
it.
That
goes
through.
Looking
for
licensing
information,
did
you
comment
at
one
point:
okay,
hey
Susan
I
commented
on
it
today,
okay,
but
I
asked
him
to
maybe
put
a
whole
documentation
when
there's
like
a
few
comments
in
his
code,
so
we
knew
what
was
happening
but
I.
A
C
Like
with
facade
I,
am
you
know,
we're
gonna
go
I.
The
team
had
a
few
concerns
about
the
native
compiling
things
that
are
in
it,
but
in
terms
of
distributing
the
code
and
I
talked
with
Brian
Warner
at
Samsung
about
it
he's
a
guy
that
wrote
most
and
there's
like
four
years
of
debugging,
yeah
anomaly
and
weird
weird
and
get
repository
stuff.
That's
in
there.
So
we're
just
gonna
boil
that
in
as
a
component
of
auger
and
that's
what
this
will
be
yeah
and
exactly
the
same
thing
you
know,
and
it's
like
his
license.
C
A
C
C
The
data
on
the
back
end
using
Brian's
work.
Okay,
it's
we're
working
to
present
it
in
augur
so
that
because
it
gives
people
something
they
want
to
see
and
all
the
things
were
doing.
I
talked
with
Brian
for
like
an
hour
last
week
and
all
the
things
we're
doing
with
highly
visualized
graphs
high,
but
high
visual
high
graphs
downloadable
data.
All
those
are
all
things
he
wants
in
the
facade.
Doesn't
it
doesn't
I'm
time
to
build
so
and
then
gotten
similar
feedback
from
Jonas
Jonas
yeah
at
VMware?
C
A
A
All
right,
well
I'll,
work
with
Matt,
we'll
probably
just
Florrick
augur
as
it
currently
exists
for
the
time
being
and
just
will
kind
of
work
independently
for
a
while,
okay
and
just
kind
of
show
you
what
we
come
up
with
them,
then
we
can
kind
of
systematically
make
recommendations
back
into
them
main
line.
Yeah
I'd
like.
A
Format,
I
think
is
gonna
yeah,
okay,
I
mean
I,
have
some
ideas
on
that.
It's
gonna
be
really
so
things
like
face
ology
and
even
getting
it
into
say,
an
SP
DX
format.
If
you
want
to
display
it
in
that
way,
there's
some
really
like
devil
in
the
details,
stuff
and
I.
Think
we
really
just
start
out
with
some
pretty
high
level
high
level
descriptions.
It's
things
like
the
number
of
files
that
have
declared
licences.
A
C
A
Yeah,
we
definitely
will
that
will
on
the
fork.
Okay.
So
when
he's
ready
for
to
talk
to
us,
we're
keen
to
talk
to
him
to
make
sure
that
rolls
in
right,
okay,
cuz
that'd
be
cool.
If
you
could
take
that
and
if
we
could
get
I'll
kind
of
follow
along
with
what
we're
more
lab
is
doing.
But
if
you
could
take
some
of
those
tools
that
have
some
initial
risk,
bootstrap
stuff,
when
you
go
to
yeah
I,
think
that
should
be
a.
C
C
The
augur
repository
moved
over
to
the
Kaos
organization.
Finally
I
slept
well
that
night,
oh
yeah,
we
were,
you
know
it
was
like
every
time
we
talked
about
doing
it.
We
sit
around.
We
go
okay.
What
are
the
things?
We're
gonna
break
one
dating
this.
What
we
got
it
down
to
a
point
where
we,
the
only
thing
we
broke,
was
Travis
CI,
and
that
was
an
easy
thing
to
fix.
So
you
know
once
our
anxiety
about
what
happens
when
you
do
an
organization
with
a
repo
was
subsided.
C
C
C
We
do
have,
we
do
have
some
of
that,
but
a
lot
of
our
documentation
is
now
hosted
in
the
app
so
we've
minimized.
Much
of
that,
but
there's
there's
still
some
things
that
are
there's
one
page
particular
I
know
it's
gonna
break
the
next
time
we
restart
our
docker
containers,
but
it's
a
page
I,
don't
think
most
people
look
at
so.
Okay.
A
A
Some
reports
might
be
bigger,
so
might
be
smaller,
but
it's
the
the
the
items
that
these
these
workgroups
find
to
be
important
and
kind
of
the
methods
behind
which
you
would
collect
that
data
which
I,
ok,
so
I've
been
totally
I.
Just
love
the
idea,
because
I
think
it's
something
that's
highly
highly
tractable.
A
So
then
I
was
thinking.
Is
there
a
way
as
these
move
forward,
which
I
think
they
are?
Is
there
a
way
for
a
community
to
run
these
reports
across
two
years
and
see
improvements?
So
what
I
mean
by
that
is
for
all
of
us
that
are
in
academia
right.
We
have
all
of
these
accreditation
bodies,
so
we
either
have
AACSB
or
we
have
abet
accreditation,
and
there
are
ways
that
you
can
understand
your
program
so
perhaps
understand
a
college,
but
one
of
the
things
that
that
abet
does
is.
A
They
also
provide
a
mechanism
by
not
only
which
to
produce
the
reports
but
which
ways
that
you
can
hold
the
reports
against
each
other
year
after
year
to
see
areas
of
improvement
for
your
own
project,
because
I
mean
we
always
have
this
problem
right
that
there's
no
best.
You
know
it's
harder.
It's
hard
for
this.
Is
that
red,
yellow
green
kind
of
thing
it's
hard
for
me
to
say
what
that
best
might
be
yeah
and
there's
people
who
want.
A
Yeah
so
I,
just
maybe
maybe
one
of
the
things
again
just
I'm
thinking
out
loud.
This
is
that
weekly
meeting
that
maybe
we
should
also
think
about
ways
by
which
you
can
run
the
report
a
yearly
basis,
or
you
know,
by
yearly
basis
and
be
able
to
compare
those
results
across
time
on
your
own
community.
So
you
see
improvement
in
DNI,
or
you
see
improvement
in
in
value
downstream
value
where
you
see
improvement
and
then
on
risk
whatever
that
might.
C
Mean
so
I
think
I
think
the
way
to
do
that
is
improvement
against
some
goal
that
you
said
mmm-hmm,
so
not
not
improvement
I
mean
you
define
what
improvement
yes
and
that's
how
a
bet
is.
You
can
find
anything
that
you
want
ya
know
what
I'm
saying
like
right
now:
yeah
it's
a
good
idea:
you're
more
lab!
Doesn't
let
you
define
that
neither
is
auger,
but
those
are
features
that
any
performance
management
system.
That's
any
good,
in
other
words,
no
performance
management
system
that
exists
today
should
do.
C
D
B
A
A
D
D
What
I
was
thinking
out
loud
is
that
even
without
gold,
comparing
across
time
is
something
I
think
naturally
will
emerge.
So
once
this
might
have
thoughts,
is
that
once
a
community
has
a
diversity
inclusion
report
or
it
doesn't
have
to
be
confined
to
only
DNI
metrics,
that
same
process
should
apply
to
all
of
our
metrics.
It's
just.
You
need
different
ways
of
collecting
the
data,
but
the
general
process
will
be
the
same,
and
once
you
have
a
document
outlining
okay,
this
is
what
we
did.
These
are
the
results,
then
doing
it
again.
D
Two
years
later,
shouldn't
be
that
difficult
there
might
already
have
established
a
repository
or
some
knowledge
base
for
how
they
did
it
initially
and
then,
including
the
previous
results
in
the
analysis.
So
one
of
the
things
that
we
have
is
reflection
part
or
that's,
something
that
I
would
strongly
advocate
for
a
word
becoming
to
look
at
okay.
So
this
is
what
we
see
is
what
we
saw.
What
we
see
this
is
what
it
means
and
to
arrive
at
meaningful
insights.
They're
looking
at
the
past,
I
think
will
naturally
emerge
now.
D
There's
a
second
component
to
that,
and
that
is:
can
we
collect
these
reports
and
aggregate
them
and
provide
comparison
data
and
by
doing
that
as
a
service,
the
chaos
project
provides.
It
produces
a
value
because
you
can
say
I
want
to
compare
myself
to
similar
projects
and
get
those
aggregated
and
because
that
is
valuable
to
the
projects.
We
will
get
the
numbers
to
do
that.
D
A
C
That's
a
really
great
question,
like
that's
a
use
case
that
probably
needs
a
short
description
and
to
be
covered
by
the
software
that
we're
writing.
You
know
how
do
we,
because
you
can
always
you
know
in
theory,
you
can
always
rerun
the
report,
because
the
data
quote-unquote
is
always
there,
but
if
you
run
a
report
off
of
something
it
might
be
good
to
just
create
a
separate
copy
of
the
data
that
generated
that
report.
C
For
that
reason,
so
I
think
if
there's
like
there's
this
notion
of
a
regular
report
like
your
TPS
reports
that
somebody's
gonna
want
and
you
want
to
save
I,
don't
think
you
just
I,
don't
think
you
want
to
save
a
PDF
of
the
data.
I
think
you
want
to
save
either
like
it
as
PDX
file
with
a
sort
of
stated
structure
or
just
flat-out,
JSON
or
CSP.
Yes,.
A
C
By
or
so
and
and
if
those
reports
are
documents
like
would
we
think
like
and
I
think
about
something
I
think
you
visit?
Is
it
or
is
it?
A
diamond
I
was
considering
the
use
case
of
preserving
data
that
was
used
to
generate
a
report
so
that
you
could
run
comparisons
using
something
that's
one
use
case,
but
the
second
use
case
that
sounds
like
is
tough
but
archiving
these
documents.
You
know
the
generated
documents
that
you
share
with
people.
Oh.
D
C
Leaning
towards
the
first
I,
don't
think
you
know
me
from
a
document
perspective.
I
might
argue
that
most
organizations
and
most
people
have
a
way
they
do
that
like.
If,
if
you
want
to
see
my
NSF
reports
for
every
grant,
I've
ever
had
I
have
them.
You
know,
but
organizationally
I
don't
have
us.
I
am
the
organization
I
yeah.
C
D
A
A
C
A
C
Used
Amazon's
glacier
for
things
that
I
don't
need
to
access.
Very
often,
it's
very
basically
a
slow,
retrieval,
limited
throughput
storage,
but
it's
about
ten,
a
tenth,
the
cost
of
what
you
paid
for
something
like
Dropbox
or
other
shared
services.
So
it's
slower,
but
you
don't
need
this
kind
of
thing
to
be
fast
and
it
might
be
something
where,
because
the
cost
is
low
and
controllable.
C
Maybe
we
can
get
a
group
of
companies
to
sponsor
this
kind
of
like
long-term
storage
as
a
separate
as
a
separate
Enterprise
and
separate
only
in
a
sense
that
people,
if
if
we
give
them
this
long-term
storage,
we
want
to
also
be
able
to
give
them
some
trust
that
it'll
be
there
in
vineyards.
So
if
we
work
out
some
kind
of
a
deal
at
Amazon
where
they
we
can
do
that,
I
think
everybody.
Let
me
give
this
a
chance
to
do
a
lot
of
things
and
help
a
lot
of
people.
It's
cool.
C
A
A
A
A
D
C
So
Ben
had
suggested
that
as
an
easier
way
to
merge
and
I
went
along
with
it
and
after
looking
at
what
we
were
doing
in
the
fork,
I
mean.
The
main
reason
is
the
opposed
to
one.
It
makes
it
a
lot
easier
for
us
to
update
from
what's
happening
in
the
metrics
repository
that
pull
request
to
a
lion
is
a
little
bit
more
straightforward.
C
We
were
somewhat
ambivalent
in
our
last
call
about
whether
ultimately,
it
was
a
fork
or
a
branch,
but
Ben
had
made
some
arguments
previously
that
a
branch
is
much
easier
to
work
with
and
understand,
and
so
that's
what
we
want
think
if
you
think
of
it
from
the
point
of
view
of
a
new
person
coming
onto
the
project,
the
that
there's
a
relationship
between
the
metrics
repository
and
the
work
group.
Gmd
repository
isn't
immediately
clear.
C
You
know,
there's
no
there's
no
natural
place,
that's
gonna
make
them
look
there
because
they
don't
really
know
what
that
is,
but
they
know
what
metrics
is
and
we
talk
about
metrics
as
the
main
repository
so
I
think
I
mean
these
are
the
these
are
the
things
that
were
discussed
in
the
last
meeting
and
I.
Think
Kevin
hasn't
already
he's
sending
out
the
notes
from
that
meeting
shortly,
and
that
was
one
of
the
topics
covered
we're
meeting
again
tomorrow,
oh
yeah,
so
what
was
it?
D
Here
the
rationale
I
heard
it
the
path
in
the
passing,
but
it
wasn't
quite
sure
what
the
main
reason
was.
I
know
that
the
alignment
was
an
issue
and
while
yes,
the
relationship
to
the
metrics
repository
is
less
obvious
in
its
own
repository.
Managing
a
branch
and
managing
of
Fork
with
regards
to
upstream
is
the
same.
C
I
think,
like
I
said
we
were
pretty.
It
was
pretty
much
a
toss-up
on
our
last
call
about
whether
to
go
with
a
fork
or
a
branch,
and
it
was
the
and
I
agree
upstream.
It
doesn't
matter,
I
think
it's
the
fact
that
we
had
gone
so
far
afield
from
what
was
in
metrics
and
we
needed
to
do
a
massive
realignment.
D
So
time
will
tell
another
observation
might
be
that
github
standard
behavior
for
pull
requests
is
to
point
it
at
the
master
branch.
And
so,
if
you
get
contributions
for
growth
much
in
decline-
and
it's
pointed
to
the
master
branch
not
to
your
growth
mission
decline,
branch,
I,
don't
know
if
that's
gonna
be
an
issue,
it's
just
something.
D
C
C
It's
it's,
not
the
easiest
part
of
their
tool,
the
pull
requests
if
you're,
not
going
master
and
master
office
straight
up,
Fork
even
going
off
of
a
branch,
is
a
little
tricky
and
you've
got
to
make
sure
it's
pointed
in
the
right
direction
and
if
you
click
the
wrong
thing,
it'll
take
you
up
up
to
the
root
directory
of
the
four
of
the
not
Fork
of
what's
been
for
it.
So
I
agree,
there's
a
there's
a
lot
of
issues
there
anyway,.
D
C
And
I
guess:
I'm,
not
apparently
I'm
not
kidding.
I,
know
I
subscribe
to
the
chaos
the
way
I'm,
the
the
new
mailing
list-
I
guess
he's
here,
so
he
Sue's
did
respond
and
he
wasn't
sure
the
branch
or
she
was
decided.
So
nothing
is
undoable,
so
that
might
be
coming
back
up.
I
didn't
see
that
we'd
sent
the
notes
out
already
or
the
reply
which
makes
you
wonder
what
stuff
with
my
mail
actually.