►
From YouTube: CHAOSS.Evolution.July.29.2020
Description
CHAOSS.Evolution.July.29.2020
A
Hit
the
record
button,
so
we've
begun:
recording
welcome
to
the
evolution
working
group
meeting
for
july
29th
2020.
I've
just
pasted
the
link
for
the
minutes
into
the
screen
cadoodle
and
I'm
now
sharing
them
for
the
purpose
of
memorialization.
On
the
video.
You
know
a
lot
of
people
have
time
to
sit
around
watching
hulu
during
the
pandemic,
but
why
not
watch
chaos?
Working
group
videos
as
well
add
us
to
your
repertoire.
A
A
That
we
had
last
time
that
that
maybe
we
wanted
to
include
it.
I
think
what
our
suggestion
was
as
a
working
group
was
that
we
would
include
a
glossary
in
the
release
of
metrics
and
when
we
took
that
suggestion
to
the
larger
chaos
group.
Following
our
last
meeting,
the
the
thought
was
well
as
we
as
we
talked
it
through
really
reviews
in
as
a
as
a
defined
thing.
A
Inside
of
our
are
the
evolution,
metrics
working
group
they're
the
only
metric
we
have
right
now.
That
is,
I
think
that
anyone
in
the
the
community
call
can
identify
is
obviously
confusing,
because
a
code
review
is
a
different
thing
than
what
we're
calling
reviews,
and
it's
also
a
common
thing,
and
neither
of
the
major
platforms
actually
calls
a
pull
request.
It
calls
what
we're
calling
a
review
review.
Github,
obviously
calls
it
a
pull
request.
A
A
So
you
know
the
discussion
that
daniel
and
I
started
to
have
on
the
community
call
and
others
who
are
here
was
that
well,
maybe
so
we
needed
to
think
about
garrett
and
whether
or
not
it's
a
significant
concern,
and
I
have
the
perspective
that
what
they
call
a
pull
request
on
garrett,
but
I've
never
used.
It
sounds
much
more
like
a
commit
that,
because
it
is
just
only
ever
one
commit,
then
what
github,
gitlab
and
other
platforms
consider
a
pull
request
which
usually
includes
more
than
one
commit.
A
So
I
don't.
I
don't
know
what
to
what
to
do.
Does
anyone
else
in
the
group
have
a
perspective
because
daniel
and
I
have
the
action
item
but
daniel's
not
here.
We
could
just
postpone
this
and
I
could
email
daniel.
B
A
No
pull
and
pull
slash,
merge
requests,
accounts
for
github
and
get
lab
and
bitbucket
which,
which
are
the
major
platforms
that
people
used.
I
was
just
on
the
asia
pacific
call
and
they
used
something
called
getty.
C
B
A
Chinese
yeah,
it's
a
chinese
exactly
what
it
is.
It's
a
chinese,
github,
gitlab
and
apparently
whatever
I
created
for
my
password,
I
didn't
save
correctly,
which
often
happens
when
I
create
a
new
password.
A
I
have
no
idea
what
that
says,
but
I
assume
it's
like
an
ok
button.
User
does
not
exist.
Okay,
sorry,
we'll
solve
that
later.
I
don't
know
off
the
top.
My
I
don't
recall
what
get
he
calls
them,
but
I
imagine
it's
in
chinese.
B
A
A
A
A
I
think
so.
A
A
There
are
there's
enough
of:
are
there?
What
are
some.
D
Of
the
other
major
projects
that
use
garrett,
I
have
to
look,
but
I
know
a
good
number
of
them
do
like
the
canonical
also
uses
gavit
yeah.
I
know
I'll
just
have
to
I'll
just
have
to
search,
but
I
know
a
good
number
of
open,
open
source
ecosystem
uses
gary
news,
garrett,
okay,
and
do
they
do
they
have
something
called
a
pull.
D
A
D
A
A
A
Yeah,
I
think
the
questions
we
have
are
largely
the
ones
that
are
documented
here,
which
which-
and
it's
really
my
questions
really.
Is
it
because
when
I
heard
them
say
it
was
a
single
request
or
a
single
commit
every
time
you
did
whatever
it
is,
garrett
calls
it
that
just
it
doesn't
sound
like
we
would
be
measuring
the
same
thing
as
a
as
a
pull
request.
A
You
know
it
sounds
like
a
fundamentally
different
workflow,
and
so,
if
we
did,
for
example,
file
counting
or
commit
counting
with
garrett,
it
seems
like
limiting
this
pull
request
or
code
review
that
they
call
it.
I
think,
possibly,
that
that,
if
that's
limited
to
only
a
single
commit,
then
that
is
a
different.
A
This
sort
of
driving
a
different
human
workflow
and
some
also
smaller
pieces.
So
in
some
respects,
that's
that's
prob.
That
may
be
a
positive
for
garrett,
because
we
know
that
very
large
pull
requests
are
not
merged
or
take
longer
to
merge
in
general
than
smaller
pull
requests.
You
know,
pull
requests
with
fewer
yeah
that
affects
your
files
and
and
from
from
a
metric
standardization
perspective.
A
I
think
one
is:
is
a
garrett
code
review
conceptually.
A
A
How
many
files
so
is
it?
Is
it
only
the
I
guess,
that's
and
then
the
second
question
is:
is
a
garrett
overview
more
like
a
commit
review
of
each
commit
commits,
and
I
guess
our
commits
collected
in
groups.
I
thought
I'd
heard.
Somebody
say
they
were.
A
Yeah
yeah,
it's
you
know
it's
it's
there's,
certainly
no.
I
mean,
I
think
I
think
the
conceptual
questions
are
where
we
were
trying
to
tease
out
if
garrett
is
a.
If,
if
this
process
on
garrett
is,
I
guess
conceptually
similar
enough
to
pull
and
merge
requests
that
comparing
this
this
code
review
process
garrett
has
with
so.
We
pulled
statistics
from
garrett
about
code
reviews
and
we
pulled
statistics
from
git
lab
and
github
repositories
are.
Do
they
really
amount
to
the
same
thing?
Are
the
comparisons
valid
or
or
is
this?
A
I
guess
a
different
kind
of
thing
and
it
might
be
just
a
it
might
just
be
that
the
question
of
how
many
it
might
just
exclude
how
many
commits
like.
Maybe
it
just
means
that
this
question
is
not
interesting,
but
everything
else
is
the.
A
A
So
maybe
the
time
to
close
and
other
considerations
like
that
are
more
remain
a
part
of
the
process
with
garrett,
and
maybe
maybe
it
doesn't,
you
know
so
I
think
getting
some
feedback
like.
Would
the
pull
request
metrics
that
we
have
to
find
be
useful
to
a
you
know
mean
anything?
What
what
would
the
would?
I
guess
would
our
existing
review.
A
Metrics,
you
mean
all
to
a
carrot
user
they
might
be,
and
then
we
have
to
consider
the
reviews
terminology
and
I
think.
A
Yeah
I
get
I
I
don't
know.
Maybe
I
mean
I
don't
know
exactly
what
to
do
with
this,
because
I
haven't
used
garrett,
so
I
think
armstrong
you're
reaching
out
and
that
sort
of
you
can
shave
that
list
of
questions
down
to
something
simpler.
But,
okay,
those
are
my
my
wonderings
and
then
I
guess
I'll
ins
I'll
put
that
in.
A
A
A
We
do
have
a
set
of
metrics
that
we're
continuing
to
advance
that
are
not,
but
we're
not
going
to
release
in
the
interim
form
until
after
this
release.
So
like
the
next
time,
we
meet
we'll
be
in
a
position
to
do
an
interim
release
again,
and
I
think
the
one
that
we
were
starting
to
work
on
last
time
was
review
acceptance
ratio
and
I
think
we
made.
A
A
A
A
A
B
A
A
A
A
D
For
the
visualization,
did
you
use
the
time
series
analysis
on
the
chart.
A
Well,
it's
it's
grouping
them
by
year,
so
it's
the
year
that
a
pull
request
so
it
only
this.
This
metric
is
only
looking
at
closed,
pull
requests
right
because
an
acceptance
isn't
decided
until
it's
closed,
and
so
it's
it's
saying
of
the
pull
requests
opened
in
these
years,
which
of
them
were
ultimately
merged
and
which
of
them
were
ultimately
not
merged
and
closed.
D
Well,
like
time,
series
analysis
is
a
model
to
understand
events
that
happens
over
a
given
period
of
time
right.
It's
like
in
machine
learning
as
well.
So
when
you
try
now
to
fix
a
fixed
time
interval
on
the
period,
so
we
cannot
see
that
we
cannot
have
additive
function
like
functionality
called
additive
model.
E
Which
now
shows
the
trend
so
right
this
this,
so
the
two
yeah.
A
This
these
two
graphs
are
more
trendy.
I
think
these
are
showing
a
seven-day
trailing
average.
A
Yes,
this
I
got
it
from
the
auger
yeah
yeah.
So
what
so,
when
you
show
when
you
anytime,
you
do
unlike
something
like
pull
requests
or
data
like
this.
If
you
do
individual
data
points
on
a
line
graph,
you
get
an
extremely
spiky
line
graph
and
so
the
way
that
this
auger
represents
the
time
series
and
tell
me
if
this
is
what
you
mean
by
time:
series
armstrong:
this
is
a
graph
that
augur
provides.
A
That
shows
a
trailing
average
of
how
many
reviews
have
been
opened
in
this
case
in
a
week,
and
then
this
is
reviews
accepted
in
a
week.
We
aren't
showing
reviews.
A
Not
there
is
not
a
not
merged
time
series,
but
I'm
not
sure
if
that
would
be
meaningful
or
not.
D
Well,
it
looks
like
it
looks
like
it
has
some
of
the
components
there.
I
just
want
to
know
like
the
time
that
we
measure,
for
example,
is
it
like
every
week
I'll
just
give
a
simple
example
like
every
monday
of
we
go
every
first
day
of
the
month
if
it
depends
like
monthly,
we
capture,
we
capture
the
events
at
that
time,
so
we
cannot
do
like
forecasting.
We
cannot
do
forecasting
is
another
yeah.
A
Yeah,
so
this
the
caption
here
could
explain
what
we're
doing.
I,
I
can
explain
what
we're
doing
we're
we're
taking
the
total
number
of
commits,
so
every
every
every
week
is
a
sunday
to
a
saturday
in
in
the
way
that
we're
calculating
it
and
we're
taking
every
so
we're
taking
those
seven
days
in
a
given
week
and
we're
totaling
up
the
number
of
commits
and
dividing
them
by
seven
and
then
providing
a
data
point
for
a
week
so
that
you,
you
can
see
the
average
per
week
of
of
well.
B
A
A
Okay,
chaos
auger
overview,
so
that
would
be
the
yeah
in
the
agar
repo.
That's
probably
right.
That
is
probably
the
average
number
of
pull
requests
we
have
in
a
week
and
we
did
have
a
spike
at
the
very
end
of
2020
a
lot
like
a
lot
of
it's
kind
of
funny.
You
see
that
trend
in
a
lot
of
projects
where
people
are
just
like
getting
stuff
sort
of
off
their
plate
before
the
end
of
the
year.
That
kind
of
end
of
year
spikiness
is
not.
A
F
A
Good,
I
don't
know
it
doesn't.
Let
me
make
it,
I
don't
understand
how
it
decides
that
you
can
assign
something
or
not,
and
so
we
were
kind
of
asked
york
for
a
grammar
lab
link
and
I'll
just
do.
A
A
The
auger
zephyr
report
is
one
of
the
things
that
we
used
to
arrive
at
this
metric.
So
I
think
I
think
this
is
ready
for
the
review
now.
I
think
it
is,
I
think
it,
but
I
do
think
we
it
would
be
nice.
A
It
would
be
nice
to
resolve
the
question
about
and
we
have
time
yes
because
we
are,
we
are
not
to
release
interim.
I
think
we
didn't
release
this
last
time,
largely
because
we
were
informed
that
we
would
be
that
kevin
would
find
me
and
harm
me
in
some
way.
If
I
made
him
deal
with
a
new
metric
as
an
interim
release
in
the
middle
of
a
regular
release.
A
To
resolve
this
question
first,
but
we
don't
need
to
so
I'll,
go
back
to
the
minutes
and
just
make
a
note
that.
A
A
A
Yeah
we
haven't
started
either
these
yet
yeah,
so
I'll,
just
I'll
go
here
and.
C
A
Like
I've
opened
a
thousand
tabs
with
this
different
user,
I
actually
call
it
pull
request.
Content
diversity.
C
B
I
can
take
issue
contributors
or
any
review
recommend
diversity.
Any
of
these
I
can.
A
Okay,
do
we
want
to
start,
but
I
think
review
comment
diversity.
Have
we
started
anything
on
that.
A
A
My
trend,
no
problem,
we
are
I'll
put
it
in
the
zoom
chat.
We're
here.
F
John,
is
there
a
github
issue
that
was
kind
of
the
driver
for
this?
I
just
wanted
to
see
some
context
around.
A
B
Perspective
of
contributors,
or
is
it
from
the
perspective
of
the
content
like
diversity
in
the
content
or
review
course
or
diversity?
In
the
comment,
reviewer.
A
The
question
arose
when
we
showed
people-
sorry
I
have
to.
Apparently
we
showed
people
this
visualization.
The
number
of
comments
on
closed
pull
requests.
So
when
we
presented
this
information
to
a
couple
of
communities
which
this
is
just
mean
comments
for
all
close
pull
requests
that
drew
out
the
question
of
well
how
many
different
people
are
commenting
on
the
pull
requests
over
a
period
of
time?
A
I
don't
know
that
we
have
a
metric
for
pull
request,
comment
rate
either
or
like
pull
request,
comment
count,
because
this
so
do.
We
have
any
metrics
that
we've
released.
That
say
anything
about
commenting.
A
Probably
oh
we've
started
developing
this
one,
so
maybe
we
should
start
here,
although
these
are
code
review
comments
which
is
this
is
this
is
not,
but
I
think
we
mean
pull
request
reviews
because
that's
yeah
yeah
this
is
this-
is
not
code
review
comment.
What
are
we
calling
it
here?
A
A
Exactly
that,
like
somebody
somehow,
when
we
developed
this,
we
actually
called
it
a
code
review
instead
of
calling
it
a
a
review
because
it
just
if
you
read
it,
if
I
guess
I
guess
this
is
what
I'm
even
not
now
it's
like.
I
don't
know
it's.
A
A
Particular
okay,
you
know
because
comments
on
a
review
are
operationalized,
but
there
is
no
like,
like
essentially
when
you
do
a
code
review
like
if
you
go
through
the
github
process
of
making
comments
on
a
pull
request,
which
would
be
in
our
definition
review.
You
are
performing
a
review,
a
code
review
and
those
comments
count
as
pull
request
comments.
A
B
B
So
so,
having
two
separate
review
doesn't
make
sense
to
me.
A
No
court,
it's
we
can't
like.
If
yeah
it's,
it's
a
review.
It's
it's
not
distinct
on
any
of
the
platforms.
A
A
That
pull
request
has
comments
and
it
is
acting
in
the
function
of
a
code
review.
But
it's
it's
still
a
pull
request
and
the
comments
are
still
pull
request
comments.
You
know
you
see
what
I'm
saying.
A
A
A
These
are
all
the
different
types
of
actions
that
can
exist
on
a
pull
request
and
there
is
not
an
action
that
says
code
review
because
code
reviews
they
can
be
requested.
They
can
be
dismissed.
A
A
A
Yeah
yeah,
so
I
think
review
comments.
We
should
probably
do
first
before
we
do
review
comment
diversity.
D
A
D
A
A
Truly
yeah
yeah.
Do
you
have
a
some
familiarity
with
jira?
Are
there
is
the?
Is
a
code
review?
Yes,
it's
it's
a
it's
a
good
review
platform.
D
So
is
that
different
than
they
have
like
pull
requests
terminology
that
they
have
most
of
those
things,
but
it's
they
have
other
functionalities
like
social
network,
social
coding
system
on
it,
yeah,
sometimes
some
certain
things
that
could
be
discussed
on
the
mailing
list.
They
can
have
this
kind
of
iteration
of
discussion
as
well.
It
has.
It
also
helps
to
keep
track
of
cross
cross
project
concern,
so
you
can
use
to
like,
for
example,
if
you
are
doing
a
review
on
a
project
that
can
affect
another
project.
D
You
know
like
this.
Crossbar
system
has
been
in
the
in
the
literature
for
a
very
long
time
when,
like
in
the
linux
kernel,
the
debian
can
be
doing
something,
and
why
already
has
the
fix
for
that,
but
these
people
are
working
independently.
No
one
knows
that
this
other
person
already
has
a
fees.
D
A
So
do
you,
I
haven't
used
gyro
in
a
very
long
time,
and
I
don't
remember
like
it's
probably.
A
Yeah
I'd
have
to
use,
I
mean
like
back
a
long
time
ago
I
used
gyra
and
bugzilla
and
those
tools,
but
once
github
came
along,
I
like
a
lot
of
people.
A
Kind
of
migrated
to
its
combination
of
tools
into
one
place
and
ease
of
use
is,
I
assume
gyro
is
a
get
tool
now,
like
the
repository.
A
A
D
Bye,
I
remember
by
2017
I
did
a
studies
that
I'm
studying
how
oh
wow
this
is
way
different
than.
A
D
D
A
A
Okay,
how
do
I
the
I
used
to
be-
I
used
to
be
able
to
get
to
a
list
of
chair
repost
for
apache,
but
now
I
don't
know
where
they
are
on
the
apache
website.
A
A
D
A
A
A
A
So
I
guess
that's
a
question
for
the
next
time
and
we're
out
of
time
and
I've
done
a
bad
job
of
being
the
time
keeper
here
for
the
two
meetings,
I've
been
the
timekeeper.
For
so
my
apologies
to
the
world
armstrong.
I
am
I'm
gonna
try
to
find
mine.
A
A
Well,
I
guess
a
question
of
jarrah
is:
if
it's
just
a
product
going
forward,
do
we
need
to
contribute
it,
and
I
think,
maybe
that's
a
question
for
the
larger
chaos
project
that
maybe
we
should
bring
up
in
our
weekly
meeting.
A
A
E
E
A
Okay
got
to
go,
I
gotta
go
see
you
later
and
the
call
and
the
recording.
Thank
you
everyone.
I
have
to
find
out
how
to
stop
my
share.
Thank
you.
Thanks
armstrong,
bye.