►
From YouTube: CHAOSS.Community.October.1.2019
Description
CHAOSS.Community.October.1.2019
A
So
off
we
go
recording
so
hi
everybody,
so
this
is
the
monthly
meeting
you
know
so
welcome
to
that.
It
feels
like
a
weekly
meeting,
sometimes
to
some,
but
this
is
officially
the
monthly
meeting
and
I
did
send
out
an
agenda
at
a
time
that
I'd
like
to
kind
of
address,
I
think
we
have
some
regular
updates
that
we'd
like
to
do
and
maybe
two
items
that
I
think
might
require
a
little
bit
of
action.
A
So
I'm
just
gonna
go
ahead
and
start
with
the
first
one,
which
is
the
Google
Doc
thing
that
Georg
sent
out
yesterday
or
something
that's
received,
a
feedback
already
there.
Could
you,
maybe
just
real
quickly,
give
an
overview
of
what
the
problem
or
concern
was
and
then,
where
we're
at
right
now,
yeah.
E
D
Now
is
it
okay
yeah?
So
basically,
what
it
happened
is
that
we
had
like
all
of
the
minutes
from
previous
meetings
like
I,
did
not
remember
80s,
so
working
in
a
similar
way
as
we
are
working
in
chaos
and
because
the
owner
of
the
document
left
organization,
that
person
belongs
to
or
kind
of
days,
that
account
was
removed.
So
all
the
documents
were
removed,
so
we
lost
part
of
the
work
we've
done
there.
Like
tens
of
pages
to
my
I
was
worried,
study
about
hey.
A
B
B
One
idea
was
what
evolution
working
group
has
done
for
a
while
is
to
create
markdown
files
after
every
meeting
and
store
the
meeting
minutes
in
the
repository,
which
I
think
is
a
good
idea.
It
just
requires
having
someone
to
do
it
after
every
meeting
and
this
I
have
proposed
to
create
a
Google
suite
organization
for
chaos.
B
A
Yeah,
okay,
any
other
thoughts
on
this.
My
one
hesitation
I,
like
the
idea
of
markdown,
but
it's
one
more
thing
to
do
and,
as
we
have
really
a
lot
of
meetings
during
the
week
from
working
groups
and
like
software
groups
as
well,
just
making
sure
that
this
is
consistently
done
every
week
or
every
other
week
crossed
all
working
groups.
It's
just
it's
a
bit
onerous.
That's
all!
That's
my
one
concern
about
the
markdown.
A
So
the
that
I
think
that
workflow,
when
using
Google
Docs
is
we
use
Google
Docs
during
the
meeting
from
what
I
understand
and
then
take
the
results
and
then
just
move
them
into
a
markdown.
So
during
the
meeting
we
would
actually
collaborate
in
a
Google
Doc
and
if
that
vanishes
in
a
week
no
big
deal
because
it's
stored
as
markdown
in
the
repository
right,
it
was
in
response
to
I.
Think
Andy
had
posted
a
comment.
A
F
A
E
C
G
D
A
C
H
In
Google
Docs,
you
take,
you
know,
choosing
not
now
mark
at
the
end
of
the
median.
It's
only
a
matter
of
seven
page
in
the
hood
requests,
but
everybody
who
attended
can
also
check
just
in
case
I,
want
to
add
anything
that
was,
and
the
same
are
down
condition
today:
email,
an
English,
because
it's
basically
a
label
format,
so
that
they're
the
role
knock
down
is
good
enough.
For
two
minutes
there
doesn't
my
to
send
so
I
understand
the
skills
you
need.
C
A
G
C
B
Okay,
so,
based
on
what
I've
been
hearing
I've
put
together
proposal
that
I
will
sent
to
the
main
list,
and
that
is,
we
can
create
a
free
Google
account
and
have
the
documents
owned
by
this
account.
That
way,
we
eliminate
the
risk
of
just
losing
the
document,
but
then,
as
an
additional
step,
all
the
workings
group
can
decide
whether
they
prefer
to
post
the
minutes
remaining
list
or
are
kept
them
in
their
repositories.
Markdown,
okay,.
I
C
E
Had
a
lot
of
conversations
about
that
within
the
kubernetes
community:
okay
and
we've-
we've
kept
Google
Docs,
mainly
because
the
kubernetes
communities
massive
right,
so
anything
that
we
anything
that
we
pick
would
eventually
get
big
enough.
That
would
get
blocked
in
China
anyway.
E
A
A
K
A
K
Like
what
Nicole
said,
I
think
we
should
have
that
in
Todd,
then
from
the
mailing
list
as
well.
A
couple
of
tools,
I
mentioned
I,
also
mentioned,
like
the
flapper,
which
some
communities
like
open
and
many
others
I
use
it's
very
good
for
visioning
as
well.
Okay,
you
can
play
back
we're
trying
to
see.
You
know
the
changes
that
at
some
point,
if
you
want
to
export
it,
HTML
PDF,
whatever
version
options
available
great.
A
Let's
see
software,
does
anybody
want
to
I
know
that
the
folks
from
grimore
lab
Nicotra
Magor
have
been
meeting
on
a
weekly
basis
to
talk
about
software
development
in
the
open,
I
think
this
has
been
really
great
from
my
perspective.
So
does
anybody
from
augur
Sean,
you're
here,
yeah
I
am
here
what's
going
on
most
lately
with
augur.
G
G
So
that's
that's
a
big
thing.
We've
also
started
to
add
to
machine
learning,
capabilities,
get
insight
worker,
and
so
that's
that's
coming
along
nicely
and
also
made
there's
some
really
great
feedback.
Two
weeks
ago
tomorrow
about
user
interface,
and
we
have
made
a
ton
of
changes
to
the
first
interface
that
people
see
on
the
auger
webpage
want
to
try
our
interface.
Can
you
show
it
boom?
I
haven't
I
have
one
that
includes
like
most
recent
things,.
A
G
In
the
repositories
and
events
that
are
outside
of
a
certain
tolerance
of
those
event,
those
trends
so
like,
if
you
have
a
need
rate
in
standard
deviation,
to
see
very
simple
statistics
for
a
set
of
metrics
will
identify
things
that
that
band
automatically
and
will
also
identify
things
that
fall
that
you
could
specify.
So
if
you
just
want
to
be
notified
when
something's
happening
and
get
more
than
ten
messages
or
whatever
issue
so.
C
G
Are
some
some
projects,
and
probably
one
of
the
big
things
you
know
is
that
the
sidebar
now
to
simplify
the
landing
pages?
What
people
might
expect,
which
would
be
the
currently
sort
of
groups
and
stuff
in
sites?
So,
like
you
notice,
the
navigation
is
much
simpler
back
to
the
way
it
used
to
be
because
some
of
the
feedback
that
we
got
was
I'd
like
to
see
the
way
that
it
used
to
be,
and
so
this
and
all
the
repositories
in
there.
G
J
G
G
G
G
A
G
That's
easy
to
do.
You
have
those
links
embedded
in
the
API
documentation.
Each
of
these
pictures
comes
from
so
okay
shouldn't
be
a
problem
to
add
it
here
and
then
insights
insights
are,
whereas
some
of
the
machine
learning
that's
taking
place,
and
this
is
we're
working
on
the
way
that
it's
windowed,
so
it
can
be
parametrized,
but
right
now
it
takes
a
year
of
data
and
gives
you
anomalies
over
the
last
90
days,
okay
and
so
for
Comcast
trickster,
it's
one
of
other
repositories
that
sharp
decreasing
issues
26
days
ago.
A
Okay,
thanks
questions
or
comments
for
Sean.
D
At
you
later,
ger,
heesu's
or
Daniel
yeah,
what's
table
reviewing,
who
were
discussing
before
did
the
only
so.
There
are
a
couple
of
documentation.
Improvements
mean,
are
things
and
then
have
the
most
peaceful
one.
Is
that
put
request
from
to
have
the
day
and
the
day
of
the
week,
information
and
her
neck
is,
was
merged
so
she's
just
a
matter
of
creating
now
the
dashboards
for
having
this
information.
Okay,.
A
A
It
comes
out
of
common,
so
I
think
this
was
is
I
think
this
is
the
discussion
that
you
were
having
last
week
with
how
I
actually
represent
this
yeah.
So
this
is
kind
of
also
for
you
don.
That
said,
in
the
grimore
lab
meetings,
they've
been
talking,
so
this
was
the
metric.
That
was
whether
or
not
people
were
working
inside
or
outside
of
business
hours.
A
You
remember
that
one
and
then
in
common
it
was
kind
of
decided
that
business
hours
are
not
a
great
term,
and
so
it
might
be
more
sensible
to
think
about
when
contributions
are
coming
just
in
terms
of
time,
eight
and
time,
and
then
abstractions
could
be
made
from
that
as
to
whether
or
not
they're
within
business
hours.
But
that's
not
the
responsibility
of
the
group
and
then
grimore
lab
for
the
last
in
their
meetings
for
the
last
I.
H
Guess
wanted
to.
They
just
wanted
to
remind
you
that
you
can
test.
They
are
cotton
that
we
are
from
the
coin.
It's
tomorrow
versus
a
bit.
We
need
to
analyze
the
opposite.
That
may
want
so
feel
free
to
report
any
problem
or
any
issue
that
we
may
find
with
it.
They
hear
that
you
can
select
those
situations
that
the
ones
gonna
write
in,
get
kicked
out,
get
and
and
meet
up,
and
then
you
can
see
some
dashboards
and
all
that
and
you
can
write
down
these
publications
in
Cabana.
H
A
H
A
H
H
That
you
want
is
go
to
the
website
and
see
this
app
and
see
it.
They
see
that
that's
pretty
Calvin
provides,
so
the
only
thing
that
you
need
to
do
is
to
select
the
repositories
you
want
to
analyze.
You
get
a
bit
of
time
because
of
the
github
API
make
may
take
some
time,
but
after
some
time,
basically
you
get
the
data
visualizations
and
everything
using.
A
A
How
much
in
using
culture-
and
this
is
kind
of
a
bigger
question,
but
in
using
cauldron
how
much
like
refinement
do
I
get
right
so
like
if
I'm
deploying
or
more
lab
locally
there's
a
ton
of
refinement
that
can
be
done.
So
does
culture
and
abstract
some
of
that
away,
or
is
it
kind
of
pretty
good
indication.
J
H
You
just
go
there
and,
and
it
works
of
course
you
can
also
deploy
it.
If
you
want,
because
we
have
everything
is
automated
so
that
you
can
also
happen,
she
brew
and
she
will
cook
books
for
put
in
there.
But
if
you
don't
want
to
just
go
to
the
website
and
the
your
stamp,
not
it
not
in
deployed
nothing
just
just
work
with
the
data.
Okay,.
B
B
D
B
J
B
B
H
B
And
then
meetups,
which
this
community
doesn't
have
meat,
okay
and
then
there's
one
dashboard
for
each.
So
if
you
wanted
to
look
at
contributors
in
more
detail,
we
can
go
to
the
contributors
dashboard,
which
then
has
again
not
as
many
visualizations
as
you
know,
from
Ramola
mm-hmm.
It
just
has
total
number.
This
is
the
active
contributors
over
time.
B
H
A
That
was
part
of
my
question
like
when
I'm
using
culture
and
like
I
know
when
I'm
using
grimore
lab
locally,
but
there's
a
lot
of
that.
Customization
and
I
was
just
wondering
when
it
was
served
as
a
service
like
this.
How
much
of
that
customization
still
exists?
That's
all
that
was
my
question,
but
it
sounds
like
a
lot
so.
J
B
A
A
A
H
C
H
For
their
standard
from
our
lab,
it
happed
them
a
bit
and
they're
just
to
show
that
it
works
with
time.
Our
idea
is
to
load
with
collections,
or
by
that
word,
so
at
some
point
they
would
like
to
have
like
the
chaos
collection,
for
instance,
except
we
have
our
metric
since
everybody
can
do
their
own
metrics,
I
also
suspect
or
share
in
their
words.
Okay,
that's
something
that
is
not
going
to
happen
in
the
next
few
weeks.
At
some
point,
we
want
to
put
that
into
them.
There's
a
whopper!
H
Now
we're
still
testing
how
quick
we
can
get.
We
can
go
get
in
repositories,
how
we
can
store
the
data
and
all
of
that,
because
it's
a
bit
a
bit
close
to
big
data
more
than
than
usual
friends,
because
we
are
having
now
thousands
of
repositories
here
and
we
try
to
have
hundreds
of
thousands
of
repository.
So
it's
okay,
it's
a
matter
of
scale
right
now.
Okay,.
L
C
L
Were
for
the
cell
for
the
ones
that
we
think
that
they
are
not
that
common,
but
they
are
useful.
For
example,
contributors
growth
is
not
only
counting
the
number
of
people
contributing
to
get
or
counting
the
number
of
people
commuting
to
the
hub.
I
mean
there
are
three
different
charts
is
all
together
in
one
time,
so
you
can
see
what
people
are
contributing
in
a
single,
so
I
think
that
makes
a
difference
from
any
other
point
in
the
market.
I
want
to
put
that
in
the
front
page
of
the
code,
yeah.
A
A
L
Have
the
numbers?
Basically,
you
have
a
tool
that
is
given
a
position.
We
are
using
a
pen
distr
for
this,
so
basically
it's
100%
free
open
source.
So
you
can
build
your
own
visualization
some
of
these
and
then,
of
course,
the
planning's
okay,
let's
put
here
also
the
cosmetic,
so
people
coming
for
the
cosmetics
and
CEO.
This
is
how
they
look
like
and
they
people
can
start
working
on
them
and
even
customize
them
for
their
own
needs.
L
A
And
I
do
think
that
when
I
was
looking
at
these
dashboards,
that
Gary
has
up
here.
I
do
think
again.
This
is
totally
mean
that
that
being
attentive
to
how
you
aggregate
things
is
gonna
be
really
important
like
what
you
show
people
it
is
your
choice
to
show
people
is
gonna,
be
really
important,
because
that's
gonna
be
the
first
thing
that
people
see
and
people
may
or
may
not
modify
things.
A
They
may
not
even
want
to
go
down
that
road,
but
they're
gonna
rely
on
what
you're
giving
to
them
would
be
my
guess
again:
I'm
not
guaranteeing
some
some
causality
or
anything
but
I.
Think
you're
gonna
have
a
lot
of
influence
just
of
what
is
in
front
of
people.
I
think
that's
important
to
keep
in
mind
no.
B
One
one
more
feat
that
I
wanted
to
highlight
is
when
one
of
the
features
of
Coldren
is
that
the
data
needs
to
be
only
collected
once
for
all
of
the
users.
So
if
I
created
a
new
dashboard
so
by
clicking
on
analyze
a
project
which
doesn't
have
any
data
source
right
now
and
I
said
I
wanted
to
analyze
chaos,
it
would
add
all
of
the
data
that
is
already
stored
in
the
database.
Sure
it's
a
month
old,
but
it's
there
and
I
can
immediately
go
and
look
at
the
data.
B
A
A
Along
those
lines
missing
something
we
were
missing
just
kind
of
a
the
core
nature
of
describing
what
a
metric
and
what
the
objective
of
that
metric
is
because
I
think
that's
that's
really
one
of
the
goals
here
of
these
templates
so
Georg
and
put
together.
If
you
take
a
look
on
page
3
again,
he
brought
this
up
last
week
as
a
way
to
bring
together
the
two
different
metric
templates
and.
B
A
Was
everybody
so
that's
cool,
so
the
idea
here
is
that
we
would
have
really
only
four
required
headings:
four
required
top
level
headings,
and
that
and
and
at
the
same
rate
to
all
of
these
headings
are
what
is
already
occurring.
So
we're
not
asking
for
more
or
less
really,
it's
just
an
opportunity
to
standardize
on
things
and
really
kind
of
highlight
these
objectives,
these
descriptions
and
objectives.
So
if
you
take
a
look
on
point
three,
so
we
would
retain
the
description
of
the
metric.
A
We
would
retain
the
objective
of
the
metric
as
two
high
level
headings
and
then
underneath
there
we
had
this
implementation
component
and
we
for
a
while
had
a
bunch
of
different
ways
that
implementations
were
being
represented
through
the
metrics,
whether
it
was
through
the
filters
and
the
visualizations
known
implementation
sample
notations
success
metrics,
as
coming
from
the
other
templates
data
collection
strategies.
We
had
a
bunch
of
different
ways
that
we
were
all
thinking
about
implementation
and
it
ended
up
being
what
felt
like
kind
of
distributed
throughout
the
metrics
template.
A
So
what
this
implementation
section
would
be
is
you
would
have
an
implementation
here
as
your
0.3
and
then
the
smaller
subheadings
are
optional
as
ways
to
represent
the
metric.
So
if
it's
in
the
case
of
say,
trace
data
metrics
that
say
live
in
risk
or
evolution,
you
may
gravitate
towards
things
like
filters
and
visualizations
tooling,
that
is
providing
the
metric
things
that
you've
been
doing
in
the
asked
and
in
the
case
of
diversity
and
inclusion,
work.
A
A
B
This
is
from
the
in
I
working
group
where,
when
where
we
have
a
different
concepts
of
a
metric,
when
we
talk
about
the
metric,
it's
something
like
speaker,
diversity
at
events
and
then
there
are
different
ways
to
measure
it
which
could
be
percent
of
women
on
keynote
lists
on
the
website.
It
could
be
a
survey
item
asking
hey:
did
you
feel
properly
represented
by
the
speakers
in
this
event,
so
there
are
different
ways
to
collect
information
about
this
high
level,
metric
that
that's
what
I
put.
H
H
Was
looking
at
it,
let
me
check
again:
okay
I'm
just
trying
to
understand,
because
I
think
that
there
is
a
bit
of
a
difference
between
what
we
call
the
matrix
in
this
case,
and
you
know
that's
why
I
was
asking
and,
from
our
point
of
view,
is
not
that
they
are
doing
a
different
thing,
but
the
way
they
are.
We
are
calling
things
in
different
ways.
It's
only
then.
H
H
For
instance,
if
we
focus
on
our
quantitative
success
metrics,
if
I,
if
I,
have
to
map
this
according
to
what
I
did
in
for
evolution,
for
instance,
the
thing
observed
what
snide
probability
impression,
blah
blah
blah
could
be
the
goal,
and
then
you
set
up
some
specific
metrics
that
help
you
to
understand
at
all,
like
how
many
diversity
access
tokens
are
available,
because
that's
a
number,
that's
something
that
you
can
measure
things
like.
What
are
the
criteria
for
qualifying?
I
I
have
a
hard
time,
understanding
that
this
metric,
the
metric
could
be.
H
C
A
B
So
I
know
this
is
where
we
have,
because
there's
no
one
single
number
that
expresses
diversity
access
tickets,
it's
something
that
we
want
to
measure
and
then
there's
different
ways.
We
go
about
measuring
it
and
that's
what
the
metrics
down
below
are.
That's
get
those
boolean
values,
the
numbers,
the
Likert
scale,
items
never.
H
C
H
May
get
confused
because
people
usually
expect
man.
In
my
experience
at
least
people
usually
expect
metrics
to
be
what
you
have
now
under
observe
website
interview,
organizers.
All
of
those
items
are
metrics,
because
basically
there
are
other
numbers
or
boolean,
so
something
that
can
be
des
no
and
then
I
understand
that
by
aggregating
all
of
that
information,
you
decide
the
level
of
inclusiveness,
for
instance,
and
that's
where
I'm
meaning.
So
in
the
end
of
course,
you
need
several
metrics
and
you
need
to
aggregate
them
in
some
way
to
learn
about
the
goal
about
the
route.
H
Code,
which
is
what
you
want
to
know
whether
it
is
project,
is
inclusive
or
not,
for
instance,
so,
but
it's
only
a
matter
of
wording.
So
that's
why
it
also
could
be
solved
by
saying
we
are
considering
like
metrics
like
this
in
this
and
just
define
that
so
that
people
understand,
if
you
prefer
not
to
change
everything.
J
H
J
J
A
Chaos
metrics
as
they
were
released
under
diversity
and
inclusion.
They
follow
the
focus
areas
with
a
goal
for
that
focus
area
around
event,
diversity
and
they
have
the
metric,
which
is
diversity,
access
tickets.
It's
really
cannot
be
answered
in
a
boolean
style
way,
but
from
what
I
understand
with
diversity
and
inclusion,
the
goal
is
to
understand
at
least
the
nice.
A
They
have
that
subheading
called
success
metrics,
so
there
is
actually
metrics
twice
in
this.
There
is
the
metric
that
is
diversity,
access,
tickets
and
then
there
is
something
called
success:
metrics
which
are
like
sub
metrics,
so
the
higher
level.
So
it
may
be
worth
simply
renaming
something
like
success:
metrics
to
success,
criteria
or
success,
approaches
or
success,
considerations
or
just
considerations,
because
then
the
metric
itself
stays.
This
is
this
could
be
discussed
in
the
working
group
itself,
but
the
metrics
day's
diversity
access
tickets
and
there
are
ways
different
ways
to
capture
that
information.
H
H
Activity
are
in
git
repositories
and
activity
initiatives
in
activity,
and
you
can,
with
all
of
that,
you
can
produce
a
single
number,
which
is
something
like
from
0
to
1
how
much
activity
and
that's
still
a
number
and
that's
still
a
metric,
and
so
my
problem
is
not
with.
That
is
the
other
way
around
is
with.
If
we
have
a
number
we
have
a
matter.
If
we
have
a
matter,
we
have
a
number.
Anything
else
is
another
very
different
thing.
H
If
we
want
to
redefine
that's
fine,
but
in
that
case
we
should
clarify
for
matrix,
we
understand
distances,
but
then
I
understand
this
is
complex
and
every
working
group
may
have
different
ways
of
looking
over
there
at
so
I'm
uncomfortable
with
any
other
decision
that
they
may
take
in
the
diversity
and
inclusion
book
anymore.
Okay,.
A
So
if
we
do
move
forward
with
this,
maybe
I'll
bring
it
to
the
working
groups
and
one
of
the
challenges,
maybe
not
challenges,
but
things
to
do
is
we'd
have
to
go
back
through
the
metrics
that
were
released
with
version
1
and
rethink
the
just
how
they're
laid
out
for
the
second
release.
That's
coming
up
with
foz
dem,
so
obviously
version
1
stays
as
it
is,
but
we
would
have
to
just
kind
of
rethink
how
we
structure
them
again.
This
doesn't
change
any
of
the
content.
G
A
Skipped
working
groups
getting
skipped
or
events
getting
skipped.
Okay,
you
don't
have
to
cover
everything
every
month,
so
I
kind
of
like
to
go
to
events.
To
be
honest
with
you
right,
oh
I,
think
chaos
Khan
is
coming
along
pretty
well
from
what
I
understand.
I,
don't
know
anybody
want
to
have
a
comment
on
that
or
the.
E
I'm
actually
working
on
the
pull
request
to
put
the
CFP
to
open
the
CFP
today
and
put
it
up
on
the
website,
so
we'll
need
some
help
from
the
Linux
Foundation,
promoting
it
once
that
I'm
still
where
I
haven't
submitted
the
pull
request.
Yet,
but
once
the
pull
request
gets
merged,
we
will
need
some
some
help
promoting
it,
but
I
think
we
can
probably
kick
some
of
that
off
on
the
mailing
list.
Okay,.
A
B
J
A
E
A
Doesn't
prevent
me
or
gear
from
asking
and
continuing
to
push
to
get
it
done,
and
the
hope
is,
is
the
community
bridge
is
relatively
new,
so
I
think
a
lot
of
this
is
still
being
worked
out,
and
so
maybe
there's
a
room
to
enable
this.
Maybe
it
just
hesitancy
some
of
the
things
we
can't
pay
for
might
be
changed
as
people
start
say
what
no
some
of
the
things
they
cannot
be
for
could
change
as
they
evolve.
What
it
is
exactly
once
they
figure
out
the
Meccan
by
twitch.
It
works,
alright
cool.
J
A
G
M
The
I'm
I
got
a
few
responses,
but
possibly
a
call
for
a
meeting
to
possibly
organize
our
thoughts
around
what
the
group
would
do.
M
J
A
Thing
so
so
Kevin
has
put
together
a
calendar
which
includes
all
of
the
chaos
meeting,
Thanks
and
I
think
we're
gonna
put
it
up
on.
You
decided
this
right,
or
at
least
fuzzy
consensus
that
we
can
go
ahead
and
put
it
up
on
the
chaos
that
community
page
and
if
people
want
to
download
this
calendar,
they
can
include
it
in
their
own
calendar.
With
all
the
caveats
that
come
with
that,
however,.
A
Like
okay,
so
the
way
I
figured
I
didn't
so
the
credit
question
was
is:
if
we
make
this
calendar
available
on
chaos
that
community
and
then
people
choose
to
use
it
right,
they
put
it
in
their
own
calendar
as
a
way
to
see
chaos
meetings.
There
are
obviously
caveats
that
come
with
that,
as
I
have
learned
over
the
past
few
weeks,
but
if
we
change,
if
there
is
a
change
to
the
calendar,
so,
for
example,
DNI
is
talking
about
changing
their
time.