►
Description
Part 1: Data Confidentiality and Privacy Needs in Scientific Computing
Trusted CI scientific researchers must participate in conversations about the challenges that they face in dealing with data confidentiality and privacy constraints in their work. Our eventual aim is to produce documentation that will help the community work toward "right sized" security and privacy solutions that enable scientific progress, at acceptable costs to usability. To accomplish this, we are seeking conversations with a diverse range of scientists to best understand the ways in which security and privacy issues hinder obtaining the data that is needed to conduct certain types of scientific research, and also to understand those researchers' scientific computing workflows in sufficient detail to best work toward what acceptable solutions to address those challenges might look like in the future.
Date: 03/06/20
Presenter: Sean Peisert
Institution: Lawrence Berkley National Laboratory
West Big Data Innovation Hub
Part 2: A Brief Update from the Trustworthy Data Working Group
Jim Basney and Melissa Cragin provide a brief update from the Trustworthy Data Working Group, a collaboration between Trusted CI, the BDHubs, and others to survey science projects to determine the spectrum of data security concerns and practices in the scientific community and to provide guidance on data security for open science, to improve scientific productivity and trust in scientific results. The working group is drafting the survey now, and we welcome your participation in the group and your assistance with obtaining survey results from a broad and diverse group of science projects.
Date: 03/06/20
Presenter: Jim Basney & Melissa Cragin
Institution: National Center for Supercomputing Applications & San Diego Supercomputer Center
Midwest & West Big Data Innovation Hubs
A
Well,
why
don't
we
start
up
I'm
John,
Goodhue,
executive
director
of
the
messages,
green
high-performance,
computing,
Center
and
I
will
be
your
moderator
today.
I
thought
we'd
start
by
just
going
around
the
virtual
room
and
just
a
quick
introduction
who
you
are
and
where
you're
from
and
I'll
I'll
start
from
my
upper
left
in
the
in
the
zoom
screen
with
Alfonzo.
D
A
A
C
A
C
C
A
A
F
A
C
Yeah
Patrick
Clements
I
manage
the
cyber
infrastructure
for
the
Romanoff's
core
grant
at
the
University
of
Vermont
and,
let's
see
the
other
Jim
welcome,
wish
hey
yeah
I
direct
research
computing
at
the
University
of
Minnesota
and
research.
Computing
includes
the
Minnesota
supercomputing
Institute
University
of
Minnesota
informatics,
Institute
and
new
spatial.
A
E
Right
now
we're
working
collaboratively
with
the
other
hopes
to
plan
for
the
all
hubs
meeting,
which
I
hope
still
happens
with
all
of
these
conference
cancellations
because
of
the
novel
coronavirus,
so
I'm
helping
a
Meredith
and
the
team
out
on
the
the
demo
plan.
For
that
and
in
the
meantime,
there's
nothing
else
big
to
report.
C
D
Yep
Meredith
actually
sent
a
note,
so
I'm
I'm,
going
to
read
from
Meredith's
note.
Meredith
is
unable
to
join
today.
She's
someplace,
where
the
environmental
conditions
just
won't
allow
she's
in
a
meeting.
But
she
writes
that
on
our
behalf.
In
light
of
the
of
koban
19
at
the
hubs
and
the
local
host
team
at
Ohio,
State
University,
that's
the
Ohio
State
University
are
discussing
contingency
plans
for
the
all
hubs
summit,
which
is
scheduled
to
happen
in
May
of
this
year.
D
Policies
and
guidance
from
any
institutions
and
organizations
engaged
in
the
hub's
community
are
emerging
and
I
will
share.
She
put
a
link
to
a
new
set
of
announcements
or
an
FAQ
from
NSF
and
I'll
post
that
it's
not
directly
relevant
to
the
hubs,
but
it
is
a
set
of
questions
that
NSF
is
answering
about
managing
all
things
related
to
NSF
and
and
the
virus,
and
that
they'll
continue
to
monitor
and
communicate
updates.
D
C
A
A
D
Nope
got
it
I
got
it.
Work
continues
across
a
number
of
the
the
OS
and
working
groups,
so
we
have
a
policy
working
group.
In
fact,
John
and
I
just
had
a
call
about
that
today
and
we're
working
on
the
intersection
now
of
ongoing
implementation
efforts
and
the
policy
development,
so
that
is
happening
we're
beginning
to
share
draft
policy
documents
with
a
few
folks
to
get
some
feedback
before
we
roll
that
out
entirely
to
the
public.
We
are
in
terms
of
use
cases,
which
is
my
area.
D
We
continue
so
from
the
original,
nine
or
ten
use
cases
that
were
in
the
original
proposal,
several
of
which
were
under
or
attached
to
different
big
data
hubs
about
half
of
those
use.
Cases
were
ready
to
go
and
collaborate,
and
so
we
are
now
loading
data
from
from
those
projects,
if
they're
not
already
loaded
we're
getting
them
loaded
into
the
OSN
into
pods
across
the
different
places
in
the
country.
D
We've
established
collaborations
with
some
third
party
front,
end
groups
for
some
of
those,
so
we're
working
with
irods
group
around
one
of
the
use
cases
we're
working
with
the
cloud
or
group
at
NCSA.
For
another
case
and
in
addition
to
those
original
cases,
we
have
both
projects
funded
by
the
Geo
Directorate,
which
is
great
because
geo
helped
fund
the
OSN.
D
We
have
some
new
projects
from
geo
that
have
pretty
large
data,
so
they
are
sitting
between
10
and
100
terabytes
per
per
bin
and
and
then
we're
looking
at
at
a
project
that
wants
about
a
petabyte
of
space,
so
we're
beginning
to
work
with
them
on
how
that
might
happen.
So
a
handful
of
new
use
cases
we're
beginning
to
roll
out,
and
this
is
all
good.
D
D
We
have
data
in
we're
getting
data
out
when
those
calls
happen
and
we're
happy
to
if
anybody
is
interested
or
there's
project
spoke
projects
in
particular
that
need
a
place
to
put
active
data
for
for
use,
so
transfer,
trance
transfer
for
compute
or
or
store
for
for
access.
Please
be
in
touch
we're
happy
to
talk
with.
A
Ya,
I
think
I'll
just
emphasize
that,
because
we
are
now
open
for
business
and
and
interested
in
engaging
I,
think
it's
fair
to
say
with
with
with
new
datasets,
and
we
did
name
the
hub's
as
a
place
in
the
proposal
that
has
kind
of
a
special
position
with
respect
to
identifying
you
use
cases
and
also
for
new
pod
sites.
So
new
places
to
to
join
the
open
storage
network.
A
A
A
A
We
have
actually
multiple
presenters
today.
I
think
it
was
Florence
who
put
us
on
to
this
Shaun
Peyser
from
Lawrence
Berkeley
labs
will
we'll
start
off
talking
about
confidential
confidant
kind
of
confidentiality
and
privacy
in
scientific
community
computing
and
Jim
Bosnia
Melissa
Craigan
will
talk
a
little
bit
about
the
trustworthy
data
working
group,
so
I'm
going
to
kind
of
without
further
ado
Shaun
hand
the
hand
the
keys
over
to
you
and
I'll.
Let
you
introduce
yourself
and
your
topic
and
how
about
it?
Thanks.
G
One
of
the
things
I
am
specifically
embarking
on
with
regard
to
that
this
year
for
trusted.
Ci
is
understanding
data,
privacy
and
confidentiality
issues
in
scientific
computing
and
I
mean
a
few
things
by
that.
The
the
primary
just
that
I'm
looking
at
this
through,
however,
is
the
notion
that
scientists
often
have
a
hard
time
getting
data
and
when
they
do
get
data
or
often
have
very
onerous
restrictions
placed
on
them.
G
In
order
to
use
that
data
and
I
am
trying
to
understand
really
what
the
what
sort
of
issues
scientists
are
facing
and
what
ways
they
are
currently
trying
to
overcome
to
address
those
needs
and
where
there
are
still
gaps
where
either
they
haven't
solved
that
problem
at
all
or
where,
where
the
restrictions
are
so
onerous,
it
essentially
halts
all
science
and
to
dig
in
this
a
little
bit
more
just
to
give
a
few
concrete
examples.
There
is
absolutely
data
that
is
regulated
it
is
covered
by,
but
it
is
covered
by
FISMA
FERPA.
G
You
know
you
name
it,
and,
and
we've
all
had
stories
where
there's
some
sort
of
regulation
that
is
placed
on
University
researcher,
and
you
know
in
some
cases
that
the
university
researcher
needs
to
build
some
sort
of
aircraft
cluster
in
order
to
actually
analyze
that
data.
In
other
cases,
perhaps
the
university
says
you
know
what
I
just
can't
deal
with
that
in
other
cases
and
I've
heard
of
this
as
well
institution
providing
the
data
will
say:
I'm
not
going
to
give
you
data
at
all.
G
You're
gonna
come
to
our
site
in
order
to
access
it,
you
don't
bring
anything
in
you,
don't
bring
anything
out
except
your
notebook
and
and
that's
it
and,
and
so
we've
seen
that
sort
of
thing
too.
There
is
also
are
also
broad
categories
of
data
that
are
not
regulated
at
all,
but
are
in
some
way
sensitive.
You
know,
depending
on
what
state
or
jurisdiction
you
live
in,
maybe
there's
something
containing
I'm
just
making
some
traffic
patterns
of
people's
driving
behavior
and
in
some
states
there
may
be
a
lot
of
sense
about
that.
G
In
other
cases
you
know
it's
not
actually
regulated,
but
it
contains
personal
information,
and
so
it's
not
creepy,
and
so
an
University
researcher
may
have
trouble
getting
that.
In
other
cases
there
may
be
data
that
is
commercial
and
proprietary.
So
a
command
entity
is
not
just
gonna
give
up
the
farm,
because
this
is
something
they
sell.
G
Yet
it's
like
okay
I'd,
like
to
provide
some
way
to
train
your
machine
learning
model
on
this
data,
but
I
don't
want
to
actually
send
it
to
you
and
I'm,
not
sure
what
it
would
look
like
to
bring
you
into
my
environment
and
and
so
on.
So
you
know
I
think
that
there
is
even
more
categories
of
sort
of
lack
of
data
sharing
as
well.
G
So
the
question
comes
down
to
then
really
trying
to
have
conversations
with
people,
both
I,
think
who
own
data
and
don't
know
how
to
make
it
more
broadly
available,
as
well
as
people
who
are
hindered
by
not
being
able
to
obtain
data
for
some
of
these.
These
different
reasons
and
I
would
love
to
interview
those
people
to
really
understand
the
situation
they're
in
what
their
research
methods
look
like.
What
does
their
computational
pipeline
look
like?
How
would
they
actually
like
how
do
they
need
to
compute
on
that
data?
G
Should
it
be
made
available
and
really
try
and
understand?
What's
acceptable
solutions
to
both
sides
might
end
up?
Looking
like
and
the
end
goal
of
this
perhaps
tour,
the
second
half
of
this
year
will
be
to
work
towards
a
some
sort
of
best
practices
document
where
I'm
sure
it's
not
gonna
be
one
size
fits
all,
but
it
hopefully
will
provide
some
different
scenarios
and
methods,
though
that
will
help
lead
to
some
solutions
and
that's
all
I
have
to
speak
about
and
I'm
welcome
to
field
Q&A.
A
Had
a
fortuitous
comment
in
a
and
I
questioned
the
crew,
his
comment
was:
was
my
favorite
category
of
data
from
student
days
was
a
couple
of
friends
who
were
informed
that
their
theseus's
were
classified
after.
They
wrote
them
because
they
brought
public
domain
data
together
in
novel
ways
that
revealed
secrets
inadvertently.
I,
don't
know
where
that
fits
in
your
taxonomy
that
the
may
be
the
more
constructive
thing,
and
it
may
have
already
done
this
already.
A
G
A
G
A
G
On
the
other
hand,
I
know
that
other
universities
like
to
say
do
you
work
for
the
DoD
and
I
think
that
there's
sort
of
always
this
overhanging
threat
that
that
could
be
involved
and
that's
that's
outside
of
the
determination
of
the
University
but
then
again,
I'm
just
kind
of
speculating
about
these
things
and
I
would
love
to
know
more
about
that
from
somebody
who
faces
it.
I'll
give.
A
G
F
Is
Jay
Sean?
Thank
you
for
the
introduction
of
your
endeavor,
so
I
think
that
there's
always
this
very
often
a
question
about
what
we
say:
data
there,
all
kinds
of
data
and
different
levels
of
data,
the
metadata
abstraction
of
data
and
so
on
so
forth
and
I
think
there's
a
there's,
a
big
difference
between
sharing
and
wanting
to
share
the
raw
data
versus
sharing
and
wanting
to
share
the
characteristics
of
the
data
and
many
many
researchers,
especially
data
scientists,
type
of
researchers.
F
G
Documents
show
how
big
companies
by
credit
card
data
on
millions
of
Americans,
and
they
quote
netweaver
down
here
at
UC
Berkeley
who
says.
Let
me
be
blunt:
this
is
BS
anonymization,
and
so
it's
one
of
the
the
interesting
things
I
think
Junior
you're,
pointing
out
that
people
are
trying
to
share
data
and
they're
trying
to
at
least
in
some
cases
are
trying
to
be
responsible
about
it,
but
still
aren't
doing
it.
G
F
I
agree,
so
so,
yes,
indeed
@dr,
so
there's
a
difference
between
anonymization
versus
that
you
use
some
kind
of
irreversible
function
to
kind
of
characterize
it.
So
right,
but
I
agree
with
you
dead.
They
are
it's
difficult
to
say
using
one
sentence
to
describe
everyone,
so
I
think
that
Casa
white,
that
is
good
thing,
I,
look
forward
to
how
you
would
categorize
different
things
by
the
way
I
will
volunteer
myself.
If
you
want
to
include
me
as
a
consumer
of
data
Oh.
G
A
A
B
Okay,
okay,
so
hopefully
you're
seeing
my
screen
here
and
so
mostly
and
I
will
be
giving
a
short
update
from
the
trustworthy
data
working
group,
which
is
a
collaboration
between
trusted
CI
and
the
BT
hubs
and
and
representatives
from
other
organizations
and
projects
that
you
hate
see
listed
here
on
the
bottom
of
the
of
the
slide
and
here's
the
URL
the
home
page
for
our
working
group
and
we're
going
to
show
that
again.
At
the
end
of
our
presentation.
And
with
that
I'll
hand
it
over
to
Melissa.
To
talk
about
our
mission.
Great.
D
So
you
know
thinking
about
open
science
and
the
the
basis
of
open
science
really
is
on
the
integrity
of
data
such
that
there's
some
basis
of
understanding
or
agreement
or
again
trust
that
the
underlying
data
and
resources
that
we're
using
for
new
uses
or
to
interoperate
with
other
other
data
resources
that
that
has
some
level
of
integrity.
That
allows
us
to
move
forward
and
have
confidence
in
the
in.
D
Not
only
the
claims
were
making
anew,
but
also
that
we
can
reliably
reproduce
results
so
the
the
context
now
so,
while
we're
thinking
about
this
under
trusted
CI
in
the
cyber
security
context,
predominantly
it
also
this.
This
cyber
security
context
sits
in
a
broader
policy
realm
now
around
the
broader
us.
Research,
landscape
and
Jim
go
ahead
and
go
ahead
and
move
to
the
next
slide
Thanks.
So
in
in
the
last
year,
or
so
it's
been
closer,
it's
been
pretty
close
to
a
full
year.
D
Now,
there's
a
emerging
effort
with
a
lot
of
activity
across
agencies
and
being
driven
out
of
the
Office
of
Science
and
Technology
Policy,
that's
part
of
the
White
House
and
the
White
House's
National
Science
and
Technology
Council,
and
they
are
driving
an
effort
to
understand
more
broadly
and
provide
policy
around
the
research
environment
and
risks
to
the
u.s.
research
environment,
particularly
related
to
open
science,
but
also
related
to
engagement
with
international
science,
in
addition
to
rigor
and
research
integrity.
D
So
I'll
just
speak
for
a
moment
about
a
couple
of
these
things,
just
because
I
think
it
gives
a
very
nice
broad
context,
very
nice
and
broad
context
for
some
of
what
we're
thinking
about
and
it
it
lets
us
connect.
If
you
will
a
lot
of
the
projects
that
are
involved
in
the
working
group
to
the
broader
efforts
coming
out
of
not
only
OSTP
but
but
now
a
lot
of
work
coming
out
of
NSF.
So
dr.
D
dragon,
Meyer
back
in
September,
put
out
a
letter
about
concerns.
That
letter
was
open
to
the
public
but
predominantly
focused
on
academic
institution
and
research
organizations
regarding
efforts
by
some
foreign
powers
to
undermine
research
activities
and
research
environments
in
the
US
and
that
US
policies
and
practices
need
to
be
evolving
in
order
to
guard
against
external
attacks
and
and
and
ethics
breaches
of
a
whole
range
of
in
a
whole
range
of
ways
in
the
system.
But
to
a
large
extent,
not
only
does
this
concern,
you
know
unreported
funding
or
unapproved
international
research.
D
There's
also
concerns
about
the
research
environment,
more
broadly,
regarding
the
integrity
both
at
the
intersection
of
computing
and
cyber
infrastructure
security.
But
also
data
and
software
integrity
and
then,
of
course,
that
broader
concept
of
trust
underneath
the
National
Science
and
Technology
Council,
there
is
a
Subcommittee
on
research
security
around
developing
best
practices
for
research
institutions
and
as
well,
there's
another
subcommittee
on
ensuring
rigor
and
integrity
in
research
and
that
research.
D
That
subcommittee
is
really
interested
in
identifying
cross
agency
principles
and
engaging
with
academic
institutions
and
other
research
organizations
and
national
stakeholders
around
priorities
and
actions
to
enhance
the
integrity,
reproducibility
and
replicability
of
of
science
and
scholarship
in
the
research
enterprise
and
so
they're,
working
on
all
manner
of
ways
to
develop,
supports
and
services
activities,
think
about
transparency
and
address
incentives
and
communication
training
and
other
things
and
those
other
areas.
Those
other
things.
Those
other
areas
that
they're
looking
at
includes
the
work
we're
doing
under
the
auspices
of
the
working
group.
D
I
just
want
to
point
out
quickly
an
NSF,
the
NSF
activity.
Interestingly,
so
this
Jayson
study
just
came
out
supported
by
my
tour.
It
just
came
out
in
December
of
2000
in
2019,
and
it
is
really
focused
on
the
entire
landscape
of
security
for
research.
But,
interestingly,
they
don't
they.
In
that
report
they
didn't
spend
a
lot
of
time
really
looking
at
the
integrity
of
data
aspect.
D
However,
the
new
appointment
by
in
a
second
research
security
certainly
covers
that
as
well.
So
I
just
want
to
give
this
sort
of
order
view
because
they'll
be
they'll,
be
new
policies
and
a
lot
of
funded
efforts
by
multiple
agencies
to
help
boost
up
the
capacity
in
and
really
start
to,
ensure
and
firm
up
security
approaches
across
the
research
landscape.
B
Great
thanks,
mosa,
and
so
in
in
within
that
context,
our
working
group
has
some
very
specific
goals.
We
would
like
to
broadly
survey
science
projects
about
that
there,
the
spectrum
of
their
data
security
concerns
and
also
understand
their
current
data
security
practices.
What's
working
well,
what's
not
working
well
and
based
on
the
responses
that
we
see
we
received
from
that
survey
produce
broadly
applicable
security
guidance
for
the
data
science
community.
So
we
have
a
one-year
timeline
for
the
working
group
we
launched
in
January
and
right
now
we're
working
on
preparing
that
survey.
B
So
we
get
a
good
breadth
of
responses,
so
we
give
ourselves
the
month
of
April
to
analyze
those
survey
responses
and
also
submit
our
baath
proposal
to
perc
and
so
that
we
can
publish
an
analysis
of
our
survey
results
in
May
and
ideally
be
prepared
to
present
those
results
at
the
all
hubs
summit
and
then
in
June.
Based
on
what
we
find
from
the
survey
will
identify
some
guidance
topics
and
begin
developing
that
that
guidance
for
the
community
will
will
meet
at
perc.
B
I
think
will
also
be
represented
at
the
esup
meeting
and
then
in
August.
We've
got
an
opportunity
to
to
meet
up
at
DEFCON
and
do
some
outreach
at
DEFCON
with
the
goal
of
having
a
draft
report
to
publish
in
September
in
time
for
the
NSF
cybersecurity
summit,
and
so
we'll
gather
input
there
at
the
summit
and
online
and
email
address
is
an
email
responses
and
mailing
lists
and
and
we'll
do
an
SG
CI
webinar
to
gather
more
input
on
the
guidance
in
that
report.
B
Update
the
reports
present
at
supercomputing
and
with
the
goal
of
publishing
the
report
and
concluding
the
working
group
work
in
December,
where
we
also
have
opportunities
to
do
our
reach
at
a
geo
and
trusted
CI.
So
so
you
can
see
here,
we've
identified
some
some
areas
of
outreach,
some
conferences
to
be
sure
to
do
outreach
to
or
outreach
at,
but
a
very
interested,
also
in
your
input
on
other
outreach
opportunities
that
we
can
put
on
the
calendar,
and
so
most
of
we'll
talk
more
about
how
you
can
help
us.
Indeed,.
D
So
picking
up
again
on
the
on
the
survey
aspect,
certainly
being
able
to
provide
us
some
input
on
the
survey
draft
and
there
it's
easy
enough
to
access
that.
If
you
can't
see
it
now,
we're
happy
to
have
you
contact
us
and
we
can
give
you
access
so
there
we
go
so
so
we
would
love
to
have
some
feedback
on
that,
we're
continuing
to
work
and
then
hopefully,
right
it'll
be
ready
to
go
into
IRB.
It
takes
a
few
weeks
to
get
clearance
and
so
we'll
be
looking
forward
to
having
that
back.
D
So
we
can
send
that
out
in
short
order.
The
next
piece
is
please
for
all
of
you
that
have
listservs
that
you
manage
or
newsletters
that
you
put
out
or
if
you're
a
part
of
communities,
or
even
you
know
you
could
post
it
out
on
your
Twitter
feed
if
you're
so
inclined
and
that's
to
help
us
solicit
survey
responses
from
the
broad
community.
We
it's
really
important,
an
important
part
of
at
a
port.
Sorry,
an
important
aspect
of
this
survey
is
to
gather
the
broadest
perspective
possible
on
on
trust
and
integrity.
D
D
D
D
D
Yes,
so,
just
in
late
November,
the
NSTC
committee
put
out
through
the
Federal
Register
a
request
for
information
on
the
broader
research
environment.
That
I
think
is
going
to
get
a
very
broad
set
of
responses
and
I
think
you
know
I,
don't
know
what
like
the
breath,
that
will
uncover
my
you
know
if
I
were
to
guess
I'd
say
there'd,
be
a
lot
of
alignment
between
the
kinds
of
responses
that
that
and
at
the
NSTC
or
OSTP
receives
from
that
request
for
information.
D
A
Research
for
results
that
are,
you
know,
deliberately
or
they're,
disingenuous
or
middle-east,
misleading,
whatever
the
adjective
is
have
been
around
for
a
very
long
time
is
but
but
you're
speaking
of
this
is
a
kind
of
a
new
problem,
and
how
is
it
new?
Is
it
different
scale,
different
techniques
or
or
really
is
it?
Is
it
just
the
same
old
stuff
and
we
need
to
look
at
it
again.
D
B
Right,
I,
I,
don't
think
there's
anything
fundamentally
new
of
though
certainly
new
newness
of
degree
in
big
data
and
larger
data,
collaborations
I'd,
say,
from
my
perspective,
I'm
very
interested
in
new
security
mechanisms
that
can
be
applied
to
help
with
the
integrity
of
the
scientific
process.
Understanding
that
security
mechanisms
are
only
one
piece
of
the
puzzle
here
and,
and
so
generally
in
part,
we
we're
not
starting
from
from
scratch
here.
We
we
want
to
understand
from
the
survey
responses
what's
already
working
well
in
our
in
the
report
that
we
will
be
producing.
B
E
Mrs.
Flores
I'd
like
to
add
a
little
to
that.
So
in
my
travels
of
talking
to
different
researchers
around
the
country
and
around
the
world,
there
are
some
that
don't
really
worry
about
cybersecurity
and
I
I
personally
find
it
troubling.
Here
they
worry
about
their
science,
and
so
part
of
the
opportunity
here
is
to
increase
awareness
and
have
them
think.
Oh,
do
I
worry
about
this.
Maybe
I.
Should
you
know
now
that
there's
a
lot
more
news
about
you
know
data
or
device
hacking
and
all
sorts
of
things
so
I
think
that's.
E
One
of
the
other
opportunities
here
is
to
increase
awareness
that
we
should
be
thinking
about
how
to
keep
the
data
trustworthy,
and
you
know
you're
right
John
there
you
know,
there's
internal
threats,
people
who
change
data,
you
know
there's
all
sorts
of
stuff,
but
the
external
threats
are
bombarding
things
more
and
so
we
than
they
used
to.
We
think
so.
This
is
an
opportunity
to
increase
awareness
of
what
they
may
need
and
then
help
them
understand.
As
Jim
said
best
practices
that
already
are
occurring
and
maybe
come
up
with
some
next
practices
too.
E
D
So,
just
to
add
a
thought,
and-
and
it
goes
right
along
with
what
Florence
was
just
saying-
and
that
is
there's
you
know
in
the
sort
of
milieu
of
open
science,
the
context
of
open
science
we've
got,
we've
got
people
that
you
know
want
all
of
science
to
be
completely
open
and
we've
got
people
practicing
open
science.
We've
got
people
that
are
you
know.
D
To
some
extent,
there
are
some
people
that
are
a
little
bit
reactionary
to
that
or
it
just
doesn't
work
with
sort
of
the
way
they
think
about
science
or
how
they
want
to
practice
science,
but
in
you
know,
sort
of
in
that
cultural
shift.
We
see
toward
open
science
I
think
right.
Sometimes
the
cyber
security
issues
and
other
security
concerns
get
get
lost
in
the
conversations
about,
and
that
and
the
movement,
a
cultural
movement
toward
open
science
and
so
I
think
right.
D
B
And
I
think
Shawn
also
talked
about
areas
where
the
the
scientist
is
being
hindered
from
sharing
their
data
or
get
access
to
data,
and
so
to
the
extent
that
we
can
increase
awareness
and
improve
the
efficiency
of
scientific
collaboration.
I
think
that's
another
big
win
for
the
working
group.
It
was
kind
of
an
analogue
to
all
is
in
the
privacy
privacy
space
as
well.