►
Description
This talk was given at IPFS Camp 2022 in Lisbon, Portugal.
A
Hi
I'm
going
to
talk
about
the
question.
What
would
be
if
the
European
open
science
Cloud
would
run
on
lip
peer-to-peer
and
that
Vision
talks
of
the
European
science
Cloud
actually
does
not
run
on
the
peer-to-peer
and
I
will
first
explain
what
the
European
open
science
cloud
is.
So
the
aim
of
the
European
open
science
cloud
is
to
create
a
web
of
fair
data
and
services
for
research
in
Europe,
with
the
goal
to
improve
the
situation
for
researchers
across
Europe,
and
it
is
it's
a
quite
large
project.
A
It
started
2015
and
there
that
was
when
the
first
idea
was
developed,
and
then
there
was
a
prototyping
phase
which
was
funded
by
was
250
million
euros
from
the
European
commission
and
now
there's
the
next
phase
from
until
2000
2030,
where
there's
a
public-private
partnership
was
a
total
investment
of
one
billion
Euro,
and
this
is
to
build
this
European
open
science
Cloud.
There
are
a
lot
of
people
involved
and
a
lot
of
countries
involved
from
Europe,
but
also
observers
from
not
within
Europe.
A
So
what
are
the
goals
of
the
of
the
European
open
science?
Cloud?
The
goals
are
that
there's
relevant
research
data
with
Rich
metadata
available
to
all
researchers,
that
we
have
a
repository
and
we
have
access
and
reusable
research
software.
Of
course,
there
needs
to
be
storage
for
research
data.
We
need
to
have
compute
power,
Access,
Control,
search
capabilities
and
something
like
automated
referencing
and
much
more.
What
researchers
need
to
work
in
a
digital
laboratory
all
across
Europe.
A
So
this
has
been
worked
out
in
details
in
the
Strategic
research
and
Innovation
agenda.
So
it's
not
a
top-down
approach.
That's
the
European
commission
says
we
will
build
the
European
open
science
Cloud
according
what
we
think
is
bad.
It's
is
best.
So
there
has
been
an
inclusive
Pro
process
to
get
information
from
from
all
the
researchers
across
Europe
what
they
want
and
what
the
goals
should
be
and
yeah.
A
There's
there's
now
the
the
implementation
phase,
which
is
which
is
precise
in
a
more
multi-annual
roadmap
so
and
we
are
currently
in
the
face
of
developing
an
infrastructure
for
Federated,
Federated
research,
infrastructure,
yeah
and
then,
and
starting
from
2023
the
as
this
research
infrastructure
should
be
productive
and
we
want
to
generate
added
value
and
then
from
2020-25
there's
the
idea
to
for
an
expansion
to
develop
impact
from
open
science
to
the
overall
community.
A
So
what
is
the
situation
today?
What
do
you
think?
Where
do
researchers
store
their
data?
Do.
Do
researchers
in
particular
store
their
data
on
ipfs?
A
A
So
it
is
not
that
prevalent,
I
guess
so.
Coming
to
the
next
questions,
do
do
researchers
use
content
hashes
to
link
their
research
artifacts,
so
you
would
probably
guess
that's
not
the
case,
but
at
least
there
is
a
one
initiative
that
that
uses
also
the
the
concept
of
content
hashing,
and
that
is
a
software
Heritage.
A
So
lady,
they
their
aim
is
to
preserve
their
the
source
code
of
research
software
and
they
thought
for
quite
a
long
while
how
to
preserve
all
the
source
code,
which
was
produced
by
researchers,
to
support
their
research
results
and
they
figured
out
that
that
a
hash
or
an
intrinsic
identifier
is
best
to
address
versions
of
research
software.
So
they
use
this.
A
This
hash,
which
was
which
you
get
from
the
from
the
software
source
code
to
directly
address
the
research
software,
and
so
they
have
a
long-term
preservation
idea
and
use
those
hashes
to
store
the
the
research
software
for
for
the
long
term
and
when
you,
when
you,
for
example,
use
the
research
software
in
a
paper,
you
can
cite
the
Slash
and
then
then
you
will
get
to
that
version
of
the
research
software
forever
and
currently
they
do
it
with
local
storage
and
they
have
UNESCO
funding,
which
is
the
best
guarantee.
A
You
can
currently
get
that
that
this
research
software
will
stay
on
with
their
online
in
the
future.
Yeah
I,
when,
when
I
saw
that
I
think
I
was
thinking
that
would
be
a
good
match
to
live
peer-to-peer
and
the
and
the
ipfs
world
and
I
thought
what
can
I
do
about
that
and
yeah
eventually
I'm
working
together
with
with
software
Heritage
as
part
of
the
of
the
fire
core
for
Eos
project?
A
That
is
one
of
those
many
projects
that
are
funded
by
the
European
commission
to
to
to
roll
out
the
European
open
science
cloud,
and
we
are
doing
a
first
experiment.
So,
for
example,
there's
a
software
Heritage
mirror
so
that
that
there
are
at
least
two
institutions
that
store
the
source
code
of
the
research
software
and
we
we
link
this
This
research,
this
this
content
hashes.
We
also
use
in
other
institutes,
especially
in
my
Institute,
which
is
about
mathematical,
research
and
mathematical,
research,
objects
and
mathematical
research
Publications.
A
A
A
So
if
you
want
to
apply
for
compute
resources,
you
have
to
talk
to
your
admin
and
you
have
to
fill
out
a
form
and
you
have
to
wait
forever
and
do
a
lot
of
paperwork,
but
you
you
could
use
crypto
incentive
mechanisms
to
to
simplify
this
procedure
and,
of
course,
there's
also
when
you
are
when
you
were
in
the
other
talks
of
of
the
conferences.
You
see
that
there's
approaches
that
are
that
are
decentralized
approaches.
A
This
questions,
which
are
the
expectations
of
the
the
European
open
science
Cloud,
so
that
you
can
participate
in
the
European
open
size
Cloud.
Even
if
you
are
not
a
big
Research,
Institute
or
or
a
country
that
says
it
wants
to
participate.
So
we
would
somehow
contribute
to
the
democratization
of
the
European
open
science
cloud
yeah.
A
My
vision
for
the
for
the
future
is
to
roll
out
decentralized
services
and
to
test
and
to
research
and
to
have
experiment
if
our
services
that
currently
run
centralized
from
centralized
system
would
also
run
at
least
for
for
testing
in
a
decentralized
environment.
A
A
So
this
has
so
my
Institute
is
is
quite
old,
so
we
have
a
long
history
and,
for
example,
in
the
second
world
war,
when,
when
I'm,
it's
a
it's
a
German
Institute,
it
was
a
little
bit
fishy
that
all
this
mathematical
Publications
were
reviewed
and
organized
by
the
Germans.
So
the
American
mathematical
Society
decided
to
create
the
second
institutes.
Now
now
we
have
at
least
two
institutes
that
do
the
same
thing,
but
it
was
the
same
idea
to
avoid
this
Central
service
provider.
So
that's
that's
one
big
Advantage.
A
We
we
see
and
when
we
look
also
at
the
situation
with
Russia
today,
it
is
always
good
to
not
depend
on
one
particular
service
provider.
We
want
to
separate
the
service
and
the
application
layer
that
we
use
only
some
crypto
value
to
to
exchange
that,
and
we
also
want
to
allow
for
optimized
executions
which,
which
you
cannot
do
if
you
have
a
centralized
server
that
way
yeah.
So,
in
the
end,
I
want
to
decentralize
my
services
and
yeah
test
test
approaches
which
for,
for
example,
Clinton
can
we
add
ipfs
as.
A
Ipfs
storage,
as
one
options
in
the
in
the
idea
of
the
European
science
Cloud
multi-cloud
project,
so
that
we
have
ipfs
as
one
storage
option
so
that
people
can
transition
seamlessly
from
the
current
infrastructure
to
to
the
decentralized
infrastructure,
is
the
same
as
for
a
software,
Heritage
mirror
with
lib
ipf
as
technology,
and
also
we
want
to
limit
or
understand
and
experience
where
we
hit
the
limits
of
current
decentralized
technology.
A
So
let
me
end
here,
and
since
this
is
a
really
a
new
version,
I
just
wanted
to
ask:
what
do
you
think?
Is
it
a
match,
Now
swipe,
left
or
right?
Thank
you.
Give
him
a
hand.
Everybody.