►
From YouTube: DDC Working Group 2018 Updates - André Cruz
Description
Originally recorded during the Berlin Developers Meetings from July 9-13, 2018.
A
Hello,
everyone
so
I
want
to
give
you
an
update
to
the
dynamic
data
and
capabilities
working
group
and
it
went
off
for
some
reason
back
up,
so
our
focus
in
the
second
quarter
was
he
oddities.
We
made
a
lot
of
investment
in
the
research
Inc
see
oddities.
We
actually
made
two
requests
for
proposals
and
there
are
two
links
there.
If
you
want
to
check
it
out
and
propose
a
solution
for
the
problems
that
are
there.
A
Also,
we
get
a
lot
of
involvement
with
the
research
community
with
some
universities,
some
research
centers
as
well,
and
also
we
have
researched
a
lot
of
topics
around
authentication
and
authorization
as
well.
So
in
brief,
I
will
explain
the
current
type
of
C
oddities
and
the
problems
they
have.
So.
A
It
doesn't
scale
so
there's
another
type
of
C
oddities
which
are
operation
based
and
especially
operations,
are
exchanged
between
replicas
and
they
are
very
thin,
usually
because
they
just
contained
the
operation
that
mutates
the
state
and
there
was
really
smaller
because
of
that
and
take
less
bandwidth.
Of
course,
the
drawbacks
is
that
you
have
to
have
the
entire
entire
set
of
operations
to
or
you
must
keep
the
entire
set
of
operations
for
new
peers.
A
You
have
to
deliver
the
whole
set
of
operations
always,
and
that
brings
us
to
a
slow
boot,
because
new
peer
doesn't
have
anything.
So
you
have
to
pass
always
the
entire
set
of
operations
in
a
single
time.
We
are
trying
to
solve
that
with
compaction
and
garbage
collection,
but
it's
a
very
real,
difficult,
a
difficult
problem
to
solve
and
also
sum
up
some
operations,
but
must
be
delivered
only
once
so
they
must
be.
A
They
are
not
really
even
content
on
some
implementations,
so
we
are
exploring
a
new
type
of
C
LEDs,
which
we
think
we
will
solve.
Many
of
these
problems,
which
is
Delta,
C,
LEDs
and
essentially
the
benefits
is
that
replicas
when
they
are
exchanging
messages
they
just
exchange
the
difference
between
the
states
and
also
the
plates
afterwards
are
based
in
mutations,
which
are
kind
of
similar
to
the
operations
on
operation.
A
Basically,
it
is-
and
this
is
this
gives
us
the
best
of
the
both
worlds
in
the
sense
that
the
states
passed
between
replicas
is
small
enough
and
the
afterwards
operations
are
also
small,
and
so
we
have
the
best
of
both
worlds.
So
we
made
an
implementation.
Actual
Pedro
Teixeira
made
an
implementation
of
Delta
Z.
Our
T's
in
JavaScript
is
on
that
repo
over
there
and
it
was
highly
based
on
the
work
done
in
this
paper.
A
Delta
states
replicated
data
types
by
Paul,
Sergio,
Almeida,
Valley,
shocker
and
Carlos
Vaquero
I
think
that
I
pronounced
that
correctly
and
these
new
si
oddities.
The
facilities
are
being
used
as
a
dependency
of
an
upcoming
library
on
the
pier
star
ecosystem,
which
is
the
pier
star
up
library
and
also
besides,
si
oddities.
A
This
library
aims
to
have
like
an
SDK
for
developers
to
build
applications
on
top
of
that,
and
it
will
solve
things
like
I
think
I
thinik
ation
identification,
collaborations
oversea
oddities
among
gondola
foundations
that
developers
need
also
we
have
invested
in
researching
around
Jason,
see
a
letís
by
Gonzalo
over
there,
and
we
made
some
research
there's
a
link
on
the
research
over
there.
I
found
check
it
out,
I'm
going
back
with
you
because
there's
a
lot
of
stuff
in
here
and
the
second
part
of
this
talk
is
authentication.
A
How
can
we
like
reliably
identify
users
in
a
peer-to-peer
application?
And
now
can
we
prove
ownership
over
over
operations
on
see
our
duties
right?
This
is
a
very
difficult
topic,
but
I
think
if
we
merge
three
foundations,
which
is
the
IDS
brief
I
will
claims
in
the
idea
of
we
are
able
to
accomplish
that
and
I
will
start
by
explaining
roughly
what
is
the
IDS.
The
IDS
are
essentially
identifiers
that
doesn't
don't
change
our
immutable
and
persistent
and
they
are
formed
by
a
scheme,
a
method
and
a
method.
A
Specific
identifier,
I
will
just
go
on
very
quickly
at
the
ID
maps
to
resolve
a
DD
ID
document
that
contains,
among
other
things,
public
keys
so
that
you
can
interact
with
the
the
the
entities
behind
like
we
define,
sign,
ensures
and
including
connections.
This
is
basically
the
the
motivation
behind
them.
A
You
can
disclose
verifiable
claims
so
that
the
the
relying
party
knows
your
name
knows
your
social
networks
and
anything
that
you
want
my
one
to
disclose,
and
so
this
is
all
cool,
but
is
it
was
about
today?
Can
use
them
today
and
they
are?
Are
they
accessible
for
regular
Internet
users?
That's
the
question
here:
I
made
a
quick,
quick
analysis
over
all
the
DMV
ID
methods
present
in
the
registry.
Visas
are
just
the
four
most
important
ones.
You
might
find
additional
information
in
that
link.
A
I
have
you
some
mock-ups,
but
you
can
see
this
the
complete
set
of
mock-ups
in
the
RFC
as
well.
So
what's
next
for
the
the
group
docking
group,
we
are
going
to
reshot
projects
such
as
you,
port
and
block
stack
and
explains
the
problems,
explain
the
problems
that
we
are
having
and
possibly
work
together
to
find
a
solution
for
those
problems.
We
need
to
make
a
plan
to
integrate
the
anything
about
identity
and
authentication
on
pier
star
up,
which
is
a
library
I
will
be
speaking
tomorrow
and
also.