►
From YouTube: Lightning Demos - QRI - Brendan O' Brien
Description
Originally recorded during the Berlin Developers Meetings from July 9-13, 2018.
A
B5
with
Clarisse
hi
everybody,
some
of
you
who's
seen
this
already
so
I'll,
keep
it
relatively
brief,
but
I
just
thought:
I'd
demo,
what
we've
been
working
on
with
query
and
specifically
I
think
for
this
crowd,
skylark
transformations
or
one
of
the
more
interesting
bits.
Is
that
really
visible?
Okay?
A
So
what
do
we
do?
What
is
query?
We
make
a
data
set
version
control
system.
We
spent
a
lot
of
time.
We
think
that
we
can
propose
a
sort
of
document
model
for
called
data,
sets
that
helps,
standardize
and
make
document
data
set,
interchange
and
interoperation
far
better.
So
all
of
this
is
built
on
ipfs.
We
use
it
and
we
use
loopy
to
be
extensively,
but
so
today,
just
really
quickly.
A
Environment
is
also
unlike
Python,
has
no
global
interpreter
lock
so
later
on.
We
could
actually
do
parallelized
versions
of
this,
but
for
now
all
this
does
is.
This
is
a
simple
example
that
is
going
to
do
a
download
step
and
don't
worry,
we've
put
a
lot
of
thought
into
when
and
you
can
and
cannot
access
the
internet,
and
but
here
we're
gonna
do
a
little
hit
of
an
API
and
then
we're
gonna
return
some
data
and
then
that
script
is
being
specified
here
as
a
transformed,
Sky
file.
A
And
so,
if
we
flip
over
to
a
terminal
and
I'm
gonna
run,
is
that
so
I'm
gonna
run
query,
add
and
I'm
just
gonna
specify
that
yellow
file,
which
is
sort
of
the
manifest
of
the
entire
data
set
document
and
you'll
notice
here
that
I'm
going
to
give
it
some
secrets
which
are
not
stored
with
the
file?
It's
also
gonna
warn
us
that
hey
you're
giving
a
script
some
secrets
you
should
robably
know
what's
going
on,
but
if
we
hit
enter
on
this,
it
will
now
run
that
transformation
scripts.
A
Take
the
output
data
and
generate
a
data
set
smash
that
on
to
ipfs
and
make
it
available
for
us.
So
if
I
use
query,
LS,
I
can
sort
of
show
these
are
the
my
data
sets
locally
and
then
we
have
our
version
of
ipfs.
Daemon
is
query
Connect,
because
the
word
daemon
is
scary
to
some
people,
so
we
so
what
do
we?
What
do
we
have
here?
We
now
have
a
ipfs
node
running
with
a
query
both
on
layer
on
top
and
so
I
can
do
just
for
this
crud.
A
We
can
do
of
great
peers
list,
and
then
we
can
show
the
ipfs
Network.
This
is
messing
with
the
connection
manager
to
actually
make
sure
you
can
see
some
of
these
support,
the
query
protocol,
and
so
the
wonderful
new
connection
manager
API,
means
that
we
can
prioritize
connections
to
query
peers
anyways
now
to
show
you
the
actual
dataset
itself,
we'll
just
hit
our
our
local
web
UI
in
front-end.
So
we
ship
this
both
as
an
electron
app
and
as
a
web
app
that
just
moves
with
your
moves
with
your
installation
query.
A
So
here
we
can
see
the
actual
body
of
the
dataset
that
was
generated,
its
structure,
which
is
just
basically
saying
that
it's
we
don't
know
much
about
the
actual
scheme
of
the
dataset,
but
it's
an
array
any
metadata
that
was
created
with
it.
We
get
a
change
log,
so
it's
a
snapshot,
so
I
can
see
everything
that
was
coming
along
with
that
and
we
secretly
snuck
in
a
visualization
template,
and
so
we
can
actually
see.
A
A
Yeah
I
know
so
we're
gonna,
do
the
exact
same
deal
and
say
yes,
and
so
all
we've
done
here
is
we're
sort
of
thinking
like
somebody
who
doesn't
know
how
to
write
scholar,
transformations
we've
set
up
a
configuration
set
of
details
and,
in
this
case
we've
just
plopped
into
the
new
configuration
instead
of
changing
the
underlying
script
itself,
and
so,
if
I
do
query,
LS
I
can
now
see
I
have
this
new
JSI
PMS
contributes
so
now
someone
has
just.
This
is
sort
of
the
ideally
someone's
just
ripping
this
off
the
internet.