►
From YouTube: Slingshot 2.2 - Closing Ceremony
Description
Visit https://slingshot.filecoin.io/ to learn more about Slingshot, and filecoin.io to learn more about Filecoin.
Join us as we celebrate the end of Slingshot Phase 2.2! We’ll recap the past month, present the final project rewards, host a live FAQ and give competition updates as we transition into Slingshot Phase 2.3.
A
We'll
do
a
little
overview
of
the
total
storage
and
the
deal
progress
from
there
we'll
transition
into
phase
2.2.
So
we
will
have
dinesh.
He
will
be
talking
about
the
judging
and
the
reward
winners,
and
then
we
will
do
a
slingshot
2.3
overview
where
pujo
will
be
taking
us
through
where
we're
going
to
be
taking
slingshot
over
the
next
few
months.
A
And
then
we
will
finish
off
with
a
live
q
a
here.
So
if
you
have
any
questions,
feel
free
to
stick
around
and
and
we'll
be
here
to
answer
them
so
again,
congrats
on
completing
case
2.2.
We
had
a
really
really
successful
phase,
as
you
see
from
the
graphs
here,
a
huge
spike
in
data
which
is
really
really
awesome
to
see
so
congrats
to
everyone.
A
B
So
first
I
wanted
to
talk
a
little
bit
about
documentation
so,
as
you
know,
documentation
to
for
providing
this
data
so
that
other
individuals
in
the
ecosystem
can
download
and
use
this
data
is
a
really
really
important
part
of
slingshot.
It's
one
of
the
the
primary
mandates
for
the
competition
and
we've
seen
some
kind
of
issues
in
a
few
projects.
I
wanted
to
kind
of
reiterate
some
of
some
tips
to
continue
to
make
your
documentation
better.
B
One
is
always
remember
to
publish
the
reference
data
source
right,
the
the
credibility
of
where
is
this
data
from
sometimes
it's
through
a
government
organization.
Sometimes
it's
through
amazon
or
google.
Sometimes
it's
through
another
authentic
source.
So
you
don't
need
to
duplicate
the
documentation
from
the
source,
but
it's
really
important
to
know
that
this
dot,
that
this
information
came
from
a
source
that
somebody
can
actually
go
ahead
and
validate
and
see
that
that's
the
data
that
they
actually
want.
B
The
number
of
people
actually
referenced,
the
slingshot
website
and
the
deals
dot
csv
that
people
uploaded,
but
that's
not
really
always
a
valid
option.
It
assumes
that
the
slingshot
website
will
be
available
which
may
not
always
be
true
or
maybe
in
a
different
form
from
the
from
the
original
directory.
B
B
A
couple
of
other
things
that
made
retrieval
specifically
challenging
is,
as
you
know,
really.
Large
files
often
need
to
be
split
up
into
multiple
chunks,
and
so
when
a
data
source
is
larger
than
32
gigs
or
64
gigs
larger
than
a
sector,
it's
very
common
practice
to
split
those
up,
and
the
number
of
teams
did
that
very
successfully.
B
But
from
a
usability
side,
how
does
the
user
actually
reassembled
those
files?
Some
documentations
were
clearer
than
others,
but
something
to
really
pay
attention
to.
If
I
don't
actually
know
the
format
of
the
file
or
how
many
pieces
there
are.
How
do
I
reassemble
these
right
if
it's
split
up
into
eight
files,
but
I
only
download
six
of
them.
How
do
I
actually
make
sure
that
I
have
the
entire
contents
of
the
file?
Is
that
through
a
hashing
strategy?
Is
that
through
another
strategy?
B
In
addition,
sometimes
I
might
actually
not
might
find
that
a
minor
is
undoubtable.
So
how
do
I
find
alternative
minors
for
a
particular
piece
of
content
and
then
on?
The
flip
side
is
that's
for
really
large
files.
There's
also
the
case
when
you're
retrieving
small
files,
sometimes
files
are
a
few
k
or
really
or
a
few
megabytes,
and
those
are
actually
combined
into
a
archive
or
a
tar
file,
and
that's
a
really
common
strategy
makes
a
lot
of
sense.
But
how
does
the
user
find
out?
B
We
saw
dramatic
improvements
from
phase
one
for
2.1
and
2.2,
but
want
to
continue
to
reinforce
the
idea
that
it's
great
the
data
is
stored,
but
it's
even
better
if
it's
usable
and
retrievable.
So
let's
go
into
the
judging
side
of
it.
We
actually
looked
at
the
submissions
from
four
different
vantage
points,
and
I
wanted
to
go
through
the
the
details,
so
so
teams
understand
where
the
what
judging
criteria
was
used.
The
first
was
deal
validation.
B
This
is
actually
purely
an
automated
process
where
we're
actually
only
counting
the
first
10
copies
as
per
the
rules
for
for
submitted
deals.
So
the
same
content
can't
be
submitted
more
than
10
times,
and
we
also
negated
deals
that
don't
contain
a
file
name.
If
I
have
a
16
gigabyte
cid,
but
I
don't
know
the
type
of
file
or
or
how
it
what
that
content
represents
it.
It's
not
usable
from
a
from
a
content
perspective,
so
I
wanted
to
so
that
was
the
actual
first
check.
B
The
second
side
was
actually
the
the
documentation
validation,
which
I
touched
on
in
a
little
bit
of
detail,
but
really
checking
for.
Can
a
third
party
independently
download
and
use
this
data,
and
so
some
of
the
tips
from
the
previous
slide
touches
on
the
criteria
used
from
that
perspective,
a
third
part
was
actually
on
retrieval
validation,
so
we
spot
checked
a
percentage
of
the
submitted
deals
in
order
to
make
sure
that
the
content
was
actually
retrievable
and
in
general,
retrieving
retrievability
was
a
challenge.
B
Some
minors
were
not
dialable,
some
were
offline
and
others
failed
for
for
a
variety
of
variety
of
reasons.
We
recognize
that.
That's
not
always
the
slingshot
participants
fault.
Sometimes
it's
other
reasons,
but
puja
will
address
this
in
a
little
bit
more
detail.
But
retrieval
is
a
big
mandate
of
why
slingshot
exists
right.
B
We
want
to
provide
the
world
with
really
really
useful
data
sets
and
how
do
we,
and
if
it's
not
retrievable
the
the
intent
or
the
purpose
of
the
competition
is,
is
heavily
diluted
and
then
the
fourth
aspect
is:
we
did
receive
complaints
from
the
from
a
variety
of
community
members
for
self-dealing
and
other
misuse
of
the
rules,
and
so
those
were
analyzed
and
processed
in
the
judging
scoring.
So
I
wanted
to
comment
on
that.
B
C
C
We
stored
so
many
more
petabytes
of
data
this
this
last
month
than
in
all
the
months
combined
so
great
work
to
everyone
there,
and
so
we
wanted
to
spend
a
little
bit
of
time
talking
about
the
next
phase
of
slingshot
2.3,
as
we've
been
calling
it
and
the
niche
covered
a
lot
of
this.
But
you
know
we
feel
like
now
that
it's
been
a
few
months
that
we've
been
in
sync
shot
phase.
C
So
if,
if
you
can
go
to
the
next
slide,
there
are
a
few
main
goals
for
for
this
competition,
and
the
first
is
to
you
know:
do
this
work
of
preserving
the
most
valuable
data
sets
that
we
know
of
that
are
public
publicly
accessible
on
the
file
coin
network
and,
as
we've
seen
over
the
last
few
months,
we've
totally
crushed
this
goal.
There
have
been
several
petabytes
of
really
valuable
data
sets
stored
in
the
network.
C
The
second
goal
for
this
competition,
as
the
niche
mentioned
is
sorry,
I
think
the
slides
went
away,
and-
and
maybe
am
I-
the
only
person
who's
saying
that
okay,
great
we're
back.
Thank
you
awesome.
So
a
second
goal,
as
as
denise
mentioned,
is
about
really
about
retrieval.
Fundamentally,
so
when
we
say
that
we
want
all
of
these
data
sets
that
are
being
stored
to
the
network
to
be
accessible
and
explorable
by
anyone.
C
You
know
when
we
started
this
competition.
It
was
actually
before
the
file
coin
made
that
launch,
and
so
much
has
changed
in
the
last
few
months.
Storage
deals
are
much
more
successful
over
overall,
just
across
the
board.
This,
like
competition,
has
been
a
really
powerful
force
for
debugging.
Some
of
the
you
know
open
issues
within
the
network
and
client
and
storage
client
software
as
well.
C
In
some
cases
we've
seen
like
100
terabyte
data
sets
they'd
have
to
reconstruct
the
whole
100
terabyte
data
set
in
order
to
unzip
it
and
actually
get
to
the
the
data
that's
stored
within
the
directory,
and
so
obviously
that
that
also
is
like
not
usable.
Most
people
don't
have
hundred
terabyte
machines
that
they
just
have
sitting
around.
C
Okay,
awesome
thanks,
okay,
so
a
few
other
common
issues
that
we've
seen
so
I
think
you
know,
the
filecoin
network
has
many
different
kinds
of
cids
and
there
may
be
some
confusion
about
what
different
cids
represent.
We've
seen
some
folks
when
they're
uploading
their
deal
information,
they
upload
dli
deal
ids,
which
are
the
cids
that
are
prefixed
with
bafy,
and
so
that
is
not
quite
the
same
thing
as
a
payload
cid.
So
people
can't
actually
use
these
cids
to
retrieve
the
data
back
out
of
the
network.
C
It's
not
the
correct
cid!
So
that's
kind
of
one
issue.
The
other
is
when
folks
have
uploaded
pcids,
which
are
prefixed
with
baga.
So
really
what
you're
looking
for
with
the
payload
cid
is
a
hash
that
starts
with
capital,
q
m
and
then
a
number
of
characters.
After
that,
I
think,
if
you're
doing
offline
data
transfer,
sometimes
the
payload
cids
begin
with
like
an
max
or
something
something
along
those
lines,
but
generally
like
that's
what
you
kind
of
want
to
look
for.
C
Those
are
the
cids
that
are
on
the
right
track
and
then
this
documentation,
you
know,
there's
just
a
lot
of
a
lot
of
the
documentation,
has
been
extremely
good
and
huge
improvements
that
we've
seen
from
teams
between
2.1
and
2.2.
But
as
denise
mentioned,
there
are
still
a
number
of
things
that
could
make
documentation
much
more
usable
and
so
really
as
we're
entering
the
next
phase.
We
encourage
you
to
think
about.
You
know
if
you're
just
thinking
about
this
from
the
perspective
of
a
data
user.
C
What
are
the
things
that
they
would
really
want
to
see
if,
like?
If
you
yourself,
you
know,
were
looking
at
someone
else's
documentation,
because
you
wanted
to
download
this
data
and
use
it
yourself.
What
would
you
need
to
see
in
order
to
actually
make
that
useful
for
you,
and
we
do?
You
know
grade
the
documentation
when
we're
doing
the
judging
process.
So
if
you
have
great
documentation
it,
it
gets
reflected
in
the
final
results
and
awards
as
well.
C
C
So
some
of
the
immediate
changes
we'll
be
making
to
the
2.3
competition
we're
definitely
going
to
keep
storage
rewards
as
a
as
a
concept,
that's
kind
of
the
bulk
of
the
where
the
rewards
comes
from,
as
denise
mentioned.
I
think
in
this
round.
It
was
like
125
000
fill
plus
like
some
of
the
leftovers
from
phase
2.1,
which
made
it
more
than
140
000
coins.
C
The
other
thing
is
we're
really
encouraging
that
if
you
do
have
any
question,
any
technical
help
requests.
Please
use
lotus
issues
and
discussions.
We
have
a
huge
community
support
team
on
the
lotus
side,
who
is
monitoring
these
discussion
forums
and
making
sure
that
they
can
get
you
your
help
as
quickly
as
they
as
possible?
Essentially,
so
we
encourage
folks
to
start
using
the
discussions
on
github.
C
There
are
some
data
sets
that
in
2.2
there
were
many
teams
that
were
storing
these
data
sets
and
one
particular
data
set
is
landsat
8..
So
for
the
next
phase
of
the
competition
we
will
probably
be
removing
a
few
of
the
data,
sets
that
now
have
many
copies
on
the
filecoin
network
and
encourage
you
to
go
out
and
search
for
more
new
data
sets
that
are
just
are
just
different,
so
we
can
increase
the
diversity
of
what
is
being
stored
through
the
competition.
C
C
Although
you'll
you'll
see
that
you
know,
no,
no
team
actually
got
20
of
the
total
rewards.
That
was
just
in
order
to
kind
of
try
to
distribute
the
rewards
more
fairly
across
all
competitors,
but
in
this
next
round
we'll
be
reducing
this
cap
to
10.
So
no
individual
team,
no
matter
how
much
data
they
store,
we'll
get
more
than
10
of
the
total
price
pool,
and
just
remember
you
know.
C
I
think
there
may
have
been
some
confusion
about
this
particular
note
last
in
the
last
period,
but
everyone
who
participates
is
always
in
slingshot
always
gets
something.
So
I
think
some
folks
were
confused
a
little
bit
about
how
exactly
this
role
works,
but
everyone
always
who
participates,
gets
something.
We
just
distribute
the
rewards
proportionally
among
all
participants
based
on
how
much
data
that
they
have
stored,
and
then
this
other
one
is
kind
of
interesting.
C
C
A
lot
of
people
are
finding
very
clever
ways
to
basically
store
data
with
their
own
miners
or
organizations
that
they
know
very
well
and
possibly
are
even
you
know,
owners
of-
and
so
you
know
this
this
rule
we
just
decided
for
the
next
round.
It
wasn't
really
serving
us
anymore
to
keep
this
rule
in
place,
so
we're
going
to
be
removing
this
rule
for
2.3,
but
you
know
we're
still
going
to
continue
to.
We.
C
And,
lastly,
like
you
know
just
thinking
like
how
do
we,
how
do
we
get
other
people
to
actually
build
on
this
data?
We're
stewing
on
some
ideas,
for
what
a
slingshot
hackathon
could
look
like
a
slingshot
hack
track
will
be
coming
soon,
but
if
you
have
any
ideas
for
slingshot
2.3
about
how
we
can
change
the
way,
the
competition
is
structured
to
enable
hackathons
and
other
application
developers
to
build
on
the
data
more
easily.
We
would
welcome
your
suggestions
there
as
well,
so
highly
encourage
folks
to
take
a
look
at
the
repo.
C
This
is
your
time.
You
know
for
the
next
two
weeks
like
we're,
we're
really
opening
the
door
to
anyone
who
has
ideas
about
how
to
design
this
you've.
Many
of
you
have
been
participating
from
the
beginning
of
slingshots.
You
I'm
sure
you
have
lots
of
ideas
about
how
this
could
be
done
differently,
and
so
we're
really
really
just
opening
this
now
to
to
anyone
in
the
community
to
participate
and
share
your
ideas.
C
We'd
love
to
hear
them,
and
then
the
other
thing
to
note
too,
is
because
we
are
sort
of
pausing
so
that
we
can
like
regroup
and
maybe
change
the
competition
structure
slightly.
You
can
keep
storing
data
over
these
next
two
weeks,
but
it
may
not
count
for
for
2.3.
If
you
know
it
doesn't
fall
within
the
new
rules,
so
it's
kind
of
like
a
bit
of
just
beware.
You
can
definitely
keep
storing
your
data
on
the
network
and
as
long
as
you're,
you
know
generally
doing
sensible
things
like
most
likely.
C
It
will
account
for
2.3,
but
we're
not
going
to
make
any
guarantees
at
this
point,
because
we
really
want
to
preserve
the
flexibility
to
design
like
the
best
competition
possible,
and
so
with
that
in
mind,
the
new
dates
for
folks
to
keep
in
mind
we'll
be
updating
the
website
with
all
of
this.
But
the
reward
deadline
for
2.3
will
be
march
24th,
so
we're
basically
just
pushing
back
all
of
the
dates
by
two
weeks.
C
Okay
and
then
the
last
thing
we
wanted
to
announce
today
is
that
you
know
we've
seen
a
huge
amount
of
really
great
goodwill
in
the
community
with
folks
helping
each
other
out.
There
have
been
some
like
really
just
awesome
people,
you
know
answering
each
other's
questions
linking
each
other
to
the
relevant
documents,
sharing
information
about
events
that
are
coming
up,
and
things
like
this
also
some
even
just
some
of
the
like
complaints
that
we've
seen
people
submit,
is
also
just
a
really
good
example
of
community
stewardship.
C
More
official
capacity
can
help
moderate
and
also
when
it
comes
time
to
final.
Judging
for
the
competition,
you
know
we'll
be
working
with
community
stewards
to
help
get
their
input
on
the
final
results
as
well.
So
we'll
announce
more
details
about
this
application
process
next
week
in
the
slingshot
announcements
channel
on
file
point
slack,
and
that
is
it.
Thank
you
so
much
for
participating
in
slingshot
and
all
of
the
hard
work
that
you've
done
so
far.
C
C
C
C
Okay,
so
kimmy
asks:
why
are
we
removing
the
self-sealing
rule
again
yeah?
So
it's
a
really
good
question.
This
one
is
really
tricky.
The
the
honest
answer
is
that
it's
self-sealing
is
like
a
really
really
difficult
thing
to
properly
enforce
and
when
we
looked
into
it
a
little
bit
with
some
of
the
heuristics
that
some
community
members
had
provided,
it
basically
became
clear
that
kind
of
the
majority
of
folks
in
the
competition
were
finding
very
clever
ways
to
subvert
this
role.
C
To
be
honest-
and
we
didn't
want
to
to
penalize
for
that
because
you
know
that
just
was
didn't,
feel
fair
and
felt
like
that
was
not
really
in
the
spirit
of
the
competition,
so
we
decided
to
just
open
open
it
up.
I
mean
I
think
most
people
are
kind
of
doing
this
anyways
and
it
really
initially
was
kind
of
to
help
the
folks
who
were
having
a
lot
of
really
hard
time
finding
minors
to
store
with.
C
That's
called
the
minor
x
program
and,
if
folks
are
having
trouble
finding
reliable
minors
to
store
with
or
having
minors
to
store
with,
there
are
there's
like
a
batch
of
you
know:
30
miners
who
are
participating
in
this
program
that
you
can
you
can.
You
know,
ask
us
for
introductions
to
or
to
put
us
in
the
channels
where
they
where
they
exist,
and
so
they
are
constantly
looking
for
data
to
accept
and
deals
to
accept,
so
that
that's
another
way
to
also
make
sure
that
you
have
other
minors
to
store
with.
B
I
was
just
going
to
remind
folks
that,
like
the
the
minor
fellowship
is
a
great
way
to
find
minors,
and
so
the
the
purpose
is
really
to
make
the
data
available
to
everybody
and
whether
it's
self-sealed
or
otherwise,
and
so
yeah,
the
in
the
past
there
weren't
as
it
was
harder
to
find
miners.
I
think
the
minor
x
program
actually
improves
that
significantly.
So,
even
if
you
don't
have
minors,
it
should
be
easy
to
find
minors.
C
No,
so
what
we
did
actually
so
again,
as
I
mentioned,
we
found
that
the
vast
majority
of
participants
were
were
doing
this
essentially,
but
so
we
didn't,
we
didn't
think
it
was
fair
to
like
penalize
essentially
the
entire
competition,
but
what
we're
doing
instead
is
that
we
gave
rewards
to
folks
who
we
basically
thought
were
like
behaving
honestly
and
so
most
of
the
folks
who
we
think
didn't
do
the
self-sealing
will
see
like
a
boost
in
their
rewards.
It's
about
like
a.
C
I
think,
it's
a
10
boost
that
we
gave
and
then
there's
a
question
about
the
30
rule
and
I
think
yeah
also
related
questions.
What
about
the
percentage
store
with
the
same
miner?
Can
it
be
100
with
one
miner
and
so
no
we're
not
changing
the
30
rule.
So
no,
even
if
you
decide
to
store
data
with
your
own
miner,
it
still
should
be.
It
still
needs
to
be
less
than
30
percent.