►
Description
This talk was given at IPFS Camp 2022 in Lisbon, Portugal.
A
All
right,
thanks
for
thanks
for
waiting
there,
we'll
we'll
jump
in
I
know
we
got
a
lot
to
cover
today,
so
yeah,
just
to
start
off.
My
name
is
Steve
lepke
I'm,
one
of
the
engineering
managers
here
with
protocol
labs
in
the
engineering
research
group,
I'm
called
Big
LEP,
at
least
on
the
Internet,
or
at
least
on
GitHub
and
real
quick,
sometimes
I
get
asked.
Where
does
that?
Come
from
that's
a
high
school
nickname
for
my
best
friend
who's?
A
Now
my
brother-in-law,
that's
a
different
story,
but
I
was
a
very
late
bloomer
in
high
school
and
so
I
think
it
emerged
from
an
eight
or
eight
inch
growth
spurt
later
on,
and
it
has
stuck,
but
honestly,
I'll
respond
to
anything.
So
again,
just
think.
Thanks
for
addressing
me,
I
yeah
I
get
to
focus
on
some
of
the
work
here.
Protocol
Labs,
that's
not
as
directly
connected
to
filecoin.
A
So
this
is
things
like
ipfs
and
the
P2P
and
ipld
like
I,
said
I'm
one
of
the
engineering
managers
and
that
really
translates
to
getting
to
be
an
encourager
and
wingman
to
a
lot
of
the
great
people
that
that
we
see
around.
Since
I
have
your
attention
a
fun
fact,
best
acquisition
of
2022
was
picking
up
a
tandem
bike.
I
can
get,
and
I
can
throw
a
third
wheel
behind
that.
Here's,
a
picture
of
me
as
my
kids,
somewhere
in
our
neighborhood.
That's
how
I
get
everyone
to
school.
A
So
if
you
want
to
get
on
that
train
happy
to
happy
to
share
I
live
in
Seattle
in
the
Pacific
Northwest
of
the
United
States,
but
I
am
a
child
of
the
promised
land
I.E
Canada
to
the
north,
and
certainly
when
the
Olympics
are
on
I,
believe
Maple
Leafs.
So
in
terms
of
what
we're
trying
to
do
in
this
presentation,
I
just
want
to
give
visibility
to
what's
happening
on
the
protocol
lab
side
with
lip
P2P
like
Max,
said
we're
not
the
only
party
here
but
we're
one
of
the
parties.
A
We
also,
hopefully
give
Clarity
to
what
you
can
expect
from
from
that
team,
and
you
know
at
the
end
we'll
talk
a
little
bit
more
about
the
road
map
and
at
least
how
we're
planning
to
apply
our
efforts
towards
the
P2P
and
really
a
presentation
like
this
is
a
good,
forcing
function
to
make
sure
that
we
actually
document
some
of
the
things
that
I'm
going
to
talk
about,
so
it
gets
in
a
place
more
durable.
So
thanks
thanks
for
letting
me
use
this
as
an
excuse
to
do
that.
A
Okay,
some
of
the
people,
you
probably
might
recognize
some
of
these
faces.
Many
of
them
are
here.
You
know
on
the
go
side,
we
have
Martin
who
will
be
presenting
later
today.
He
leads
that
effort,
there's
Julian,
who
wasn't
able
to
make
it.
We
have
Marco,
who
we
already
acknowledged,
is
not
here
Marco's
a
special
case
and
that
he
makes
commits
across
multiple
implementations
most
recently
and
go
in
JS,
but
has
contributed
to
rust
as
well,
and
maybe
some
others
on
the
JavaScript
side.
A
We
have
Alex
or
aikengrain
who's,
already
done
a
couple
of
great
presentations
here
this
weekend
and
he
is
supported
with
by
some
Engineers
from
chainsafe,
particularly
Cayman
and
Marin,
who
involved
in
all
of
our
maintenance
efforts
and
then
on
Rust.
We
have
Max,
we've
got
Elena,
Thomas
Zhao,
again
multiple
people
there.
So
those
are
our
core
engineers.
A
So
that's
that's
the
people,
but
before
moving
on,
we
do
want
to
make
it
clear
like
that.
May
have
looked
like
a
lot,
but
really
a
lot
more
is
needed.
More
people
are
welcome.
That
doesn't
you
know
that
can
mean
employment
with
us,
but
there's
also
a
lot
of
opportunity
outside
of
protocol
Labs.
So
if
you
have
great
ideas
or
if
you
want
to
just
even
be
involved
in
the
maintenance
efforts
like
there
is
there's
various
funding
options,
please
please
come
and
talk
with
us.
A
As
you
see
in
this
presentation,
one
of
the
key
one
of
our
key
goals
is
to
be
bringing
more
people
in.
We
can't
have
you
know
the
groups
of
people
on
that
slide
and
others
in
the
ecosystem.
Being
the
only
people
who
are
holding
up
the
project,
we
need
more
okay,
so
the
team
at
lib
in
the
P2P
focused
on
sorry,
the
team
in
Andres
PL
Andrew,
has
focused
on
the
P2P
like
what's
our
purpose,
this
is
my
mental
model.
This
isn't
like
vetted
with
the
team.
A
I
think
it'll
stand,
but
you
know
this
probably
needs
some
refining,
but
it's
really
to
ensure
that
lip
P2P
is
a
solid
foundation
for
web3
to
build
on
today.
Okay,
great,
take
care
of
today,
but
also
making
sure
we
can
keep
up
with
the
load,
bearing
that
we
know
is
coming
tomorrow
as
more
and
more
applications
move
to
you
know
into
peer-to-peer
networks.
A
So
I
don't
want
to
be
like
the
bridge
on
the
left
which,
which
failed
I
want
to
be
able
to
be
the
stress
tests
of
ridges,
where,
as
you
put
concentrated
load
that
we
can
handle
that
and
that's
you
know
that
that's
really
what
we
need
to
work
towards,
so
to
motivate
that
really
encourage
folks
to
watch
a
talk
that
Juan
gave
in
Paris
back
in
April
kind
of
talking
about
the
long-term
view
of
Lupita
P
again.
This
is
like
40
or
50
minutes
of
really
good
content.
A
I
will
do
a
quick
speed
run
through
a
couple
of
the
things
in
it,
but
he
does
a
much
better
job
of
presenting
it
again
that
one's
really
worth
watching
for
kind
of
how
I
arrive
at
what
we're
doing
so.
First
thing
to
be
first
thing
to
call
out
right
is
lip.
P2P
is
really
critical
for
actualizing
a
lot
of
the
values
that
we
have
in
web3.
A
You
know
there
are
a
set
that
have
been
like
shared
amongst
the
community
and
as
soon
as
you
want
to
start
taking
action
on
a
lot
of
those,
you
need
the
P2P
to
make
it
happen.
So
we're
really
at
the
foundation
there-
and
you
know,
as
Max
already
shared
like
lip
P2P-
is
underpinning
a
lot
of
these
great
ecosystems,
and
so,
even
if
projects
don't
realize
that
they're,
depending
on
the
P2P
by
building
on
ethereum
or
ipfs
or
polka
dot
or
filecoin
like
you
now
have
a
lip
P2P
dependency.
A
So
we
have
a
lot
of
impact
with
this
project.
You
know
the
third
Point
here
is
that
to
do
well-executed,
lipid
to
P
or
sorry
well
executed,
open
source
work.
It
takes
a
lot
of
effort
across
a
lot
of
areas.
You
know
some
of
those
are
called
out
on
the
right
and
you
can't
just
do
some
of
those
you
got
to
be
doing
applying
effort
across
all
of
them
and
so
in
lipidus
piece
case.
A
There
is
definitely
some
areas
where,
like
we
need
extra
help,
you
know
we're
good
on
the
funding
side,
but
a
lot
of
it
has
to
do
with
people
coming
in.
You
know
in
maintenance
in
development
and
then
all
the
coordination
that
goes
on
with
how
you
bring
a
large
group
of
people
together.
So
a
lot
more
is
needed
there,
and
so
we
kind
of
used
in
that
talk.
A
But
you
know
if
you
invert,
that
and
kind
of
draw
what
lip
P2P
is
like
today,
it's
really
wide
and
there's
not
very
many
people
down
here
at
the
bottom,
and
that
is
one
of
the
key
things
we
need
to
increase.
So
obviously
we
want
to
have
more
people
using
the
PDP
for
sure,
but
also
increasing
the
rate
at
which
more
people
are
contributing
to
the
base
so
that
we
can
take
on
take
on
that
load.
A
The
team
spends
a
good
amount
of
effort
in
a
power
empowering
other
people
to
contribute
to
those
implementations
or
supporting
and
bootstrapping
other
language
implementations
as
they
come
about.
Some
of
what
defines
this
group
is.
There
are
a
lot
of
lip
PDP
knowledge
experts.
They
have
like
the
historical
knowledge
or
some
domain
knowledge
that
are
in
the
group
and
are
generally
just
used
in
making
connections
across
groups,
which
is
great,
and
you
know,
Max
hit
this
as
well.
A
But
one
one
thing
really
want
to
make
clear
is
like
the
goal
the
team
is
not
shooting
to
be
is
not
sorry.
Their
goal
is
not
to
be
The
Gatekeepers
or
sole
decision
makers
around
the
P2P.
This
is
meant
to
be
a
open
communal
affair,
so
some
of
the
some
of
the
values
for
the
team
against
some
of
these
are
still
in
progress
and
being
actualized.
But
since
we're
talking
about
the
team,
I'll
call
them
out
one
specs
first
I
think
with
P2P
project.
A
Does
this
better
than
some
of
the
others
you
know
surrounding
it,
but
want
to
keep
leaning
into
that
in
order
for
the
team
to
scale
it
needs
to
be
responding
with
URLs,
we
can't
have
big
long-winded
discussions
buried
in
a
chat
room.
Ideally,
that's
text
that
is
getting
put
into
some
durable
place
that
we
can
respond
with
URLs
first.
So
that's
a
key
principle.
It's
easy
for
this
team
to
get
pulled.
A
Lots
of
different
directions,
so
limiting
our
work
in
progress
is
a
value
as
you'd
expect,
with
an
open
source
project
that
we
default
to
being
working
in
public,
and
you
know
I
guess
one
of
my
litmus
tests
is
that,
if
we're
doing
a
good
job
is,
would
our
future
selves
six
months
from
now
curse
us
for
the
decisions
we're
making
today
or
if
all
the
team
was
gone
and
a
new
group
was
in?
Would
they
be?
You
know,
pleasantly
surprised
and
delighted
by
how
we're
making
things
simpler
and
easier
for
folks.
A
So
those
are
some
of
the
guiding
values,
the
kind
of
the
rhythm
of
the
team.
This
is
maybe
a
little
hard
to
read.
There's
some
things
that
happen
on
a
weekly
basis
like
some
of
our
reporting,
some
stand-ups
and
some
triage
meetings,
there's
some
things
that
are
more
bi-weekly.
There's
that
rust,
there's
a
general
lip
P2P
Community
call
which
alternates
with
a
rust
Community
call
as
well.
We
sometimes
get
time
with
Juan
and
then
on
a
monthly
basis
as
part
of
being
in
the
engine.
A
Pl's
engineering
research
group
there's
a
public
all
hands
that
we
partake
in.
So
that's
that's
one
of
our.
You
know
that
all
gets
published
to
YouTube.
That's
one
of
our
mechanisms
for
reporting
out.
There
is
also
this
weekly.
We
call
it
a
sit,
rep
or
situational
sorry
situation
report,
which
is
kind
of
our
where
we
go
through
all
of
our
initiatives
and
talk
about
what
happened
last
week
and
what's
coming
and
those
those
are
public
and
we
drop
those
into
the
lip
P2P
implementers
channel
on
each
week.
A
A
A
Some
of
the
nice
attributes
are
the
the
URLs
are
very
hackable,
so
you
can
kind
of
get
the
view
that
you
want
and
it
lets
you
store
extra
metadata
like
who
are
some
of
the
close
collaborators
that
you
want
to
stay
on,
top
of
so
that
you
don't
drop
their
issues
so
that
tool
is
available
and
we
we
use
it
as
you'd
expect.
These
groups
are
responsible
for
their
releases.
Their
release
processes
are
all
documented
in
their
repos.
You
can
read
about
them,
I
believe
you
know
js
on
the
JSI.
A
We
use
release,
please
with
conventional
commits
for
generating
the
releases
and
I
think
rust
is
going
to
be
moving
in
that
direction
as
well.
You
know
one
call
out
is
right
now
in
for
these
implementations
anyway,
we
don't
have
a
standard
release.
Cadence,
you
know,
I
listed
the
number
of
major.
You
know
major
releases
that
we
did
this
year,
so
releases
are
happening
about
every
five
to
eight
weeks.
A
A
A
Security
incidents
or
security
related
matters
certainly
take
first
priority,
and
you
know
if
there
are
issues
observed
in
production
networks
we
dive
into
those
as
well
for
chat.
Most
of
the
team
lives
in
file
coin
slack,
but
that
is
bridge
to
the
lip
P2P
channels
and
ipfs
Matrix
and
ipfs
Discord.
But
really
the
kind
of
like
the
home
of
the
team
publicly
is
in
the
lip.
A
P2P
implementers
Channel
and
you
know
really
just
can't
scale
by
getting
direct
DM,
so
don't
be
offended
if
people
respond
back
to
you,
moving
it
to
a
public
Channel,
the
outputs
for
the
team,
like
we
said
as
much
as
possible,
we're
trying
to
live
in.
We
put
things
in
the
specs
repo
if
it's
repo
specific.
So
if
it's
implementation,
specific
we
put
in
there's
corresponding
GitHub
repos
and
also
there
is
a
decent
amount
of
the
teams,
organizing
and
note-taking.
That
happens
in
notion
just
given
that
there
is
a
browse
experience
there.
A
It's
easy
to
export
out
to
markdown
for
GitHub
and
hack
MD.
We
get
automatic,
backlinks
and
good
organization.
Information
with
databases
and
I
want
to
give
a
quick
shout
out
on
the
doc
side,
because
I
know
this
has
been
an
area
like
lacking
for
a
long
time.
You
know
the
team
has
been
making
it
a
priority
now
to
as
they
do
releases,
to
have
the
corresponding
docs
changes.
A
A
Okay,
some
of
the
next
things
here
are.
We
really
need
to
get
some
meaningful
metrics
about
some
of
these
important
things.
So
if
we
want
to
be
bringing
more
in,
you
know
bringing
more
maintainers
in
having
more
contributions
from
outside
some
of
the
folks
here
at
PL
like
we
need
to
get
that
measured.
So
we
can
see
if
our
efforts
are
actually
improving
there,
so
that
that's
one
thing
you
know
it's
harder
as
a
library
to
prove
that
your
work
is
having
impact,
but
our
our
team
needs
to
take
us.
A
Go
A,
Step,
Beyond,
just
releasing
software.
We
need
to
make
sure
that
software
is
actually
having
impact
and
being
used
so
that
it's
getting
on
the
road
maps
of
other
projects.
Etc-
and
you
know
it
doesn't-
have
to
be
this
corset
of
Engineers.
We
need
to
be
bringing
other
people
in
that
are
helping
more
on
the
community
support,
and
so,
if
you
go
to
the
discussion
forums
at
discuss.lib
p2p.io,
there
just
isn't
the
vibrant
discussion
that
you'd
ideally
want
to
be
seeing
happen
there.
A
So
all
areas
of
improvement-
if
you
have
other
ideas
in
mind
that
you'd
really
like
to
see
this
team
be
doing.
Please
please
share
I,
look
forward
to
hearing
from
you
just
to
close
things
off,
want
to
talk
about
Road
mapping,
and
so,
as
we've
been
working
on
this
the
last
few
weeks,
there
are
a
few
tenants
for
the
project
that
kind
of
distilled
down
I.
Don't
think
any
of
these
would
be
a
surprise,
but
just
kind
of
articulating
them
this
this
way
and
wanted
to
share
it.
A
A
So
so
some
of
our
I
think
I
would
take
a
lot
of
inspiration
from
is
what
we
see
in
other
large,
open
source
projects,
particularly
like
programming,
languages
and
browsers
how,
when
you
see
their
benchmarking
and
test
Suites,
you
get
this
overwhelming
sense
of
like
oh
wow.
They
take.
They
take
this
really
seriously,
there's
a
degree
of
rigor
here
of
like
why
wouldn't
I
build
on
this
thing,
and
that's
that's
really
really
where
we
want
to
be
moving
with
P2P,
you
impart,
so
that
people
feel
confident
building
on
it.
A
But
that's
going
to
be
one
of
the
ways
that
we
scale
in
bringing
more
people
in
is
that
we
can
have
lots
of
people
contributing
and
we
can
know
the
potential
ramifications
on
a
performance
or
interoperability
regard
as
more
contributions
make
its
way
in.
So
that's
that's
an
area
of
focus
and
you'll
be
seeing
that
on
the
on
the
roadmap,
so
in
terms
of
navigating
the
p2p's
roadmap,
there's
actually
five
road
maps
to
be
looking
at,
but
it's
I
I
think
it'll
make
sense.
A
A
This
lives
under
the
specs
repo,
but
it
you
know
it
didn't
it
didn't
answer
like
well,
what's
going
to
be
delivered
when
and
which
implementation
and
that's
really
kind
of
some
of
the
effort
we've
been
taking
on
over
the
last
month
or
so
so,
each
of
the
implementations
have
a
markdown
file
for
the
roadmap,
some
roadmap.md.
It
lists
some
key
Milestones
that
users
or
stakeholders
would
really
care
about.
A
We
think
it
provides
Target
completion
dates
either
in
the
month
or
quarter,
there's
usually
a
tracking
issue,
so
you
can
follow
along
with
that
work
and
then
some
sort
of
issue
where
you
can
follow.
You
can
give
feedback
on
the
roadmap
as
a
whole.
So
we
have
that
across
the
project
not
going
to
go
into
them
here
you
can,
you
can
link
to
all
of
the
you
can
get
to
all
of
these
from
the
specs
repo.
A
We
have
them
for
gojs
and
rest,
but
then
also
an
important
call
out
is
we
have
a
dedicated
roadmap
for
the
p2p's
testing?
This
is
in
the
test
plans,
repo,
which
we'll
be
getting
to
be
hearing
some
more
about
this,
but
this
really
complements
what's
going
on
and
go
rust
in
in
JS,
and
it
really
hits
on
the
stability
and
performance
tenants
that
I
referred
to
earlier
and
it's
it's
a
shared
priority
across
all
the
implementation
teams.
A
So,
and
you
know
what
we
can
do
that
today
with
web
transport,
we
also
want
to
get
webrtc
delivered,
so
we'll
be
seeing
the
browser
to
go
in
Rust
side
at
the
end
of
the
by
the
end
of
the
year
and
then
browser
to
browser
in
q1
and
next
year
on
the
interoperability
end,
you
know
Max
and
his
slide
walk
through
all
some
of
the
different
modules
or
dimensions
of
with
P2P.
We
need
to
make
sure
all
that
all
those
permutations
work
across
all
the
implementations
across
multiple
versions.
A
There
is
some
improvements
on
go
the
P2P
to
reduce
round
trips,
that'll,
be
landing
by
the
end
of
the
year,
getting
quick
into
to
rust,
we'll
be
landing
and
then
also
adding
it
in
the
node.js
context
for
JS
the
P2P
and
q1,
and
similarly
want
to
be
able
to
have
a
dashboard
that
shows
real
life
Network
conditions
of
how
the
P2P
performs
compared
to
HTTP.
So
that's
some
of
what's
coming,
it's
not
exhaustive!
Please
dig
into
some
of
the
other
documents
I
referenced
and
with
that
I'll
say
a
huge.