►
From YouTube: IPFS All Hands 🙌🏽 📞 Oct 9, 2017
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
It
has
been
a
couple
of
weeks
since
we
had
the
last
Holland
so
in
forward
to
hear
everyone's
a
bit
like
what
I
brought
is
working
on.
As
always,
we
have
the
gender,
please
do
add
items
to
the
agenda.
You
want
to
be
named
up
to
the
call
protocol
and
the
note
they
note
take
care
for
today
these
David
Bishop.
Thank
you
very
cool.
So
let's
get
started.
A
B
A
We
will
have
yeah,
thank
you
for
asking
I
kind
of
like
mentioned
that
on
the
issue,
so
essentially
most
of
the
people
that
TP
wait
in
this
call
are
traveling.
So
it's
really
hard
for
you
to
tell
the
life
traveling,
like
literally
inside
a
plane,
so
it's
hard
for
them
to
be
here
and
I
discussing
road
map.
We
want
well
that
call
as
soon
as
possible
and
we
are
shooting
for
next
week.
A
A
C
A
B
B
Credentialing
of
physicians,
as
well
as
verifiable
claims
of
vaccinations.
One
of
the
best
examples
of
that
is
after
the
hurricane
actually
came
through
Florida.
There
is
a
massive
refugees
of
people
fleeing
the
hurricane
to
Tennessee
the
state.
The
governor
Tennessee
had
an
executive
order,
number
66
that
suspended
this.
The
laws
that
surrounded
the
practice
of
medicine,
so
that
allowed
for
any
physician
in
any
state
to
take
care
of
any
patient
as
long
as
they
were
refugee
of
their
hurricane.
So
that
was
the
use
case.
B
I
tried
to
actually
using
using
AG
dimensioned
project
in
D,
which
is
hyper
Ledger's,
identity
management
and
member
services,
and
looking
at
the
code,
it
was
just
filled
with
the
the
skulls
of
bad
poor
security
practices
and
ultimately,
I
did
install
it,
and
but
it
was
just
fluff.
It
actually
was
like
a
command
line,
Alice
and
Bob
and
actually
didn't
have
any
sort
of
robustness.
I
decided
to
build
my
own.
B
The
use
case
here
is
really
using
these
decentralized
public
key
infrastructure
in
order
to
get
the
public
keys
of
the
patient
and
the
provider
in
order
to
create
a
peer-to-peer
protocol
so
that
the
physician
can
document
vaccination
records
into
their
EHR
and
patient
gets
the
vaccination.
But
the
patient
is
able
to
get
a
verifiable
claim
that
they
actually
got
that
vaccination.
In
this
case
they're.
If
they're
refugee
fleeing
a
hurricane,
they
don't
have
their
their
yellow
card
of
all
their
vaccinations.
B
So
the
syntax
for
the
D
ID,
spec
decentralized
identities,
was
actually
came
out
of
one
of
the
rebooting,
the
web
of
trust
meetings
back
last
year
and
I
think
actually
one
was
there
and
actually
but
I.
Think
batalik
and
Drummond
read
actually
pens
the
the
specification-
and
here
it's
basically
it's
a
variation
of
the
you
RN
methods,
Beck
Uniform,
Resource
name,
which
is
a
published
I
Anna
protocol
instead
of,
for
instance,
this
could
be
you
RN
ISBN,
but
in
this
case
that's
D,
ID,
sovereign
as
being
the
method.
B
Think
it's
just
I
wanted
it.
It's
a
it's
a
method
on
top
of
IP,
NS
and
so
I
think
and
I,
and
this
is
I-
think
we're
just
for
I
need
some
feedback.
If
this
is
going
to
be
a
good
use
of
IP
NS
for
one
node,
because
it
really
is
just
serving
up
one
DDO.
If
it's
that's
that
the
and
so
I
penis
is
really
sort
of
could
be
used
for
serving
up
a
lot
more
than
just
one
JSON
document,
and
so
I
think
it's.
B
D
B
And
the
idea
is
that
one
actually
have
the
public
keys
I
actually
created
a
pub/sub
broom
that
actually
uses
a
concatenation
of
each
of
the
public
keys
in
order
to
create
web
RTC
communication,
and
once
you
actually
have
that
I
guess,
it's
actually
technically
pub
flood
actually
reading
the
documentation.
So
it
really
still
needs
to
be
a
specific
peer-to-peer
communication
channel.
But
I
was
used
to
using
the
pub
sub
room
that
Hector
created
and
I'm
good
and
it's
stalling.
E
B
Cash,
what
div
stands
for
decentralized
identifiers
and
in
the
sovereign
framework
they
actually
they're
gonna
have
an
entire
block
chain,
there's
also
also
one
by
Manu
scoonie,
which
is
Varys
one.
That's
the
crude
digital
bazaar,
so
we
actually
have
to
add
the
rebooting,
the
web
of
trust
conference.
We
actually
had
a
whole
bunch
of
different
ways
of
resolving
either
over
sovereign
and,
ultimately,
it's
a
namespace
resolution
and
I
think
that's
I
think
the
next
thing,
which
is
that
it
you
basically
needed
to
register
this
method,
spec
and
figure
out
how
to
resolve
it.
B
B
But
basically
it
creates
a
room.
The
the
room
is
a
pub/sub
room.
Once
you
actually
have
that
channel
of
communication,
there's
just
a
messaging
good
back
and
forth.
It
says:
hey
what
I'm
a
requesting
verifiable
claims.
Do
you
have
any
verifiable
claims
and
which
is
all
just
the
code
that
I
stole
from
Project
Indy
I
was
just
you
know.
That's
I
think
that
the
next
step
is
that
once
we
resolve
the
DDO,
then
it's
a
matter
of
like
showing
the
verifiable
claim
and
and
using
it
and
verifying
it,
verifying
the
signatures
using
the
public
key.
C
E
B
A
A
nice
fire
just
just
before,
like
you,
were
cheering
I,
think
just
a
browser,
and
so
we
could
see
the
entire
capture
of
the
browser
right
now.
I
think
you
are
sharing
the
entire
screen
and
so
like
for
everyone.
It
does
like
a
screen.
Although
smaller
resolution
and
yours,
we
cannot
see
the
entire
slippered
of
codes
that
you're,
showing
on
the
slide
all
right.
If
you
could
just
share
the
browser,
every
would
help
you
and
follow
your
you
know
what
I
thought.
B
B
Here's
an
example
of
a
verifiable
claim
about
a
specific
D
ID,
in
this
case
the
proof
vaccination
credential,
with
the
claim
being
a
specific
vaccination
code.
This
example
is
just
a
flu
shot
and
then
I
also
did
a
smart
contract,
just
a
simple
proof
of
existence,
contract
or
the
vaccination,
and
it
has
a
signature
that
actually
can
validate
the
signature
and
who
the
signatures
of
about.
Then
here's
a
proof
of
licensed,
credential
credential
that
proves
the
physician
is
licensed
and
in
this
case
I
just
made
it
up.
B
This
is
where
actually
I'm
on
mania
w3
task
force
for
a
verifiable
claims,
we're
actually
working
out
the
attributes
of
describing
how
do
the
robustness
other
of
the
attributes
to
describe
a
licensing
credentials.
So
I
think
I
think
this
is
a
better
solution
to
the
whole
decentralized
public,
key
infrastructure
that
is
truly
self
sovereign,
that
I
own
and
doesn't
require
a
blockchain
I
think
to
resolve
that
I.
Think
it's
really
steps
our
messaging
of
how
do
you
actually
pass
the
verifiable
claims
back
and
forth
and
then.
B
A
She's
awesome,
thank
you.
So
much
for
the
presentation.
I
have
a
couple
of
questions
actually,
like
I
was
literally
talking
with
another
group
I
can
hour
ago
about
the
IDs
and,
like
their
experience,
this
a
mega
time
and
they
are
working
on
like
fake
news,
and
so
they
are
trying
to
develop
like
a
language
for
fight
claims.
A
It
seems,
like
you
hear,
like
a
great
head
start
when
we
say
that
like
identity
is
completely
self-serving,
sovereign,
how
kin,
like
the
responsibility
for
said
that
I
didn't,
but
just
for
between
two
organizations
managing
like
heaps
I,
have
something
unique
that
validates
my
identity,
though
it
it's
like
who
governs
a
record.
What
can
I
transfer
that
all
of
it
and
to
another
like
blockchain
or
even
just
like
to
a
centralized
service
like
an
anniversary
or
or
a
government
institution
desired?
Is
there
any
way
to
do
that
or
yeah.
B
So
ultimately,
it's
about
key
management,
and
it
requires
that
central
authority
to
actually
the
post
a
public
key
and
use
a
service
that
resolves
that,
basically,
that
you
can
validate.
Let's
say
it's
the
state
of
Tennessee
for
position,
licensing
or
could
be
I'm
on
the
ibms
American
Board
of
Medical
Specialties.
B
The
next
thing
is
actually
is
the
resolving
of
the
verifiable
claims
in
a
way
that
is
resolvable
across
different
methods,
and
that
means
it's
like
alright,
so
I
I
used
I
P
ID,
but
then
you
can
get
it.
My
verify.
The
claims
on
ipfs
so
or
I
see
that
you're
using
sovereign
or
Varys
or
bit
bt
r
TK
before
bitcoin,
stupid
I
was
talked
about.
So
there's
a
way
of
actually
resolving
across
blockchains
that
actually,
you
can
get
those
be
Republic
claims
and
verify
the
signatures
in,
and
so
this
is
ultimately
about
namespace
resolution.
B
So
how
do
you
designate
a
flag
and
say
IP,
ID
or
Varys,
one
which
is
going
to
be
V,
1
or
e
th
for
aetherium
its?
We
actually
also
talk
about
who
registers
ETH
for
ethereum,
who
owns
it
and
who,
basically,
as
part
of
the
method,
spec
specifies
how
you
resolve
the
video
and
exchange
the
verifiable
claims.
The
Evernham,
who
does
sovereign
always
actually
had
an
interesting
thing.
Was
they
actually
basically
wanted
to
a
zero
knowledge?
B
Proof,
you're
passing
a
binary
to
the
prover,
the
person
who
owns
the
identity
and
yet
this
gif
to
solve
a
math
problem
in
order
to
then
give
it
back
to
the
once.
You
actually
solve
that,
then.
Actually
it's
proof
of
the
verifiable
claim.
But
it's
a
right
now
on
the
verifiable
claimed
task
force,
we're
actually
still
just
working
on
aged
over
21
as
being
through
the
use
case,
and
that
I
think
is
still
where
we're
at.
A
G
G
B
Yep
and
I
think
one
of
the
is
brilliant
mark
Miller,
the
famous
computer
scientist,
was
actually
at
the
rebooting
weather
trust
as
well
as
Tim
berners-lee.
So
just
when
I
just
had
to
pause
and
like
you
know,
I
had
intellectual
conversation
with
Tim,
berners-lee
and
I
was
just
wowed,
although
I
actually
he's
very
much
HTTP
focused,
that's
what
he
thinks.
B
That's
all
and
I
think
the
HTTP,
of
course,
being
just
one
one
protocol
among
many
in
the
ipfs
stack
so
but
yeah
so
I
think
certainly
ID
identities
and
and
I
think
the
other
thing
is
actually
smart
agents
that
actually
we
talked
about
at
the
meeting
was
that
Alice
and
Bob,
but
then
there's
the
idea
of
having
a
dummy
bot
that
actually
Ellis
gives
to
the
dummy
bot
in
order
to
store
and
actually
transfer
a
asset
and
delegate
that
to
an
agent.
So
I
think
it
really
is.
B
G
And
if
I
could
add,
Johnny
I
think
that
the
point
you're
talking
about
particularly
the
smart
agents,
might
be
a
next
paradigm
for
smart
contract,
so
I'm
going
to
say
logic
built
into
the
overall
networking
system
that
might
actually
be
superior
to
the
etherium
approach.
So
mm-hmm
I
think
it's
really
fundamental.
What
you're
talking
about
cool.
A
B
Yep
sure,
at
the
end
of
the
rebooting,
the
web
of
trust
we
actually
were
all
forced
to
basically
is
about
pushing
out
code
and
specs,
and
so
so
I
did
it's
here.
My
github
I
submitted
and
then
project
india's
under
hyper
ledger.
It's
a
hyper
larger
project
and
I'll
find
it
it's
under
hyper
ledger:
github,
slash,
project,
indie
and
I'll.
Put
that
in
there.
A
All
right,
okay,
thank
you
cool!
So
let's
go
back
to
the
agenda.
I,
don't
think
any
new
item
was
added.
We
still
have
30
minutes.
Would
people
like
to
go
around
and
just
give
an
update
on
what
they've
been
working
on?
I
guess,
Charlie,
just
shared
these
update,
you
know
a
lot
of
cool
stuff.
Do
others
want
to
do
the
same
house
again.
A
G
Well,
Jason
who's
on
the
call
as
well
from
Phoenix
is
part
of
a
group
here
in
Phoenix.
That's
working
on
some
naming
addressing
numbering
and
identity
applications
using
IP
FST's
using
the
JSI
PFS
and
web
RTC,
and
we've
we've
had
a
little
all
over
the
last
month
or
so
because
of
some
other
distractions.
But
our
intention
is
to
work
on
exactly
what
japanese
talking
about,
and
you
know
again,
I
see
this
as
a
key
area
for
innovation
and
something
that
could
also
be
an
opportunity
to
use
file
coin.
G
G
G
Yes,
it's
been
a
long
time
coming
and
yes,
the
mass
adoption
isn't
there
yet
and
I
think
this
naming
addressing
a
numbering
is
a
key
missing
piece
for
mass
adoption
of
WebRTC,
so
whatever
browsers
don't
support
it,
in
my
view
that
the
direction
is
definitely
towards
ubiquitous
web
RTC
capability
and
if
you
are
interested
there
is
a
conference
next
June
in
the
UK
called
calm
Khan,
which
is
going
to
be
a
very
small
group
of
participants
that
are
going
to
be
focused
on
WebRTC
application
and
innovation.
So
I
think
it's
comm,
co,
n
dot
IO.
A
H
Yeah
sorry
yeah
I,
don't
want
to
have
to
find
another
way
to
test,
rather
than
all
the
goo,
the
small
group
programs
that
are
used
by
Shauna's,
because
it's
not
it
would
be
a
lot
of
work
and
I.
Don't
think
it's
worth
it
because
it
will
make
the
tests
be
different
between
go
ipfs
and
GSI
PFS,
so
yeah.
H
A
A
When
we
know
it's
a
change,
we
change
the
tests
but
like
it
would
be
great
if
we
could
actually
have
a
way
to
use
the
sharpness
test
to
tell
us
when
there
is
a
breaking
change
so
that
you
can
update
the
version
accordingly,
we
are
also
getting
more
into
editions
of
ipfs.
We
now
see
that
there's
a
nick
in
DSC
implementation.
There
is
a
working
progress,
python
implementation
and
in
the
me
long
term
we
probably
will
see
a
rest
implementation.
A
So
it
would
be
useful
right
now
to
start
preparing
the
sharpness
test
to
be
just
like
this
consumable
by
our
basic
patience
so
that,
yes,
we
can
use,
go
binaries
to
do
the
tests.
But
if
that's
the
case,
then
those
binaries
should
live
in
their
own
repose
be
compiled,
be
pulled
into
the
docker
image
and
another
image
should
know
how
to
pull
just
a
profess.
The
ipfs
rest
at
if-else
pipe,
if
at
and
so
on.
H
A
E
A
Probably
can
I
would
strongly
prefer
unless
there
is
like
a
strong
Walker
for
it
to
have
all
of
the
Charlotte
tests
like
ipfs
Turner's,
it's
on
the
ripple
without
the
setup
like
it
would
be
great
if
we
could
like
just
point,
the
Charlotte
stares
sounds
like
test
is
binary
and
I
kill,
just
like
run
the
test
again
that
binary
and
tell
us
like
what
is
broken,
what
doesn't
work
and
so
on.
Instead
of
like
having
the
tested
ABS,
you
know
this
repose
and
which
is
okay.
H
H
H
E
So
I'm
I've
been
teaching
a
course.
I've
been
pretty
bogged
down
with
that
and
that's
sort
of
been
my
story
for
the
last
couple
of
months,
but
but
I'm
working
on
finishing
still
up
on
this.
A
cluster
feature:
happy
FS
cluster,
to
support
basic
auth
for
simple
authentication
and
and
working
on,
going
to
do
a
call
with
Debbie
this
week
to
talk
about
writing
up
a
bit
swap
spec,
which
will
be
really
cool
and
also
continue
to
work
on
a
bit
swap
research
and
right
now,
I'm
very
close
I'm.
E
A
F
Classic,
okay,
so
everybody
not
too
much
to
report
I
just
got
back
off
a
plane.
Last
night,
it's
still
kind
of
tired
I've
been
mostly
focusing
on
a
few
issues
that
are
still
showing
up
without
the
MS
cluster,
specifically
with
the
rap
state.
When
you,
when
you
join
a
cluster
and
then
people
leave
and
then
things
are
pinned,
and
then
people
come
back
so
yeah
and
currently
on
yeah,
just
working
on
some
details.
There
anybody's
interested
further,
let
me
know,
but
why
don't
little
details
with
that
yeah.
A
B
D
Is
this
kind
of
like
ipfs
pub/sub,
API
implementation,
where
third
third
party
auditors
can
contribute
and
listen
into
a
measure
of
IOT
devices
that
communicates
some
sending
types
of
you
know
co2
emissions
like
that,
and
if
anybody
has
any
input
on
how
to
execute
something
like
that,
some
kind
of
decentralize
API
execution,
reusing,
pub/sub
or
RPC
or
whatever
it
is,
might
be
more
than
happy
to
hear
about
that
and
I'll
post.
Some
information
about
the
act
upon
in
the
chat.
A
Awesome
yeah
couple
of
ideas,
so
in
its
most
simple
form,
you
should
be
able
to
just
like
to
subscribe
to
a
topic
that
everyone
else
is
like
publishing
to
like.
Think
of
it
as
like
a
log
like
if
everyone,
if
every
single
device
is
publishing
to
like
this
log
thing,
then
you
can
I
know
which
topic
they
are
publishing
subscribe.
You
get
events
of
course
like
right
now
and,
as
Johnny
pointed
out
earlier,
the
implementation,
the
underlying
implementation
of
cops
having
a
PSS,
is
what
we
call
what's
up.
A
So
you
just
like
who
averages
the
peer
connectivity
that
site
defense
has
to
a
bunch
of
peers
and
use
it
HT
to
collect
all
the
peers
that
are
publishing
on
the
topic
to
get
the
messages
on
a
topic.
It
does
do
some
relay,
but
it
preemptively
tries
to
die
out
to
all
the
peers
that
are
interested
on
the
topic,
and
so
it
is
not
necessarily
X
killable
for
like
millions
of
nodes
like
in
order
to
be
scalable
for
millions
of
nodes,
which
is
typically
what
you
get
from
IOT.
A
You
have
to
have
some
kind
of
like
spanning
tree.
Where
know
that
about
sharing
message
with
each
other,
and
then
you
have
like
to
have
strategies
to
make
sure
that
no
no
they
cooks
is
the
others
and
that's
something
that
we
are
actively
researching
on.
There
is
a
repo
and
we
peer
to
peer,
slash
research
observed
which
has
a
bunch
of
papers
and
there's
people
working
on
that.
A
That
being
said
again
for
perfect
use
cases
and
I
for
trying
the
API
out,
you
can
do
this
like
subscribe
on
a
topic
that
they
don't
establish
into
and
therefore
replicating
their
data.
When
one
of
the
examples
that
was
built
in
the
past
that
used
this
tool,
this
technique
was
the
three
pi
IFT
be
and
the
annotations
for
triple
I
apartments.
A
It
was
shelled
around
june
this
year.
I
think
there
is
like,
and
as
a
demo
of
add,
Andrew
and
federal,
showing
other
words
and
essentially
like
you,
have
multiple
organizations
publishing
annotations
to
pops
out.
You
see
the
additional
developments,
but
when
the
annotations
to
persists
so
then
you
have
like
a
couple
of
other
extra
nodes
that
are
subscribing
on
the
same
topics
and
just
like
pinning
all
the
changes.
All
the
annotations
that
were
created
by
the
browser
nodes.
This
way
you
have
like
the
real-time
ish
way
of
like
everyone,
sees
the
latest
annotations.
A
D
F
A
C
I
So
I've
been
mostly
kind
of
jumping
in
and
fixing
some
of
the
political
issues,
specifically
progress
functionality
and
in
Jessica's
API
I've
also
started
to
work
on
the
IPO
de
aetherium.
Integration
with
Herman
and
and
come
on
is
being
picking
up.
Some
go
so
yeah,
it's
pretty
much
it.
We
finally
map
the
storage
tree
or
cry
or
or
it's
pronounced
now.
A
All
right
so
I
think
I
am
the
last
one
essentially
like
I've,
been
like
out
on
just
figuring
out
everything
that
we
were
just
like
writing
down
everything
that
needs
to
be
worked
on
when
ipfs
land
and
seeing
a
lot
of
discussions
on
it,
I've
been
out
serving
a
lot
of
discussions.
We
petrol,
who
is
working
on
pierpass
on
what
needs
to
be
the
nor
prepare
back
to
work.
What
are
the
things
that
we'll
be
able
to
extract
from
people
for
others
to
build
applications
on
top
from
mic
methods
of
Education?
A
Just
this
charity,
that's
on
top
of
a
DFS
that
other
people
can
use,
is
just
a
key
value
store
and
so
on
been
working
on
preparing
the
workshop
so
like.
If
someone
is
going
to
be
at
most
fast,
you
know
the
month.
Morlin
will
be
there
just
like
showing
people
how
PFF
works
internally
and
I
quill
about
a
small
workshop,
but
everyone
will
be
able
to
build
like
a
small,
discrete
application
on
top
of
us
quickly.
Fun
if
you
are
in
one
and
make
sure
to
to
join
us.
A
Yeah,
like
I,
have
a
huge
backlog.
So
many
issues
I
know
a
lot
of
people
having
to
heat
up
an
email,
I'm
working
as
fast
as
I
can
to
get
for
everyone,
I'm.
Sorry,
if
you're
blocked,
let
me
know
Tracy
sure
like
super
blocks,
because
you
need
an
answer
for
me
and
I'll
try
to
provide
you
in
information
as
soon
as
I
can
I'm
trying
to
understand
like
what
are
the
top
priority
ones
right
now,
and
that's
it
for
me.
A
B
B
I
even
was
actually
thinking
about
reusing
activity
pub
to
actually
as
the
framework
for
communication
back
and
forth,
but
I
think
there's
probably
some
sort
of
need,
for
you
know:
hey
I'm,
talking
about
di
D,
verifiable
claims
back
and
forth,
or
I'm
actually
doing
a
distributed
geo,
carbon
marketplace
and
actually
what
what
you
know?
What
what
messages
do
you
speak
sort
of
actually,
once
you
actually
get
into
a
room
or
a
channel,
there's
actually
a
need
for
a
protocol
to
actually
to
chat
about
all.
A
Right,
okay,
so
there
is
multiple
answers
to
the
question,
because
there
is
multiple
ways
to
approach
it
right
now:
ipfs
pub
sub
room
he's
like
one
of
the
many
persons
we've
been
building
on
top
of
EFS,
and
it's
like
what's
here
now
kind
of
like
developing.
Is
this
framework
where
there's
a
lot
of
innovation?
A
So
we
kind
of
like
develop
these
things
on
top
and
we
expose
them
to
the
like
general
community
and
like
then,
we
kind
of
try
to
understand
if
they
are
useful
or
not
like
how
they
are
useful.
What
are
the
pain
points?
What
should
we
in
the
end
push
to
pour
like
what
is
this
innovation
that
is
happening
unusual
and
that
we
should
push
the
poor
that
should
be
implemented
in
all
of
the
ipfs
implementations
in
multiple
languages?
So
that
is
one
IP
first
box
that
room?
A
Yes,
like
you
get
this
room
that
you
subscribe
to
and
then
just
broadcasting
information,
but
the
underlying
pub/sub
that
ipfs
exposes
tells
you
that,
like
you
get
messages
that
you
subscribe
on
a
certain
topic
and
and
for
example,
you
can
just
publish
on
a
topic
like
you
can
just
publish
on
a
topic
like
ids
of
peers
that
are
interesting,
but
you
don't
even
have
to
transmit
the
information
through
the
pub
sub
channel.
If
you
were
to
transmit
the
ideas
that
are
interested
on
a
certain
topic,
then
you
can
buy
all
to
those
nodes.
A
Once
you
die
out
to
those
nodes,
the
primitives
that
we
peer-to-peer
offers
you
to
die
ohm
is
actually
to
like
you
can
buy
all
just
for
the
sake
of
a
link
and
then
I
quote
you
dial,
you
start
bid
swapping.
We
have
node,
you
start
DHT,
we
don't
know,
but
you
also
can
dial
on
a
protocol.
You
can
do
like
dial
PID
or
appear
a
multi
other,
and
then
you
can
select
a
string
which
is
a
multi
product
and
that's
how
beats
warp
and
HD
works.
A
He's
like
when
you
dial
to
appear
you
open
multiple
streams
over
a
same
connection,
and
you
say
on
this
theme:
I
want
to
speak
website
on
each
team.
I
want
to
speak
bits
out
on
the
extreme
LSP
DHT,
and
you
can
do
that
for
your
own
protocol.
You
can
say,
oh
I,
want
to
use
the
same
underlying
connection
and
speak
another
protocol
there
as
a
specific
wire
format
and
because
it
will
be
on
its
own
isolated
multiply.
Extreme
like
can
do
whatever
it
wants.
Like
you
can
define
your
own
wire
format.
A
You
can
even
have
a
multiple
wire
formats
working
over
the
same
connection.
In
fact,
I'd
be
chopped
already,
because
there
was
a
migration
from
IP,
FS
0
for
1
to
2,
or
something
that
now
we
have
Pete
swap
1
0,
0
and
B
2
up
1,
1,
0
and
I.
The
ffs
node
will
check
will
try
to
dial
on
both
bit
swaps
and
I,
get
the
notes
for
the
latest.
A
It
uses
the
latest
and
it's
the
one
that
supports
IPL
denotes
the
the
other
version
is
like
just
my
code:
I
go
and
we're
just
like
a
change
per
box,
and
so
it
gives
you
the
size
power
of
future-proofing.
Your
system
gives
you
a
way
to
upgrade
the
wire
formats
without
necessarily
having
to
search
within
a
very
well
it's
time
to
change.
A
A
If
you
point
me
at
questions
like
the
same
way
that
you
just
asked
here
through
issues
and
then
I
can
like
point
you
to
all
of
these
repos
and
all
these
discussions,
and
all
these
notes,
or
even
just
like-
create
a
new
documentation,
because
it
helps
us
a
lot
each
time.
I
people
ask
questions
about
all
these
students
work
between
then
understand.
Ok,
we
need
to
write
more
examples
about
this
or
more
diagrams
and
so
on.
B
B
Itself
of
actually
that
has
structure
and
I
think
that's
where
I'm
at
right
now,
I
think.
Oh,
you
worry
about
the
plumbing
down
here.
I'm
worried
about
content
inside
it
actually
like
it,
tears
talking
back
and
forth.
Actually,
hey
I'm
talking
this
protocol
and
and
and
how
do
we
actually
go
back
and
forth
and
actually
the
way
I
implemented
it
ultimately
was
actually
I
had
to
pub/sub
rooms.
This
one
was
a
server
client,
this
one's
client
server
and
basically
I'm
sending
encrypted
stuff
this
way
and
you're
sending.
Could
you
know
its
way,
I
solved
it.
A
I
Sorry,
the
pub
sub
room-
it's
in
Crete
I
was
recently
working
with
it
as
well,
and
so
I
was
taking
out
the
code
in
and
you
can
talk
to
specific
peer
as
well.
So
you
can
actually
specify
a
peer
that
you
want
to
send
messages
back
and
forth.
I'm,
not
sure
if
that's
what
you
guys
are
talking
about,
but
that's
what
are
then
they're
actually
supports
them.
Yeah.
I
Can
use
a
signal
server,
but
once
you
have
a
connection
you
can
just
watching
once
you
know
about
the
peer,
you
can
just
directly
connect
it
as
well.
If
that's
your
supports
done
so
just
passing
the
PID
there's
a
specific
method
and
in
pub
sub
room
that
you
can
pass
a
pure
ad
and
that
way,
you'll
establish
a
direct
connection
to
that.
To
that
be
writing
you'll
be
able
to
send
messages
back
and
forth,
but
then
that's
with,
and
that's
where
that
was
the
signal.
Server
signals,
Harry
I
believe
it's
just
for
discovery
now.
A
Another
level,
if
I
may
sorry
I
mean,
like
think
about
it.
I,
like
you,
have
a
thing
that
just
like
establishes
connections
like
sockets
between
peers.
Once
you
do
that
you
upgrade
a
connection
to
a
string
multiplexer
so
that
you
can
have
like
these
multiple
channels
for
every
peer
when
you
reach
the
pub
sub
level,
either
pub
sub
room
or
even
like
the
ipfs
pub.
So
you
don't
think
anymore,
like
oh,
you
get
the
connection
you
just
think.
A
Oh
I'm,
connected
to
the
peer
or
I
can
find
more
ways
to
connect
to
other
peers,
and
so
that
can
be
to
a
web
RTC
channel
that
can
be
to
a
TCP
socket
that
can
be
through
a
web
sockets.
But
it's
like
at
that
level
like
the
IP
fest
pops
at
room
doesn't
really
care
in
learn
for
it
to
work
in
a
browser
today.
The
transport
that
we
support
is
web
RTC
and
the
WebSockets
and
we're
back
to
see
does
use
a
signaling
server.
That
also
gives
the
discoverability
property.
So
like
that's.
A
Why,
when
you
open
a
node
with
WebRTC
in
a
lot
like
it
just
starts
connecting
to
other
nodes,
because
all
of
them
are,
according
to
this
point,
that
as
information
about
more
nodes,
but
what
we
want
to
upgrade
is
to
enable
IP,
fest
notes
to
be
this
random,
appoints
for
any
other
node.
So
I
connect
to
a
bootstrapper,
now
I'll
buy
DFS
and
then
I
asked
hey.
Do
you
have
more
peers
that
are
using
lowercase
e?
A
If,
yes,
tell
me
about
it,
and
so
that,
like
any
nodes
in
the
in
the
matter,
can
be
a
relevant
point
for
any
other
else,
so
that
you
don't
have
to
like
use
a
specific
brand
of
a
point
as
it
is
working
today,
it
is
part
of
like
the
depth
great
path.
We
cannot
just
escape
the
fact
that
we
really
need
a
renewable
point
like
for
WebRTC.
You
need
a
place
to
exchange
the
signaling
data
like
that,
the
offerings
between
peers
so
that
they
can
open
ports
in
the
browser.
That's
we
cannot
expect
yeah.
D
B
A
I
I
have
one
one
small
thing
to
mention:
I
did
run
a
couple
of
meetups
here
in
Costa
Rica
and
they
present
I
profess
and
talk
about
I
profess
to
a
few
people.
And
surprisingly,
a
lot
of
people
are
actually
where
I
guess.
Not
surprisingly,
any
more
people
are
eager
to
start
using
it
and
I'm
thinking
about
running
a
couple
more
once
I
have
a
chance
and
maybe
maybe
even
building
a
few
something
that
uses
I
profess,
but
there
is
will
and
there
people
are
aware
and
and
they're
paying
close
attention
to
what's
going
on
here.