►
From YouTube: IPFS - go-ipfs
Description
Details: https://github.com/ipfs/pm/issues/174
Notes: https://public.etherpad-mozilla.org/p/ipfs-sept-12-go-ipfs
https://ipfs.io
A
B
Mm-Hmm
pub,
so
it's
pretty
neat
and
that's
what
we
worked
on
them.
Even
I've
been
up
late,
a
lot
with
one
and
doing
all
sorts
of
work
on
that
front,
and
we've
been
spending
a
lot
of
time,
trying
to
figure
out
what
we're
going
to
do
for
demos
for
DEFCON
2
and
about
what
like
what
we
need
all
in
all
for
that.
So
most
of
our
time,
I'm
gonna
go
with
all
of
our
kind
has
been
spent
getting
pub/sub
and
j
SMS
in
our
up.
B
So
that's
what
we've
been
doing
as
far
as
other
miscellaneous
changes
on
ago,
I
PFS,
it's
been
a
little
quiet,
kubas
been
moving,
so
he's
less
active
than
normal
I've
been
doing
pub/sub
stuff.
So
that's
that
and
the
0
other
thing
is
that
we've
been
getting
closer
and
closer
and
closer
to
having
the
DHT
extracted,
which
is
a
cool
thing,
also
really
isn't.
B
Something
really
long
yep
yeah
yep,
so
that
is
a
quick
update
of
how
we're
doing
so
in
terms
of
milestones.
If
I
want
to
frame
things
that
way,
we
have
finished
the
rc4
milestone.
I
should
click
close
on
that?
Oh
that's
close.
We
are
pretty
close
to
closing
the
j
s.
Go
interoperability
milestone,
which
is
cool.
Like
we've
got
all
the
multi
stream
stuff
we've
got.
I
figured
out
that
really
stupid
nasty
stream
multiplexing
bug
with
phoebe
stream
last
night
took
us
like
what
seven
hours
did
well
a
long
time.
B
There
were
changes
on
both
sides
of
the
spectrum,
one
of
so
on
a
ghost
side.
They,
the
doctor
team,
whoever
was
working
when
that
didn't
they
were,
they
were
using
a
priority
queue
for
ordering
packets
that
you
know
for
ordering
the
frames
that
they
receive,
but
they
weren't
actually
setting
the
priority
on
the
incoming
streams.
So
that
was
causing
weird
things
where
the
data
packets
were
having
a
higher
priority
than
the
actual
sin
stream
packets
that
created
the
stream
that
those
data
packets
were
four.
B
So
we
were
just
dropping
data
packets
because
we
don't
know
what
streets
report
on
and
on
the
GS
side
there
was
all
sorts
of
weird
whole
stream
stuff,
I,
don't
know
what
I
don't
know
what
actually
was
wrong
there
or
what
we
actually
had
to
fix.
But
there
was
just
some:
it's
JavaScript
man,
man
like
come
on
yeah,
so
that
was
that
was
fun
and
also
the
JavaScript.
B
Speedy
transport
was
just
setting
random
priorities
on
things
which
hey
I,
don't
know
just
like
it
just
pick
a
number
said:
hey.
This
is
priority
7,
and
this
thing
this
thing's
part
III
just
because,
because
I
felt
like
it
so
that,
like
it
technically
works
as
long
as
we're
respecting
the
priorities
on
the
other
side,
which
one.
B
B
B
E
E
B
C
Ok,
I've
two
months
and
related
things:
one
is
hosting
one
of
our
websites
using
IP,
NS
and
IPL
d,
dot,
IO
and
multi
formats
that
I
or
hosted
through
a
pianist
right
now.
So
the
means
we
don't
have
to
change
dns
records.
Every
time
we
deploy
a
change
which
is
great,
I'm
gonna
get
metrics
for
how
well
that
performs
today
or
latest
tomorrow.
C
Avid
have
done
already,
so
I
only
need
to
set
up
graphs
and
figure
out
what
and
actually
useful
metric
would
actually
be
and
this
so
we
can
probably
strike
off
that
Matt's
done
and
all
the
other
one
is
concerning
the
core.
Api
I
have
a
pro
request
up
now,
which
does
well
it's
a
first
step
for
the
core,
API
and
parts
of
the
Gateway
use
it,
and
it
will
be
really.
I
don't
need
it
much,
but
just
a
quick
review.
Hey
is
this
giving
in
to
the
right
direction.
That
would
be
great.
C
Yeah,
that's
it
from
my
side
and
I'm
gonna
try
and
get
Kevin
Atkinson
involved
in
it.
I
guess
I
already
made
a
few
remarks
on
another
issue
about
it
and
we'll
see.
There's
enough
to
do.
The
API
is
big,
so
actually
that
stuff
is
not
using
the
API
is
that
we
have
already
properly
spect
out,
which
is
the
object,
API
and
a
couple
of
others,
and
it
said
it
uses
the
UNIX
FS
API.
So
that's
the
basic
cat
ad
LS
command,
and
how
does
this
packing
work?
D
C
D
A
D
C
That
that
reminds
me
of
that
another
point
about
that
for
now,
I've
placed
the
go
interface
for
all
that
in
go
a
PFS
and
the
go
a
PFS
rico.
Just
for
these
of
getting
started
with
it
and
against
when
we're
extracting
the
Gateway.
You
want
to
extract
that
interface
respective
repo,
where
the
jazz
stuff
is
as
well
together
with
the
press.
F
B
G
Alright
cool,
so
I
was
hearing
on
the
live
stream.
So
sorry
for
my
five
seconds
away,
larks,
let's
get
the
like
at
least
one
comment
like
files,
all
right
files,
cat
or
something
integrated
with
the
current
I
PFS
I'd
receptacles,
for
because
we
might
need
to
polish
a
little
bit
unlike
what
we
want
to
do
exactly
in
terms
of
like
pulling
the
interface
into
go.
G
I
PFS
there
has
been
like
some
discussions
if
we
should
like
have
the
interface
file
inside
the
idea
in
difference,
I
think
that's
called
repo
and
then,
like
the
girl,
I'd
be
fast
food
imported
or
if
that
should
be
just
a
speck
and
then
every
language
should
like
create
the
interface
file
for
their
own
implementation,
but
definitely
like.
Let's
try
for
one
endpoint
before
I
trying
to
do
for
all
the
end
points.
G
At
the
same
time,
even
just
happy
fest,
we
always
like
to
one
by
one
like
we
start
with
the
object
and
a
movie
went
to
the
block,
then
we
got
this
form.
Then
we
got
the
files,
because,
if
I
classes,
polishes
apology
API
is
way
faster
and
it
also
gets
things
marched
ways
to
there
because
they
don't
get
walked.
In
some
other
comment:
oh
you're.
G
C
A
A
C
C
A
C
A
C
I
would
still
need
to
deal
with
the
unix,
FS,
stuff
and
I.
Don't
want
to
have
the
Gateway
depend
on
that.
I
want
to
ever
depend
exclusively
on
the
core
API
and
just
to
keep
the
deeper,
because
otherwise
the
dependency
tree
is
gonna
is
going
to
be
a
mess
and
the
extractions
going
to
be
really
hard.
Okay,.
A
G
B
H
B
B
H
Yeah
I
have
a
hunch
from
listening
white.
Go
to
what
you
walk
through
that
the
generated
roadmaps,
accusing
haads
roadmap
generator
might
be
useful,
I'm
gonna
play
with
that
in
the
future,
because
I'd
like
basically,
you
ran
through
stuff
that
would
have
been
on
that
on
that
generated.
So
maybe
we'll
play
with
that
in
the
future.
But
for
now
I
like
that
before.