►
From YouTube: IPFS All Hands 🙌🏽 📞 Aug 28, 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
B
C
Thank
you,
yeah
welcome.
We
don't
have
a
lot
of
urgent
items
today,
but
if
you
do
have
anything
you'd
like
to
talk
about
please
edit
and
we
get
to
it
in
a
bit.
Otherwise,
let's
start
with
a
little
I.
Could
you
there
must
typing
furiously
the
V?
Thank
you.
So,
let's
start
with
with
around
what
everybody's
good
to
in
in
the
past
week
and
I,
don't
know
if
we
have
any
fixed
order
for
it.
So
I'm
just
gonna
start
I'm,
just
gonna
pick
someone
Victor.
D
D
So
I
currently
are
looking
into
a
bug
with
these
four
piers
in
JSI
TFS
brought
in
some
additional
chests
some
tests
or
not
specific
enough.
So
we
introduced
a
bug
where
we
listed
all
the
addresses
is
one
address
for
pure
fixing,
thoughts
and
time
I
discovered
on
different
bug,
I
managed
to
introduce
small
new
key
into
thoughts
as
well
other
than
thoughts.
I.
Also
noticed
a
week
or
two
ago,
that's
we
haven't
published
again
the
source
code
for
our
discourse,
information
on
the
configuration
and
everything
and
I
also
notice.
D
D
F
What's
up
guys,
I
started
teaching
a
class
last
week,
so
I've
been
pretty
bogged
down
with
that,
but
I'm
also
I've
been
working
on
cluster
a
bit
I'm
just
getting
it,
but
the
are
pushed
through
and
once
I
finished
what
I'm
working
on
with
cluster
I'm,
going
to
start
working
on
more
tests
lab
and
bit
swap
related
stuff
from
the
most
work.
H
H
Is
showing
up
on
my
screen
stuff,
Thank,
You
Lars,
so
I
have
been
goofing
off
last
week.
I
was
on
vacation,
but
we're
working
on
some
ways
to
record
calls
as
well
as
video
and
store
them
in
ifs
and
utilize
file
coin,
once
that's
available.
So
what
we're
working
on
here
in
Phoenix
is
more
towards
the
application
side
and
I'm
just
getting
back
on
the
grid,
but
we'll
keep
you
guys
posted
as
to
how
that
progresses,
and
hopefully
we'll
have
a
demo
and
then
not
too
just
a
future.
G
G
I
That
that's
everyone,
so
I've
been
working
on
various
things.
The
most
recent
one
was
my
work
on
crypto
because
turns
out.
Web
crypto
is
giving
us
more
heartache
than
we
had
hoped
for
so
I've
been
working
on
that
today
and
doing
research
and
I
think
some
things
experimentations,
the
other
things
I've
worked
on
a
fortune
eight
year.
I
There
is
a
release
of
Adria
we're
currently
working
through
various
issues
of
operating
the
various
refers
to
that,
but
it
is
much
better
now.
So
hopefully
it
will
be
less
painful
once
that
migration
has
happened
other
than
that
oh
yeah,
I,
finished
upgrading
for
anybody
who
was
interested
in
Bitcoin,
I,
finished
upgrading,
go
I,
peeled
the
BTC
to
be
able
to
ingest
segwayed,
compare
a
seg
woodbox,
because
that
was
broken
since
the
new
editions
of
the
new
format
was
changed
a
couple
days
ago.
That's
working
now
again!
So
that's
nice,
yeah.
A
I
flew
to
Tennessee
to
see
the
Eclipse
and
though
I
passed
through
Nashville,
twice
I
failed
to
see
Johnny
punch.
He
graciously
was
gonna
host
me
at
his
house,
but,
and
you
know,
camping
in
the
woods
longer
than
planned.
So
but
it
was
awesome,
so
yeah
I
took
some
time
off
this
last
week,
so
I
haven't
been
doing
so
much
right
now,
I'm
in
Boston
to
co-host
a
workshop
tomorrow
with
people
from
edgy
than
nominal
data
governance
initiative,
where
they
were
behind
the
grassroots,
organizing,
with
a
to
rescue
events
and
and
all
of
those
efforts.
A
And
so
this
workshop
is
all
based
around
showing
people
how
to
use
tooling
they've
built,
which
uses
ipfs.
How
to
use
that
tooling
to
hold
copies
of
the
data
that
you
rely
on,
how
to
find
whether
there
are
already
copies
of
data
that
you
rely
on
stored
somewhere
on
the
destroyed
distributed
web
and
also
how
to
add
information
about
those
data
on
to
the
district
onto
the
distributed
web.
A
So
it's
basically
starting
to
build
metadata
patterns
and
cultural
patterns
and
incorporating
institutions
into
using
the
distributed
web
to
hold
and
redistribute
data
that
previously
was
held
only
in
centralized
ways,
so
that
and
yeah
I've
been
working
on
a
bunch
of
stuff.
There's
the
discussions
in
in
discourse
about
content
policies,
which
leads
to
interesting
considerations
of
like
long
term.
How
do
we
want
to
be
approaching
that
as
a
community
and
how
do
we
want
to
be
stealing
that
so
I've
been
working
on
some
of
that
stuff
behind
the
scenes?
J
Alright,
my
turn
so
I
started.
Writing
some
notes
on
the
things
that
I
was
going
to
mention,
so
they
did
forget
anything.
So
first
thing,
that's
really
mentioned
I
feel
us
are
working
on
really
cool
awesome
like
compiling
crypto
primitives
from
rusts
to
Azzam,
so
that
we
can
script
in
the
browser
and
essentially,
if
you
are
a
side
effects
user
today
and
you're
using
jessica's
in
your
browser.
One
of
the
things
that's
really
important
to
know
that
you
realized
basically
this
last
week,
is
that
if
you
load
a
webpage
from
an
on.
E
J
From
an
IP
model
maintain
s:
if
s
tries
to
do
any
crypto
primitive,
it
will
fail
and
it
will
fail
with
a
very
cryptic
error,
because
we
we
completely
missed
the
fact
that,
like
the
browser
just
simply
locks
at
API,
if
the
resource
is
being
loaded
with
all
HTTP,
and
so
you
get
somewhere
about
like
private
key
something
or
generate
key
function
is
out
there.
We
are
working
on
it
like.
J
We
now
are
aware,
and
that's
why
I
like
reading
this
new
optional
by
just
just
getting
cooler
primitives
to
compile
from
rusts
to
JavaScript
to
get
something
that
is
fast
and
if
that's,
what
we
need.
Another
solution
is
using
javascript
shims
and
what
we
want
to
have
in
the
short
term
like
spare.
Morale
is
just
have
better
error
logging
so
that
the
users
understand
what's
going
on.
J
We
saw
at
least
like
for
issues
of
people
like
getting
this
problem
and
it
was
like
actually
an
hour
the
same
time
for
weeks
like
no
one
ever
this
issue
and
and
suddenly
everyone
started
white
clothing,
just
IP,
perhaps
for
my
beads,
and
it
was
very
hard
to
reproduce,
because
every
local
host
test
or
like
ipfs,
that
I
you
/
IP
fest.
So
much
that
we
did
always
work
perfectly
but
yeah
I'll
post.
A
link
to
the
description
of
the
problem
on
the
nodes
go
ahead.
J
Essential
problem
is,
if
you
load
a
web
page
that
uses
Jessop
FS
from
a
non
secure
endpoint.
That
means
you
are
loading
a
web
page
from
HTTP
and
not
HTTP.
The
browser
like
it's
a
web
crypto
standard
definition
will
lock
the
web
crypto
API
and
not
let
you
call
any
web
crypto
cops
so
simply
cannot
use
those
primitives.
They
literally
are
not
there.
It's
not
even
like
the
browser
tells
you
web
crypto
is
locked
because
it's
an
results
in
an
unsecured
endpoint.
It
just
says
okay,
this
is
this
is
an
exist.
I
L
H
L
I
J
Alright,
so
this
is
an
issue.
There
is
a
solution.
We
do
the
crypto
in
JavaScript
or
we
do
the
crypto,
which
will
be
in
as
ingest
by
comparing
you
from
rest
and
week.
We
got
this
just
called
work
up
a
surprise
and
if
you
are
getting
sir,
this
is
the
issue,
and
there
is
an
issue
now
on
the
notes
for
you
to
track
the
solution
cool,
so
the
next
one
that
I
have
to
show.
You
is
ideally
deep
types.
J
So
last
Thursday
we
had
the
first
meeting
of
what
we
are
calling
like
a
howl
ends
for
IPL
david
ives.
There
is
like
multiple
verticals
that
people
have
been
showing
interests
to
discuss,
taco,
to
implement,
to
really
do
more
research
of
one
on
IPL
deal,
and,
and
so
now
we
have
a
call
every
thursday
and
for
no
three
p.m.
EDC
at
3
p.m.
UTC
and
part.
J
We
that
call
we
identified
things
that
need
to
be
done.
I
know
that
Steven,
my
turnip
I,
don't
like
he's
worked
there,
but
what
I
did
since
last
Thursday
was
adopting
all
of
the
proposals
for
presenting
the
format's
back.
So,
as
you
know,
we
have
this
interface
formats
back
that
enables
people
to
implement
more
formats
to
be
supported
by
IP
LD
from
the
AXI
war
doctor
above
atrium
I
know
that,
because
he's
now
working
on
it,
it's
pretty
awesome,
and
now
we
are
giving
a
new
update
to
that
spec.
C
C
M
Yeah
we
okay,
so
I
didn't
get
that
much
done.
Last
week
I
well
the
coop
sue
discovered
some
men,
the
middle
vulnerability,
with
the
current
state
of
my
quick
implementation,
so
I
decided
that
the
best
way
to
move
forward
with
this
is
to
first
do
all
the
interface
changes
in
Joel
idiot
appear
and
then
much
the
quick
specific
stuff
later
so
I'm
hoping
that
we
can
get
the
interface
changes
reviewed.
N
Hey
guys
so
last
week,
I
spent
some
time
working
on
still
working
on
circuit
kind
of
trying
to
get
everything,
cleaned
up
and
ready
for
release.
Unfortunately
didn't
get
as
much
time
as
I
expected
to
work
in
it.
I
was
gonna
have
a
demo
for
today,
but
that's
not
gonna.
It's
gonna
happen,
fortunately,
but
I
am
shooting
for
getting
everything
ready
and
prepped
for
possibly
release
either
this
week
or
next
week
or
whenever
we're
going
with
the
next
JavaScript.
O
Yeah
so
I
took
some
time
off
last
week
to
go
drive
over
somewhere
and
watch
the
Sun
go
away
for
a
minute
was
pretty
cool,
definitely
recommended
after
that
been
getting
back
and
mostly
just
spending
time
reviewing
people's
code
there's
a
lot
of
other
people's
code
through
you,
it's
pretty
great,
pretty
happy
with
it
all
and
I
look
forward
to
continuing
to
review
other
people's
code
because
there's
still
a
lot
of
it.
You
guys
are
fast,
don't
stop.
P
Was
sorry
last
week,
I
was
working
on
a
proposal
for
reorganizing
a
the
interfaces
in
go
idea
vest
so
that
the
the
dag
in
the
block
store
and
everything
fits
together.
You
can
find
this
on
the
ipfs,
slash
notes,
repository,
it's
issue,
255
I'd
like
comments
and
feedback.
The
current
issues
are
that
when
we
try
to
improve
or
implement
up,
I
feel
of
these
selectors.
P
The
current
system
just
won't
work
because
exchanges
and
things
like
that's
what
need
access
to
IP
LD
at
that
point,
not
block
or
not
the
block
service,
because
they
think
you
look
like
query.
I
bill
t,
so
we
can
have
to
reorder
stuff
anyways.
I've
also
noticed
that
a
lot
of
people
are
complaining
about
various
parts
systems
and,
like
this
piece
is
broken,
but
I
haven't
seen
any
like
overarching
plans
to
try
to
fix
everything,
so
basically
description,
Lily,
what's
wrong.
What
or
what
is
always
fixing
in.
Q
Q
B
E
L
Last
week
was
kind
of
weird,
because
I
was
reviewing
lots
of
code.
I've
noticed
one
problem
with
one
PRI
was
leading
people,
it
has
improved
coverage,
collector,
go
ipfs,
there's
a
small
problem.
There
were
because
I'm
trying
to
do
garbage
collect.
So
there
was
waste
problem
in
garbage
collector
between
MF
s
and
garbage
collector,
but
it
was
very
small.
The
window
is
very
small,
but
with
how
I
do
garbage
collection
currently
did
when
there
was
much
much
bigger,
so
I
have
to
refactor
that
pout
part.
L
C
Thank
you
so
up
in
a
bit
sick
last
week,
but
I'm
getting
better
I
fixed
the
bug
in
the
Gateway
deployment
that
was
kind
of
annoying
because
well
it
was
buffering
responses
to
disk,
and
so,
if
somebody
down
on
a
large
file,
then
all
these
gigabytes
were
buffer
to
disk
for
just
a
few
seconds
and
that
led
to
interesting
oscillations
in
free
disk
space,
but
that's
fixed
I'm.
Writing
a
bit
on
path.
C
H
C
C
C
J
This'll
be
past,
zero
25.2
was
released.
It
was
a
patch
release,
so
no
big
like
issue
with
a
lot
of
flair
because,
mostly
like
things
stay
the
same
but
better,
there
were
fixes
that
were
bubbled
up
the
we
peer
to
peer
tree
to
beat
swap
and
then
pictures
that
happen
in
bit
swap
that
were
below
up
to
just
a
few
FS.
We
already
got
some
reports
from
our
users
that
now
they
cb2
up
faster
and
also
more
reliable.
M
O
J
Because,
like
so,
we
have
been
getting
a
lot
of
reports
and
I
just
want
to
be
clear
on
that
all
by
people,
seeing
a
lot
of
browser,
resource
consumption
and
and
there's
like
two
main
reasons
for
that.
One
of
them
is
web.
Rtc
like
where
RT
C
is
a
monster
consuming
with
of
the
browser
like
in
every
single
plication
I
like
open
there.
Today,
video
chat,
I'll,
keep
you
in
and
so
on.
They
will
consume
a
lot
of
users
from
the
browser
and
we
use
WebRTC.
J
We
open
a
bunch
of
connections,
and
then
we
you
like
easy
to
say
like
all
of
these
tiny
packets
with
multiples
of
them
right
and
because
we
do
the
routing
the
main
Trev.
It
means
that
the
browser
main
tread
released
becomes
stretched
to
process
everything
that
it
needs
to
process
in
order
to
decide
which
things
to
send
to
which
connection.
J
Unfortunately,
like
most
browser
vendors,
only
see
web
RTC
as
a
video
and
audio
transport,
and
they
like
some
special
code
paths
that
benefit
those
use
cases.
But
still
again,
like
you
put
your
regular
web
video
chat
app
and
you
will
see
a
lot
of
reserves
being
consumed.
So
that's
that's
an
issue
that,
like
takes
me
to
the
next
point
of
connection
closing
the
other
issue.
We've
like
browsers
that
consume
a
lot
of
with
Jessica
passed
up
a
lot
of
resources.
J
It
is
really
the
fact
that,
like
people
are
putting
their
jessica
festival
on
the
main
trail
and
like
again,
the
main
trail
is
like
a
single
event
loop
and
like
we
do
a
lot
of
operations.
We
do
a
lot
of
stuff
like
we
do
quit,
though
we
do
like
browsing.
We
do
like
there
is
so
much
stuff
going
on
that.
J
J
I
G
J
It
is
two
things,
so
what
were
suggesting
is
like
extensions.
Actually,
you
have
a
process
for
themselves
right
like
they.
If
they,
even
if
they
process,
questions
that
you
can
restart
in,
doesn't
affect
the
main
page
like
the
app
of
the
user
can
still
run
and
free
though
you're
right
like
there
is
definitely
do
resource
consumption
and
also
do
to
promote
gracias
following
policies.
J
So
if
you
haven't
read
an
issue
like
if
you
put
a
tab
or
if
you
put
a
window
like
your
standard,
app
on
your
OS
on
front
of
chrome
and
if
it
covers
the
scene,
if
you
can
part
of
the
tab,
chrome
will
stop
that
just
like
blocking
right.
So
I'll
have
the
the
stuff
that,
like
the
node,
is
trying
to
do
to
be
like
very
fast
to
like,
for
example,
receive
one
place
and
process
which
blocks
it
as
to
send
it
back.
If
you'll
stop.
J
And
now
you
have
like
thousands
of
messages
waiting
for
that
thing
to
the
process
right,
so
it's
just
like
madness
for
for
the
for
the
nodes
to
two-and-oh,
and
that
gives
which
to
some
weird
situations,
Firefox
is
different
though,
and
and
this
is
again
connected
to
yes,
it
is
very
hard
to
do
these
things
in
the
browser,
but
we
can
be
so
much
better
in
a
way
that
we
manage
resources
right
now.
We
are.
J
We
have
a
very
good
grass
if
policy,
where
we
open
all
the
connections
that
we
can
and,
as
we
know
from
go
ipfs,
we
need
to
find
a
way
to
close
connections
graciously
and
effectively,
because
the
ones
that
aren't
used-
and
you
just
happened-
to
notice
that
in
Jessica's
faster,
because
it's
a
constrained
environment
with
a
lot
of
rules
and
restrictions
and
like
a
lesser,
a
smaller
number
of
connections,
makes
a
bigger
impact
than
like,
for
example,
go
ipfs
that
has
access
to
full
machine
resources.
So
yeah.
J
The
next
item
on
the
agenda
that
I
added
is
connection.
Closing
and
I
open
an
issue
describing
these
problems
and
more,
and
this
should
be
like
one
of
the
priorities
for
us
to
get
then
at
least
like
it
might
not
be
perfect,
but
like
we
definitely
should
not
let
the
node
crash
by
running
out
of
resources.
J
I
would
argue
that
it's
better
to
say,
hey,
like
can't
really
open
a
new
connection
already
have
so
many
might
pick
one
to
drop,
then
just
like
weather
knows
kill
itself,
because
you
just
becomes
too
eager
to
open
connections
to
other
nodes,
as
Jeremy
once
put
it
I.
Guess:
ipfs
is
a
social
butterfly
like
what
stuck
with
everyone.
J
A
A
J
Well,
because
video
are
connections
have
special
flex
inside
the
browser's.
Let's
say
this
connection
should
not
ever
stop,
which
is
all
the
same
for
other
channels.
I
agree.
We
we
should
like
just
again
raise
this
problem,
something
that
they
need
to
fix,
but
it
is
something
that
we've
been
talking
with
them
for
more
than
a
year
now
for
sure,
as
a
substantial
problem
of
enabling
peer-to-peer
on
the
web.
E
J
J
So
you
can
see
my
screen.
This
is
just
the
browser
brave
browser,
a
repo
we
in
a
branch
called
feed
ipfs
and
if
we
locate
the
extensions
folder,
we
look.
There's
like
three
extensions
here.
One
is
brave
brave.
The
brave
browser
basically
is
an
extension
around
itself
and
then
there's
like
a
torrent,
extension
and
ipfs
extension.
And
if
you
look
at
the
IP
path
extension
you
have.
J
J
What
this
is
doing
is
all
these
steps,
starting
in
your
notes,
waiting
for
the
middle
to
get
ready
connecting
to
the
bootstrapper
nodes
and
carrying
this
ash
and
again
once
it
gets
that
B
sash
doing
the
bit
swap
thing.
It
replies
back
using
this
car
back
here
and
yeah
like
so
they
ask
that
I
was
welding.
You
just
hate
right,
it's
like
a
very
simple
old
world
of
Jessica
fests
running
and
this
force
falls
directly.
What
was
your
proposal?
J
You
have
to
declare
all
of
the
endpoints,
that's
that
African
dial
before
hands
before
it
loads
and
there
is
a
way
to
automatically
load
new
things,
so
so
yeah
again
telling
Jess
if
you're
spreading
on
the
Brit
browse
here
expect
next
week.
Next
thing
I
want
to
come
back,
will
actually
buy
clothing,
several
ashes
and
webpages,
and
so
on.
As
I
said,
I've
been
very
focus
on
this
content
security
policy
days,
but
yeah
it
is
happening
who
were
also
integration.
K
J
Questions
there's
a
like
a
big
PR
like
so,
if
you
go
to
my
user
account,
there's
a
fork
and
then
a
PR
directly
to
the
main
repo
on
the
brave
organization
and,
like
the
PR,
basically
describes
the
happy
cat
venture
I've
been
on
just
like
to
figure
out
how
to
ads
the
protocol
and
learn
how
to
solve
the
CSP.
How
to
get
the
Chrome
extension.
To
give
me
the
ability
to
register
your
protocol-
and
that
was
another
so
journey,
I
see
that
you
have
your
hand
up
so.
O
J
Right
now,
it
is
I
actually
already
requested
the
team
to
give
me
a
streaming
handler
right
now.
The
only
thing
that
move
on
which
is
for
collection
supports,
so
that's
exactly.
It
was
appointing
yeah
well
from
what
I
understand
it
is
totally
doable
to
do
this
streaming
thing.
They
just
abandon
sure
it's
a
security
concern
from
the
team
like
they
they
were
yeah
I,
haven't
received
a
clear
message
on
that:
I
get
to
not
be
a
problem.
It
should
be
just
like
when
today,
but
yeah.
J
Another
thing
that
I
want
is
to
be
able
to
also
like
send
a
webpage.
For
example,
I
would
love
for
an
IP
fest
file
to
build
building
and
like
just
like
clothes.
First,
the
webpage,
the
back,
shows
a
progress
bar
or
something,
and
they
just
see
now
that
page
to
share
that,
because
it's
been
all
done
like
if
you,
if
you're
downloading
a
PDF,
for
example,
you
want
some
signal
to
the
user.
J
A
J
The
there
is
so
using
your
electron
or
using
any
electron
works
if
you
use
on
the
render
process,
so
that
is
equivalent
to
a
browser
page
on
the
electron,
which
is
more
or
less
what
is
happening
here.
It's
very
similar
to
these
contexts
that
I'm
running
inside
the
browser.
What
doesn't
work
yet
is
running
just
ipfs
in
the
node.js
process,
also
known
as
the
main
process
of
electron.
J
Another
solution
is
just
to
remove
all
of
the
native
dependencies
and
I
just
use
JavaScript
pure
JavaScript
and,
like
ever
build
of
Giuseppe
ifs.
That
might
be
slower
and,
like
you
know,
it's
like
covering
sin
might
be
slower,
but
it
works
for
the
native
the
main
process,
the
node.js
process
of
electron.
If
people
want
to
run
JSI
PFS
they're,
the
the
other
thing
that
is
interesting,
the
notes
in
the
conversation
of
electron
is,
if
you
want
to
run
on
the
node.js
process
and
knows
about
defense.
Another
thing
that
you
can
do
is
run
a
go.
J
Ip
best
note
we've
at
the
SAFS
api
client.
So
that's
what
orbit
did
for
a
long
time
right
like
orbit,
was
always
running
a
go.
Ipfs
nodes
we
just
BFS,
API,
client
and
electron,
has
a
very
cool
feature
where
you
can
pass.
You
can
grab
an
instance
of
an
object
and
say
it
sends
this
to
the
other
process
and
what
it
does
creates
automatically
an
RPC
that
links
any
call
that
you
do
on
the
renderer
process
to
the
node.js
process.
J
So
so
very
long
time
like
orbit,
was
running
and
like
doing
everything
on
the
render
process,
but
the
nose
was
associated
and
running
on
a
little
J's
process.
I
can
write
more
na
ki
fee.
If
you
I'd
like
to
have
a
meeting
and
people
are
asking
for
this,
I
can
write
more
notes
all
these
options
that
could
be
helped
yeah.
A
Always
more
even
just
there's
this
other
set
issue:
I
just
posted
these
in
the
chat
in
the
zoom
chat.
There's
the
issue
a
fourth:
you
have
been
gradually
updating
and
then
there's
a
new
discussion
on
discourse
about
it,
so
just
toss
in
some
sort
of
reference
point
or
suggestion
for
the
people
who
are
trying
to
build
stuff
this
week.
What
should
they
do
rather
than
because.