►
From YouTube: ⚡️ⒿⓈ Core Dev Team Weekly Sync 🙌🏽 May 13, 2019
A
Okay,
hello,
welcome.
It
is
it's
made
of
13
fitness,
the
Jas
core
dev
team,
weekly
sync
I
am
Alan
your
host
very
nice
to
see
you
all
hope
you
are
well.
It
is
time
that
time
again,
you
know
that
when
we
do
that
thing
that
we
always
do
every
week
and
tell
you
what
we
did
last
week,
what
we
gonna
do
this
week
and
what
we
currently
blocked
on.
So
let's
do
that.
My
name
I
got
here.
A
First,
so
I'm,
gonna,
quick,
I'm
gonna
quickly
go
first
and
tell
you
about
what
I
was
doing
last
week,
which
was
mostly
working
on
content
for
ipfs
camp
I.
Put
some
sneak
peak
images
there.
If
you
look
at
the
actual
preview
working
on
like
a
dag
visualize
a
visualizer,
so
you
can
just
drop
your
file
on
the
on
the
on
the
thing
and
it
will
draw
a
graph
of
the
resulting
dag
that
gets
built
for
further
your
particular
configuration
so
like.
If
you
I,
guess,
add
stuff
you
get
you
use
like
a
balanced
dog.
A
If
you
use
NFS
you'll
be
using
a
trickle,
dag,
and
so
this
tool
will
allow
you
to
kind
of
toggle
between
them
for
any
one
given
file
and
change
tweak
all
the
dials
and
stuff,
but
it
basically
allows
you
to
see
how
things
are
laid
out
in
ipfs
and
you
can
also
drop
multiple
files
on
it,
and
then
it
will
show
you
how
the
it
will
point
to
the
any
deduct
blocks
that
you
have
there
so
demonstrate
the
deegeu
ping
powers
that
ipfs
has,
which
is
super
rad
anyway.
That
is
super
early
work
in
progress.
A
That's
why
it
is
pink
because
the
better
colors
haven't
come
along
yet
anyway.
So
that's
I
was
working
on
that
last
week,
so
that's
kind
of
fun.
Still
what
allowed
to
do
for
that,
because
that
sucked
up
and
hold
on
my
time,
but
it
was
really
fun
so
I
kept
doing
it.
I've
also
released
the
JSF
FS
HTTP
client
version
31
thanks
to
Dirk
for
all
of
the
work
on
the
ref
stuff,
which
is
now
in
there.
A
So
that's
in
there
now
and
there
is
a
pull
request,
which
is
on
the
cusp
of
being
merged
for
GSI
PFS,
which
implements
the
refs
endpoint
and
that
will
be
available
in
0:36,
speaking
of
which
I
started
the
release
process
phone
0:36.
So
hopefully,
at
some
point
this
week,
or
maybe
early
next
week,
hopefully
this
week,
we'll
get
a
new
version
with
a
bunch
of
things:
click
on
the
link
to
see
the
list
of
stuff.
A
That's
not
interesting,
I
enable
greenkeeper
on
Jess
ifs
and
the
client
and
my
interface
I'm,
just
cool
just
to
see
how
it
works.
It
seems
to
be
okay.
So
far,
it's
pretty
pull
request
II,
so
that's
been
fun
and
what
else?
But
other
than
like
the
camp
content
I've
mainly
been
working
on
the
the
tweak
which
will
make
any
CID
V
once
the
ideas
that
we
create.
If
you
to
string
them,
they
will
become
base
32
encoded
strings
rather
than
base
58,
and
so
the
that
just
is
was
basically
a
small
PR
suggests
the
ID.
A
But
everything
depends
us
the
HSC
ID,
so
just
been
kind
of
propagating
that
through
and
there
we
had.
There
is
an
issue
with
like
IP
OD,
because
we
can't
really
release
a
new,
a
breaking
change
of
IPL
B
because
already
released
a
breaking
change
of
IPL
D
to
add
the
async/await
stuff,
so
we're
having
to
bring
that
stuff
into
a
new
release.
So
it's
just
it's
just
fun
times,
but
it
should
be
good
eventually.
A
A
B
A
Yeah,
it's
happening
it's
just.
It
just
means
as
a
bunch
of
dependencies
that
we
need
to
get
merged
beforehand
and
I
talked
to
Alex
about
what
he's
doing
cuz
he
much
of
his
work
that
he's
doing
for
the
eunuchs
of
s,
importer/exporter
and
MF
s.
Work
also
will
need
to
be
merged
in
some
way
and
used
so,
and
he
thinks
he
will
be
able
to
get
that
done
before
the
end
of
this
week.
A
B
B
That
there's
an
issue
but
maybe
shouting
support
around
and
FS.
There
was
around
actually
shouting
support
around
HTTP
gateway,
because
that
one
is
need
support
for
shadow
directories
which
were
was
missing.
It
was
like
a
blind
spot.
Can
you
can
you
link
me
to
the
that
I
will
put
it
in
my
update.
So,
okay.
A
C
Yeah,
it's
good
to
be
back
sure,
yeah,
so
I'm
in
my
egg
wash
so
this
might
be
a
little
bit
choppy
but,
as
Alan
said,
I've
been
working
in
a
bunch
on
the
raft
endpoints,
so
I
still
got
there
just
fixing
up
a
couple
of
things
in
the
unit
tests
and
then
I've
also
been
working
behind
garbage
collection.
So
we
had
a
bit
of
I'll
have
a
link
to
the
issue,
but
basically
we
decided
to
go
with
the
same
algorithm
as
go
users
to
go
right:
EFS,
which
is
a
mark
and
sweep
algorithm
and
I.
C
B
B
A
A
D
Last
week,
sort
of
work
on
ipfs
camp
course
that
will
be
featuring
some.
The
PDP
concepts
continued
work
on
that
this
week
as
well.
Last
week
we
released
JJ
asleep
Peeta
Peeta
update
that
includes
mist
signing
for
pub/sub.
So
anybody
who
has
that
whenever
go
decides
that
they
want
to
release
their
breaking
change,
we
won't
lose
Interop.
D
D
We
may
need
to
look
at
like
keep
alive
for
the
HTTP
API
for
the
daemon
in
order
to
prevent
socket
hang-ups
from
happening
for
the
CLI
until
we
can
get
better
latency
on
the
network
for
Lib
p2p,
which
is
the
next
series
of
things
that
I'll
be
working
on.
We
released
the
lip
PD
p.m.
DNS
update
that
included
the
compatibility
layer
for
gota
Allen
added.
So
that's
there.
Also
the
mdns
spec
got
merged
in
the
p2p,
so
hopefully
we
can
start
collaborating
on
getting
go.
D
Nsj
implementations
to
the
right
place
so
that
they
also
work
with
rust
and
then
I
started
on
the
lip
PDP
natural-bristle
work,
which
is
starting
with
the
identify
push
protocol
in
Jas,
which
is
already
an
NGO
which
lets
us
as
our
addresses
change
with
NAT
negotiation.
Well,
I
should
be
able
to
push
those
address
updates
out
to
our
peers,
so
they
can
come
and
requests
on.
Our
new
addresses
this
week
continue
on
the
ipfs
core
camp
coursework,
the
DHT,
PR
and
then
I'll.
Also
look
at
starting
Auto
NAT
when
I
get
identify,
identify
push.
D
Protocol
auto
net
is
a
protocol
Inlet
p2p
that
lets
us
communicate
with
peers
in
the
network
to
figure
out
what
our
net
status
is
and
also
our
observed
address
better,
which
is
why
we
need
the
identified
protocol
so
that,
once
we
have
updated
addresses
that
we
can
for
those
off
to
everybody.
Does
anybody
have
questions
I.
A
D
So
pops
up
message
signing
basically
what
was
going
on
is.
We
would
send
a
message
in
pub/sub
and
you
don't
know
where
it
came
from,
there's
no
way
to
verify
who
it
came
from.
So
what
we
do
with
signing
is
we
actually
take
the
message
and
you
use
our
private
key
and
you'll
actually
sign
that
whole
message
and
then
on
your
signature
to
it
and
then,
when
we
send
that
along
to
peers,
peers
will
be
able
to
verify.
D
A
E
According
to
the
previous
version
of
the
interface
connection,
which
was
not
really
a
connection,
and
it
makes
me
like
think
and
also
analyze
the
go
implementation
and
after
that,
I
ended
up
making
a
new
proposal
for
a
completely
different
interface
connection
and
this
game.
Also
with
the
interface
trim.
I
have
the
link
for
the
issue
with
the
proposals
Jacob
already
provided
some
feedback,
I
think
Alan
and
Derrick
already
saw
it.
E
I
have
been
iterating
according
to
the
feedback
and
I
think
it's
getting
in
a
really
better
shape
than
it
was
before
other
than
that
I
applied
to
the
IP
in
Liberty
web
RTC,
star
8,
the
previous
version
at
security
and
performance
problems
and
I
related
it
in
the
context
of
islands
of
played
for
the
Jessup
fs1.
Also
yeah.
Today,
I
got
the
interrupt
test
for
Ibanez
over
pups
that
finally
green
again
in
Darwin
margins,
so,
finally,
that
PR
is
merged
so
for
this
week,
I've
been
also
carrying
up
straightening
the
reviews
for
the
gospel
implementation.
E
The
current
status
is
that
the
the
prsti
completely
drain
the
code
seems
good
according
to
the
specs
and
according
to
the
implementation
of
go,
I
tried
to
review
with
all
those
in
mind
and
basically
we
are
I
think
in
the
finish
line
to
to
merge
that
the
pr.
Basically,
what
we
are
missing
now
is
interrupt
tests
with
flood
syrup
and
interrupt
tests
with
bola
p2p.
E
I've
been
discussing
this
with
my
period,
and
in
do
it
also
with
we
meet
again
and-
and
we
will
probably
start
by
merging
the
current
state
of
the
pier
after
my
last
review
and
it's
right
from
there,
because
the
PRI
is
getting
really
big
and
massive
to
review.
So
I
hope
to
get
it
initial
version
of
the
super
submerge
this
week.
Also
I
want
to
update
the
interface
connection
for
requests
with
the
new
proposal
and
with
just
for
being
used
and
also
create
the
interface
stream.
E
D
D
The
four
pub/sub
we're
probably
gonna,
need
to
look
at
updating
DJ's
configuration
because
right
now
it's
like
flood
subs
hard-coded
in
there,
so
we're
probably
going
to
need
to
make
it
so
that
pub/sub
is
actually
configurable
and
then
rollout
configuration
changes
for
pub
sub
to
JSI
PFS
so
that
people
can
actually
choose
what's
up
or
gospel
sub.
Rather.
A
A
G
I'm
having
a
few
errors
on
the
I
pianist,
observe
tests
and
that's
because
I
needed
to
actually
include
the
recursive
through
defaults
in
these
be
ours,
or
else
it
would
be
really
hard
to
make
HTTP,
which
tests
will
go
and
JSP
files
which
tests,
which
tests
with
the
API
in
J's
work
everything
work
together
and
those
days
fast,
so
I
needed
to
include
that,
but
that
breaks
observe
I'm,
trying
to
figure
out
what's
the
problem
in
there,
but
I
think
that
will
be
the
last
problem.
I,
hopefully
encounter
on
these
stuff.
G
Changing
the
seal
I
test
a
little
bit,
I
discussed
this
with
Alan,
because
I
found
that
most
of
the
tests,
at
least
on
the
IPS
subsystem,
are
kind
of
duplicated
in
there,
and
we
actually
don't
need
to
to
test
everything
again
because
we
already
test
in
the
interface
core
partially
on
Jessup
EFS
core
and
also
through
the
HTTP
API,
so
testing
it
again
on
the
CLI
is
kind
of
there's
no
purpose
to
do
that.
But
we
still
have
some
logic,
some
simple
logic
that
which
we
would
like
to
test
about
CLI
specific
stuff.
G
So
that
took
me
a
while,
because
this
really
is
not
actually
easy
to
test
right
now.
But
I
found
a
way
which
is
pretty
simple
and
I
will
push
that
those
changes
to
the
PR.
Hopefully
you
guys
will
see
it
and
give
some
feedback,
so
we
can
roll
out
to
the
other
systems.
I
also
started
working
on
the
DNS
resolver,
there's
still
much
to
do
there,
but
it's
I
started
the
work.
G
I
did
a
bunch
of
stuff
for
a
PFS
camp
Corsi
with
Vidal
and
Jim,
and
hopefully
this
week,
I'll
be
able
to
finish
the
file.
Dom
API
requests,
so
we
can
release
HTTP
clients.
We
that
stuff
in
I
will
also
probably
be
doing
a
couple
more
sessions
with
Jim
and
Lydell
for
the
ipfs
camp
and
I'll
be
attending
in
the
DTN
in
Berlin.
That's
that's
me.
Any
questions.
A
B
Try
to
be
brief,
so
there's
a
new,
stable
release
of
I
profess,
companion
and
I
mention
it
because
it
includes
a
prototype
of
embedded
HTTP
gateway
in
bread
that
can
actually
open
with
Wikipedia,
which
is
quite
cool,
and
that's
embedded
note.
So
no
third-party
software,
external
node
and
stuff,
like
that,
some
related
releases
and
work
around
IP
first
camp
course
see
with
Hugo
and
Jim
and
sort
of
maybe
not
like
it's
not
a
hard
block,
but
the
work
Alex
did
for
adding
sharding
support
to
J's
API
first
gateway.
B
I
retested
those
changes
and
those
changes
actually
are
with
activist
companion
to
make
their
Wikipedia
org.
So
it's
like
very
good
and
I,
fully
adores
reviewing
those
peers
and
there's
a
Irie
based
another
PR
4j
31st
Gateway
updates
to
add
IP
NS
names
by
support,
but
that
depends
on
all
the
Guinness
peers
that
will
eventually
Lunt
and
the
next
week
will
be
IP
first
comp
and
compile
brave
switcheroo
and
that's
all.
A
H
I'm
still
deep
in
a
single
Weiland
for
UNIX
a
person
on
the
first,
the
Tesla
passing
in
order
modules,
the
interface
no
surpassing
the
Interop
tests
are
almost
passing.
The
last
piece
of
the
puzzle
is
the
trickled
out
the
triple
deck
chairs
implementations
used
to
have
no
like
kind
of
nothing
in
common
at
all
would
to
go
on.
It
doesn't
produce
the
same
structures
all
of
the
same
hashes,
so
I'm
making
a
do
that
which
is
cool.
H
It's
also
like
a
whole
bunch
you're,
just
pausing,
you
know
stairs
actually
slow
and
any
one
is
way
faster,
though
still
creating
incorrect
co
DS
at
the
moment,
but
it
will
soon
be
fast
and
correct,
but
is
faster
and
correct.
That's
going
to
be
me,
I'm
also
going
to
DT
and
Kampf
at
the
end
of
the
week
same
as
vodka
and
did
a
bit
on
I.
Just
can't
planning
for
the
the
p2p
course
was
Jacob
and
rail
and
unimaginably.
Something
like
that.
That's
gonna,
be
me.
Thank
you
very
much.
A
Okay,
perfect
all
right.
We've
still
got
two
minutes,
isn't
on
how
many
questions
for
Alex,
real,
quick,
otherwise,
we'll
move
on
to
questions
questions
so
I
did
someone
have
any
questions
for
Alex
I,
just
clicked
away
from
the
three
okay
all
right?
Next
up,
we've
got
so
general
question
someone.
Should
we
do
an
iterative,
non-breaking,
eventually
I
think
it
arises
in
modules
such
as
the
DHT.
E
Yeah
I
asked
that,
because
Kumar
visited
the
PR
who
were
basically
extracted
some
of
the
work
that
directed
to
us
to
a
new
PR
which
does
not
have
breaking
change
and
I,
don't
know
if
I
should
spend
time
in
reviewing
that
now
or
if
I
should
just
wait
until
the
most
priority
yard
slant,
and
we
do
our
previous
approach.
I
know
what
your
opinions
my.
A
Gut
feeling
is
that
if,
if
it's
just
been
done
by
someone
randomly,
then
we
can
look
at
it
and
get
it
in
like
I
I.
It's
essentially
kind
of
I
mean
it's
good
work,
but
there
we're
gonna
have
to
undo
a
bunch
of
it,
I
believe
when
we,
when
we
completely
switch.
So
there
are
some
some
wasted
work
there
and
I'd
rather
not
look
for
that
to
happen.
E
Yeah
I
have
more
or
less
the
same
opinion
as
you
I
think
that
we
should
not
spend
that
much
time.
But
if
the
review
does
not
take
that
much
time
and
after
that,
we
only
have
to
remove
promise
if
I
or
a
few
things
I
think
like.
Maybe
it's
worth
to
move
because
after
that,
it
will
be
easier
also
to
get
the
module
to
the
final.
Why?
Okay.
D
A
Yeah
I
mean
the
DHT
is
a
tricky
one,
because
it's
it's
having
a
lot
of
active
development
right
now.
It's
only
just
come
into
being
and
yeah
this
it
it's
being
changed.
Quite
a
lot
and
yeah
in
in
hindsight,
I
think
the
actually
converting
it
to
async/await
was
probably
a
little
bit
soon,
but
yeah
I.
Guess
it's
good
to
be
prepared.
A
Mattituck
about
this,
because
we're
not
gonna
have
that
when
not
for
it,
like
I,
didn't
ever
expect
that
for
every
module
we'd
be
able
to
do
our
plan
of
we
do
everything
at
the
bottom
and
slowly
propagate
everything
up
and
nothing
stays
sat
there
for
a
long
time
and
it's
just
the
ideal.
If
we
have
to
do
this
for
someone
else,
then
let's,
let's
do
it,
that's
fine
and
but
like
I,
just
want
to
avoid
doing
too
much
work.
That's
going
to
just
be
undone
later
yeah.
A
If
it's,
if
it
makes
sense,
then
absolutely
yes,
let's
do
it
so
I
would
say:
let's,
let's
look
at
the
work
and
and
get
it
merged.
If
we
can.
A
I
So
the
MF
s
tutorial
is
getting
so
close
to
feeling
like
it's
done.
There
are
a
couple
little
bugs
near
the
end,
but
you
can
now
read
through
the
entire.
You
know
text
of
the
lessons
and
what
the
exercises
are
to
get
a
sense
of
what
everything
will
be
so
I'm
very
open
to
feedback
there.
Also
a
lot
of
you
on
this
call
are
working
on
my
PFS
camp
materials,
Lydell
reached
out
to
me
and
schedule
some
time
to
talk
about.
I
You
know
talk
through
what
the
plans
are
see,
whether
anything
fits
in
front
of
school
content,
whether
it
does
or
not,
I'm
happy
to
have
that
conversation
with
any
of
you.
So
please
don't
hesitate
to
reach
out
to
set
something
up,
so
we
can
think
through
that
together
and
I
can
give
you
a
sense
of
kind
of
the
limitations
of
the
platform
and
what
works
well.
There.
A
Cool
Thank,
You
Terry,
all
right,
that's
cool!
That's
it
apart
from
other
notes,
so
I'm
out
on
Friday,
if
I
say
what
he
said,
that
he's
partially
out
on
Wednesday
and
Friday
and
I
think
that's
all
we
have
for
today.
So
thank
you
very
much
for
joining
us
and
I
will
see
you
next
week
for
another
round
of
what
are
you
doing
in
ipfs?
Have.