►
From YouTube: ⚡️ⒿⓈ Core Dev Team Weekly Sync 🙌🏽 July 16, 2018
Description
Weekly Call - https://github.com/ipfs/pm/issues/650
A
Okay,
hi
everyone
welcome
to
the
desk
or
dev
weekly
sync
up
for
the
Jessa,
because
just
a
few
meetings
of
July
16,
the
last
time
we
synced
up
with
this
two
weeks
ago.
My
speaks
of
stuff
cotton
Monza,
we
love
code,
but
here
we
are
again
to
doing
things
today:
special
cortex
Inc,
because
we
are
trying
out
pure
path.
We
already
found
a
bunch
of
problems
this
morning
we
already
fixed
a
bunch
of
things,
but
we
are
still
turned
a
little
bit
with
it.
A
A
A
A
A
Yes,
now
that
we
can
use
that,
we
don't
need
to
like
us
anymore,
so
so
I'm
working
on
it
as
I
kind
of
like
my
old
flash
new
pet
project
per
se
like
interesting
demo
for
the
web
community
I've
been
reviewing
out
yours
people,
please
give
me
feedback
on
your
cares.
Let's
get
it
done
this
week.
We
need
to
have
it
finished
by
Wednesday.
This
is
this
really
important?
Everyone
should
have
done
in
frozen
or
cares
by
you.
A
You
might
have
seen
comments
for
me
on
spreadsheet
or
github
repos
I
read
the
release
of
Guardian
vests
I
did
the
release
dance
of
like
the
NPM
modules.
So
now,
if
you
enjoy
the
best
image
controller,
you
get
0
for
16,
not
0
for
15
anymore,
and
with
that
I
started,
doing,
interrupts
testing
like
sometimes
still
pass
but
I
know
I'm.
A
Seeing
a
lot
of
failures
from
girl-
specifically,
it's
always
go
to
jails
of
files,
bigger
than
1
Meg,
and
so
the
other
way
around
it
doesn't
happen,
but
you
have
to
just-
and
we
know,
there's
a
lot
of
like
stream
website
updates.
So
we
need
really
to
spend
some
time
debugging
what's
happening
there
yeah.
A
Essentially,
we
also
need
to
find
ways
to
encourage
the
Galactica's
team
to
run
in
topdesk
before
doing
a
release,
because
this
didn't
have
to
happen
this
way,
but
more
on
that
soon
yeah
and
then
we
have
an
important
event
in
another
month.
This
is
EPS
Emmett,
throw
we
recommend
people
like
attending,
or
at
least
watching
a
live
stream
and
we're
going
to
have
two
workshops
there.
So
I
am
I
designing.
A
They
are
not
cutting
the
content,
helping
others
like
to
the
weber,
and
then
we
have
the
jessica
satire
website
which
has
made
a
lot
of
progress
but
still
needs
to
get
done
so
I've
been
reviewing.
It
I'm
really
blocked
on
the
central
testing
like
right
now,
like
I'm
as
good
as
anyone
else.
I
could
really
do
my
get
like.
There
is
a
lot
of
things
there.
A
We
know
that
the
three
monsters
was
it
things
that
change
in
my
PFS
and
and
we
need
to
fix
that
because
it
means
like
we
cannot
download
files
larger
than
one
make
from
quite
DFS
in
useful
time,
like
they
finish,
but
it's
just
taking
like
one
order
or
two
orders
of
magnitude
more
time
than
it
was
before.
It's
really
obnoxious
and
then
next,
let's
finish,
the
cures
continue
just
because
we're
putting
over
the
workshops,
I'm
visiting
London,
so
hi
people
so
I'll
be
around.
A
If
you
want
to
hang
up
and
and
hack
together
and
yeah,
I
want
to
explore
like
getting
a
piece
a
little
sites
that
we
cannot
share
with
the
rest
of
community,
so
that
people
can
access
publishing
websites
to
IP
passed
very
easily
and
also
every
night
chatting
a
lot
like
the
idea
of
creating
a
workshop
er
for
IP
bus.
Let's
see
if
we
can
like
make
something
interesting
in
LA
of
being
together
but
yeah.
That's.
C
A
E
Okay,
so
me
so:
I
went
to
the
Deaf
meeting
in
Berlin
and
I
met
all
of
your
beautiful
faces
for
the
first
time
in
real
life,
and
not
all
of
them
a
lot
of
them,
and
it
was
great.
So
it
was
super
awesome
to
see
you
all
and
yeah.
So
yeah
like
David,
said
release
it
Jess
I
professor
0:30,
live.
It
almost
went
without
a
hitch
up
until
the
last
minute,
but
yeah
it
went.
It
went
okay,
it
went
pretty
well.
So
that's
that's
super
exciting
news.
E
You
can
create
it
and
initialize
it
and
not
start
it.
You
can
create
it
and
initialize
it
and
start
it
and
all
of
those
kind
of
combinations.
The
way
the
code
was
structured,
the
pre-start
was
being
called,
but
in
most
of
those
combinations,
pre-start
was
being
called
twice.
So
that's
probably
not
great
for
that
for
the
node
in
terms
as
well.
E
I
don't
know
if
it
was
messing
up
some
other
bits
and
pieces,
but
in
terms
of
performance
I
guess
we
don't
want
to
be
recreating
things
twice,
so
hopefully
mine
a
performance
bump
by
kind
of
addressing
that,
and
so
that's
that
was
what
I'm
doing
this
week
after
the
meeting
I'm,
not
anything
at
the
moment.
At
the
weekend,
I
did
a
little
hack
on
Jeff's
idea.
E
Fests,
you
add
a
CID
base
option
for
to
allow
people
to
actually
get
space
for
two
encoded,
C
IDs
out
of
ifs,
when
you
add
things
to
it
and
that's
important,
because
if
we're
going
to
have
C
IDs
in
web
browsers
as
as
an
origin,
then
they
need
to
be
case
insensitive
and
face.
Vertice
looks
to
be
the
kind
of
best
option
for
that
at
the
moment,
so
we,
the
idea,
is
to
eventually
move
to
base
32
encoded,
Co
DS
by
default
and
version,
and
by
virtue
of
that,
having
v1c
IDs.
E
But
this
why?
What
I
packed
on
was
the
first
there,
which
actually
leaves
a
default.
They
are
but
allows
you
to
add
things
to
my
PFS
and
specify
command
line
option
so
that
when
you
get
your
output
back,
the
see
IDs
that
are
printed
to
your
console
or
delivered
to
your
browser
are
in
encoded
using
base
32
and
mod
codec.
E
Yes
and
my
next
stop,
my
next
kind
of
actions
are
to
finish
off
that
there's
a
whole
bunch
of
stuff
around
that,
while
I
was
doing
it
so
I
just
need
to
clean
all
of
that.
Bad,
especially
pinning
because
pin
English
is
very
much
assuming
base
58
coded
hashes
in
the
pen
second
set
file,
so
that's
gonna
need
some
some
stuff
done
to
it,
and
but
it
it's
gonna,
be
a
big
drop
and
something
that
I
will
be
doing
over
the
quarter.
E
Yes,
I
profess
API
tests,
that's
kind
of
working
progress,
they
are
getting
better,
I,
think
they're
getting
better
and
yeah
and
then
and
then
once
they're
kind
of
stable
to
a
kind
of
something
I'm
happy
with
and
I'll,
be
able
to
then
look
into
why
they
take
so
long
to
execute,
and
so
I
want
to
identify
where
that
bottlenecks
are
and
then
make
a
plan
for
dealing
with
that.
And
then
the
other
okay
I've
got
is
to
create
a
attend.
E
A
Oh
by
the
way
next
and
I
guess
easily
created,
everyone
should
be
the
things
that
you'd
planned
for
next
week,
not
necessarily
the
entire
quarter.
Otherwise,
the
fact
becomes
like
you,
you
will
focus,
but
but
it's
good
I
know
that
you're
thinking
about
all
these
things,
I
guess
like
the
feedback,
is
let's
like
pick
a
couple,
rather
than
all
the
key
results
at
the
same
time.
Otherwise
you
might
ever
be
overwhelmed
on
the
bottlenecks
for
execution
time.
I
think
like
if
you
like,
to
check
my
issue
on
a
capacity
CTL.
A
A
Like
cut
the
the
execution
time
by
50%
with
a
demon
generation
and
then
focus
all
your
attention
on
the
best
32
things
and
make
it
happen,
it
will
be
very,
very
viable
for
a
lot
of
the
discussions
by
like
seeing
it
happen
in
quiet.
It's
not
just
like
an
idea.
It's
just
like
an
RC.
It's
actually
care
that
works,
that
people
can
see
working
on
the
browser.
The
the
other
thing
like
in
a
bonus
point
for
that
soul
in
their
fur
is
the
fact.
A
If
you
put
that
support
for
UNICEF
s4r
all
leaves
right
now
we
have
draw
eunuchs
in
fest
notes,
which
are
like
just
a
buffer
with
the
eunuchs
it
has
wrapping.
But
if
you
have
just
like
a
row
leaf,
which
is
like
the
full
block
he's
done
out
of
that,
not
only
saves
overhead
but
also
gives
this
compatibility
with
files
added
to
the
URL
store
of
go
offense
that
sound
so
like
is
just
like
a
simple
trick,
because
we
decided
check
the
type
of
the
leaf.
A
F
All
right
so
I
was
also
at
the
Deaf
meetings
in
Berlin
and
there
we
create
discussions
about
two
made,
a
deep
dive
about
graph
sync,
and
it
was
great
so
there's
also
like
some
communication,
I
flipped,
the
crew
requests
with
the
details.
But
if
disgusts
me,
I
also
found
a
vulnerability
in
ipfs
which
is
not
fixed
yet,
therefore
not
disclosed,
but
it's
a
good
one.
So
we
found
the
studio
discussions
and.
F
We
also
the
discussion
about
IP,
LD
and
IP
on
research.
It's
also
an
issue
with
this
stuff.
There
you
can
just
read
it,
and
next
I
will
be
a
pre-press
stuff
for
the
D
Web
Summit
and
yeah
also
prepare
for
the
foss4g
conference,
which
is
another
conference
in
of
august,
but
you
will
do
this
in
parallel,
because
yeah
it
needs
to
be
some
type
of
time
idea.
A
A
D
We
came
up
in
the
in
Berlin
and
we
already
have
a
plan
a
likes
to
ask
that
vid
to
have
the
the
copy
that
is
still
missing
until
Friday,
because
we
kind
of
want
to
on
the
next,
not
this
one,
but
the
next
week
on
the
Tuesday
on
Tuesday
will
like
to
have
like
a
complete
production,
ready
version
of
the
site.
So
we
have
some
kind
of
buffer.
D
A
A
Let's
continue
using
that
document
about
the
first
iteration,
actually
bag
documents
to
keep
updated
like
both
sites.
Like
you
tell
me
what
you
have
discussed
in
plan
and
I'll,
give
you
feedback
right
here
on
the
dock.
That's
why
it
works
very
well
because,
like
I
get
all
the
implications
on
my
books,
but
yeah
you're.
A
D
Sure
one
thing
that
which
talks
that
probably
I'll
write
that
on
the
Google,
Docs
Michelle
just
say
if
it
takes
too
much
time
the
gateway,
serviceworker,
graphics
and
section.
We
can
probably
like
add
to
the
features
section,
a
small
like
a
new
item
for
the
gateway
and
that
will
probably
faster
just
to
get
the
this
thing,
ships
and
next
week,
but
you
probably
will
sync
with
Agata.
So
it's
just
suggesting
if
it
that's
faster,
I,
think
it's
a
might
be
a
solution.
Yeah.
A
D
A
B
Last
week,
step
meetings
in
Berlin,
we
also
released
some
updates
Lupita
Peaks.
There
was
a
bug
in
the
cat
DHT
when
I
was
trying
to
read
when
there's
no
bytes
there.
So
if
you
run
into
any
issue
with
poll
reader
where
you
start
seeing
like
read
0
instead
of
X
bytes,
let
me
know
and
I
can
help
with
that.
B
So
I
think
that
hasn't
been
touched
in
about
five
months,
but
there
are
more
conversations
around
that.
So
probably
try
to
resurrect
and
get
a
final
answer
on
that.
Hopefully
this
week
or
next
and
then
I
will
be
working
this
week
on
looking
through
the
peer
delivery
or
no
that's
wrong
pure
discovery
and
content,
routing
modules
to
figure
out
what's
needed
to
get
those
all
wrapped
up
for
this
quarter
and
to
catch
up
on
PRS
that
are
open
from
last
week.
B
A
So
we
like
placeholders
of
all
the
discussions
that
you
remember,
because
otherwise,
if
you
like
are
waiting
to
finish,
one
bring
them
and
then
the
other
one
in
them,
you
might
just
like
flew
out
of
all
them
that
happen,
and
so
just
kind
of
like
leave.
Notes
around
I'd
like
to
remind
you
that
you
have
more
notes
about
certain
topic
so
that
we
don't
we
don't
lose
that
information
and
there
is
it
any
more
questions
for
Jacob
cool.
Ok,
we
have
four
minutes
left
I,
think
we
have
to
one-time
passcode
god
you're
next,
hello.
G
So
I've
been
working
in
the
NS
I
managed
to
finish
the
remaining
optional
parameters
that
I
was
laps.
I
also
had
this
logging
and
rankly
and
cache
as
well,
then,
with
peer-to-peer
Oh
cards
were
also
froze.
I
created
today
appear
for
the
interface
ffs
core,
with
the
IPS
test
as
well,
and
as
almost
all
of
you,
I
Finland,
being
Berlin
Wall,
Street,
I'm
kind
of
lost
right
now,
with
the
interrupt
test
as
well.
Basically,
I'm
trying
to
test
I
Janus
range
interrupt,
but
I
have
two
problems.
G
One
is
when
I
try
to
use
create
a
repo
that
was
used
in,
go
and
then
try
to
use
it
in
jest.
It
is
simply
broken
and
curiously,
there
is
a
test
for
it,
but
there
is
being
skipped
right
now.
I
tried
to
figure
out
why
I
went
to
Europe
and
I
saw
that
Dimitri
was
the
last
person
traded
that
so
I
couldn't
manage
to
talk
to
him
as
well.
G
I'm,
not
sure
why
it
is
skipped
and
I
also
taught
in
Berlin
with
Victor
instead,
but
they
also
couldn't
help
me
and
I
into
the
other
problem.
Is
he
the
she
has
to
go
report
then
I
can
use
the
repo
in
that
interaction.
But
my
solution
is
not
working.
I,
don't
know
if
it's
a
problem
of
the
data
store
is
not
exactly
the
same
or
not,
and
I.
G
A
Good,
so
a
lot
of
the
problems
that
you
described
is
essentially
because
we
don't
have
like
a
proper
up
to
expect
it
has
repo
and
then
that's
beaten.
Many
of
us
like,
if
you
fun,
you
find
notes
all
verticals,
saying
god,
you
just
broke
the
program
or
my
exchange
without
noticing,
and
so
we
were
always
playing
this
like
cat-and-mouse
game.
Do
we
we
got
our
that's
it
like
we
like
chasing
to
IP.
A
First
repo
changes
without
having
tests
with
other
big
notifications,
just
like
make
it
very
painful
for
us
to
keep
moving
after
when
we
actually
didn't
need
to
change
the
report
all
and
so
like
on
0
for
16.
There
was
another
repo
change,
and
actually
this
one
really
affects
you,
because
they
change
the
location
of
like
DNS
records,
and
so
this
is
where
we
actually
now
need
a
really
update
from
Steven
and
Steven
knows
this
very
well
like
he
knows
exactly
like
what
changed
there.
A
What
we
also
need
in
order
for
us
to
move
forward,
is
there
a
repo
migrations
tool,
so
you
finally
meet
you
on
justify
passage
describes
this
like
we
need
a
tool
to
do.
River
migrations:
there
is
one
for
desktop,
so
there
is
one
written,
a
girl
that
works
for
any
the
best
of
knows.
The
thing
is
the
browser
in
the
browser.
You
have
nothing
to
report
to
the
repo
migrations
and
every
time
we
do
change
the
repo.
A
What
happens
is
people
just
like
open
30,000
issues
on
just
the
first
thing
that
people,
because
now
their
demon
is
saying
that
the
repo
is
ocupado
anymore,
which
is
true
the
repo
when
it
changes
our
money
more?
What
we
need
to
have
is
a
tool
to
migrate,
the
repo
that
is
inside
the
browser
which
doesn't
exist
right
now
and.
C
A
Kind
of
like
now
like
between
sort,
then
like
what
to
do.
First,
it's
actually
like.
We
need
to
understand
what
change
your
IP
is
with
your
tests
and
you
can
locally
to
update
the
repo
version
for
the
whole
ecosystem.
We
need
the
repo
migration
school
and
they'd
see
something
that
perhaps
you
can
take
on
as
well.
Again,
they
killed
is
already
written
for
a
girl.
You
can
find
it
on
MSP
for
migrations,
repo
Mike.
You
just
have
to
translate
it
to
JavaScript.
That's
it
make
sense.
A
A
The
I
liked
what
changed
from
0
for
15
to
0
for
16,
like
what
was
the
change
specifically
on
Ibanez
records
and
it's
something
about
namespaces.
They
was
level
to
me
it
just
like
it
does.
He
told
me
I
carry
briefly
in
like
less
than
10
seconds,
so
I
don't
know
the
details,
but
they
do
have
already
the
repo
migrations
done.
So
you
can
even
likely
expect
what
the
tool
does
like
the
FS
before
migrations
to
us
and
see
what
changed
exactly.
A
H
So
one
thing
I've
done
in
these
two
weeks
as
built
a
tool
called
peer
Tyler
which
basically
publishes
local
host
over
HTTP
and
the
other
thing
I
did
is
I
found
a
few
vulnerabilities
I,
don't
know
if
they
are
really
that
much
of
a
deal,
but
they
could
be
and
for
part
of
them
and
that's
all
I
did
in
these
two
weeks.
It
sounds.
A
A
A
Time
like
use
the
the
private
security
Channel
like
it's
called
a
responsible.
This
soldier
program,
like
eh,
where
you
like
you
responsibly,
disclosure
upon
the
beauty
to
prep
a
set
of
people,
the
fix
happens,
and
then
we,
you
responsibly
disclosure
that
to
everyone
with
the
fix
already
them,
so
that,
like
no
one
gets
beaten
right,
because
if
just
professor
was
out
run,
bike
was
already
won't.
Allow
and
running
he
like
millions
of
services.
Now
you
just
like
put
a
huge
bounty
on
everyone.
A
A
I
A
I
So
that's
that
was
new
information,
so
yeah!
That's
why
I'm
adding
support
for
that.
So
I've
got
to
work
in
UNIX.
Fs
I
Casca
I've
just
finished
I,
just
called
the
test
passing
for
em
FS
with
that
as
well,
because
what
it
then
transpires
in
order
to
use
your
leaves,
you
have
to
have
support
for
CID
version.
One
and
I
was
just
kind
of
passing
everything
through
and
it
wasn't
really
working
properly.
I
So
I've
got
that
working,
probably
the
manifest
now
so
now,
I'm
running
the
Jo,
so
ipfx
tests
and
then
interrupt
tests
and
I'm
sure
doesn't
notice
that
so
that
should
be
done.
Someone
basically
it's
kind
of
majority
of
the
work
is
there
after
that,
I'm
gonna
pick
up
the
NPN
on
my
PFS
stuff,
which
would
be
quite
good,
so
I
want
to
okay.
So
we
need
to
run
a
service
to
keep
the
NPM
registry
up-to-date.
I
J
A
I
J
I
A
A
A
I
A
You
mean,
like
raw
leaves,
a
scene,
something
that
doesn't
have
like
any
unique:
the
first
dot
Marshall,
just
like
the
the
raw
IPL
dinos
yeah,
just
the
raw
peel
Deena.
Oh,
it's
excellent,
perfect,
okay!
So,
let's
work
for
for
Allen
when
I
have
a
nice
to
do
too
cool
alright,
so
that's
it
I
hope.
I
didn't
confuse
people
with
other
dogs
any
other
thing
or
if
your
timing
is
over,
we
need
to
finish
here.
Some
people
have
to
go
to
the
next
call.
Thank
you
all
for
coming.
This
was
a
great
update.