►
From YouTube: ⚡️ⒿⓈ Core Dev Team Weekly Sync 🙌🏽 September 24, 2018
Description
A
A
All
right
so
yeah
everybody
make
sure
you
put
your
name
on
the
list,
looks
like
we
have
a
couple
new
people
potentially
here
we'll
go
ahead
and
kick
off
updates
and
then
it
looks
like
we
have
a
demo
from
Allen.
If
we
have
time
do
that
the
end
and
then
we'll
cover
questions
as
well,
so
I'll
go
ahead
and
start
off
last
week
did
a
lot
of
retrospective
and
okay,
our
work
so
scoring
okrs
quarter
and
then
preparing
for
the
new
quarter.
There's
a
draft
of
the
Lib
PDP,
okay,
ours.
I
will
link
to
that.
A
You
get
to
look
at
that
I'm,
hoping
to
get
most
of
the
comments
back
today.
I'm
gonna
move
that
over
into
the
spreadsheet
tomorrow,
and
then
we
can
just
do
comments
from
there.
I
worked
on
some
testing
of
the
data
source.
Standardization
error
codes
for
JSI
PFS,
there's
I
think
two
outstanding
poll
requests
for
that
with
JSI
p,
FS
and
IP
FS
m
FS,
but
that
should
resolve
our
problems
going
forward.
A
There
I've
got
a
couple
of
PRS
out
for
delegator
routing
time
out
defaults
as
part
of
making
sure
that
we're
not
spamming
the
delegate
nodes
with
requests
that
never
time
out
and
then
just
submitted
a
PR
to
fix
stability,
builds
that
got
merged
in
so
I'll.
Do
a
pop
fix
release
of
that
today
this
week,
I'll
be
working
on
delegate,
routing,
PRS,
the
finished
q4,
okay,
ours
and
then
supporting
anything,
that's
needed
for
the
datastore
update
and
then
I'll
continue
on
with
working
on
finishing
the
lippy
TB
switch
incoming
connection
staging
updates.
C
Everything
sounds
pretty
solid
to
me.
I'm
super
excited
to
see
the
delegate
appear,
auditing
and
content
routing
and
like
overall,
like
the
LEP
to
Pierce,
which
do
you
like
I,
know
you
plan
to
finish
it
this
week.
Do
you
have
a
demo
in
mind
that
you
are
also
preparing
at
the
same
time,
or
is
that
not
in
your
schedule
for.
A
The
switch
at
all
it'll
depend
on
when
that
gets
done,
but
I'd
like
to
you
I,
think
the
update
to
the
PDP
itself
would
be
pretty
quick,
and
so
I
ideally
like
to
get
that
or
we
can
do
the
various
states
spinning
from
that.
But
I
think
it
will
be
tight
with
all
the
OPR
stuff
this
week
to
get
that
one
gonna
make
sense.
D
D
So
if
you
want
to
follow
it
over
there
and
I'm,
not
blocked
and
I
were
working
on
this
stuff,
but
I
also
will
do
some
IPL
D
project
maintenance
because
I'm
you
do
the
grousing
stuff,
yeah
I
still
need
to
like
the
releases
clean
up
things,
get
things
into
shape,
so
I
will
mix
this
up
with
a
few
little
bit
of
crafting
stuff
and
then
I
will
hopefully
finally
record
my
foss4g
talk
that
I've
planned
to
record
for
the
past
two
weeks.
Hopefully
you'll
happen
this
week.
C
D
C
On
grassing
I
know
that
there's
like
a
lot
of
discussion
in
the
spec,
and
you
mentioned
that,
like
the
prototype
might
be
out
of
date
already
I
would
like
to
propose
one
thing:
is
there
like
any
prototype
of
graphs?
Inc
he's
always
like
workable
with
a
TFS
so
that
we
can
like
grab
all
the
use
cases
present
like
what
they
do,
I've
been
doing
and
and
have
real
tests
on
the
network.
D
So
so
it's
not
so
I
saw
originally
like
like
like
half
a
year
ago,
I
had
a
version
which
worked
with
I
filled
with
IPs,
but
it
didn't
work
out
so
well.
So
this
current
version
only
works
with
only
p2p
and
but
the
plan
certainly
is
to
basically
get
it
as
soon
as
possible
in
five
FS,
but
it's
still
like
yeah
in
the
future.
Sometime
Ellen.
B
B
B
D
E
Would
actually
really
like
to
see
so
you
said
that
it's
changed
a
bunch
like
in
all
this
time.
I
would
actually
like
to
see
those
changes,
and
why?
Because
a
lot
of
those
changes
probably
speak
to
fundamental,
like
architectural
issues
that
we're
going
to
run
into
later,
and
it
would
be
nice
to
know
kind
of
the
reasoning
of
like.
Oh,
we
tried
this
we're
going
to
go
down
this
route,
but
actually
it
won't
work
or
won't
scale
to
this,
for
whatever
reason,
I
definitely
I
think
help
us
out
a
lot
later.
E
D
D
F
Hello,
so
last
week
I
worked
in
the
retrospective
in
Dhaka,
our
planning,
then
I
started
looking
at
the
awesome.
Ditch
endeavor
I
basically
realized
the
current
PRA
I
studied,
found
several
bugs
and
then
I
started,
creating
the
PRS
for
fixing
those
bugs
then
I
also
continued
making
some
tests
for
the
DHT
interrupt
and
I've
also
found
some
problems.
There.
F
I
will
probably
take
this
week,
but
I
didn't
have
time
yet
then
I
also
made
two
small
parts
for
ending
the
new
Lattisaw
reactors
that
Jacob
implemented
I'm
still
blocked
on
the
IP
and
stuff,
and
there
is
also
PR
for
Jessica
fest
blog
for
review
that
we
will
be
needed
for
the
fest
CPR.
That
Jacob
did
then
for
this
week.
F
C
I'll
I'll
try
to
get
to
these
PR
to
give
you
some
feedback,
make
sure
that
you're
not
blocked
I'm,
not
sure
if
this
is
like
a
technical
design
or
code
review
that
you
never
jump
to
it.
I've
been
like
jumping
here
and
there
and
having
converse
with
you
through
some
peers,
giving
some
suggestions.
I
think
we
are
in
sync
on
dos
right
or.
C
G
F
Yeah
I
still
have
to
make
some
tests
because
I
have
another
pair
which
fixes
the
interoperability
problem
for
local
that
was
introduced
because
I
think
it's
a
problem
that
will
also
occur
with
the
DHT,
because
the
goal
appear
to
be
records
in
the
GS.
The
peer
to
peer
records
are
different
for
both
and
that's
one
of
the
things
that
I
want
to
get
into
this
week.
C
Awesome
yeah
I
was
going
to
say
like
if
the
interrupt
was
already
there
like.
I
would
also
suggest
to
record
a
demo,
because
that's
going
to
be
super
exciting
to
see
a
penis
names
getting
propagated
across
website
like
it
will
also
give
Pedro
and
all
the
people
that
are
building
tabs
on
top
of
ipfs
a
lot
of
new
ideas.
B
B
You
know
like
half
a
second
and
so
I
had
I
just
had
to
fix
that,
but
I,
because
I
haven't
looked
at
the
proxy
for
a
whole
bunch
of
time,
I
had
to
kind
of
bring
it
up
to
up
to
scratch
first,
so
this
is
sucked
up
a
bunch
of
my
time,
not
related
to
kind
of
core
dev
jest
stuff.
Anyway,
that
I've
tracked
down
a
problem.
B
B
It's
almost
finished,
I
just
need
to
write
a
couple
of
tests,
not
a
couple
of
the
few
tests
and
and
then
it
should
be
ready
for
like
a
look
through
at
least
and
then
I
was
off
on
Friday,
so
it
was
kind
of
a
short
week
for
me.
So
I
didn't
get
done,
get
a
lot
done,
I'm,
sorry
anyway.
So
next
up
is
yeah.
I
need
to
review
some
IPS
pr's
like
I,
know
Alex
and
a
keychain,
PR
and
I
want
it.
B
I've
been
wanting
to
look
at
his
repo
config
PR
for
a
long
time
as
well,
and
then
yeah
I
wanted
to
finish
off
that
CID
base
pull
request,
but
also
do
the
do.
Ok
are
planning
and
also
do
some
presentations
for
that
as
well
this
week.
So
this
is
going
to
be
a
big
week
for
me
and
I'm
on
holiday
next
week,
so
yeah
burn.
C
Like
I
guess
done
a
question,
awesome,
always
amazing
work
and
I
guess
I
could
one
of
the
things
that
eyelid
is
also
asking.
Everyone
is
like,
if
you
have
anything
and
you
nice
to
back,
make
sure
to
do
it
as
soon
as
possible
because
it
is
not
going
to
be
around
until
your
question
so
perhaps
like.
If
you
have
like
a
an
issue
or
a
pill,
request
review
that
T.
C
H
So
the
P
all
I
had
open
about
array
types
and
generators
for
documentation
got
merged
yeah.
So
that's
fixed
a
bug
in
that
which
is
great
I,
most
Jacobs
pirata
MF
s,
but
it's
awaiting
some
work
to
be
done
on
the
main
chair
ipfs
because
of
a
module
release
started
refactoring
M
come
on
I
profess
to
transfer
modules
between
workers
using
C
IDs.
H
At
the
moment,
it's
all
MF
s,
space,
which
is
great
and
it
worked,
but
it's
kind
of
a
cheat
because
because
everything's
done
by
pods,
you
can't
just
get
it
from
another
machine
which
kind
of
sort
of
defeats
objective
using
my
KF
som
ree
fracturing
it
to
you,
see
IDs
as
well
as
part
which
will
be
really
cool.
Tell
you
how
stuff
unblocked
on
a
storage
solution
for
NPM
on
my
PFS
in
in
the
deployments.
H
At
the
moment
we
have
a
bunch
of
docker
containers,
running
workers
for
for
the
app,
and
they
only
have
a
finite
amount
of
storage,
which
will
no
doubt
run
out
when
I'm
asleep
and
people
are
using
it
in
CI,
and
then
it
would
be
really
tedious
when
that
all
falls
over.
So,
there's
a
peer
open
for
converting
it
to
use
the
s3
data
store,
which
should
hopefully
yeah
you
know,
negate
some
of
those
problems.
So
next
up
the
manner
finish
doing
this
refactoring
and
then
I
want
to
start
looking
at
this.
H
This
idea
of
an
IEP
MPM
install
command.
So
instead,
at
the
moment,
with
NPI,
my
PFS
either
use
the
public
registry
mirror
kind
of
thing
that
we've
got
or
you
spin
up
a
demon
on
your
a
machine
and
then
start
like
doing
NPM
stuff,
which
is
fine,
but
there's
quite
a
high
friction,
so
just
having
a
command
that
kind
of
basically
in
there
is
how
NPM
does
install,
but
it
will
do
all
this
kind
of
plumbing
stuff.
H
In
the
background
for
you
and
Pullo
is
in
from
my
PFS,
which
will
be
awesome,
it's
a
real
shame
that
we've
got
like
you
know
everything
some
acronym
like
NPM,
because
it
just
makes
try
to
come
up
with
names,
for
these
commands
really
hard.
You
know
something
you're,
not
gonna
trip
over
when
you're
saying
I
mean
no
problem.
C
Wanna
be
first
to
become
the
default
registry
that
we
catch
our
awesome
work.
There
I
heard
that
we
are
getting
a
peanut
soon,
so
we
can
start
publishing
the
route
of
the
registry
again
so
see
like
take
a
look
at
that,
because
it's
coming
I
got
it
already
support.
If
and
emlib
vests
can
still
support
both
be
personal
entity
Fessenden's
you
can
I
play
with
both.
C
I
have
that
feature
ready
so
that
when
that,
when
done
just
because
you
just
like
who
John
is
just
calling
that
feature
another
thing
they
like
kind
of
like
right
now,
we
kind
of
have
to
spin
these
NPM
on
ipfs
daemon
to
use
as
the
registry
endpoint,
but
the
reality.
If
this
is
me
divergent
into
this,
I,
don't
work
but
I
want
I,
don't
want
to
go
there.
I
just
want
to
point.
C
The
idea,
which
is
the
reality,
is
empty,
install
kind
of
like
hooks
for
just
like
regular
paths,
and
so
it
would
be
great
if
we
could
expose
the
NFS
folder
to
that
normal
IP,
fest
gateway,
because
then
he'll
develop
a
fest
he's
just
like
a
module
that
fetches
models
from
NPM
and
puts
them
on
a
manifest.
But
then
any
nodes
can
be
an
install
point
in
via
registry
for
the
NPM
CLI.
C
E
Yeah,
did
you
get
a
chance
to
look
at
the
tink
thing
that
NPM
put
out
just
a
really
flaky
yeah
yeah?
They
just
did
like
a
blog
post
about
it,
but
basically
they're
experimenting
with
like
an
entire
new
way
of
doing
package
installs
and
relating
to
the
registry
and
I
just
want
to
make
sure
that,
like
as
the
sort
of
future
of
how
all
this
stuff
works,
is
in
flux
that
we're
not,
you
know
into
like
bound
forever
to
the
older
model.
H
A
H
I
I
I
I
Another
thing
is
about
serviceworker,
it's
about
the
streams
and
not
being
able
to
use
fetch
or
HML
HTTP
requests
inside
service
calls
and
stuff
that's
fixed,
but
it's
not
shipped
yet
I
didn't
make
the
port
quest
yet,
but
it's
working
also
did
retro
and
planning
for
the
okie
ours.
I
got
back
to
the
missing
tie
fire
for
the
big
data
stuff.
I
It's
already
just
making
the
like
core
instances
connecting
to
other
types
of
instances
or
demons
or
whatever
and
also
stuff
running
inside
the
serviceworker.
That's
pretty
much
done.
I
just
need
to
finish
up
some
stuff
and
push
it
to
the
to
eat
up
and
also
I'm
still
blocked
on
the
IPF
SETI
shanks
ad
and
also
on
the
ipfs.
If
you
guys
can
go
there
and
reveal
that
stuff
would
be
great,
the
chrome
fix
is
also
these
or
requests
are
just
extracted,
so
the
guys
from
the
companion
can
ship
stuff
faster.
I
If
we
can
just
merge,
that's
the
simpler
pollock
quest,
my
they
have
like
the
shells
files
new
app,
they
are
releasing
that
needs.
This
fix.
I
also
have
a
couple
of
other
peers
that
are
blocking
me,
but
just
minor
stuff
for
documentation
on
CTL
also,
don't
know
what
to
do
about
the
bridge
and
pure
ID
stuff.
If
there
it
can
help
me
out
here
if
I
should
just
release
CTL
without
that
and
just
revert
it
or
just
ship
it
like
it
is
now
because
I
would
like
to
really
isn't
need
this.
I
C
I
follow
up
on
your
questions
for
me,
so
the
pre
gen
IDs
were
shipped
they
are
there,
like.
You
can
use
a
flag
to
say
that
you
don't
want
to
generate
ideas
on
the
fly
that
like
basically
makes
cloning
and
ipfs
nodes,
either
Jess
or
go
way
faster
because
we're
not
to
like
to
the
RSA
dance
that
curl'd
was
tested.
I
So,
explaining
and
better
this
the
situation
is,
there's
not
actually
a
flag
you
just
like
if
he
is
the
on
disposable
mode.
It
runs
through
the
code
path
using
Prashant
IDs,
but
I
run
a
basic
benchmark
and
it
does
basically
nothing
or
in
terms
of
performance
or
speed
gains
or
whatever.
That's
where
my
concern
is
interesting.
C
C
I
C
So
so
yeah,
if
they,
there
is
no
other
feature
that
people
are
looking
for
no
need
for
release.
There
is
another
feature
what
you
can
consider
realizing.
Otherwise,
let's
get
it
fixed
first,
the
rest
on
sounds
good,
like
we
are
now
one
time
I'll
just
like
go
very
quickly
to
my
update.
Basically
I've
been
doing
like
okay,
our
dance
all
over
the
place
and
preparing
IP
fest
days
and
thinking
about
the
girl
core
hack
week.
That's
going
to
happen
in
Honda
and,
and
so
my
time
has
been
sucked
by
all
those
like
planning
management
things.
C
C
F
I'm
not
sure
it's
the
right
forum,
but
it
is
to
know
well
today,
two
people
thought
to
me
about
problems
with
WebSockets
tar
bottle
night
being
the
bottleneck
off
of
pierced
are
based,
apps
and,
and
so
is
that
there's
a
rendezvous
protocol
or
RC
that
has
been
open
for
some
time.
I
know:
there's
a
chicken-and-egg
problem
in
the
implementation.
I
would
like
if
there
is
possible
to
have
an
update
on
that
and
what
are
the
plans
and
whether
I
can
help
to
unblock
something
or,
if
there's
any
something
else
that
that's
blocking
this
I.
C
J
So,
with
the
rendezvous
RFC,
there
is
a
few
comments
in
the
proquest
have
to
be
adjusted
in
the
spectrum
itself.
None
on
the
on
the
go
so
and
there
is
incrementation,
and
there
is
also
a
testing
server
you
can
use
and
there
is
a
JavaScript
implementation,
but
it
seems
to
be
a
bit
inefficient
and
it
has
like
it's
kind
of
over
engineered
at
this
point.
I
think
you
can
take
a
look
at
it,
it
should
be
I'm.
Gonna,
repeat
the
only
me
section
is
anyone.