►
From YouTube: ⚡️ⒿⓈ Core Dev Team Weekly Sync 🙌🏽 August 20, 2018
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
A
B
Hi,
yes,
so
the
first
item
is
actually
something
from
last
week,
but
we
have
any
time
and
I've
got
a
small
script
called
deck
builder,
and
so
it
takes
ins
text
file
as
an
input,
and
then
it
builds
up
a
deck.
So
that's
useful.
If
you
want
to
build
some
like
more
complex
deck
somehow
and
it
has
intentionally
no
documentation
because
I
really
want
to
change
it
quite
a
bit.
But
I
thought
it's
a
good
backup
to
operate
to
github.
So
it's
there
and
you
would
get
the
idea.
B
It's
pretty
small
and
then
I
did
wrote
the
fiddle
graph
sing
into
sub
P
to
P
stuff,
which
ended
in
a
power
in
a
few
pulling
Chris,
making
the
documentation
of
material
because
yeah
it
just
got
started
and
and
I'm
not
long
and
next
week.
Hd
is
this
Thursday
on
I
will
be
in
Tanzania
at
a
conference
that
possibility
conference
and
I've
only
await
here,
September
9,
so
just
for
information,
and
so
I
will
be
somewhere
in
Africa.
B
B
A
Cool
cuz,
yeah,
okay,
alright
sure
it's
there,
any
other
questions.
C
So
last
week
we
did
a
release
of
multi
stream
select
that
should
prevent
people
from
seeing
already
pipes
error,
so
the
root
cause
of
that
hasn't
been
identified.
But
it's
likely
we're
trying
to
read
when
somebody
has
said
we're
no
longer
reading,
but
at
least
this
should
prevent
the
crashing
from
happening.
So
if
you
see
that
happening
scream
at
me,
so
I
can
figure
out
what
else
is
going
on
there.
C
We
got
a
release
of
lid
p2p
to
fix
an
issue
that
Alan
resolves
with
multi
adders,
throwing
errors
instead
of
calling
back
there's.
So
that's
fixed
yay
for
stability,
I've
added
a
small
fix
for
JSF
ipfs
API.
It
was
the
way
the
HTTP
and
HTTPS
modules
were
including
requests.
It
made
it
impossible
to
use
knock
for
mocking
out
API,
which
I
need
for
some
upcoming
work
for
delegate
routing.
So
that's
just
a
quick
small
change
with
how
they're
being
required
to
allow
not
to
work
properly.
So
the
Lib
p2p
Factory
is
ready
to
get
merged.
C
A
No
that's
cool
Thank,
You
Jacob.
That's
all
sounds
good
I'm.
Looking
forward
to
a
lippy
tepee
switch
refactor
cool
showing
move
on
then
I
said.
Let
me
know
of
a
question:
no
good,
ok,
good!
So
who's!
Next,
oh
wait!
That's
me!
Okay!
So
I
had
a
busy
week
if
you're
on
the
All
Hands
I
said
some
words
about
the
a
module
for
discovery
in
transport
using
a
Lib
D
web,
which
is
this
library
for
Firefox.
That
will
enable
web
extensions
to
do
more
pizza,
pier
stuff.
A
You
should
take
a
look
at
the
video
that
is
in
the
linked
from
the
hackpad.
What
else
so
yeah?
There
was
a
whole
bunch
of
places
that
I
found
that
the
where
the
inputs
weren't
being
validated
probably
properly,
and
that
would
have
meant
that
the
demon
would
explode
and
crash
rather
than
so
in
they
meant
that
errors
were
being
fro
and
rather
than
called
sent
via
the
callback.
So
that's
good
and
that's
sort
of
towards
my
okay
are
of
having
a
demon
run
for
10
days
without
crashing.
A
There
was
an
issue
with
object,
patch
RM
link,
which
a
fix
I,
think
that
was
because
a
dependencies
something
in
the
dependencies
changed
in
a
patch
version
and
the
co
vmj
ipfs
was
using
stuff
that
was
considered
private,
like
it
had
an
underscore
in
front
of
it.
So
anyway,
the
tests,
the
master,
started
to
fail
and
then
I
fixed
it
up.
So
that's
good
and
yeah.
A
A
Yes,
IPS
api
erased
a
new
major
version
of
that,
so
that's
got
a
fix
for
object,
get
which
uses
the
new
data
encoding
argument
is.
That
is
a
major
version,
because
the
go
Ike
if
it
breaks
the
possibility
of
altogether
ipfs
nodes,
and
you
can't
you
need
to
use
zero
4.17
for
it
to
work
so
not
clocked
on
anything
at
the
moment.
Next
thing
I
need
to
do
is
to
Rison
you,
yes,
so
you're,
really
new
jet
ipfs
with
the
preload
disabled
fix
in
it.
A
B
B
E
F
Sure
I
flow
with
our
to
get
the
data
for
encode
in
PRN.
We
talked
about
us
a
little
bit
ready
just
briefly.
What
it
is
is
in
Jason
all
strings
of
Unicode
when
the
objects
has
been
turned
into
Jason
by
the
go
demon
it
uses
text
encoding
for
the
beta
field
and
that
just
spits
out
raw
bytes,
some
of
which
can
be
invalid,
unicode
characters.
F
So
then,
when
you're
trying
to
deserialize
that
from
a
string
back
into
an
object
and
then
turn
that
property
into
a
buffer,
you
get
all
kinds
of
rubbish,
and
you
can't
can't
get
the
data
out.
Let
you
put
in
so
what
the
data
encoding
argA
does
is,
let
you
say,
give
it
to
me
in
base64
instead,
which
is
safe
to
turn
it
into
a
string
and
back
out
of
string.
So
that's
cool,
so
we
got
an
TFS
a
while
ago
at
sign.
F
G
F
F
To
do
that
and
just
falls
over,
so
that
was
fun,
so
changing
the
building
code
to
delete
node
modules,
folders
deeply
nested,
a
mortgage
holders
I've
been
doing
some
profiling
on
case
IVFs
found
a
spend
an
awful
lot
of
time,
creating
jason
objects
or
just
objects
that
will
then
return
from
to
Jason.
So
what
was
Volker
to
get
APR
into
IP,
OD,
Doug
pb2,
just
too
lazy.
F
So
when
you
are
requesting
the
jason
objects,
then
it
will
actually
do
the
calculation
if
it
hasn't
been
done
before,
which
is
cool,
put
a
another
PR
in
to
go
ipfs
to
add
the
parents
flag
to
the
right
command.
So
you
don't
have
to
create
a
directory
before
you
write
to
a
file
in
that
directory.
You
can
just
say:
parents
called
like
make
derp,
so
that's
been
approved,
but
it
hasn't
been
measure.
F
A
Clearly,
the
classman
thanks
Alex
can
I
just
say
that
the
change
to
dag
PB
is
what
is
the
thing
that
was
breaking
Jo
psyche
FS,
but
it
shouldn't
have
been
using
the
underscore
properties.
So
it's
not
your
fault
at
all.
Just
my
problem
to
fix
so
yeah,
don't
use
stuff
cos
tried
it.
A
H
We've
traditionally
used
these
these
objects
with
a
slash
in
them
and
we
something
to
sort
of
reserve
space
for
that.
But
ideally
we
should
just
be
CID
instances
and
it
would
thank
a
lot.
The
API
a
lot
simpler
to
use
so
there's
a
PR
to
beg
seaboard,
to
use
this
kind
of
form
and
don't
already
uses
it,
and
this
will
kind
of
make
its
way
into
an
RF.
So
you
see
Malcolm.
H
Yeah
I'll
just
update
the
notes
on
them
right.
Thank
you,
yeah.
It's
currently
on
my
on
my
private.
It
had
on
private.
You
know
my
my
personal.
You
know
private
repo
Oh
in
that
vein,
there's
a
couple
issues
that
people
might
be
interested
on
in
the
EFS
flash
community
repo
one
is
a
just
about
sort
of
blogs
and
Twitter
and
like
different,
we
have
like
a
million
communication
channels,
we're
trying
to
maybe
create
a
new
YouTube
channel.
H
They
can
encompass
everything
and
another
one
is
a
pull
request
to
create
a
project
lifecycle
which
would
basically
call
out
kind
of
what
levels
of
maturity
that
we
think
different
projects
are
at,
so
we
have
like
300
repos,
and
some
of
them
are
sort
of
softly
deprecated.
All
the
way
to
you
know
we're
supporting
this
forever
and
we
don't
really.
G
A
H
Yeah,
this
is
like
a
slightly
more
detailed
version
of
that
Rob,
giving
a
lot
of
feedback
in
into
this
as
well,
but
basically
like
we.
We,
especially
as
we
start
to
spin
up
sort
of
community
engineering
and
even
sort
of
developer
relations.
We
need
a
good
signal
to
say,
like
this
is
ready
for
people's
attention,
rather
than
trying
to
push
attention
towards
projects
that
we're
actually
not
really
ready
for
that.
Yet
so,
there's
a
lot
more
work
on
kind
of
the
earlier
lifecycle
and
also
we
need.
H
We
need
a
different
way
than
just
saying
deprecated,
to
call
out
certain
things
like
we
we
have.
We
have
projects
that
we
are
still
actively
using
and
supporting,
but
we
know
we
want
to
move
past,
and
so
we
don't
want
to
encourage
new
people
to
adopt
them
that
they
don't
have
to,
but
it's
not
actually
deprecated
we're
actively
using
and
porting
it.
So
we
there's
some
there's
some
stuff
in
there
to
try
to
differentiate
like.
H
G
Muted,
thank
you.
Yeah
I
have
a
question
so
I
have
a
long-term
kind
of
you
know,
month
long
goal
to
create
a
demo
app
that
I
can
use
it
the
conferences
and
workshops
and
stuff
that
would
basically
it's
a
go,
live
p2p.
Some
kind
of
demon
wrapper
around.
Go
lip
you
to
Pete.
That
then,
would
connect
to
a
Jas
Lupita,
a
browser-based
Jas
lippy
tepee,
and
they
would
share
a
pub
subtopic
and
I'm.
Just
wondering
is,
and
I
also
am
hoping
to
incorporate
rust.
As
a
third
tier
in
this
system.
A
Yeah
I
would
I
would
talk
to
Travis
a
little
bit
about
the
work
he's
been
doing
with
the
test
bed
stuff,
but
the
I
guess
the
biggest
known
issue
is
that
JSI
VFS
doesn't
have
a
DHT
and
kind
of
tool
to
be
able
to
kind
of
find
content
and
stuff
I
mean
Jacob.
Can
you
elaborate
on
that.
C
C
C
So
one
of
the
things
that
we
actually
have
slated
for
this
quarter
as
well,
which
probably
not
going
to
get
done
just
because
blockers
is
the
actual
getting
lit
p2p
interoperability.
So
we
don't
know
for
sure.
What's
state
that
specifically
is,
but
we
know
we
have
j/s
nodes
talking
to
go
nodes
and
that
worked
that's
functional,
so
we
have
the
ability
to
do
it.
It's
just
a
matter
of
what
surprises
you
might
find
along
the
way.
Okay,.
G
A
H
Have
one
more
thing
that
I
just
remembered
the
organizers,
the
cold
front
conference,
just
ping
me
about
have
it
they
want
something
like
an
introduction
to
distributed
web
kind
of
stuff
talk,
I'm,
not
available
to
go
to
I!
Think
it's
in
Copenhagen
around
that
time.
But
if
somebody
else
has
a
good
sort
of
intro
T
web
talk
that
they're
interested
in
doing
ping
me
and
talk
about
it.