►
From YouTube: ⚡️ⒿⓈ Core Dev Team Weekly Sync 🙌🏽 September 17, 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
Hello
good
morning,
everyone
or
good
evening,
depending
on
your
geographic
location,
on
this
planet,
how
everyone
is
doing:
oh
good,
cool,
good
cool.
So
do
we
have
a
note-taker
birthday.
A
Jacob
Jacob
I'm
going
to
create
like
a
custom
award
just
for
you,
that's
the
best
one
ticket
in
the
world,
cool,
okay.
So
welcome
to
the
Jess
Cordell
team,
weekly
recap
of
September
17
2018.
Today,
we
as
usual,
will
do
the
run
of
updates
before
getting
started.
I
just
want
to
raise
attention
to
something
that
was
discussed
on
SRL,
which
is
okay,
are
plenty.
Please
do
read
through
that
issue.
This
is
not
new
to
this
team,
like
we
have
done
this
in
the
past.
It's
actually
this
two-step
retrospective,
first
and
okay,
our
next
there.
A
The
goal
is
to
have
it
done
by
the
end
of
this
month.
So
like
okay
as
frozen
so
the
sooner
we
start
like
discussing
and
massaging
the
key
results,
and
also
that
we
have
the
retrospectives
out
the
the
better,
because
the
the
rest
of
the
team
would
also
like
to
know
what
are
the
plans
for
this
team
before
they're,
like
so
on
the
24th?
And
so
it's
actually,
if
we
get
our
ideas
out
there
sooner,
it
also
helps
all
the
other
working
groups
rely
on
jet.
A
A
Alright,
let's
go
to
the
realm
of
updates.
I
can
start
first.
My
dad
is
very
simple.
Came
back
from
time
off
already
tried
to
like
go
up
in
a
bunch
of
issues
and
get
feedbacks
to
the
latest
wiki's
and
ask
lots
of
questions.
I
also
kicked
off
the
LKR
planning
for
next
one.
That's
pretty
much.
What
I've
done
last
Friday
any
questions
for
me.
A
B
B
There's
a
bunch
of
PRS
open
some
of
the
datastore
stores,
it's
blocked
by
the
interface
datastore
PR,
on
getting
merged
and
released
on
DVD
and
Friedel
are
the
only
one
who
have
access
to
publish
that
right
now,
but
that
fixes
an
issue
where
we're
not
using
consistent
error
codes
and
so
a
bunch
of
different
places
in
the
in
JSI
PFS.
We're
checking
to
see
if
the
repos
not
created
yet
then
create
it
so
from
not
getting
the
key
initially
do
that,
but
the
problem
is
we're
getting
several.
B
The
libraries
are
returning
different
types
of
error
codes,
so
those
PRS
all
unify
that
also
been
working
on
the
connection
state
machine
for
Lippe
to
be
switch.
That's
going
well
I'm
working
on
the
listener
side
of
that,
so
that
that
all
kind
of
gets
more
sane
and
that
should
help
fix
issues
where
we
have
it
right
now.
B
We're
a
dialer
can
actually
request
an
unencrypted
connection
from
a
listener,
so
that
should
fix
all
of
that
PR
I
just
released
an
update
for
WebRTC
started
a
to
fix
an
issue
with
the
chrome
leaking
RTC
peer
connections,
so
that
error
won't
get
thrown
anymore.
It
obviously
doesn't
fix
the
core
issue,
but
we
shouldn't
see
that
in
the
console
logs
anymore
and
then
this
week,
I'm
gonna
do
an
update,
Lars
and
I
agreed
on
some
time
out
stuff
for
the
delegate
routing.
B
A
B
So
there
are
people
dialing
to
gateways
over
DNS,
also
people
who
are
dialing
to
their
own
nodes
so
like,
for
example,
Mike
was
going
and
setting
up
his
own
ipfs
node
and
then
die
link
to
that,
but
he
couldn't
dial
over
to
it
on.
So
there
was
several
requests
for
that.
So
it
was
a
pretty
quick
fix
which
went
ahead
and
did
it.
A
B
It
doesn't
mean
we're
in
a
better
condition,
so
it's
basically
caching
right
now.
What
happens
is
any
time
that
error
is
getting
thrown
it's
getting
thrown
all
the
way
at
the
stack,
because
we're
not
catching
it.
So
all
we're
doing
now
is
just
catching
that
so
it
doesn't
solve
the
problem
that
should
get
better.
Hopefully
we
should
be
able
to
improve
that
when
we
work
on
the
switch
FSM
stuff,
cuz
I'm
trying
to
do
some
that
are
reasonability
there,
but
yeah
it's
not
fixed.
It's
just
preventing
that
from
getting
from
understood.
A
B
A
A
C
I
created
the
PR
for
GSM
FS,
with
the
full
implementation
of
agonize
over
pursuit
and
some
tears
lighter,
like
CTO
PR
for
the
agonist
tests
and
then
I
found
a
problem
with
interrupt
for
enhance
integrated,
also
appear
for
Jessica
fest
and
interrupt
s
mostly
for
Aquinas
intercepts
so
I'm
blocked
in
a
lot
of
be
ours
right
now.
All
of
them
need
to
review
and
release
some
of
them
regarding
against
Europe
and
others.
C
I've
been
a
super
pup,
so
I
detailed
to
what
each
one
needs
to
be
done
and
for
the
next
week
all
I
want
to
fix
all
the
feedback
available
for
the
Block
peers
and
I
will
continue
working
on
the
edged,
interrupt
and
they're
interrupted
for
IP
network,
except
for
the
DHD.
Interrupt
I
have
a
call
with
call
to
discuss
some
things
about
the
HT
and
I
already
found
the
work
and
I
have
a
work
in
progress.
We
are
with
some
tests
and
the
that's
it
for
me.
A
Awesome,
thank
you.
So
there's
a
lot
of
things
here,
especially
all
of
the
interrupts
there's
like
that
massive
PR
on
Jessup.
If
s
getting
that
very
based
will
enable
you
to
run
the
internal
tests
on
IP
passenger
up
for
the
HT,
which
then
I
could
help
you
debug
a
lot
of
things.
There
John
had
some
pretty
good
results
when
he
tested
it
himself,
so
we
need
to
just
basically
spend
a
couple
of
hours,
realizing
that
PR
and
and
retesting
we've
at
IP
fest
level.
A
They
drop
at
least
he's
looking,
whether
or
not
that
might
help
your
work
as
well.
Abbas
cool.
The
other
thing
is
really
like
this
happiness.
Over
perhaps
up
it's
a
super,
exciting
I
saw
these
pull
requests
and
I
I
need
to
understand
better,
like
what
was
the
decision
to
like
push
down
the
broadcast,
all
that
DNS
records
by
creating
a
data
store
over
pub/sub,
because
what
this
means,
like
anything,
that
you
sort
to
the
data
store.
You
are
now
broadcasting
over
pub/sub,
which
is
not
necessarily
the
behavior
that
we
want.
A
A
C
Just
one
finger
the
data
store,
pub/sub
thing:
it's
not
a
data
store
and
it
receives
the
local
repo
data
store
where
it
does
is
to
have
the
same
interface
like
the
Putin
gets
functions
in
order
to
allow
from
the
outside
to
use
the
data
store,
which
will
allow
us
to
use
the
DHT
in
the
pub.
So
at
the
same
time,
we
don't
need
a
single
Putin.
Yet.
A
D
Hello,
it's
me
cool,
so
what
did
they
do?
Last
week
last
week
at
the
beginning
of
the
week,
I
released
CSI
PFS
0.32.
Finally,
it
took
a
while,
but
it
it
finally
went
out
the
door,
including
blog
posts
and
everything.
So
that's
good,
then
so
I
basically
spent
a
lot.
The
rest
of
the
week
kind
of
looking
looking
at
issues
merging
some
things
that
I
was
then
able
to,
but
I
was
also
trying
to
work
on
the
CID
base
stuff,
which
I've
got
a
lot
a
lot
further
with
about
I
guess.
D
Halfway
through
the
week,
I
I
saw
that
there
was
a
so
Kyle
had
requested,
like
a
tool
for,
like
just
general
purpose,
sort
of
CLI
based
CID
wrangling
so
like
to
convert
you'd,
see
IDs
to
basically
to
list
out
properties.
Do
some
formatting.
If
you
need
to
pull
out
bits
of
see,
IDs
and
things
like
that,
and
so
there's
a
pull
request
for
go
ipfs
adding
this,
and
it
was
relatively
easy
to
set
up
something
in
Jay
s.
D
So
I
added
that
as
I
like
it's
a
it's
like,
it
can
be
a
standalone,
see
if
I
told
you
just
install,
but
it
also
because
it's
just
all
yarks
based
it
integrates
pretty
easily
with
Jace
IVFs.
So
that's
that's
kind
of
there's
a
pull
request
for
adding
that
suggests
IVFs
as
well.
It's
just
a
simple
CLI
tool:
kind
of
to
smooth
over
the
the
transition
of
to
base
32
V
one
stuff,
I.
D
Guess
if
I
were
to
summarize
that
so
yeah
I
made
a
bit
of
progressive
CID
base,
adding
an
option
to
to
to
output
CID
base
32
or
whatever
base
you
like
I've
got
a
call,
request,
open,
I,
guess
it
would
be
cool
to
get
some
feedback
on
it.
I've
taken
an
approach
which
is
probably
a
bit
controversial
because
it
basically
introduces
breaking
changes
such
as
like
where
a
hash
was
returned,
as
a
string
would
now
be
returned
as
a
CID,
so
I'm
not
entirely
sure.
D
That's
the
best
way
to
go
and
looking
at
the
the
the
go
discussion
for
the
same
thing,
I
think
they
decided
it
wasn't
the
best
way
to
go
as
well.
So
that's
relatively
easy
to
back
out
and
go
go
the
way
they
they
wanted
to
go.
But
the
detail
is
in
the
pull
request.
So
please
take
a
look
if
you're
at
all
interested
so
yeah.
That
was
me
this
week.
Basically
I
blocked
on
Jenkins.
Most
of
the
time
this
guy
he's
yeah
like
I.
D
Do
it
Jenkins
takes
up
a
lot
of
my
time
in
a
moment,
running,
builds
and
when
it
fails,
because
of
something
that
isn't
the
J
s
code
tests
failing
legitimately,
because
there's
something
wrong.
That's
just
a
massive
time
sink
for
me
and
so
like
NPM,
just
in
NPM
installing
at
the
start
of
things
and
row,
workers
running
out
of
space,
which
happens
almost
every
day,
it's
kind
of
annoying
and
then
compounded
by
like
there's
I,
think
there's
a
job
that
runs
every
evening
to
clean
up
some.
D
Some
stuff
on
the
the
workers,
but
that
causes
like
a
branch
indexing
thing
which
basically
runs
all
the
outstanding
pull
requests.
So
anything
you
had
passing
before
that
was
waiting
to
be
merged
could
possibly
become
like
not
passing
because
because
it
might
fail
the
second
time
around.
So
that's
that's
kind
of
annoying
for
me
and
but
like
in
on
the
whole
I
think
I
feel
like
the
GSI
PFS
tests
are,
are
feeling
a
lot
more
stable.
D
Now,
there's
a
couple
of
errors
which
I'm
seeing
kind
of
frequently
like
database,
not
opening
this
one,
but
like
other
than
that
it
seems
to
be
NPM
issues
or
workers
out
of
space
issues.
So
so
that's
good,
and
this
week
I
wanted
to
work
more
on
the
CID
based
stuff.
Get
that
finished.
I
need
to
create
interrupt
tests
who
check
that
the
block
store
can
get
CI
get
data
out
with
see
IDs
that
you
give
to
it
from
either
version.
A
Any
questions
cool.
Thank
you
for
the
bed,
my
question
so
everything
the
question
I
really
have
is
given
these
problems
from
chickens.
They're
like
like
stealing
our
time
from
the
developments.
Do
we
feel
that
the
developer
experience
team
is
aware
of
these
problems
I'm
on
top
of
it
is
that
I
can
estimate
that
day
time
for
them
to
be
fixed,
or
should
we
be
trying
to
help
to
make
to
get
these
problems
solved?.
D
I,
don't
get
the
feeling
that
it
it's
being
looked
at
with
an
incredible
amount
of
urgency,
just
because
I
opened
a
lot
of
bunch
of
issues
I'm,
usually
on
IRC,
saying
something
if
I
can,
if
I
notice,
something
up
is
up,
I
will
open
an
issue
but
I'm,
not
really
sure
of
a
way
of
because
looks
like
it's
basically
Victor
right
and
I
am
aware
that
he's
super
tight
with
time.
So
you
know
I,
don't
want
to
just
be
complaining
at
him
the
whole
time
so
yeah
I,
don't
know
what
the
I
don't
know.
D
If
this
is
the
situation,
we're
in
now
is
a
lot
better
than
it
was
before.
I
wasn't
really
involved
with
the
project
when
it
we
weren't
using
Jenkins
that
that
he'll
release
so
I,
don't
know
if
this
is
like
way
better
or
if
this
is
worse
or
or
what,
but
it's
all
I
know
is
that
it's.
It
is
sucking
up
a
whole
bunch
of
my
time
and
that
I'd
rather
be
doing
development
work
or
reviewing
or
merging.
A
Thank
you,
yeah,
like
in
I,
believe,
like
the
situation
is
going
for
multiple
months
now
in.
So
it's
really
like
it's
all
about
like
complaining,
it's
just
about
like
creating
away
for
the
team
to
know
if
issues
are
being
ll
prioritized
and
if
there
is
any
hopes
of
getting
it
done
soon,
because
if
the
situation
is
completely
out
of
hand
because,
like
a
couple,
people
cannot
handle
all
the
tasks.
Then
that
is
fine.
A
You
just
like
something
that
we
have
to
acknowledge
and
like
it
means
like
we
have
to
like
increase
the
team
or
I
find
some
more
capacity
in
some
other
way
in,
rather
than
trying
to
hope
that
my
capacity
appears
from
somewhere
and
so
that
what
our
issues
in
one
week.
So
that's
what
I
was
trying
to
like
get
a
Samsung
cool.
Thank
you
any
other
questions
for
Alan
Oh.
A
F
F
F
Hopefully
you
can
get
to
to
them
as
quick
as
possible
and
they
for
you
to
beat
I,
don't
know
if
you
already,
you
can
now
add
like
big
files,
big
Nalin
in
two
ways,
either
the
normal
way,
which
is
like
non
chunk.
Just
one
request
means
the
whole
data
you
can
go
as
as
far
as
your
browser
can
handle
it,
and
one
of
the
main
issues
I
have
right
now
is
defining,
which
is
the
max
value
for
the
demon
to
handle,
like
the
max
lights,
for
one
requests
for
the
body
of
that
request.
F
That's
on
the
on
the
podcast
you,
so
you
keep
feedback
on
that
and
also
you
can
add,
like
Terrell,
the
Korean
live
data
in
the
shanked
way,
and
you
can
define
the
shanks
and
just
add
away,
and
it
will
hopefully
work
out
well
and
at
the
very
point,
I'm
blocked
in
a
couple
things.
I
have
a
small
small
stuff
related
to
dogs
for
the
at
the
FST.
Cereal
also
thing
is
regarding
the
stream
HTTP
library.
F
Gp
had
to
like
fixed
a
cup
of
stuff
in
there
to
make
the
non
shank
a
adds,
work
that
what
was
what
was
making
him
chrome
crash
when
you
like
uploaded
a
hundred
megabytes,
but
regarding
service
workers,
there's
still
stuff.
That
needs
to
be
fixed
on
this
thing
which
to
be
and
to
figure
out
how
to
do
it,
because
I
was
thinking
about
just
making
a
new
one
based
on
direct
package.
F
But
then
I
found
out
that
the
maintainer
is
working
with
us
to
collapse
and
stuff,
so
I'm
gonna,
probably
like
try
to
contribute
back,
but
that
will
take
more
time.
So
I
need
to
figure
out
what
to
do
with
that
and
next
week
or
this
week,
I'll
try
to
release
the
CDL
apart
with
JSF
Rosario,
with
embedded
examples
and
also
the
OTR
stuff.
A
Sweet
okay,
so
I
believe
I,
pretty
much
I
got
the
whole
gist
of
it.
Although
I,
like
some
cuts
like
something,
is
weird
with
audio
at
my
headphones
today,
because
I
can
always
see
you
guys
will
be,
the
video
is
portrayed
it's
just
since
it
cuts
so
I'll
follow
up
on
this
pull
request
and
Ike
read
the
context
or
some
oricon
I'm
getting
I,
prefer
CPI
and
just
be
best
like
to
support,
adding
very
big
files.
A
G
The
Interop
test
and
kind
of
what
what
that
was
going
to
be
but
I
found
or
I,
was
looking
through
the
the
ipfs
interrupt
repository
and
a
lot
of
that
is
JavaScript,
which
going
to
make
sense.
But
I
was
curious
if
the
original
windows
ok
or
was
kind
of
set
up.
If
we
want
to
continue
down
the
route
of
using
javascript.
G
A
A
F
G
So
I
was
just
talking
about
plugins
emerged.
Well,
it's
just
working
they're
missing
some
documentation
which
I'm
going
to
work
on
work
on
and
between
couple
things,
but
currently
I
wanted
to
get
some
more
information
on
the
actual
IP
BT
interrupts
test
harness
and
what
we
wanted
that
catch
me
look
like
what
was
the
expectations
on
that
that
I
have
worked
up.
Some
worked
on
a
little
bit
and
go
but
I
noticed
everything
in
the
interrupt
currently
is
all
written
in
JavaScript
and
us
trace.
B
B
Don't
know
menu,
yeah
I
think
we're
probably
gonna
need
to
table
that
offline
I
know.
There's
been
a
lot
of
discussions.
We
had
some
discussions
at
the
Berlin
meetup
around
what
we
want
to
do
for
Interop
and
obviously
with
Interop
tests
as
they
are
now
like.
The
go
team
is
just
not
going
to
use
that
because
it's
all
in
JavaScript,
so
the
more
we
can
do
to
change
that
over,
so
that
IP
TB
is
running
that
whatever
that
means
I
think
that
is
the
better
way
to
go.
Okay,.