►
From YouTube: ⚡️ⒿⓈ Core Dev Team Weekly Sync 🙌🏽 February 11, 2019
A
A
Cool
okay:
what
do
we
do
now?
Well,
normally
we
do
a
weekly
update
and
everybody
tells
each
other
what
they
did
last
week,
what
they're
going
to
do
this
week
and
what
they
currently
blocked
on.
So,
if
you
haven't
already,
then
please
put
your
weekly
update
in
back
pad.
You
know
where
it
is
just
put
your
name
and
done
and
block
and
next
and
we
shall
slowly
scroll
through
and
talk
through
what
we've
done,
what
we're
doing
next
so
right
at
the
top.
A
What
else
have
I
been
doing
as
I've
been
working
on
Eid
version?
One
base
42
by
default,
I've
added
a
whole
bunch
of
requests
there
and
the
DHT
PR.
That's
been
merged
Thank
You
max
ago,
and
everyone
else
who
worked
on
that.
It's
gone
in
it's
there.
So
what
I
did
was
I
did
a
release.
0.35,
it's
a
pre-release,
so
the
warning
there
are,
there
are
dragons
and
things
I've
noticed
already.
A
The
lock
file
has
been
unable
to
update
lock
within
still
threshold,
which
is
a
bummer,
because
that
actually
quits
your
daemon,
which
is
which
is
not
not
great,
hang
on
a
second
people
are
messaging
me,
as
you
can
have
to
put
on,
do
not
disturb
yeah
and
but
the
other
problem
which
I've
noticed,
which
is
actually
I.
Think
the
thing
that
we
really
need
to
fix
is
the
CPU
pegged
at
100%,
and
that
is
the
reason
why
it
can't
get
to
locking
updating
the
lock
file.
A
So
what
else?
Yeah,
the
IP
FS
HTTP
client,
the
tests.
There's
this
one,
weird
trick
that
you
can
do
in
go
ipfs
if
you
specified
a
test
profile
when
you're
doing
an
IP
FS
in
it,
it
I,
don't
know
exactly
what
it
does.
I
didn't
couldn't
find
any
documentation
on
the
exact
specifics
of
what
the
test
profile
does
yet,
but
it
made
the
test
go
well
fast.
A
That
would
help
them
to
move
over,
so
we
need
to
tell
them
somehow
that
they're
using
the
old
version,
there's
already
a
deprecation
notice
on
the
old
module,
but
that's
clearly
not
enough
so
I'm
going
to
do
something
about
that.
This
week,
like
I,
said
before,
I
want
to
look
at
the
HT
testing
and
I'm,
hoping
to
continue
with
the
CID
version
132
by
default
endeavour
and
I've
got
basically
look
I'm
listing
out
the
pull
requests
that
as
I'm
opening
them
on
the
issue
I've
linked
there.
A
B
Everyone
so,
as
I
once
told
you
that
HTTP
our
was
finally
ready
and
merged.
I
also
created
three
pull
requests
for
controlling
the
number
of
open
connections,
and
it
is
also
configurable
and
config
file
of
Jessup
FS.
Basically,
there
is
just
when
PR
there
is
the
main
care
for
Jessica
fest.
It
still
needs
a
review
and,
besides
all
this
I
also
started
continued.
B
So
for
the
for
this
week,
I
want
to
help
Alan
with
the
HT
a
test
and
also
get
the
danger
of
poor
request
shipped,
which
needs
to
be
that
one
that
I
previously
described
and
the
one
that
I
did
before
there
is
the
NPM
goal
is
to
create
dependency
which
currently
in
PR
that
I
have
only
supports
Linux,
but
I
will
also
make
the
changes
before
merging
it.
To
get
the
CIA
grain
in
all
the
environments
and
also
create
the
HT
interrupt
tests.
B
Then
I
also
want
to
enable
enable
the
guard
gossips
of
implementation,
the
chain
safe
systems,
guys
mostly
my
career
ever
now,
an
initial
implementation
in
a
pro
request,
I
want
to
review
it
this
week
and
also
improved
the
documentation
for
jslabbee
to
observe
and
add
support
in
the
Damon
and
Damon
client
for
web
serve
in
order
to
allow
them
to
afterwards
do
the
interrupt
tests
for
the
implementation,
and
that's
it
for
me.
Any
questions
I.
A
A
C
Last
week
released
a
hotfix
for
the
s3
datastore
to
improve
some
error
handling
around
the
AWS
SDK
because
it
likes
to
call
back
multiple
times.
So
we
added
once
around
that
to
prevent
it
and
there's
also
some
improvements
to
the
sample
lock
file
in
the
examples
for
whoever
would
like
to
walk.
There
are
three
buckets.
C
Updated
some
Docs
for
the
ager
pre-release,
nothing
big.
There
then
mostly
worked
on
pull
in
Plex
of
basic
benchmarks.
There
I've,
basically
working
in
Lib
p2p
test
Suites,
so
I'm
just
going
to
be
working
this
week
on
finalizing
that
and
doing
testing
up
through
GSI
PFS
to
make
through
make
sure
everything
is
good
to
go.
But
we
did
see
some
pretty
nice
performance
improvements
over
the
current
loop
e2b
implementation,
so
we'll
be
working
on
that
also
be
working
with
Alex
on
a
blog
post
from
April
on
I
P
of
s
and
the
s3
datastore.
D
So
if
it
falls
over
it
should
then
just
like
come
back
up
without
wedging
the
box,
which
is
what
it
was
doing
before,
which
was
gone
a
tedious,
so
yeah
I
need
to
track
down
what
is
causing
us
memory,
because
it's
kind
of
the
only
thing
that's
changed
is
changing
the
version
of
ipfs,
sure
yeah,
that's
gonna,
be
fun.
Yeah
did
some
small
releases
of
UNIX
fests,
nothing
massively
interesting
they're,
so
not
blocked.
Only
hang
tomorrow
we're
having
our
first
package
manager'
weekly
sync.
D
So
if
anyone
wants
to
come
to
that,
I
can
stick
with
details
somewhere
for
people
to
see.
Yeah
I'm
gonna
be
leaking
other
package
managers
roadmap
as
well
doing
that
blogpost
with
Jacob
and
then,
if
the
IP
ID
changes
go
in
I'll
stop
mostly
PRS
that
allow
the
UNIX,
fest
and
MF
s
stuff
to
use
it
and
then
it'll
be
nice
to
start
doing
the
async
iterator
ports.
Although
those
libraries
as
well
that's
gonna,
be
me
any
any
questions.
A
E
Yeah,
it's
just
you've
been
with
something
so
am
I
was
working
on
the
still
on
the
J
as
I
feel
the
API
things
it's
close,
so
I
got
I
fixed
other
things
that
anacs
mentioned
now
it's
time
to
fix
all
the
things
that
Ellen
mentioned,
and
so
I
was
so
proud
of.
Mixing
promises
with
callbacks,
but
Ellen
is
not
really
a
fan
of
this,
so
I
will
probably
promise
if
I
all
the
stuff
yeah
but
Yuri
while
I
was
doing.
E
This
Alex
was
pointing
out
that
there's
a
problem
with
using
async
functions
within
the
async
waterfall
stuff,
so
it
won't
probably
Transpower
it
for
browsers,
and
while
doing
this,
I
found
out
that
there
is
no
good
way
of
running
the
tests
against
the
actual
build
we
are
doing
for
the
browsers.
So
the
thing
that
we
distribute-
it's
not
the
thing
we
test
against
there's
up.
You
are
open
which
I
just
put
into
my
notes.
E
This
was
what
I
was
typing
about
and
while
I
was
doing,
this
I
found
out
that
there's
tests
in
interface
I
picked
the
score
that
currently
don't
pass
in
the
browser.
If
you
enable
the
production
bill,
yes,
so
this
took
a
lot
of
time
and
then
I
do
yeah,
so
yeah
and
I
do
some
I
thought
this
stuff
that
it's
not
interesting
from
this
meeting
and
I'm,
not
logged
and
next
will
be
a
Duke
finishing.
The
chase.
I
pulled
this
stuff
and
hopefully
it
merge
this
week.
A
F
I
filed
a
PR
for
the
GS
ipfs
repo,
some
things
no
I
just
kind
of
went
at
it.
So
if
there
are
things
are
like?
Oh,
what's
this
just
let
me
know
and
I'll
dress
it.
The
other
thing
was
that
they're,
like
also
like
sort
of
little
dependencies
in
terms
of
tasks
so
like
FS
datastore,
it
relies
on
which
the
sort
of
relies
on
data
store,
Court
but
I,
just
kind
of
followed.
The
pattern
from
the
news
story,
interface.