►
Description
IPFS Newsletter: https://tinyletter.com/ipfsnewsletter
A
B
So
role
was
a
conferences
all
last
week,
but
I
took
the
opportunity
to
mess
around
with
some
DHT
testing
incident
get
my
secret
there
and
my
the
cloud
run.
Google
cloud
run:
surrealist
environment
I
was
playing
with
and
figured
out
how
to
hook
in
some
of
the
visualization
work
we've
been
doing,
and
so
now
I
can
run
the
tests
and
get
like
thousands
of
separate
traces
with
its
full
sets
of
events,
and
then
click
on
them
and
see
the
visualizations.
B
So
I
think
it
needs
a
little
bit
more
review,
but
I
think
when
it
lifted
a
bit
more
feedback
on
it,
and
then
maybe
you
make
a
little
video
we
can
circulate
so
the
the
visualizer
itself
is
also
under
development.
So
I,
don't
know
how
much
I
don't
know
want
to
know,
don't
know
if
I
want
to
post
screenshots
of
that
publicly.
Yet
so.
C
A
D
Distinctly
the
same
as
it
was
last
week.
Basically,
you
know
that
yes,
which
is
released
I,
just
need
to
update
winter
up
here
with
the
newest
version,
which
I
will
do
maybe
today
or
tomorrow,
and
it
will
be
ready
to
emerge
after
that
also
I
still
blocked
in
the
GFS
integration
for
code
review,
and
that's
it.
E
F
So
I
just
got
the
web
party
signaling
working
over
a
relay
today,
so
I've
got
a
proof
of
concept
of
that,
so
I'm
going
to
work
on
abstracting
that
out
and
updating
the
spec
and
commenting
there
on
a
path
forward
for
that
I'll
push
some
code
up
tomorrow
of
that
all
working
and
then
I
also
gonna,
look
at
leveraging
other
peers.
We
talked
about
the
last
week
for
ice
because
they're
currently
we're
using
the
default
servers
that
simple
peer
uses
which
are
set
Google
and
Twilio
ice
servers.
G
Eat
it
yeah
so
landed
the
make
I
can't
have
summer
puffs
up
like
work
independently
PR,
so
that
that
works
now
asterisk
yeah.
So
this
means
that,
if
you
know
it
now
takes
like
one
DHT
query
to
get
your
pup's
up
your
arcane
s
records
instead
of
us
just
waiting
for
the
timeout,
which
is
both
not
hitting
the
number
of
quorum
peers
and
also
not
returning
very
fast
kind
of
the
worst
of
both.
G
Unfortunately,
the
bootstrapping
that
is
built
in
that
is
already.
There
is
too
low
for
one
DHT
query:
it's
just
our
DHT
is
too
slow
for
the
timeout
we
set,
and
while
we
could
just
fudge
with
the
numbers,
we
want
to
make
pub/sub
discovery
better
anyway,
which
is
where
my
time
will
be
going
next
right
now,
I'm
working
on
putting
some
utility
tools
for
discovery,
so
that
we
can
basically
just
like
make
things
usable
and
make
them
composable
like
we
have
for
content
routing
and
there's
an
issue
link
there.
H
So
I
want
to
get
that
spec
for
the
DNS
I
do
know.
Somebody
Nancy
out
this
week,
probably
a
little
bit
scope
down
without
the
local
mdns
parts
and
also
finish
the
definition
of
the
deploy
process
for
the
name
servers.
So
we
can
actually
test
the
the
code.
It's
an
actually
actual
running
and
name
servers.
It's
about
it.
A
Okay,
next
one
migration
to
multinational
he's,
Davis
tour
is
still
pending
I,
not
seeing
any
package
manager
updates
up
its.
What
it's
the
least
was
something
here.
A
C
D
So
Alan
he's
not
here,
but
I
can
talk.
I
have
been
working
on
it
last
week,
mostly
so
I
created
two
three
new
PRS
for
the
gesture
which,
besides
migration
and
basically
the
status
now,
is
that
I'm
kind
of
blocked
in
a
set
of
peers,
because
I
need
to
have
Alan's
reviewing
the
interface
connection,
which
will
be
probably
the
biggest
blocker
at
the
moment
and
will
allow
us
to
get
March
for
five
other
modules
and
regarding
the
the
statues
we
are.
Seventy
two
percent
of
the
PRS
are
now
or
merged
or
open.
C
E
A
I
Just
in
general,
looking
at
kind
of
are
okay
ours
for
the
quarter
as
as
kind
of
encompassed
in
the
project
operations.
Chunk
of
things
seems
to
be
aiming
for
for
a
really
good
end
of
quarter
cadence,
but
just
wanted
to
get
like
a
general
hey
we're
about
halfway
through
the
quarter.
How
are
folks
feeling,
what's
folks,
take
on
things?
What
what
are
the
things
that
people
feel
are
most
at
risk
or
need
extra
help?
The
people
could
maybe
jump
in
on
Oh
testing.
A
Yeah
my
concern
is
like
we
want
to
be
able
to
start
cutting
police's,
probably
in
the
quarter
and
I.
Don't
know
how
close
we
are
at
about.
You
I
told
the
group
kittib,
a
close
one
think
people
can
jump
on
it
doesn't
require.
B
Last
week,
I
did
some
finally
got
some
ability
to
get
some
good
numbers
out
of
the
DHT
stuff,
and
this
is
the
version
that
I
was
testing.
You
had.
The
tracing
stuff
is
way
way
way
faster
than
the
current
master
stuff,
which
I
imagine
this,
because
there's
a
lot
of
unmerged
things
that
are
gonna
go
into
the
next
release.
Yes,.
A
C
B
A
A
G
A
A
Like
to
do
but
like
not
that
when
people
make
that
mistake,
yes,
some
people
think
it
was
sake,
but
not
that
when
people
yeah
it
would
motivate
it
release
but
like
if
we
do
do
impactful
is
that
I'm
sneaking
it
for
people
lacking
context
here?
Something
will
turn
on
auto
relay
and
really
hot
at
the
same
time,
and
it
will
make
themselves
part
of
relays
and
then
they
will
get
tossed.
The
idea
here
is
to
make
this
not
possible
and
say
it
looked
like
you
can
like
you
want
to
be
a
relay
or
a
sorry.
A
A
There's,
no
reason
these
have
to
be
in
the
same
client,
but
it's
like
it's
really
all
that's
just
because
they
often
names,
but
it's
really
confusing
and
usable
trailer
song
without
realizing
it
and
unfortunately
because
they
told
the
network
that
they
are
this
every
day.
It's
really
hard
to
convince
everybody
shields
continues
to
be
docile
forever.
Basically,
yeah.
A
Yeah
I
mean
a
small
feature
that,
but
if
they
want
it
would
be
like
a
tool
for
doing
that
like
this
is
something
we've
wanted
for
a
while
anyways,
so
kind
of
like
a
maybe
I,
don't
know
if
we
go
under
key
or
ID
there's.
Some
security
concerns
there,
but
yeah
basically
did
I
think
we
wanted
this
eventually.
But
it's
not
like
it's
burning.
The
thing
that
made
my
little
bit
actually
would
be.
A
C
Yeah
I
think
one
more
question:
I
just
wanted
to
get
clarification
on
like
how
we
go
to
a
process
for
approving
or
just
merging
a
PR
like.
When
can
we
kind
of
made
our
own
yeah.
A
C
A
A
A
Point
to
a
person
should
look
at
it.
That's
just
my
comment
there
I.
Maybe
we
should
make
that
more
explicit,
now
yeah
in
the
past,
and
so
it
just
been
like
well,
if
someone
gives
it
a
review
and
you
can
probably
just
merge
it
like
unless
it's
like
some
weird
comment-
get
a
change.
A
lot
of
it
is
just
like
if
it's
a
breaking
change,
pulling
more
people
who
like
have
no
context,
if
it's
not
a
break,
engaging
but
small
little
bit,
bug
fix,
get
someone
will
get
it.
It
doesn't.
C
A
Be
doing
that
I
do
that
because,
like
I'm
trying
to
push
a
bunch
of
stuff
through
all
at
once
and
like
I
hate,
wavering
I
should
not
be
doing
that
I.
Also
to
get
that
out
of
this
and
I
would
go
over
work
for
reviews.
I,
yeah,
yeah
I
mean
I,
don't
think
it's
you
I
think
I've
seen
it
a
few
places,
but
it
depends
on
what
you're
doing
so
like
if
it's
a
like.
If
this
is
like
a
really
tiny,
adopt
change
like
that
or
the
CI
clicks,
it's
fine.
A
It
was
like
a
test
fix
it's
fine.
If
it's
a
new
test
is
fine.
It's
only
like
if,
like
if
you're,
making
an
API
change.
You
definitely
want
a
review
if
you're
making
like
the
book.
That's
it's
good
idea
to
get
a
review.
If
you're
like
adding
a
feature,
you
definitely
water.
If
you
stuff
like
that,
we're
like
it
was
just
like
really
over
time.
You
kind
of
like
what
things
invisible
things.
Don't
don't
do
what
I
do
it
like
I'm
trying
to
get
better
at
this,
but
I'm
so
lazy,
sometimes
yeah
but
yeah.