►
Description
Meeting description here: https://github.com/ipfs/team-mgmt/issues/992#issuecomment-647586048
For more information on IPFS
- visit the project website: https://ipfs.io
- or follow IPFS on Twitter: https://twitter.com/IPFS
Sign up to get IPFS news, including releases, ecosystem updates, and community announcements in your inbox, each Tuesday: http://eepurl.com/gL2Pi5
A
A
A
B
C
B
Are
not
necessarily
groundbreaking
by
themselves,
but
are
really
useful
for
us
going
forward.
Click
in
particular
it's
gonna,
be
really
nice
for
both
performance
for
both
lookup
speed
performance,
but
also
for
udp-based
hole
punching
when
we
get
a
chance
to
Plymouth
us
like
that's
awesome
and
base.
36
will
allow
us
to
do
elliptic
curve
keys
by
defaults,
which
will
then
allow
us
to
do
more
sophisticated
provider
records
which
will
then
cut
down
on
the
number
of
provides
you
have
to
do
to
the
DHT
and
allow
you
to
have
sort
of
more
sophisticated
finding
of
data.
A
Nope
alright
next
is
content
routing.
We
can
effectively
close
this
with
six
release.
The
team's
gonna
be
working
on
transitioning,
so
we're
gonna
be
going
through
our
q3
priorities
this
week
and
then
we'll
create
some
new
initiatives
based
on
that
I'm
to
go
over
in
next
week's
meeting
next
up,
subdomain
gateway,
Lydell.
D
Yep
so
I
hope
to
finish
before
this
called,
but
they
did
not
so
expect
the
later
this
week,
the
PR
in
which
we
try
to
find
a
way
to
work
with
in
TLS
I'll
I'm,
replacing
it
on
top
of
0.6
and
I'm
decreasing.
The
scope
I
talked
with
Stephan
and
we
overall
figured
out
that
all
the
magic
should
not
be
present
in
the
initial
version,
so
I'm
simplifying
everything
we
won't
be
automatically
replacing
the
root.
D
We
will
be
simply
trying
it
will
be
checking
if
the
CID
in
the
text,
if
text
representation
of
a
CID
fits
in
the
dns
label.
If
it
does
not,
we
will
try
base
46.
If
that
does
not
work,
we'll
simply
return
error,
and
that
would
be
like
small
but
significant
improvement
and
a
it
will
solve
like
it
will
solve
the
problem
of
IP
NS
with
elliptic
curve
keys,
and
we
won't
introduce
any
like
magic.
That
would
surprise
people.
We
can
always
decide
to
do
that
in
a
separate
PRF
I
feel
for
now.
D
A
A
C
Allow
me
to
unmute
there
we
go
yeah,
so
UNIX
FS
exporting,
is
landed
and
done,
and
the
milestone
report
for
the
grant
funding
is
submitted
and
merged.
So
you're
welcome
to
look
at
that
I
link
to
those
in
the
notes.
We
have
a
blog
post
coming
out
soon
that
we're
hoping
to
get
on
the
ipfs
blog
as
well
as
this
week
in
rust.
C
So
we'll
be
doing
benchmarking
and
things
like
that
on
both
regular
regular
old
hard
drives
and
things
like
SD
cards
for
raspberry,
PI's
and
stuff,
like
that,
so
drinking,
more
holistic
approach
to
you
know
where
things
are
stored
and
how
things
are
retrieved
and
stuff
like
that
and
performance
will
be
the
focus
there.
That's
about
it
from
us.
E
E
A
H
H
D
D
There's
like
a
space
of
unknown
that
we
would
like
to
clarify
before
we
move
full-speed
ahead,
so
I
created
a
document
link
is
in
the
notes.
I
created
a
document
which
documents
all
the
needs,
this
integration
introduces
and
how
well
it
shows
the
changes
that
will
be
introduced
and
then
what
needs
to
happen
on
the
backend
to
come
out.
Those
changes
so.
D
Namely
at
least
someone
from
NGO
activist,
ideally
from
JSF
EFS,
as
well,
just
to
have
more
eyes
on
it.
Go
through
that
document.
Comment
on
proposed
solutions,
nearly
in
every
place
where
we
have
some
need.
I
propose
like
two
scenarios:
one
is
delegating
more
work
to
go
ipfs.
Second,
one
is
doing
more
stuff
in
userland
in
javascript,
so
we
would
appreciate
some
feedback
on
that.
How
much
stuff
could
happen
in
go
back
DFS?
B
H
Another
one
is
so.
There
are
a
lot
of
intermediate
failures
in
Genesis
ipfs
that
makes
life
really
hard
because
I
don't
know
what
our
regressions
versus
what
are
known
failures.
I
would
like
to
want
you
to
try
to
figure
out
maybe
different
strategy
or
keeping
the
tests,
while
keeping
the
trade
green.
There
are
a
few
things
that
Mozilla
has
been
doesn't
have
to
be
sure.
H
H
I
think
that's
great,
but
I
think
it
generally.
It
will
be
good
to
have
some
sort
of
system
where
we
can
know
like,
for
example,
what
Mozilla
was
doing.
If
there
was
a
known
intermediate
test,
it
would
get
marked
and
then
see.
I
would
pick
it
up
and
show
you
in
lines
that
you
can
ignore
this,
because
that
each
mediately
fails
otherwise
it's
gonna,
it
might
be
fixed
now,
but
I
imagine
over
times.
H
A
B
Just
had
like
a
thought,
so
every
so
often
run
into
issues
with
basically
trying
to
use
our
sink
without
amount
of
limitation
and
how
things
aren't
going
to
work,
because
our
sink
relies
on
like
touching
the
files
in
your
file
system.
And
so,
if
you
use
the
file
store,
then
they
get
locked.
And
so
you
can't
update
them.
And
if
you
don't
have
or
trust
your
mount
implementation,
then
you
can't
really
use
that
to
do
the
updates
either.
B
I'm
wondering
if
there
is
any
like,
if
there's
any
interest
in
seeing
if
we
can
sort
of
import
and
export
data
from
the
file
store
easily
by
like
returning
hit
like
locking
control
and
then
returning
control
to
the
user
right.
Because
part
of
the
reason,
if
you
add,
like
a
humongous
amount
of
data
like
if
s
that
it
takes
a
long
time,
is
because
it
has
to
just
copy
it
like
it's
literally
just
hard
drive,
right.
Speed
and
the
file
store
gets
around
this.
Because
you
just
leave
the
data
as
it
is.