►
From YouTube: 🖧 IPLD Every-two-weeks Sync 🙌🏽 2023-01-30
Description
An every two weeks meeting to sync up on all IPLD (https://ipld.io) related topics. It's open for everyone and recorded. https://github.com/ipld/team-mgmt
A
Welcome
everyone
to
this
week's
ipld
sync
meeting:
it's
generally
the
30th
2023
and
as
every
two
weeks
we
go
over
the
stuff
that
people
have
worked
on
and
then
discuss
any
open,
Agenda
items
or
also
if
there
are
questions
or
any
other
things
and
yeah.
So
everyone
attending,
please
put
your
name
on
the
attendees
list
and
yeah,
so
I
will
go
through
the
list
and
I
started
with
myself.
I
have
a
quick
update.
A
Finally,
in
a
rust
IPL
D
we
have
a
deck
PPS
spec,
compliant
implementation,
the
things
freedom
for
implementing
this,
and
it
took
a
long
time
and
also
the
things
for
the
colleague,
fixtures
and
now
yeah,
that's
kind
of
cool,
because
it's
been
I
think
an
issue
that
has
sparked
me
for
a
long
time.
And
finally,
we
have
it
good
yeah,
that's
all
I
have
next
on.
My
list
is
Rod.
B
I
think
I
I've
got
a
grab
bag
of
all
these
things
and
I'm,
just
gonna
read
through
them
one
by
one
and
the
links
are
in
the
notes
that
will
be
in
GitHub
if
you're
looking
at
this
afterwards.
But
okay,
there's
a
new
release.
New
releases
of
gokai
out
yesterday
mainly
just
updating
dependencies
in
a
few
very
minor
things,
but
the
notable
thing
here
is
that
we're
putting
binaries
finally
on
the
downloads
on
the
releases
page.
B
So
if
you
want
to
just
grab
binaries
for
for
the
car
command
line,
you
can
grow
in
there
and
I
I
would
suggest
that
it's
actually
really
worthwhile.
If
you
do
anything
with
cars,
if
you
use
them,
the
go.
Car
is
where
we
put
most
of
our
just
random
utilities
that
we
use
to
inspect
and
debug
car
files.
So
the
car
command
is
really
useful.
It's
got
an
LS,
it's
got
version,
it's
got
a
debug
which
is
really
interesting.
B
It'll
actually
show
you
a
a
text
version
of
all
the
block
contents
and
you
can
even
rewrite
blocks
using
that
pretty
funky.
B
So
have
a
look
at
that
and
if
you've
got
other
stuff,
you
want
to
put
in
the
car
command
line
because
we're
all
just
we
all
just
sort
of
shove
stuff
in
there
go
multi-coding
has
a
new
release.
It's
just.
It
was
supposed
to
be
auto
updating
and
it
wasn't
so
it's
now
a
day
with
the
latest
multiconnectic
table
and
we'll
be
able
to
keep
that
up
to
date.
Automatically
now
I
have
a
a
PR
up
for.
B
B
Let
me
put
another
Link
in
the
thing
there
we
go
so
multi-codic
table.
The
idea
is
we
have
the
adjacent
format
and
CSV.
We
leave
the
CSV,
that's
not
going
away.
People
rely
on
that.
We
get
added
Json
form
where
we
can
put
more
information
and
we
can
be
more
descriptive
so
and
so
people
often
want
new
things
in
the
table,
but
we
can't
really
change
the
CSV
format
without
adding
stuff
to
the
end
of
it.
It's
just
a
bit
unwieldy.
B
So
this
is
this
adds
Json
form.
It's
just
discussion
at
this
stage
with
a
proposal
up
with
some
notes
about
that.
B
B
Car
is
not
a
new
things
so
much
because
it's
just
you're
just
really
writing
out
the
bytes
kavi
2
is
different
but
copy
one
you
can
just
stream
out,
but
this
this
does
it
with
the
Block
store
interface,
so
you
can
use
it
as
a
a
go
Block
store
for
where
you
need
to
plug
a
block
store
in
so
as
long
as
you're
only
doing
rights,
then
you're,
fine,
but
it'll
also
support
has
and
get
size.
So
it'll,
probably
it's
about
everything
but
get
and
delete,
which
is,
is
pretty
interesting.
B
So
you
can
pretend
that
there's
a
block
store
here,
but
it's
really
just
spitting
bytes
out
over
the
network
and
keeping
track
of
which
cids
have
been
written.
Js
Carr
thanks
to
Hugo
and
others
now
has
a
car
buffer
reader,
which
is
a
fully
synchronous
router.
So
you
just
give
it
the
bytes
and-
and
you
can
read
out
of
it-
you
could
do
that
before,
but
you
had
to
use
the
await
to
do
it
and
it
would.
B
It
would
just
because
the
interface
allows
for
synchronizing
an
asynchronous,
but
this
one
is,
if
you
just
want
to
completely
skip
asynchronous
work
at
all
and
just
give
it
a
bites
and
get
back
a
reader.
So
that's
now
released
just
notice
on
go
ipld
tool,
I
I've
archive
that
repo.
We
just
don't
have
enough
bandwidth
to
maintain
it.
B
We
have
some
people
showing
up
to
want
to
do
things
with
it,
which
is
unfortunate,
but
there
are
a
few
people
using
it,
but
it's
just
so
such
an
early
stage
project
and
nobody's
maintaining
it
and
I'm
pretty
much
the
only
one
that
can
show
up
and
say,
hey.
B
B
And
finally,
there
is
a
a
discussion
it's
going
on
in
JS
car,
but
it
might
not
be
the
right
place
to
do
it,
but
the
discussion
on
what
a
validated
Block
store
interface
might
look
like
for
Js,
and
the
reason
it's
in
jsr
is
that
you
came
up
with
a
car
gets
used
as
a
block
store.
Where
you
can.
You
know,
read
and
write
blogs,
and
it's
not
validated
so
it
doesn't.
B
It
doesn't
check
that
the
block
matches
the
CID,
which
is
a
bit
of
a
problem
in
JavaScript,
because
the
main
issue
with
doing
that
automatically
is
that
having
the
hash
functions
available
is
really
hard.
So
we
can,
you
know
we
can
get
easy
ones
in
every
environment,
so
the
Char
2256
is
available
in
the
browser,
any
node,
but
anything
novel
and
you're
in
trouble.
B
Whereas
in
go
you
just
pull
in,
go
multi-hash
and
and
all
the
the
standard
used
ones
are
in
there.
So
there's
a
desire
for
having
the
interface
having
the
Block
store.
Do
the
validation.
But
what
does
it
look
like
to
make
those
hash
functions
available?
And
what
do
you
do
when
you
don't
have
them
available?
B
B
You
know
you
have
an
interface,
that's
got
to
get
and
put,
and-
and
you
want
to
validate
on
the
gets
so
anyway,
that's
there's
an
issue
there
linked.
So
that's
me
anyway.
Nathaniel.
C
Thank
you
Logan
and.
B
I
here
we're
at
on
brexit.
C
Maybe
a
little
bit
louder
than
that
current
apologies
I'll.
Let
us
know
if
you
can't
hear
us
but
yeah
so
I'm,
the
new
member
of
the
three
box
team.
Looking
to
help
out
the
tribute
in
many
ways
we
rust
and
specifically
looking
at
Adecco,
say
implementation
in
ipob.
B
C
I've
got
that
PR
Walker
I
just
saw
your
review
that
came
in
last
night,
which
is
great
so
at
least
three
last
night
and
so
yeah.
The
review
sounds
very
generally
just
added
to
a
new
repo
and
intentionally
use
the
sort
of
a
civilization
directly,
as
opposed
to
reusing
Tax,
Board
I'm.
C
But
I'll
go
look
at
it
and
take
a
look
but
yeah
happy
to
help
out
how
I
came
yeah.
A
Cool
thanks
yeah.
It's
just
one
comment
because,
like
I
feel
really
sorry
that,
basically,
you
spend
so
much
work
on
implementing
it.
This
way
and
now
I
say
well.
This
might
not
be
the
right
way
to
do
it,
but
just
like
I
also
have
basically
so
little
time
to
maintain
the
whole
thing,
so
I
kind
of
like
yeah,
the
documentation,
isn't
good
and
it's
still
hard
to
figure
out
where
to
find
things
but
yeah
so,
but
still
like
I
really
hope
to
be
helpful
and
so
on.
A
So
if
okay,
do
you
have
anything
just
ping
me.
A
C
I'll,
take
a
look
at
the
repo
and
I'll
probably
have
questions,
but
I'll
send
them
your
way,
and
it's
the
best
way
to
do
that
via
slack
or
just
on
the
pr
either.
It's
fine.
If
you
need.
D
Yeah
thanks
again,
I
guess:
I'll
go
next,
not
much
on
this
side,
except
for
that's
one.
Here,
that's
been
around
for
a
little
bit
Rod.
If
you
do
get
a
chance,
if
you
can
take
a
look,
it
was
It
was
kind
of
going
in
a
different
direction
towards
the
end,
but
it
was
mostly
done
for
the
first
approach.
So
when
you
do
get
a
chance
to,
let
me
know
what
you
think
about
the
second
approach,
the
first
approach.
I
know
you
probably
lost
a
bit
of
context.
D
You
know
a
little
bit,
but
I
could
change
the
code
back
to
the
initial
version
that
was
more
tested
and
that
performance
numbers
and
everything
the
second
one
is
definitely
way
more
impressive,
because
it
kind
of
touches
the
the
guts
of
the
note
implementations.
D
Them
that
was
one
of
the
things
we
discussed
about
is
like
you
and
I
discussing
the
drain.
It's
definitely
more
generic,
definitely
more
intrusive.
The
previous
one
was
more
Standalone
would
have
been.
We
just
needed
to
figure
out
the
package
so.
B
Yeah
I
have
to
apologize
because,
to
be
honest,
I
didn't
know
it
was.
The
ball
was
in
my
court.
I
didn't
realize
that
it
was
waiting
for
review.
So
that's
that's
definitely
on
on
my
list
but
yeah.
So
this
this
adds
plain
map
and
plain.
D
The
last
few
couple
of
commits
were
a
little
more
prototyping
and
I
wanted
them
to
be
just
because
I
kind
of
wanted
to
see
what
was,
but
the
general
approach
did
seem
to
work,
updating
the
base
notification.
B
I
was
having
a
conversation
with
someone
yesterday
on
on
in
guy
purely
Prime
about
they
had
the
expectation
that
they
could
use
it
like,
like
the
Jason
and
martial
and
get
back
a
you
know,
map
map,
string
interface
and
it's
not
really
how
it
works,
but
this
kind
of
gets
us
a
little
bit
closer
to
that.
Doesn't
it
yeah.
D
And
yeah
there
is
another
PRN
adder
I'll
have
a
look
back
at
it
again
to
see
how
it
would
fit
with
this
stuff
now,
and
if
it's
in
that
actually
does
try
to
have
a
more
like
intuitive
DX
with
maps
and
lists
and
stuff.
B
Great
yeah,
apologies
again,
but
I'll
definitely
I'll
I
will
try
and
get
this
reviewed
and
commented
on
this
week.
B
No,
it's
it's!
It's
it's
been
in
the
list
and
it
comes
up
every
you
know.
Every
time
we
do
a
review
of
the
list
and
I
was
just
in
my
head.
It
was
like
we're
waiting
for
motion
to
get
some
time
to
get
back
to
this,
but
I
didn't
realize
it
was
actually
on
me
to
review
it.
So
apologies
yeah.
A
Cool
all
right
anything
else.
E
Question
before
before
you
go
before
you
go
into
the
rusty
weed,
a
good
attack
by
crabs.
I
I
was
wondering
like
two
two
weeks
ago,
I
vaguely
remember
at
the
end
of
the
meeting.
You
said
something
about
like
two
different
sort
of
mutually
exclusive
proposals:
one
by
you,
one
by
mauve
and
I,
was
like
looking
into
the
notes
and
I
couldn't
find.
The
link.
I
was
just
wondering
if,
if
you
remember
what
I'm
talking
about
from
that
description,
yeah.
A
Yeah
so
basically
to
this
dv2
is
yeah,
so
I
have
the
link
should
be
in
the
old
hack
pad,
so
there
was
basically
so
there
was
The
Proposal
I
have
and
then
I
discovered,
like
I,
think
right
after
the
meeting
or
the
day
after
I
discovered
that
it's
basically
the
same
thing
as
having
two
cads
kind
of,
and
there
is
another
proposed
that
I
would
like
to
write
down,
which
I
haven't
found,
the
time
which
would
be
different,
I'm,
not
sure
if
moves
has
a
different
one.
D
E
Misunderstood
I
thought
there
was
a
ipip
r
or
issue
by
mauve.
That
was
like
an
alternate
solution
to
the
problem
or
a
solution
for
the
problem
in
a
different
like
problem
space
or
something
but
I
scratched
that
part.
If
it's
not
rigidity
builds
yeah.
B
No,
we
are,
we
are
we're
poking
around
in
different
spaces
of
the
that
that
problem
set
and
we're
we're
really
we're
trying
to
come
up
with
various
Solutions.
So
there
are
conflicting
ideas
here,
but
nothing's
yeah.
E
B
A
But
I
can
certainly
link
to
the
to
the
one
that
I
think
I've
looked.
Oh.
A
D
E
A
So
yeah
it's
just
like
it's
just
like
busy.
Yes,
so
my
conclusion
was
it's
basically
a
two
cads
in
the
end,
so
basically
so
it
was
while
I
was
writing
it.
It
all
made
sense
and
then
I
will
put
it
again
and
was
like
okay,
that's
basically
just
two
cids
and
why
not
them
using
two
cities?
It's
just
yeah
might
be
easier
and
yeah,
but
there's
the
other
idea.
I
have
which
makes
sort
of
short
version.
A
A
This
way
kind
of
to
be
able
to
add
your
custom
things
or
interpretation
to
things
might
be
an
idea,
but
it's
yeah.
It
needs
I
need
to
do
like
more
thinking
and
write
it
down,
but.
E
A
Even
you
put
it
just
the
CID
and
it
understands
the
CID
codec
yeah,
the
CLD
multi-coding,
and
then
knows:
oh,
it's
a
CID,
oh,
if
it's
a
CID,
you
only
want
to
get
the
bytes
back
or
you
introduce
your
new
multi
codec,
which
is,
let's
say
streaming,
video
Codec
and
it
has
then
some
metadata
after
it.
But
it
knows
how
to
interpret
what
comes
next,
because
your
implementation
of
ipfs
understands.
A
Oh,
it's
the
multimedia
streaming
coding
kind
of
like
having
an
abstraction
on
top
of
like
the
Codex
really
and
then
yeah
go
from
there,
but
yeah,
which
is
already
like
a
concept
we
kind
of
have
because
we
have
those
model.
Coatings
and
multicodic
plus
thing
could
be
more
general
purpose,
but
yeah
I
would
like
to
like
if
I
would
do
something
with
context
today,
like
honestly
yeah,
if
I
would
do
something
today,
I
would
probably
just
use
two
cids
and
be
done.
It's
my
current
point
of
view,
but
yeah
we'll
see.
C
E
C
I'm
happy
to
I
think
I've
got
a
pretty
good
direction.
I
just
took
a
peek
at
what
you're
suggesting
effectively
is,
instead
of
using
the
Daga
keyboard.
Drive
macro
use
certain
directly
to
make
all
the
bindings
and
then
there's
a
mapping,
I've
sort
of
vaccine
board,
which
makes
a
lot
of
sense
so
yeah,
I
guess.
My
first
simple
mechanical
question
is
this:
this
new
repo
also
makes
sense.
We
wanted
to
live
in
the
ipld
OR.
A
Like
what
whatever
works
for
you,
so
I
personally,
would
prefer
to
it
if
it
would
be
outside.
To
be
honest,
because
it's
kind
of
like
so
so
I
or
like
at
least
kind
of
like
my
expectations
is,
is
that
if
it
lives
under
the
iPod
organization,
it's
kind
of
like
the
people
from
involved
in
this
all
this
take
care
of
it
yeah
it's
what
I
personally
expect
from
those
repos
and
if
it's
not
really
the
case,
then
I
think
it
would
be
better
if
it
would
be
outside
and
also
like.
A
So
actually,
the
the
whole
plus
iPod
D
repo
lived
for
a
long
time
somewhere
else,
and
it
was
then
just
basically,
we
have
like
abandoned
and
we
took
it
over
and
then
we
moved
into
IPL
lead
for
yeah.
It
was
easier,
but
I
would
be
happy
to
really
have
things
separate
because
I
think
like
yeah.
Ideally,
we
end
up
with
also
like
other
colleagues
that
people
just
do
and
live
somewhere
else.
A
B
B
C
B
Sounds
great
yeah
just
just
to
agree
with
the
talk
with
it.
Anything
in
the
ipld
organ
same
with
multi-formance.
Is
that
there's
an
assumption
that
that
we
will
show
up
and
fix
things
and
and
just
we're
either
too
stretched
or
don't
have
the
expertise?
That's
why
I've
been
archiving
a
lot
of
stuff
recently,
just
because,
like
last
week,
I
archived
a
closure
script
and
a
and
a
c-sharp
implementation
of
multi-formats,
because
people
are
showing
up
with
pull
requests.
B
I
don't
have
expertise
to
do
it
and
we
don't
have
anyone
else
around
to
to
help
review,
don't
know
who
to
trust
with
that
stuff.
So
we
can't
just
publish
this
stuff
and
say
and
and
sort
of
have
it
under
the
official
banner
and
people
assume
that
it's
all
good
without
being
able
to
know
that,
so
we
would
prefer
I
think
that
people
do
stuff
out
and
about-
and
we
sort
of
you
know
it's
better
for
the
this
ecosystem
to
be
more
fractured
rather
than
centralized
just
for
its
own
health.
Yeah.
C
B
It's
good
to
see
the
go,
the
go,
dag
Jose,
it
seems
to
be
I,
don't
know
if
it's
actually
actively
maintained,
but
it's
pretty
minimal.
Yeah.
D
The
good
one
there's
an
alternative
ceramic
and
relaxation.
B
Well,
I
think
it's
isn't
it.
It
does
the
basic
work,
but
it
doesn't
do
a
lot
of
the
utility
work.
I
think
the
the
JavaScript
one
is
where
you
can
really
interface
with
it
and
create
new
data
as
a
user,
it's
much
easier,
whereas
the
go
one
is
much
more
just
consuming
the
data
and
yeah
yeah
well,.
A
And
also,
if
you
put
it
on
your
own
repository
but
basically
have
a
question
or
something
if
you'd
also
like
feel
free
to
ping
me
on
kit
and
just
like
I'm
happy
like
also
like
to
help
on
other
repositories.
Obviously
so
just
yeah
because,
like
especially
like
the
service
stuff,
is
like
I'm
still
like,
even
if
I've
used
it
enough
for
years,
I'm
still
struggling
like
it's.
Sometimes
it's
just
so
hard
to
get
it
working
so
yeah.
If
you
have
problems
there,
it's
just
yeah.
A
It
is
so
difficult
like
if
you,
if
you
feel
lost
it,
is
that
difficult,
so
yeah
so
feel
free
to
ping
me
if
you
yeah,
oh
yeah,
need
any.
B
Help
you
can
put
me
in
as
well
I
I
I've
messed
around
a
bit
with
the
deck
Jose
stuff,
so
I
I
do
have
it
in
my
head
and
I
actually
I'm
doing
some
stuff.
That's
dag
Jose
adjacent
at
the
moment,
just
for
fun.
But
it's
interesting
to
see
the
work
I'm,
just
not
as
useful
with
rust.
So.
A
Yeah,
it's
so
busy
I
I
would
be
more
useful
with
the
rust
rather
than
rather
than
Tech
crochet,
which
I'm
not
really
familiar
with
so
yeah,
so
yeah.
So
perfect.
You
have
two
people
for
for
both
problems,
so
perfect,
great,
all
right,
cool
yeah.
If
there
isn't
anything
else
and
then
I
lost
the
meeting
feel
free
to
stay
around
for
the
after
party.
If
there's
anything,
you
want
to
talk
about
which
shouldn't
be
live
stream
and
then
yeah
see
you
all
again
in
two
weeks
bye.
Everyone.