►
From YouTube: 🖧 IPLD Weekly Sync 🙌🏽 2019-10-14
Description
A weekly 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
B
A
Okay,
but
those
pestering
works
I
can
hear
myself
and
as
every
week
we
go
over
the
stuff
that
we've
done
and
then
discuss
additional
items
that
we
might
have
and
I'd
like
to
have
fun,
no
take
any
volunteers
or
is
it
the
same
volunteers
every
week,
I
guess
and
hope.
So
so,
I
start
with
myself,
so
yeah
I
spend
some
time
on
really
understanding
surveying
the
rust,
a
civilization.
A
You
do
realize
this
utilization
library
because,
like
it's,
so
it's
a
suffering
super
complex
and
I
needed
to
do
this
in
order
to
implement
this
utilization
for
the
support
staff
for
the
so
in
order
to
be
able
to
visualize
tags
in
Seymour.
There's
the
the
issue
linked
in
the
notes
and
to
find
out
more
details
about
this
and
then,
finally,
that
day
PVA
got
merged.
It's
also
great
and
I
learned
a
lot
about
rust,
error,
handling,
yeah,
that's
also
I,
think,
very
importantly,
good
and
then.
A
Finally,
there
is
a
new
github
help
help
her
to
in
order
to
publish
our
notes
on
github.
It's
also
linked,
and
if
you
go
to
this
page
and
click
on
great
PR,
it
will
be
creative.
Pr
with
all
the
information
needed
for
this
meeting-
and
we
will
be
good,
which
is
a
huge
time-saver
for
me-
it
was
fun
creating
because,
like
new
travel
script
on
the
browsers,
basically
after
the
ie6
is
so
easy
to
do,
I'm
so
happy.
A
Alright!
Next
on
the
list
is
Turkey.
C
So
I,
probably
don't
have
a
ton
to
report.
I
spent
a
bunch
of
time
in
the
last
week
at
the
etherium
DEFCON,
which
was
super
fascinating
and
I'm,
not
sure
that
I
have
a
very
well
distilled
reports
from
it
yet
other
than
they
talked
to
quite
a
few
people
about
guilty
things
and
about
other
serialization
R&D,
that's
going
on
in
the
etherium
community.
C
They
are
solving
some
of
their
own,
very
specific
problems
and
very
specific
ways,
but
we
did
try
to
establish
some
common
ground
just
like
here.
In
vocabulary,
development,
I'm,
hoping
good
stuff
comes
about
in
the
future
that
we'll
see
a
post
around
a
few
docs
pr's,
a
bunch
of
code,
gen
stuff
that
is
maybe
first
mentioned
a
couple
weeks
ago,
got
enough
clean
up
that
it
finally
landed
on
master.
So
it
just
got
a
clean
workspace.
Again.
Thank
goodness
excited
about
that.
C
Then
this
just
turns
out
to
be
like
a
really
tricky
problem
because,
like
obviously
economics
are
tricky,
but
also
getting
this
to
be
of
good
performances
like
continuing
to
be
super
super
tricky
as
well.
I
learned
some
new
things
about
how
the
compiler
optimizes
memory
usage
recently,
for
example,
if
I
have
like
this
certain
struct
layout,
where
several
things
are
going
to
be
converted
into
pointers,
sometimes
the
compiler
will
amortize
that
whole
thing
until
one
move
to
the
heap,
but
in
doing
so
it
might
drag
more
structure
in
it
than
you
actually
needed.
C
Keeping
track
of
all
the
moving
bits
here
is
just
really
heavy.
So
much
like
going
into
the
phases
of
like
development,
I've
lost
track
of
how
many
different
attempts
I've
made
at
this,
and
what
draft
number
I'm
on
but
I.
Think
at
this
point,
I've
gotten
one
that
will
be
worth
carrying
on
so
we'll
see
how
that
goes
for
the
rest
of
this
week.
A
B
Yeah
so
I
finished
up
the
code
gen
refactor
in
JavaScript,
so
that's
great
does
almost
everything
that
we
do
in
a
field
these
schemas
more
than
we
actually
had
working
in
the
last
gen
before
I
rewrote
it
and
it
you
know
it
has
everything
that
we
need
for
UNIX
opposed
to
YouTube,
which
is
what
kind
of
driving
it
with
and
it's
way
faster
way
way
way
faster,
he's
a
new
one.
Is
it
significantly
nicer
and
faster
because
I
got
whatever
you
da
stuff
and
just
cleaner?
B
So
that's
gone
really
well
now,
I'm,
just
back
to
you!
Next
52
should
have
like
a
full
reader
writer
by
the
end
of
the
month
in
JavaScript
for
UNIX
52.
At
that
point,
I
will
actually
take
the
schema
and
publish
it,
and
then
we
can
have
a
long
round
and
debate
over
all
of
the
exact
pieces
of
the
schema,
but
I
want
to
make
sure
that
there's
a
working
implementation
before
we
do
another
round
of
feedback,
because
it's
changed
significantly
since
the
last
time
that
we
did
so
implementations
seem
to
drive
a
lot
of
this.
B
So
I
want
to
make
sure
that
the
implementation
is
complete.
First
I
committed
to
doing
an
NFS
abstraction.
On
top
of
that,
that
won't
be
the
entire
ipfs
API,
but
it
will
be
basically
like,
like
a
mutex
on
top
of
all
the
serialization
of
reading,
you
have
to
be
pretty
close
to
the
serialization
and
reader
in
order
to
do
really
efficient
operations.
B
So
I
wanted
to
just
knock
that
out
and
so
it'll
be
just
like
the
other
UNIX
fe2
library
it'll
be
very
reusable
outside
of
my
PFS,
not
valid
IP
FS
like
its
own
kind
of
abstraction,
but
it
will
just
be
one
single
sort
of
mutex
that
you
can
use
on
top
of
it
to
concurrently
write
and
read
files
into
into
it
and
yeah.
That's
about
it.
For
me,
I
did
have
a
question
for
rod,
though
part
of
this
will
should
probably
be
the
hamp
stuff.
B
So
it's
sort
of
like
there's
like
a
to-do
in
there
right
now
about
that.
We
don't
necessarily
have
the
HAMP.
We
need
the
HAMP
to
be
done,
like
you
know,
within
the
next
month
or
two,
but
it
would
be
nice
to
at
least
stage
it
so
that
it's
coming
in
when
we're
starting
to
integrated
and
that
BFS
we
don't
use
that
feature
or
have
to
start
doing
integration
before
that's
ready.
D
Think
it's
pretty
good,
there's
just
one
that'll
change
with
storing
keys
that
I'd
like
to
make
my
own.
My
outstanding
concern
was
really
trying
to
unify
with
the
file
coin
hampt,
but
you
know
it's
just
been
so
hard
to
get
engagement
with
from
them
on
that.
So
I
feel
that
they
might
just
have
to
be
some
divergence
there
and
I.
Don't
know
exactly
what
the
divergence
is.
I
suspect
it's
mainly
just
gonna,
be
how
things
are
laid
out
on
the
block.
D
They've
made
particulars
and
that
might
just
have
to
be
like
we
could
probably
unify
that
by
having
alternating
schemas.
Like
you
know,
Eric's
vision
of
schema
matching
four
blocks.
We
can
match
a
particular
schema
that
fire
coin
is
using,
and
then
we
have.
We
have
a
not
only
one,
that's
the
default
or
something
like
that,
and
that
that's
really
that's
really
it,
though,
aside
from
that
I
think
it's
ready
it's
and
it's
just
these
apps
any
questions
of
what
are
the
best
default
values,
and
we
just
really
don't
know.
D
D
The
the
yeah
that
they
representation
in
the
JSON
representation
was
the
outstanding
thing
because
of
the
discrepancy,
and
that's
just
for
enums
in
general.
Do
we
go
with
the
Union
approach
or
do
we
go
with
the
members
approach?
So
just
some
thought
on.
That
would
be
good
and
maybe
you've,
maybe
you
think,
you've
said
enough
in
there,
but
so
just
in
clarification
we
could
Eric
at
some
point.
D
E
Not
really
I'm
still
working
on
identity
in
the
w3c
work
for
the
verifiable
credentials
using
IP
LD
published
the
IP
NS
I'm,
trying
to
do
some
CRD
T's
in
the
IPL
d2
Indian
and
I
select
a
team
recently
or
I
guess
back
in
Barcelona
about
what
on
influent
at
at
the
IP
NS
level,
but
I'm,
just
keeping
my
ear
to
the
ground,
listening
to
what's
going
on
with
IP
LT
and
and
how
to
implement
it.
For
the
stuff
I'm
working
on
for
identity.