►
From YouTube: GUI and In Web Browsers Weekly, 2019-04-10
Description
Meeting Notes: https://github.com/ipfs/team-mgmt/pull/937
IPFS Mirror: https://ipfs.io/ipfs/bafybeigbfg5zqfhowpxqmoanpkzrx7vdy5v6dsbe4ha36cdryovjacxeya
A
Welcome
to
the
IDF
s
in
web
browsers
and
GUI
team
weekly
sync
cool
more
places
than
ever
on
the
call
it's
very
good
news,
so
we
will
start
off
with
a
round
of
show-and-tell
about
things
that
we
worked
on
in
the
last
week
and
then
we
also
go
back
to
the
agenda,
which
includes
switching
to
CID
v12,
face
30
team
language,
repinning
sharing
and
saving,
and
an
NPM
won't
give
us
at
discussion.
But
first
top
of
the
list
he's
very
organized
Lydell.
B
B
Just
let
me
realign
all
the
windows
yep,
so
I've
added
optional
support
for
translations
of
blog
posts
to
aquafers
blog,
so
there's
a
PR
with
more
details.
But
if
you
just
basically
scroll
to
the
very
bottom,
you
will
see.
There's
there's
probably
I!
Think
I
can
just
show
you
because
it's
already
merged.
B
So
if
you
right
now,
I
think
there's
only
one
post
translated
but
Chinese
simplified.
But
if
you
go
to
like
ipfs
weekly
level,
you
can
see
that
there's
a
section
that
says
this
post
is
also
available
in
different
languages
and
there's
just
Chinese
version,
and
if
you
go
there
it
will
load.
You
can
see
location
bar
there's
a
language
code,
graphics.
There
is
like
link
back
to
English
and
other
versions.
B
There's
a
section
in
Chinese.
It's
sort
of
like
a
disclaimer
that
sometimes
technical
terms
and
concepts
are
hard
to
translate
from
English
to
other
languages.
And
if
you
want
to
double
check,
if
you
are
unsure
about
something
just
original
English
version,
that's
sort
of
a
fail-safe
we've
built
into
falcone
websites
and
I
feel
it's
quite
useful.
And
if
you
go
to
page
source
and
go
to
the
header,
you
will
see
there
are
all
the
meta
headers
pointing
at
the
canonical
version.
The
default
language
and
all
the
available
translations
are
also
listed
there.
B
B
But
I
think
the
most
interesting
part
is
update
on
brave,
so
I
wrote
a
sort
of
more
details,
as
usually
in
my
weekly
dump
of
Chrome
socket
related
stuff,
but
I
feel
I
am
able
to
make
a
quick
demo.
So
basically,
it's
still
on
this
very
field
with
chrome,
sockets
branch,
but
it's
not
ready
to
be
merged
upstream.
B
So
I
have
like
the
pre-built
version
of
companion
here
from
that
branch
and
I
love
that
it
is
brave
and
you
can
see
it,
it's
a
fresh
install.
So
it's
just
how
it
would
look
like
when
a
new
person
is
installing
a
companion
in
brave.
So
you
can
see
that
what's
interesting
is
that
ipfs
companion
is
already
set.
You
are
connected
to
eight
peers
and
if
you
check
you
can
see
that
you
are
actually
running
JS
ipfs
and
it's
Jess
idea.
France
is
exposing
both
gateway.
B
Api
and
all
the
usual
manual
entries
are
available,
and
you
can
basically
just
open
web
UI,
and
you
can
see
here
that
the
web
UI
loaded
from
the
gateway
exposed
by
the
browser
accession
itself
and
it
just
works,
so
it
connected
to
the
gate
report
it
connected.
It
has
access
to
the
Gateway
port,
it's
using
the
API
from
the
browser
extension
itself.
You
know
and
all.
B
Pages
work
and
I
can't
even
like
upload
entire
I
guess
I
can
maybe
maybe
this
one
entire
directory
I
hope
there
is
nothing.
It
should
be
family
friendly.
So
right
and
what's
interesting,
is
that
all
the
content
that
is
locally
available
you
can
explore
using
between
actually
being
Explorer
and
when
you
open
on
the
ipfs
gateway
it
will
redirect
to
the
local
gateway,
and
this
listing
is
from
the
browser
extension
itself
and
you
can
just
load
the
image
directly.
B
Is
there
right
and
there's
also
like
refreshed
settings
page
where,
instead
of
regular
embedded
note,
you
have
this
embedded
plus
chrome,
sockets
and
three
instead
of
the
regular
one?
This
probably
needs
a
lot
of
refinement,
but
right
now
you
can
tweak
ports
here
and
those
ports
will
be
automatically
populated
to
browser
extension.
So
you
just
change
them
here
and
it
will
change
everywhere.
B
I
also
picked
custom
ports,
so
you
can
run,
go
ipfs
and
Jessica
TFS
on
the
same
machine
without
the
need
for
changing
default
ports
just
to
remove
alliances
for
like
regular
people
playing
with
it.
So
the
problem
is
I.
I
did
some
like
troubleshooting
around
this
and
content.
Discovery
was
very,
very
slow.
Yesterday,
it's
better
today
and
there
are
small
issues
with
loading,
specific
content
over
DNS
link
and
Charlotte
directories,
but
so
this
stuff
is
ongoing
efforts
to
refine
it.
B
My
and
my
plan
for
the
next
week
is
to
investigate
how
we
can
improve
content
discovery
and
clean
up
all
the
patches
cause.
I
was
focusing
on
the
UX
of
companion,
making
sure
all
the
like
options.
If
you
are
black,
you
can
see
that
this
is
a
EFS
resource
and
all
the
regular
options
are
available.
You
can
even
like
pin
or
copy
stuff,
so
I
profess
companion
is
fully
recognizing
embedded
node
I
settled
it.
B
No
there's
a
feature,
parity
sort
of
with
external
node,
and
the
next
step
is
to
clean
up
a
few
remaining
absolute
patches
and
I
have
time
to
sync
with
rave
on
Friday
on
next
steps.
Regarding
the
user
experience
and
I
also
started
talking
with
infra.
How
can
we
improve
content
discovery
because
we
could
either?
B
We
could
either
set
up
our
own,
like
WebSocket,
based
more
boost
up
nodes,
more
crit
nodes,
specific
to
the
Bridgeview
for
the
bridge
users,
or
we
could
just
invest
some
time
into
people
to
be
transport,
and
that
way
it
would
just
work
because
it
would
interrupt
with
goal
notes.
So
that's
my
plan
and
on
Friday
I'm,
also
meeting
with
extension,
developer,
advocate
and
chrome
to
discuss,
like
upcoming
changes
in
like
WebEx
nation
echo
system
I've
been
talking
for
too
long.
Sorry
about
that
kubernetes,
yeah
I
can
show
I
think
you
want.
C
Questions
for
ladle
I
have
a
question:
do
we
know
if
we
are
planning
to
ship
with
chrome
sockets
enabled
by
default
in
brave
and
the
second
question
I
had
to
that
ops
bundle
them
up
here?
Is
you
talked
about
the
content,
discovery
performance,
stuff
kind
of
almost
weather
report
like
it
was
better
today
than
yesterday?
Do
we
do
he
do
we
know
where
that's
coming
from,
or
is
it
really
just
straight
up?
We
have
unpredictable
number
performance,
some
aspects,
ignasi's
yeah.
B
B
One
is
specific
for
Firefox,
because
there
are
some
additional
api's
for
Firefox
and
there's
a
separate
one,
the
generic
one
for
all
the
chromium
based
browsers
and
basically,
what
I
did
last
week
was
to
ensure
that
this
generic
build
can
work
in
both
chrome
chromium
based
browsers
and
in
brave,
and
the
idea
for
like
launching
it
it's
sort
of
like
soft
path.
First,
we
want
to
just
publish
this
like
runtime
feature
detection
with
like
regular
into
regular,
better
channel
without
any
adding
any
additional
UX
to
the
brief
itself.
B
So
the
problem
with
the
discover
content
discovery
is
that
we
are
not
able
to
fetch
any
content
unless
one
of
eight
nodes
that
we
are
connected
to
has
already
like
that
data
in
which
we
have
kept
riddled
nodes
which
preload
that
content
we
want
to
through
those
nodes,
but
that
did
not
work
as
expected
yesterday.
So
I'm
talking
with.
In
fact
that
was
like
a
problem
on
my
end:
can
we
improve
it?
There
are
other
options
who
won.
The
dimension
is
to
use
p2p
transport.
That
interrupts
will
go
and
other
nodes.
B
A
A
A
A
Basically,
so
all
our
documentation
now
talks
about
using
the
command
line
tool,
so
it's
nice
that
we
can
now
recommend
it.
So
basically,
it's
still
desktop
get
my
my
tools.
We
can
simplify
the
install
instructions
on
the
website
which
are
currently
terrible
because
right
now
go
to
IP,
desktop
I
know
it's
like.
Oh,
you
should
install
up
your
fence
and
redirect
and
then
you're,
like
I
shouldn't,
know
your
occasions.
Do
I
give
this
it's
like
Oh
download
at
our
CD
into
it
run
some
scripts,
the
you
know
it
does,
and
maybe
it
works.
A
Maybe
it
doesn't.
We
ran
through
this
with
area
tilde
and
on
a
stock
install
of
Mac
OSX.
Today
this
script
does
not
work,
so
these
scripts
are
extra
terrible
because
there
are
a
level
of
complexity
if
you're
a
beginner
you're
not
going
to
get
through
them
and
if
you're
advanced
you
already
know
how
to
do
it.
So
the
big
win
means
we
can
skip
that
I
pivot
bad.
So
this
is
me
just
stepping
out
and
it
did
the
thing
there's
a
layer
of
indirection
in
there
so
that
basically
we
link
to
a
shell
scripts.
A
Out
to
your
memo
about
when
you
sort
of
note
X,
we
couldn't
figure
out
the
local.
That's
why
it
prompted
us
for
this
weird.
You
must
allow
access
the
system.
Events
like,
let's
that's
nonsensical
assistant
man,
and
it
turns
out
that
that
is
the
permission
required
to
have
you'll,
have
to
be
able
to
launch
it
startup,
which
is
interesting
because
they
never
actually
seem
to
prompt
me
for
that.
At
least.
We
now
have
a
root
cause,
so
we
can
start
there
unpack
if
there's
a
solution
to
it.
A
A
It
had
a
bunch
of
other
nice
to
have
seen,
but
there
is
a
new
release
of
go
IP
offense
and
that's
less
relevant
that
there's
a
new
release
of
go
up
if
this
hold
on
20.
That
is
when
he's
candidate
already
with
it.
She
fixes
a
bunch
of
bugs
and
last
time
it
took
us
about
three
weeks,
two
to
three
weeks
to
get
okay,
it's
best
not
released
after
and
I.
Think
that's
release
and
people
started
raising
issues
on
desktop
saying
this
is
terrible.
I
want
the
latest
GOG
your
best.
A
What
is
activist
just
of
doing
you
must
release
things
immediately.
So
this
we're
not.
You
know
desperate
to
pander
to
these
people,
but
it
is
nice
to
have
a
release,
schedule
and
cables,
but
somewhat
mirrors
the
latest
and
greatest
go
give
us
I'm
sure
it's
not
attending,
but
he
is
working
on
a
new
release
of
J
ahsoka,
that's
over
35,
but
it
got
stuck
the
Colonel's.
Basically,
problems
with
DHT
performance
means
that
announcing
agile,
professed
release.
A
That
says:
hey,
look,
everyone,
it's
got
the
DHT
and
then
then
trying
it
and
basically
having
a
terrible
time
didn't
seem
like
a
good
thing
to
push
on
the
world.
So
the
the
DHT
work
is
being
backed
out
of
the
out
of
the
five
release
so
that
they
can
get
out
of
the
door
because
it
has
a
bunch
of
other
fixes
in
the
drone
and
then
I
think
digitally
are
going
to
the
over.
Basically
is
what
he's
saying
other
bug
fixes.
A
Gui
team,
okay,
arts,
the
we
spoke
about
swine
squawking
about
it
and
we
came
up
with
dish
list
of
things.
The
conversation
in
the
chat
is
not
for
me
to
see
so
please
just
do
you
have
volatile
taking
notes,
so
no
one
wanted
taking
notes
so
Dietrich
voluntary
particular.
Thank
you
very
much.
That's
okay.
The
notes
section
is
definitely
more
relevant
for
the
agenda
items
such
as
people
saying
what
we've
done.
A
Anyone
who
was
at
Lisbon
might
recognize
this
list
for
the
blue
team.
It
was
a
good
list.
We
spoke
about
it
a
lot.
It
didn't
contain
anything
significant
about
the
package
managers
endeavor,
which
is
the
number
one
priority
for
the
entire
organization.
This
issue
has
noticed
by
the
project
overlords
and
we
responded,
and
so
now
we
never
added
a
get.
A
thousand
users
actively
re,
hosted
in
modules,
quiet
and
p.m.
when
I
give
this
map
as
a
top
priority
for
the
Gary
thing,
it
was
a
subject
that
came
up
in
Lisbon,
but
it
wasn't.
A
We
weren't
clear
how
to
approach
it
in
a
way
that
would
be
super
valuable
and
we
got
excited
about
other
things,
but
when,
when
people
were
like
hey
come
on
guys,
we
didn't
really
have
a
leg
to
stand
on,
so
things
have
been
realigned,
but
we
don't
know
what
to
do
yet.
There's
a
couple
of
things
that
I
need
from
folks
on
is
cool.
Hey,
that's
a
very
little
feedback
on
this
list,
and
this
is
what
the
GUI
team
will
be
working
on
for
the
next
three
months.
A
So
please,
oh,
wait
because
loads
of
feedbacks,
just
plus
ones,
all
over
the
place,
yeah
great,
a
lack
of
comment-
means
everyone's
up
for
it.
Super
cool
we
didn't
take
out.
The
the
thing
that
we
got
excited
about
was
have
a
service
that
makes
it
super
easy
to
deploy
websites.
Okay,
that's
it's
not
out
of
the
list,
but
it
has
been
dropped
down
to
a
p2.
So
it's
something
that
I'm
personally
excited
to
work
on
so
eventually
in
spare
cycles.
Who
you
make
progress
with
that?
A
Alex
raised
an
issue
whereby
we
were
doing
a
bad
job
of
preserving
the
base
that
the
IDS
work
avoided
in
in
the
word.
Fix
is
99,
which
is
the
emergency
services
in
the
UK.
There's
no
subject
so
very
much
like
it's
the
issue
number
that
I
got
so
obviously
the
IP
of
the
Explorer
is
my
baby
and
I
get
very
upset
when
people
discover
problems
with
it.
A
I
like
so,
it's
like
here
is
a
glaring
problem
with
the
IP
of
the
Explorer
and
you
pass
it
a
base
30
to
use
the
ID
and
it
will
force
it.
It
will
brutally
coerce
it
into
a
face
with
the
opt
C
flavor,
which
is
exactly
not
what
you
want
from
this
tool,
because
it's
expressed
purpose
to
take
one
give
it
and
to
have
it
explained
to
you
information
about
the
CID.
A
You
gave
it
not
for
it
to
pretend
it
was
Miller
CID,
but
instead
another
problem
was
deeper
deeper
in
stack
and
basically,
when
you
pass
a
string
to
a
JSC
ID
object,
it
was
the
thing
that
was
losing
information
about
the
base
you
had
provided
so
I
think
I
mentioned
was
briefing
last
week,
but
the
problem
in
the
old
world
was
looking
past.
A
CID
object,
Australian
that
was
included
in
misleading.
It
would
not
preserve
information
and
return
it
to
the
default
density
anyway.
A
A
The
documentation
so
that
beginners
might
move
verifying
figuring
out
how
to
use
Jack's
ready,
because
the
whole
thing
is
quite
an
odd
look
at
that
Geo.
We
can
come
back
to
that
when
we
look
around
and
p.m.
do
any
question
to
me
and
Faline
on
I
mean
definitely
so:
okay
yeah,
so
I'm.
Sharing
this
the
thing.
D
Where
am
I
yeah,
so
we
shipped
the
NiO
project
structure
that
I
talked
about
last
week
for
those
who
don't
know
what
I'm
talking
about.
Basically,
we
componentized
some
of
the
main
components
for
school
and
now
they're
really
components
because
they
weren't
and
what
that
makes
it's
it
give
us
the
ability
not
for
us,
but
for
everyone
who
is
going
to
to
have
tutorials
the
process
is
so
much
easier,
I'm
not
going
to
show
it
again
because
I
already
showed
it
last
week,
but
it
is
now
and
it's
it's
live.
D
D
For
example,
this
wasn't
a
link.
So
if
you
want
wanted
to
go
to
the
mini
landing
page,
you
had
to
come
to
the
tutorials
and
then
choose
the
tutorial
again,
but
right
now
this
is
a
link
so
you're
inside
the
data
structures.
On
any
lesson,
you
just
click
here
and
you
go
to
the
index
of
that
lesson.
So
that's
a
bit
that
helps
something
so
I
made
some
other
fixes
yeah.
D
But
my
my
main
goal
and
what
I've
been
looking
at,
is
the
code
validation
until
now
that
wasn't
a
big
problem,
but
right
now
is
being
a
problem
with
the
new
on
what
Terry
and
I
are
working.
But
it's
a
new
lesson
of
the
IP
FSM
arrest,
so
we're
adding
files
were
creating
new
directories
what's
happening
right
now,
yeah
I
have
to
go
to
this
French.
D
So
what
is
imagine
you
had
a
file?
This
is
a
lesson
that's
being
worked,
but
what's
happening
is
when
this
error
that
you
see
here
this
was
not
supposed
to
be
here,
because
this
error
is
being
is
being
IVFs
is
erroring
and
instead
of
this
being
spit
out
to
us-
and
we
can
say:
okay,
this
error
occur.
So
just
let's,
let's
put
a
message
out
that
the
user
will
understand.
That's
not
happening.
This
is
going
over
an
algorithm
and
it's
being
automatically
appearing
here
so
basically
had
no
control.
D
D
Looking
at
this
from
what
I've
have
been
looking,
I
think
this
will
will
be
a
huge
refactor
to
to
make
it
work,
but
I
think
we
have
to
do
this
because
a
user
that
doesn't
know
a
lot
about
ipfs.
These
errors
are,
they
aren't
met.
They
aren't
clear
what
you
have
to
do,
so
we
really
need
to
fix
this.
So
I
think
I'll
continue
to
work
on
this.
This
is
a
priority
to
make
the
files
upload
license
worth
having
I
think.
E
D
Yeah,
but
the
thing
is
that's
our
goal,
but
the
problem
right
now
is
how
this
is
wire.
So
the
way
this
is
what
buyer
gave
you,
this
wasn't
taken
on
consideration.
So
that's
the
main
problem.
Okay,
if
you
need
some
else
just
be
alright,
thanks,
I'll,
take
a
look
and
probably
I'll,
have
some
idea.
Well,
I
can
make
this
better.
F
Something
I
just
want
to
call
out
that's
cool
the
project
restructuring
that
Diego
took
the
lead
on
is
we
ended
up
getting
to
a
point
where
we
realized
it
made
sense
to
build
in
now.
This
concept,
of
course,
is
like
it's
not
fully
structured,
so
we
can
kind
of
create
one
in
the
sense
that
we
envision
long
term,
but
there's
you
know
an
array
of
things
that
we
want
to
appear
on
the
homepage,
as
featured
and
there's
an
array
of
things
that
we
want
to
appear
in
the
list.
F
G
E
E
We
add
some
back-and-forth
in
that
discussion,
I.
Think
right
now.
We
finally
convinced
Alan
on
a
compromise,
so
hopefully
that
will
be
merged
quickly.
I
just
need
to
finish
up
some
of
the
feedback
on
that.
Pr
also
did
some
work
on
the
CLI
moved
around
some
any
of
em
FS
commands
that
were
in
the
wrong
place,
added
a
safer
way
to
handle
exits
or
force
exits.
I
still
need
to
fix
the
test,
because
some
of
them
are
failing
and
still
run.
E
The
JSP
FS
files,
API
I,
will
continue
my
research
on
IPS
ipfs
over
whatever,
hopefully
I'll
figure
out,
then
the
DNS
part
quickly.
I
have
some
questions
then
to
to
find
answers.
We
also
have
a
meeting
next
week
with
the
other
guys
that
are
looking
at
IPS
like
alien
and
Pedro
and
stuff,
like
that
and
I
will
fix
JSF
EFS,
because
right
now
we
cannot
resolve
ibn
as
the
path
we'd
basically
domains,
and
that
will
be
helpful
for
IP
a
nice
and
helpful
helpful
for
browsing
the
intubation,
brave
integration
and
yeah.
That's
for
now.
A
C
Mpm
are
the
package
management
priority,
but
I
am
sharing
the
e.
Oh
here
we
go
this
one.
If
you
look
at
this
issue,
which
I
will
also
link
that
screen
is
motion,
it's
not.
C
Shared
a
repeat
warning:
oh
that's:
alright,
Freda,
yes,
not
a
single
click,
it's
a
double
click
or
a
quad
click
if
you're
want
to
make
really
sure
so.
Yeah
support,
ipfs
camp
p0
support
package
managers,
p0
and
a
whole
bunch
of
things
around
reducing
surprises
for
web
developers,
but
only
couple
hundred
p0.
C
But
it's
been
cool
to
see
the
overlap
between
many
things
that
we
need
to
support
a
BFS
camp
like
view
and
the
things
that
we
need
to
support
package
managers
like
being
faster
and
all
of
these
things
also
already
making
progress
well
week
after
we
got
out
of
list
fitter
and
made
all
these
decisions,
and
only
a
few
days
after
we
made
more
decisions
after
getting
back
from
Lisbon.
One
of
the
things
I
did
last
week,
too,
is
record
the
video.
C
A
C
I
could
I
could
rerecord
it
in
a
whole
bunch
of
times
to
moving
back
to
the
things
that
I
actually
did
did
a
few
things
around
the
collaborations
work
some
stuff,
that's
not
necessarily
too
directly
to
web
browsers
working
group,
but
involves
some
web
browsers.
We
are
talking
to
a
bunch
of
friendly
collaborators
that
want
to
put
that
are
using
IP,
that's
already
and
want
to
put
it
in
more
places.
So
those
are
the
things
that
is
a
ecosystem
growth
engineer
that
I'm
really
excited
by
we'll
have
more
details
about
that.
C
But
there
are
things
like
I
talked
to
somebody
at
Microsoft
about
their
side,
tree
project
which
uses
IP
FS
for
identity
services,
you're
interested
in
the
identity,
work
which
I
think
from
our
project
perspective
has
been
a
little
bit
deprioritized
for
this
year,
but
is
still
ongoing
in
a
lot
of
different
areas.
A
couple
of
different
things
to
you
around
kind
of
like
community
post
kind
of
stuff,
so
harvesting,
Stack,
Overflow
posts
being
aware
of
what
type
of
questions
people
are
asking.
C
B
A
G
Just
keep
my
daily
chart.
I
finally,
have
a
protocol
AI
email,
so
you
don't
have
to
email,
Jimmy,
Jimmy,
calm
anymore,
although
you
can
still
do
that.
I
set
up
a
little
subdomain
for
Pierre
pad
network
and
put
links
to
these
little
hashes
and
things
that
sort
of
come
and
go
and
I've
got
like
a
little
experiment
going
on
there,
I'm
not
done
with
it
yet,
and
why
don't
he
just
touch
base
about
ipfs
camp
and
help
planning
out
what
the
building
the
the
how
to
build
apps
with
I
profess?
G
H
H
Kind
of
spurred
me
to
do
a
quick
mock-up
of
what
it
might
look
like
if
it
was
more
of
a
dashboard
arrangement.
Sort
of
the
notion
is
like
Oh,
crazy
ideas.
Let's
do
some
weird
and
then
and
then
hear
about
all
the
reasons
why
that
sort
of
approach
was
never
done,
and
then
you
know
maybe
kind
of
you
know
it
helps
me.
This
is
mostly
about
helping
Erica
understand
that
all
this
works,
but
also
maybe
you
know
it-
will
help
help
me
formulate
some
actual
recommendations
that
are
incremental.
H
You
know
there's
like
functional
thing,
sort
of
a
dashboard,
readout
thing
and
then
sort
of
an
educational
thing
and
then
settings
so
they're
kind
of
different
different
beasts
as
and
and
they
also
are
the
sections
you
know
some
sections
I
might
want
to
rarely
visit
like
settings.
You
know-
and
so
this
this
minimizes
it
but.
H
H
If
you
have,
if
you
have
some
some
thoughts
there
and
then
the
I
did
a
little
survey
about
ipfs
desktop,
primarily
I
again
just
the
idea
here
is
just
educate
me
plus
to
hopefully
remind
us
all
about
like
what
is
the
the
user
that
we
have
in
and
how
useful
is
what
we're
making
for
them
and
what
little
incremental
things
can
we
do
to
push
push
things
towards
being
a
little
bit?
You
know
as
useful
as
possible
and
a
little
better.
A
F
I
just
figured
it
out
Oh
share,
so
one
thing
that
I
just
did
is
made
part
of
school
works
in
social
cards.
So
this
is
what
it
look
like
on
Twitter
and
Facebook,
maybe
we'll
decide.
We
want
different
pictures
or
words
at
some
point,
but
this
is
much
better
than
not
having
these,
and
this
is
something
that
we're
gonna
have
to
fix.
Once.
F
If
I,
you
think
you
have
people
in
your
networks
that
care
about
offline
first,
whether
or
not
they
care
about
decentralized,
feel
free
to
share
that,
and
we
will
make
the
big
announcement
of
the
location.
Soon
lots
of
other
protocol
UX
stuff
in
the
works
do
go,
has
been
awesome.
So
we've
spent
a
lot
of
time
kind
of
brainstorming
and
working
together
to
make
stuff
happen.
There
I
built
a
react
component,
there's
an
app
it's
in
react.
It
works
for
my
class
more
to
come
on
me
actually
know
and
react.
F
So
the
files
tutorial
I
think
we
might
have
talked
a
little
bit
about
it
on
this
call
last
week
that
I
was
feeling
stuck
between.
Are
we
doing
em
FS?
Are
we
doing
not
MSS
so
I'm
feeling
better?
Because
I've
picked
the
clear
direction
which
is
to
make
this
tutorial
will
be
about
em
FS
in
part,
because
a
few
people
agree
that
it's
the
most
kind
of
beginner
focused
content
and
in
part
because
blobs
until
Hugo
fixes
it
aren't
supported
in
the
non
em
FS
methods
in
the
browser.
F
So
we
have
lessons
now
and
files
right,
falsa,
LS
and
then
I
added
one
yesterday
on
file
dot,
mcdr,
which
I
just
haven't
done
the
validation
for
it's
a
very
interesting
learning
process,
because
I
get
like
I
look
at
the
documentation
and
then
I
write
what
seems
like
it
would
probably
be
the
truth
like
translating
the
this
much
of
the
repo
to
sentences
about
how
you
would
do
something
and
then,
when
I
try
to
do
the
validation
I'm
like?
Oh,
that's,
not
how
that
works
right.
F
Alright,
let's
try
again
I
won't
talk
about
the
validation
challenges.
Cuz
you
already
hit
on
that
Chrissa
PR,
if
you're
interested
so
next
step,
actually
doing
the
launch
for
camp
and
then
more
work
on
this
and
my
first
lesson.
This
is
the
best
thing
in
terms
of
content
that
I'm
working
on
there's
lots
of
UX
stuff.
They
were
so
in
the
middle
of,
and
then
better
instructions
for
tutorial
authors,
Allium,
hoping
that
you
will
have
some
time
to
poke
around
in
the
structural
changes
that
Diogo
made,
which
we
did
already
put
up.
F
There's
more
that
will
come
when
the
file
upload
is
added
that
changes
it.
But
we've
already
made
some
improvements
to
the
instructions
for
adding
your
tutorial
a
piece.
That's
missing.
There
is
just
what
is
good
instructional
methodology
like
one
thing
at
a
time
or
those
kinds
of
concepts,
and
you
so
you
mentioned
a
couple
of
things
when
we
were
in
Lisbon.
One
of
them
was
helping
people
understand
what's
good,
educationally
and
one
was
making
sure
that
people
can
build
tutorials
easily.
F
So
I'm
curious
how
much
of
your
ideas
for
like
restructuring
the
project,
to
support
that
building
we've
already
knocked
often
the
changes
we
made
or
if
you
had
any
other
ideas
that
we
should
also
be
implementing
now
before
other
people
build
for
camp
I,
think
you'd
volunteered
to
kind
of
be
the
guinea
pig
and
go
first
for
the
next
set
of
building.
So
if
you
have
feedback
that
stuff
that
we
should
be
changing
sooner
rather
than
later
to
support
camp
that'd
be
great.
A
It
was
mostly
me
volunteering
to
be
the
guinea
pig
and
we
did
a
bunch
of
work
for
a
few
years
back
when
we
had
to
build
a
workshop
and
give
it
to
people,
and
so
can
only
I
can
offer
from
experience,
but
and
mostly
look
I'll,
be
a
bikini
pig
and
say
this
doesn't
make
sense
great,
that's
Jessica.
Did
you
want
to
share
a
thing.
A
I
I
B
So
it's
like
sort
of
a
follow-up
to
that
problem.
We've
just
CID
solved,
but
taking
it
step
further.
Basically,
right
now,
when
people
opt
in
to
adding
content
to
ipfs
using
CID
v1,
they
get
CID
v1
in
base,
58
I,
think,
and
then
we
everywhere
we
usually
say
that
base
42
is
the
future.
But
if
it's
the
future,
why
the
default
representation
of
CID
v1
is
still
based
58.
So,
like
my
idea,
is
that
basically,
can
we
like
just
change
the
default
representation
of
the
adg-1
everywhere?
B
A
That's
perfectly
reasonable.
You,
you
sort
of
in
charge
of
wrangling
forward
progress
on
the
endeavor,
so
I
think
you
get
to
say
we.
We
should
do
this
and
rattle
rattle
sabers
and
make
sure
that
there's
no
burning
reason
I.
Think
Alan
would
have
a
more
insight
into
what
blocks
that
right
now,
but
his
process
has
been
change
all
of
the
foundational
building
blocks,
so
they
don't
set
on
fire
when
they
do
receive
for
base
30
to
encode,
it's
the
ID,
so
that
when
we
come
to
actually
flip
the
switch.
B
The
thing
is
that
I
believe
this
would
be
part
of
preparing
our
libraries
and
entire
record
system
for
nightly
has
come.
It
would
just
remove
this
element
of
surprise
that
people
we
say
you
can
use
CATV
one
today.
Jujhar
can
opt
in
to
that
and
remove
the
problem
of
converting
it
between
different
bases,
so
I
probably
create
issues
for
go
ipfs
and
jails.
Ipfs
yeah.
A
A
D
A
Yeah,
let's,
let's
change
the
default
in
JC
ID
and
see
what
happens
all
right.
One
way
of
doing
it.
A
C
I
have
one
ask
which
is
real
quick
if
anybody
has
links
to
the
issues
or
previous
discussions
around
this,
if
you
could
drop
them
in
there,
that
would
be
great
review
and
talk
about
it
until
at
least
my
my
issue
with
another
really
short
one,
which
is
if
we
are
gonna.
If
it
is
time
to
do
a
care
of
you,
maybe
we
should
schedule
a
a
part
of
this
meeting
to
be
dedicated
to
that,
so
that
we
can
share
in
this
meeting
very
explicitly
in
the
part
of
the
meeting.
A
Shoes,
maybe
we
should
have
a
column
from
like
waiting.
Another
week
seems
an
ideal
this
this
time
tomorrow
to
talk
about
cars.
Would
anyone
be
up
for
that
interested
I
can't
see
everyone
so
I'm
going
to
stop
sharing
my
screen.
Okay,
I'm
gonna
send
out
an
invite
first.
Another
call
this
time
tomorrow
to
do
immune.
Okay,
I'll
check
it's
three
minutes
to
go.
I
really
want
to
see
hugosthere
Hugo.
Would
you
do
us
the
honors.
E
E
Hopefully
a
hundred
milliseconds,
but
that's
another
matter.
This
still
very
early,
there's
still
some
bugs
to
be
fixed
on
some
areas
of
the
cold,
because
some
data
store
stuff
I'll
still
need
to
go
through
all
the
use.
Cases
like
the
gateway
result
pointing
a
dns
to
the
gateway
and
the
gateway
redirecting
through
the
DNS
link
and
stuff
like
that,
but
yeah.
This
is
basically
doing
all
the
validation
that
boots
validates
everything.
E
E
Yeah
the
next
step,
this
is
basically
working
through
CloudFlare
workers,
so
we
type
in
s
/
workers,
what
yeah,
so,
basically
it
uses
the
work
assist
the
court,
the
workers
as
the
API,
so
it's
highly
distributed.
This
is
perfect
for
the
browser,
because
all
the
HTTP
requests
to
resolve
an
IP
NS
will
be
really
fast
because
it
they
will
be
really
near
the
end
user
and
the
worker
basically
saves
the
records
on
the
key
value
store
that
each
workers,
each
worker,
as
access
to
and
yeah.
That's
basically
how
this
works.
E
The
next
step
will
be
DNS
right,
that's
gonna,
say
interesting
dependency,
yes,
selves
yeah!
It's
the
the
way
to
be
able
to
do
this
without
saying
to
the
end
users.
Okay,
you
need
to
run
some
demon
to
answer
the
NS
requests
or
stuff
like
that
needs
to
be
a
solution
like
this.
This
is
not
like
the
final
solution,
but
it
works
and
it's
self-certifying.
A
E
A
E
We
always
like
the
demon
will
try
all
the
methods
that
it
has
available
to
them.
So
we
can
try.
I've
been
s
over
the
DHT,
can
try
a
penis
over
the
dnas
I've
been
s
over
whatever
and
basically
the
first
one
wins
and
people
can
opt
out
or
opt
in
to
each
each
of
them.
So
I
think
that's
the
trick,
like
the.
A
The
lip
p2p
stack
is
designed
to
be
configurable,
so
you
choose
the
pieces
and
transports
and
discovery
mechanisms
that
make
sense
for
your
app
I'm
unaware
of
like
how
pluggable
the
IPS
resolution
mechanism
is
right
now,
but
it
should
be.
It's
not
really
pluggable
right
now,
it's
a
lot,
but
it
should
be
right
yeah.
It
should
be
alright!
That's
here!
That's
your
angle
like
this
is
a
totally
valid
implementation
of
what
should
be
a
strategy
pattern.
A
A
So
the
point
there
was
like
oh,
but
we
don't
want
to
have
any
points
of
centralization
and
we're
like.
Oh,
we
have
two
battles
to
fight,
and
one
is
that
like
if
s
needs
to
work
in
an
offline
scenario,
but
and
in
an
off
in
a
fully
like
the
there's,
a
net
split
or
there's
like
the
backbone
has
been
broken.
A
You
will
tolerate
a
few
seconds
of
lookup
time
using
a
DHT,
but
if
you're
also
competing
against
HTTP
and
centralized
services,
you
should
use
mechanisms
that
that
also
work
quickly
where
there
is
where
there
is
those
options.
Work
fast,
as
you
can,
when
you
have
all
of
the
things
available
and
work
when
everything
is
on
fire,
yeah.
E
Music
plays
it's
like
that.
We
just
need
for
now
a
near-future
solution
that
it
actually
works.
So
you
gets
people
to
start
using
this
stuff
and
publishing
websites
with
this
stuff.
Just
like
that,
the
dead
guys
did
it.
They
started
with
something
that
works,
and
then
they
are
evolving
to
something
more
decentralized.