►
From YouTube: IPFS GUI and Browsers Biweekly 2020-06-16
Description
Meeting notes: https://github.com/ipfs/team-mgmt/pull/1163/files
About IPFS GUI and Browsers Biweekly: https://github.com/ipfs/team-mgmt/issues/790
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
Tada,
so
if
anyone
is
interested
in
tracking
that
there's
the
second
metal
shoe
with
initial
description
of
how
we
imagine
that
integration
will
look
like
those
are
very
broad
strokes,
but
super,
exciting
and
I
think
another
one
is
that
the
latest
version
of
ipfs
companion
lets
people
to
opt
in
to
using
the
latest
web
UI.
So
not
everyone,
but
some
folks
were
complaining
that
when
you
click
on
there,
when
you
open
web
UI
from
companion,
it
will
pick
the
version
that
is
hard-coded
in
Jes,
ipfs
or
goid
preference.
A
So
if
you
want
to
run
the
latest
and
the
greatest
from
web
UI
that
I
give
as
IO-
and
you
trust
your
DNS
resolver
and
you
set
up
all
the
course
headers,
then
you
can
toggle
the
switch,
and
my
previous
companion
will
open
the
latest
web
UI
build
from
from
its
rep
rep.
Oh
it's
as
far
as
I
know.
It's
the
latest
version
from
the
master
so
be
wary
of
that.
A
D
A
So
next
steps
for
opinion,
KPI
integration,
long
story
short-
we
want
to
add
support
for,
like
third-party
pinning
services
such
as
Panetta
and
others,
and
add
UI
to
web
UI
and
I'd.
Give
us
desktop
to
be
able
to
quickly
pin
something
remotely
and
we've
identified
that
there's
a
problem
of
state
right
now.
Web
UI
itself
is
just
a
single
page
application,
which
is
effectively
which
you
point
at
the
remote
API
and
that
remote
API
represents
a
node
and
you
can
switch
between
nodes.
A
So,
apart
from
like
the
language
that
you
choose
and
the
eccentric
country,
metrics,
there's
no
state
no
cookies
in
web
UI,
which
could
be
leveraged
for
storing
the
list
of
pinning
services.
And
if
you
define
pinning
service
in
as
desktop
which
embeds
web
UI,
you
would
want
to
see
the
same
list
of
defined
pinning
services
when
you
open
web
UI
from
a
Kieffer's
companion
right.
A
So
we
need
to
persist
the
state,
the
list
of
paying
services
in
the
node
in
go
ipfs
somehow
and
ensure
the
is
to
you
to
create
the
config
namespace
for
pinning
remote
pinning
services.
It
makes
sense
because
if
we
have
that
in
the
config
then
go
ipfs
itself
could
start
leveraging
that
information
and,
for
example,
you
could
have
when
you
pin
something
from
the
command
line,
it
could
automatically
pin
something
remotely
as
well.
A
But
that's
just
like
initial
idea,
we'll
probably
have
a
document
with
a
sketch
of
api's
that
we
need
from
the
UI
side
we'll
have
like
a
probably
a
separate
call
about
that.
We
just
can
refer
after
this
call,
but
it's
just
like
to
given
status.
Update
on
this
endeavor
is
there
anything
to
add
Jessie
Caravelle.
C
D
A
F
F
A
F
Yeah
so
essentially
we
have
another
ipfs
implementation
that
works
over
message
protocol
and
it
implements
a
subset
of
IHS
ipfs
api.
At
the
moment,
dog
api
under
ipfs
is
fully
implemented
and
it
passes
in
most
of
the
tests.
A
few
tests
are
disabled
because
they're
doing
some
argument,
normalization
that
I
had
not
implemented
but,
as
I
said,
it
does
pass.
I
kept
as
good
ad
function
had
been
also
fully
implemented
and
all
the
tests
passing
except
few
today
de
sel,
because
they
depend
on
IPS
that
object
or
ipfs.
Don't
start.
F
Oh
sorry,
false
start
it
guys
that
are
not
implemented
so,
like
we
run
those
tests,
as
this
work
required
bunch
of
changes
in
other
libraries
within
the
dependency
trace
setting
to
make
you
and
erase
usable
in
place
of
buffers.
So
they're
like
multiple
levels
of
dependency,
so
they're,
slowly
propagating
once
they
make
its
and
I
will
push
them
to
IP
as
well
yeah.
So
with
the
MV,
I
only
have
ipfs
cut
left
and
once
that's
done,
I'll
be
requesting
review
and
it
will
be
a
lot
of
fun
because
it's
a
lot
of
changes.
A
And
for
like
additional
context
is
there's
a
lot
when
you're
at
least
started
working
on
this
we've
identified
a
lot
of
problems
when
they
need
for
passing
data
or
passing
content
identifiers
between
different
browsers
in
context,
the
need
for
like
serialization
and
deserialization.
Not
everything
was
being
just
like
passed
in
the
browser
context.
So
a
lot
of
yeah.
F
F
So
a
bunch
of
things
were
that
still
an
issue,
but
in
the
meantime,
I
added
a
serialization
deserialization
face
on
both
sides,
and
there
are
ongoing
discussions
on
trying
to
make
some
changes
in
the
library
surrounding
to
not
have
that
step,
but
I'm,
not
sure
if
they
will
ever
be
made
or
not.
Yeah.
A
That's
I
mentioned
it,
because
we've
initially
hit
that
prob
of
multiple
layers
of
either
cell
realizing
this
realizing
overall
performance
impact
or
when
we
introduced
winter
ipfs
experiment-
and
we
had
this
like
all
the
postmessage
probe
seeing
in
place.
I
know
there
are
like
right
now,
it's
disabled
in
companion,
but
it
may
get
back
at
some
point
in
some
form,
but
there
are
like
meta
mask
is
probably
a
heating
symbol
at
sample.
It's
also
like
a
browser
extension
and
they
have
multiple
contexts
and
they
inject
window.
A
What
free
and
window
et
room
in
two
pages
so
there's
like
probably
the
same
problem
space
I,
know
brave,
is
using
fork
of
meta
mask
and
it's
bit
older.
It
doesn't
support
those
new
experimental
plug-in
systems,
but
at
some
point
it
may
be
that
if
we
have
a
95
geofence
in
brave
and
they
decide
to
implement
that
API
is
those
api's
from
like
window
web
three.
We
may
once
again
get
to
the
point
when
those
performance
issues
relevant
again.
A
F
Say
it's
not
really
a
performance
issue
as
much
as
it's
a
complexity
issue,
but
yeah
I
I,
don't
think
the
performance
of
it
will
be
as
impactful.
It's
probably
negligible
and
crown
scheme
of
things,
but
complexity.
Overhead
is
really
high
and
obviously
likes.
It
feels
like
fighting
the
browser
environment
rather
than
playing
with
it.
So
if
we
can
make
changes
to
it
to
play
nicer,
that
would
be
I
seem
good,
so
hopefully
we
can
make
it
there.
F
F
C
That's
cool
next
one's
also
me
I've
been
joining
so
that
you
probably
all
saw
the
work
that
Microsoft
launched
service
that
is
instance
some
of
the
decentralized
identity,
efforts
and
standards
implementation
of
the
IDs.
They
are
also
working
with
the
larger
group.
That's
combined
w3c
and
decentralize
identity
foundation
work
on
kind
of
when
they
were
we're
calling
encrypted
data
bolts.
C
This
idea
of
personal
storage,
user,
centric
storage
that
travels
around
the
separation
between
user
storage
of
data
and
the
services
and
applications
that
might
operate
against
that
data,
and
now
this
is
called
the
secure
data
storage
spec.
At
this
point,
this
group
is
meeting
weekly.
The
meetings
are
open
if
people
are
interested
I
link
to
a
couple
of
interesting
things,
we
have
some
some
folks
like
some
of
the
Microsoft
people.
Are
there
and
Johnny
crunch?
C
Is
there
and
a
few
other
folks
that
consistently
bring
up
Ori
steel
who's
been
following
some
of
the
issues
in
our
repos
recently
relating
to
this
work
that
have
been
pushing
ipfs
as
a
huse?
So,
every
time
there's
a
HTTP
centric
part
of
the
conversations
of
one
of
those
folks
will
pop
their
hand
up
and
say.
Well
how
would
this
work
with
IP
of
us
and
it's
great,
because
it's
really
quickly
re
oriented
the
thinking
of
that
group
to
really
consider
Network
agnostic
protocols
and
standards
right
now,
these
reading
these
meetings
are
all
weeds.
C
They're,
basically
like
at
this
point,
they're
like
where
what
is
the
application?
What
are
the
layers
so
like
what
the
one
issue
I
looked
at
all
right
link
to
is
a
proposal
for
one
way
of
defining
what
the
layers
are
and
there's.
It's
actually
really
interesting,
because
it's
a
lot
like
this
already
a
spec
and
everybody
kind
of
knows
where
they
want
to
get
to
in
the
bigger
picture,
but
they
are
really
working
on
like
language,
application,
stack
and
layer
norms.
C
At
this
point,
but
I
think
the
salient
bit
for
us
is
more
that
that
storage
is
expected
to
be
content
addressable
as
as
a
as
a
function,
and
that
that
is
very,
very
interesting
for
us
in
a
long
term,
and
some
of
the
partners
who
are
working
on
really
expressed
an
interest
of
ipfs
should
really
should
be
like
the
normative
default
there
as
an
example.
So,
even
if
it's
not
named
and
expect
a
lot
of
the
prototypes
and
examples
we'll
use
IP
fest.
C
So
it's
a
pretty
big
win
already
just
in
that
culture
of
having
this,
but
people
start
to
think
in
a
not
HTTP.
You
know
blinders
way,
they're
starting
to
think
and
network
agnostic
out
of
the
box.
When
approaching
these
types
of
standards,
it
big
opportunity,
yes
for
us
on
the
long
term,
but
interesting
for
the
work
that
we're
doing
in
browsers.
C
I
think
I
think
I,
remember,
I,
think
it's
just
community
group
and
then,
but
they
had
to
work
out
like
the
IP
and
patents
ease
to
be
able
to
make
it
a
joint
decentralized
identity,
foundation
and
w3c
effort.
So
they've
kind
of
like
at
the
broader
level,
figured
out
how
those
two
organizations
work
together,
so
I
think
eventually
will
probably
be
more
formalized
and
though
III
see
it's
not
just
I.
Think
it's
just
a
community
group
in
this
one.
A
All
right,
we
are
at
the
end
of
the
agenda
and
moving
to
highlights.
If
you
have
any
topic
that
you
want
to
add
to
agenda,
that's
a
good
time.
I
will
be
talking
a
little
bit
about
the
first
item.
The
first
item
is
in
the
past
weeks
we
had
a
release
of
ipfs
companion,
and
then
we
had
a
release
of
ideas
companion,
so
we
got
to
13
Oh,
which
shipped
revamped
welcome
screen
with
all
new
on
updated
on
burning
materials.
A
So
big,
thank
you
for
Jessica
for
doing
all
the
work
I
just
shipped
it
with
a
release.
So
now,
if
you
want
to
go
to
the
welcome
screen,
you
can
click
on
this
cube
and
you
are
back
to
the
welcome
screen
before
you've
seen
it
only
once
and
install
now.
You
can
get
back
to
it
and
another
thing
that
happened
is
all
the
doc
links
point
to
the
new
doc
website,
which
is
a
frequent.
E
A
Totally
I
was
about
to
say
that's
now
you
can
instead
of
like,
depending
on
this
one,
we
probably
need
to
add
it
here
new
Doc's
website.
We
also
link
everywhere
to
it
from
that
landing
page
and
small
bug
fixes,
but
in
the
2:14
one
there's
this
opt-in
for
like
this
web
UI,
and
we
also
got
some
new
languages
I
think
it's
a
I
think
this
really
shipped
Arabic.
However,
we
did
not
do
the
proper
audit
of
ipfs
companion
or
any
other
of
our
interfaces.
A
E
Yes,
this
is
super
quick,
so
the
ipfs
GUI
usage
survey
that
we
did
like
a
month
ago.
I
have
finally
gone
through
all
of
the
vervain
responses
for
that,
as
well
as
the
like.
Almost
two
thousand
verbatim
responses
from
the
companion
uninstall
survey
that
gets
kicked
back
to
people
who
try
to
categorize
some
of
that
stuff.
There
is
a
link
there.
I'm
not
gonna,
share
my
screen
because
right
now
it's
just
got
everything
in
in
some
graphs
and
I
want
to
add
some
additional
notes,
but
I
will
update
this
group
TL
DR.
E
E
It
also
provides
some
really
strong
backing
for
support
in
the
GUI
apps,
for
both
pinning,
whether
that's
third-party
pinning
services
and
then
also
a
lot
of
talk
about
the
use
case.
If
I
run
five
I
run
five
nodes
in
my
house
and
I
want
to
make
a
tiny
private
cluster
form,
for
example.
So
watch
has
face,
but
that's
those
are.
Those
are
mainly
the
big
two
takeaways
I've
seen
so
far.
E
Some
I
mean,
obviously
the
feedback
that
we
saw
for
companion
is
on.
People
who
use
companion
and
I
haven't
got
any
deeper
into
it.
That
said,
we
did
see
a
limited
amount
of
feedback
of
just
there
is
no
beginner
friendly
stuff
for
the
other
graph
is
as
well.
You
know,
yeah,
it's
gonna
be
an
order
of
magnitude
higher
than
stuff
that
people,
just
like
randomly
they're
like
looking
up
ipfs,
and
they
see
this
extension
and
they're
like
okay,
cool
but
but
I
mean
you
know,
it
is
still
a
valid
point.
You
know
we're
still.
B
A
For
additional
context,
most
of
companion
users
are
chromium
users,
so
we
have
over
31,000
weekly
users
who
installed
companion
from
Chrome
Web
Store,
and
we
have
around
3,000
of
users
in
Mozilla's
add-on
store.
So
it's
like
10
10
times,
chromium
users
and
that's
Google
Chrome,
but
also
brave
Microsoft
edge.
The
new
one
and
even
Oprah
for
desktop,
like
all
those
browsers,
are
chromium
based.
E
A
C
C
A
Just
yeah
so
like
the
Welcome
page,
is
very
clear.
Now,
that's!
Okay!
That's
one
more
step!
You
need
to
either
start
your
note
or
install
app.
If
it's
desktop,
however,
like
we
probably
should
be
clearer
up
front
before
someone
instance
it
because
some
people
just
close
the
tab,
anything
that's
opens
on
the
install
people
tend
to
close
it,
definitely
something
to
improve
raphaël.
Do
you
want
to
give
a
quick
update
on
the
accessibility
fixes
in
your
web
UI
yeah.
D
Now
you
go
to
the
content
of
the
page
first,
since
the
users
want
the
content
of
the
page
first
and
also
fix
a
lot
of
focus,
focus
elements
missing
a
lot
of
feedback
as
well
buttons
all
that
good
stuff,
so
yeah
when
the
next
version
of
Yui
ships,
please
test
it
out.
If
any
of
you
are
power
users,
when
I
mean
power,
users
I
mean
like
keyboard-only
users
that
really
like
to
use.
D
Please
provide
feedback
as
well
open
issues
on
it
and
we'll
also
be
asking
feedback
from
an
in
visually
impaired
group
on
this
court,
since
our
user
has
a
disk
or
group
where
she
and
other
people
go
there
and
talk
about
and
provide
feedback
for
apps.
So
that's
a
good
point
to
keep
in
mind
in
the
future.
C
C
A
And
so
that's
a
good
timing
that
you
can
now
obtain
to
the
latest
web
UI.
So
if
anyone
wants
to
test,
you
are
running
ipfs
desktop.
Well,
you
can
install
companion
and
open
web
UI
from
companion
with
that
toggle
toggle,
and
then
you
will
use
the
latest
one
with
referrals
fixes
and
submit
any
feedback
on
that
for
regular
users.
A
D
C
A
A
So
that's
like
a
requires
manual
step.
Even
if
you
toggle
that
toggle
in
companion,
you
will
still
have
to
do
this
manual
configuration
but
yeah
web
UI.
That
ipfs
IO
is
the
latest
version
from
the
master
branch.
So
if
we
want
someone
to
provide
feedback
on
accessibility
fixes
and
in
future,
like
ETA
I,
think
on
the
pinning
services
that
will
be
the
URL
to
use
yeah,
so
it
does
not
require
bumping
with
Goa
Kieffer's
version
or
JCB.
First
version
I.
A
E
F
Actually
I
saw
the
one
question:
I
was
gonna.
Ask
Dietrich
earlier
appetite
for
the
end
of
doing
in
those
discussion
about
the
secure
storage.
Is
anyone
from
the
protocol
apps
or
ipfs
greater
community
in
world
to
kind
of
inform
the
group?
So
they
don't
make
wrong
assumptions
about
how
I
give
a
storage
works.
Yeah.
C
Johnny
French:
is
there
every
one
of
those
meetings
and
he's
he's
usually
pretty
good
at
bringing
up
those
types
of
concerns,
but
but,
like
I
said,
I
think
more,
the
more
people
that
are
present
the
more
it
creates
an
environment
and
atmosphere
where
our
concerns
are
considered
so
I
would
recommend
that
folks
just
hop
in
and
lurk
even
like.
That's
pretty
much
said.
C
I
do
I,
don't
really
say
anything
other
than
introduce
myself
every
now
and
then,
when
they
do
intros
or
whatever,
but
just
being
there
to
ask
the
question
or
to
recognize
when
a
question
or
some
design
choice
might
be
going
down.
The
the
non-network
agnostic,
crap
Network,
specific
networking
exclusive
is
worth
doing
so
and
more
really
like
this
is
really
early
work
to
like.
There
are
some
implementations
of
variations
of
this,
but
nothing
of
the
spec.
C
Yet
so
it's
it's
not
kind
of
high
priority
or
emergency,
but
the
presence
allows
us
to
shape
which
which
direction,
but
so
do
maybe
just
worth
hanging
out
if
you're
interested
in
in
this
stuff.
It's
also
kind
of
interesting
to
to,
if
you
haven't
done,
that
kind
of
stuff
before
sit
in
these
meetings
and
watch
how
the
proceedings
go
and
watch
how
how
they
handle
the
needs
and
interests
of
diverse
groups,
of
stakeholders
who
have
various
competing.
Sometimes
implementations
and
competing
feature
requests
for
things
like
identity.
B
A
Alright
folks,
I
won't
keep
you
around
for
long
much
longer,
mostly
because
I
have
a
such
a
long
block
of
meetings.
I
think
everyone
will
deserve
some
like
freeing
up
some
time.
So
I
ended
this
one
like
executive
decision
to
finish
this
week
sooner
than
later,
and
invite
everyone
same
time
in
two
weeks,
I.