►
Description
On our October 16 call we explored insights gleaned from conversations about the crossover of DWeb and Offline First at Offline Camp Oregon 2019. https://offlinefirst.org/camp
You can subscribe to the Offline Camp publication on Medium for upcoming articles summarizing many of our discussions: https://medium.com/offline-camp
Learn more about the monthly calls of the IPFS Local Offline Collaboration Special Interest Group: https://github.com/ipfs/local-offline-collab/issues?utf8=%E2%9C%93&q=+is%3Aissue+label%3A%22%F0%9F%93%9E+Monthly+Call%22+
A
Hey
folks
welcome
to
the
October
edition
sure.
Oh
yes,
based
on
they've,
suddenly
apparel
in
this
call.
It
is
definitely
October
the
October
edition
of
the
ipfs
local
offline
collaboration,
special
interest
group.
Whatever
this
thing
is
called
where
we
talk
about
offline
first
meets
t
web
kinds
of
topics,
so
we're
happy
to
have
everybody
here
and
we
have
a
couple
of
people
I.
Think
that
haven't
been
on
this
call
before
so.
A
If
we
can
just
go
around
really
quickly,
introduce
yourself
and
kind
of
what
you
work
on
and
why
you're
interested
in
this
space
of
making
stuff
work,
offline
I
can
get
us
started.
I'm
Terry,
I,
currently
work
at
protocol
labs
and
I
am
working
on
a
platform
called
proto
school
which
teaches
decentralized.
A
Web
protocols
tries
to
keep
it
beginner
friendly,
but
I
got
here
by
way
of
the
offline
first
community
and
the
organizer
of
an
event
called
offline
camp,
which
we'll
talk
about
more
in
a
minute
and
I've,
also
done
stuff
with
couch
TV
and
progressive
web
apps.
In
that
the
more
traditional
side
of
cyberspace
I
will
go
this
way
on
my
screen.
Deitrick
you're
up.
B
Hi
everybody,
my
name
is
Dietrich
Ayala
I
work
at
PL,
I'm
working
on
I,
give
fests
in
web
browsers,
mostly
and
interested
in
offline
topics.
Lately
in
that
world,
we've
been
talking
about
things
like
offline
co-hosting
of
Wikipedia
and
some
of
the
things
that
I
added
a
agenda
item
to
talk
about.
After
often.
C
And
light
up,
I
worked
PL
on
IP,
first
in
web
browsers,
maintaining
browser
extension
and
recently
working
on
embedded
Jase
ipfs
in
brave,
without
fail
party
software
and
I
really
interested
in
making
that
side
of
our
technical
stuff
stuck
supporting
like
offline
use.
Cases
which
I
believe
I
will
have
some
notes
later.
A
D
Hello
hi,
hello,
I'm,
Janice
I
am
working
at
UCL
and
I'm.
Also,
an
advisor
on
protocol
labs
I
have
been
working
in
the
area
of
mobile
communications
and
delayed
alarm
that
works
for
almost
decade.
It's
been
very
hot
interest
in
my
list,
so
I'm
really
interested
to
see
what
is
going
on
in
terms
of
developments.
We've
got
a
few
proof-of-concept
prototypes
that
are
working
in
this
kind
of
mode.
A
E
Oh
hi,
everybody
from
the
Oregon
forest
that
I'm
definitely
in
right
now:
I'm
Michael
burns
I
work
at
PL
as
well.
I
came
from
an
open
source,
I'm
kind
of
self
posting
background
and
have
some
interest
in
like
data
sovereignty,
and
so
the
the
offline
first
kind
of
mindset
meshes
with
that
rather
nicely.
So
that's
my
perspective
on
the
world
and.
F
Sure
so
I
work
at
protocol
during
labs,
particularly
on
Google
ipfs,
and
what
I
do
is
I
like
to
think
about
how
to
integrate
that
with
other
systems.
So
like
the
operating
system,
or
maybe
your
service,
or
something
like
that,
and
that
is
something
that
I
do
day
to
day
and
I
like
to
come
to
these
meetings,
because
I
want
to
see
what
other
people
are
doing,
that
kind
of
falls
in
line
with
you
know
some
of
our
values
and
the
goals,
and
things
like
that.
F
A
G
A
Yes,
I
muted.
Thank
you
soon
for
that
hopeful,
pop-up
reminder
when
I
was
noisy
tonight.
Talking
to
all
of
you,
I
was
hope.
There
are
a
couple
other
people
who
I
was
hoping
to
have
shared
there.
There
are
insights
today,
but
they
had
conflicts,
so
we
can
certainly
catch
up
with
them
on
our
next
call.
A
If
they're
able
to
join
us,
then,
but
Michael
and
I
were
both
at
offline
camp
which
took
place
the
last
weekend
of
September
in
the
great
state
of
Oregon,
where
Michael
hails
from
yes,
very
fancy
and
the
event
just
for
folks
who
haven't
heard
about
it
before
is
focused
on
offline
first,
more
broadly
so,
making
technology
that
works
in
bad
network
conditions,
whether
that's
intermittent
or
primarily,
just
offline.
Sometimes
it's
about
syncing
with
couch
TV
and
pouch
TV,
and
sometimes
it's
about
IOT,
and
sometimes
it's
about
mesh
networking
for
disaster
recovery.
A
All
discussion
based,
not
presentations,
so
we
were
nominating
as
a
group
what
we
wanted
to
talk
about
and
breaking
into
smaller
groups,
so
we've
each
attended
some
of
the
sessions
over
the
course
of
our
time
at
camp,
and
then
we
are
working
on
having
folks
who
attended
camp
write
up
some
rays
of
most
of
those
discussion
topics.
So
if
you
follow
our
medium
publication,
which
I
have
the
link
to
in
the
notes,
then
you
can
stay
tuned
and
hopefully
see
a
little
bit
more.
A
That
content
as
time
goes
on
and
if
you
see
a
topic
in
one
of
those
articles
that
I
think
would
be
nice
to
surface
on
a
call
and
I'll
chat
about,
please
feel
free
to
propose
something.
This
is
all
audience
proposed.
Topics
for
these
cause
so
feel
free
to.
Let
me
know,
but
I
thought
we
could
just
kind
of
Michael
if
you
want
to
maybe
start
and
share
some
of
the
conversations
that
you
found
particularly
interesting
or
that
stood
out
to
you-
and
this
doesn't
need
to
be
just
us
sort
of
summarizing.
E
Yeah,
so
just
to
kind
of
emphasize
this
is
my
first
offline
first
camp
and
first
kind
of
introduction
to
that
community.
So
I
kind
of
went
in
white.I,
wide-eyed
and
bushy
tailed
and,
as
is
true
about
that
conferences
like
there
were,
is
a
pretty
small
one,
Theory
schedule.
So
there's
a
couple
dozen
people,
all
of
them
kind
of
view,
their
own
background,
their
own
context
of
getting
here,
all
of
them
crazy,
passionate
about
this
stuff,
and
so
we
we
quite
literally
just
nerd
it
out
for
hours
a
couple
of
really
cool
talks.
E
We
had
do
some
highlights
a
representative
that
works
with
homeless,
trans
troubled
recently
or
frequent
jail
inhabitants.
Like
the
the
rough
edges
of
people
who
are
struggling
in
the
New
York
metro
area
came
and
talked
about
tools.
They
need
the
the
kind
of
workflow,
the
kind
of
the
whole
new
set
of
challenges
of
kind
of.
How
do
you
teach
someone
how
to
do
a
computer
or
use
a
smartphone
app
when
they're
not
necessarily
technically
literate
beyond?
E
You
know
they
know
about
the
internet
or
whatever,
but
they
also
have
like
more
pressing
things
going
on
in
their
lives.
They
don't
have
lots
of
spare
cash;
they
don't
have
necessarily
a
home
or
place
to
keep
like
longings,
don't
necessarily
trust
all
the
Civic
resources
around
them,
etc,
etc.
Those
are
really
incredible.
Talk,
a
good
chunk
of
which
got
notes
and
possibly
recorded
so
I
think
we'll
be
able
to
expand
on
some
of
that
and
I
won't
just
keep
talking
about
it.
These
are
two
topics
that
were
really
really
interesting.
E
If
there's
a
way
of
rewarding
people,
who
can
kind
of
summarize
give
you
some
context
on
the
inner
turmoil
and
and
discussions
of
a
community
without
having
to
dig
into
every
person's
back
and
forth
a
lot
of
that
in
the
context
of
Twitter
and
secure
scuttlebutt
and
then
finally,
the
peer
tech
side,
I
hope
I'm
not
going
too
long
or
too
high
level.
There's
a
great
discussion
on
a
number
of
people
worked
in
the
field
in
rural
and
disconnected
communities,
particularly
in
Africa,
and
so
there
was
just
an
open
design
discussion
of
okay.
E
E
What
do
you
call
it
native
and
all
the
the
scaling
issues
and
all
the
things
that
come
of
that?
But
this
was
a
person
had
that
problem,
they're
gonna,
leave
camp
and
then
go
start
working
on
their
new
prototype
and
they
come
up
with
a
rough
design
of
what
they
wanted
to
do,
and
it
was
an
open
discussion
of
okay
if
you
had
to
go,
build
a
chat
app
for
people
who
potentially
had
feature
phones
in
the
past,
but
possibly
not
have
no
reliable
telecom
connection,
but
possibly
have
an
unreliable
one.
E
What
does
that
look
like?
What
did
that
text?
Deck?
Look
like
how
do
you
deploy
updates?
How
do
you
like
get
discovery
work?
All
of
those
like
ui/ux
things
that
you
can't
just
go
find
on
reddit
or
after
news
and
like
click,
the
link
and
join
might
give
a
valid
email
address
and
you'll
be
signed
up
right
away,
just
a
really
interesting
set
of
technical
challenges
that
open
up
so
many
interesting
doors,
but
that
need
to
be
addressed
kind
of
on
day.
One
right.
E
You
have
to
get
your
design
right
before
you
start
committing
to
sending
people
and
installing
features
and
all
these
things,
because
if
you
have
to
sign
a
new,
build
and
distribute
that
and
it's
all
on
USB
stick
updating
every
single
client
becomes
really
problematic,
etc,
etc.
It
was
a
crazy
good
discussion.
I
could
happily
keep
going,
but
I
suspect
I
I'm
talking
too
long,
so
I
will
stop
there,
but
a
great
high
level
and
like
diverse
range
of
topics.
I
was
really
really
happy.
A
Zoom
again
tells
me
I'll
share
a
couple
of
quick
thoughts,
and
then
we
can
circle
back
and
dive
further
into
any
of
these
topics,
that
people
are
interested
to
hear
a
little
more
about
yeah
after
Michael
and
I
actually
remember
more
about
them.
One
of
the
things
that
was
interesting
to
me
is
the
person
who
Michael
mentioned,
who
came
in
was
talking
about
the
like
the
transient
and
trans
and
troubled
kind
of
folks
in
the
New
York
City
area.
That
person
was
in
a
different
discussion
about.
A
About
developing
world
applications
for
iPhone,
first
and
one
of
them,
it
so
far
at
offline
camp,
most
of
the
people
who
come
to
camp
who
are
working
pretty
much
exclusively
on
applications
that
are
used
in
the
developing
world,
whether
that's
healthcare,
etc
are
folks
who
are
based
in
Europe
or
the
United
States,
and
aren't
really
tied
themselves
to
the
culture
there.
So
there
was
a
lot
of
talk
about
like
how
do
you
understand
the
culture
that
you're
building
for
enough
to
build
the
right
solution
for
them?
A
And
do
you
need
to
be
hiring
UX
professionals
or
user
researchers
who
are
already
there
to
eat
it?
Like
you're,
probably
asking
the
question:
it's
the
wrong
way,
if
you're,
not
a
member
of
that
community,
so
they
were
expressing
that
this
is
like
this
very
specific
cultural
group
within
New
York
and
the
cultural
group
in
the
developing
world
situations
have
that
similar
thing
of
the
people
right
now
who
are
trying
to
build
solutions
for
them?
Aren't
the
people
who
quite
understand
the
problems
thoroughly
and
it's
it's
very
interesting
to
see
how
that
plays
out.
A
I
think
that's
something
we
run
into
a
lot
enough
at
first,
because
for
me,
the
real
life-like
day-to-day
experience
of
wanting
to
stuff
to
work
while
I'm
offline
is
like
oh
I'm
on
the
subway.
It
would
it's
obnoxious
that
my
game
doesn't
work
right
now,
those
kinds
of
things
not
critical,
critical
stuff,
whereas
there
are
a
lot
of
folks
who
are
suffering
from
a
lot
more
infrastructure
issues
where
we
don't
think
about
the
chain
like
I,
don't
think
about.
A
Oh
the
power
went
off
and
now
there's
no
router,
and
then
that
caused
like
this
trickle
down
of
effects
from
an
infrastructure
side.
So
that's
always
one
of
the
things
that's
most
interesting
for
me
to
hear
and
then,
as
I
said,
I
didn't
get
into
a
lot
of
discussions.
I
was
running
around
so
much,
but
one
of
the
ones
that
I
was
in
part
of
was
about
user
experience
of
offline
first,
which,
because
of
the
nature
of
the
group,
turned
out
to
be
more
about
the
user.
Experience
of
see
web
and
I.
A
There's
also
a
life
I
state
that
people
miss
and
it
really
affects
the
way
that
you
need
to
build
an
application
to
use
cache
data
first
before
you
try
to
get
new
stuff.
So
your
device
thinks
it
has
a
useful
connection,
but
it
doesn't
and
it
keeps
trying
to
get
the
new
information.
It
won't
load
cache
data
because
it
thinks
it's
connected,
but
it's
not
giving
you
anything
useful.
So
that's
kind
of
this
scale
that
I'm
used
to
thinking
about
in
terms
of
what
offline
means
and
I'm
used
to
thinking
about
user
patterns.
A
For
like
okay,
your
your
data
is
safe
locally,
but
it
hasn't
been
connected
kind
of
to
the
cloud
yet
or
to
whatever.
The
central
database
is
the
shared
database.
Whatever
I
mean
thinking
about
messages
for
that,
but
in
this
conversation
it
really
became
apparent
that
the
whole
concept
of
like
what
it
means
to
be
offline
or
what
it
means
to
be
online
is
totally
different.
When
you
get
into
peer-to-peer.
Is
it
like
is
having
one
other
peer
with
you
now
your
online?
Is
it
about
a
number
of
peers
that
you
have
available
to?
A
But
I
will
leave
that
to
the
author
of
the
summary
to
explain
better,
because
that
was
only
in
for
part
of
the
conversation,
but
that
was
interesting
that
it
felt
like
a
whole
different,
like
a
whole
different
discussion
about
what
it
means
to
be
on
and
off
line
even
there,
even
though
there
are
certainly
some
similar
implications
for
how
you
have
to
convey
state.
Maybe
it's
different
states
that
exist
that
you
have
to
convey,
so
that
was
one
of
the
topics
that
was
interesting
to
me.
A
D
E
There
are
notes
and
I
will
be
summarizing
from
memory,
but
there
we
did
record
all
the
stuff,
so
we
can
dive
a
little
deeper.
Some
of
the
takeaways
were
Bluetooth
like
doesn't
work
in
a
a
production-ready
like
host-to-host,
all
the
there's,
a
bluetooth,
I
think
words,
no
like
service
discovery
or
some
various
subsets
of
the
Bluetooth
protocol,
which
are
generally
is
acknowledged
by
this
developers.
I
have
not
used
it
firsthand,
just
didn't
work.
E
So
when
we
were
talking
about
designing
a
chat
stack,
it
was
essentially
go
Android
only
and
use
whatever
they
have
like
a
local
peer
discovery,
API,
which
uses
like
audio
tones
outside
of
human
perception
and
then
Bluetooth
and
then
Wi-Fi
direct
as
a
kind
of
cohesive
tech.
Spec
Apple
has
a
similar
thing,
which
is
same
same
but
different
and
incompatible,
and
the
idea
was
just
give
up
pick
one
platform
and
go
all
in
on
that
platform.
E
E
Obviously,
there's
lots
of
CouchDB
love
a
way
of
syncing
states
and
like
once,
once
the
network
side
works,
the
software
side
seems
simple,
but
the
hardware
side
of,
like
we
have
two
handsets
that
we
want
to
share
any
form
of
data
with,
is
like
just
above
broken,
so
the
biggest
demo,
hopefully
I'm,
not
like
spoiling
any
big
reveals
on
our
media
blog.
The
coolest
demo
shows.
A
E
Super
cool
person
gave
a
demo
of
you.
Have
two
phones
like
how
do
you
get
them
to
talk
to
each
other?
If
you
don't
know
their
tech
stack,
you
don't
have
an
app
install
so
with
the
little
local
webpage,
which
uses
full
screen,
2d
barcodes
on
one
screen
and
the
camera
on
the
phone
on
the
other
screen.
Can
you
point
about
each
other?
E
So
the
screen
that's
showing
barcodes
flashes,
maybe
five,
a
second
ten,
a
second
whatever,
and
he
essentially
frames
of
data
which
is
checkable
because
of
the
2d
barcode
and
then
a
camera
that
just
parses
all
those
of
the
library,
and
it
is
shocking
how
reliable
a
cross-platform
that
is
compared
to
the
other
like
service
discovery,
mdns,
yada,
yada,
yada
tech
pieces,
which
like
work
in
demos
or
if
you
have
the
same
platform
app
but
doesn't
work.
If
you
have
an
old
Android
and
a
brand
new
iPhone
or
some
some
mix
and
metric,
that's
it.
A
Yeah
I
can
expand
on
that
a
little
bit
and
I'm
actually
gonna
share
my
screen
as
part
of
this,
just
to
show
you
that
this
resource
exists.
So
as
people
have
written
for
the
offline
can't
medium
publication,
I've
done
some
categorizing
and
some
of
it
would
get
at
the
question
that
you're
asking
so,
for
example,
in
the
left.
So
this
time
then
in
that
past,
but
we
usually
have
a
huge
contingent
of
folks
who
are
really
excited
about
how
it
should
be
in
pouch
TV
for
mobile
sync.
A
So
this
could
be
as
part
of
a
progressive
web
app,
for
example.
So,
certainly
like
that
mobile
sink
thing,
if
you're
dealing
with
a
traditional
web
app
as
opposed
to
decentralized,
that's
a
pattern
that
you'll
find
pretty
well
documented
and
then
the
use
of
service
workers
as
part
of
a
PWA
is
well
documented.
A
crop
like
PWA.
That
term
is
much
better
known
than
offline
first
and
so
progressive
web
apps
and
service
orders
our
pattern
for
traditional
web
apps
that
are
much
easier
to
find
information
on
and
I
think
you
would
on
70
web
stuff.
A
A
It's
really
cool,
let's
see
if
I
can
hunt
that
down,
we've
had
a
lot
of
people
talk
about
developing
world
use
cases
and
there
I
think
infrastructure
is
the
biggest
thing
that
I
in
my
everyday
life
don't
anticipate,
but
the
other
piece
that's
not
exclusively
related
to
this,
but
disaster
recovery,
which
could
be
could
obviously
happen
in
the
States
as
well,
but
I
think
a
lot
of
folks
in
other
places
have
mesh
networking
situations
that
they're
using
that
work
really
well
for
communicating.
You
know
when
we
have
a
big
disaster.
A
The
phone
lines
go
down
just
from
overuse.
It
seems
like
so
there's
that
kind
of
stuff
that
you'll
find
here
so
a
few
different.
There
are
a
few
different
kind
of
common
themes
that
come
out
like
that.
I
do
think
I
feel
like
there
was
an
effort,
or
at
least
the
suggestion
of
an
effort
to
kind
of
codify
a
bit
more
like
what
counts
as
offline.
A
First,
what
you
know
one
of
the
things
I
encountered,
because
the
maybe
because
the
audience
was
saudi
web
heavy
was
like
hanging
too
much
onto
this
word,
but
like
the
combination
of
offline
and
first
in
that
order,
I
think
there's
a
lot
of
misperception
about
with
term
means
and
as
someone
who
has
meant
that
the
team
at
neighborhoodie
that
coined
the
term
I
can
share
just
my
perspective
on
it,
which
mine
kind
of
grows
out
from
a
progressive
web
app
perspective.
But
it's
not
about
being
able
to
use
the
thing
offline.
A
Maybe
one
thing
is:
one
resource
is
more
expensive
to
them
than
another.
So
we've
talked
about
ideas
like:
can
you
set
a
switch
where
they
can
prioritize
bandwidth
or
prioritize
battery,
or
something
like
that?
Is
it
if
you
know
how
those
things
get
drained
by
your
application?
So
we
have
ideas
like
that
from
a
design
perspective,
but
it's
really
like
loading
cached
resources.
First,
to
put
the
user
experience
first
on
those
subsequent
visits
is
the
way
that
I
describe
offline
first
to
people.
A
You
will
find
some
articles
in
the
medium
publication
from
past
events
about
things
like
pirate
boxes.
I,
don't
know
if
you've
seen
that
technology
before
or
like
there
are
some
devices
that
are
meant
to
like
here's,
your
little
Internet
in
a
box
for
your
smaller
community.
So
there's
stuff
like
that.
That's
been
talked
about.
That's
definitely
in
kind
of
deep
web
world
for
like
just
making
my
own
little
thing.
So
does
that
help
at
all?
With
your
question,
yeah
Michael,
you
got
mine.
E
You
just
reminded
me
this
is
probably
even
further
away,
but
a
really
interesting
problem.
I
keep
coming
up,
particularly
in
the
designing
of
chat
app,
but
just
generally
interesting
that
there
is
so
much
overlap
between
like
the
progressive
web
app
and
the
web
development
world
and
offline
first
I
would
have
thought
like
native
mobile
developers,
who
want
to
be
able
to
be
offline.
First
would
be
a
more
natural
matchup,
but
you
get
some
interesting
limitations
like
like
weird
same
origin
policies.
If
you
are
a
web
app
that
is
not
a
mobile
app.
E
Certain
api's
aren't
available
to
you.
You
are
otherwise
limited
and
further
sandbox,
so
you
get
issues
accessing
all
the
cool
device
features
but
kind
of
the
mentality
and
experience
a
lot
of
people
who
are
interested
in
offline.
First,
it's
coming
from
the
web
depth
world
and
so
something
like
I
know,
jazz
or
those
variants
are
more
familiar
than
Kotlin
or
Objective,
C
and
Swift
or
whatever.
So
one
clearly
Firefox
OS
should
have
taken
over
the
world
and
that
would
have
solve
some
problems.
E
A
E
D
E
I'm
going
from
memory
so
I'm
definitely
forgetting
details.
If
it's
a
native,
app
and
side-loaded,
you
can
distribute
that
app
like
a
zip
file
and
I
believe
Android
will
handle
it
and
be
able
to
let
you
install
it
normally
if
it's
a
PWA
again
I
think
it's
the
same
version
issue,
but
I
could
be
confusing
conversations,
but
there
are
handles
separate.
They
I
think
it's
tied
to
an
SSL
cert.
So
do
you
think
you
need
to
make
a
web
request?
E
Thank
you,
a
DNS
record
that
has
that
cell
cert,
with
your
local
phones
not
going
to
a
bit
less.
It's
jailbroken
and
you've
done
a
budget
set
up
so
in
on
PWA
world.
You
need
to
authenticate
against
the
cloud
to
be
able
to
pull
that
down,
whereas
a
native
app
is
a
sign
bundle,
and
so
you
can
move
that
around
exactly
if
it's
a
sign
but.
E
The
majority
of
the
developers,
though,
aren't
fans
of
native
and
they
come
from
a
react
world
and
want
to
use
a
PWA
or
a
web
stack.
They
don't
want
to
learn,
Kotlin
or
swift
or
whatever
that,
and
then
people
with
their
apps
may
be
a
threat
solution
on
a
technical
level.
But
in
terms
of
like
friction,
getting
started,
they
want
to
live
in
a
PW,
a
web
stack
world
and
have
the
benefits
of
data.
B
So
I
mean
we
hit,
we
hit
this
wall
at
Firefox,
OS
and
and
we
had
to
basically
invent
packaged
apps,
which
were
a
zip
file
filled
with
your
web
app
and
your
manifest
and
then
signed
to
be
able
to
get
that
that
side,
load
ability
and
to
be
able
to
get
away
from
that
same
origin
restriction.
The
point
which
the
install
flow
and
update
flow
happens
through
that
dedicated
network
connection
through
a
browser.
B
So
in
order
to
achieve
that,
ap
Kanis
that
so
much
the
developing
world
I'm
go
up
to
the
corner,
get
a
SD
card,
that's
loaded
with
300
apps,
just
dump
it
on
your
phone
they're,
all
pre-signed,
cetera
or
not,
as
often
is
the
case
as
well.
That's
how
a
lot
of
times
the
side
loading
you're,
throwing
away
a
lot
of
those
for
those
privacy
and
security
restrictions
anyway.
They
so
the
authenticity
isn't
necessarily
guaranteed
because
it's
a
binary
that
you've
got
from
a
strange
SD
card
and
put
on
your
device.
B
So
there's
a
little
bit
of
like
the
mirage
of
security
there,
especially
in
the
side
loading
instances,
but
the
use
case.
The
need
of
the
use
case
tends
to
trump
that
in
the
field
so
and
the
from
the
web
standpoint
they've,
the
entire
model
has
hinged
every
bit
of
trust
in
that
same
origin
policy
and
the
ssl
assert
chains,
and
I
a
lot
of
times
to
the
detriment
of
adoption
in
these
types
of
offline
use
cases.
It
storage
is
a
good
example.
B
So
some
of
the
constraints
that
you're
talking
about
around
native
versus
web
are
around
storage.
Even
in
JSI
PFS.
We
do.
We
basically
build
a
local
data
store
inside
of
a
webpage.
That
is
blown
away
completely.
If
you
hit
stores
quotas,
because
we
don't
want
to
ask
for
permission
to
have
more
storage,
so
we've
traded
off
the
persistence
of
that
data
for
a
perceived
user
experience
problem
around
having
to
ask
permission
for
something
from
a
web
page.
A
Remembered
the
thing
I
had
forgotten
about
the
conversation
about
like
when
you
would
use
pwace
versus
native
just
thinking
about
it.
We
were
originally
assuming
that,
when
you're
using
a
native
app,
you
know
there
are
certain
places
in
the
settings
of
your
phone
where
you
can
go
and
see
like
the
data
usage
from
a
specific
app
and
you
have
a
list
of
all
apps
and
we
had
been
assuming
that
maybe
with
a
PWA
which
is
really
running
in
web
browsers
or
even
though
that's
it
in
from
you.
A
But
it
would,
but
it
wouldn't
show
as
its
own
usage.
But
somebody
during
the
session
pulled
it
up
and
did
see
the
things
they
knew
to
be.
Pwa
is
coming
up
in
that
list,
which
was
interesting,
so
I'm
not
positive,
whether
that
is
universal
or
specific
to
the
kind
of
device
that
had
that
it
was
treated
that
way
that.
B
A
E
So
there
have
been
some
tickets
about
this
on
my
PFS
repo,
it's
probably
reasonable
to
say
that
ipfs
is
not
an
offline
first
and
I.
Think
there'd
be
some
advantages
to
trying
to
do
that.
How
that
skirts
or
addresses
the
like
same-origin
PWA
is
a
little
blurry
to
me,
but
it
seems
like
there's
some
advantages
there
of
like
being
able
to
pin
and
resolve
content
you're
an
offline.
Therefore,
ipfs
is
a
open
book
sort
of
things.
E
A
E
The
technical
side
is
not
the
hard
part.
The
UX
is
by
far
the
more
tricky
part
of
like
breaking
while
guard
monopoly
is
having
an
experience
like
secure.
Scuttlebutt
is
a
very
popular
thing
in
some
of
these
conferences.
Onboarding
someone
to
secure
scuttlebutt
is
a
non-trivial
thing
and
there's
about
two
term.
They
need
to
know
about
and,
like
gigs
of
data
needs,
to
be
processed
between
two
users.
E
If,
like
a
more
seasoned
user,
says
hey,
let
me
join
you
to
my
whatever
scuttlebutt
averse
and
do
all
those
things
so
like
the
technical
side
is
not
there,
but
it
is
relatively
close,
like
you
can
make
it
work
with
some
issues
of
api
compatibility
and
stuff.
What
is
not
seemingly,
there
is
the
UX
and
kind
of
terminology
and
design
language
for
what
it
means
to
do
that
stuff
and
then,
like
you,
just
need
to
be
aware
of
all
this,
so
that
the
user
knows.
E
Oh,
it's
failing
because
there's
above
not
it's
failing,
because
things
online
it
isn't
or
whatever.
So
there
there's
too
much
state
that
user
needs
to
maintain
to
be
able
to
have
a
reasonable
experience
using
this
tech
like
it
is
not
part
of
the
phrase
but
like
mom
friendly
or
like
a
tech
illiterate
friendly
though
it
totally
could
be
like
all
the
pieces
are
there,
it
does
not
have
not
has
not
been
built
yet
that
way.
Yeah.
A
One
of
the
things
that
came
up
in
that
UX
discussion
was
on
a
thing:
I
call
the
normal
internet
with
you
when
you're
having
a
computer
problem.
They're,
often
like
the
diagnosis
tools,
they're
like
okay,
try.
This
did
it
work.
If
not,
this
might
be
the
problem
and
I
think
people
were
people
who
are
more
familiar
with
T
web
than
me
or
saying:
okay.
Well,
it
could.
This
could
be
the
thing,
that's
the
problem.
It
would
be
like.
Oh
you,
you
have
93.
A
Whatever
is
something
that
doesn't
mean
anything
to
me
as
someone
who's
new
to
do.
You
have
like
you
need
to
take
it
one
step
past
where
you
think
it
isn't
like
what
is
the
implication
of
that?
What
should
I
do
next,
because
of
whatever
this
random
statistic?
Is
that
you're
presenting
to
me
that
I
don't
understand
so?
Is
it
that
I
need
to
okay?
I?
A
A
I,
don't
know
how
to
tell
who's
like
whose
fault
it
is
that
I
can't
get
the
thing
and
so
I
think
there's
a
lot
of
just
one,
a
second
like
vocabulary,
not
just
for
offline
first
applications
for
ipfs
or
do
you
web
stuff.
But
for
me
the
vocabulary
for
D
web
stuff
in
general
is
very
inaccessible,
very
hard,
very
unfamiliar.
It
needs
a
lot
of
work
to
come
down
to
like
normal
human
I,
don't
mean
English
as
opposed
to
Spanish
I
mean
English
as
opposed
to
tech
jargon.
It's
that's
one
of
the
biggest
things.
A
B
We've
been
having
this
same
conversation
in
how
we
kind
of
packaged
up
browser
related
features
for
ipfs,
like
there's
a
lot
of
things
that
we
can
do
that
align
with
with
that
that
meet
a
lot
of
really
common
offline
or
partial
offline.
In
these
cases,
but
and
some
of
the
stuff
that
Lytle
has
in
the
agenda,
you
know
as
a
reflection
of
these
conversations
that
we've
been
having
and
from
it
from
UX
standpoint,
like
the
work
that
we've
done
so
far,
and
things
like
that
people,
companion
or
desktop,
are
really
orient
around.
B
I
confess
as
a
technology
and
people
who
are
I
profess
core
contributors
or
technical
users,
but
not
around
the
things
that
people
need
to
do
if
they
want
to
save,
share,
publish,
read
those
core
use
cases.
So
when
people
are
looking
for
something
that
I
want
to
be
able
to
save
my
competing
pages
offline
and
share
them,
they
don't
Google
NTFS.
A
A
So
if
anyone
is
interested
in
helping
to
participate,
that
would
be
great
and
one
need
that
the
community
has
in
particular
right
now
is
we
have
a
lot
of
authors
assigned
for
these
articles
on
medium
and
we
don't
have
enough
people
who
have
time
to
cat
herd
and
follow
up
with
those
folks.
So
that's
one
thing:
anybody
watching
this
if
you'd
like
to
volunteer
to
be
an
official
cat
herder
for
the
medium
publication,
please
let
me
know
and
I
could
give
you
all
the
training
and
some
experience.
A
So
subscribing
to
our
medium
publication
would
be
another
way
to
ensure
that
you've
got
notified
when
this
happen,
but
you
personally
dietrich,
I
will
make
sure
you
know
when
yeah
and
obviously
I
will
tell
people
in
the
school
district.
Do
you
want
to
jump
over
to
your
topic
quickly?
I
don't
know
what
the
length
of
your
entitles
topics
is.
I'm.
B
It's
an
organization
that
does
a
lot
of
work
in
a
lot
of
different
countries
surround
making
sure
that
people
have
access
to
middle
care
and
they
work
with
governments
and
NGOs,
and
one
of
the
projects
that
they
have
right
now
is
a
project
called
meta
capped,
which
is
I
initially
just
started
as
a
set
of
printable
PDFs
for
documenting
when
there's
an
intersection
between
doctors
in
the
field
that
there
needs
to
be
an
intersection
with
a
criminal
justice
system.
So
a
lot
of
times.
B
These
are
things
like
sexual
abuses
and
things
like
that
that
are
happening.
The
two
countries,
where
they've
kind
of
piloted
this
work,
if
it
was
gonna,
be
thei
Opia
and
they
have
now
developing
and
funding
for
an
application
to
be
able
to
work,
to
go
to
record
evidence
from
that
needed
to
get
from
doctors
into
the
criminal
justice
system
in
these
countries,
and
they
are
very
unique
offline
requirements.
For
this
needs
to
be
things
like
a
lot
of
times.
It
needs
to
be
encrypted
a
lot
of
times.
B
It
needs
to
be
able
to
be
passed
between
multiple
messengers
without
them
be
able
to
see
the
contents
necessarily
and
it
almost
they
almost
all
of
the
time.
These
are
situations
where
internet
Access's
it
is
not
available
in
large
parts
of
an
average
and
regular
day.
Things
like
also
in
infrastructural
requirements
like
we're
talking
about
earlier
things
like
not
know
power
and
things
like
this
very
slow,
slow
connections.
If
any
connection,
so
it's
a
very
interesting
set
of
requirements
and
I
thought
I.
B
C
C
So
when
you
install
I,
will
use
brave
as
a
good
example,
because
in
if
we
have
embedded
jazz
I'd,
give
us
note,
you
don't
need
to
install
and
if
a
party
software,
so
the
idea
is
when
you
install
a
profess
companion,
you
will
have
web
UI
pre
cached,
so
it
will
work
in
offline
environment
and
also
making
the
initial
onboarding
experience
much
more
pleasant,
because
it's
already
there
does
not
need
to
be
loaded.
It's
22
Meg's,
so
it's
the
first
initial
load
may
take
time,
especially
if
your
network
is
not
too
fast.
C
The
second
item
is
ipfs
companion
right
now,
it's
able
to
recover
from
some
broken
links,
but
there's
contributed
PR,
which
is
very
close
to
landing
in
the
next
companion.
Basically,
the
idea
is
that
companion
is
already
identifying
ipfs
resources.
Like
content
addressed
things
on
websites,
people
run
their
own
public
gateways.
Some
people
just
use
ipfs
content
addressing
in
paths
on
their
website.
The
problem
is
what
happens
if
the
Gateway
goes
offline
or
DNS
gets
censored
or
stuff
like
that.
C
Browser
like
HTTP
request,
will
just
fail,
even
though
the
content
is
still
available
on
ipfs
network,
so
companion
pretty
soon
we'll
be
able
to
recover
from
those
specific
error
codes
when
DNS
lookup
fails
when
like
there's
a
server
error
or
like
content,
is
blocked
for
legal
reasons
there.
It
will
like
seamlessly
with
time
to
enable
it
by
default.
It
will
automatically
try
to
recover
and
open
it
from
the
public
gateway,
so
it
should
work
with
our
gateway.
C
If
user,
that
is
not
running
click
on
node,
and
if
it's,
if
they
are
running,
then
it
will
automatically
get
redirected
to
local
node.
So
you
will
not
get
blocked.
You
will
not
get
censored
and
even
if,
like
the
service
called
curse
of
anthos
link
will
no
longer
be
dead,
so
that's
coming
soon.
I
probably
will
demo
those
things
next
time
we
see
each
other
I'll,
miss
Cole
and
finally,
I
wanted
just
to
mention,
there's
an
experiment
in
MF
s
based
co-hosting.
C
So
generally,
the
idea
is
to
leverage
existing
KPIs,
namely
EMFs,
existing
IP,
FS
api's,
and
try
to
model
a
very
simple
protocol
scheme
and
more
like
a
set
of
conventions
for
co-hosting
stuff
without
any
third
party
applications
or
new
api's,
and
specifically
to
the
discussion
we
had
before
on
this
call
coming
from
the
use
case.
So
if
you
want
to
house
Wikipedia
Wikipedia
is
like
six
over
six
hundred
gigabytes,
English
Inc
with
Wikipedia
on
part
when
you
unpacked
it
without
videos,
just
text
and
pictures.
It's
like
650
gigabytes
in
size.
C
So
not
everyone
is
able
to
like
house
entire
thing
and
if
people
visit
a
website
with
like
IP,
first
companion
or
other
browser
which
supports
distributed
protocols
and
how
that
person
can
opt
in
to
contributing
to
co-hosting
Wikipedia
or
to
co-hosting
a
part
of
Wikipedia.
So
there's
a
discussion
about
concept
called
I
called,
like
lazy
co-hosting.
So
the
idea
is
that
you
could
mark
a
website.
Let's
say
Wikipedia
to
be
co-hosted
and
there
would
be
two
modes
like
one
would
be
like
the
full
copy.
C
But
that
requires
a
lot
of
space
and
time
to
fetch
it
and
the
lazy
mode
is.
You
are
contributing
you're
co-hosting,
but
only
the
parts
you
already
visited.
It's
sort
of
already
built
into
ipfs.
The
only
missing
piece
is
keeping
those
blocks
around
like
pinning
them
or
like
implicitly,
pinning
them
and
that's
what
co-hosting
spec
is
basically
providing
a
means
of
formalizing
a
way
to
keep
those
blocks
around
for
the
websites
that
you
care
and
we
have
like
a
very
early
draft
of
the
spec
and
eventually
we
want
to
experiment
in
what
like
internment.
C
C
Sort
of
like
related
link,
I
dropped
related
to
discussions
about
like
audio
based
signaling,
and
things
like
that.
We
have
on
in
the
lipid
2p
notes,
repo,
there's
a
issue
where
we're
sort
of
like
a
drop
and
maybe
take
a
closer
look
at
some
alternative
discovery
and
transport
protocols.
I'm
interesting,
especially
for
those
things
that
could
be
implemented
or
work
in
web
browsers.
So
vasco
looked
at
the
well
web.
C
Pluto
spec
is
right
now
and
like
we've
sort
of
identified
missing
pieces,
we
also
have
like
link
to
audio
signaling
for
WebRTC
I,
believe
we
someone
mentioned
that
the
few
it's
back
on
this
call.
So
if
you
have
any
new
links
over
time,
just
that's
the
place
when
you
can
drop,
drop
them
and
kinda
related
to
pwace
is
sign
HTTP
exchanges
and
handle
HTTP
exchanges,
so
sign.
Http
exchanges
are
for
a
specifically
like
sign,
signing
a
specific
request
and
make
it
eat
the
couples
from
origin.
C
C
Chrome
is
planning
to
implement
bundles
HTTP
exchanges,
which
is
I,
believe
very
relevant
to
pwace,
so
you
could
like
bundle
all
the
resources
rub
them
in
into
this
bundle
format,
and
when
you
load
it
in
like
mobile,
chrome
or
a
desktop
Chrome,
it
will
look
like
it
came
over
HTTP
from
what
server,
but
in
fact
that,
instead
of
like
HTTP
over
TLS,
the
signature
was
a
part
of
the
bundle
and
there's
an
issue.
I
dropped
in
the
notes
about
like
oh,
where
we
are
so
far
with
experimentation
around
signed
and
bundled
HTTP
exchanges.
C
B
A
Maybe
next
time,
so
speaking
of
next
time,
I
looked
at
the
calendar
today,
we
usually
have
this
on
the
third
Wednesday
of
the
month
and
I
realize
that
on
the
third
Wednesday
of
November,
no
one
from
protocol
out
so
to
be
able
to
attend,
including
me
so
I
would
propose.
Unless
people
know
they
have
conflicts
that
we
do
this
one
on
November
13th
instead
of
November
20th
next
month,
and
that
keeps
us
away
from
both
that
conflict
and
day
before
Thanksgiving.
A
So
if
I
don't
hear
any
strong
objections,
I
will
scoop
that
on
the
calendar
and
as
always,
I
I
am
create
an
issue
for
the
meeting
and
feel
free
to
draw
up
ideas
and
therefore
topics
you
want
to
talk
about
next
time.
Ladle.
If
there's
any
of
this,
that
you
want
to
go
into
in
more
detail,
we'd
be
happy
to
have
you
do
that
and
then
we
got
a
really
cool
note
from
Nicolas
pace,
who's
doing
some
cool
stuff
right
now,
which
is
like
he's
not
here
today.
A
So
you
will
see
what
we
have
in
store
for
us
next
time,
but
thank
you,
everyone
for
coming
and
keep
an
eye
on
that
medium
publication.
We
hope
to
have
more
on
this
session
summaries
coming
out
soon.
Also
people
gave
some
quick
fashion
talks.
Some
of
them
had
to
do
with
the
or
offline
first
stuff,
and
some
of
them
had
to
do
it
completely
random
things,
but
they're
all
really
cool,
and
some
of
the
folks
would
like
to
share
those
videos
with
you
and
some
articles
around
them.