►
Description
Browsers & The Web Platform - presented by Dietrich Ayala at IPFS þing 2022 - Roadmapping Next Steps - https://2022.ipfs-thing.io
A
A
Light
I'll
give
us
an
update
on
where
we
are
with
HTTP,
generally
compatibility
and
support.
What
we
can
do
with
new
Gateway
features
that
have
recently
shipped
move
talk
to
us,
and
this
is
kind
of
interesting
ways
of
thinking
about
how
to
operate
on
ipfs
data
from
web
platform
apis.
How
do
you
do
it
from
within
the
web
page
as
opposed
to
fighting?
A
What's
around
the
web
page
for
me
to
talk
to
us
about
capiloon
and
their
approaches
towards
integrating
ipfs
in
a
mobile
OS
and
one
of
the
highlights
there
was
end
user
interface
for
ucan-based
permissions
at
the
operating
system
level.
So
watch
that
talk
for
sure
David
showed
us
what
we're
going
to
do
with
manifest
V3.
A
Now
that
it's
reducing
our
capability
space,
that
we
have
in
ibps
extensions
and
what
a
road
forward
might
look
like
for
ipfs,
based
extensions
to
be
able
to
flower
and
make
a
thousand
of
them
bloom
for
specific,
vertical
content
use
cases
like
maybe
Wikipedia
or
misinformation,
be
able
to
save
things
from
Twitter
leading
right
into
ilya's
talk
about
web
recorder
and
he
demoed
live
for
us,
basically
like
an
automated
archiving
of
browsing
Twitter
to
ipfs
in
real
time,
which
was
pretty
amazing.
A
The
talk
from
Martin
kind
of
gave
us
a
view
into
what's
coming
in
the
near
future
for
browser
connectivity.
So
a
lot
of
what
we
fight
with
in
the
browser
is
the
fact
that
we
don't
have
the
transports
available
that
we
need
for
a
folded,
PDP
connectivity
and
implementation.
There
are
some
interesting
opportunities
there
and
those
will
be
available
as
pretty
soon
we
had
two
bonus
talks.
Fleek
showed
up
and
gave
us
preview
into
what
they're
going
to
be
shipping
later
this
year,
which
is
interestingly
aligned
with
a
lot
of
stuff.
A
We've
been
talking
about
this
week
and
then
Daniel
gave
us
the
ipfest
201
materials
talk
about
how
we
take
that
to
the
next
level,
the
number
of
problems
and
frictions
that
none
of
these
are
going
to
be
super
surprising.
We
really
do
need
to
fix
gateways
that
are
still
not
working
for
some
basic
use
cases.
So
how
do
we
get
folks
like
Ilia
and
Febreze
in
the
same
room
with
people
that
are
Gateway
operators
to
make
sure
that
they
can
actually
just
pull
files
through
uploading?
A
We
try
to
identify
that
list
of
of
operations
like
basic
high
level
flows
of
how
you
use
ipfs
in
these
different
contexts.
From
that
came,
some
learnings,
like
you
know,
really,
the
like
one
of
the
biggest
takeaways
is
that
at
the
most
basic
level,
oftentimes,
these
transport
constraints
shouldn't
have
to
be
in
our
way
like
it's
you're
it
starving
person
in
a
desert
made
of
marshmallows
there's
so
much
bandwidth
and
and
connectivity
in
places
where
we
can't
get
things
to
work.
A
So
where
are
we
fighting
against
the
transports
as
opposed
to
kind
of
either
working
with
them
or
just
reimagining,
where
those
like
architecture,
boundaries
are,
and
one
of
the
things
that
you
know
we
saw
often
was
was
we
are
able
to
get
Beyond
those
constraints
and
get
things
to
work
when
we
stopped
trying
to
fight
the
platform
and
when
we
really
thought
about
reimagining,
where
where
ipfs
needs
to
live
in
the
overall
architecture?
A
So
in
browsers,
maybe
it's
not
inside
web
content,
and
maybe
it's
just
outside
of
web
content
in
new
and
interesting
ways
that
we
haven't
really
thought
about
before.
So
a
couple
of
short-terms
actions
came
out
of
this
spec
reviews
of
the
new
connectivity
stuff
for
Lupita
b
super
important
and
that's
something
that
you
can
contribute
to
now.
A
Getting
web
recorder
and
copyland
these
people
that
are
building
on
top
of
ipfsnar
gateways
in
the
same
room
with
the
Gateway
operators,
so
they
can
debug.
Why,
like
you,
should
be
able
to
fetch
a
650
Meg
file
through
the
Gateway?
It
shouldn't
really
be
that
much
of
a
problem,
but
it's
not
working.
A
Why
immediately
provide
feedback
and
and
summarize
the
requirements
that
we
have
in
the
web
platform
space
for
that
minimum
viable
ipfs
specification,
a
proof
of
concept,
implementation,
that's
already
kind
of
been
done
by
the
DAC
house,
folks
of
a
client-side
racing
Gateway.
So
right
now
the
racing
Gateway
implementation
is
on
a
service
worker
somewhere
behind
a
domain
which
may
not
work
for
various
use
cases
that
we
have
in
local,
first
and
local
first
applications
and
things
like
that.
A
So
that
would
also
be
something
that
would
be
useful
for
a
number
of
these
different
projects
and
then,
finally,
David
and
Ilya,
using
like
the
web
recorder,
use
case
to
really
push
on
what
an
ipfs
extension
template
an
example,
ibfs
extension,
how
that
would
function
and
how
it
would
work
some
of
the
Milestones
that
came
out
some
of
the
specification
stuff
and
really
in
the
spirit
of
you'll
notice.
A
Here,
a
lot
of
draft
specification
specification,
driven
development
array
using
that
process
to
be
able
to
Define
how
ipfs
should
work
together
in
these
spaces
in
a
way
that
is
not
implementation,
specific
is
language,
agnostic,
transport,
agnostic
and
can
really
yield
more
uniform
developer
surfaces
across
the
different
places
that
we
want.
These
things
to
work
and
really
pushing
out
a
little
farther
like
focusing
earlier
on
files,
which
is
the
most
common
use
case
and
then
pushing
out
more
complex
data
operating
scenarios
a
little
farther
out
into
the
future.