►
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
so
all
right,
I'm,
not
Nam,
no
I'm
on
for
or
fortunately
he
has.
B
Better
things
to
do
as
he
had
his
firstborn
just
born.
A
So
I'll
be
taking
over
one
of
his
presentations
and
I
think
Ian
well
take
over
the
other.
There.
A
Breakout
session
with
me,
but
yeah
yeah,
so
this
is
about
the
prefetch
processing
model
and
trying
to
look
at
unbrink
it
or.
A
Is
currently
used
both
to
pre-fetch
HTML
for
documents
the
user
may
soon
navigate
to
and
at
the
same
time
also
used
for
low
priority
resources.
A
A
A
Page
or
something
along
those
lines,
yeah
so
yeah,
there
are
a
lot
of
differences
in
the
way
that
different
vendors,
Implement
prefetch
and
a
lot
of
differences
like
when
trying
to
test
the
current
behavior
for
chromia
prefetch
works
well
for
both
same
site
navigations
as
well
as
for
sub
resources.
So
it
works
well
for
both
the
document.
B
Cases
along
the
sub
resource
use
case
for
cross-site
documents.
There
is
yeah
a
special
item,
documented
code
path
or
a
hack
as
I
think
more
shortly,
where,
as
equal
document
puts.
B
Origins
partition
and
the
resources
origin
partition,
rather
than
in
the
site
that
is
initiating
that,
like
a
rather
than
the
partition
of
the
site,
that's
associated.
B
Way
to
solve
the
cross
origin
same
document,
not
the
cross,
origin
navigation,
but.
A
A
B
B
That
is
used
for,
like
it
essentially
is
a
signaling,
a
sign,
Exchange
prefetch.
B
Yeah,
a
special
prioritization
for
prefetches
after
so
they're,
only
being
prefetched
after
the
load
event
and
being
done
one
at
a
time
to
make
sure
that
they're
not
logging
the
backwards.
There
are
no
specific
cash
semantics.
There
is
a
Moz
prefetch
header
that
enables
sites
to
know
that
this
resource.
B
Header
that,
like
the
purpose
of
prefetch
header,
that's
being
used
and
the
accept
header
is
just
Stars
whatever
kind
of
resource.
B
Flag,
it's
a
simple
low
priority
fetch
without
any
particular
semantics.
A
The
issues
yeah
there
are
essentially
yeah.
This
five-minute
rule.
It's
it's
getting
in
the
way
of
catching
semantics.
A
B
Create
correctness
problems
in
some
cases,
they're
like
major
differences
between
the
same
side
of
the
behaviors,
and
that
is
not
great.
There
is
like
a
purpose
header
versus
mozilla's,
prefet,
signaling
header
versus
nothing
in
webkin,
and
that.
B
Accept
headers
can
break
caching
when
the
resources
sent
back
to
the
very
accept.
A
A
Put
up,
try
and
Define
a
reasonable
task
forward
that
is
not
compatible
with
with
any
one.
Implementation
is
currently
shipping,
but
potentially
something
we
can
agree
on
to
agree
to
converge
on
which
essentially
doesn't
give
any
time-based
cash
semantics
or
like
special
treatment.
B
The
signal
to
the
servers
that
prefetch
is
being
done
and
ignore
very
accept
when
the
prefetch
is
consumed
in
the
browser.
B
And
as
attribute
other
than
the
answer
document
thing,
the
developers
don't
currently
Define
what
the
destination
would
be.
So
the
accept
headers
for
prefetch
will
not
match
the
destinations
except
headers,
unless
you
know
we're
very
lucky,
but
there's
a
effort.
Images,
for
example
the
simplify
formats
and
but
basically
yeah
we'll
have
different,
accept
headers
for
Destination
and
ignoring
that
in
the
browser's.
Cache
will
enable
us
to
also
supports
a
very
accept
resources
without
any
particular
negative
side
effects.
A
Yeah
that
let's
discuss
this
at
the
end
but
yeah
we're
actually
at
the
end.