►
From YouTube: WebPerfWG triage call August 1st 2019
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).
B
D
D
So
basically,
the
problem
is
that
the
back
forward
says
this
is
true.
This
is
the
navigation
type.
Whenever
you
go
through
a
history
traversal,
but
as
it
turns
out,
every
navigation
will
go
through
traverse
the
history
in
the
spec,
and
so
one
of
the
things
that
we
noted
was
it's
possible
that
the
definition
meant
traverse
the
history
through
a
Delta,
but
I
think
that
that's
not
the
only
issue.
Obviously
here
it
was
pointed
out.
B
D
C
B
C
B
B
A
D
So
the
context
for
this
is,
we
have
Firefox,
implements
the
back
back
forward
cache
and
so
which
operates
in
the
way
that
the
back
forward,
cache
sort
of
it
can
be
specified
in
in
the
standards.
And
so
one
of
the
things
that
we
want
to
do
is
make
sure
that
we're
giving
the
right
value
back
for
navigation
timing
specs
when
we're
retrieving
the
pager
back
forward,
cache
that.
D
What's
interesting
is
we
we
looked
at
that
because
we
put
in
a
test
to
specifically
just
our
implementation
of
this
and
I
put
it
in
our
sites
in
our
browser,
specific
stuff,
but
Boris
pointed
out,
oh
well,
why
don't
we
have?
You
know?
Why?
Is
it
this?
A
web
platform
tests
and
I
thought?
Well
that's
odd,
because
we're
Pat
singing
the
web
platform
tests
and
had
this
bug.
A
D
E
B
B
A
B
B
B
B
A
A
So
this
is
an
issue
that
Nikolas
raised
well,
while
trying
to
implement
that
bit
on
the
chromium
side
of
things,
and
we
need
to
like
yeah.
We
need
to
properly
define
it
before
implementing
add
tests
and
that
and
see
if
the
current
implementations
are
passing.
Those
tests
I
believe
that
right
now
only
Safari
passes
the
tests,
which
are
less
rigorous
than
that,
so
it
will
probably
require
some
at
least
some
changes
once
we
will
define
it
and
like
in
Firefox
and
maybe
Safari.
A
B
A
A
A
207
yeah,
that
is
something
that,
like
I,
basically
have
multiple
batches
in
different
places
that
contain
tests
spec
language
as
well
as
the
not
the
implementation
but
close
enough.
I
just
need
to
wrap
all
these
together
and
into
you
know
something
that
means
shippable,
so
yeah.
This
is
already
assigned
to
me.
I
see
and
okay,
you
just
I'm
very
active
the
other
day
and,
as
we
speak
sorry
all
your
yeah.
A
C
C
B
B
E
E
So
I
commented
earlier
there's
a
in
a
table:
Specter
gin,
generic
tasks,
sources
fold
them
when
is
the
domani
patient
task
force
and
add-ons
user
interaction,
task
force
and
other
ones,
network
task
force
and
the
last
one
is
history,
traversal
task
force
tasks,
so
I
think
what
we
probably
have
to
do
is
for
each
oh
I
could,
basically,
you
probably
have
to
tie
the
deadlines
to
the
concept.
Basically,
like
the
text
could
say
something
like
if
there
was
new
tasks
scheduled
in
in
these
task
sources,
them
their
lines
at
pizzelle.
A
A
A
B
E
Well,
I
think
the
regatta
so
who
makes
PR
or
no
I
think
the
one
thing
we
need
to
do
is
somebody
who
has
already
implement
this
API
to
see
what
people
pose
make
sense
or
no
right,
like
I,
think
one
before
writing
in
a
changing
spec.
We
need
to
make
sure
whatever
actions
we
make
is
actually
compatible
the
existing
implementations.
Otherwise,
you
know
who
knows
like
it.
A
Change
would
make
sense.
Yeah
there's
around
78
I
can
try
to
craft
something
along
the
line
like
craft,
the
language
that
the
Rio
skate
proposed
here,
see
if
it
fits
into
the
current
PR
and
get
the
people
who
implemented
our
request.
Idol
call
back
in
Chrome,
which
is
Ross,
was
also
on
the
issue.
Get
them
to.
You
know
give
this
a
thumbs
up
from
our
perspective.
I
don't
know
if
Firefox
like
who
would
be
the
right
Mozilla
person
to
do
the
same.
A
D
Yeah
I
was
shaking
my
head.
Sorry,
I
I,
just
message
to
Benjamin
asked
him
the
same
like
who
do
we
think
would
do
this,
but
honestly
I
don't
know
so
we,
but
we
could
take
an
action
to
figure
out
who
would
be
responsible
for
that
and
make
sure
that
we
do
this
because
I
think
Ryan's
case
approaching
is
a
very
good
one
like
we
need
to
make
sure
it
makes
sense.
The
vehicle
measures.
E
B
E
A
B
C
B
C
A
E
E
A
A
Anyway,
there
are
a
bunch
of
tests
where
safaris,
failing
and
the
other
browsers
are
not
so
maybe
this
is
a
progression
of
one
of
them.
Let's
say
like
real
skate
when
you,
let's
make
sure
that
if
this
turns
up
to
be
something
that
is
not
WPT
tested,
such
a
test
will
be
added
as
part
of
the
fix.
Does
that
make
sense.
E
A
B
A
B
Great
and
so
I
think
HR
time
we
already
did
publish
this
is
just
an
open
issue
that
came
in
a
trip
times
in
great
shape,
and
so
we've
just
cruised
through
the
bottom
of
the
agenda,
just
on
open
specs
and
we've
got
25
minutes
long.
So
do
you
want
to
crack
another
spec
you
off
I
know
you
mentioned
long
task,
I!
Think
I'm
in
good
visibility.
B
B
C
B
B
A
A
C
A
B
B
B
A
A
B
F
B
F
B
F
F
B
F
F
B
B
B
B
F
F
G
A
A
F
A
B
A
B
E
F
F
A
A
A
In
Section
four
processing
model
user
agent
implementing
the
long
tasks
API
would
need
to
include
long
tests
in
supported
entry
types
or
window
contacts.
This
allows
developers
to
the
tech
support
for
long
tasks.
That
was
pretty
like
that.
Before
that
there
was
a
registration
stack
and
this
note
removed
it.
So
it's
well
dove
like
there
is
a
note
saying
that
it
should
mean
window
contacts.
There's
the
registry
that
actually.
C
A
B
C
F
E
D
B
F
B
There's
not
one
yes,.
B
C
I
think
it
would
be
great
to
hear
from
Nick
so
there's
another
issue
which
is
26
right
below
it,
which
is
performance
online
support
which
basically
highlights
the
exact
same
thing.
So
what
what
Nick
is,
after
there
is
buffered,
it
would
be
great
to
I,
don't
think
we
have
Nick
on
the
call
today,
but
maybe
something
we
could
revisit
with
him
just
to
get
feedback
from
folks
that
have
actually
deployed
this.
B
A
C
B
B
All
right
yeah,
we
have
a
thumbs
up
rios
okay.
Did
you
say?
Yes,
it
is
you
talking
about
all
of
us,
8th,
August,
8th,
yes,
okay,
so
we've
got
a
yessiree,
yo,
oven,
Elya,
you
guys
are
houses
yeah,
okay,
so
Tim
can't
make
it,
but
I
think
we
have
quorum
and
move
forward
with
that
for
poor
router
crew,
as
well
as
the
others.