►
From YouTube: IETF-TOOLS-20230912-1800
Description
TOOLS interim meeting session
2023/09/12 1800
A
A
Of
people
joining
at
the
moment
as
you
join,
if
you
have
not
already
shared
video
and
are
capable
of
doing
so,
please
do
we'd
like
for
this
to
be
a
as
close
to
around
the
table
meeting
as
we
can.
A
Please
scan
at
least
the
outline
of
the
notes.
Remember
you
can
open
the
Outline
by
clicking
the
expand
all
in
the
right
hand,
View.
A
A
We
have
a
pretty
full
agenda
for
today.
Does
anybody
have
any
changes
that
they'd
like
to
make
to
the
agenda
that
they
don't
feel
comfortable
making
by
just
editing
the
notes
document.
A
A
Kept
his
work
in
Python
2,
the
statement
work
that
he's
working
under
has
a
supporting
the
project
to
Python
3,
but
it
looks
like
we
have
a
good
chance
of
making
an
interim
release
earlier
on
just
making
some
modifications
to
the
python
2
version,
so
we
can
deploy
it
in
the
dual
chain
as
it
currently
exists.
B
The
notepad
is
empty.
Is
that
intentional?
Normally
we
have
notes
in
there
say
again,
the
the
notepad
is
empty.
Oh.
A
A
Liz,
if
it's
convenient
or
somebody
else
on
the
call,
could
you
go
to
the
notes
page,
that's
linked
off
of
the
data
tracker
agenda
and
just
put
in
a
go
see
the
real
notes
over
here
message.
Please.
A
C
A
Serve
as
secondary
name
servers
for
us,
but
we
ran
into
trouble
when
we
put
those
name
server
records
into
DNS.
We
didn't
have
a
way
to
get
those
signature
over
those
NS
records.nt.org.
A
We
have
an
open
ticket
with
cloudflare,
where
they're
looking
at
what
they
could
do
differently.
They've
offered
us
a
solution
that
would
work
with
ostensibly
would
work
with
what
they
offer
right
now,
where
identity,
digital
would
sign
their
name
server
records
themselves,
and
we
would
pass
this
stuff
into
the
dot
org
domain
by
way
of
cloudflare
we're.
We've
asked
identity
digital
if
they'd
be
willing
to
do
that,
and
waiting
to
hear
back
for
the
time
being,
we've
removed
the
secondary
name
servers
from
DNS.
A
A
D
Sure
so,
at
this
point,
all
of
the
pieces
that
we're
moving
into
the
data
tractor
have
been
moved.
We've
got
the
appeals,
the
ID
statements,
the
minutes
and
agendas.
So
that's
all
done.
The
other
thing
tools
wise
that
we're
waiting
for
is
the
RSS
feed
of
IAB
tagged
blog
post
in
the
iatf
blog
I.
D
Think
casara
was
working
on
and
I
meant
to
follow
up
with
you
on
that
last
week
and
I
did
not
get
a
chance,
but
that's
that's
I,
think
the
biggest
thing
and
then
to
the
side
of
that
Liz
and
I
met
with
miria
about
a
week
and
a
half
ago
to
go
through
the
site
as
the
the
actual
temporary.ib.org
site
talk
about
the
contents,
she
wanted
some
changes
to
the
information
architecture,
so
we've
been
moving
things
around.
D
Based
on
that
conversation,
Liz
has
been
working
on
content
updates,
based
on
those
conversations,
and
most
of
those
changes
are
done.
So
we
are
really
really
close
on.
This
I
feel,
like
I've,
said
that
a
couple
of
times
but
I
mean
it
this
time.
A
Because
I
think
you
I've
mentioned
already
further
down
in
the
notes
that
you
have
the
feed
ready
to
go.
We
just
need
a
new
deployment.
C
C
E
A
Yeah,
it's
the
it's
actually
how
the
websites
are
going
to
talk
to
each
other.
The
blog
is
going
to
live
entirely
on
the
iatf
website,
with
the
AI
IAB
website
will
show
elements
out
of
the
blog.
So
it's
the
IAB
website,
that
is
the
consumer
of
the
RSS
feed.
A
So
the
last
thing
I'd
like
to
add
here
is
that
the
next
steps
for
moving
things
into
the
data
tracker,
since
we
moved
the
IB
agendas,
statements
and
appeals,
we've
got
the
framework
for
moving
those
kinds
of
artifacts
into
the
data.
Tracker
and
I
will
be
moving
the
iesg
variants
of
those
things
in
the
next
few
weeks.
A
I
believe
when
I
finish
that
move,
we
will
have
taking
care
of
the
last
blocking
items
for
shutting
down
www.6.ietf.org.
A
Someone
move
on
over
the
last
few
weeks.
We
have
tried
a
different
way
for
deploying
the
irtf
website.
We
have
a
proposal
in
front
of
Colin
right
now,
he's
reviewing
where
the
website
is
being
served
out
of
blob
store
in
Azure,
just
because
Azure
was
had
the
the
feature
set
that
we
needed
to
be
able
to
make
it
easy
for
Colin
to
deploy
new
versions
of
the
website
and
have
good
previews
of
the
work
in
progress
from
the
repository
that
he
already
had.
A
So
when
he
finishes
this
review,
unless
he
finds
a
showstopper
and
and
his
initial
thoughts
are,
that
he's
not
expecting
to
find
a
show
stopper,
we
will
change
the
deployment
over
to
this
mechanism
and
I
expect
we'll
be
doing
that
in
the
next
couple
of
weeks.
A
So
it's
a
good
prototype
for
simple
websites
that
can
be
served
out
of
out
of
blob
stores
that
don't
have
any
you
know,
back-end
logic
that
needs
to
run
somewhere.
So.
C
A
Nom-Com
feedback,
for
example,
to
a
Django
user
record
instead
of
the
person
that
was
associated
with
that
Django
user
record.
This
has
caused
us
grief
when
we've
been
cleaning
up
things
with
split
persons
and
split
users.
A
We
have
a
bit
of
work
completed
that
will
change
these
models
so
that
everything
in
the
data,
tracker,
hangs
off
of
person
and
the
only
thing
that
knows
about
user
is
that
link
from
the
data
tracker
person
to
user
not
entirely
true,
they're,
still
a
little
bit.
That's
buried
under
the
covers
with
the
oidc
gearing,
but
the
primary
business
logic
that
we
have
when
we're
worried
about
documents
and
appeals
and.
A
It
can't
yet
know
about
all
of
the
email
addresses
that
a
data
tracker
user
has
so
this
notion
that
we've
gotten
the
data
tracker
where
you
can
log
into
the
data
tracker
with
any
email
address
that
the
data
tracker
knows
you
about
is
not
something
that
we
can
achieve
with
the
IMAP
server
we'll
get
there
eventually,
but
because
of
this,
the
few
people
that
were
doing
that
needed
to
shift
to
using
the
the
actual
user
that
was
attached
to
their
person,
which
involved
them
changing,
basically
redoing
the
their
mail
client
set
up
and
re-downloading
lists
and
whatnot
they've
all
done
that
their
users
have
been
fixed
up.
A
We're
now
going
to
go
through
the
very
large
number
of
users
that
we
have,
most
of
which
are,
as
I
mentioned,
likely.
Bots
that
have
no
person
have
never
really
had
any
login
and
purge
them
and
start
to
tighten
the
code
so
that
we
don't
have
creation
of
these
useless
persons.
F
A
And
they
can't
find
the
person
associated
with
that
author,
that's
listed
in
the
draft
or
if
somebody
makes
a
nom-com
nomination
for
someone
that
they've
provided
an
email
address
or
a
name
variant,
that
the
data
tracker
doesn't
already
know
about
we'll
end
up
with
these
duplicate
person,
records
that
have
to
be
merged
and
we're
merging
the
merging
tends
to
happen
around
meetings
when
people
run
into
trouble
with
attending
and
we're
now
to
the
point
where
we're
merging
Ryan
would
you
say,
like
five
on
average
five
or
six,
each
meeting
cycle.
A
So
it's
a
big
tricky
effort.
We
expect
to
be
completing
this
over
the
next
two
or
three
weeks
again.
So
if
you
have
anybody
sees
any
Fallout
from
it,
keep
in
mind
that
we
might
be
doing
this
in
raise
red
flags.
A
During
117,
several
of
us
got
together
and
sprinted
towards
building
a
standing
dashboard
that
could
be
used
to
show
the
progress
of
work
in
the
ietf
Greg
did
a
bunch
of
of
wireframe
prototyping
started
talking
through
that
with
the
isg
see
what
the
isg
wanted.
There
will
be
continuing
conversations
with
the
isg
and
we're
planning
to
do
another
Sprint
again
later
this
later
this
week.
A
We
think
we've
got
a
good
piece
of
technology
for
building
up
these
dashboards
and
hopefully
we'll
have
something
in
place
that
we
can
actually
be
showing
people
during
ietf,
118.
Greg.
You
have
anything
you
want
to
add
to
that.
G
I
didn't
get
any
further
feedback
than
what
was
shared
in
the
room,
so
okay,
I,
don't
think,
there's
anything
further
to
follow
up
at
the
moment,
as
Robert
said.
Well
me
sharing
something
when
we
have
it.
A
All
right,
my
next
Point,
this
will
probably
turn
into
an
action
item
for
I.
Guess:
it'll
be
Eric
again
since
he's
the
person
we
have
on
from
the
isg.
A
A
A
We
are
using
cloudflare
to
shed
the
load.
We
have
the
load
down
to
something
very
small,
with
a
very
aggressive
rate
limiter,
but.
A
It
would
be
nice
to
to
find
out
why
we
have
this
file
and
who
should
be
on
the
hook
for.
B
Thanks
yeah
I
know
the
background
to
this.
This
is
the
what
happens
if
Ayanna
fails.
B
That
means
that
we
have
copies
of
various
things
from
my
Anna
on
our
website,
and
we
did
actually
discuss
this
with
the
community
some
years
ago,
when
we
got
rid
of
a
whole
bunch
of
Ayanna
things
that
we
were
keeping
I,
think
it
was
the
time
zone
database
actually
that
we
were
keeping
copies
of
and
I
mean
I
I
I
understand
the
original
reason
for
it,
but
and
I.
B
Think
Russ
was
the
person
that's
responsible,
for
it
he's
not
on
the
call
right
the
but
Ayanna
is,
is
is
designed
to
accept
this
level
of
requests,
in
fact,
orders
of
magnitude
greater
than
this
for
these
sorts
of
things
and
us
having
a
copy
of
it
in
case
Ayanna
fails
is
one
thing
but
are
serving.
The
copy
is
a
very
different
thing
and
I
don't
believe
that
we
should
be
serving
the
copy,
so
I
think
we
can
and
I
also
don't
think
we
need
to
worry
about
keeping
the
copy.
B
Personally,
you
know
right
since
that
time
Ayanna
is
now
globally
distributed,
has
all
sorts
of
different
things.
They've
invested,
you
know
tinted,
Williams
and
stuff,
so
they're
an
order
of
magnitude
different
from
the
way
they
were
before
so
I
think
we
already
have
an
agreement
in
principle
to
turn
these
things
off.
I
can't
find
the
reference
just
yet,
but
I
think
it's
on
tools,
discuss
lists
and
I
would
suggest
we
go
ahead
and
do
that.
A
All
right
Liz:
could
you
capture
that
we
are
going
to
go
through
the
process
of
turning
this
off
I'll
contact,
Astor
p,
and
let
them
know
that
they
should
change
the
link
that
they've
got
to
go
directly
to
Ayanna
or
you
know,
drop
it
as
a
mirror
out
of
their
software.
A
I
think
it
was
an
explicit
that
I
we,
we
exposed
a
great
deal
of
what
used
to
be
just
to
Diana
on
www.itf.org,
and
it
sounds
like
this
was
a
subset
of
that.
C
A
B
Just
to
add
sorry,
it
was
when
a
Roman
ran
the
consultation
on
turning
off
FTP
that
we
covered
this
I
believe
and
there
were
or
70
80
more
90
messages
on
that.
So
I
think
that's
reasonably
well
covered.
C
A
A
Looming
issue,
that
is,
a
few
incidents,
have
raised
its
importance
with
the
quality
of
the
data
that
we
have.
We
had
this
metadata
associated
with
an
email
address
about
whether
that
email
address
is
active
or
if
it's
the
primary
email
address
for
someone.
We
have
large
parts
of
the
data
tracker
when
it
sends
mail
will
refuse
to
send
mail
to
an
email
address.
That's
inactive
and
we'll
hunt
for
the
most
recently
active
one
or
the.
A
And
use
that,
instead,
if
somebody
is
trying
to
send
mail
to
an
inactive
address,
this
isn't
implemented
everywhere
that
the
data
tracker
sends
mail.
The
submission
tool
is
notorious
for
just
sending
mail
to
you
raw
bits
that
come
in
as
author
email
addresses.
So
there's
there
are
issues
that
we
have
to
go
modify
the
way
the
data
tracker
sends
mail
to
always
pay
attention
to
active
or
not,
which
we
plan
to
leverage
on
dealing
with
having
the
data
tracker.
Do
the
right
thing
when
it
knows
a
participant
is
deceased.
A
When
you're
dealing
with
like
a
draft
submission
with
a
deceased
author,
we
had
an
problem
with
this
most
recent
nam-com,
where
someone
didn't
end
up.
C
A
In
that
initial
set
of
choices
and
didn't
end
up
getting
notified
because
the
data
tracker
was
using
an
address,
that
was
actually
a
bad
address
that
had
come
in
by
way
of
a
co-author,
submitting
an
internet
draft
and
getting
the
email
address
wrong,
and
this
didn't
get
straightened
out
until
after
the
challenge
period
ended.
So
that
person
didn't
end
up
on
this
nom-com.
A
We
need
to
put
some
pressure
on
making
sure
that
these
addresses
are
correct.
I
intend
on
the
page
for
someone
volunteers
for
a
nom
nom
come
to
make
it
very
clear.
These
are
the
addresses
that
the
day
your
tracker
will
try
to
contact
you
at
these
addresses
won't
be
used
so
that,
at
least
at
that
stage
there
is
a
little
bit
of
Polish
added
to
the
quality
of
the
data.
A
We've
got
a
very
large
number
of
Records
out
there
that
have
been
acquired
by
way
of
Internet
draft
submission
when
there
wasn't
a
person
already
in
place
and
a
user
to
go
with
that
person
where
we
scooped
up
the
email
address,
because
we
needed
an
email
address
to
potentially
reach
the
author
and
that
email
address
will
end
up
in,
say
the
draft
aliases,
but
it
is
otherwise
marked
inactive
right.
A
So
the
question
is
going
forward:
do
we
just
Mark
these
email
values
that
come
in
by
way
of
Internet
drafts
always
active?
Do
we
try
to
do
anything
retroactively
to
the
ones
that
are
inactive
for
the
people
that
we've
got
their
users?
We
previously
discussed
I
believe
on.
One
of
these
calls
having
the
data
tracker
at
whenever
they
access
use
the
messaging
framework
to
periodically
tell
them.
You
know,
with
the
yellow
box,
hey
by
the
way.
A
F
F
C
E
A
Don't
do
anything
with
them
automatically,
it's
not
so
much
that
there
are
too
many
of
them.
It's
the
the
doing
anything
with
them
automatically
is
dangerous
because
you
have
these
transient
periods
where
you
have
bounces,
because
somebody's
ISP
has
done
something
silly.
A
And
we
could
with
mailman
3,
you
know
somebody
does
bother
to
log
in
if,
if
we've
disabled
an
address
for
excessive
bounces
anywhere
in
the
system,
we
could
you.
B
Thanks
so,
as
a
you'll
probably
know,
I
send
a
a
survey
out
once
a
year,
an
IHF
Community
survey
and
I
send
it
to
every
mailing
address
that
is
registered
to
one
of
our
mailing
lists,
which
is
about
55
000
email
addresses
I,
get
about
5,
000
hard
bounces
from
that
and
I
get
a
few
hundred
more
like
about
500
more
or
something
that
are
then
have
repeated
hard
bounces
from
that,
and
therefore
mailman
you
know
removes
them,
and
so
what
I
do
is
I.
B
Take
the
the
full
list
that
I
get
given
from
Robert
I
create
a
populator
mailing
list
with
it.
I
then
use
the
mailing
list
to
send
out
the
survey
and
then,
when
I'm
done
with
the
survey,
I
export
that
list
of
of
members
again
and
the
difference
between
that
and
what
I
was
originally
given
are
the
five
plus
thousand.
You
know
who
are
definitely
dead.
Email
addresses,
I,
think
I've
done
it
twice
now.
B
I've
got
you
know
pretty
much
the
same
set
of
things
so
there's
for,
for
example,
there
is
a
a
particular
ISP
or
something
career
that
must
have
collapsed
and
there
are
like
100
plus
you
know
just
associated
with
that.
So
I
think
we've
got
some
good
data
somewhere
on
this,
but
we've
definitely
got
a.
You
know
at
least
a
one
in
ten
problem
here.
A
And
it's
a
subset
of
those
addresses
that
would
be
associated
with
an
account
in
the
data
tracker,
I
suspect.
A
Okay,
I'll
move
along.
A
We've
had
a
discussion
with
a
few
people.
Russ
Roman
I've
talked
a
little
bit
about
this
with
Jay
as
we
change
I.T
change,
our
I.T
infrastructure.
One
of
the
mechanisms
that
we've
used
to
sign
internet
drafts
isn't
going
to
work
anymore
because
the
cam
just
the
Machinery,
won't
be
in
place.
A
And
we'll
wait
for
Jay
to
come
back
because
I'm
expecting
him
to
see
something
here
and
if
we
don't,
then
great
less
work
for
us
to
do.
If
we
do
then
we'll
work
with
the
security
ads
to
come
up
with
a
way
that
is.
A
A
So
during
these
conversations
it
came
up
that
what
one
of
the
primary
motivations
for
signing
these
things
in
the
first
place
and
what's
in
RFC
54.85,
was
to
make
it
easier
on
the
Secretariat
when
responding
to
requests
about
internet
drafts.
When
was
this
the
internet
draft
that
you
published
and
when
it
when
was
it
made
available
to
the
community?
C
A
All
right,
so
these
conversations
will
continue
and
we'll
do
the
right
thing,
with
Consulting,
with
the
community
Etc
and
and
changing
what
we're
doing.
B
From
now
from
a
legal
view,
I
did
signatures
and
nonsense.
No,
they
don't.
B
They
they
would,
they
do
for
documents,
but
for
for
us
everything
is
a
subpoena.
That
is
then,
a
you
know,
a
notified,
a
declaration
from
us
as
to
what
document
existed
at
what
time,
and
that's
the
only
way
that
any
of
these
lawyers
want
to
do.
It
is
that
declaration
from
us,
so
the
the
art,
the
signature,
doesn't,
do
anything.
A
A
So
I
would
like
I
think
to
try
and
I
know
Eric.
This,
isn't
your
favorite
way
of
doing
things
but
to
just
give
people
a
few
minutes
to
read
everything:
that's
in
the
FYI
section
and
if
you
have
any
questions
to
to
raise
them.
A
E
When
you
started
to
ask
the
question,
I
was
looking
for
my
unmute
button,
because
I
was
about
to
say
something
there
I
I
think
you
are
in
the
same
isg
selectional
as
isg
is
so.
Yes,
we
will
go
to
the
yes,
no,
no
opinion
and
with
three
buttons
to
be
selected.
If
you
don't
select,
none
of
them,
you
are
not
content.
In
the
opinion,
no
opinion
is
kind
of
an
opinion
itself.
E
That's
all
we
want
to
say
you
were
right
on
an
email
that
I
want
to
make
Express
this
pretty
clear
and
I
will
make
it
pretty
clear
this
week
over
email
to
two
list.
Yes,
that's
the
one
that
the
thing
we
wanted
to
do
right
to
be
minute.
Yes,
okay,
I
will
do
it.
Please
write
a
note
here.
Welcome.
A
A
And
there's
a
lot
in
this
list.
If
you
look
at
what
happened
with
data
tracker,
we
had
a
lot
of
of
code
Sprint
activity,
some
things
like
dark
mode
came
in.
A
Okay,
I'll
repeat
this:
exercise
for
the
bib
XML
service.
A
I'll
make
a
comment
here
again
that
I'm
still
struggling
with
trying
to
make
it
easier
for
the
Secretary
of
the
RPC
to
correct
all
their
names,
as
they
appear
in
the
the
entries
for
specific
versions
of
a
draft.
I.
Think
it's
going
to
require
an
actual
new
set
of
data.
Tracker
views
to
be
created
to
allow
this
editing
to
happen
sensibly
and
to
stick
so
still
a
work
in
progress.
A
A
A
One
of
these
is
changing
the
structure
of
the
XML
to,
as
mentioned,
avoid
stripping
the
tea
tags.
All
of
these
things
are
probably
good
fodder
for
people
to
pay
attention
to
when
they're.
Considering
the
do
we
republish
re,
you
know
make
new
XML
for
already
published
XML.
Do
we
re-render
and
when
do
we
re-render
so.
A
B
C
B
It
is
it
worthwhile
explaining
to
people
the
importance
of
it
being
a
node.js
climasm
Library.
A
Yeah
I
think
we
touched
on
this
on
a
previous
call,
but
it's
worth
touching
them
again.
The
idea
for
ID
Nets
being
Rewritten
entirely
as
a
node.
You
know
a
JavaScript
library
thing
is
that
it
will
also
live
in
the
directly
in
the
editors
that
we're
building
for
the
RPC
that
might
have
good
reuse
for
authors
in
the
long
run.
A
We
might
also
be
able
to,
though
we
will
also
be
able
to
leverage
it
to
do
early
work
on
ID
submission
before
somebody,
you
know,
is
actually
bothering
the
back
end
with
the
submission
the
front
end
can
will
have
given
submitters
some
feedback.
A
And
the
last
item
we've
got
on
here,
I
haven't
seen
Joe
on
screen.
It
looks
like
he
is
connected
to
the
call,
though
that
changes
as
Joe
reported
here,
the
the
well.
You
should
just
let
him
talk
Joe
if
you're
here.
A
A
So
one
of
the
things
that
I'm
planning
to
do
on
future
calls
is
to
have
to
bring
more
things
like
this
issue,
with
the
freshness
of
the
email
addresses
up
for
discussion
during
during
these
calls,
when
we
have
large
clusters
of
issues
around
problems
and
in
the
data
tracker
features
that
people
want,
that
aren't
easy
to
implement.