►
From YouTube: IETF 92 - Newcomers' Orientation
Description
(Session begins around 9 minutes in.)A tutorial introducing the history of the IETF, its purpose and processes. See http://www.ietf.org/meeting/92/92-newcomers.html for more information
A
Well,
good
afternoon,
welcome
to
this
new
newcomers
session,
my
name's
Scott
Brandner
I'm,
one
of
the
Greybeards
of
the
IETF,
as
you
can
tell
there's
a
few
of
us
gray
beards
around.
This
is
trying
to
give
you
a
little
bit
of
information
about
what
the
IETF
is.
What
we
think
we
do,
we
how
we
do
it
and
what
we're
here
for
we've
got
an
agenda
some
a
little
bit
a
little
bit
of
history
and
what
we
think
our
purpose
is
how
we
get
how
work
gets
done.
A
What
you'll
see
this
week
in
terms
of
how
work
gets
done,
the
IETF
role
in
the
world?
How
is
how
its
structured,
how
its
management
is
selected
and
process
and
procedure
of
how
material
gets
put
out?
Therefore,
as
standards
typical
working
group
session
and
some
stuff
on
intellectual
property
rights,
not
because
it's
wonderful
stuff,
but
you
need
to
know
it.
Patents
are
a
pain,
so
the
IETF
ITF
doesn't
exist.
A
It's
not
a
legal
entity,
but
it
was
organized
and
formed
in
1986
out
of
some
US
government
activities
that
were
designed
to
continue
the
creation
of
technology.
For
what
was
then
the
ARPANET.
The
ARPANET
was
the
first
of
the
large-scale
packet
data
packet
data
networks,
packet
switching
networks
started
to
be
deployed
in
1969,
but
the
US
government
decided
they
wanted.
It
was
not
a
static
thing.
They
wanted
to
be
able
to
continue
development
of
protocols
for
it.
A
The
biggest
protocol
at
the
that
was
evolved
after
1969
was
tcp/ip,
which
was
rolled
out
on
the
ARPANET
in
January
of
1986
83.
These
are
various
government
activities.
The
IETF
has
been
for
most
of
its
early
life
completely
ignored,
and
that
was
great
because
we
didn't
have
to
answer
to
anybody.
We
did
what
we
thought
was
the
right
thing
to
do.
Rather
than
being
told
what
the
right
thing
to
do
was
which
a
lot
of
standards
bodies
are
in
the
I'm
doing
we're
not
government-approved.
We
don't
have
formal
government
relations
of
officials
within
the
IETF.
A
That
is
a
problem.
That's
it's
a
benefit
and
a
problem.
It's
not
just
with
governments
but
with
other
organs.
Other
Sanders
organizations,
the
IHF
mantra,
is
its
people,
not
companies.
It's
people,
not
organizations
you're
here
representing
yourself,
and
what
what
knowledge
you
have
and
what
capabilities
you
have.
A
The
fact
you
work
for
a
particular
company
or
a
particular
organization
is
irrelevant
if
somebody
gets
up
to
microphone
and
IETF
meeting
and
says
I'm
here
from
Cisco
and
Cisco
thinks
it
will
be
laughed
at,
because
everybody
knows
that
Cisco
doesn't
think
companies
don't
think
people
think
so
wear
your
hair
as
individuals.
You
may
may
have
a
lot
of
baggage
with
you,
but
you're
here
as
individuals
and
it's
your
own
capabilities
and
intellect
and
knowledge
and
technical
ability
is
what
wits
counts,
not
where
you're
from
not
what
what
name
is
on
your
badge.
A
It's
really
it's
it's
a
meritocracy
that
can
be
very
frustrating,
particularly
for
folks
that
are
coming
from
government
government
entities
or
from
other
standards
bodies,
because
when
they're
speaking,
they're,
not
just
speaking
for
themselves,
they're
speaking
for
a
large
large
group,
sometimes
a
great
deal
of
technical
effort
for
a
particular
from
a
senator's
body.
But
we
don't
can't
really
can't
tell
the
we
can't,
there's
no
way
for
us
to
judge
the
quality
of
that
thinking.
So
it's
really.
We
have
to
look
at
what
the
knowledge
that
the
comments
that
are
made.
A
We
have
to
look
at
that
directly
and
judge
judge
what
we
think
based
on
what
that
said.
So
that's
that's
an
advantage
because
we're
getting
the
best
technical
stuff
out
it's
a
disadvantage,
because
there
are
times
when
we
can
really
learn
from
what
a
government
and
what
government
representatives
say
or
a
standards
body
representatives
say,
but
we
just
don't
know
how
to
do
that
and
you'll
you'll
may
run
into
that
from
time
to
time
this
week.
A
So
we're
not
government
approved
governments
that
approve
our
output,
there's
no
formal
role
for
governments.
We
were
funded
by
the
US
government
up
until
1997
at
least
the
Secretariat
was
the
primary
you'll
see
later
later
on.
The
primary
funding
for
the
IETF
is
the
people,
the
organizations
that
the
people
who
come
here
and
participate
in
it
belong
to
Harvard
pay.
Some
of
my
gives
me
my
time
to
do
work
on
this.
Other
companies
pay
for
their
time
and
expenses.
A
A
And
we
reject
Kings
presidents
and
voting.
We
were
believed
in
the
rough
consensus
and
running
code
and
I'll
go
through
each
of
those
things
in
sequence,
as
we
get
to
them
so
ITF
overview.
We
do
internet
standards,
not
all
internet
standards,
but
most
of
the
internet
says
most
of
the
plumbing
standards
that
we
do.
We've
got
some
surface
things
like
the
page,
descriptor
language,
HTML,
that's
done
by
the
w3c
I've
got
things
like
the
Ethernet
standard
or
the
tea
or
the
ATM
stander,
or
the
how
you
put
photons
on
fiber.
That's
not
us!
A
A
We
don't
legally
exist,
it's
just
an
act
or
organized
activity
of
the
Internet
Society
Internet
Society
is
a
legal
entity,
but
the
IETF
isn't
by
itself
a
legal
entity.
It's
just
an
organ
is
just
a
a
process
as
much
as
anything
else
we
have
somewhere
between
1,000
and
1,500
people
at
meetings.
There
were
thirteen
hundred
and
thirty-five
I
think
people
pre
registered
for
this
meeting,
which
is
bigger
than
most.
It
was
the
surprise.
A
Here's
the
chart
of
time.
Over
time
we
started
out
at
21
attendees
at
a
company
called
Lync
a
bit
in
San
Diego.
Four
of
those
21
people
are
still
active
in
the
IETF
I've
seen
I've
seen
one
so
far
today
and
I
expect
to
see
three
others
the
other
three
during
this
week.
So
that's
a
good
percentage.
Wise
people
have
been
there
from
the
very
beginning.
We
peaked
in
the
middle
of
the
internet
boom
of
twenty
eight
hundred.
That
was
a
pretty
heady
time.
A
A
What's
our
purpose
purpose
is
to
develop
and
maintain
standards
and
technologies
used
to
provide
Internet
service
and
use
to
provide
services
over
the
Internet,
so
the
Internet
service
itself,
Internet
service
providers,
its
routing
transport
things
things
on
that
that
level,
but
also
applications
that
are
providing
services
such
as
streaming
audio
streaming,
video
things
like
that.
We
want
to
make
sure
that
the
technology
can
perform
the
functions
that
it
needs
to
buy.
Don't
to
do
that,
we
have
to
understand
what
those
functions
are.
A
They
have
to
understand
what
that
what
the
actual
requirements
are
for
a
technology
before
we
can
standardize
it.
We
support.
We
make
sure
that
it
can
support
at
the
scale.
The
internet
is
one
of
those
things
that
the
concept
of
going
viral
is
not
a
new
one
in
the
internet,
its
new,
in
the
sense
that
that's
that's
YouTube
or
Twitter
or
Twitter,
or
something
but
for
the
history
of
the
internet,
back
it
back
into
the
into
the
mid-90s,
which
it
goes
much
further
back.
A
That's
when
the
web
started
and
when
people
started
really
starting
the
general
population
started
using
it.
You'd,
go
from
unknown
to
a
million
hits
a
day
overnight
in
two
days.
It
would
be
you'd,
be
doing
that
and
so
you're
at
your
systems
and
your
technologies
have
to
be
at
a
scale
to
that
level.
Very
quickly
can't
fall
over
just
because
you
get
over
your
being
you're
a
success.
Death
by
success
is
not
an
option.
We've
got
to
do
better
than
that
make
sure
the
technology's
secure.
A
This
is
even
more
true
in
the
last
few
years
since
the
Snowden
revelations,
the
IETF
got
pretty
jazzed
up
about
the
Snowden
revelations,
and
you
will
see
that
the
IAB
put
out
a
RFC
which
equated
pervasive
surveillance
with
an
attack.
It's
an
attack
on
the
population
of
the
Internet,
and
so
the
IETF
spending
quite
a
bit
of
time,
trying
to
be
ensure
that
it's
that
the
Internet
itself
is
resistant
to
that
kind
of
attack,
so
encryption
by
default,
and
things
like
that.
A
You
want
to
also
it
might
help
to
be
able
to
manage
the
stuff
you
deploy
it,
but
you
can't
manage
it.
Then
it's
not
gonna
be
much
you
much
use,
so
it
has
to
be
manageable
and
we
produce
standards
plus
other
documents.
Well,
what's
a
standard,
quite
a
few
years
ago,
back
in
the
mid
nineties,
when
the
ITU
international
telecommunication
union
was
just
beginning
to
understand,
there
was
this
internet
thing
out
there,
but
really
it
wasn't
really
upped
on
what
it
was.
A
I
was
the
liaison
between
the
IETF
and
the
ITU
and
I
went
over
to
Geneva
for
a
meeting.
There
was
no
internet
connectivity
in
the
building.
You
could
dial
out
to
an
Internet
service
provider.
That's
all
you
could
do
was
it
was
a.
It
was
a
modem
room
in
the
basement
you
could
dial
out
from
so
the
then
head
of
the
IAB
and
I
gave
a
presentation
after
hours
on
what
IETF
was
and
what
the
internet
was,
and
we
had
five
or
six
hundred
people
in
this
room
and
it
was
after
hours,
so
it
was.
A
It
was
pretty
dedicated.
The
very
first
question
I
was
asked
when
we
were
finished.
Our
presentation,
which
was
somewhat
similar
to
this
one,
was:
how
can
you
call
your
thing
standards
if
no
government
mandates
their
use?
Well,
that's
a
different
concept
of
standard
than
we've
got
our
concept
of
standard.
Is
we
put
it
out
there?
A
We
call
it
a
standard
because
it's
gone
through
the
right
sequence
of
processing
to
get
it
to
be
defined
as
a
standard,
but
that
standard
is
irrelevant
unless
people
use
it,
where
there's
no
mandatory
use,
there's
no,
we
don't
have
any
protocol
police
which
is
telling
you
what
to
do
and
what
not
to
do.
Unlike
some
regulate
some
governments
which
have
regulations
saying
you
must
use
this
standard
or
not,
we
don't
do
any
of
that
sort
of
stuff.
Our
standards
are
only
useful
standards
if
you
and
others
decide
to
use
them.
A
We
are.
We
publish
our
standards
as
rfcs,
we'll
talk
more
about
rfcs
in
a
bit,
there's
no
formal
recognition
for
IETF
standards.
They
some
some
requests
for
for
quotes
or
regulatory
requirements
do
refer
to
them,
but
we
don't
have
any
miss,
there's
very
few
that
do
that,
because
if
we
don't
exist
and
how
can
they
refer
to
our
standards
as
something
that's
meaningful
and
it's
a
catch-22
there,
it's
a
it's
a
difficult
thing
to
do.
The
lack
of
formal
government
approval
is
a
problem.
A
In
some
areas
there
are
parts
of
Europe
which
art
which
cannot
by
regulation
refer
to
IETF
standards,
because
we
don't
meet
their
definition
of
a
standards
body,
but
we
don't
submit
our
standards
to
anybody
else
to
approve
their
dis
done
when
we're
done
with
them.
They're
done
and
you
use
it
either
use
or
me
you
don't.
A
A
To
me,
a
document
editor
in
general.
The
working
group
chair
is
not
the
document
editor
because
then
the
working
group
can't
make
a
unbiased
judgment
on
the
quality
of
the
work,
so
the
working
group
doc
they
documented-
is
somebody
different
than
the
working
group
chair
in
most
cases.
Occasionally
that
does
not
that's
not
the
case.
Almost
the
case
it
is
they
have
it
edit
individual
documents.
The
editing
is
exactly
that.
A
It's
editing,
it's
not
authoring
in
the
sense
of
it's
your
document,
you
are
supposed
to
be
reflecting
if
you're
a
documentary
you're
supposed
to
be
reflecting
the
consensus
of
the
working
group
and
in
fact,
we've
had
situations
when
I
was
an
area
director
for
a
walk
and
I
had
situations
where
the
document
editor
was
not
reflecting
the
working
group
consensus
and
we
had
to
fire
the
editor
and
give
the
document
to
somebody
else.
That's
a
very
messy
and
ugly
situation.
You
don't
want
to
get
into
it,
but
it
does
happen.
A
The
the
working
groups
are
collected
together
in
two
areas,
mostly
for
managerial
efficiency.
Something
in
one
area
is
there's
overlap
between
what
working
groups
are
in.
What
area
is
that?
Something
that
should
be
in
the
applications
area
was
in
the
transport
area
when
I
was
transport
ad,
just
because
I
like
the
topic
and
that
and
the
air
plication
serie
directed
didn't,
but
it
could
have
gone
either
way.
So,
there's
it's
a
fuzzy
arrangement.
Some
of
the
some
of
the
working
groups
are
pretty
clear.
A
The
iesg
is
the
body
within
the
IETF
that
makes
the
decision
on
when
something's
a
standard
they're
the
deciding
body
they
review
a
technical
review
of
the
proposals
and
if
they
think
the
proposal
meets
nuf
is
up
is
is
good
enough
has
all
our
problems
resolved
and
that
they
can
and
they
check
to
see
the
consensus
was
reached.
Then
they
would
have
proved
it
as
a
as
a
business
on
the
Status
track
or
informational
or
whatever
other
type
of
document.
A
There
is
also
an
internet
architecture
board,
that's
the
IAB
and
they
provide
general
architectural
guidance.
So
a
little
bit
of
background
here
that
might
be
amusing
in
1992
we're
actually
in
1991
the
IAB
made
some
pronouncements
of
the
future
of
the
internet
without
consulting
the
IETF
general
populace.
The
general
populace
didn't
like
this
very
much
and
rebelled
and
basically
did
a
palace
coup.
At
that
time,
the
IAB
was
the
decision
body.
They
were
the
ones
to
decide
whether
something
was
going
to
be
a
standard
or
not.
Well,
they
worked.
A
There
was
a
working
group
form
to
pass
on
working
the
ass
on
working
group
that
looked
into
this
and
decided
to
reorganize
the
IETF,
and
it
kicked
the
IV
out
of
authority,
put
the
isg
in
place
of
it
and
told
the
IAB
that
it
was
guidance,
not
direct,
not
decision
anymore.
It
moved
from
the
it
is
now.
That's
it's
now
in
that
role
of
a
of
an
advisory
body
rather
than
the
decision
body.
So,
and
we
did
that
through
an
IETF
working
group.
A
So
we
do
working
groups
not
only
for
technology
but
for
our
own
processes
as
well
area
directors.
Each
area
has
one
two
or
three
area
directors.
That's
you
this
used
to
say
as
of
a
few
weeks
ago,
it
said
two
area
directors,
except
for
the
general
area,
which
has
one
which
is
the
ITF
chair,
but
the
AI
is
G
has
reorganized
in
the
last
few
months,
and
as
month,
things
together
and
he's
experimenting
with
different
ways
to
organize
the
working
groups
within
within
areas
and
so
they're
right
at
the
moment.
A
The
general
area
has
one,
and
one
of
the
other
technical
area
has
one
and
then
some
of
the
others
have
three
most
of
them
have
two
and
there
they
want
people
that
are
responsible
for
creating
working
groups.
Disbanding
working
groups,
working
group
charters,
making
sure
the
working
group
is
generally
doing
its
job.
If
not,
then
they
go
have
a
discussion
with
their
working
group
chairs
to
get
get
it
to
work.
They
also
do
a
technical
review
of
any
proposals
that
come
up
from
working
groups,
so
the
area
director
suppose
an
important
role
there.
A
They
they
are.
They
has
a
technical
role
because
they
have
to
evaluate
the
technology,
that's
being
proposed
for
a
standards
track
or
from
informational,
and
then
they
also
have
to
be
managerial
in
that
they're.
Managing
all
this
efforts
and
the
iesg
of
every
mention
is
the
deciding
body
it
is
cross
area.
Technical
Review.
This
is
one
of
the
things
that's
fundamentally
different
about
the
IETF
than
many
other
standards
bodies,
a
lot
of
other
standards
bodies.
A
The
work
of
the
developing
a
standard
is
takes
place
in
a
working
group
or
a
study
group,
or
a
wrapper
tors
group
in
a
small
technical
group
that
with
people
who
are
focused
on
that
particular
technology
topic,
once
it's
done
in
there
all
the
rest
of
the
steps
in
the
process
are
process
steps.
Did
you
send
out
the
right
ballot?
Did
you
do
this?
Did
you
do
that,
but
they're
not
evaluative?
They
don't
go
in
and
evaluate
the
technology
of
those
higher-level
steps.
They
just
make
sure
that
the
processes
are
followed
ITF.
A
Does
it
differently
when
a
working
group
finishes
finishes
a
proposal
hands
it
up
to
the
iesg
through
their
Area
Director
Area
Director.
First
reviews
it
to
sees
whether
it
makes
technical
sense
and
whether
it
makes
linguistic
sense
just
and
whether
the
thing
makes
any
sense
at
all.
We
can.
Can
you
read
it
and
understand
it
if
you
can
read
it
under
if
the
area
director
thinks
you
can
read
and
understand
it,
then
it
goes
up
to
the
iesg.
The
isg
then
does
a
cross
area.
A
Technical
Review
each
of
the
isg
members
is
from
a
different
technical
area,
security,
routing
transport.
Whatever
real-time
activities
all
of
those
different
areas,
they
all
evaluate
the
same
documents.
So
it
comes
up.
The
working
group
done
every
security
expertise.
It
gets
to
the
isg.
The
security
area,
directors
says
the
area
directors
say
this
is
not
going
to
fly
because
it
can't
it
doesn't
have
enough
security
or
the
operational
area.
A
A
Apple
is
telling
me
something:
I
didn't
want
to
know,
but
that's
Apple,
for
you
so
does
cross
Harry
review
and
if
there's,
if
that's
like
it,
then
it
passes
and
I'll
go
into
a
little
bit
more
part
how
that
process
works
in
a
bit
man.
The
iesg
manages
the
general
process
of
the
IETF
approves
working
groups
and
things
like
that.
A
So
how
does
it
actually
get
done?
How
does
it
work
get
done
almost
all,
but
not
all
that
standards
come
out
of
working
groups.
You,
the
working
group,
does
the
thinking
about
a
standard?
Does
the
somebody
might
propose
a
document
and
the
document
proposed
document
document,
as
proposes
in
the
form
of
what
is
called
an
internet
draft
and
the
working
group
reviews
it
and
iterates
on
it
so
that
successive
generations
of
internet
drafts,
the
BGP
update,
took
26
iterations
of
the
draft
to
get
it
right,
and
that
was
mostly
because
the
technology
kept
changing.
A
It
got
finding
frogs,
they
had
to
fix
and
had
to
fix
the
technology
as
I
went
along.
So
these
working
these
working
group
documents,
the
internet
drafts
mostly
are
done
by
people.
The
document
editors
within
the
context
of
a
working
group
and
at
the
direction
of
the
working
group
itself,
so
the
editor
is
supposed
to
be
reflecting
the
working
group
consensus
if
revision
after
revision
and
then
when
it's
when
they're
working
group
believes
it's
ready.
It
then
kicks
it
up
to
the
area
director.
A
But
they
arrest
the
is
the
read
the
rest,
the
is.
She
doesn't
review
it
immediate
late.
The
first
thing
that
the
rest,
the
is
she
does,
is
send
out
what
is
called
a
last
call
an
IETF
wide
last
call
this
is
to
anybody,
that's
on
the
IETF,
announce
list
and
people
in
other
standards
bodies
who
are
subscribed
to
a
mailing
list
that
we
maintain
that
anybody
can
comment
on
this
anybody
out.
So
the
idea
here
is
to
get
people
outside
of
the
expertise
of
the
local
people,
the
people
that
develop
the
standards.
A
You
may
know
something
that
the
people
who
develop
the
standard
didn't
know
when
we
free
free
content,
some
niche
of
expertise,
we'll
look
at
a
proposal
and
say
this
is
broken
because
of
X,
because
this
won't
work.
Because
of
this
thing
they
didn't
know,
and
they
didn't
think
of
so
it
has
to
go
through
that
process.
Once
it's
gone
through
the
last
call
process
and
the
isg
evaluates
the
results
of
the
last
call
and
their
own
cross
area,
Technical
Review
and
either
kicks
it
back
to
the
working
group
for
more
work
or
exibições.
A
Yes,
one
thing
I
meant
forgot
to
mention
of
the
previous
thing.
I
said
that
most,
but
not
all
standards
come
from
working
groups.
You
can
write
your
own
if
you
come
up
with
this
phenomenal
idea
for
a
pinochle
playing
protocol,
you
can
write
it
up,
write
it
up
and
if
you
you
write
it
up,
you
publishes
an
internet
draft.
We'll
talk
about
inner
drafts
drafts
in
a
bit,
publish
the
internet
draft,
you
circulate
it.
You
tell
people
on
the
IETF
list
about
it,
may
get
excited
about
it.
They
read
it.
A
They
offer
suggestions,
you
modify
it.
No
working
groups
involve
it's
just
you
doing
it
and
you
keep
doing
that
and
then,
when
you
think
it's
ready,
you
can
go
around
shop
it
to
various
area,
directors
and
say:
I
think
this
is
ready
for
to
being
a
standard
and
if
you've
got
an
area
director
that
believes
in
it
they
can
pick
it
up.
Panic,
Rock
work
through
that
process,
same
process
that
goes
to
an
IETF
lot
working
group.
A
It
goes
from
I,
ATF
last
call
and
then
across
area
Technical
Review,
and
all
of
that,
the
one
difference
is
that
for
a
working
group
document,
it's
a
two-week
last
call
is
G
sends
out
a
last
call
saying
tell
us
within
two
weeks:
what
do
you
think
for
an
individual
submission?
It's
four
weeks,
so
it's
going
to
take
you
four
weeks
to
get
that
pinochle
playing
protocol
through,
but
you
can
still
do
it.
There's
not
a
requirement
to
go
through
a
working
group.
It's
just
generally
done
that
way.
A
A
When
I
was
area
director
for
transport,
we
had
a
number
of
birds
of
a
feather
sessions
or
called
buffs,
which
were
simply
informational
to
provide
context
for
the
for
the
ietf
about
some
issue.
We
did
one
on
on
quality
of
service
over
the
Internet
didn't
lead
to
a
working
group.
It
led
to
just
more
information
about
what
could
be
done
and
what
the
expectations
were.
A
We
had
somebody
in
had
brought
in
people
from
big
banks
and
from
big
companies
and
from
little
researchers
to
tell
us
what
is
what
they
would
like
to
see
in
a
Internet
that
they
could
control
a
quality
on.
We
had
all
kinds
of
stuff
input
from
various
people
and
the
most
memorable
was
the
guy
from
FedEx
said
that
what
he
wants
out
of
the
Internet
is
McDonald's.
He
wants
the
same
crappy
burger,
no
matter
where
he
goes.
He
wants
predictability
exactly
the
same,
no
matter
where
he
is
in
the
world.
A
He
gets
a
but
McDonald's
burger.
He
gets
the
same
thing
he
wants
out
of
the
internet,
no
matter
what
day
it
is,
but
what
time
of
day
it
is,
he
gets
the
same
quality
of
service.
He
gets
the
same
and
user
experience.
That
was
a
memorable
way.
To
put
it,
of
course,
that's
from
a
company
whose
business
is
doing
something
other
than
what
everybody
else
does
delivering
faster
and
more
expensively
jew-jew,
so
working
groups
and
I
both
can
create
a
working
group.
A
So
one
of
the
things
working
AB
office
4
is
to
allow
the
area
director
to
decide
whether
there's
enough
interest
in
the
topic
and
enough
support
for
the
topic
that
a
working
group
could
be
successful
there
enough
worker
bees,
that'll
work
on
it.
I
remember
one
time
when
I
was
area
director
I
did
a,
we
did
a
Boff
and
where's
300
people
in
the
room
or
something
it
was
a
popular
topic
at
the
end,
I
got
up
and
I
said:
okay
show
of
hands.
A
How
many
people
think
this
should
be
an
IETF
working
group
it
about
half
the
people
put
up
their
hands.
You
will
see
this
week
that
that's
actually
a
very
good
turnout.
Most
of
the
time,
it's
a
very
small
fraction
of
the
people
in
the
room
express
an
opinion.
So
then
I
asked
a
reverse
question:
how
many
people
think
this
should
not
be
an
IETF
working
group
and
about
half
the
people
put
up
their
hands?
A
Well,
that
led
me
to
believe
there
wasn't
consensus,
even
rough
consensus,
that
this
should
be
a
working
group,
so
it
did
not
become
a
working
group
it
at
that
point.
It
did
a
few
years
later
when
there
was
more
consensus
on
what
it
should
be
doing,
so
the
the
Boff
is
trying
to
understand
whether
it's
ready
for
primetime
ready
to
be
work
done.
Sometimes
the
Boff
will
result
in
this
is
still
research.
We
don't
know
how
to
do
this.
A
The
technology
is
not
yet
developed
enough,
in
which
case
the
IRT
F
I'll
talk
about
that
in
a
bit.
The
IRT
F
is
the
proper
place
for
it.
We
only
put
things
in
the
IETF
working
groups,
which
we
know
we
can
standardize.
We
know
we
enough
about
the
technology.
It's
ready
for
actual
deployment,
it's
ready
for
actual
development.
A
Most
of
the
work
is
done
in
working
groups,
few
off
and
individuals.
Most
of
the
work
is
done
during
working
groups.
The
face-to-face
meetings
like
this
week
are
short
and
have
to
be
to
the
point.
In
many
standards
bodies,
a
working
group
meets
in
a
face-to-face
way
for
a
week
to
two
weeks.
At
a
time.
That's
not
the
way
it
works
here.
An
average
session
is
between
an
hour
and
two
and
a
half
hours.
Occasionally,
a
working
group
will
have
more
than
one
session
in
a
week.
A
That's
not
enough
time
to
write
the
document
in
real
time.
The
document
is
written
on
the
mail
on
that
using
the
mailing
list,
succession
of
internet
drafts
commentary
on
the
mailing
list.
What
goes
on
at
this
this
week
goes
on,
and
on
IETF
meeting
is
people
getting
together
to
worry
about
the
high
order
issues,
issues
that
weren't
resolved
in
the
mailing
list?
Should
they
should
the
bits
be
in
this
order
or
that
order?
That's
a
discussion
you
have
in
them
in
the
working
group
session.
A
You
won't
get
a
discussion
of
the
overall
technology
because
it's
the
specific
point,
specific
open
issues
is
what's
going
to
be
discussed
there
as
I
say,
the
meetings
are
quite
short
an
hour
an
hour
to
two
and
a
half
hours.
You
can't
do
a
lot
in
that
period
of
time
in
terms
of
actually
generating
a
document.
Some
of
the
working
groups
do
have
a
second
section
of
their
meetings
where
people
propose
new
activities
for
the
working
group,
and
you
can
you'll
you'll
see
some
of
that
this
week.
A
Almost
always
IETF
working
groups
come
from
the
bottom
that
become
bottoms
up.
A
number
of
people
get
together
and
say.
This
is
something
we
would
like
to
work
on.
We
have
an
idea
here.
They
write
an
internet
draft,
they
publish
internet
draft
under
their
own
name,
not
under
the
eye
and
not
under
working
group
name
under
their
own
name,
and
they
say
here
it
is:
let's
stock
up
talk
about
it.
Let's
have
a
bought.
It
there's
something
here
we
can
talk
about
and
from
that
grows
a
working
group.
A
I
say
you
don't
need
a
buff
to
generate
a
working
group.
A
narrator
I
can
do
that.
Just
by
looking
at
the
document
and
seeing
the
discussion
of
mainly
us
and
say:
okay,
I'm
gonna
make
a
working
group,
but
most
of
the
time
it
goes
through
this
process
of
doing
a
boss.
Very
occasionally,
an
area
director
will
go
and
pick
some
people
and
say
we
need
this
technology.
We
need.
A
A
We're
trying
to
do
in
that
by
using
that
mechanism
having
a
charter
like
that,
we're
trying
to
risk
to
focus
the
working
group
on
their
work,
they
need
to
get
done
rather
than
going
off
on
various
excursions
into
various
other
things.
We
want
them
to
work
on
the
things
that
we
want
them
to
work
on
and
what
they
agreed
to
work
on
and
in
theory,
and
more
often
than
not,
when
a
working
group
God
gets
its
job
done,
publishes
its
documents,
it's
disbanded,
there's
no
need
for
it
anymore.
Once
it's
done
now.
A
This
is
a
great
shock
to
some
people.
Many
most
standards
groups,
you
become
a
working
group
chair
and
that
you
were
rewarded.
If
you
do
a
good
job,
you're
rewarded
and
you
get
another
working
group
or
you
get
this
working
group
continued
with
a
new
charter.
We
don't
generally
do
that.
I.
Remember
one
point:
when
I
one
of
the
what
my
working
groups
did
a
wonderful
job
got
everything
done
and
I
called
up
the
the
chair
and
said
great
job.
A
It's
the
first
meeting
of
the
physical
meeting
I
walked
into
was
that
particular
working?
That's
still
going
on,
because
there's
still
much
still
options
that
you
want
to
add
to
dynamic
host
configuration
protocol,
but
most
working
groups
tend
to
fade
out.
Then
they
go
away.
So
here's
they
there's
a
visual
of
the
what
I've
talked
about
might
have
a
Boff
chair
works
to
it.
Chair
works
up
a
charter
with
the
get
a
chair.
A
Just
come
up
with
a
Charter,
get
get
a
process
in
place,
get
to
make
sure
the
working
groups
there
read
the
worker
bees
are
there
area
director
looks
at
it
if
the
area
director
Gries
pops
it
up
to
the
isg,
which
then
asks
the
IAB
for
their
advice.
If
it
also
asked
this
new
work
list,
new
work
list
is
a
as
a
whaling
list.
We
maintain
between
different
standards
bodies,
any
standards
body
can
subscribe
to
this,
and
so
we
set
up
saying
we're
about
to
work
on
this
new
topic
and
here's
this
Charter.
A
What
do
you
think
and
that's
specifically
there
so
another
standards
body
and
say
whoops
I'm
already
working
on
that
we're
already
working
on
it
over
here?
You
don't
need
to
do
it
or
we
want
you
to
work
together
with
us
or
liaison
with
us
or
do
something,
but
don't
ignore
us
and
we
get
that
occasionally,
but
but
just
truths
to
be
said.
The
reaction
time
is
not
quite
right
for
most
standards
bodies
we
put
out
this
list.
Note
that
says
isg
is
considering
this
new
working
group.
A
Please
let
us
know
within
two
weeks
whether
it's,
okay,
whether
you
have
any
questions
and
most
standard
most
standards
bodies,
can't
say
hello
in
two
weeks.
They
don't
know
how
to
how
to
do
that
process
wise.
They
don't
have
any
way
to
do
that.
So
it's
we
do.
We
occasionally
get
a
way
to
say:
wait
a
sec,
I
gotta
talk
to
folks
about
this
hold
on,
but
most
of
the
time
don't
get
a
lot
of
response,
because
it's
too
fast
to
return
time.
A
So
working
group
session,
which
you'll
see
this
week
is,
as
I
said,
only
a
few
hours.
So
it's
focused
on
the
big
things
specific
unresolved
issues.
Things
like
that.
You
have
microphones
like
this
in
the
middle
of
the
room.
This
this
is
a
room.
That's
will
be
used
for
working
group
sessions
this
week
and
you'll
see
this
old
tag
on
the
microphone
a
little
little
sign
on
the
microphone.
That
is
to
remind
you
to
state
your
name.
A
Every
time
you
get
to
the
front
of
the
microphone,
you
may
have
been
there
one
minute
before,
but
the
people
out
there
in
audio
land
don't
know
that
and
the
person
taking
the
minutes
doesn't
know
that
so
every
single
time
you
went
to
the
microphone
say
your
name
and
the
other
thing
is
talk
to
the
microphone.
Don't
talk
to
the
guy
that
asks
you.
The
question
talk
to
the
microphone.
It's
amazing
how
poor
geeks
are
dealing
with
microphones
they
throw
over
here.
They
talk
over
here
microphones
over
here.
A
You've
got
to
pretend
to
love
the
microphone,
get
really
close
to
it,
really
like
the
microphone,
but
talk
to
the
microphone.
Don't
talk
to
your
friend
okok
around
it
talk
to
a
microphone
and
close
if
you're
tall
lean
down
if
you're
short
and
the
microphone
down
but
get
to
the
microphone.
I
can't
emphasize
that,
because
it's
so
poorly
adhered
to
it
should
be
an
easy
concept,
voices,
voices
and
well-known.
A
You
are
welcome
you're
new
here
fine,
some
of
you
have
been
comport
than
one
I
ATF
meeting
and
just
coming
for
the
abuse
in
this
session.
But
if
you're,
if
you
had
something
to
say,
say
it
just
because
you're
new
doesn't
mean
you
don't
have
something
to
say
but
read
the
documents
understand
what
the
issues
are.
Don't
make
it.
Everybody
understand
the
true
fool:
make
sure
that
you
actually
have
something
useful
to
say
when
you
say
it,
but
don't
hesitate
if
you
do
have
something.
A
That's
why
you're
here
is
to
participate
to
be
to
help
us
with
create
better
stand,
and
you
only
do
that
if
you
can,
if
you
speak
up,
there
is
blue
sheets
for
all
sessions,
not
this
one,
because
it's
not
a
formal
standard
session,
but
there's
what
are
call
blue
sheets,
because
they're
blue
paper
and
they're
to
record
who
is
in
the
room
part
of
the
openness
you
need
to
do.
If
you're
a
standards
body,
not
just
the
internet,
centers
body,
every
standards
body
you
need
a
record
of
who
was
participating
in
the
process.
A
We
maintain
mailing
list
archives,
so
you
can
see
who
send
up
mail
some
messages
on
the
mailing
list.
We
maintain
these
archives
of
people
who
are
in
the
room.
So
when
the
blue
sheet
comes
around
to
you,
they'll
be
passed
around
on
a
clipboard
sign
it
as
legibly
as
you
can
and
put
down.
You
know
just
what
it
asks
for,
which
is
your
your
affiliation.
It
does
not
ask
for
your
email
address,
we're
not
going
to
create
a
spam
list,
what
a
record
of
who
is
there?
A
A
A
This
week
to
127
sessions,
actually
just
now's
120,
it's
actually
less
than
that,
and
when
I
made
up
the
slides
or
127
sessions,
there
have
been
at
least
three
cancellations
since
then,
so
there's
three
less
112
are
actually
a
little
less
112
unique
sessions.
Seven
vos
7i
RTF
sessions
there's
a
technical
plenary
on
Monday.
This
is
this
is
from
the
earlier
agenda.
It
is
no
longer
the
ITU
planet.
Penny
pot
is
no
longer
being
the
tie
level.
Summary
of
that
is
no
longer
on
the
agenda.
A
I,
don't
know
what
happened
to
it,
but
it's
no
longer
on
the
agenda
as
of
this
morning,
but
there
is
a
I
piece
tack,
evolution
so
mom
and
then
there's
an
operations
and
administrative
pulmonary
on
Wednesday.
Those
are
interesting
to
go
to
as
newcomers.
The
technical
plenary
is
tries
to
bring
up
some
technical
issues
and
have
a
discussion
about
them
and
trying
to
inform
the
community
about
some
technical
situation.
A
Bruce
Schneier
did
I,
did
a
speech
and
the
technical
plenary
a
relatively
short
after
the
the
after
the
Snowden
revelations
on
how
important
it
was
for
the
IETF
to
try
and
create
a
secure
Internet,
and
that
was
a
pretty
and
encouraged
a
invigorating
speech.
It
was
pretty
carefree,
pretty
directive
that
what
do
you
thought
we
should
do
this
time,
it's
more
on
some
specific
technical
issues
and
ones
which
will
you
find
may
find
pretty
interesting.
The
operations
administration
plenary
is
more
in
line
line.
A
Some
may
generally
answer
okay,
but
not
always,
then,
on
the
on
the
Wednesday
plenary,
this
to
open
mics.
One
with
the
ia
OC,
which
I'll
talk
about
in
a
minute
today,
which
is
the
business
month
of
the
business
managers
for
the
IETF
and
the
others
for
the
IES
G,
which
we've
talked
about
already
and
then
on
Thursday
afternoon,
there's
the
evening,
there's
a
what
is
called
a
bits
and
bytes:
that's
a
people.
A
Companies
buy
tables
to
display
their
wares,
display
their
what
they're
doing
technology
wise
and
to
an
order
to
encourage
people
to
come
this
free
booze
wine
and
beer
and
free
snacks,
and
so
it's
it's
definitely
worth
the
price
of
admission.
Even
if
you
don't
do
anything
but
just
drink
the
beer
and
eat
the
snacks.
A
A
If
you
provides
propose
something
and
ten
percent
of
the
people
in
the
room
think
that's
a
terrible
idea
and
ninety
percent
think
it's
a
great
idea.
It'll
proceed
don't
have
to
satisfy
everybody.
Many
standards
bodies
require
unanimity.
They
require
what
Margaret
Thatcher
once
said
was
that
consensus
is
when
everybody
is
equally
dissatisfied.
Well,
we
don't
require
that
you
can
have
some
very
dissatisfied
people,
but
we're
gonna
move
on
we're
not
going
to
be
stuck
by
a
few
people,
not
thinking
this
is
the
right
path,
we're
supposed
to
listen
to
them,
we're
their
air,
their
ideas.
A
Let's
make
sure
we
understand
why
they're
against
this
particular
proposal,
but
in
the
end,
it's
the
rough
consensus
now
what
level
of
consensus
that
depends
on
the
situation,
if
it's
a
question
of
that
makes
no
technical
difference
its
own
or,
let's
say
the
bid
water
on
a
wire
and
the
argument
is
aesthetics,
then
it
can
be
a
pretty
rough
consensus.
I
mean
75
percent
people
can
say
one
way
in
25
the
other
and
we'll
still
move
ahead.
A
It
also
allows
the
fact
that
if
let's
say
you're,
you're
saying
I'm
not
gonna
support
this,
unless
you
put
in
my
silly
idea
now,
I'm
sure
you
don't
have
any
silly
ideas,
but
let's
just
take
that
for
an
example,
you
have
a
silly
idea.
The
rest
of
the
working
group
can
look
at
and
say
that's
a
silly
idea,
we're
not
going
to
put
it
in
in
other
and
many
standards
bodies.
They'd
have
to
put
it
in
to
satisfy
you
to
get
you
to
agree.
A
We
don't
do
that
which
pens
means
our
Sanders
tend
to
be
a
little
simpler,
a
little
less
geegaws,
a
less
cruft
on
them.
They're,
not
perfect,
but
there's
a
little
less
of
that
and
that
helps
we
don't
do
voting.
We
can't
do
voting.
We
don't
know
how
many
we
don't.
We
know
any
constituency
we're
going
to
be
members,
don't
legally
exist,
don't
have
any
members,
we
do
a
consensus.
We
get
its
understanding
consensus
in
a
room.
We
do
that
a
couple
of
different
ways,
but
I'm
mailing
once
we
do
that
by
just
watching
the
traffic.
A
That's
a
very
hard
thing
to
do
by
the
way,
understanding
consensus
on
a
mailing
this,
because
one
fast
typist
with
a
with
an
agenda
can
destroy.
The
mailing
list
can
wipe
that
out,
but
in
the
room
at
least
you
get
the
general
idea.
So
let's
try
that
here.
I've
got
a
simple
case,
which
is
how
many
people
travel
more
than
300
miles
to
get
here
today
show
of
hands
how
many
people
travel
less
than
300
miles
to
get
here.
So
the
rough
consensus
is
an
IETF
meeting
is
more
than
300
miles
away.
A
That's
how
we
determine
notice,
I,
didn't
count.
It
I
looked
at
the
general
tenure
of
the
room,
there's
somebody's
counting
hands,
then
they're
not
getting
the
point.
The
point
is
whether
it's
obvious
to
just
an
observer,
but
let's
take
a
special
special
case,
so
you
think
it
you
think.
The
right
answer
is
a
your
boss,
who
is
sitting
over
there
thinks
the
right
answer
is
B.
Now.
Are
you
going
to
stick
up
your
hand
with
it
for
a
well?
It
might
be
a
tough
time
back
at
the
old
factory.
A
A
Hum
if
it
was
less
than
300
miles,
so
we
know
as
more
than
300
miles,
but
I
have
no
idea
who's.
Who
said
what
so
your
job
is
safe,
so
you'll
see
that
you'll
see
both
of
those
used.
The
idea
is
to
get
see
whether
there's
an
overwhelming
feeling
in
the
room
or
what
in
in
in
one
direction
or
another,
trying
to
gauge
the
rough
consensus.
A
Not
unanimity,
but
rough
consensus,
whatever
is
gauged
in
the
room,
is
not
final
until
it's
ratified
in
the
mailing
list,
because
there's
a
lot
of
full-time
ITF
participants
who
never
come
to
a
meeting
they
can't
afford
to.
There
are
students.
They've
got
real
day
jobs.
They
can't
afford
to
get
away
from
so
they're
working
entirely
from
from
home
or
from
their
office.
A
Ietf
documents
got
all
ITF.
Documents
are
open,
any
working
group
documents
and
he
or
dozens
resultant
standards,
they're
all
open
anybody
can
copy
them.
Anybody
can
reproduce
them
and
that's
fine.
In
the
early
days
of
the
IETF
company,
called
SR
I
used
to
take
copies
of
rfcs,
packaged
them
up
in
books
and
sell
them
some
libraries,
because
libraries
needed
real
physical,
tangible
things.
They
need
a
real
physical
books.
Nobody
else
did
because
you
could
just
download
it
from
me
from
the
ARPANET.
A
You
can
get
it
for
free
over
the
air
beneath,
but
the
fact
they
were
doing.
That
was
just
fine.
They
were
making
money
on
it,
cool.
That's
fine!
No
problem,
we've
seen
a
few
cases
that
more
recently,
but
they
are
all
open
there
for
everybody
and
whatever
they
want
as
long
as
they
treat
them
entirely.
As
is
you
can
take
any
IDF
document
and
translate
it
into
any
language.
You
want
to
that's
a
blanket
permission.
A
A
That's
that's
not
part
of
it,
but
they're
open
other
than
that
many
standards
bodies,
don't
let
you
get
a
hold
of
the
working
group
documents
because
it's
a
mechanism
by
which
they
are
getting
you
to
join
the
company
join
the
standards
group.
If
you
don't,
if
you're,
not
a
member
the
status
group,
you
can't
see
the
working
document.
So
therefore
you
can't
affect
them.
Well,
our
documents
are
all
open.
Anybody
can
get
them
to
work
on.
A
Rfc's
are
the
archival
publications
once
they're
published
they're,
never
changed
if
you've
got
RSC
XYZ,
it's
the
same
one,
it's
public
that
was
published
ten
years
ago,
the
same
text.
If
we
need
an
update,
we
do
that
by
publishing
another
RFC
with
another
number.
This
is
different
than,
for
example,
Ethernet.
If
you
go
get
the
Ethernet
standard,
the
ethernet
standard
today
is
very
different
than
the
easter
net
standard
was
20
years
ago.
The
documents
call
the
same
thing,
but
it's
a
different
document.
It's
internally
there,
that's
not
the
way
we
work.
It's
the
same
thing.
A
Our
language
is
English.
You
are
permitted
translating
anything
you
want,
of
course,
but
the
official
language
is
English
up
until
very
recently,
everything
was
in
ASCII
plain
ordinary
printable
characters,
no
no
accent
acute
sore
anything
else.
It
was
simply
a
ski
that's
changing
there's!
It's
now
have
now
been
formally
adopted
that
we
can
have
a
fancier
xml-based
standard
which
allows
you
to
get
better
better
things.
A
We
were
getting
a
lot
of
derision
about
being
a
ski
only
because
we
couldn't
draw
pictures
well,
actually
you
can
there's
something
called
ASCII
art
which
you
draw
pictures
using
vertical
bars
and
minus
signs
and
left
arrows,
and
it's
you
got
to
have
really
good
drugs
to
do
some
of
that
art.
Some
of
that
art
is
absolutely
amazing,
but
it's
not
something
that
easy
to
easy
duck
easy
to
do
and
it
doesn't.
It
doesn't
read
all
that.
Well,
so
it's
pretty
hard
to
read
so
now
we
can
have
better
pictures,
but
Dave.
A
The
archival
version
is
still
going
to
be
a
text-based
version
at
the
XML
base,
but
it's
a
text-based
version.
We
can
still
read
the
RFC
as
we
did
in
1969
and
how
many
standards
bodies
can
say
that
when
they
use
word,
which
now
has
three
generations
of
incompatible
generations
of
words
since
then,
so
we're
we
were
tardy
but
we're
consistent
internet
draft.
That's
the
initial
document,
that's
anybody's
thoughts.
There
is
no
barrier
to
publication
for
internet
drafts
other
than
the
jury
greet.
You
agree
to
the
intellectual
property
rules.
A
Everything
else
is
anybody
can
publish
anything
they
want
to.
And
if
you
look
at
the
internet
drafts,
you
will
see
that
anybody
can
publish
anything.
They
want
to
there's
some
very,
very
strange
documents
out
there
ones
that
I
certainly
wouldn't
want
to
put
my
name
on,
but
their
input
to
the
process,
sometimes
they're,
just
informational,
just
straight
information
of
the
community.
It's
a
company
publishing
their
own
internal
technology
standards
and
that's
fine.
A
They
we
don't
claim
any
ownership
over
it
unless
it's
put
into
a
working
group
that
it's,
that
is
simply
your
own
ego,
publishing
it
for
your
own
ego.
You
know
we
get
much
publication
credit,
because
it's
not
even
peer
reviewed.
It's
just
computer
reviewed
to
make
sure
it's
in
the
right
format
and
it
has
the
right
boilerplate
on
it.
All
RFC's
must
pre
exist
as
internet
drafts.
A
That's
in
order
to
make
to
meet
some
of
the
IPR
issues
and
we'll
talk
about
IPR
issues
in
a
bit,
except
for
some,
some
particular
Ayana
based
or
RSC
editor
based
rfcs
and
I'll
show,
but
talk
about
them
in
a
bit.
Internet
drafts
have
names
and
you
can
tell
a
little
bit
about
the
status
of
them
just
by
looking
at
the
file
name.
All
of
the
file
names
begin
with
draft.
A
The
ones
that
are
IETF
working
groups
then
have
IETF
so
draft
IETF,
followed
by
a
working
group
name.
The
ones
that
are
individuals
are
draft.
Your
last,
the
author's
last
name
followed
by
some
other
information
like
what
the
topic
is
ends
with
a
vertebra,
the
version
number
and
a
dot
txt.
So
it's
in
text
form
it's
it's
ASCII
text,
some
examples
of
that
this
is
bgp-4
26.
A
That
is
an
IETF
working
group.
Id
are
the
inter
domain
routing
working
group,
but
there
were
26
versions
of
that
before
it
became
an
RFC.
They
took
a
lot
of
iterations
Brandner,
RFC
39
71,
six
revision
of
a
update
of
the
IETF
intellectual
property
rights
rules
that
I've
written
are
a
B
RFC
format,
requirements
0
3.
This
was
a
document
from
the
IAB
saying.
How
do
we
go
move
away
from
purely
text?
Rfc's?
That's
what
led
us
to
the
the
XML
based
RFC's.
A
What
is
an
RFC
well
used
to
stand
for
requests
for
comments
and
their
very
early
days.
They
were
that
you
publish
it,
and
people
would
comment
on
it.
Then
they'd
published
another
one.
That
role
has
been
taken
up
by
an
internet
drafts,
so
an
RFC
is
an
RFC.
It
is
not
request
for
comments.
It's
a
brand
name,
that's
a
trademark
or
something,
but
it's
we
didn't
trademark
it
because
too
many
other
people
use
it.
But
it
is
not
meant
to
be
a
request
for
comment.
A
So
if
you
see
somebody
like
a
news
news,
article
saying
RC,
mumble,
mumble
mumble,
that's
requests
for
comments
mo
mo
mo
mo
moany.
You
know
they
don't
understand
the
idea.
It's
an
RFC.
That's
all
it
is
it's
by
the
time
it
gets
to
be.
An
RFC
is
too
late
to
comment
on
it,
so
we're
certainly
not
going
to
call
it
request
for
comments.
A
No
over
7,000,
RCS
I,
don't
know
exactly
how
many
you
look
at
the
index.
You
can
see
what
the
biggest
number
is,
but
many
the
rfcs
and
the
early
days
didn't
get
published.
They
were
promised
to
be
published
and
they
would
somebody
promised
to
deliver
something.
So
a
number
was
set
aside
for
them,
but
it
were
never
actually
published.
So
there's
gaps
and
I,
don't
know
how
many
gaps
are
so
there's
over
7,000
RFC's
when
I
started.
My
first
RC
is
RC
1244,
so
it
goes
back
way
to
quite
a
ways.
A
Not
all
our
Caesar
standards,
sometimes
you'll
see
particularly
press
press
releases
from
companies
saying
this
is
an
IETF
RFC,
which
now
it's
a
standard
and
really
all
it
is,
is
somebody's
private
musings,
a
company
published
in
their
internal
documentation
as
RFC
s
but
which
we
do
because
it
can
be
useful
to
the
community.
But
it's
not
a
standard.
There's
been
no
IETF
review
of
it.
A
It
was
just
published,
so
here's
a
bunch
of
the
types
we've
got,
standards
track
obsolete
requirements,
documents,
policies,
poetry,
events,
surf,
writing
of
boating
of
writing
a
poet
the
poem
white
papers
asking
for
information,
corporate
documentation,
history
process
documents
like
the
ITF
standards
process
itself
as
RCS,
and
then
the
one
that
people
like
most
mostly
is
the
April
Fool's
Day,
RFC's,
April,
Fool's,
Day
ours.
Rfc's
are
ones
that
look
like
real
technical
documents,
but
there's
something
wrong.
The
first
one
was
the
telnet
randomly
lose
option.
Turn
this
option
on
it
would
randomly
lose
characters.
A
Jon
Postel
published
this.
It
was
I
forget
who
it
was
from,
but
he
I
don't.
No
one
really
knows
why
I
decided
to
publish
it
other
than
it
was
a
hoot.
It
was
fun
and
it
looked
exactly
like
the
real
thing.
It
just
wasn't
something
you
would
generally
implement,
but
it
was
there
now.
I
have
a
couple
of
actually
three
or
four
in
that
series.
Now
one
of
them
was
the
omniscience
protocol
requirements
and
that's
after
a
US,
senator
Orrin
Hatch
said
that
he
wanted
to
be
able
to
do.
A
A
That's
clunky
I
was
going
in
bed
with
Orrin
Hatch,
which
is
an
image
I
did
not
want
to
have,
but
they
so
Slashdot
thought
I
was
serious
about
it
and
just
trashed
me
something
awful
which
was
perfect,
because
if
you
do
it
right,
they
think
it's
real.
Some
will
get
the
joke.
They'll
look
at
the
they'll,
look
at
the
publication
date
and
get
the
joke,
but
others
won't
and
that's
perfect.
A
There
have
been
implementations,
but
so
you'll
see
that's
coming
up
on
April
4
I've
built
a
plethora
we
may
or
may
not
weave
that's
up
to
the
RFC
editor,
the
independent
stream
editor,
whether
there's
a
whether
there's
a
document,
that's
good
enough
to
be
published
as
this
we
don't
know
yet
this
year,
whether
there's
going
to
be
any
this
slide
is
here
because
many
people
have
heard
that
the
ITF
has
a
three-stage
standards
process.
It
no
longer
does
it's
a
two-stage.
We
have
a.
A
We
have
a
separate
series
of
standards
called
best
practices,
which
is
what
the
the
dot
the
for
example.
The
standards
process
itself
is,
is
documented
as
a
BCP
best
current
practice.
Otherwise
you
have
two-step
standards
track
technical
standards
track.
This
is
what
is
called
a
proposed
standard.
Now
that
doesn't
mean
it's
proposed
as
a
standard.
It
is
a
standard
at
the
proposed
level.
It
is
something
which
is
equivalent
to
the
output
of
most
standards
bodies.
It
is
something
that
is
as
far
as
we
know,
a
completely
right.
All
the
problems
have
been
examined
and
resolved.
A
They
may
not
be
any
implementations
there
might
be,
but
there
may
not
be
and
that's
the
output
of
most
standards
bodies
in
that
form.
You
don't
have
to
actually
build
anything
to
actually
put
out
a
standard,
so
proposed
standard
is
in
that
level.
And
if
you
get
multiple
interoperable
implementations,
then
you
can
move
up
to
the
full
standard
level.
Now
I
say
this
is
running.
This
is
goes
back
to
David,
Clarke's
rough
consensus
and
running
code
running
code.
Is
there
if
you
get
multiple
interoperable
implementations,
starting
from
different
implementations,
then
what
do
you
know?
A
Most
people
look
at
and
say:
oh
well,
that
means
it's
popular
people.
Multiple
people
implemented
it.
That's
actually
not
our
concern.
Our
concern
is
that
you
implementing
it
from
the
specification
and
you
implementing
the
specification,
create
two
implementations
that
leaner
operate.
By
doing
that,
we
have
debug
the
standard.
We've
done
the
text
itself.
The
text
must
have
been
clear
enough
for
both
of
you
to
implement
and
have
an
improper
rate.
So
they're
running
code
is
not
about.
Popularity
is
about
clarity
of
standards,
not
something
that's
a
meteor
parent.
A
We
have
other
air,
RFC's,
informational,
experimental
historical.
You
can
only
tell
the
current
status
by
looking
at
the
index
because
we
don't
change
the
document.
Once
it's
published
it's
published
as
proposed
standard,
it
may
be
replaced
by
another
version
of
the
same
thing
as
proposed
standard.
It
may
be
updated
to
full
standard,
but
you
won't
know
that
by
looking
at
the
anything
but
the
index,
if
you
look
at
the
document
itself,
that
document
was
published.
It
says
standards
track,
doesn't
tell
you
what
level
set
it
doesn't
tell
you.
A
It's
been
superseded
or
declared
historic,
because
it's
because
it
turned
out
to
be
a
really
bad
idea.
We've
got
a
few
of
those.
You
have
to
look
at
the
index
to
tell
that
the
RFC
editor
is
a
function
that
publishes
the
RFC
s.
They
used
to
be
Jon
Postel
an
individual
and
many
of
the
early
RFC's
were
done
by
him
personally.
He's
had
he's
lead
editor
on
a
number
of
them
and
he
did
the
ran.
They
added
a
function
for
many
years.
He
got
a
small
team
of
people
working
with
him.
A
That
function
is
now
been
taken
over
by
the
Secretariat.
Here,
it's
a
few
more
people
in
just
John
John's
John's
gone
now,
unfortunately,
was
a.
We
ran
a
an
open
RFP
to
get
companies
to
bid
on
the
stuff
that
process
and
we've
got
a
group
right
now
as
part
of
the
Secretariat.
That's
doing
it.
So
that's
a
separate
function.
That's
that's
edit,
edits
the
documents
making
sure
they're,
they're,
readable
and
in
the
in
the
right
format
and
then
publishes
them.
A
The
RFC
production
center
takes
input
from
takes
our
FCS
from
internet
drafts
from
various
places
and
publishes
them
in
RFC's.
So
that
kind
of
that
includes
the
IETF
itself.
The
is
iesg,
maybe
the
IRT
F
and
an
independent
stream
editor
we'll
talk
about
that.
Second
yeah,
the
independent
stream
editors
next
John
used
to
publish
our
FCS
from
people
who
sent
him
Internet
drafts
to
publish
one.
Before
we
had
the
internet
drafts.
A
He
published
our
seas
from
text
files
because
he
felt
that
it
was
important
to
not
have
a
filter,
a
controlling
filter,
saying
of
the
IES
IETF
saying
these
are
the
only
things
that
are
allowed
to
be
published.
If
you
want
something
where
you
come
up
with
this
great
idea,
you
bring
it
to
working
group
and
they
say
man.
We
don't
like
that
idea.
A
You
still
want
to
be
able
to
publish
that
as
an
RFC
won't
be
a
standards
Ric,
because
standards
track
comes
through
the
IETF,
but
you
can
publish
it
for
information
say
this
is
my
idea
and
I,
and
in
10
years
I'll
show
you
you'll
come
back,
want
it
whatever
you
want.
Why
ever
you
want
to
do
it?
You
can
do
that.
So
the
our
independent
stream
is
a
safety
valve
on
the
process.
It
overcomes
that
any
kind
of
rejection
that
the
IETF
per
se
does.
A
It
also
provides
a
venue
where
we
can
publish
standards
from
other
bodies
which
we've
done
or
company
corporate,
which
is
not
output
of
an
IETF
working
group
or
IETF
deliberative
process.
It's
simply
information
for
the
community,
so
the
independent
stream
editor
takes
care
of
all
of
that
and
also
takes
care
of
publishing
the
parts
of
the
April
Fool's
Day
RCS
see
all
right.
A
This
is
a
pic
go
for
the
ITF
submissions
on
the
on
the
RFC,
so
it
goes
from
the
working
group
or
the
individual
to
the
is
G
last
call
to
the
Internet
to
the
IETF
community.
But
the
is
G
has
concerns
it
bounces
it
back
to
the
working
group.
Otherwise
it
published
pushes
it
off
to
the
publication
series
non
I,
80
F,
which
is
the
independent
stream,
goes
from
the
individual
to
the
independent
stream.
Editor
industry
stream.
A
Editor
has
a
review
board
that
it
that
he
uses
to
sound
sound
out
whether
this
document
is
right
for
the
IETF
and
well
right
for
the
internet
community,
and
if
you
got
a
recipe
for
brownies,
even
if
they're,
very
special
brownies,
that's
not
something
we're
going
to
publish
as
an
RFC.
You
got
a
yeah,
it's
gotta
be
technically
relevant
to
the
to
the
Internet.
A
The
IAB
and
I
RTF
have
their
own
ways
to
do
this,
so
that
set
of
the
plumbing.
So
now,
why
are
we
here?
What
are
we
doing?
What
what's
our
role
when
I
first
started
working
in
the
IETF
in
the
early
night
in
early
1990,
the
saying
was
above
the
wire
and
below
the
application.
We
didn't
define
wires
and
we
didn't
define
what
was
on
the
screen.
We
defined
the
stuff
in
between
the
media
work.
Well,
it's
getting
a
little
harder
to
do
that,
particularly
at
the
bottom
layer.
A
So
what's
a
wire,
let's
say:
Ethernet,
that's
a
nice
thing,
except
we
can
now
simulate
Ethernet
over
IP.
We
can
simulate
frame
relay
over
IP
IP.
We
can
frame
the
ATM
over
IP.
We
can
simulate
SONET
over
IP,
so
you
can
have
IP
over
ATM
over
sonnet
over
IP
over
MPLS
over
IP.
So
what's
the
wire
anymore,
so
it
gets
a
little
fuzzy
and
direction.
A
It
is
also
because
we're
bottoms
up
you
come
you
come
in
and
you
have
this
great
idea.
You
come
in
with
a
wonderful
new
protocol.
Let's
say
it's
for
streaming,
video
of
some
kind
or
let's
say
more
pointedly,
it's
telephony,
and
it's
a
way
that
you
can
run
cell
cellular
telephones
from
Wi-Fi
base
stations
without
modifying
the
cellular
telephones.
It's
gonna
be
a
little
hard
to
do.
A
But
let's
say
you
come
up
with
that
that
steps
on
a
number
of
different
standards
bodies,
but
if
you
bring
it
into
the
IETF
and
you've,
got
enough
people
liking
it.
We
might
move
ahead
with
it,
which
is
going
to
get
really
touchy
and
when
our
relations
with
some
of
the
other
standards
bodies,
so
many
other
standards
bodies,
look
at
and
say,
wait
a
second
you're
stepping
on
our
turf,
and
you
shouldn't
be
doing
that.
But
the
other
way
is
true.
A
Everything
is
moving
to
the
Internet,
all
telephony,
all
video
stream,
all
video
distribution,
both
distribution,
everything's,
moving
via
Internet,
everything's
bits
and
everything's
bits
running
over
IP.
That's
just
the
way.
It
is
that's
not
the
way
people
thought
it
was
going
to
be,
but
that's
the
way
it
is
so
those
spot.
Those
traditional
standards,
bodies
that
were
let's
say,
working
in
telephony,
no
longer
have
a
the
field
to
themselves.
They
have
this
other
upstart,
the
IETF,
who
is
doing
stuff
in
the
same
space
and
doing
with
a
different
philosophy.
A
So
other
stos
look
at
and
say
well
this
internet
thing
we
can
use
that
we
have
to
because
they
don't
worry,
everybody
else
is
going
to
we're.
Gonna
have
to
use
it,
but
we
could
make
it
better.
By
doing
this,
we
can
make
a
better
for
voice
or
better
for
video.
We
can
make
it
different
and
make
it
better
one
of
the
things
about
the
Internet
Protocol
suite.
Is
it's
not
very
good
for
any
one
thing,
it's
designed
to
be
flexible.
It
is
not
designed
to
be
best,
it's
not
the
best
way
to
do
telephony.
A
It's
not
the
best
way
to
do
video.
It's
not
there.
Even
the
best
way
to
do
books.
It's
a
way
you
can
do
all
of
those
things.
So
when
a
standard
body
who's
only
focuses
on
voice,
looks
at
the
internet,
they
say
it's
not
good
enough
for
voice.
It's
not
high
enough
quality,
or
it's
got
too
many
gaps
in
it
or
something
like
that.
They
want
to
improve
it.
A
So
many
of
them
have
taken
Internet
protocols
and
changed
them
on
their
own
without
understanding
wonder
the
basis
of
it,
and
so
we
get
to
fights
with
other
standards
bodies
from
time
to
time
about
that-
and
this
particular
one
with
the
ITU
was
pretty
fierce
for
a
number
of
years
where
the
ITU
had
taken
MPLS
one
of
the
IETF
technologies
and
modified
it
to
make
it
better
for
transporting
traditional
telephony,
and
they
did
it
in
a
way
that
we
thought
was
broken.
A
plus
factor
wouldn't
interoperate,
with
what
we've
got.
A
So
we
had
a
significant
fight
back
and
forth
all
kinds
of
politics
involved,
and
then
it's
resolved
pretty.
Well,
that's
we're
now
working
together
pretty
well
and
it's
the
IETF
is
responsible
for
modifying
its
own
protocols.
The
ITU
is
not
modifying
our
protocols
anymore,
but
it
was
a
hard
time
getting
here
and
it
will
happen
again
to
other
with
other
standards
bodies.
A
So
here's
the
top-level
view
of
the
whole
constellation
here
stuff
in
green
as
the
IETF
and
the
other
stuff
around
it
is
around
it.
Internet
Society
is
very
important
to
us.
The
Internet
Society
was
formed
significantly
after
the
IETF
was
the
Internet
Society
was
formed
1992
the
IETF
was
formed
in
1986
and
I.
Don't
a
society
was
formed
to
be
the
legal
home
for
the
IETF
and
to
continue
developing
country
workshops
that
a
fellow,
a
malarial
and
Weber
had
done
very
important.
A
Things
promote
the
open
development
evolution
use
of
the
internet
for
a
benefit
of
all
people
throughout
the
world.
There's
no
goal
there,
I
mean
that's
that's
as
goals,
go,
that's
a
pretty
good
goal
and
do
explain
it
to
the
world
the
the
world,
but
the
IETF
looked
at
the
Internet
Society
and
said
where'd
you
come
from
and
it
wasn't
until
1996
that
the
IETF,
through
a
working
group
process,
agreed
to
come
under
the
legal
umbrella.
The
period
of
four
years
after
the
Internet
Society
was
formed.
A
The
IETF
just
said
we
don't
care
you're
you're,
offering
us
an
umbrella.
We
don't
think
we
need.
Well,
then
some
rain
happened
and
there
was
it
it
got.
There
was
reason
to
move
in,
but
we
went
through
a
deliberative
process
in
a
working
group
within
the
IETF
to
make
the
decision
to
become
part
of
to
come
under
the
Internet
society's
umbrella
and
it's
under
the
umbrella.
Now
all
of
all
the
money
goes
through
the
Internet
Society,
where
it's
legal
spokesperson,
the
legal
spokesperson
for
the
IETF
is
the
Internet
Society.
A
The
eye
sock
president,
is
on
the
IAB
calls
and
although
Dutch
doesn't
have
any
extra
power
there,
as
part
of
the
part
of
the
word
work
there,
the
IAB
is
chartered
by
the
Internet
Society
and
the
Internet
Society
president
picks
the
chair
of
the
NomCom.
So
it's
and
the
IETF
picks
board
members
for
the
Internet
Society.
It's
definitely
a
worthwhile
group
to
associate
with
if
you're
interested
they
you
can
join
as
an
individual.
Your
company
can
join
as
an
individual.
You
can
do
it
for
money
or
for
free
for
a
company.
A
A
Irt
F
is
another
thing
that
wasn't
quite
in
that
green
bucket
there
of
green
section
on
the
previous
page.
This
is
at
the
time
the
IETF
was
formed.
The
internet
research
task
force
was
formed.
Also
there
were
a
number
of
other
research
groups.
At
the
same
time,
these
the
only
two
that
are
survived
research
task
force
is
looking
at
things
that
aren't
ready
for
primetime
they're,
not
ready
for
standardization.
When
we
had
an
anti-spam
research
group,
we
know
we
don't
know
how
to
do
anti-spam.
A
So
that's
research,
maybe
someday
it'll,
be
something
good
to
actually
standardize
on,
but
there's
not
today.
These
are
the
research
groups
of
crypto
forum,
delay-tolerant.
Networking,
though
actually
I
think
that's
been
moved,
there's
also
a
working
group
in
that
space
congestion
control.
Looking
at
the
future,
the
bottom
one
is
particularly
interesting
proposed
protocol
considerations.
Research
group:
this
is
to
write
up
what
are
the
kinds
of
things
if
you're
concerned
with
human
rights,
what
are
the
kinds
of
things
that
you
should
or
should
not
do
in
a
protocol?
A
That
would
make
it
easier
or
harder
for
a
despot
organ
despot,
a
government
to
in
to
invade
human
rights.
It's
very
early.
It's
a
proposed
worry
is
basically
a
Boff.
That's
not.
It
has
not
has
not
been
chartered
as
though,
as
a
research
group,
yet
they'll
be
meeting
this
week
and
they
might.
There
might
be
some
very
interesting
things
that
they
did.
A
They
read
IR
TF
today,
RFC
a
few
years
ago
on
privacy
considerations
on
what
you
should
do
in
your
protocols
to
make
them
more
privacy
protecting-
and
this
is
more
on
the
Human
Rights,
the
area
of
free
speech
and
the
like.
Maybe
I,
talked
about
the
IV.
Already
it
is
the
the
Purdue
that
does
the
all
the
architecture
guidance.
It
also
does
liaisons
outs
about
groups
outside
the
IETF.
Why
so
liaison
to
other
standards?
Development
organizations
formally
done
through
them?
A
They
are.
The
IAB
does
help
review
working
group
proposals
and
things
like
that.
They
undertake
specific
activities.
These
workshops
they
also
organize
their
work
into
these
programs,
where
it's
trying
to
make
it
so
that,
as
I
use,
gmia
be
members
turnover,
you
can
still
get
those
programs.
These
particular
activities
continuing
on
there's
a
number
of
them.
You
can
look
at
under
under
the
IAB
website
on
that
the
I
Anna,
the
I
Anna,
is
not
part
of
the
ITF,
but
we're
dependent
on
it,
but
so
are
others
used
to
be
once
a
very
early
on.
A
It
became
clear
that
when
you
developed
a
standard
for
a
protocol
that
runs
over
in
those
days,
the
ARPANET
there's
a
field
in
that
protocol
in
the
in
the
in
the
underlying
protocol,
it
says
14
or
25
that
tells
the
higher-level
protocols
what
application
is
being
run
here.
It
doesn't
make
any
difference
at
all
what
that
value
is
as
long
as
every
degrees
in
it.
So
you
have
to
have
a
record
of
place
that
that
those
are
recorded.
The
internet
assigned
numbers
authority
was
the
place
that
was
reported.
A
That
was
a
fellow
named
Jon
Postel,
the
same
jon
postel
that
was
doing
the
RCS.
He
was
the
artist
that
he
was
the
Ayana
he
individually
was
the
Ayana.
He
got
some
people
to
help
him
later
on,
but
starting
that
was
the
idea
and
he
recorded
what
the
protocol
parameters
were.
25
means
email,
22
means
SSH,
he
recorded
those
and
put
them
in
a
little
table.
A
On
my
desk
in
the
psychology
department
for
20
years,
you
want
to
use
a
human
friendly
name,
my
domain
name,
which
can
be
translated
into
an
IP
address
when
it's
needed
to
be.
Somebody
had
to
record
who
was
running
the
name
servers
for
those
domains,
so
you
had
a
root
domain
and
in
there
you
had
subdomains
of
dot-org
com
dot.
A
Frus.
Somebody
had
to
run
that
little
database
to
say
what
the
IP
addresses
were
for
the
name
servers
for
those
underlying
subdomains.
Yes,
John
did
that
too.
John
did
all
of
those
things.
He
developed
a
little
group
at
ISI
to
do
that,
and
then
he
decided
he
needed
to
spring
that
out
to
the
the
general
world
not
just
be
part
of
a
US
government
funded
thing.
It
was
more
important.
A
The
internet
was
too
big
for
that,
so
he
proposed
to
what
he
called
institutionalize
it
and
came
up
with
a
proposal
which
has
been
adopted
as
ICANN.
The
Internet
Corporation
for
Assigned
Names
and
numbers
that
function
in
within
ICANN,
which
is
doing
the
protocol
parameters
is
being
done
at
the
request
of
the
IETF
through
an
a
memorandum
of
understanding.
A
All
of
the
every
RFC
has
got
to
have
an
internet
drafts,
as
proposed
to
be
an
RFC,
has
to
have
some
section
in
there
saying
whether
the
I
Ana
has
to
pay
attention
to
this
or
not.
Do
you
have
to
assign
parameters
and
things
like
that?
The
IETF
is
no
longer
in
the
part
dealing
with
a
part
of
the
IANA
which
deals
IP
addresses
or
domain
names,
as
they're
done
elsewhere.
The
regional
registries
deal
with
IP
addresses
and
the
domain
name
community
deals
with
main
names.
The
IETF
just
deals
with
the
underlying
technology.
A
So
ITF
management,
all
volunteers,
the
IETF
chair
area,
directors,
working
group,
chairs,
document,
editors,
IV,
members,
IASE,
members,
they're,
all
volunteers,
none
of
them
are
paid.
Even
their
expenses
are
not
paid
by
the
IETF
they
are
paid
for
by
their
the
companies
they
work
for
or
by
themselves.
It's
it's
not
we
don't.
We
don't
do
any
anything
which
is
which
is
supported
from
the
IETF
side.
It's
a
lot
of
work.
The
Area
Director
job
is
between
1/2
and
3/4
time
there
having
been
one
for
a
long
time.
It
certainly
is
IV
jobs.
A
Third
time,
my
chair
is
full
time.
The
last
few
chairs
have
been
full
of
their
voice.
This
is
their
job
being
the
ietf
chair.
The
companies
I
work
for
give
them
up
to
the
ietf
for
that
function
for
a
number
of
years
to
perform
that
function,
because
those
companies
think
it's
worthwhile
either
PR
reasons.
So
we
got
the
chair
or
because
they
just
think
it's.
The
internet
attack.
A
We
do
pay
for
the
RFC
publish
a
publication,
we
do
pay
for
the
Secretariat,
speaking
of
Secretariat
they're,
the
people
and
their
purple
shirts
running
around
the
person
who
gave
out
the
handouts
to
you
as
you
came
in
works
for
the
Secretariat,
and
that
was
done
in
an
RFP
and
company
called
AMS
is
the
the
company
that
got
the
jump.
The
job
I
mentioned
the
I
asseh
this.
This
is
why
I
mentioned
the
IOC.
My
assay
is
the
IETF
administrative
support
activity.
A
Well,
it's
basically
that's
the
home
for
the
IOC,
which
is
worries
about
the
ITF
budgets
and
things
like
that.
The
budget
actually
goes
to
the
Internet
Society,
but
they
are
the
ISO
and
the
IOC
produce.
That
budget
propose
that
budget
to
the
Internet
Society
Internet
Society
decides
to
work
the
fund
funds
it
most
of
the
money,
comes
from
meeting
fees.
Actually,
from
a
very
long
time
ago,
people
coming
to
IETF
meetings
paid
if
it's
paid
some
money
starting
in
94
or
something
like
that.
It
started
to
pay
money.
A
A
A
good
chunk
comes
from
the
Internet
Society,
a
third
or
so
of
the
budget
comes
from
Internet
Society
and
that
mostly
comes
from
the
PIR
of
the
registry
that
runs
org
and
from
Corp
and
from
organizational
memberships
in
the
Internet
Society
combination
of
those
things
the
is
a--
just
like
the
internet.
Society
has
no
authority
of
what
so,
over
ever
over
the
tech.
A
The
is
includes
the
IETF
Administrator
director,
that's
Ray,
that's
a
picture
array
right.
There
is
the
one
employee
of
the
IETF
directly
and
it's
actually
a
direct
employee
of
the
Internet
Society
buddy
is
assigned
to
the
IETF
and
then
the
IEEE,
the
ITF
administrator
Oversight
Committee,
which
an
eight-member
body
I'm
on
it
right
at
the
moment.
A
The
rights
to
those
are
put
in
the
trust
any
code
that
the
IETF
contracts
for
gets
put
into
the
rights
to
that
is
put
into
the
trust,
the
trademark,
the
domain
names.
All
of
those
things
are
put
into
the
IETF
trust.
It
is
not
a
patent
pool,
it
is
simply
a
place
to
hold
intellectual
property
rights
and
you
can,
in
all
of
the
trust
documents
are
public,
of
course.
So
how
do
we
select
the
moment?
The
members,
the
isg,
the
IAB,
the
ioc
etc?
Well,
we
do
it
with
participation
from
the
community.
A
We
don't
vote
because
we
don't
have
a.
We
don't
have
any
way,
knowing
what
the
constituency
is.
So
we
do
is
ask
for
volunteers
from
the
community
to
form
a
nominations
committee:
it's
called
nominations,
but
it's
all
stronger
than
so
there's
more
of
a
selection
committee,
the
you
put
you
if
you
volunteer.
A
Well,
you
can't
yet
you
have
to
have
actually
been
to
three
of
the
last
five
face-to-face
meetings
in
order
to
be
able
to
be
qualified
to
the
author
ID
to
volunteer
for
the
nominations
committee,
but
take
all
of
the
nominees
and
denominations
that
volunteered
select
a
ten
of
them
to
form
a
nama
to
nominations.
Committee,
a
chair,
that's
selected
by
the
IETF,
the
Internet
Society
president
and
they
go
off
and
they
figure
out
who
should
be
taking
these
roles?
A
They're,
the
they're,
not
the
final
say
but
they're
the
semi-finals
say
once
upon
a
time
at
the
early
early
selection
process
was
the
names
the
volunteers
were
written
on
little
pieces
of
paper
and
dropped
into
a
hat
and
taken
to
a
local
priest
who
then
picked
out
the
members
of
the
committee,
but
that
was
deemed
as
I.
Don't
know,
bias
a
ball
or
something
we
have
a
ATF
member
who
decided
that
that
was
simply
not
good
enough.
A
They
came
over
there
RFC,
and
this
is
the
latest
iteration
of
it
3797,
which
is
way
over
the
top.
But
it's
sort
of
indicative
of
the
quality
work,
though
what
you
do,
if
you're
the
I
when
the
NomCom
chair
is
you
say
on
July
on
July
7th
I'm,
going
to
pick
these
people
so
on
July,
7th
I'm,
going
to
take
the
quantity
of
stock
sold
on
these,
these
nine
stocks
and
this
the
value
they
dot.
A
A
It
done
and
that's
IETF
chair,
the
is
G
members
really
I
entered
it
at
the
area,
directors,
in
other
words
IAB
members
and
the
IOC
members
the
and
each
one
of
this
is
approved
by
a
higher
level
body,
so
that
the
is
G
nominees
are
approved
by
the
IAB.
The
IAB
upon
nominees
are
approved
by
the
Internet
Society
Board
of
Trustees.
A
Okay,
a
chair
that
chair
I
think
a
chair
is
approved
by
the
IAB,
so
it's
a
and
and
yes,
those
bodies
can
reject
one
or
more
members
of
that
slate,
and
it's
at
has
happened
usually
because
the
somebody
in
the
confirming
body
knows
something
that
something
on
the
nominations
committee
didn't
know
and
they
send
it
back
for
more
work.
See
dots
is
one.
That's
missing
the
dots,
the
colored
dots
on
people's
badges,
indicate
activities.
A
A
Most
people
are
nice,
but
usually
these
people
have
signed
up
to
be
nice.
So
if
you
ask
questions
of
hurt
somebody
with
a
smiley,
then
if
you
don't
get
Smiley's
back
then
something
wrong
they
shouldn't
have
signed
up.
We
have
an
appeals
process.
We
have
to
have
an
appeals
process.
That's
just
part
of
the
rules
said
if
you're
a
standards
body,
you
have
to
have
an
appeals
process.
A
So
if
you
bring
in
your
technology-
and
the
working
group
chair
says
yeah
so
we'll
use
somebody
else's,
you
can
say,
but
the
consensus
was
with
me
and
you
can
appeal
that
decision.
Any
decision
from
the
working
group
chair
on
up
you
can
appeal
now.
This
is
not
a
vacuous
thing.
There's
not
that
many
Appeals,
but
of
the
appeals
that
have
been
somewhere
between
a
quarter
and
a
third
have
been
successful
where
the
isg
or
a
working
group
chair
has
talked
been
told.
They
did
the
wrong
thing
and
they
have
to
go
redo
it.
A
So
this
it's
a
worthwhile
thing
to
do.
If
you
feel
wronged
I
hope
you
don't
feel
wronged
on
your
first
meeting.
That
would
be.
That
would
be
sad.
So
let's
try
it.
Let's
try
not
do
that,
but
it's
that
the
appeals
process
is
there.
If
only
if
the
appeal
is
of
the
process
itself,
you
think
that
the
mechanism
by
which
the
IETF
functions
the
standards
dot
the
document
RFC
2026
and
the
other
documents,
which
is
the
process
documents.
A
A
You
will
hear
applause
when
the
I.
If
the
IAB
announces
that
there
have
not
been
appeals,
there's
that's
an
applause
situation.
That's
a
good
thing!
Well,
we've
had
we've
had
appeals.
We've
had
Appeals
for
actions
we
had
of
Appeals
for
non
actions.
Appeal
was
that
the
isg
did
not
decide
about
a
document.
What
they've
been
in
front
of
them
for
months
that
sort
of
thing
intellectual
progress,
the
last
big
topic
I'm
going
to
talk
about
it's
a
big
area,
there's
two
different
subcomponents
once
copy
copy
right
and
the
other
is
patents.
A
Copyrights
is
pretty
straightforward.
You
write
you
write
that
pinochle
playing
protocol
all
right
and
your
net
draft.
When
you
publish
that
internet
draft,
you
were
giving
the
IETF
permission
to
publish
the
draft
logical.
Otherwise,
why
send
it
to
us?
If
you're
not
going
to
give
us
permission
to
publish
it,
then
it
doesn't
make
any
sense.
They
wouldn't
you
wouldn't
have
sent
it,
but
you
also
give
us
permission
to
publish
it
as
an
RFC.
A
Unless
you
explicitly
say
you
can't
do
that
and
that's
there's
a
mechanism
by
which
you
say
that,
and
you
give
us
permission
to
modify
it
in
within
the
IETF
standards
process.
In
fact,
you
give
us
permission
to
have
it
modified
anywhere,
but
it's
only
within
the
IETF
standards
process
that
we
exercise
that
those
are
the
things
you
give
us
a
none
revocable
right
to
publish
a
none,
revocable
right
to
use
it
in
the
standard
process.
A
Unless
you
specifically
disclaim
that
once
you've
done
that,
though
it's
yours,
you
can
take
that
protocol
and
turn
it
into
the
great
American
or
a
great
SWA
lenient
novel,
and
that's
just
fine.
You
can
take
it
to
a
different
standards
body
and
that's
just
fine.
You
retain
all
rights
other
than
the
non
exclusive
publication
rights
and
modification
rights
you
give
to
the
IETF
they're
all
yours,
so
we're
not
some
standards
bodies
require
you
to
hand
over
full
copyright
to
the
standards
body.
A
A
The
worst-case
scenario
is
somebody's
in
that
standards
body,
watching
the
progress
of
a
standard
directing
that
process
in
such
a
way
that
the
resulting
standard
overlaps
one
of
their
patents,
and
they
don't
tell
you
about
it.
This
is
not
as
has
happened,
it's
happened
in
the
IETF,
it's
happened
elsewhere.
It's
happened,
many
other
places
more
frequently
elsewhere
than
in
the
idea,
but
it's
has
happened
here.
A
That's
called
a
submarine
patent
and
what
happens
is
when
they
so
many
standard
gets
published,
people
start
implementing
it
and
then
sometime
down
the
road
they
start
getting
sued
sued
for
implementing
the
standard,
and
that's
a
really
ugly
thing.
So
we
want
to
avoid
that.
So
we
have
a
disclosure
requirement
if
you
put
in
an
internet
draft
that
has
something
in
it
that,
if
implemented,
would
infringe
your
patent
I
buy
your
here,
I
mean
a
patent
that
you
own
personally
or
that
your
company,
that's
sponsoring
you
owns,
or
at
least
can
benefit
from.
A
So
you
may
have
a
license
to
use
this
patent
license
to
sue
over
the
patent.
You've
got
any
of
those
things,
then
you
must
disclose
the
fact
that
you've
got
the
patent
when
you
put
in
the
internet
draft.
That's
the
first
thing.
Second
thing
is,
if
you're
in
sitting
in
a
working
group-
and
you
see
somebody
else-
put
in
a
an
internet
draft
that
has
your
intellectual
property
rights
in
it,
a
patent
that
you
have
in
it
and
you
participate
in
any
way
in
the
activity
of
that
working
group.
A
Then
you
must
disclose
the
fact
you've
got
the
patent.
We
wrote
it
just
that
way
so
that,
if
you're
in
the
room-
and
you
see
that
somebody
else
is
put
in
a
an
internet
draft-
that
has
your
patented
material
in
there
or
more
particularly-
has
material
in
it
from
a
patent
application
and
your
company
does
not
let
you
talk
about
patent
applications.
You
can
just
remain
silent.
A
Don't
do
anything,
don't
talk
to
anybody,
don't
don't
participate
in
the
activities
of
the
working
group
and
you're
okay,
but
if
you
do
anything
that
would
push
that
technology
one
way
or
the
other
either
for
your
patent
or
against
your
patent,
then
you
have
to
tell
us
about
it
and
then
the
working
group
is
up
it's
up
to
the
working
group
to
decide
what
to
do
with
that.
We
specifically
do
not
have
the
rule
that
we
can't
standardize
something
that
has
patents.
We
decided
not
to
invoke
that
rule.
A
There
are
other
groups
like
w3c
which
have
such
rules.
We
do
not.
We
leave
it
up
to
the
working
group.
There
are
working
groups
where
every
single
proposed
technology
to
cover
a
particular
problem
has
patent
on
the
robust
header
compression
for
IP
over
cellphones.
Every
technology
had
a
patent
on
it,
so
the
working
group
can
look
at
it
and
decide.
The
technology
is
important
enough.
Then
we're
going
to
pick
one
of
these
things
we're
going
to
move
with
it.
A
A
No
well,
when
you
signed
up
for
they
come
to
this
meeting.
You
saw
what
was
called
the
note
well
note-
and
this
is
this-
is
the
this
is
a
note
that
a
couple
lawyers
and
I
wrote
back
late
90s,
it's
been
distorted
from
its
original
purpose.
The
original
purpose
was
to
tell
you
what
a
contribution
was.
A
That's
what
the
no
well
note
says
there
is
a
mentoring
program.
If
you
want
to
be
a
mentor
EE,
you
contact
the
mentoring
program,
the
ITF
regulars
will
take
you
under
their
wing
and
help
you
see
what's
see.
What's
going
on,
that's
told
this
is
a
remarkably
effective
solution
that
this
this
works.
Remarkably,
there
are
other
tutorials
today,
you're
in
the
first
one,
so
you're,
not
in
the
second
one,
because
that
happens.
A
At
the
same
time,
there
is
a
yang
advice:
that's
a
language
for
writing
management
standards
for
management
protocols
for
configuring
routers
and
things
like
that.
As
a
yang.
It's
a
transport
area
overview
at
starting
at
3
o'clock.
There's
a
document
as
a
instructions
on
writing.
Our
FCS
internet
draft
language
internet
drafts
in
this
new
markup
language-
if
you
don't
want
to
use
what
I
use,
is
the
old
ascii-based
thing?
A
There
is
a
optional
newcomers
dinner,
there's
an
informal
dinner
to
chat
about
how
your
your
experience
is
to
date.
Of
course,
that's
Monday
afternoon,
so
your
experience
is
limited,
but
you're
welcome
to
do
it.
If
you
want
to
do
that,
then
you
should
go
to.
You
should
send
mail
to
Maddie
at
that
email
of
that
email
address
and
say
you
would
like
to
come
there,
heaven
or
they'll,
be
meeting
at
the
IETF
registration
desk
at
7:30
and
wandering
over
to
some
local,
reasonably
priced
restaurant.
A
What's
next
you're
here
give
us
give
us
the
benefit
of
your
knowledge.
Your
your
insight
join
mailing
list.
Go
talk,
look
read
the
documents
understand.
What's
going
on
comment
on
them,
help
help
us
become
part
of
us,
become
part
of
art,
of
making
this
internet
better.
Don't
standard,
don't
settle
for
a
standard,
second-rate
stuff,
make
sure
we're
doing
the
right
thing,
make
sure
rate
making
the
best
possible
technologies.
A
And
finally,
this
is
not
in
your
handouts,
because
I
was
told
about
it
about
an
hour
before
the
session
they've
now
set
up
a
doodle
poll
or
a
Survey
Monkey
poll
asking
what
you
thought
of
this
session.
They
can
do
that
after
each
of
the
tutorials,
so
that's
Survey,
Monkey,
/
s,
/
92,
new
comers
and
this
little
five
five
question
survey
where
they're
asking
you,
what
do
you
want
and
with
that
do
you
have
any
questions?
If
you
have
questions
is
the
microphone.
A
Richard
hansen,
BBN,
why
are
there
that's
a
good
start?
You
got
your
name
BBN.
Why
are
there
two
microphones?
It's
usually
just
a
split.
They
split
the
audience
up,
because
people
don't
want
to
come
over
come
down
to
the
front.
The
microphone
is
used
in
two
words:
one
is
it
so
everybody
can
hear
you
and
the
other.
It's
actually
a
pacing
mechanism
so
that
we
don't
have
everybody
talking
at
once.
We
get.
If
you
didn't
have
the
microphones
and
people
would
talk
to
each
other
across
the
room.
Well,
the
microphones.
A
That's
a
very
good
question.
That's
very
quite
the
question
is
whether
we
have
a
mecca
for
interoperability
dot,
particularly
there's
a
lot
of
documentation
on
how
people
have
done
in
the
past.
The
original
concept
is
that
you
should
check
each
one
of
the
features
of
the
specification
to
be
sure
that
it
was
implemented
and
it
interoperates
and
in
fact,
you're
supposed
to
remove
any
feature
which
was
not
implemented
or
does
not
interrupt
before
you
move
the
technology
along.
A
So,
for
example,
we
had
OSPF
was
a
as
a
routing
protocol
and
the
first
version
of
OSPF
had
seven
different
priorities.
Nobody
won't
only
one
company
ever
implemented
the
priorities.
So
when,
when
OSPF
moved
along
the
standards
track,
the
priority
the
different
priorities
were
removed.
So
it's
all
one
priority
now
so,
but
that
was
done
pretty
much
you
just
the
individual
who
decided
who
is
going
to
Shepherd.
A
A
The
open
net
foundation
that
would
be
a
question
for
the
IAB
I,
don't
recognize
that
we
have
a
liaison
with
them.
But
that's
it's
a
question
for
the
IAB,
and
so
the
IAB
will
be
that
it'll
be
microphone
for
them
or
open
mic
session
for
them
on
tomorrow
after
tomorrow
evening,
and
you
could
ask
that
they're,
okay
well,
I
think
we're
supposed
to
vacate
the
room
about
now.
So
thank
you
all.