The fragmented nature and asymmetry of local and remote file access and network access, combined with the current lack of robust authenticity and privacy, hamstrings the current internet. The collection of disjoint and often ad-hoc technologies currently in use are at least partially responsible for the magnitude and potency of the plagues besetting the information economy, of which spam and email borne virii are canonical examples. The proposed replacement for the internet, Internet Protocol Version 6 (IPv6), does little to tackle these underlying issues, instead concentrating on addressing the technical issues of a decade ago. This paper introduces CANE, a Content Addressed Network Environment, and compares it against current internet and related technologies. Specifically, CANE presents a simple computing environment in which location is abstracted away in favour of identity, and trust is explicitly defined. Identity is cryptographically verified and yet remains pervasively open in nature. It is argued that this approach is capable of being generalised such that file storage and network access can be unified and subsequently combined with human interfaces to result in a Unified Theory of Access, which addresses many of the significant problems besetting the internet community of the early 21st century.
arXiv:0710.5006v1 [cs.NI] 26 Oct 2007
CANE:
The
Con
ten
t
A
ddressed
Net
w
ork
En
vironmen
t
P
aul
Gardner-Stephen
(gardners info
eng. inders.edu.au)
17th
No
v
em
b
er
2021
Abstra t
The
fragmen
ted
nature
and
asymmetry
of
lo
al
and
remote
le
a ess
and
net
w
ork
a ess,
om
bined
with
the
urren
t
la
k
of
robust
authen
ti it
y
and
priv
a y
,
hamstrings
the
urren
t
in
ternet.
The
olle tion
of
disjoin
t
and
often
ad-ho
te
hnologies
urren
tly
in
use
are
at
least
partially
re-
sp
onsible
for
the
magnitude
and
p
oten y
of
the
plagues
b
esetting
the
information
e onom
y
,
of
whi
h
spam
and
email
b
orne
virii
are
anoni al
examples.
The
prop
osed
repla emen
t
for
the
in
ternet,
In
ternet
Proto
ol
V
ersion
6
(IPv6[13
℄),
do
es
little
to
ta
kle
these
underlying
issues,
instead
on en
trating
on
addressing
the
te
hni al
issues
of
a
de ade
ago.
This
pap
er
in
tro
du es
CANE,
a
Con
ten
t
A
ddressed
Net
w
ork
En
viron-
men
t,
and
ompares
it
against
urren
t
in
ternet
and
related
te
hnologies.
Sp
e i ally
,
CANE
presen
ts
a
simple
omputing
en
vironmen
t
in
whi
h
lo
ation
is
abstra ted
a
w
a
y
in
fa
v
our
of
iden
tit
y
,
and
trust
is
expli itly
de ned.
Iden
tit
y
is
ryptographi ally
v
eri ed
and
y
et
remains
p
erv
asiv
ely
op
en
in
nature.
It
is
argued
that
this
approa
h
is
apable
of
b
eing
gen-
eralised
su
h
that
le
storage
and
net
w
ork
a ess
an
b
e
uni ed
and
sub-
sequen
tly
om
bined
with
h
uman
in
terfa es
to
result
in
a
Uni ed
Theory
of
A
ess,
whi
h
addresses
man
y
of
the
signi an
t
problems
b
esetting
the
in
ternet
omm
unit
y
of
the
early
21st
en
tury
.
1
In
tro
du tion
What
are
the
real
problems
with
the
in
ternet
to
da
y?
Is
the
in
ternet
user
base
most
a utely
pressed
b
y
the
imp
ending
exhaustion
of
the
IPv4[1
℄
address
spa e?
Is
it
p
erhaps
that
the
urren
t
In
ternet
Proto
ol
V
ersion
4
(IPv4)
has
trouble
with
sending
more
than
a
few
giga-b
ytes
in
a
single
session?
Ma
yb
e
it
is
that
65,536
p
orts
p
er
host
just
isn’t
enough?
Or
is
it
that
getting
allo
ated
an
IP
address
and
nding
the
grass
ro
ots
net
w
ork
servi es
y
ou
require
on
a
new
net
w
ork,
su
h
as
DNS,
is
to
o
hard?
While
these
are
problems
with
the
urren
t
in
ternet,
they
are
not
impassible
road
blo
ks
in
the
progress
of
the
so
alled
information
so
iet
y
or
information
e onom
y.
These
problems
ha
v
e
b
een
mitigated
b
y
new
te
hnologies
su
h
as
net
w
ork
address
translation
and
PPP
.
Indeed,
IPv6
primarily
targets
te
hni al
1
issues
whi
h
ha
v
e
already
b
een
mitigated.
Although
these
issues
ma
y
remain
imp
ortan
t,
they
are
no
longer
the
most
pressing
issues
fa ing
the
in
ternet
to-
da
y
.
Th
us,
there
is
no
univ
ersally
omp
elling
adv
an
tage
or
reason
to
mak
e
the
transition
from
IPv4
to
IPv6,
explaining
wh
y
IPv6
is
still
not
in
extensiv
e
use
a
de ade
after
its
release.
If
it
is
not
the
te
hni al
issues
addressed
b
y
IPv6
that
are
the
bugb
ear
of
the
in
ternet
then
this
logi ally
implies
that
there
are
other
problems
whi
h
need
to
b
e
iden
ti ed
and
addressed;
IPv6
fails
to
address
the
real
problems
fa ing
the
in
ternet
in
the
21st
en
tury
.
The
stri t
traditional
de nition
of
the
in
ternet
as
the
global
in
ter-net
w
ork
of
omputers
is
to
o
narro
w
and
te
hni al
to
b
e
ee tiv
e
here.
I
suggest
that
the
a
more
useful
de nition
of
the
in
ternet
is
the
global
in
ter-net
w
ork
of
agen
ts
(h
uman
or
arti ial)
omm
uni ating
using
omputers.
This
dieren e
highligh
ts
the
realit
y
that
the
outstanding
problems
of
the
urren
t
in
ternet
lie
in
the
in
ter-
a tion
of
agen
ts
(h
uman
or
arti ial),
not
omputers.
The
in
ternet
should
mak
e
it
simple
for
p
eople
using
omputers
to
do
what
they
w
an
t
to
do.
If
they
w
an
t
to
a ess
and
mo
dify
data,
then
they
should
not
b
e
hamp
ered
b
y
their
lo
ation
or
that
of
the
data.
They
ma
y
also
w
an
t
to
do
this
with
appropriate
se urit
y
and
priv
a y
.
All
other
uses
of
the
in
ternet
(a ording
to
our
de nition
of
it)
are
deriv
ativ
es
of
this.
Consider
the
follo
wing
t
w
o
examples:
(a)
publishing
w
eb
pages;
and
(b)
email.
The
rst
example,
publishing
w
eb
pages
(in
its
simplest
form)
an
b
e
mo
d-
elled
as
pro
viding
globally
unrestri ted
read-only
a ess
to
a
olle tion
of
les.
In
the
urren
t
in
ternet
this
is
implemen
ted
using
a
sp
e ial
program
(the
w
eb
serv
er)
to
mak
e
the
pages
a
v
ailable,
and
another
sp
e ial
program
(the
w
eb
bro
wser)
to
read
the
pages.
The
se ond
example,
email,
an
b
e
mo
delled
as
writing
a
le
y
ou
ha
v
e
re-
ated
(the
message)
in
to
a
w
ell
kno
wn
writable
lo
ation
in
the
re ipien
ts
storage
(their
email
in
b
o
x).
In
the
urren
t
in
ternet
this
is
implemen
ted
using
a
sp
e-
ial
program
(an
email
omp
osing
and
sending
program)
whi
h
omm
uni ates
to
another
sp
e ial
program
(the
email
deliv
ery
serv
er)
using
a
sp
e ial
proto
ol
(usually
SMTP),
whi
h
pla es
the
message
in
to
a
sp
e ial
kind
of
storage
(the
re ipien
ts
email
in
b
o
x,
whi
h
usually
exists
outside
the
name
spa e
of
the
re ip-
ien
ts
regular
This content is AI-processed based on open access ArXiv data.
