December
23 , 2005
| Part Two: The Telephone and its Disconnects
Before we can begin any discussion of a “repair”
of the telephone (in truth, it is not the telephone that needs
repair, but rather that we, who design them, must repair our
thinking), it is necessary to prick consciousness with a few
probes, interventions in the form of three questions, which
will help to illuminate the enormous weight of subconscious
and unrecognized prejudices which accompany the modern mobile
telephone. These questions outline the telephone’s “disconnects,”
where logic and reason fail, and prejudice, embodied, is revealed.
Question One: Why does
a mobile telephone have a keypad?
Automatic
dialing technology is about eighty years old [23]; before
that, calls were switched by an operator at a switchboard, who
physically connected trunk lines to make the circuit. Mechanical
automation obsolesced the operator, and the rapid advance of
computerized switching systems in the 1960s gave birth to DTMF
“touch
tone” dialing.[24] It is this “touch tone”
keypad which we have come to consider as the “natural”
interface to the telephone, even though it is not quite as old
as this author. In the years before the microprocessor revolution,
when AT&T stamped out hundreds of millions of absolutely
identical, identically dumb, touch-tone telephone handsets,
users had to peck the number into the keyboard every single
time they placed a call.
How often do you hand tap a number
into your mobile phone? I’ve been doing some informal
polling, and the answer seems to be, “About once a week.”
Although we might be using our mobile phone tens of hours a
week, we only tap a number into it once a week. If that’s
the case, why is half the interface of the mobile
telephone composed of a keypad?
We all know what the keypad is used
for, tens of times a day: SMS.
Text messaging is the killer app for the mobile telephone, and
the keypad is the interface to that service. Yet text messaging
via keypad is so slow, so fraught, even with predictive text,
that it’s amazing we bother to do it at all. It’s
a clear indication that the need to send text messages outweighs
our frustrations with the text message interface. Nevertheless,
that resentment persists, and grows with every text message
sent. It’s part of the reason why THE NEW INVENTORS audience
gave up that groan.
Why do we use the keypad for SMS?
Because that’s the interface we’ve got. And why
is that the interface we’ve got? Because this is a telephone,
and the one, absolutely uniform feature of the telephone interface
is the dialing keypad. People need to dial numbers. But no one
actually does dial numbers anymore. They’re all in our
address books. We don’t even remember numbers anymore.
I can remember my mobile and my office and my voice mail numbers.
Everything else is managed by my mobile phone – which
is as it should be. As we transition to a VOIP world, the concept
of the phone number will be entirely meaningless, replaced by
a SIP
URL. At that point, we’ll never type numbers into our
telephones.
The interface to text messaging, the
keypad, shapes all communication through the device. Consider
these contrasting examples: A few weeks ago I learned
how to compose text messages on my Macintosh, shipping them
over to my phone for delivery. It was immediately liberating;
I had the full benefit of punctuation and a dictionary larger
than the predictive text library. As an immigrant to the world
of text messaging, I found relief. On the other hand, the younger
generation has gone
in the opposite direction, adapting their communication
forms to the interface, creating a rich linguistic pidgin, which
reads like shorthand.
The information pressure behind text
messaging has been so intense that we have seen the emergence
of new SMS-based languages[25]. This pressure is pushing the
mobile telephone through a series of mutations; some look like
the Blackberry, which features a full, if miniature QWERTY keyboard.
Others favor other interface modalities, such as speech recognition.
The keypad itself is a vestigial organ. It will wither away
then disappear entirely, resurrected
as a ghost, in a virtualized, on-screen interface, for those
rare occasions when we type a number into the telephone.
Question Two: To how many
networks is the mobile telephone connected?
The telephone is essentially a network
terminal; it provides services to the user by making, managing,
and destroying connections to the network. In Australia, we
readily acknowledge that our mobile telephones sit on two networks:
GSM,
which handles lightly-digitized analog traffic[26]; and GPRS
(Global Packet Radio Service), which is a packet-switched data-only
network with full TCP/IP capabilities[27]. As third-generation
“3G”
cellular networks become pervasive, the analog-only network
is being dropped in favor of a all-digital, high-speed, packet-switched,
TCP/IP network connection. These are the networks we connect
to with our mobile telephones. But these are not all.
The average model mobile telephone
for sale today – anything that sells for more than AUD
$100 – actually presents at least four
network interfaces. In addition to the standard GSM/GPRS network
interfaces, the mobile telephone is also equipped with an IrDA
(Infrared Data Association) network interface. The IrDA interface
is a very short range, line-of-sight, point-to-point networking
protocol, which generally fades away after about 50cm of distance.
Historically, IrDA has been used to allow devices to share small
chunks of data, such as address cards, URLs, and the like. It
features a relatively slow transmission speed – reaching
an upper limit of 57Kbps – and hence is not suitable for
large file transfers. While IrDA is an inexpensive interface
to implement in a small-profile device such as a mobile telephone,
its shortcomings have kept it from widespread use.
The fourth network interface, and
by far the most interesting one, is Bluetooth.
Bluetooth is a low-power wireless networking technology which
provides a sphere of coverage of about 10 meters in radius around
the Bluetooth device. Bluetooth has been around since the late
1990s, launched with great fanfare, followed immediately by
a very public failure as many Bluetooth devices proved to be
incompatible. The market took care of these problems quickly,
and with the introduction of version 1.1 of Bluetooth in early
2001, the market for Bluetooth devices began to take off. Nearly
all mobile phones which cost more than $AUD 100 have Bluetooth
network interfaces, and, at present, five
million Bluetooth are manufactured each week[28]. This includes
mobile phones, laptops, PDAs, wireless headsets, etc.
At present Bluetooth isn’t used
for very much; it has become a more-reliable replacement for
IrDA, because it does not rely on line of sight, but rather,
physical proximity, something that’s much easier to manage.
Bluetooth is used to keep devices synchronized, or as a file
transfer protocol – a way to get photographs off the phone,
and appointments onto it. Yet Bluetooth is capable of far more
than this. Its link layer (layer 3), L2CAP,
is capable of managing “piconets”
of eight devices – one master and seven slaves. These
piconets can overlap in physical and logical space, so the same
device can be a slave in one piconet and master of another.
This means that when Bluetooth devices are brought together,
they can easily form a complex network topology. Furthermore,
this network topology need not be entirely local, with all traffic
restricted to the piconet; any Bluetooth device could act as
a gateway, routing traffic directed through it to the Internet,
or to other piconets, as required.
We haven’t seen anything like
this show up on our mobile telephones. Although these devices
have the advantage of some very sophisticated networking technology,
we’re simply using Bluetooth to push discrete blocks of
data around. We are not treating these Bluetooth devices as
nodes within a packet-switched network. Why? Because we have
not recognized the power of this highly versatile network interface.
We see the mobile telephone purely as a terminal on the GSM/GPRS
networks, when it is actually a terminal on at least four different
networks. As mobile telephones acquire 802.11 “WiFi”
capabilities – they already are, to satisfy the demands
of VOIP users – these handheld network terminals will
present five network interfaces. Yet we continue
to act as though these other interfaces simply do not exist.
Question Three: What is
the mobile telephone doing?
The modern mobile telephone –
such as my SonyEricsson
K750i – is an impressively powerful device. It has
a gigabyte of non-volatile memory, a fairly large RAM scratchpad
for program execution, an operating system (Symbian
OS), a Java Virtual Machine (J2ME),
and a host of custom ICs dedicated to digitizing audio signals,
converting digitized signals to audio, managing the UHF connection
to the cellular network, driving the Bluetooth radio, and so
forth. The K750i also has firmware which performs the vector
mathematics needed in three-dimensional transformation matrices,
so that I can play any number of 3D games.
Why would we play games on our mobile
telephones? By this we don’t mean to question the validity
of mobile entertainment, be it music or video or games of strategy
and skill. This is a big industry, earning billions of dollars
a year as people find new ways to use their mobile phones to
fill the otherwise empty moments in their lives. We are not
questioning the human desire to be entertained every single
moment of the day (although
perhaps we should)[29], but rather, we are asking if this
is an appropriate use of the mobile telephone. The mobile telephone,
like our desktop and laptop computers, suffers from consistent
underutilization; we rarely keep it busy. Even when we are engaged
in voice communication, newer mobile telephones use only a small
portion of their capabilities in call management. The mobile
telephone represents a tremendous computational resource which
is almost entirely unutilized. Hence, the phone is free for
games and other entertainments.
Why is the mobile telephone so underutilized?
Once again, we see the vestigial behavior of analog fixed-line
telephony. Fixed-line telephones did nothing until the network
sent a call to the handset, or until the user picked up the
handset to make a call. The duty cycle for the fixed-line telephone
was entirely driven by users, as the only actors within the
network. This basic assumption drives the design of mobile telephones:
the devices are essentially passive, waiting to be activated
by the network or the user. But why should this be? There’s
no essential purpose served by such passivity – far from
it. But the mobile telephone has been cursed by its ancestry,
and this curse has kept it from reaching its full potential.
This is the most important thing we must unlearn, if we are
to repair the telephone.The mobile telephone is only a passive
device because we have designed it so.
We believe it a necessary
precondition for telephone repair that we treat the mobile telephone
as an entirely active device, a network terminal which has been
designed from its outset to facilitate management of and communication
with the social network of its owner-user. The mobile
telephone is already the de facto device for digital social
network management; voice calls and text messaging are arguably
the most significant components of the electronic communication
within our social networks. The ephemeral nature of synchronous
voice communication and asynchronous text messages means that
these informational transactions are not captured by existing
digital social networks, which, in turn, means that we unconsciously
underestimate their importance, because they are not counted
(except on our monthly bills), and are not tracked, except within
the mobile handset. If we transform the mobile telephone into
an active device, and design it to be conscious of the electronic
communication which takes place through it and around it, we
have a device which can gather a wealth of data – a “data
shadow” – from which we can build emergent models
of a dynamic digital social network. The mobile telephone is
the only device which is well-suited to the task of feeding
our ever-hungry digital social networks; it is the only device
capable of recording our lives as they are lived. The
mobile telephone should be fully realized as an active device
which takes note of our digital social interactions, using this
information to assist us in improving the quality of these interactions.
This is the core design principle
that we recommend be put into practice; without this step forward,
nothing else is possible, and the mobile telephone will remain
an overpowered, underutilized twin of its mechanical-analog
ancestors. Because of the growing importance of ad-hoc digital
social networks, there is a growing pressure to consistently
improve and reinforce the connections within our social networks.
Should we succeed in transforming the mobile telephone into
the instrumentality of our social networks, that transformation
will release this pressure, driving the mobile telephone forward
into a fantastic array of mutations and forms. It will be a
Cambrian
Explosion of communication, brought into being we connect
our need to our capability.
The impact of such a transformation
would be immediate and profound. When users stop fighting the
interface, and find, instead, that the interface enables social
network management, these users will enter into a new ontological
accommodation with the device. It will, in short order, become
entirely indispensable. It will not be thought of as a device
for voice communication, or even as a terminal for text messaging;
it will be the portal into the user’s social network:
the physical, proximal and ubiquitous connection into the sphere
of human connection. There is precedent for such a rapid transformation:
in twenty-four months the
web browser grew from its origins as a hacker curiosity to become
the indispensable information age tool.
Our analysis of the requirements
for this transformation of mobile telephony from passive to
active modes indicates that the technological infrastructure
for such a revolution is already in place; this is a revolution
in software, not hardware; a revolution in usability, not deployment.
The handsets and networks are fully ubiquitous. We need only
learn how to design software to fit the needs of the network’s
users. To do that, we must experiment, play, and listen to the
users.
References
23. http://en.wikipedia.org/wiki/Automatic_telephone_exchange
24. http://en.wikipedia.org/wiki/Touch_tone
25. Journal of Applied Linguistics, 22, January 2002,
p. 481
26. http://en.wikipedia.org/wiki/GSM
27. http://en.wikipedia.org/wiki/GPRS
28. http://www.bluetooth.com/news/releases.asp?A=2&PID=1521&ARC=1
29. Neil Postman, Amusing Ourselves to Death, Penguin
Books, New York, 1986, p. 43
Continue
to Part Three >>
Mark
Pesce
is the co-creator of the Virtual Reality Modeling
Language (VRML) - the first
3D interface to the internet - and the founder of the Interactive
Media Program at USC's School of Cinema-Television. In 2000,
Ballantine Books published Pesce's The
Playful World: How Technology is Transforming our Imagination,
which explored the world of interactivity through a detailed
examination of the Furby, LEGO’s Mindstorms and the Playstation
2. In late 2003, Pesce was invited to the Australian Film
Television and Radio School, with a mandate to redesign the
curriculum to incorporate the new opportunities offered by
interactive media.
Read
Mark's blog: hyperpeople.
This work is licensed under the
Creative Commons Attribution-NonCommercial-ShareAlike License.
To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/2.0/
or send a letter to Creative Commons, 559 Nathan Abbott Way,
Stanford, California 94305, USA.